USF Libraries
USF Digital Collections

Effect of CdCl₂ treatment on CdTe and CdS solar cell characteristics after exposure to light for 1000 hours

MISSING IMAGE

Material Information

Title:
Effect of CdCl₂ treatment on CdTe and CdS solar cell characteristics after exposure to light for 1000 hours
Physical Description:
Book
Language:
English
Creator:
Rangaswamy, Ashok
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
stability
dark current
cadmium telluride
light soaking
CdCl2
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The CdTe solar cell is a leading candidate for cost-effective thin-film solar cells having demonstrated small area cell effciencies of 16.4%. A Key issue associated with CdTe thin film photovoltaic modules is the analysis of degradation behavior of the device. The analysis is complicated as changes due to degradation may be reversible. Solar cell measurement techniques were used to understand the changes in device parameters after light soaking for 1000 hours. An automated measurement setup was implemented as part of this thesis work. The main objective of this thesis was to study the effect of CdCl₂ heat treatment on the device stability. The temperature for this heat treatment was varied from 360oC to 400oC. Cells were stressed under illumination at both short circuit and open circuit conditions. It was found that the increase CdCl₂ heat treatment temperature slowed down the degradation rate.This was true for both short and open circuit stress conditions. Also short circuit stress condition slowed down the degradation of the device when compared with the open circuit condition. It became evident that the recombination current mainly got affected when the device was said to be degraded.
Thesis:
Thesis (M.S.E.E.)--University of South Florida, 2003.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Ashok Rangaswamy.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 71 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 52809421
notis - AJJ4778
usfldc doi - E14-SFE0000064
usfldc handle - e14.64
System ID:
SFS0024760:00001


This item is only available as the following downloads:


Full Text

PAGE 1

EectofCdCl2TreatmentonCdTeandCdSSolarCellCharacteristicsafterExposuretoLightfor1000HoursbyAshokRangaswamyAthesissubmittedinpartialfulllmentoftherequirementsforthedegreeofMasterofElectricalEngineeringDepartmentofElectricalEngineeringCollegeofEngineeringUniversityofSouthFloridaMajorProfessor:Chirs.S.Ferekides,PhDDonMorel,PhDY.L.Chiou,PhDDateofApproval:July11,2003Keywords:CdCl2,lightsoaking,CdTe,darkcurrent,stabilitycCopyright2003,AshokRangaswamy

PAGE 2

AcknowledgmentsIwishtodirectmyspecialthankstoDr.ChrisFerekidesforhisguidancewithkind,wisdomandexperience,Dr.DonMorelandDr.Chiouforbeingavailablewithknowledgeandusefuladvice,andtoRobert,Harish,Prasana,SuYu,Zhao,Barri,Vishwanath,Varuna,Mike,TrungandSwethafortheirhelp.Thisresearchissup-portedbyNREL.

PAGE 3

TableofContentsListofTablesiiiListofFiguresivAbstractvii1Introduction11.1Objectives................................. 2 2SolarCellsUnderIllumination32.1InterfaceStates.............................. 6 3LiteratureReview73.1EectofCopperDiusionondeviceStability............. 8 3.2EectofCdCl2TreatmentonDeviceStability............. 9 4SolarCellCharacteristicsandCharacterizationTechniques114.1Current-VoltageCharacteristics..................... 11 4.1.1Current-VoltageCharacteristicsinDark............ 11 4.1.2CurrentVoltageCharacteristicsUnderIllumination...... 13 4.1.3RollOverinIVCurvesDuetoBackContactBarrier..... 15 4.1.4DynamicResistance....................... 16 4.1.5FillFactorLossDuetoBackContactBarrier......... 16 4.2SpectralResponse............................. 17 i

PAGE 4

4.3CVMeasurement............................. 17 5Experimental205.1DeviceStructure............................. 20 5.2MeasurementSetup............................ 21 6ResultsandDiscussions256.1EectofCdCl2HeatTreatment..................... 25 6.1.1EectofLightSoakingonVocandFF............. 25 6.1.2RecombinationCurrent...................... 30 6.1.3OptimumTemperatures..................... 37 6.1.4EectofMicroDefects...................... 39 6.1.5Summary............................. 39 6.2CuxTeAsBackContact......................... 40 7Conclusion437.1RecommendationsforFutureStudies.................. 43 References45AMeasurementAutomation48BJVCharacteristics52 ii

PAGE 5

ListofTables3.1CdTeFailureModesandPossibleCauses[8]................ 7 6.1EectofMicroDefectsonResults..................... 39 B.1AverageEstimatesofandSeriesResistanceafterLS@OC........ 52 B.2AverageEstimatesofandSeriesResistanceafterLS@SC........ 52 iii

PAGE 6

ListofFigures4.1TypicaldarkJVcharacteristics....................... 13 4.2CrossoverinIVcurves........................... 15 4.3PhotonAccountingforaCdTeSolarCell[19]............... 18 5.1DeviceStructure............................... 21 5.2SampleBed.................................. 22 5.3MeasurementSetup............................. 22 6.1EectofCdCl2HeatTreatmentonDeviceParameters.......... 26 6.2TemperatureProleDuringStressPeriodTemperatureoCVsTimemin. 26 6.3DegradationBehaviorofDevicesHeatTreatedHTat360oCandLightSoakedLS@O.C.............................. 28 6.4DegradationBehaviorofDevicesHTat360oCandLS@ShortCircuitS.CCondition............................... 29 6.5DegradationBehavior:HT@400oCandLS@O.C............ 31 6.6DegradationBehavior:HT@400oCandLS@S.C............ 32 6.7JVCharacteristics.HTat360oCandLSatOC.TopandMiddle:DarkJVandBottom:LightJV.......................... 33 6.8JVCharacteristics.HTat360oCandLSatSC.............. 34 iv

PAGE 7

6.9JVCharacteristics.HTat400oCandLSatOC.............. 35 6.10JVCharacteristics.HTat400oCandLSatSC.............. 36 6.11DarkJVCharacteristics.HTat380oCandLSatTop:OCandBottom:SC....................................... 38 6.12SummaryofDegradationBehavior..................... 40 6.13DegradationbehaviorofsampleswithCu2TeasBackcontactwithThick-nessLeft:70oAandright:60oArespectively................ 42 A.1FrontPanelofVIProgramforStabilityTesting.............. 49 A.2SnapShotofBlockDiagram........................ 49 A.3CircuitCongurationofHardwareSetup................. 50 A.4InputPanelforMeasurementIntervalSpecication............ 51 B.1DegradationBehavior:HT@380oCandLS@O.C............ 53 B.2DegradationBehavior:HT@390oCandLS@O.C............ 54 B.3DegradationBehavior:HT@380oCandLS@S.C............ 55 B.4DegradationBehavior:HT@390oCandLS@S.C............ 56 B.5CrossOverEectforSamplesLS@OCandHT@Top:360oC,Middle:380oCandBottom:400oC........................ 57 B.6CrossOverEectforSamplesLS@SCandHT@Top:360oCMiddle:380oCandBottom:400oC......................... 58 v

PAGE 8

B.7JVCharacteristics.HTat380oCandLSatOC.TopandMiddle:DarkJVandBottom:LightJV.......................... 59 B.8JVCharacteristics.HTat380oCandLSatSC.............. 60 B.9JVCharacteristics.HTat390oCandLSatOC.............. 61 B.10JVCharacteristics.HTat390oCandLSatSC.............. 62 vi

PAGE 9

EectofCdCl2TreatmentonCdTeandCdSSolarCellCharacteristicsafterExposuretoLightfor1000HoursAshokRangaswamyABSTRACTTheCdTesolarcellisaleadingcandidateforcost-eectivethin-lmsolarcellshavingdemonstratedsmallareacellecienciesof16:4%.AKeyissueassociatedwithCdTethinlmphotovoltaicmodulesistheanalysisofdegradationbehaviorofthedevice.Theanalysisiscomplicatedaschangesduetodegradationmaybereversible.Solarcellmeasurementtechniqueswereusedtounderstandthechangesindeviceparametersafterlightsoakingfor1000hours.Anautomatedmeasurementsetupwasimplementedaspartofthisthesiswork.ThemainobjectiveofthisthesiswastostudytheeectofCdCl2heattreatmentonthedevicestability.Thetemper-atureforthisheattreatmentwasvariedfrom360oCto400oC.Cellswerestressedunderilluminationatbothshortcircuitandopencircuitconditions.ItwasfoundthattheincreaseCdCl2heattreatmenttemperaturesloweddownthedegradationrate.Thiswastrueforbothshortandopencircuitstressconditions.Alsoshortcir-cuitstressconditionsloweddownthedegradationofthedevicewhencomparedwiththeopencircuitcondition.Itbecameevidentthattherecombinationcurrentmainlygotaectedwhenthedevicewassaidtobedegraded. vii

PAGE 10

Chapter1Introduction Theonlyreasonfortimeissothateverythingdoesn'thappenatonce...AlbertEinsteinThestabilityofsolarcellsacquiresimportancesinceitisgenerallyassumedthatterrestrialphotovoltaicsystemsmusthaveausefullifeofatleast20years.Thinlmfabricationprocessesarebeingdevelopedaslowcosttechniqueforproducingterrestrialsolarcells.Semiconductorswithbandgapsofaround1.5eVareoptimumintermsofeciency.HenceCdTeisthematerialofchoice.Publisheddata[1][2]showthatCdTe/CdSsolarcellsmayexhibitinstabilityundercertainconditions.Ingeneral,degradationofsolarcellsoccurswhentheyarelightsoakedforpro-longedtimeand/oratelevatedtemperatures.Thereasonsfortheobserveddegra-dationmaybejunctiondegradation,degradationoftheelectricalcontacttoCdTeandshunting.Thedegradationisfrequentlyrepresentedbyperturbationofcurrentvoltagecharacteristics,thedevelopmentofrolloverathighforwardbias,adecreaseinopencircuitvoltageVoc,anincreaseintheseriesresistanceandadecreaseinthellfactorFF.ThemostsuspectedcauseforcellinstabilityisthediusionofCufromthebackcontactintotheCdTeregionandtheCdTe/CdSinterface.AsmallamountofCu 1

PAGE 11

isoftenaddedtobackcontactstoeectivelyp+dopetheCdTesurfaceandallowtheformationofbetterohmiccontactstoCdTe[3][4].AtthecelljunctionCuwasproposedtoformrecombinationcentersandshuntingpathways,limitingthelifetimeofthecell.OtherprobabledegradationmechanismsinCdTe/CdSsolarcellsareoxidation,electromigrationanddiusionofnativeandotherimpurities.1.1ObjectivesTheobjectivesofthisworkwere: toConstructandautomateameasurementsetupneededforstabilitytestingofCdTe/CdSsolarcellsundervariousstressconditionsand tostudytheeectofprolongedilluminationandCdCl2heattreatmentontheperformanceofCdTe/CdSsolarcells. 2

PAGE 12

Chapter2SolarCellsUnderIlluminationItisnecessarytorstunderstandthedevicebehaviorunderilluminationtoknowtheeectofprolongedilluminationondevicestability.Asolarcellisgenerallyap-ndiode.Inourcase,CdSwasusedasthen-typesemiconductorandCdTeasthep-typesemiconductor.ThejunctionformedbetweenCdSandCdTeiscalledheterojunction.Ifthelightpassesthroughthesubstraterst,thendeviceissaidtohaveasuperstratestructure.CdTesolarcellsforthisexperimenthadthesuperstratestructureasshowning.5.1.Whentheradiationisreceivedfromthesun,itgetsabsorbedinthesemiconductorswhichgenerateschargecarriersinthem.Themostofthesecarriersareseparatedbythejunctionduetoitsinternalelectricaleldandthencollectedatthecontactsofthedevice,thusdeliveringpowertoanexternalloadconnectedtothesolarcell.Thefollowingregionsofasolarcellareofinteresttoexplaintheelectronicprocessinvolved: ThebackcontacttoCdTe:canformbarrierwithCdTe,possiblyincreasingseriesresistanceandrollover ThebulkofCdTe:whereelectron/holepairsaregeneratedbytheabsorptionoflightoutsidethedepletionregion 3

PAGE 13

Theheterojunction:wherethechangeininterfacestatesandhighrecombina-tiontakesplace,increasingthedarkcurrent ThebulkofCdS,whichcontributestoseriesresistance,butdoesnotcontributetoJsc.TheidealizedjunctioncurrentinthedarkisgivenbyI=I0[exp[qV=AkT])]TJ/F15 11.955 Tf 11.955 0 Td[(1] .1 whereI0isthereversesaturationcurrent,kistheboltzmanconstant,qiselectriccharge,TistemperatureinokelvinandAisthediodefactor.SomeheterojunctionsolarcellstructuresshowavariationinIoandAwithil-luminationlevelandwavelength.Thisisbecause,whentrappingcentersatornearjunctionarenotingoodthermalcommunicationwithconductionorvalencebandsslowstates,theiroccupancyandhencechargecanbechangedbyillumination.Alsotheionizeddonororacceptordensityinthenorptypesemiconductormaychangeuponillumination,whichinturnmodiesthedepletionlayerwidth,theshapeofthejunctionbarrier,andnallythejunctiontransport[7].Underillumination,theequation2.1becomesI=I0[exp[qV=AkT])]TJ/F15 11.955 Tf 11.955 0 Td[(1])]TJ/F18 11.955 Tf 11.955 0 Td[(IL .2 WhereIListhelightgeneratedcurrentdeterminedbytheprocessesinCdTeduringillumination. 4

PAGE 14

Equation2.2isvalidonlyiftheshuntRshandseriesRsresistancesofthesolarcellapproach1and0respectively.Inpracticaldevices,thisisnottrue.Equation2.3includestheeectoftheseresistances.I=I0[exp[qV)]TJ/F18 11.955 Tf 11.955 0 Td[(IRs=AkT])]TJ/F15 11.955 Tf 11.955 0 Td[(1]+[V)]TJ/F18 11.955 Tf 11.955 0 Td[(IRs=Rsh])]TJ/F18 11.955 Tf 11.955 0 Td[(IL .3 HeretheshuntresistanceRshmayaccountfortheeectsofthecurrentpathsprovidedbyimperfectionslikepinholes,surfacerecombination,threedimensionalimperfectionsinthejunctionetc.Rsisduetobulkresistancesoflayerspresentinthedevice,frontandbackcontactresistancesetc.Ifthethedeviceiskeptopeni.e.noload,thepotentialdierenceacrossthefrontandbackcontactsismeasuredastheopen-circuitvoltage,Voc.Whenthecontactsareshortedwithzeroresistanceload,theresultantphotogeneratedcurrentisknownasshortcircuitcurrentIsc.Jscisthecorrespondingcurrentdensity.Atacertainload,theoutputpowerismaximizedPmax.TheratiobetweenPmaxtotheproductofVocandJsciscalledthellfactorFFwhichisthemeasureofthe"squareness"ofthecurrent-voltagecharacteristicsunderillumination.Rshaectsthecurrent-voltagecharacteristicsinthelowvoltageregionandRsathighcurrentregionofcurrent-voltagecurve. 5

PAGE 15

2.1InterfaceStatesOpticalabsorptionbystatesattheinterfaceandinthebulkmaterialnearthejunctionproduceschangesinthejunctionprole,andhencechangesinIoandAuponillumination.TheinterfacestatesdensityNssaectsthecarriertransport.Thesesurfacestatesactasarecombinationcenters,whichcanprovideatunnellingpathforthecarriers[6].Theincreaseininterfacestatesdensityduetoilluminationwillleadtoadecreaseinthebarrierheightbandconsequentlyenhancethesaturationcurrent.Thecon-sequencesofthedecreaseinpotentialbarrierheightareanincreaseinbothdarkcurrentandseriesresistance,andadecreaseinopencircuitvoltage,photocurrentandconversioneciencyofthesolarcell[6]. 6

PAGE 16

Chapter3LiteratureReviewItisimportanttounderstandthefailuremechanismsthatleadtocelldegradation.Table3.1listsmajorfailuremodesinCdTe/CdSsolarcellsalongwiththepossiblecauses. Table3.1CdTeFailureModesandPossibleCauses[8] FailureModes PossibleCauses Mainjunction,increasedrecombination Diusionofdopants,impuritiesetc.,electromigration Backbarrier,lossofohmiccontact Diusionofdopants,impuritiesetc.,electromigration Shunting Diusionofmetals,impuritiesetc. Thesemechanismscancausethefollowingchangeinthecurrent-voltageI-Vcharacteristics: anincreaseinRocandRollover"inI-VcurvesatvoltageslargerthanVoc,possiblyfromthebackcontactbarrierformation, areductioninVocand 7

PAGE 17

anincreaseintheslopeofI-Vcurve,throughJsc,possiblyfromincreasedshuntingorfromincreasedrecombinationinthespacechargeregion[9].RocistheslopeoflightI-VcurveatVoc.ItisfoundthattheformationofbackcontactbarrierleadstohighvaluesofRocandRollover".AhighseriesresistancecanalsoleadtoanincreaseinRoc.ANonohmicbackjunctioncanbedueto[10] lossinacceptordensityinthep+layerand Oxideinterfacelayer.Alossinacceptordensityismainlycausedbydiusionofimpurities.Here,Cuwithdiusioncoecientof5E-14cm2=secat100oCandactivationenergyof0.66eVisaprimesuspect.Thedegradationalsodependsonbias,asthechargestateofvacanciescanaecttherateofdiusionofanatom.Also,grainboundarieshavesomeeectondiusion,electromigrationofchargedatoms,andlocalcompensationeectsindegradation[10].3.1EectofCopperDiusionondeviceStabilityEventhoughCuformsbetterohmiccontactatthebackcontactinterface,itcandegradethecell,bydiusingthroughtheCdTelayertothejunctionandCdS,asCuisthefastdiuserinsinglecrystalCdTeDiusionCoecientD3E)]TJ/F15 11.955 Tf 11.444 0 Td[(12cm2.Cuasaninterstitialion[Cu+i]givesshallowdonorstateorsubstitutesCdatomtoform 8

PAGE 18

adeepacceptorstateinsinglecrystalCdTe.Inpolycrystallinelms,thediusionofCopperisevenmoreasthesurfacebondsareweakerandtwosurfacesareavailableatgrainboundaries.ElectriceldmayalsoenhanceCudiusion."Rollover"afterlightsoakingindicatesthedecreaseinCudensitynearthebackcontactandpossibleincreaseofCudensitynearthejunction[10].WhencellswithandwithoutCuininthecarbonbackcontactwaslightsoaked,onlycellswithCushowedsomerecoveryafterrestedindarkfor6-12months.ThissuggeststhatCucanreturntotheCdTe/backcontactinterfaceundersomecircumstances[10].AlsowhentheCucontentinthebackcontactwasincreased,VocandFFdegradedfaster[10].OnepossiblemechanismwhichdrivestheCuintojunctionandCdSmaybegrainboundarydiusionandsurfacereactiontoformCu-Sbonds.ThisprocesswouldbeaidedbyS/TeinterdiusionandalsoduetothefactthatCu-SbondisstrongerthanCu-Tebondasderivedfromtheheatsofformation[10].ItistobenotedherethattheinterdiusedCdSxTe1)]TJ/F19 7.97 Tf 6.586 0 Td[(xregionadjacenttotheCdSacquiresalowconcentrationofCucomparedtothelessTe-richCdS[10].WhilethebuiltinvoltageslowstheconcentrationgradientdrivenCudiusion,forwardbiasand/orlightlowersthatbarrierfordiusion[11].3.2EectofCdCl2TreatmentonDeviceStabilityAfterCdCl2treatment,ClmovesviagrainboundaryGBdiusionthroughtheCdTelayer.TheaccumulationofClneartheCdSinterfaceisduetothegreaterGB 9

PAGE 19

areainthesmaller-grainCdSlayercomparedtotheCdTe.ThisisalsoduetothefactthatClofatomicradius167pmisexpectedtosubstituteforS0pminCdSoverTe7pminCdTe[2].Thisaccumulationimprovestheinitialperformance.ThenetacceptorconcentrationneartheinterfaceincreasesbyincreasingthetemperatureofCdCl2heattreatment[2].AlsostructuralchangesinCdTethinlmsoccuronlywhenCdCl2ispresentduetorecrystallizationandsubsequentgraingrowth.Recrystallizationisthefunctionoflattice-strainenergy,andinitialstrainenergyincreasesduetoCldiusion[12].IthasbeenshownthatduringvaporCdCl2treatmentdiusionofCdSintotheabsorberlayerproceedsfasterthanthediusionofCdTeintowindowlayerandisenhancedbyreactiontemperature[14]. 10

PAGE 20

Chapter4SolarCellCharacteristicsandCharacterizationTechniques4.1Current-VoltageCharacteristics4.1.1Current-VoltageCharacteristicsinDarkUnderthermalequlibrium,whenthediodeiszerobiasedorreversebiased,nocurrentowsduetothepotentialbarrier.Whenitisforwardbiased,barrierpotentialisreduced,holesorelectronsfromthepornregionareinjectedtocreateexcessminoritycarrierinnorpregion.Inlowinjection,themajoritycarrierconcentrationdoesntchangesignicantly.Butminoritycarrierschangeseveralordersofmagnitude.Thetotalcurrentinthejunctionisthesumoftheindividualelectronandholecurrents.Sincetheelectronandholecurrentsarecontinuousfunctionsthroughthejunction,thetotalpncurrentwillbesumofminoritycarrierdiusioncurrents.Sinceitisassumedthattheelectriceldatspacechargeedgesiszero,thereisnominoritydriftcurrent.Soforidealp-njunctiondiode,currentdensityequationindarkis,J=Js[expqV kT)]TJ/F15 11.955 Tf 11.955 0 Td[(1] .1 whereJs=qn2i[1 Nas Dn n0+1 Nds Dp p0] .2 HereJsiscalledidealreversesaturationcurrentdensity. 11

PAGE 21

Butthisidealdiodeequationneglectsanyeectsoccurringwithinthespacechargeregion.Sinceothercurrentcomponentsaregeneratedwithinspacechargeregion,theactualI-Vcharacteristicsdeviatefromtheidealone.Theseadditionalcurrentsaregeneratedfromtherecombinationprocesses.Underreversebias,sincenumberofelectronsorholesaresaidtobezerointhespacechargeregionastheyaresweptaway,electronsandholesaregeneratedtoreestablishthermalequilibrium.Thisreversebiasgenerationcurrentshouldbeaddedtogetthetotalreversesaturationcurrent.Itisgivenby,J0=Js+Jr0 .3 WhereJ0andJsaretotalandidealreversesaturationcurrentdensities.Jr0isthereversegenerationcurrentinthespacechargeregionandisgivenby,Jr0=qniW 2O .4 WhereWisthedepletionwidthandOtheaveragecarrierlifetime.Underforwardbias,someexcesscarriersareinjectedintothespacechargeregion.Recombinationcurrentdensityisthen,Jrec=Jr0exp[qV 2kT] .5 Henceingeneral,thediodeequationbecomesJ=J0[expqV AkT)]TJ/F15 11.955 Tf 11.955 0 Td[(1] .6 12

PAGE 22

Figure4.1TypicaldarkJVcharacteristics whereAisthediodefactor.Foralargeforwardbiasvoltage,A1whendiusiondominatesandforalowforwardbiasvoltage,A2whenrecombinationdominates.Thereisatransitionregionwhere1
PAGE 23

Butinrealdevices,thereisacross-overofdarkandlightJ-Vcharacteristics.Suchcross-overrelatestochangeindiodeparameterswithillumination;thelightmaycausealoweringofthejunctionbarrier.AlsothereischangeindI=dVatsmallreversebiasasthelightintensityincreases.Sinceconductionisnegligibleatthisbias,thechangeinslopemustbearesultofavoltagedependentcurrentsource.ThecrossovercanbeexplainedbythecollectionfunctiongV,thefactorthatincludesthebiasdependanceoflightgeneratedcurrent.ThiseectisincludedtoJas,J=J0expqV AkT)]TJ/F15 11.955 Tf 11.955 0 Td[(1)]TJ/F18 11.955 Tf 11.955 0 Td[(gVJL .7 wheregVisthecollectionfunctionandJListhevoltageindependentlightgeneratedcurrent.Sinceboththecurrenttransportandjunctionbarriermaybesensitivetolight,thediodefactorAoftentakesondierentvaluesastheilluminationintensityandwavelengthisvaried.LightsaturationcurrentJOLisrelatedtodarksaturationcurrentJ0as,JOL=g gVocJ0 .8 JOLisslightlyhigherthanJ0duetocollectionfunction.Thereisalsothechangeinjunctionbarrierwhencellsarelightsoaked[17]. 14

PAGE 24

Figure4.2CrossoverinIVcurves 4.1.3RollOverinIVCurvesDuetoBackContactBarrierTherolloverofIVcurvesoccursbecausethejunctionvoltagesaturatesathighbias.Thissaturationoccursbecauseofabackcontactbarrier.Thesaturationvoltagesforrolloverunderdarkandilluminationaregivenby[18],Vsdark'nkT qlnJc Js .9 andVslight'Vsdark+nkT qln[1+Jc Js] .10 respectively.WhereJcisthecontactsaturationcurrentdensityandJ0isthetotalreversesaturationcurrentdensity. 15

PAGE 25

4.1.4DynamicResistanceThederivativedV=dJisusedtounderstandlightanddarkJVcharacteristicsmoreclearly.Duetothenonlinearityofthephysicalphenomenonofthesolarcell,thecurrentvoltagecharacteristicsisnonohmic.ThisbehaviorisshownintheplotofthederivativedV=dJwithrespecttovoltage.Theminimumresistancecorrespondingtoachosenvalueofhighforwardcurrentsay0.08A=cm2canbetakenasRs,thoughthisquantitydoesn'treecttherealvalueofseiesresistanceofthedevice[20].Howeverthisquantitityisusefultotheextent,canrepresentthedegradationofthedevicequantitatively.TheshuntresistanceRshcanbeapproximatedbytheslopeofthecurrentvoltagecharacteristicsinthelowvoltageregionsay,-1.7Vto-1V.4.1.5FillFactorLossDuetoBackContactBarrierThenon-idealFFisgivenby[18],FFFFoV1oc[1)]TJ/F18 11.955 Tf 16.013 8.088 Td[(kT qVocln[1+JL Jc]] .11 where,FFoV1oc[1)]TJ/F15 11.955 Tf 13.15 8.088 Td[(lnV1oc V1oc][1)]TJ/F15 11.955 Tf 17.766 8.088 Td[(1 V1oc][1 1)]TJ/F15 11.955 Tf 11.955 0 Td[(exp)]TJ/F18 11.955 Tf 9.299 0 Td[(V1oc]andV1oc=qVoc)]TJ/F15 11.955 Tf 11.955 0 Td[(V nkT: 16

PAGE 26

WhereVistheforwardbiasatthebackcontactwhenlightcurrentowsunderreversebias.WhenthebackcontactdiodesaturationcurrentJcislargerthanorequaltothelightcurrent,theFFlossduetothebackcontactisonlyafewpercent.Incontrast,considerableFFlossesoccur,ifJcisconsiderablysmallerthanJL.4.2SpectralResponseThereareavarietyofpossibleopticallossesbeforephotonsreachasolarcell'sabsorber,andthereareadditionallossesfromnonradiativerecombinationorfromphotonsexcitingtheabsorber.Figure4.3isoneexampleofthefractionofphotonsofeachwavelengththatcontributedtophotocurrentandthefractionsthatarelostineachofseveralways.Thelossincludesthereectionfromthecell,theabsorptionoftheglasssubstrate,theabsorptionoftheSnO2contact,andtheabsorptionoftheCdSwindowlayer.SomephotonsarelostduetodeeppenetrationatCdTeregion.4.3CVMeasurementInformationlikedopingprole,depletionwidth,acceptorconcentrationandbarrierheightofthejunctionbcanbeobtainedfromCapacitance-VoltageCVcharacter-istics.AstheCdSlmhascarrierconcentrationapproximately1E+16to1E+17electrons/cm2severalordersofmagnitudehigherthanthatofCdTeapproximately1E+13to1E+15holes/cm2,thedepletionlayermainlyspreadsintotheCdTeab-sorberlayertoseparatethephoto-generatedcarriers.Henceonlythepermittivityof 17

PAGE 27

Figure4.3PhotonAccountingforaCdTeSolarCell[19] CdTelayerisusedtocalculatethedepletionlayerwidthWd.ThiswidthchangesastheexternalbiasVisappliedacrossthecontactsofthesolarcell.ThedepletionlayerwidthWdatparticularbiasViscalculatedusingWd=or C=A .12 Whereoispermittivityoffreespace8.854E-14F/cm,risrelativepermittivityofCdTelayer0.2,CisthecapacitanceFatthatbiasVandAistheeectivecontactareaofthedevice.ThenetacceptorconcentrationisthenfoundfromNA=2 [qor][d[C2=A2]=dV] .13 whereqiselecticcharge,1.6022E-19Coulombs. 18

PAGE 28

Theaccuracyofpredictingrealvaluesusingequations4.12and4.13maybeaectedduetofollowingreasons: Theequation4.13isvalidonlyforhomogeneouslydopeduncompensatedp-typeCdTe.ForpartiallycompensatedCdTelayer,thevalueequalsNA)]TJ/F18 11.955 Tf 12.222 0 Td[(NDasNA>ND. Theequationscompletelyneglectminoritycarriersandassumetotaldepletionofmajoritycarriersinthespacechargeregion[SCR].ThisisvalidonlywhentheSCRisreversebiasedandsubstrateisuniformlydoped. Alsotheseequationsarebasedonasinglejunctionmodel.ButCdTe/CdShasseparatecapacitancecomponentsforthemainjunctionaswellasthejunc-tionduetobackcontactschottkydiode.Thistwodiodecircuitalsoaectstheaccuracyofvalues.Thoughthesesourcesoferrorsaecttherealdeviceparametervalues,arelativecomparisonofthedeviceparametervaluesmaybevalidingeneral. 19

PAGE 29

Chapter5Experimental5.1DeviceStructureThedeviceusedforthisstudyhadthesuperstratestructurethatislightentersthroughsubstraterst,whichis7059Borosilicateglassasshownintheg.5.1.Thedepositionproceduresaredescribedindetailelsewhere[22].Inbrief,thefrontcontactSnO2wasdepositedbytheMOCVD[MetalOrganicChemicalVaporDeposition]techniqueandwasdepositedaslow)]TJ/F18 11.955 Tf 12.942 0 Td[(=high)]TJ/F18 11.955 Tf 12.942 0 Td[(bilayer.ThesheetresistanceofSnO2wasbelow10=2.Cadmiumsuldeofapproximately1000AwasdepositedusingChemicalbathdepositionCBDmethod.TheCdTedepositionwascarriedoutbyClosedspacesublimationCSSmethodat550)]TJ/F15 11.955 Tf 12.363 0 Td[(600oCassubstratetemperature.TheCdTe/CdSstructurewasthenheattreatedinthepresenceofCdCl2.ThemainobjectiveofthisthesisworkwastoanalyzethetheeectofCdCl2heattreatmentondevicestability.Forthis,theCdCl2annealingtemperaturewasvariedfrom360to400oC.ExcessCdCl2wasthenremovedbyetchingthesamplesinmethanol/brominesolutionfor8secondswhichresultsinsmoothTerichsurfaceforcontacting.CudopedgraphitepasteorsputteredCu2Tewasappliedasbackcontactandheattreatedat270oCfor25minutesinvacuum. 20

PAGE 30

Figure5.1DeviceStructure 5.2MeasurementSetupThemeasurementsetupshowning.5.3consistsofavacuumchamberwhichcanbeevacuatedandbacklledwithultrahighpurityN2gaspriortothebeginningoflightsoaking.AlldevicesundertestwerefabricatedunderidenticalconditionsexcepttheannealingtemperatureoftheCdCl2treatment.EightidenticalcellsforeachannealingtemperaturewerestressedunderonesunAM1.5illuminationintensityatshortcircuitand4atopencircuitconditions.Thecellswereplacedonthecopperplatesamplebedasshowning.5.2.Thesampleswereplacedinsuchawaythatthefrontcontactglasssidefacingthelightandbackcontactwasincontactwithathinfoil.Theheatistransferredfromthedevicestocoolingwaterwiththehelpofhighthermalconductivitycompound. 21

PAGE 31

Figure5.2SampleBed Figure5.3MeasurementSetup 22

PAGE 32

ThelightintensitywascalibratedtoapproximatelyAM1.5conditions15%.SampleswereundercontinuouslightstressandN2ambientduringthemeasure-ment.Thesampleswerekeptatdarkandlightcycleof4hourseach.Thetemperatureofthecopperplatesamplebedwascontrolledusingwatercirculation.ThelightsourcewasmadeusingGE12V,71W25oBeamMR16lamps.Twosetsof10lampseachwereusedtoilluminatethesamplesinthefrontandbackrowofthesamplebed.ThesampleleadswereconnectedtotheKeithleySourceMetervialowresistancecopperelectricalwires.Theleadswereattachedtothefrontandbackcontactswhichissilverandindiumrespectivelyusingconductivesilverepoxy.Abriefheattreatmentof5minuteswasgiventothesamplesat100oCforgoodadhesion.Thesampleswerekeptatvacuumdesiccatorbeforelightsoakingtoavoidpossibledegradationcausedbyhumidity.Acleansodalimeglasswasusedtopressthesamplestothesamplebedforbetterheattransfer.Thelightwasdiuseduniformlythroughoutthesamplebedusingquartzplates.ThetemperaturewascontinuouslymonitoredandcontrolledusingthermocouplesandEuroThermControllerswhichcontrolthewaterow.Thesystemwasautomatedduringthisworkandthedetailsaregivenintheappendix.Routinecurrent-voltageIVmeasurementsweredonewhenthesampleswerebe-ingstressedinsidetheoven.AKeithley2400sourcemeterwasusedforbothVoltage 23

PAGE 33

sweepandcurrentmeasurement.Thevoltagesweepwasfrom-2Vto+2Vwithstepsof0.01V.ItshouldbementionedthatthefollowingeventscouldhaveanimpactontheJVresultspresentedinthisthesis. Thelampsfailed6-7timesduringlightsoakingwhichintroducedextendeddarkperiodandinturncouldleadtounpredictablerecoveryinJVparametersofsomecells. Therewasawaterleakinsidethechamberafter200lightsoakedhours,thoughthecellsdidnotgetwet.Capacitance-VoltageCVandCapacitance-frequencyCFmeasurementsweredoneusingHPimpedanceanalyzer4145A.Spectralanalysisfordesiredbandwidthofwave-lengthsofphotonswasdoneusingOrielCornerstonemonochromatormodel74100withlightsourceofGE400W/120Vquartzlinelampmodel43707.Siliconreferencewasusedtoadjustthelightintensitypriortomeasurement. 24

PAGE 34

Chapter6ResultsandDiscussions6.1EectofCdCl2HeatTreatmentThesampleswerepreparedasdescribedinsection5.2.ThetemperatureofCdCl2heattreatmentwasvariedfrom360oCto400oC.ThedependanceofthedeviceparametersontheCdCl2treatmentwasgivening.6.1.TheOpencircuitvoltageVocwasintherangeof800-850mVfordierentannealingtemperatures.ThellfactorFFhadamaximumforthecellsannealedat390oCanddecreasedforbothhigherandlowertemperatures.ThedeviationinFFwaslargerathightemperatures.TheresultswereingoodagreementwiththatofOkamatoet.al[23].Thesampleswerethenlightsoakedundertheconditionsspeciedinsection5.2.Forthisexperiment,eightcellswereselectedforeachannealingtemperature.FourofthemwerestressedatOpencircuitOCandfouratshortcircuitSCconditions.Itshouldbementionedthatthemaximumoperatingtemperatureduringthelightsoakingperiodcouldbe60oC10oCreferg.6.2dependingonthelocationonthesamplebed.6.1.1EectofLightSoakingonVocandFFFigure6.3showsthechangeinVocandFFforsamplesannealedat360oCandlightsoakedatOC.The1stVocdatapointwastakenatoperatingtemperature,which 25

PAGE 35

Figure6.1EectofCdCl2HeatTreatmentonDeviceParameters Figure6.2TemperatureProleDuringStressPeriodTemperatureoCVsTimemin. 26

PAGE 36

isthereasonforitbeingapproximately70-80mVhigher.Excludingthiseect,theVocwasincreasedinitiallyupto10hoursoflightsoakingandstayedconstantessentiallythereafterforthesesamples.Near1000hoursoflightsoaking,theVocappearstodecrease.TheFFwasnearlyconstantupto100hoursoflightsoakinganddecreasedthereafter.ThemeasurementsatroomtemperatureatAM1.5simulatoraftertakingthesamplesoutfromthestabilityovenarealsogivenforcomparison.Figure6.4showsthechangeinVocandFFforsamplesannealedat360oCandlightsoakedatSC.TherewasnearlynodierenceintheVocbehaviorwhencom-paredtotheOCstresscondition.WhereastheFFwasnearlyunchangedupto300hoursoflightsoaking,anddecreasedthereaftersample5-20A-1.Sample5-20B-1showedadeviationfromthisbehavior.Itcanbespeculatedthatthedeviceswerenotidenticalasassumed,thoughthereasonsarenotclearatthistime.AstheoperatingtemperatureincreasedduringtheONcycle,theVocmeasuredduringthe4thhouroflightsoakingwasalwayslowerthanthatof1sthour.Figures6.5showsthechangeforsamplesannealedat400oCandlightsoakedatOC.ThechangeinVocwasessentiallythesamewhencomparedto360oCsamples.ButtheFFwasimprovedandstayedunchangedupto400hoursoflightsoakinganddroppedthereafter.Figure6.6showsthechangeinVocandFFforsamplesannealedat400oCandlightsoakedatSC.TherewasessentiallynochangeinVoc.WhereastheFFwas 27

PAGE 37

Figure6.3DegradationBehaviorofDevicesHeatTreatedHTat360oCandLightSoakedLS@O.C 28

PAGE 38

Figure6.4DegradationBehaviorofDevicesHTat360oCandLS@ShortCircuitS.CCondition 29

PAGE 39

decreasedinitiallyandstayedessentiallyconstantthereafterupto1000hoursoflightsoaking.Itisnotclearthatwhatcauseddatascatteringnear1000hours.6.1.2RecombinationCurrentTheincreaseincurrentatsmallvoltagesbelow0.5VfordarkJVcurvescanbeidentiedaseither"shunting"orincreaseinrecombinationcurrents.FromtheJ-Vitappearsthatthedarkcurrentatlowvoltages.3Visdominatedbyshunting.Butinthe0.4-0.75Vrange,italsoappearsthattherecombinationcurrenthasincreased.However,thisdarkshunting"doesnotappeartoleadtolightshuntingastheslopeofthelightJVishigher.Forsamplesannealedat360oCandlightsoaked@OC,themajorincreaseinrecombinationcurrentand/orshuntingoccurredinitiallywithin100hoursoflightsoakingreferg.6.7.Whereas,theSCstressconditionlimitedthisincreaseinitially,thoughchangesafter1000hoursoflightsoakingweresamereferg.6.7.Thecellsshownwerefromthesamesubstrate-20A-1forbothSCandOCstressconditions.Forsamplesannealedat400oCshowningures6.9and6.10,therecombinationcurrentand/or"darkshunting"wasincreasedafter100hoursoflightsoakingatOCsimilarto360oCcase,thoughchangeswerelesscomparedtothatofthe360oCsam-ples.Inthiscasealso,S.Cstressconditionprovidedgoodcontroloverrecombinationcurrentchange.Tosummarize,thechangeintheCdCl2heattreatmentorstressconditionessen30

PAGE 40

Figure6.5DegradationBehavior:HT@400oCandLS@O.C 31

PAGE 41

Figure6.6DegradationBehavior:HT@400oCandLS@S.C 32

PAGE 42

Figure6.7JVCharacteristics.HTat360oCandLSatOC.TopandMiddle:DarkJVandBottom:LightJV. 33

PAGE 43

Figure6.8JVCharacteristics.HTat360oCandLSatSC 34

PAGE 44

Figure6.9JVCharacteristics.HTat400oCandLSatOC 35

PAGE 45

Figure6.10JVCharacteristics.HTat400oCandLSatSC 36

PAGE 46

tiallydidnotaectthechangeinVocduringlightsoaking.WhereastheincreaseinthetemperatureofCdCl2heattreatmentsloweddownthedropinFF.AlsowhenthedevicewasstressedatSC,thedropinFFwasdelayedwhencomparedtothesamplesatOC.TheincreaseinrecombinationcurrentwererapidiftheCdCl2heattreatmentwasatlowtemperature.Also,theSCstressconditionsloweddownthechangeintherecombinationcurrentand/ordarkshunting"comparedwithOC.6.1.3OptimumTemperaturesThechangeinVocandFFforsampleswithannealingtemperaturesof380oCand390oCduringlightsoakingperiodwasshownintheappendix.Poorcontactforsample5-13B-1resultedinlossofdata.Thedegradationratewasminimumforthisoptimumannealingtemperaturesof380and390oC.Onlyafter750hoursoflightsoaking,therewasachangeintherecombinationcurrent/darkshunting"region.Stillthesechangeswerelowcomparedwiththe360oCsamplesatOC.Incaseofshortcircuitcondition,therewasnochangeevenafter1000hoursoflightsoaking.Thisimpliedthatthedeviceannealedatoptimumtemperaturegavegoodperformanceduringlightsoaking.Tosummarize,thedevicesannealedatoptimumtemperatureexhibitedsmallerincreaseintheirrecombinationcurrentand/ordarkshunting.AlsotheSCconditionappearstocausesmallerchangesinallinstances. 37

PAGE 47

Figure6.11DarkJVCharacteristics.HTat380oCandLSatTop:OCandBottom:SC. 38

PAGE 48

6.1.4EectofMicroDefectsMicrodefectssuchaspinholeshadagreatimpactontheperformanceinthedeviceshownintable6.1.Thedevicethatwaslightsoakedfor1000hourshadthelowVocandFF.Afterbreakingthisdevice,theVocandFFofonepartincreaseddramatically.Thisshowsthatevenaverysmalllocalizeddefectinadevicecancausetheentiredevicetobefailed. Table6.1EectofMicroDefectsonResults Device VocmV FF% Deviceaswhole 270 29 Firsthalf 810 64 Secondhalf 160 28 6.1.5SummaryThechangeinVocandFFunderOCandSCstressconditionsaresummarizeding.6.12byaveragingsimilardevicesundersamestressconditions.ThechangeintheCdCl2heattreatmentorstressconditionessentiallydidnotaectthechangeinVocduringlightsoaking.WhereastheincreaseinthetemperatureofCdCl2heattreatmentsloweddownthedropinFF.AlsowhenthedevicewasstressedatSC,thedropinFFwasdelayedwhencomparedtothesamplesatOC.Alsothe 39

PAGE 49

Figure6.12SummaryofDegradationBehavior devicesannealedattheoperatingtemperaturesof380oCand390oCgavethebetterperformanceduringlightsoaking.Theeectofthestressconditiononthedevicedarkcurrentsdecreased,astheCdCl2annealingtemperaturewasincreased.Thedegradationcanbeattributedmainlytothechangeinrecombinationcurrents.Seriesresistanceswerecalculatedasexplainedinsection4.1.4.Theresultsaregivenintheappendix.6.2CuxTeAsBackContactEventhoughadetailedstudyoftheeectoftheCuxTebackcontactondevicestabilitywasnotdone,somecellswerestressedduringtheinitialphaseofthiswork.Thedevicestructurewas7059glasssubstrate/SnO2/CBDCdS/CSSCdTe/Cu2Te. 40

PAGE 50

TheCuxTecontactwasappliedusingoptimizedproceduresdescribedelsewhere.Briey,CuxTewassputteredontoCdTeafterthesurfacewasetchedinBr2/Methanol.ThethicknessoftheCuxTewasvaried.FollowingCuxTedeposition,Mowasde-positedtoathicknessof1m.ThedarkandlightJVcharacteristicsforsampleswithCu2Teof60Aand70Arespectivelywereshowning.6.13.TheJ-Vofof70Ashowsthatthedevicehasabackcontactbarrier,whichafterlightsoakingfor1000hoursincreasesseverelylimitingtheFF.TheJ-Vof60Ashowsnobarrier.After1000hours,therewasadropinVocwithoutrollover.Thisconveysthatthereisaconditionfor"right"amountofCutoatleastlimitthedegradationofthebackcontact. 41

PAGE 51

Figure6.13DegradationbehaviorofsampleswithCu2TeasBackcontactwithThicknessLeft:70oAandright:60oArespectively. 42

PAGE 52

Chapter7ConclusionLightsoakingexperimentofCdTe/CdSsolarunderdierentstressconditionswasconductedupto1000hours.Thecellsaremadeatdierentprocessconditionstoobservetheeectofprocessconditionsondevicestability.ThechangeintheCdCl2heattreatmentorstressconditioncausedapproximately5%changeintheVocregardlessofstressorprocessconditions.WhereastheincreaseinthetemperatureofCdCl2heattreatmentsloweddownthedropinFF.AlsowhenthedevicewasstressedatSC,thedropinFFwasdelayedwhencomparedtothesamplesatOC.WhentheCdCl2heattreatmentreachedoptimumtemperatureof380)]TJ/F15 11.955 Tf 9.734 0 Td[(390oC,thedegradationwasminimum.Thedegradationcanbeattributedmainlytothechangeinrecombinationcurrent.7.1RecommendationsforFutureStudiesAsthedurationofthisinvestigationistooshorttomakecompleteanalysisofstabilitybehaviorofCdTe/CdSsolarcells,thefollowingstudiesarerecommendedtobeconductedinfuture. Therangeofstressconditionscanbewidenedtocharacterizetheireectonstabilityofdevicesproducedundervariousprocessconditions. 43

PAGE 53

Advancedanalyticaltechniquesmustbeusedtorevealthecourseforthein-creasesinthedarkcurrents. SincethebackcontactalsocontainsHgTe,theEectofHgondevicedegrada-tioncanalsobeanalyzed. 44

PAGE 54

References 1. S.E.Asheret.al,DeterminationofCuinCdTe/CdSDevicesBeforeandAfterAcceleratedStressTesting,IEEEPVSpec.Conf.,Vol.28,pp.479-48200 2. D.Cahenet.al,OvercomingDegradationMechanismsinCdTeSolarCells,FinalReport,WeizmannInstituteofScience,Israel,NREL1998-2001 3. D.L.Btzneretal,ThinSolidFilms,Vol.387,No.1-2,pp.151-15401 4. S.S.Hegedusetal,AnalysisofStressInducedDegradationinCdS/CdTeSolarCells,NCPVProgramReviewMeeting,Denver,CO,April16-192000 5. Su-HaiWeiet.al,ElectronicstructuresanddefectphysicsofCd-basedsemicon-ductors,IEEEPVSpec.Conf.,Vol.28,pp.483-48600 6. E.HRhodericket.al,J.Appl.Phys,D:App.Phy.,Vol.4,pp.1589-160171 7. AlanL.FahrenbruchandRichardH.Bube,FundamentalsofSolarCells,AcademicPress,NewYork,pp.2341983 8. T.J.McMahonet.al,ProgresstowardaCdTeCellLifePrediction,NCPVPho-tovoltaicsProgramReview,CP462,pp.54-6199 9. R.CPowelet.al,StabilityTestingofCdTe/CdSThinFilmPVmodules,IEEEPVSpec.Conf.,Vol.25,pp.785May,1996 10. K.D.Dobsonet.al,StabilityofCdTe/CdSThinFilmSolarCells,SolarEnergyMaterialsandSolarCells,Vol.62,pp.295-235000 11. D.Grecuet.al,Appl.Phys.Letter,Vol.75,No.6,pp.361-363999 12. H.R.Moutinhoet.al,StudiesofRecrystallizationofCdTeThinFilmsAfterCdCl2Treatment,IEEEPVSpec.Conf.,Vol.26,NREL997 13. K.M.AlShibani,EectofIsoThermalAnnealingonCdTeandthestudyofelectricalPropertiesofAu-CdTeSchottkyBarriers,PhysicaB.,Vol.322,pp.390-39602 45

PAGE 55

14. B.EMcCandlessetal,25thPVSCConference,IEEE,Vol.25,pp.7811996 15. R.Ludeke,ASurveyofOpticalandElectricalPropertiesofThinFilmsofII-VISemiconductingCompounds,J.VacuumScienceandTech.,Vol.8,pp.199-207971 16. H.Oumousetal,OpticalandElectricalPropertiesofAnnealedCdSthinlmsObtainedfromaChemicalSolution,ThinSolidFilms,Vol.386,pp.87-902001 17. FredrickBuch,PhotovoltaicII-VICompoundHeterojunctionsForSolarEnergyConversion,PhDDissertation,StanfordUniversity,June76 18. A.NiemegeerandM.Burgelman,J.Appl.Phys.,Vol.81,No.6,pp.2881-2885997 19. JamesR.Sites,SolarEnergyMaterialsandSolarCells,Vol.75,No.1-2,pp.243-25103 20. DonaldThomasMorgan,AnInvestigationintoDegradationofCdTeSolarCells,M.S.Thesis,ColoradoSchoolofMines,pp.20-29998 21. DonaldA.Neamen,SemiconductorDevicePhysics:BasicPrinciples,ThirdEdi-tion,TataMcGrawHill,pp.238-31802 22. C.SFerekideset.al,ThinSolidlms,Vol.361-362,pp.520-5262000 23. T.Okamotoet.al,ThinSolidFilms,387,pp.6-10001 46

PAGE 56

APPENDICES 47

PAGE 57

AppendixAMeasurementAutomationAutomationoftheStabilitymeasurementsystemfulllsthefollowingneeds Consistencyandaccuracyofmeasureddata, Reductioninmeasurementtimeandhasslesinvolved, EasyCongurationofStressConditions, MassiveDataCollectionand Dataloggingandmonitoringofstresstemperature,humidityandstressperiod.ThehardwarepartconsistsofsolidstaterelayswhichiscontrolledbyPCIdigitalIOcardviaoptocouplers.TherelayisinNCpositionwhennotenergized.TherequiredstressconditionscanbeconguredinthispositionviaterminalsprovidedfromNCandGNDwhichisconnectedtonsideofdiodetobetestedoutputs.Thepossi-blecongurablestressconditionsareopencircuit,shortcircuit,maximumLoadand,forwardandreversebias.Whentherelayisenergized,thedeviceundertestDUTisconnectedtomeasurementdevice.Caremustbetakenwhileprogrammingrelaystoavoidsimultaneousactivationofmorethanonerelay.Thedipswitchisprovidedtoisolatearelayfromexperimentincaseofcircuitfailure.Totally48relaysarecon-trolledbyasingle48DIOcardwhichconsistsof2-8255PPIchips.DUTisthuscontinuouslymaintainedinthedesiredstressconditionexceptforthedurationinfewsecondsduringmeasurement.Keithley2400sourcemeterisusedtosweepthevoltagefrom-2Vto+2Vinstepsof0.02VandmeasurethecurrentduringI-Vmeasurement.Omega44-Mdataloggerisusedtomonitortemperatureandhumidity.Itiscon-nectedfromthestressovenviaRS232interface.SoftwareisprovidedbyOmegatocollectthedata.Itcontainsinternaltemperatureandhumiditysensorsandtwoexternalthermistors.Samplingintervalcanbevariedfrom0.5sec.ThesoftwareisdevelopedusingLabView.Thefrontpanelofmainviprogramanditsblockdiagramofdataowduringdarkcycleareshown.Someofthesalientfeaturesofthesoftwareare Dayandnightcyclearemonitoredandlogged, Logicisentirelybasedonrealtimetestcondition, Themeasurementsareavoidedincaseoflampfailure, 48

PAGE 58

AppendixAContinued FigureA.1FrontPanelofVIProgramforStabilityTesting FigureA.2SnapShotofBlockDiagram

PAGE 59

AppendixAContinued FigureA.3CircuitCongurationofHardwareSetup Thetotallightsoakedhoursarecalculatedwithmillisecondaccuracy, Daytransitionwillnotcauseproblems, Measurementtimingscanbecustomdenedforanynumberoftimesneeded Thecodeisoptmisedtoavoidmemoryleaksand Thedeviceparametersarestoredinwelldenedmannerforeasyretrievalfordataanalysis.Themeasurementtimingsmustbegiveninascendingorder.Alsotheintervalshouldbegreaterthanthesinglemeasurementsetperiodof8+5minutestoavoidmissingofdatapoints.

PAGE 60

AppendixAContinued FigureA.4InputPanelforMeasurementIntervalSpecication

PAGE 61

AppendixBJVCharacteristics TableB.1AverageEstimatesofandSeriesResistanceafterLS@OC. HToCafterCdCl2treatment LSPeriod Rs;light[)]TJ/F18 11.955 Tf -45.141 -14.446 Td[(cm2] Rs;dark[)]TJ/F18 11.955 Tf -44.916 -14.446 Td[(cm2] 360 Initial 3.0275 3.975 LSfor1000Hrs 3.38 3.535 380 Initial 2.525 6.23 LSfor1000Hrs 2.99 1.405 400 Initial 1.882 2.81 LSfor1000Hrs 3.36 3.25 TableB.2AverageEstimatesofandSeriesResistanceafterLS@SC. HToCafterCdCl2treatment LSPeriod Rs;light[)]TJ/F18 11.955 Tf -45.141 -14.445 Td[(cm2] Rs;dark[)]TJ/F18 11.955 Tf -44.916 -14.445 Td[(cm2] 360 Initial 2.73 3.85 LSfor1000Hrs 3.53 3.1 380 Initial 2.39 11.62 LSfor1000Hrs 4.10 4.95 400 Initial 2.11 2.63 LSfor1000Hrs 3.74 5.45 52

PAGE 62

AppendixBContinued FigureB.1DegradationBehavior:HT@380oCandLS@O.C

PAGE 63

AppendixBContinued FigureB.2DegradationBehavior:HT@390oCandLS@O.C

PAGE 64

AppendixBContinued FigureB.3DegradationBehavior:HT@380oCandLS@S.C

PAGE 65

AppendixBContinued FigureB.4DegradationBehavior:HT@390oCandLS@S.C

PAGE 66

AppendixBContinued FigureB.5CrossOverEectforSamplesLS@OCandHT@Top:360oC,Middle:380oCandBottom:400oC

PAGE 67

AppendixBContinued FigureB.6CrossOverEectforSamplesLS@SCandHT@Top:360oCMiddle:380oCandBottom:400oC

PAGE 68

AppendixBContinued FigureB.7JVCharacteristics.HTat380oCandLSatOC.TopandMiddle:DarkJVandBottom:LightJV.

PAGE 69

AppendixBContinued FigureB.8JVCharacteristics.HTat380oCandLSatSC

PAGE 70

AppendixBContinued FigureB.9JVCharacteristics.HTat390oCandLSatOC

PAGE 71

AppendixBContinued FigureB.10JVCharacteristics.HTat390oCandLSatSC


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 006 m d
007 cr gn|
008 030807s2003 flu sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000064
035
(OCoLC)52809421
9
AJJ4778
b SE
040
FHM
c FHM
049
FHME
090
TK145
1 100
Rangaswamy, Ashok.
0 245
Effect of CdCl treatment on CdTe and CdS solar cell characteristics after exposure to light for 1000 hours
h [electronic resource] /
by Ashok Rangaswamy.
260
[Tampa, Fla.] :
University of South Florida,
2003.
502
Thesis (M.S.E.E.)--University of South Florida, 2003.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 71 pages.
3 520
ABSTRACT: The CdTe solar cell is a leading candidate for cost-effective thin-film solar cells having demonstrated small area cell effciencies of 16.4%. A Key issue associated with CdTe thin film photovoltaic modules is the analysis of degradation behavior of the device. The analysis is complicated as changes due to degradation may be reversible. Solar cell measurement techniques were used to understand the changes in device parameters after light soaking for 1000 hours. An automated measurement setup was implemented as part of this thesis work. The main objective of this thesis was to study the effect of CdCl heat treatment on the device stability. The temperature for this heat treatment was varied from 360oC to 400oC. Cells were stressed under illumination at both short circuit and open circuit conditions. It was found that the increase CdCl heat treatment temperature slowed down the degradation rate.This was true for both short and open circuit stress conditions. Also short circuit stress condition slowed down the degradation of the device when compared with the open circuit condition. It became evident that the recombination current mainly got affected when the device was said to be degraded.
653
stability.
dark current.
cadmium telluride
light soaking.
CdCl2
773
t USF Electronic Theses and Dissertations.
949
FTS
SFERS
ETD
TK145 (ONLINE)
x sv 8/6/03
4 856
u http://digital.lib.usf.edu/?e14.64