![微電子工藝-課件_第1頁](http://file4.renrendoc.com/view2/M02/02/20/wKhkFmYMRLWATyESAABDaND1jK0026.jpg)
![微電子工藝-課件_第2頁](http://file4.renrendoc.com/view2/M02/02/20/wKhkFmYMRLWATyESAABDaND1jK00262.jpg)
![微電子工藝-課件_第3頁](http://file4.renrendoc.com/view2/M02/02/20/wKhkFmYMRLWATyESAABDaND1jK00263.jpg)
![微電子工藝-課件_第4頁](http://file4.renrendoc.com/view2/M02/02/20/wKhkFmYMRLWATyESAABDaND1jK00264.jpg)
![微電子工藝-課件_第5頁](http://file4.renrendoc.com/view2/M02/02/20/wKhkFmYMRLWATyESAABDaND1jK00265.jpg)
版權說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權,請進行舉報或認領
文檔簡介
微電子工藝1微電子工藝1PhotoLithographyTextBookSemiconductortechnologyisverycomplicateandneedcarefulhandlingwhenAllEngineersdoinghisownjobinCleanroom.I’llprepareallkindsoftechnicalmaterialforthispurpose.Itcoversfrombasicitemstoadvanceditems.IhopethismaterialswouldbeusefulforallLithographyEngineers(Process&Equipment)I’mveryappreciatetohaveachancetosummarizethisdata.Icansaytoeveryonethatit’snotperfectmaterialbutbestone.Thankyouforyourhelp.2002.4.1AlbertChang
PrefacePhotoLithographyTextBookPre精品資料精品資料你怎么稱呼老師?如果老師最后沒有總結(jié)一節(jié)課的重點的難點,你是否會認為老師的教學方法需要改進?你所經(jīng)歷的課堂,是講座式還是討論式?教師的教鞭“不怕太陽曬,也不怕那風雨狂,只怕先生罵我笨,沒有學問無顏見爹娘……”“太陽當空照,花兒對我笑,小鳥說早早早……”微電子工藝-ppt課件Lithographykeywordsdefinition1[1]Designrule:(D/R)Definition:It’susuallydefinedthat“Halfpitchsizeofgatemask(Gatelineisthemostcriticallayer)Thisisthereferenceofsemiconductortechnology:25Logic,22DRAM,18DRAMetcKeyfactorstodecideDesignrule:Lightsource,Lensperformance,Resist,Enhancedprocesstechnologyetc[2]NA(NumericalAperture):Definition:Light-gatheringPower==>beproportionatetodiameterofLens(Maxvalueis1)ThisisthebarometerofResolution(Designrule):DoF=k2*(λ/NA2),R=k1*(λ/NA),KeyfactorstoincreaseNA:Lenspolishingtechnology(toControlAberration)[3]PR(PhotoResist):Definition:Photoactivatedmaterials:Usemajor3componentsandhave2typesDifferenttechnologybydifferentlightsource:g-,I-,KrF,ArF,X-ray,e-beametcThisisthe2ndimportanttechnologytoincreaseResolution[4]ARC(Anti-reflectivecoatingmaterial):Definition:Akindoffilmthatreducelightreflectiveproblem(BARC),Standingwaveeffect(TARC)BARC,TARCforeachwavelength(I-line,KrF)==>BARCisusefulforL/S,TARCisforHoleprocessTARC:Watersolublematerial,BARC:Organicmaterial(Useoxideetchantavailable)DemeritforTARC:Defectsource(Bubbleissues),forBARC:AddetchprocessandparticlesourceMerit:GoodCDcontrolandreducepatternnotchproblemThesmallerwavelength,thebiggerreflectiveproblemDyedresistisonekindofARCprocessLithographykeywordsdefinitiLithographykeywordsdefinition1[5]AberrationDefinition:thestateofAbnormalopticalcondition,path;It’srelatedtolenspolishing&assemblyaccuracyTherearearound6kindsofAberration:(1)SphericalAb.,(2)Coma,(3)FieldCurvature,(4)Distortion,(5)Astigmatism,(6)ChromaticAb.Thisisnoteasytoimproveafterlensassemblyfinished:BecausethiscomesfromlensitselfThisisthereasonwhyLensevaluationspecificationissoimportant[6]OAI(Off-AxisIlluminator):Definition:Akindofspecialilluminationmodetoimproveprocessperformance(DOF).Ingeneral,thereare3kinds:smallσ,Quadruple,AnnularThistechnologyisveryusefulandeasytouse,butbadthroughput(~30%Intensitydecrease)[7]PSM(Phaseshiftmask):Definition:Phaseshiftershift180olightphase,it’scontrolledbythicknessofshifterandrefractiveindexThismaskenhancepatternresolution:Thereare2kinds:AttenuatePSM(HalfTone),AlternativePSM(Levenson)HalfTonePSM:GoodforHolepattern,butithassideLobeissue(Controlledbymaskbiasandtransmittanceratio)LevensonPSM:Stillithaslotsofproblem.It’susefulforrepeatingpattern;Stillit’snotsopopular.Conventionalmask:WejustcallitBinarymask[8]ContrastCurve:Definition:Thetoolofresistsensitivity(D0orE0)anddevelopersolubilitycheck(Contrast:γvalue)E0isdecidedbyResistsensitivity(Dose-to-clear):Eop(Dose-to-print)=1.6~2.2xEoγvalue=Indexofdevelopingpower;Thehigher,thebetter(Clearedgecut)LithographykeywordsdefinitiMaskTechnologyOverview[1]Maskprocessoverview(Binary&AttenuatePSMmask)BlanketMask←Cr+CrOX←Quartz←Resist←Cr+CrOX←Quartz←Resist←Shifter←Cr+CrOX←Quartz←Resist||||||||||||||||←Cr+CrOX←Quartz←Resist←ShifterExpose&Deve-beamwritere-beamresist&Dev←Cr+CrOX←Quartz←Resist←Cr+CrOX←Quartz←Resist←ShifterCrEtch←Cr+CrOX←Quartz←Resist←ShifterShifterEtchPR&Crstrip←Cr+CrOX←Quartz←Quartz←Shifter(1)BinaryMask(BIM)(2)AttenuatePSM(HT-Allarea)2MaskTechnologyOverview[1]MaMaskTechnologyOverview[1]MaskprocessoverviewBlanketMask←Cr+CrOX←Quartz←Resist←Shifter←Cr+CrOX←Quartz←Resist←ShifterExpose&Dev←Cr+CrOX←Quartz←Resist←ShifterCrEtch←Cr+CrOX←Quartz←Resist←ShifterShifterEtchPRstrip←Quartz←Shifter||||||||(3)AttenuatePSM(HT-Partialarea)
←Quartz←Shifter←Cr+CrOX←Cr+CrOX←Resist←Quartz←Shifter←Cr+CrOX←Resist|||||←Quartz←Shifter←Cr+CrOX←Resist←Quartz←Shifter←Cr+CrOX←Quartz←Shifter←Cr+CrOX2ndcoating2ndE-beamDevCrWetEtchPRstrip2MaskTechnologyOverview[1]MaMaskTechnologyOverview[2]PrincipleofAttenuatePSManditscharacteristic(1)PrincipleEI+-+-SideLobeAreaThisisthereasonwhyedgeprosogoodandenhanceDoF(2)CharacteristicsMasktechnologyisnotsodifficultcomparewithlevensonWecanchoosetransmittanceofshifter(8%,4%etc)Especiallythisisgoodforcontacthole,ButSideLobeisbiggestproblem(Almostclearedthisproblemnow)ToprotectsideLobeproblem,weusepartialattenuatePSMmask(ScribelineforBinarymask)Forthefuture,thisismoreusefultoolstoenhanceprocesswindowRecentlywecanuseforL/Sprocesstoo.HowtoreducesideLobe:Optimizetransmittance&resistdissolutionrateadjustment2MaskTechnologyOverview[2]PrMaskTechnologyOverview[3]ComparewithAlternativePSMandBinarymaskonwaferLENSBinaryMaskE-fieldI-fieldCrQzLight00+onwaferonwafer0AlternativePSMCrQzLightShifterLENS0+-onwafer
AlternativePSMmaskissues
Maskdesignandfabricationisnoteasy(especiallyshifter)Wecangetfantasticresolution,butthereareghostpatternatpatternedge(Lowenergy).Tosolvethisproblem,highperformanceNegativeresistweneed.ThisisnotgoodperformanceforisolatedpatternStillthistechnologyisnotpopular.2MaskTechnologyOverview[3]CoPhotoin-lineprocessflowoverviewDefinition:PhotolithographyIt’spatternprintingprocessontheresistcoatedwaferbyUVexposureandpatterndesignedmask(Reticle).WaferFilmFilmDepositionWaferFilmPhotoresistResistCoatWaferFilmPhotoresistUVExposurehvReticleWaferDevelopingFilmWaferEtchGasEE/FOOLBaseLine/FBMix&Match(System)PatternSizeControlPatternProAlignControlbetweenlayerPhotoJOBFACTORProcessconditiondependsonwhichlayer:Substratecondition,CD,Topology,OLtarget,EtchtargetetcKeyprocessfactor:Tpr(A),UseBARCornot,SB,TARCornot,WEE,EE,FO,OLoffset,PEB,Dev.Dippingtimeandsoon.Keymachinefactor:Cupexhaustpressure,Waterjacketcontroltemp,Chamberpressure,Temp,focus,Plateexhaust,PEBtempAccuracy.[1]GeneralFlow3Photoin-lineprocessflowovePhotoin-lineprocessflowoverviewHMDScoatCooling1PRCoat(**)SoftBakeCooling2WEE($)EGAalignExposureHardBakeDevelopingP.E.B($$)Scope($$$)OLmeasureCDmeasureIonImpEtchReworkDepositionEXPOSURETRACKPRstripIN-LINECoolingBARCTARC(***)BAKE(*)stepper(*)AbsolutelyweremovemoistureonthewaferbeforeBARCcoating:Secondaryreaction(**)EBRisnotabsoluteprocess:WecanuseWEEinsteadofEBR==>Needrecipetuning(***)TARCshouldbecoatedafterPRbake(Softbake):DonotTARCbake(Secondaryreaction)($)WecanchangefrombeforeexposuretoafterexposureforI-line,butDUVcannot.($$)PEDcontrolisveryimportanttomaintainprocessperformanceforDUV,butI-lineisn’t.($$$)ADIprocedureisimportanttoreduceADIloss:Scope==>OL==>CDBAKE3Photoin-lineprocessflowovePhotoin-lineprocessflowoverviewHMDSCoatToenhanceadhesionbetweenwaferandresist:MakehydrophobicpropertyUseN2bubblinginthevacuumchamber:Particleissuehappenwhenlowvac.Usually130oCplatetemp.isoptimumconditionCooling1Hightemp.ofwaferisnotgoodforcoatinguniformityUsuallyroomtemp(23oC)isoptimumcondition:Dependonresistanditstks.PRCoatSoftBakeCooling2Usesteppingmotorchemicalpump:ControldispenseamountanditsspeedProcessEBR/backrinsetoremovedummyresist:WEEcandoinsteadofEBRUniformityfactor:PRtemp,wafercool,cupexhaust,T&H,Chucktemp,Resistsolventremove,relaxtheresiststressSettingtemp:80oC~130oC==>SolventB.Pisaround150oC,thereforeit’simpossibletoremovesolventperfectly.Toremovemisalignproblem:Wafermag.,scale,rotetcRoomtemp.isOK(23oC)Havetomaintainsametemp.foreachwafer.[2]CoaterFlow3Photoin-lineprocessflowovePhotoin-lineprocessflowoverviewP.E.BStandfor“PostExposureBake”==>I-line:ReduceStandingWaveeffect==>DUV:DiffusePhotoAcid(Coreprocess)RootcauseofStandingWave(I-line)撩碟醞==>DifferentconcentrationofdissolvedPACbecauseoflightwavePEBtemperatureofDUVdecideCDanditsprodependonresistNeedtoevaluateoptimumTemp:BelowPACdissolutionTemp(130oC)I-line(110oC~120oC),DUV(110oC~130oC)areoptimumTemp.CoolingToimproveDevelopuniformityOptimumTemp:23oC(roomtemp)/60secDevelopHardBakeImprovePRhardening&Adhesion,Removewater&Solvent,Butthisisnotnecessary.(I/Ilayerisnecessary:Outgassingissue)Usuallyunder130oCNormallyweuse2.38%moleconcentration:I-line/DUVForspecialDUVprocess,weuse1.914%TMAHDeveloper.ThisisonlyforhighspeedDUVresist(Localsupply)Bake[3]DeveloperFlow3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)ADHESIONToenhanceadhesionbetweenwaferandresist,wetreatvaporizedHMDS(HexaMethyleneDiSilazane)onthewaferandmakeithydrophobicstateHydrophilicSurfaceSubstrateSiSiSiSiOOOO(H3C)3Si(H3C)3Si(H3C)3Si(H3C)3Si⊿T-nNH3HydrophobicSurfaceLowContactAngleHighContactAngleSubstrateSiSiSiSiOHOHOHOH(H3C)3SiSi(CH3)#NH(H3C)3SiSi(CH3)#NHBubblingN2N2HMDSINHMDSOUT~~N2+HMDSBubblingSystemHMDSitselfisLiquidstate.WecanmakevaporizedHMDSbyN2bubblingsystem,andtreatthisgasinthevacuumCh.[1]Adhesion(HMDS)*Wecancheckwafersurfaceconditionbycontactangle*Needtocleanthevaporizedgassupplypipetoremoveparticlesource3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)*StaticmodeisusefulforHighviscosityResist,andDynamicforlowviscosity[1]Adhesion(HMDS)HowtocoatSpinoncoatingmethod:2kindofmethods(1)Dynamic:Resistdispenseonrotatingwafer,(2)Static:Dispenseonstoppedwafer企2企3企1t2t1t3t4SPINSPEEDTIMERESISTDISPENSINGEDGEBEADREMOVINGRESISTCoatingCYCLE(Conventionaltype)r1r2r3r4r5ThiscycleisconventionaltypecycleCurrentlyallresistpumprecipeisnotthistype.1.DispenseRPMislow2.ThicknessisdecidedbyHi-RPM3.Resistviscosityisimportanttodecidethicknesstarget==>Hi-RPMcancoverunder3000Athicknesst1:RESISTDISPENSETIMEt2:HIGHRPM(Decidethickness)t3:SIDE/BACKRINSETIMEt4:SOLVENTDRYTIMEt=√rpmkS2t:RESISTthicknessS:%ofresistsolidk:SPINCOATERConstant[2]PRcoatingcycle(Conv.)3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)*Weneedspecialresistpumpsystem(steppingmotorpumpandprewetdispensesystem)*Needveryhighspintoreduceresistamount.(~3500RPM)[1]Adhesion(HMDS)HowtocoatSpinoncoatingmethod:2kindofmethods(1)Dynamic:Resistdispenseonrotatingwafer,(2)Static:Dispenseonstoppedwaferω2StaticSPINSPEEDTIMERESISTDISPENSINGBRRESISTCoatingCYCLE(Solventprewettype)ThiscycleissolventprewettypecycleSolventprewetenhancesPRspreadout.1.Dispenseamountislow2.ThicknessisdecidedbyHi-RPM,buttherearelimitation(NeedtochangePRviscosityifwewanttochangeover1000Atarget)3.Resistviscosityismostimportanttodecidethicknesstarget4.Needtoremoveprewetsolventalmost.(justenhancePRspreadout)[2]PRcoatingcycle(Prewet)ω1ω1SolventPreWetEBRBRBR3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)[3]PRThicknessuniformityparameterParameterAveragethickness:Viscosity&spinspeedUniformity:Resist&Environmenttemp,wafertemp,motorflangeetcParameterWhenCoolingtempishigherthanopt.WhenPRtempishigherthanopt.Whencupexhaustishigherthanopt.Whencuphumidityishigherthanopt.Whenmotorflangetempishigherthanoptimum.CenterEdgeWhencuptempishigherthanopt.NoinfluenceOnlyChuckedgebecamethickthicknessInawaferWithinwafermapResisttempvsWFavgtks0204060802223242526Resisttemp(℃)
PRtks(A)020406080100120212223242526CoolingtempvsWFavgtks
PRtks(A)Cooltemp(℃)3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)[4]PRThicknesstargetfactorCurrentPRTargettrendThintargettrend:Toenhanceprocesswindow.PRselectivityimproved(StillmetallayerusethicktargetChoosevalleyorhillofS/WCurve(bare):StableCDuniformity==>Recentlyfilmwaferevaluatetrend(1stlayer)StandingWaveCDBalance(Bulkevaluation)EtchSelectivityEnhanceResol.LowPRselectivityfilm:Thicktarget(e.gMetallayer)I/Icondition榆Tocovertopologyinfluences,weneedbulkevaluationDRAMgatelayerisnecessary(Yield&device’sspeed)HighenergyI/IlayerneedthickfilmtocoverbreakthroughThroughputThinPRfilmisnecessarytogethighresolutionTodecreaseExposeenergy,thinPRfilmisbetter3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)[5]SolventdistributioninPRfilmSolventconc.vsBAKETempSolventconcentrationtrendinPRfilmBAKETempvsRESISTThicknessResistconsistsofResin,PACandover80%ofsolvent.SolventconcentrationiskeyfactorofPRthicknessSolventCon.SOFTBAKETempSOLVENTcon.BEFORESOFTBAKEAFTERSOFTBAKESUBSTRATERESISTFILMLOSSRESISTTHICKNESSSOFTBAKEtemp125~130﹎ReciprocalproportionbetweenbaketempandsolventconcentrationBeforesoftbake:Lotsofsolventininternalresistfilm,andsteeplydecreaseneartheresistsurfaceAftersoftbake:Evaporatedsolventininternalresistfilm,andit’llbepeakneartheresistsurfaceSoftbake==>SolventEvaporation==>Thicknessdown,DensityupWhenwebakeoverTgtemp,PACwillbedestroyedandplasticflowwillhappeninresistfilm.Sothicknesswillbesteeplydecreased.GLASSTRANSITIONTEMPERATURE3Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)[6]Developmentparameter(1)DeveloptimevsRprDeveloptimevsprofile鼻衛(wèi)除vs鼻樓紫Rpr(exparea)time(sec)PROFILE5060409070Optimumtimetime(sec)Tprtime(sec)E3E2E1E1<E2<E3WecanuseDRMsystemforthisone.ItdependonEE.Usuallyit’sunder20secdippingtime.Thisisthestandardtooltodecidedevelopdippingtime.UnderDevelop:ScumandtailOverDevelop:TopprofilecollapseUsually40~60secdippingisstandarddipping(dependonPR)EEvs.RprrelationshipCDwillbedecreasedbydippingtimebutsaturationtimeisaround30secdip.0t13Photoin-lineprocessflowovePhotoin-lineprocessflowoverview(Detail)[7]Developmentparameter(2)DevelopNormalityvsCONTRAST滲NORMALIZDTprDeveloptimeLowNormalityDev:NeedlongDevtime,gethighcontrast.ThisisgoodforperformanceHighNormalityDev:ShortDeveloptempvsDevelopspeed(rate)DEVELOPRATETEMPERTUREDeveloptemperatureisoneofimportantitemtocontroldeveloprate.Ithasoptimumtemp.ButTomaintainconstanttemperatureisnoteasy.Itdependonspraytypeandnozzletype.UsuallyEngineerneedtofindoutoptimumtemp.Butusuallywecanuseroomtemperaturebecausethistemp.isthemoststabletemperature.ABCNormality:A<B<CCONTRAST:C<B<AE/E:C<B<A3Photoin-lineprocessflowoveExposureProcessFlow[3]Exposeshot&Area(Shotdesign)ShotStepper:31.1mmmaximumfieldsize,Xmax=22mm,Ymax=NAOnwafer
Scanner:Xmax=26mm,Ymax=33mmOnwaferscribelineCrReticleDesignAlignKeyTegMonitoringPadReticleAlingkeyCrGuardRingchipchipchipchip1234DoubleExpoAreaExpoDoubleExposeArea:BothofonecoveredbyCr.GuardRing:ProtectblindsettingerrorXmaxisbecauseofReticlealignkeySHOT*WidthofScribelineisdecidedbydesigner,buttherearelimitwidth:Usually90~120umscribeline*Maskalignkey:Machinedependency==>ASML,Nikon,CanonSVGLetctotaldifferent4ExposureProcessFlow[3]ExposExposureProcessFlow[5]ExposingtechnologyOverviewConventionalAnnularQuadrupleMaskWaferFLEXNA/
OptimizationHighResolutionMaterialResistProcessTechnologyOff-AxisIlluminationPhaseShiftMask/OPCMultipleExposureOthersMaskPhase018001800180OPC*d=
/2(n-1)4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyOffAxisilluminator(1)Illuminatorsendtiltlighttolenssitebyspecialaperture(Centerclosed)Aperturecancut-1stlightamong-1,0,+1stdiffractedlightOnlyuse0&+1storderlightforimageexposingComparewithConventional,DiffractionAngleissmall:HiR&wideDoFIntensitydegradebecauseof-1storderdiffractedlight(~30%)TheoryOAIApertureMask-1storder+1storder0orderNAlensWaferLensSin
=n伙/Pn:0,
1,
2颻..P:PitchSize
:DiffractionangleConventionalApertureMask-1storder+1storder0orderNAlensWaferLens
4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyOffAxisilluminator(2)ConventionalAnnularQuadrupleEffectImproveResolutionImproveD.O.FItcan’tprintsmallsizepattern(>0.2um)becauseoflargediffractionAng.BadResolution/D.O.F盪OAI45
PatternDefinebad碳榆PatternDeformationBigeffectofLensAber.ThroughputdegradeLimitationLimitationAperturetypeConventional4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyOffAxisilluminator(3)QUADRUPOLEANNULARLOW
HIGH
LightpathLightpath4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyPSM(1)onwaferLENSConventionalMaskPhaseE-fieldIntCrQzLight00+onwaferonwaferPhaseE-field0PSMMaskCrQzLightShifterLENSInt0+-onwaferPSM(PhaseShiftMask)ThisisforResolution&DoFenhancement.Itusephaseinterference4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyPSM(2]TheoryDifferentrefractiveindexmaterialuseasshifter(n=1.5==>shift豇OpticalPathLength=nxdOpticalPathDifference=
dxn-
dx1=伙
/2MaskSubstratePhase"0"Phase"180"PhaseShifterDirectionofLightPropagationElectricfieldR.I(n)=1.5R.I=1.5R.I(n)=1.04ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyPSM(3]0.250.500.75RESOLUTION(um)EQUIPMENTG-lineI-lineI-lineDUVI-linePSMCrmaskDOF(um)EQUIPMENT2.01.51.00.5CrmaskPSMPSMeffectHighResolutionWideDoFG-lineI-lineI-lineDUVI-line4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyPSM(4)CrmaskLevensonPSMOutriggerPSM180oshifterQuartzQuartzQuartzCrosstargetGood&BadGeneral(Bin)撩ResolutionLimitEasymakingForLine&SpaceForContactPSMtype&itscharacteristicHalf-tonePSMForContactResolutiongreatHardmakingResolutiongoodMostHardmakingResolutionimproveEasymaking4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyPSM(5)Limit&itssolutionProcess/EquipNeedsPSMdesign啗&SimulationMasklayoutDummyline&patternTransmittance&PhaseMaskmachineproblem撿
LowperformanceofinspectionMCLowperformanceofdefectrepairMCExposuremachine(Stepper)Needgoodill.unif.AtLow
&N.ANeedgoodill.unif.AtHigh
&N.AHighperformanceresistHighqualitynegativeresistLayoutlimitationbecauseofproximityeffect-It’snotsuitableforLOGICdeviceProcesscontrolNeedtoughMaskCDspecNeedDefectfreemaskProcessdefectminimizeNeedHighresolutione-beamMCDummypattern<0.75um
(Levenson&OutriggerPSM)4ExposureProcessFlow[5]ExposExposureProcessFlow[5]ExposingtechnologyOPCAddorremovedummypatternatpatternedge,soastoprintalmostdesignedpatternonthewaferTargetPRPatternPRPatternConventionalMaskOPCMaskCrReticleCrReticleSmallMarginMarginEnlargeHoleHole*Thistechnologyisveryusefulbutit’sverydifficulttomakemask(Defectfree)4ExposureProcessFlow[5]ExposExposureProcessFlow[6]StepperBasicNANA(NumericalAperture)?Aperture(NAill)CollimatorLensReticleProjectionLensApertureStop(NAlens)WaferNA=d*sin
NAisthediameterofaperture(Itrestrictlightpath)Thereare2kindofNA:illuminationNA&LensNA.UsuallyNAmeansLensNA.CurrentallmachineadoptedvariableNA:UserchooseNA.NA=d/2fImagePlaneLens玄fdNA=n*sin
n:AirR.I(
1)4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasic(Coherence)<1>
(Coherence)?Aperture(NAill)CollimatorLensReticleProjectionLensApertureStop(NAlens)Wafer
=NAill/NAlensLightinterferencebetweenpath1andpath2,3lightistherootcauseofimageplanedistortionWhenwereducelensaperturediameter,wecancutpath2&3light(Reducelightinterference)WhenitcomestoCoherence,idealis0,usually0<<1(PartiallycoherentNA豇ImagePlanePath1Path2Path3LensNA幙4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasic(Coherence)<2>relationshipBigGOODHIGHGOODSmallContrastIntensityUniformitySpecialFrequency
=1
=0
=0.2
=0.6(conventional)
=0==>GetidealImageMTF(MultipleTransferFuntion)Smallisnotalwaysgood.WehavetochooseoptimumDifferentchangeimagefield(mismatchisoneofmisalignreason)NAill=ChangebyHWapertureNAlens=Freelychange
4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicResolution&DoF<1>Resolution&D.O.Fhavereciprocalproportionof/NA.It’sverydifficulttoimprovetheseperformance.Togettheseperformance,weuseHiperformanceResist,PSM,OAIandsoon.
R:Resolution
DOF:DepthOfFocus
:WavelengthofExposureSource
NA:NumericalAperture
k1,k2:ProcessFactordecidedbyPR,ARC,PSMOAI,OPC...RelationbetweenRandDoFR=k1*
/NADOF=k2*
/(NA)2NABigGOODSmallResolutionD.O.FGOOD
BigGOODSmallResolutionD.O.FGOOD4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicResolution&DoF<2>RelationbetweenRandDoF
(nm)365248193NA0.630.60.6K1K21.091.161.31R(um)0.350.250.18DOF(um)1.000.800.70OVERLAY(um)0.1000.0750.0450.630.60.6
vsD.O.FvsRESOLUTION00.20.40.60.811.2
(nm)(um)193248365D.O.FRESOLUTIONNAvsD.O.FvsRESOLUTION0.30.50.70.91.11.31.51.7NA(um)0.430.520.540.63D.O.FRESOLUTIONR=k1*
/NADOF=k2*
/(NA)24ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicLensAberration<1>AberrationInopticsystem,itmakesabnormalfocalplane.Actuallyallkindoflenshaveaberration.kindsMonoChromaticAberrationChromaticAberrationSphericalAberrationComaAberrationAstigmatismFieldCurvatureDistortionImagebecomeblurImagebecomedistort*LenspolishingaccuracyandassemblyaccuracyismajorreasonofAberration*Insemiconductorbusiness,mostimportantpointisHowtochooselowaberrationmachine.Becauseaberrationcannotimprove,it’snotrepairableproblem.*EspeciallyCOMAisbigprobleminOLcontrol(PatternalignandOLdataunmatch)4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicLensAberration<2>*ThisaberrationismoreimportantoneinLogicdevicesFab.Becausetherearemanydifferentpatternsizecircuitinonemask.*Evaluatespec:MinimumsizeCD(0.18~0.20)and0.5umpatternSphericalAberrationFocalpointisdifferentbetweenpath-Aandpath-Bontheopticaxis.Bending:BestForm(Lensformthatminimizesphericalaberration)dependoninputlightUseGradientIndexLens(differentrefractiveindex)ImagePlaneCDFocusWecangetdifferentbestfocusbetweenpatternsize(CD)Thesmallergap,thebetterBestFocusGapHowtocheckonwaferBendingHowtoremoveLowrefractiveGradientIndexLensHighrefractiveHowtoremove4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicLensAberration<3>*Easytofindoutthisaberration:WecanmeasurePatternslopeonright/leftside(ortop/bottom)*Thisonegivemoreseriousattackinfinepattern(Under0.2um):ShouldminimizeCOMAlevel.ComaObjectisnotontheOpticaxis.Itisnotfocalimagebutcometontheimageplane.Thisaberrationcanbeminimizedbylensassembly.BendingCombinationofLensUseAplanatopticsystem(Spherical&Comaisperfect)ImagePlaneHowtocheckonwaferIn-FieldHoleLineOff-axisComa:NodirectionalAberration,NocorrectableOn-axisComa:directionalAberration,CorrectableonebybendingglassDirectional爾薑
寞徹Object4ExposureProcessFlow[6]SteppExposureProcessFlow[6]StepperBasicLensAberration<4>*Easytomaketroubleatfieldedgearea*Especiallythisoneisfataltroubleincontactprocess:DifferentshapeindifferentpositionAstigmatismIncoherentfocalplanebetweenTangentialandSagittalImage.ImageplaneisinConicoidarea.conicoidareaSagittalsurfaceTangentialsurfaceTangentialPlaneSagittalPlaneHowtocheckonwaferCheckbestfocusatHorizental/Vertical/Sigittal
PatternHoleshapedistortedbythisaberrationeventhoughatBFBestFocusNormalAbnormalAstigmatism4ExposureP
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請聯(lián)系上傳者。文件的所有權益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會有圖紙預覽,若沒有圖紙預覽就沒有圖紙。
- 4. 未經(jīng)權益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲空間,僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護處理,對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對任何下載內(nèi)容負責。
- 6. 下載文件中如有侵權或不適當內(nèi)容,請與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準確性、安全性和完整性, 同時也不承擔用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。
最新文檔
- 26《好的故事》說課稿-2024-2025學年語文六年級上冊統(tǒng)編版
- 1場景歌說課稿-2024-2025學年統(tǒng)編版語文二年級上冊
- 2024年秋一年級道德與法治下冊 第二單元 我和大自然 5 風兒輕輕吹說課稿 新人教版
- 18古詩三首浪淘沙(其一)說課稿-2024-2025學年六年級上冊語文統(tǒng)編版
- 8 設計制作小車(二) 說課稿-2024-2025學年科學四年級上冊教科版
- 23《月光曲》說課稿-2024-2025學年語文六年級上冊統(tǒng)編版
- 1 24時計時法(說課稿)-2024-2025學年三年級上冊數(shù)學人教版001
- 2023九年級道德與法治上冊 第三單元 文明與家園 第五課 守望精神家園第2框 凝聚價值追求說課稿 新人教版
- 2025北京市飼料采購合同新
- 2025建造船舶所要用到的合同
- 農(nóng)產(chǎn)品貯運與加工考試題(附答案)
- 學校財務年終工作總結(jié)4
- 2025年人民教育出版社有限公司招聘筆試參考題庫含答案解析
- 康復醫(yī)學治療技術(士)復習題及答案
- 《血管性血友病》課件
- 2025年汽車加氣站作業(yè)人員安全全國考試題庫(含答案)
- 2024年司法考試完整真題及答案
- 高三日語一輪復習日語助詞「に」和「を」的全部用法課件
- 2024年山東省高考政治試卷真題(含答案逐題解析)
- 2024年執(zhí)業(yè)藥師繼續(xù)教育專業(yè)答案
- 2024-2025學年人教版七年級數(shù)學上冊期末達標測試卷(含答案)
評論
0/150
提交評論