Tibor Grasser E ditor Hot Carrier Degradation in Semiconductor Devices Hot Carrier Degradation in Semiconductor Devices Tibor Grasser Editor Hot Carrier Degradation in Semiconductor Devices 123 Editor TiborGrasser InstituteforMicroelectronics ViennaUniversityofTechnology Wien,Austria ISBN978-3-319-08993-5 ISBN978-3-319-08994-2(eBook) DOI10.1007/978-3-319-08994-2 SpringerChamHeidelbergNewYorkDordrechtLondon LibraryofCongressControlNumber:2014952459 ©SpringerInternationalPublishingSwitzerland 2015 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartof thematerialisconcerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation, broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,andtransmissionorinformation storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology nowknownorhereafterdeveloped.Exemptedfromthislegalreservationarebriefexcerptsinconnection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’slocation,initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer. PermissionsforusemaybeobtainedthroughRightsLinkattheCopyrightClearanceCenter.Violations areliabletoprosecutionundertherespectiveCopyrightLaw. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant protectivelawsandregulationsandthereforefreeforgeneraluse. While the advice and information in this book are believed to be true and accurate at the date of publication,neithertheauthorsnortheeditorsnorthepublishercanacceptanylegalresponsibilityfor anyerrorsoromissionsthatmaybemade.Thepublishermakesnowarranty,expressorimplied,with respecttothematerialcontainedherein. Printedonacid-freepaper SpringerispartofSpringerScience+BusinessMedia(www.springer.com) Preface Together with bias temperature instabilities and time-dependent dielectric breakdown, hot carrier degradation has been at the forefront of critical reliability issuesforhalfacentury.Inearliertechnologies,deviceswereoperatedatrelatively high voltages in which highly energetic (“hot”) carriers are created in a rather straight forward manner. Using some lucky-electron arguments, where a solitary “lucky”hotcarrierisabletocausedevicedegradation,simpleyetaccuratereliability models could be constructed. In modern scaled technologies, however, the true origin of hot carrier degradation is much more subtle, requiring more detailed knowledge of the multi layered physics of defect creation. A lot of research has been carried out in this field during the last 15 years, triggered significantly by the pioneering work of the group of Karl Hess. During recent years, the rapid introductionofnewmaterialsandothertechnologicaloptionshasraisedanumber ofnewissuesandchallengeswhichhavetobeaddressedurgently. Whilealotofprogresshasbeenmadeintheunderstandingofdevicedegradation brought about by hot carriers, the topic is far from being fully understood, in particular when challenges in future technologies must be resolved. As such, I felt that a thorough and comprehensive collection of the state of the art would be a valuable resource for scientists and engineers working on this phenomenon. I have therefore invited leading authors in the field to summarize their current understanding and review the state of the art in greater detail than is possible in regularjournalandconferencepublications. The book is structured in three parts and encompasses characterization, defect/device modeling, technological impact, and circuit/compact modeling aspects. In the opening chapter, McMahon et al. (GlobalFoundries) provide an overview of modeling attempts going beyond the simple lucky-electron picture. They summarize the theoretical foundations and contrast these models to those often used in industry to eventually arrive at a qualification scheme compatible with industrial needs. In the next chapter, Rauch and Guarin (State University of NewYork/IBM)describethegroundbreakingenergy-drivenhotcarrierparadigm, v vi Preface which acknowledges the fact that the energy distribution of the carriers plays a crucial role in degradation. They provide simple and effective approximations to the carrier energy distribution function and demonstrate how they can be used to accuratelymodelhotcarrierdegradation.InthechapterbyBravaixetal.(ISEN/ST Microelectronics), the authors build on the energy-driven paradigm by Rauch and LaRosaandtheworkoftheHessgroupondefectbreakagedynamicstoconstruct a refined hot carrier degradation model. They compare their model with simpler models and validate it for numerous technologies and use cases. Based on these fundamentalcontributions,Tyaginov(TUWien)summarizeshiseffortsincreating a comprehensive TCAD model for hot carrier degradation which utilizes a solver for the Boltzmann transport equation for the accurate determination of the carrier distribution function. A detailed study of the impact of the various contributors to thecarrierenergydistributionfunctionandthepeculiaritiesofthedefectgeneration kineticsaswellastheirimpactondegradationisprovided. As outlined above, detailed knowledge of the carrier distribution function is essential for accurate hot carrier degradation modeling. Unfortunately, this distribution function is the solution of the seven-dimensional Boltzmann transport equation and as such very difficult to obtain. Although this has been a standard TCAD problem for many decades by now, an efficient and user-friendly solution scheme for this highly complex problem remains a challenge. Zaka et al. (GlobalFoundries/University of Udine/IMEP/Institut d’Optique Graduate School/ST Microelectronics) suggest a highly efficient semi-analytic solution scheme for the Boltzmann equation, which is capable of considering full-band aspects as well as various challenging scattering mechanisms including impact ionization and carrier-carrier scattering. In the next chapter, Bina and Rupp (TU Wien)describetheireffortsincreatinganefficientdirectsolverfortheBoltzmann equation based on a spherical harmonics expansion of the distribution function, undertheinclusionofthePauliPrinciple,impactionization,andelectron-electron scattering. Contrary to the conventionally used Monte Carlo approaches, this approach allows a deterministic solution of the Boltzmann equation, which is extremelybeneficialfortheeliminationofthenoiseintheall-importanttailsofthe distributionfunction. While recovery of hot carrier degradation is typically neglected for reliability assessment,itcanbeshownthatthedegradationisnotfullypermanentandcanbe recoveredbyincreasingthetemperature.RecentresultsaresummarizedbyPobegen (K-AI)whodemonstratesthatthedistributionofreactionbarriersisconsistentwith results of electron-spin resonance measurements on P centers, which are silicon b danglingbondsattheinterface.Itisfurthermoreshownthatquitedifferentresults areobtainedafterbiastemperaturestress,indicatingthatthelinkbetweenthesetwo degradation modes is not yet fully understood. In the final chapter of the first part of the book, Aichinger and Nelhiebel (Infineon) provide a detailed tutorial on the charge-pumping technique, which is the most commonly used method to analyze interfacestatesinMOSdevicesandassuchofimmensevaluetoourunderstanding of the time-dependent evolution of the defect profiles. The various suggested modifications of the method are summarized, using hot carrier degradation as an example. Preface vii Inthesecondpartofthebook,theopeningchapterbyFrancoandKaczer(imec) studies hot carrier degradation in high-mobility SiGe and Ge channel MOSFETs, in which a more severe degradation is expected due to the smaller bandgap comparedtoSi.Theysuggestandstudygatestackoptimizationmethodswhichare demonstrated to reduce not only hot carrier degradation but also bias temperature instabilities.Next,Choetal.(imec)investigatehotcarrierdegradationinFinFETs, which are the likely end-of-the-road map CMOS architecture. Given the small channel volume and the poor thermal coupling to the substrate, FinFETs (as well asSOItechnologies)arepronetoincreasedself-heatingeffects,whichareshownto unfavorablyinteractwithdegradationmechanisms. Lateral double-diffused MOS (LDMOS) transistors have been an important component in the microelectronics industry for decades. Reggiani et al. (Univer- sity of Bologna/Texas Instruments) present their TCAD approach aimed at the understanding of hot carrier degradation in these complicated structures in terms of the safe operating area. Using a drift-diffusion approach, the impact of device geometryandinparticularofthecornersaroundshallowtrenchisolationsisstudied andtheaccuracyofthemethodologydemonstratedviacomparisontoexperiment. The chapter of Alagi investigates the applicability of a dispersive rate-limited modelingapproachtothecaseofLDMOSFETs.Particularcareistakentocapture the degradation for varying bias conditions, which is essential for understanding the behavior of a device in a circuit. Given the large dimensions, the rates can be successfully described by an extended lucky-electron description, and a compact modelsuitablefortheimplementationintostandardcircuitsimulatorsissuggested. In the final chapter of this part, Chakraborty and Cressler (Georgia Institute of Technology)lookintohotcarrierdegradationinsilicon-germaniumheterojunction bipolar transistors (HBTs), the understanding of which has significantly evolved during the last few years. The authors review experimental evidence, summarize thephysicsofdegradationforthesedevicesbasedonverticalcurrenttransport,and eventuallydevelopandvalidateanaccurateTCADmodelingapproach. Thethirdpartofthebookisdevotedtocircuit-relatedaspectsofhotcarrierdegra- dation.Inthefirstchapter,Huardetal.(STMicroelectronics)developabottom-up modelingapproachforcircuitreliabilitypredictionforgeneralstresspatterns.The modelisvalidatedindetailwithaparticularfocusontheinteractionwiththebias temperature instability, and the authors demonstrate how this methodology can be used to determine accurate design margins. The chapter by Schluender (Infineon) focusesontheidentificationoftherelativeimpactofhotcarrierdegradationandthe biastemperatureinstability.Itissuggestedthatdependingontheapplicationfield,a circuitcanbemorepronetooneofthesedegradationmodes.However,anumberof exceptionsarehighlightedwhichdemonstratethatnoconclusionsonthedominance ofonemechanismcanbeprovidedforthegeneralcase.Inthelastchapter,Scholten et al. (NXP) discuss compact modeling approaches to hot carrier degradation and how to guarantee that the conventional DC degradation models remain accurate under transient conditions. The methodology is successfully validated for three ratherdifferentdevices,namely,HBTs,MOSFETs,andLDMOSdevices. viii Preface Isincerelyhopethattheinformationprovidedinthesechaptersprovesusefulto scientists and engineers working in this challenging field by accurately capturing thestateoftheart.Furthermore,itishopedthatthisbooktriggersfurtherresearch intothiselusivephenomenon. Wien,Austria TiborGrasser May2014 Contents PartI BeyondLuckyElectrons FromAtomstoCircuits:TheoreticalandEmpiricalModeling ofHotCarrierDegradation..................................................... 3 WilliamMcMahon, YoannMamy-Randriamihaja, BalajiVaidyanathan,TanyaNigam,andNinadPimparkar TheEnergyDrivenHotCarrierModel........................................ 29 StewartE.RauchandFernandoGuarin Hot-CarrierDegradationinDecananometerCMOSNodes: FromanEnergy-DriventoaUnifiedCurrentDegradation ModelingbyaMultiple-CarrierDegradationProcess....................... 57 AlainBravaix,VincentHuard,FlorianCacho,XavierFederspiel, andDavidRoy Physics-BasedModelingofHot-CarrierDegradation ....................... 105 StanislavTyaginov Semi-analyticModelingforHotCarriersinElectronDevices.............. 151 AlbanZaka,PierpaoloPalestri,QuentinRafhay,RaphaelClerc, DenisRideau,andLucaSelmi TheSphericalHarmonicsExpansionMethodforAssessing HotCarrierDegradation........................................................ 197 MarkusBinaandKarlRupp RecoveryfromHotCarrierInducedDegradationThrough TemperatureTreatment......................................................... 221 GregorPobegen Characterization of MOSFET Interface States Using the ChargePumpingTechnique .................................................... 231 ThomasAichingerandMichaelNelhiebel ix