Springer Handbook oƒ Semiconductor Devices Rudan Brunetti Reggiani Editors 123 Springer Handbooks SpringerHandbooksmaintainthehigheststandardsofreferencesinkeyareasofthephysical andappliedsciencesforpractitionersinindustryandacademia,aswellasgraduatestudents. Designedtobeusefulandreadabledeskreferencebooks,butalsopreparedinvariouselectronic formats, these titles allow fast yet comprehensive review and easy retrieval of essential reliable key information. Springer Handbooks cover methods, general principles, functional relationshipsandfundamentaldataandreviewestablishedapplications. All Springer Handbooks are edited and prepared with great care by editors committed to harmonizingthecontent.Allchaptersarewrittenbyinternationalexpertsintheirfield. Indexed by SCOPUS. The books of the series are submitted for indexing to Web of Science. Massimo Rudan • Rossella Brunetti (cid:129) Susanna Reggiani Editors Springer Handbook of Semiconductor Devices With1335Figuresand71Tables Editors MassimoRudan RossellaBrunetti DepartmentofElectrical,Electronic DepartmentofPhysics,Informatics andInformationEngineeringandAdvanced andMathematics-FIM ResearchCenteronElectronicSystems UniversityofModenaandReggioEmilia UniversityofBologna Modena,Italy Bologna,Italy SusannaReggiani DepartmentofElectrical,Electronic andInformationEngineeringandAdvanced ResearchCenteronElectronicSystems UniversityofBologna Bologna,Italy ISSN2522-8692 ISSN2522-8706(electronic) SpringerHandbooks ISBN978-3-030-79826-0 ISBN978-3-030-79827-7(eBook) https://doi.org/10.1007/978-3-030-79827-7 ©SpringerNatureSwitzerlandAG2023 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,broadcasting,reproduction onmicrofilmsorinanyotherphysicalway,andtransmissionorinformationstorageandretrieval,electronicadaptation, computersoftware,orbysimilarordissimilarmethodologynowknownorhereafterdeveloped. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublicationdoesnot imply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevantprotectivelawsand regulationsandthereforefreeforgeneraluse. Thepublisher,theauthorsandtheeditorsaresafetoassumethattheadviceandinformationinthisbookarebelieved tobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsortheeditorsgiveawarranty, expressedorimplied,withrespecttothematerialcontainedhereinorforanyerrorsoromissionsthatmayhavebeen made.Thepublisherremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalaffiliations. ThisSpringerimprintispublishedbytheregisteredcompanySpringerNatureSwitzerlandAG. Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland Foreword by Chihiro Hamaguchi The first semiconductor device, named “point-contact transistor,” was demonstrated on the 17th of December 1947 by John Bardeen, Walter Brattain, and William Shockley at Bell Laboratories.Aboutamonthlater(January1948),Shockleyinventedthe“junctiontransistor.” Thefirstdemonstrationofaradioreceivermadewithpoint-contacttransistorswasperformed inDecember1947.Thedesignofsolid-statedevicesforelectricalamplificationwasorganized atBellLaboratoriesundertheguidanceofShockley,whowasalsointerestedinanewdevice tocontrolsurfaceelectronswithametalgate.AFET-liketransistorhadalreadybeenpatented intheUSAbyJuliusEdgarLilienfeldinJanuary1930;theideawasnotrealizedbecauseof the high density of surface states in silicon. The difficulty was resolved by using an oxide surface layer processed in high-temperature vapor atmosphere, as proposed by C. Frosch in 1954;thedepositionofametalgateontheoxidelayerwasachievedbyJ.AttalaandD.Kahng atBellLaboratoriesin1959.First,PNPandNPNjunctiontransistorswerecommercializedand usedtodesignsolid-statecircuits;however,theSi-MOSFET,inventedafterwards,revealedits superiorityoverthejunctiontransistor,inviewofdesigningandmanufacturingdigitalcircuits. Somanypeoplewereinvolvedintheearlydevelopmentofintegratedcircuits(ICs);among them,JackKilbyofTexasInstruments(hybridintegratedcircuit,1958)andRobertNoyceat Fairchild Semiconductor (monolithic integrated circuit, 1959) played a very important role in the opening of IC technologies. The advancement of IC technology is well explained by Moore’s law: “the number of transistors in a chip of integrated circuits doubles every two years.”Moore’sexpectationexpressesquitewellthehistoryofICsfrom1970till2020.Thisad- vancementhasbeensupportedbymanynewtechnologies,materials,processes,circuit-design strategies, and others. Starting from the simple planar technology, structures are now more complexandthreedimensional.Thetechnologicalrevolutioninthefieldofthesemiconductor devicesradicallychangedmanyaspectsofthesocialorganization(communications,economy, transportation, and all the primary necessities of everyday life), so that in the present time, humanbeingsfinditdifficulttosurvivewithoutICs. v vi ForewordbyChihiroHamaguchi Semiconductor-deviceoperationsarebasedonfundamentalelectricalandopticalproperties of the materials, as anyone can easily agree about just by leafing through this Springer Handbook of Semiconductor Devices. The knowledge of energy band structures is essential for understanding any device operation. In other words, energy band calculations provide knowledge on electronic and optical properties of semiconductors. Many semiconductor propertiesarewellunderstoodonthebasisoftheeffectivemassesofelectronsintheconduction bandandholesinthevalencebands,inadditiontothe(directorindirect)bandgap.Pioneer datafromcyclotronresonanceexperimentsinGeandSiwereinterpretedbythetheorybased onk·pperturbationmethodbyG.Dresselhaus,C.F.Kip,andC.Kittelin1955,andrevealedthe many-valleystructureoftheconductionbandandthespin-orbitsplittingofthevalencebands intheseelementalsemiconductors.TheenergybandcalculationinthefullBrillouinzonefor 14 semiconductors of the diamond and zinc-blende structures was reported for the first time byM.L.CohenandT.K.Bergstresserin1966bymeansofthepseudopotentialmethod.More advanced theoretical approaches have been reported afterwards, and some of them are dealt withinthishandbook. This Springer Handbook of Semiconductor Devices provides a quite useful framework suitabletoguidetheinterestedreaderacrossamassiveamountofinformation,andachievesan almostcompletereviewofthebasicandadvancedknowledgeofsemiconductordevices.The editorsplannedtocoverthepresentstatusandfuturedevelopmentofsemiconductordevices, includingphysics,structures,processing,designing,andsimulation.PartIisconcernedwith thetechnologicalaspectsofsemiconductordevices,anddealswiththemanufacturingofsemi- conductormemorydevicestogetherwiththetechnologyofprocessing,measuringparameters, advancedlithography,anddielectricmaterials.Inaddition,FinFETsandadvancedtechnologies forfuturematerialsanddevicesarediscussed.PartIIdealswithbasicsemiconductordevices and applications, including Si-based devices such as MOS devices, MOSFETs, capacitors, and CCDs. Part II also deals with the topics of sensors, semiconductor lasers, and solar cells.PartIIIdealswithnew-generationdevicesandarchitectures,suchasresonant-tunneling devices,bioelectronics,spin-baseddevices,quantum-computingdevices,andotherinnovative structures.PartIVisdevotedtodevicemodelingandprocesssimulation,wherebothclassical device-simulationtoolsandmoreadvancedmethodsaredescribedindetail. Eventhoughmanykindsofsemiconductordevicesarestilldesignedandproduced,andnew andadvanceddevicesareexpectedtoappearinthefuture,thishandbookcanbeconsidereda milestone reference, since it covers a large spectrum of devices and surveys their structures and properties in depth. The editors selected the topics cleverly, and the contributors are specialized andoutstandingscientistsfromallovertheworld.Therefore,onthebasisofmy experienceinthisfield,IconsiderthisSpringerHandbookofSemiconductorDevicesthemost importantguidebookforresearchers,inindustriesandacademicinstitutes,aswellasgraduate students.Icongratulatetheeditorsandalltheauthorsforbringingsuchanexcellenthandbook topublication. ProfessorEmeritusatOsakaUniversity ChihiroHamaguci,Ph.D. IEEELifeFellow March2021 Foreword by Herman Maes I was born in the year 1947, the blessed year so to speak of the revolutionary invention of the point-contact transistor by Shockley, Bardeen, and Brattain. Perhaps this explains my fascinationtothisdayforsemiconductors,thephysics,thedevices,andthetechnology,inthat order.Ihadthepleasuretoadmirethisfirst“transistor”attheBellLabsinMurrayHill,New Jersey.ImoreoverhavehadtheprivilegetomeetJohnBardeenandattendafewofhisclasses during my postdoctoral research stay at University of Illinois Champaign-Urbana in 1974– 1975.Ievenhadabrief,accidentalencounterwithWilliamShockleyin1974.Igraduatedas anelectronicengineerattheKatholiekeUniversiteitLeuven,Belgium,in1971atatimewhen the semiconductor industry had in fact only just come out of the starting blocks. Intel just launched its Intel 4004, the first processor-on-chip consisting of 2,300 transistors. We were still within the first phase of the semiconductor saga, a few years after Gordon Moore had reportedhisfindingsontheevolutioninsizeofthefirstintegratedcircuits(IC)andlaunched his predictions, later known as Moore’s Law. Only on the basis of a few data points, Moore concluded and predicted that the number of transistors in ICs on a chip would double every year with innovations in lithography as the main driving force to enable transistor scaling. Howmorevisionarycouldamanbe?Andexactlythesekindsofboldstatementshavebeenso typicalandhave,asweexperienced,infactbeendrivingthedevelopmentofthisfascinating semiconductortechnologyfieldoverthepastnearly65years.Thisiscommonlyreferredtoas self-fulfillingprophecy. During our engineering training in the late 1960s, a somewhat older professor, close to retirement, actually strongly doubted whether those semiconductors would eventually break through and therefore still spent a lot of attention to the usefulness, properties, and applicationofvacuumtubes.Fortunatelyforus,therewasthisother,young,dynamic,visionary professor, Roger Van Overstraeten (the founder of imec in 1984), who had obtained his vii viii ForewordbyHermanMaes Ph.D. from Stanford University and who had returned a few years earlier to our university. He was so enthusiastic and inspiring about the latest developments and opportunities of semiconductor science and technology that he could convince many of us to fully engage ourselves in that direction. My fascination was aroused and from then gradually turned into passion and hunger for knowledge. So, we started to look for suited books. In these early years, appropriate textbooks were still scarce. The books that were being produced mostly focusedonspecifictechnologicalandsolid-statephysicsissues.Ipersonallyrememberafew ofthesewhileincollege,theSEEQseriesonthebipolartransistor,AndyGrove’sPhysicsand Technology of Semiconductor Devices, and of course Simon Sze’s standard work Physics of SemiconductorDevices,whichsawitsfirsteditionin1969.Overthepast50years,numerous newhandbookshavebecomeavailable,manyofthemcoveringindetailspecificdisciplinesor aspectsofthestronglyexpandingfieldofsemiconductordevices.However,atthisearlystage there was a need of books providing a broad-scope and balanced mixture of state-of-the-art, prospect,insightandcontextofallinherentaspectsofsemiconductortechnology,devicesand applications,offeringhighleveltreatmentswithmorethanmereencyclopedicknowledge,with attentiontointerdisciplinaryaspectsandtothemutualsupportofthevariousdisciplines:such books,infact,remainedratherscarce. After about 45 years in the twentieth century and 20 years in this twenty-first century, theindustrializedworldhasbeenmarkedbyextremelyrapidtechnological(r)evolutions.We witnessed the incredible and far-reaching developments of the semiconductor industry with growing amazement and disbelief, even to those who were involved in the process from the first rank. This was largely due to the extremely successful scaling endeavors by research and industry. The Intel 4004 processor that we referred to earlier was realized in the 10 μm technology node, today we have reached the 5 nm node (a feature size scaling by 2,000), allowingfabricationofchipswithover50billiontransistors!Semiconductortechnologytoday is a very mature, established industrial discipline. The entire semiconductor market is close to 500B$ and expected to further grow year after year, with even a doubling in 10 years’ time. This industrial progress went hand in hand with (and was also steered by) profound societalchangesinitsmostdiverseaspectsandledusintotheInformationAge.Semiconductor devices,ICs,andproductsplayedanundeniablyprominentroleinthesetremendoustransitions. They are now used in every corner of our “digital” society and are taking an increasingly crucialandessentialroleinalmostalleconomicsectors,informationtechnology,multimedia, communication,automation,robotics,transport,medicalandhealthsector,games,education, environment,energy,cryptography,smartbuildings,andsmartcities. But we are today on the verge of a new disruptive landslide. The transition from the InformationAgetotheDigitalAgeistakingplace,withexpectedrevolutionsintheinterfaces forspeechrecognition,augmentedreality,virtualreality,andartificialintelligence,justtoname afew.Ourbusinessesandtheeconomywillhavetorealizethatthistransitionwillnotcome gradually, in small steps. After all, a characteristic of a revolution is that the change comes abruptlyandisaccompaniedbybreakingwiththepast.Andifwehadthoughttohavereached the pinnacle of technologically possible developments, we realize that many even more far- reachingtechnologicalbreakthroughswillbeneededinthefuture. The semiconductor industry is facing a new era in which device scaling and cost re- duction will no longer continue to proceed on the path followed for the past few decades. Further advanced nodes, even when feasible, may no longer bring the desired cost benefit. Moreover, R&D investments in new lithography solutions and devices below 5 nm nodes are rising substantially. Therefore, the insatiable and ever-increasing demand in the digital era forces the industry to come up with ever more innovative and sophisticated technology solutions to bridge the gap and improve cost/performance while at the same time adding more functionality through integration. The so-called More-than-Moore devices (including ForewordbyHermanMaes ix MEMSand sensors, CMOS Image Sensors,power devices, THz devices) represent thisnew functional diversification of technologies, combining performance, integration, and cost not merelylimitedtoCMOSscaling.Theirroleisexpectedtobecomemoreandmorepredominant. Semiconductordeviceandtechnologyengineersandscientistswillthereforehavetocontinue thequestforingeniouslynewelectronicdevicesandfunctionalities.Weareinfactonthecusp ofbigchanges. Morethanever,semiconductorresearchers,scientists,andengineerswillhavetomaintain the widest possible view of these rapidly evolving fields. They need to be constantly and continuously curious, interested, amazed, fascinated, and triggered by the evolution and the developments in their own fields but also in the domains adjacent but often complementary to their own specialization. They therefore need to acquire thorough yet easy-to-digest, appropriate, timely, and reliable knowledge on the state of the art but also on the newest developments and trends in these domains. And above all, they count on obtaining this informationfromspecialistsandexpertsinthesefields. A handbook is a book capable of being conveniently carried as a type of ready reference work,providinginformation,facts,insights,andinstructionsonaparticularfield,subject,or technique,designedtobeeasilyconsultedandprovidinguseful,quick,butsolidandreliable answersinacertainarea.Itcoversandoffersscientificbasics,fundamentals,methodsofre- search,generalprinciples,functionallinks,andreviewofestablishedorpotentialapplications. Moreover,eveninrapidlyexpandingfields,theinformationandcontentneedtobesufficiently timeless. SpringerHandbooksrespondtosuchneedsfullyandsuccessfully.Theystrivetoprovideand maintainthehigheststandardsofreferencesinwell-selectedfieldsandkeyareasofphysical and applied sciences and technology. They are intended for practitioners in academia and industry, as well as graduate students. They are allowing fast yet comprehensive review and easyretrievalofreliablekeyinformation. ItisSpringerHandbooks’traditiontocomeupattherighttimewithnew,well-considered initiatives for launching appropriate reference books. Recent examples have been the hand- booksonnanotechnology,robotics,mechanicalengineering,andautomation. Today certainly is the right time to launch this Springer Handbook on semiconductor devices,consideringthetransitiontotheevenmorechallengingpost-scalingera. Conceivingandrealizingsuchahandbookisadaring,challenging,andambitiousendeavor. Itisamajorundertakingandrequiresvision,significantorganizationalskills,andcommitment. Onealsoneedstohaveanetworkofexpertswholookuptotheeditor. This Springer Handbook on semiconductor devices you have in hand fully meets the strict requirements and high-level standards of the Springer Handbook series. It provides comprehensive coverage of numerous technological and fabrication aspects of advanced semiconductor devices, focusing not only on advanced “traditional” structures but also on ingeniouslynewandoftenfuturistic,beyondCMOS,andMore-than-Mooreconceptsformany applications.Rightlyso,strongattentionwaspaidtotheindispensableadvancedmodelingand simulationmethodsforpredictiveandsupportingpurposes,aswellasanoutlooktowardnear or far future needs. This handbook is not the first one covering semiconductor technology and device disciplines, and it will undoubtedly not be the last one. But the differentiator, the USP of this handbook according to me, is this mixture of broad scope, thoroughness of treatmentwithstillsufficientaccessibility,thewiderangeofconceptsformanyapplications, andtheforesight.And,worthnoting,allchaptershavebeenwrittenbyinternationalexperts, innovators, and leaders in their field. The reasoned interpretations of developments in each areaarecommendableandoftenunique.Theeditorsdidawonderfulandtremendousjobin notonlyselectingthe45rightfulandrelevanttopicsbutalsoininvitingtherightspecialiststo treatthesetopicsandsharetheirexpertiseandexperiencewiththeknowledge-hungryreader.