LEL—1511DE8 00371CRYSTALLOGRAPH MARTENSITEAN STABILIZATIO AUSTENI - PDF document

LEL—1511DE8 00371CRYSTALLOGRAPH MARTENSITEAN STABILIZATIO AUSTENITMehme Sarikay(Ph.DLawrenc Berkele LaboratorUniversitBerkeley Californi 9472OctobeNOTICPORTION 7 AR ILLEGIBLEha bit reproduced fron availabl th broades possibl avalllbilityThi wor wa supporte b th Director Offic o Energ ResearchOffic o Basi Energ Science Materia Scienc Divisio o thU.S o Energ unde Contrac No DE-AC03-76SF00098A DISTRIBUTIO O THI DOCUMEH I UHUMVIE CRYSTALLOGRAPH O "LAT MARTENSITEAN STABILIZATIO O RETAINE AUSTENITMehme SarikayPh.D Scienc anMinera EngineerinProfesso G ThomaChairma o CommitteABSTRACTransmissio electro microscop ha bee use t studth morpholog an th crystallograph o "lat martensitei lo an mediu carbo steel i th "as-quenched an200° tempere conditions Th steel hav microduplestructure o dislocate lat martensit ( b ca a 0. um wit fairl paralle boundarie an continuouthi film (^20 A o retaine austenit a th lat interfaces o lath crystal o martensite th packet whic ar derive fro differenvariant o th sam austenit grain Th existenc oresidua paren austenit enable on t perfor microdiffractio experiment wit smal electro bea spo size tdirectl determin th orientatio relationship (OR betweeaustenit an martensite Al thre mos commonl observeORs,namel Kurdjumov-Sachs Nishiyama-Wassermann an operat withi th sam sample Thi i b attribute t th increas i th numbe o variant availabl durin th nucleatio proces wher th nucleatio ar individua marteusit laths Relativ orientation o th lath chang fro a lo angl boundar (1-10 perpendicula t ) t a abrup chang ith orientatio b a larg rotation again abou , whic i a commo directio i a packe o laths Th habi i austenit i {111 (paralle t {110 . an thylon directio o th lath i i , (paralle tConsiderabl C partitionin int th v-film (^ wt%an t th a'/y-interfac (^ wt% ha bee reveale b convergen bea electro diffractio an atom-prob studies largel occur durin th transformationalthoug som C diffusio int y i unavoidabl durin rootemperatur holding Austenite trappe betwee th lath i chemicall (b lowerin th M temperature an (b pinnin th high- interfac probabl b i highl deforme t accommodat thstresse create a a resul o th shea transformatio oth surroundin matri (mechanica stabilization) Consequently th overal stabilizatio i determine b thcooperativ operatio o al o th abov mechanisms CRYSTALLOGRAPH O "LAT MARTENSITE ANSTABILIZATIO O RETAINE AUSTENITTABL O CONTENTPagABSTRAC 1 1 GENERA INTRODUCTIO 1 2 MATERIAL AN METHOD 4 2.1 COMPOSITION AN HEA TREATMENT .. 4 2.2 PREPARATIO AN THINNIN 5 2.2.1 Foil fo TE 5 2.2.2 Techniqu fo CBE 6 2.2.3 Specime Preparatio fo FIM-AAnalysi 7 3 AN CRYSTALLOGRAPH O LATMARTENSIT I LO ALLO LO CARBO STEEL ... 13.1 INFORMATIO AN INTRODUCTIO . . 13.2 AN SUBSTRUCTUR 13.2.1 Morpholog 1 13.2.3 Retaine Austenit 13.3 CRYSTALLOGRAPH 23.3.1 Relativ Orientatio o Packet . . 2 Orientation o AdjacenLath i a Packe 23.3.3 Determinatio o OrientatioRelationship betwee Martensitan Austenit 43.3.4 Habi Plan Analysi 5^ -i Pag3.4 AN DISCUSSIO 63.4.1 Morpholog o Lat Martensit .. 63.4.2 63.4.3 Formatio o Lath 63.5 CONCLUSION 7PART-II FIL RETAINE AUSTENIT 74.1 INTRODUCTIO 74.2 IDENTIFICATIO AN TH OCCURRENC ORETAINE AUSTENIT 74.3 CHEMICA ANALYSI O RETAINE AUSTENIT . . 84.3.1 Experimenta Technique 84.3.2 Convergin Bea Electro DiffractionAustenit Lattic ParameteMeasuremen 84.3.3 Measurement o Solut Partitioninint Retaine Austenit b Fiel IoMicroscopy-Ato Prob Analysi .. 94.4 DISCUSSION - MECHANISM O AUSTENITSTABILIZATIO 114.4.1 Chemica Stabilizatio 114.4.2 Therma Stabilizatio 124.4.3 Mechanica Stabilizatio 124.4.4 Summar 124.5 MECHANISM O DECOMPOSITIO O RETAINEAUSTENIT 124.5.1 Mechanica Instabilit 12 Therma instabilit 134.6 CONCLUSION 14APPENDICE 14-ii Pag5.1 A COMPUTE PROGRA T PLOSTEREOGRAPHI PROJECTIO AN COMPOSITSTEREOGRAPHI PROJECTION FO MARTENSITAN AUSTENIT HAVIN DIFFEREN ORIENTATIORELATIONSHIP 145.2 VARIATIO O TH LATTICPARAMETER O AUSTENIT AN MARTENSIT WITALLOYIN ADDITION AN A ATTEMP TMEASUR LATTIC PARAMETE O MARTENSIT BCBE TECHNIQU 145.3 APPENDIX-3 CRITICA DIFFUSIO TIM DURINLAT GROWT AN TH PREDICTIO O THGROWT RAT O MARTENSIT LATH 155.4 DIFFUSIO COEFFICIENT OCARBO I MARTENSIT AN AUSTENIT 155.5 APPENDIX-5 EFFECTIV DIFFUSIO DISTANCFO CARBO DURIN TH TRANSFORMATIOWITHI M -M RANG 15s x 5.6 APPENDIX-6 INTERSTITIA ATOM-DISLOCATIOINTERACTIO 156 ACKNOWLEDGEMENT 167 16 1 1 GENERA INTRODUCTIODurin coolin man metalli an nonmetalli systemunderg martensiti transformations i.e. shea o thcrysta lattic o a paren phase stabl a hig temperatures for a ne phas a th firs o fina stetoward restorin thermodynami equilibrium Althougmartensit formatio i on o th mos importan phenomeno(1-4i Material Scienc an ha bee widel studied thdetail o th mechanism ar stil no wel understoobecaus th resultin structura feature an othe characteristic ar almos alway investigate afte thtransformatio i completedMartensiti transformation define a a firs ordesolid-stat structura chang whic i displaciv andiffusionless i dominate i kinetic an morpholog bth strai energ arisin fro th shear-lik displacements Th transformatio occur i numeroumetalli ' ' an cerami system an bear grea( 2scientifi an technologica significance ' Th reactio involve th nucleatio o th ne phas an growt o int th paren phas durin coolin fro a higtemperature ' ' ' o durin isotherma holding' Sinc n atomi diffusio i involve thparen an produc phase hav basicall th sam composi(1-7tion reaction whic usuall involve volumchang ' occur b shea o certai plane an direction i eac crysta b th cooperativ movemen oth atom alon th semi-coheren interface ' A a resul paren an produc crystal hav fixe orientatio(2 28relationship ' includin a unrotate an undistorteplan i th paren phas (habi plane whic i on o thessentia feature o th crystallography '"'Anothe characteristi o th reactio i tha th productunles subsequen decompositio t stil othe phase haoccurred wil transfor bac t paren phas upo heatin(1 21ideall withou an hysterisis.Martensiti transformatio wa firs recognize istee system an th ter "martensite wa give t th(29phas forme i quenche carbo steels Mos subsequeninvestigation hav bee performe o stee allo systems' ' I general tw type o martensite arclassifie i steels; an "lathmartensites studie hav concentrate o th formation, morphology,(14-18,30-34(4,6,15,17,26,31 othe propertie oplat martensites ' O th othe han lat martensiteattracte littl attentio unti recen years(presumabl becaus o th hig M temperatur an lac odistinctiv microstructur a compare t th characteristi"lenticular plate o martensite whic for i alloy witlo M temperature Th smal siz o th averag individua martensit lath compound th difficult o 3 morphologica an crystallographica studies Th advancemen o transmissio electro microscop (TEM haaccelerate studie o th lat martensit whic occur iman technologicall importan steels.US.17,30,34,35,39,40Part- o thi investigatio wa devote t th identifica an classificatio o th importan morphologica ancrystallographi feature i a serie o lo an mediu allo steel b transmissio electro microscopy Part-I thi fil retaine austenite whic ha bee t b a integra par o lat martensiti structureswa studie i detai an th mechanism o stabilizatio waexamine i orde t understan it occurrenc an clarifit behavio durin temperin o deformation 4 2 MATERIAL2.1 COMPOSITION TREATMENTTh experimenta alloy researc techniqueuse investigatio hav bee developedurin rang progra designi40—45ver demandin structura applicationsSteel wit roo temperaturcontai carbo conten betwee t 0. wtallo conten aroun wt (seTablAllo Alloyin Element M47454234Sample wit dimension (o 10mm b 15m b 10mwer austenitize on a tubfurnac a argo atmospher decarburi2atio int o brine Som samplewer tempere1 0.06 + 2S2 0.10 + 2S3 0.17 + 2S4 0.3 + 3C + 2M + 0.5M5 0.3 + 3C + 2N + 0.5M 5 2.2 PREPARATIO AN THINNIN2.2.1 Foil fo TE'A. Foil fo transmissioelectro microscop (TEH studie wer prepare b a "twinjet technique Approximatel 15mi slicewer "dimet cu fro th hea treate sample an thechemicall thinne t a thicknes o 4 t 5mil i a 4-vol (hydrofluori acid)—CH,COO (aceti acid solutioi a magneticall stirre bath Standar werpunche ou b "spark-cutting an thinne t 1. t 2.5milb grindin throug a serie o Si paper usi-. extremprecautio no t defor o overhea th samples Th discwer cleane wit aceton an electropolishe wit"chrome-acetic solutio (75g Cr_0 - 400m CHJCOOH an 20mdistille i absolutel necessar t homogenizth solutio whic ca b use man time i store wel bstirrin fo a hou a 40-50°C Fina polishin wa dona 20° a 40-5 volt an 20-50m curren (belo 10° anabov " performatio th foi iquickl rinse wit wate an the wit CH_0H(nethy alcohol drye wit paper Foil ca b store eithe we (i a 20 proo ethy alcohol o dr (i a foi holde kep i a dessicato containin dr othe electro medi wer als use suc a 10 HCIO (perchloriacid i CH a -2 t -30°C 60 CH HC10C a 0° an bot a th sam curren anvoltag setting (thes solution ar goo t us onl 6 2.2.2 Techniqu fo CBED Th foil foretaine austenit lattic paramete measurement bConvergen Bea Electro Diffractio (CBED wer prepare ba newl develope "dipping metho whic produce foil iwhic. th martensit phas i preferentiall etche an thretaine austenit film ar expose t th perforrationsThi metho use perchlori acid-aceti acid-glycero solu a -25° wit th sam voltag an curren setting aabov simila t th conventiona windo techniqu exceptha n "painting i involved Th specime slic (4-5milthick i manipulate b a pai o tweezer t chang thdirectio o th sampl t giv th bes irregula thi Afte washin i alcoho selecte thi region cu b a shar razo blad an carefull place i"sandwich grid an store fo furthe use (Th samslic ca b repolishe t obtai mor foil i necessary.C (99.999% an N (99.09% foils t calibratlattic paramete measuremen b th CBE technique werelectropolishe usin a 25 HNO (nitri acid i CH(methanol solutio a -2 t 30° temperature C an Nsample wer originall homogenize (C a 550° an N a an furnac coole t ensur non-defec structureFoil wer examine b differen transmissio electromicroscop technique i differen microscopes Somconventiona brigh fiel (BF an dar fiel (DF imaginan wea bea dar f analysi wer don usin a 7 Philip EM30 microscop tiltin experiments microdiffractio an convergen bea electro diffractio (CBEDexperiment wer performe wit a Philip EM40 TEM/STE whic wa als equippe wit Energ. DispersivX-ra Analyze (EDX an Electro Energ Los Spectromete(EELS) operate a lOOkV Fiel Io Microscop (FIM"tips examine i a RCA-1.2me hig voltag microscop a 80 keV2.2.3 Specime Preparatio fo FIM-A AnalysisTh "two-stag polishing techniqu wa use t preparfiel io microscope-ato prob (FIM-AP fro specimen 15.0m squar pin whic wer cu frohea treate samples Th firs stag use a "doubl layerpolishin solutio o 25 perchlori aci i acedi acifloatin o carbo tetrachloric (CCl. t etc a "neck ith pin Th "necked pi i polishe i 2 perchlori acii 2-butoxy-ethano solutio unti i separate t for 2 tips i continue momentaril i orde t removan deforme layer fro th specimens Bot stage opolishin ar carrie ou a 25 D.C a temperature whicd no excee 20°C Afte cleaning th tip ar thefurthe etche i HC (hydroge chloride t obtai a sharedg (se th optica micrograp i Fig 37) The ar thestore i a desiccato fo futur useDirec chemica analysi o th phase i th structurwer mad usin ato prob microanalyze whic consis 8 o fiel io microscope equippe wit time-of-fligh masspectrometers Th tw particula AP-instrument (thOxfor th U.S Stee one use i thiinvestigatio ar simila i design experimenta condition wit th Oxfor instrumen wil b describeher ( ful descriptio o th instrumen itsel ca bfoun elsewhere e.g. referenc 48)A needle-shape ti prepare a describe abov imounte o a cryogeni (coole wit liqui N_ goniometestage 75m micro-channe plat intensifie anphosphorou scree assembl has a centra hol (1.4mr idia. whic serve a a entranc apertur t th spectromete fligh tub whic i typicall o l-2 i lengt(Oxfor use highl magnifie imag o atom aspecia position o th ti i projecte ont a viewinscree usin a "imagin gas (Ar) Fo chenica analysisth syste i evacuate (^1 torr an hig frequency voltag pulse applie t th ti (se belo fo th experimenta release b "fielevaporation fro microconstituent o th selecte area o ti pas throug th prob hol (whos diamete dictate diamete o th circula are analyze o thThe struc a particl detecto whic identifie everarrivin io accordin t it m/ (mas t charge ratifro th fligh times Th fas digita time (159. MHzi interface wit a PD 11/1 micro-compute whic control 9 th experimenta cycle record th data an provide thon-lin outpu o th accumulate spectr durin thexperimentA typica analysi i carrie ou unde th followinconditions- a syste bas pressur o x 1 torr_ 9 - a backgroun pressur durin analysi o 1 torri.e. wit th imag ga removed- th specime cryosta coole i liqui nitrogen- ar effectiv dat collectio apertur o 2 A a puls fractio o 15%- a puls repetitio rat o 5 Hz- io collectio rate o betwee 1 an 1 ionpe pulse- a neo (o Ar imag ga pressur o 4 x 10 torrTw genera mode o analysi ar possible selecteare o rando are (sequentia th firsmod specifi region ( retaine austenit fil o ainterface ar chosen an analyze o a point-to-poinbasis th sequentia mod a rando are i chose anth ion fro a colum o materia paralle t th specime ar analyze (se Fig 40) Th siz o th are oanalysi (typicall 10-3 A diam size i determine b thsequentialit o th apertur an th magnificatio o thFIM A materia i remove compositiona fluctuatio o element du to e.g. retaine austenite a interfac 1region carbide ar recorde i turn Th resultin datchai o tim serie ca the b processe usin statisticatechniques example simpl graphin involve takin a a ever 5 ion a eac ste (see e.g. Fig 36)Not tha th numbe o ion ca b relate t th distanc fro th tip Simpl geometr show tha fo a prob diamete o 1 A th are covere fo th probcorrespond t abou 3 atom i a iHenc a ever 3 atom 2. A plana materia i removefro th ti whic correspond t 8 A thicknes fo ever ioiis O cours thi valu i ver sensitive anchange dependin o th factor suc a th crystallographiorientatio o th tip th radiu an th shap o tip etc 13 PART-MORPHOLOG AN CRYSTALLOGRAPH OLAT MARTENSIT I LO ALLO LO CARBO STEEL3.1 INFORMATIO AN INTRODUCTIOTw basicall distinc morphologie o martensit fori ferrou alloys; an "lathmartensite mainl dependin o th composition Platmartensite (wit plat siz fro fe t man micronsoccu i hig alloy e.g. 12(9,10,39,53 Mg alloy carbonnitrogen ' ~ ' steel wit M temperature usuallbelo roo temperature Lat martensite ar foun i steelcontainin relativel lo alloyin element (tota 10-1 wt%o i lo allo (5- wt%)-mediu carbon'60 nitrogen M generall� 200°CExtensiv studie o th variou aspect o platmartensite hav clearl show tha th "plates" whic fordurin cooling isotherma holding (20-22,32,60ar lenticula wit a "midrib (consistin o extremel fintwins a th cente lyin alon {225 o {259 austenitplane (habits).shea durin th transformatio i accomplishe b twinwhic ar usuall o {112 , plane i , directionTherefor bot side o th midri i consiste o fin (50-20 A structura twins.Plat martensiti structure contai a considerabl amoun 1o retaine austenit (usuall 1 vol o more a islandbetwee th plates. crystallographiaspect hav bee studie i dept (b X-rays an variouorientatio relationship betwee y an a' hav beeestablishe (specifi t certai allo systems ' th result discusse i term o curren phenomenologi theorie o martensit transformation ' Finallnumerou studie o nucleatio characteristic ' growt kinetics ' ' ' hav als bee conductedDespit th above-mentione accomplishment b mantinguishe worker i th fiel inconsistencie stil exis tha plat martensit continue t attrac muc attentionLat martensites despit thei wide-sprea occurrenci th technologicall mor importan lo carbon-lo allo no attracte comparabl attentio untirecently. th shap o th lath hano bee establishe clearly ' ' I i know tha allat martensite for athermall an tha M temperature arusuall wel abov ambient Ther i n stud o nucleation(1 3 33onl a fe dealin wit growt kinetic ' ' wit mosconcentratin o th morphologica features ' ' ' Th inhomogeneou shea occur b sli s n micro-twinoccu an th parallel}opiped-shape lath contai a higdensit o dislocations ' ' ' Sinc n retainea'and stan fo produc martensit (bc o bet phas any an A stan fo paren austenit (fee phase 1austenit ha bee found th crystallographi studie werdor. indirectl o martensite U5,33,36,39,83-87 maiconcer ha bee th relativ orientation o adjacen lathwhic wer identifie t hav lo angl o hig angl{3 3 83—87boundaries ' o the ar interprete a beineithe i twi relate o nea twi relate orientations ' 3 8 3 84' ' ' Th macroscopi habi plan wa identifie t beithe nea(85 els i th vicinit o {111 .I th curren investigatio empathesi wa place oth characterization b TEM o morphologica an crystallographi feature (relation betwee th .individua martensitunits lath an th large unit "packets" i thlat martensiti structure i lo carbo (0.0 t 0. wt%-lo allo (tota abou 5 wt% steels Sinc th presen(4 88alloy contai retaine austenit withi th structureth importan aspect o th crystallograph o martensite a orientatio relationship an habi plan analysescoul b establishe directl o austenit b som recentldevelope microdiffractio techniques Th result wil b o th basi o th overal strai minimizatiowhic i a characteristi o martensite an dictate botth morpholog an th crystallograph o thi clas ophas transformations 13.2 AN SUBSTRUCTUR3.2.1 Morphology Th microstructur o "as-quenchedlo C martensiti steel i compose o severa distinccomponent o rathe differen dimension an shape a i Fig 1 Th larges ar th equiaxe prioaustenit grain abou 100-20 y diamete i size Thecontai irregula packet abou 20-3 vi i siz whic ar(15compose o man "lath martensit crystals Individua"laths eac packe hav fairl paralle boundarie anar aligne nea t a habi plan varian o th parenaustenite I mos instances th "laths ar separate bcontinuou thi film (^20 A thick o untransformeaustenit (retaine ar singl crystalo martensite an hav a roughl parallelopipe shape i.e.a b c Typica dimension ar a = 0.2 - 0.5 vimb = 1- p an c = 10-3 y fo 0.3 alloys Th "lathswer foun t b somewha wide an longe i 0.1 alloys3.2.2 Th substructur o th lathconsist mainl o dislocation wit a densit estimate a1 3 1 cm/c . Becaus o thi hig dislocatio density i igenerall impossibl t resolv th individua dislocation(Fig Henc detaile dislocatio analysi coul no bdon i an o th alloys Th C conten o thes alloy ito lo t produc muc microstructura twinnin an thewer observe o {112 , an onl locall i th 0. wt c alloys 1'PrioGrnMartensitPacket5 AtFig Genera microstructur o th alloy a a lomagnification Thi exampl b optica microscop showth configuration o lath an packet i a singl preaustenit grainXB 821- 1*.-Fig Genera microstructur o th alloy i th asquenche condition (a Th heav dislocatio densit(strai contrast an th effec o autotemperin withith lath ar reveale an th lat boundarie arclearl resolve ( M Ih sam a abov i 2 N containin alloy 1I i interestin t compar th tw packet oimpingin martensit laths labelle A an B whic appea(89i Fig 3 Th lath i packe (A ar "twi related"(i.e. contrast—thi i discusse i Sectio3.3.2) presenc o regularly-space twin i thsmal lath wit retaine austenit a thei boundarie ipacke (B suggest tha packe (A forme firs (sinci contain twin-relate laths probabl du t autocatalyti nucleatio jus belo M ) Th lath i packe (Bprobabl evolve late an encountere th lath i packe(A) th impingemen o th tw packets twinforme t accommodat th stresse create i thi regionI i generall note tha twins usuall i everothe lath ar simila i bot dimension an separationMicrodiffractio characterizatio o suc area revealetha th twin ar lyin o {112 plane o martensitesimila t transformatio twins ' ' Th micrograp i Fig 3 als indicate tha som"secondary lath formed afte impingement alon th growtdirectio o th origina crystals Thi configuratiosuggest tha thes smal crystal o martensit formeafte th impingemen o th primar laths Not als thath end o th lath i packe (B ar curve i thi region Fig Detail o a packe boundar wher th packetA an B interven (Allo .J 825-416 1Thi i observe ver commonl an mus b relate t thgrowt mechanis o th lath (se th discussio sectioo PagSinc th M o th alloy ar quit hig�(300°C) i tim fo som C redistribution i.e."autotempering ' o th martensit matri t forC-cluster alon dislocation o sometime the eve groint carbides Thes effect ca b muc mor pronounceher tha i hig C steel wit ver lo M temperaturesTh foil studie a hig resolutio b wea bea dar fiel(WBDF techniqu i th ai quenche conditio sho thevidenc o suc clusterin withi th martensite Foexampl i Fig 4 th dislocation ar show i a heavilstraine matri (a) C-cluster see i th D imag (brevea highl advance stag o autotemperin (ma b inter a a earl stag o e o Fe, carbid precipitation3.2.3 Retaine Austenite A integre. par o thmicrostructur i th generall continuou thi film ountransforme retaine austenite o abou 50-45 A i thickness th lat boundarie (Fig 5) The ar trappe th lath an ar highl deformed Retaine austenit i als presen a th packe an prio austenit graiboundarie a thi films However ther i n retaineaustenit presen whe th adjacen lath ar twin-relate(packe B i Fig 2) Carefu microscop i require tidentif thes film i thi kin o comple structure Th 2Fig (a B an (b WBD image showin C clusterin (a a earl stag o tempering a th dislocationwithi a lat i a as-quenche sampl (2-MXE 825-415 2Fig B (a an D (b micrograph revea thretaine austenit a thin continuou film a th lat ( Ni allo i th 200° tempereXB 794-455 2presenc o thes remnant o th paren phas withi thmicrostructur dictate th morpholog an propertie o thiclas o steels Therefore a separat detaile sectio(Part-II i devote t th identificatio an stabilizatiomechanis o retaine austenite3.3 CRYSTALLOGRAPH3.3.1 Relativ Orientatio o Packets A noteearlie group o u t 5 paralle singl crystal omartensite havin commo growt habits appea a packet ith TEH A exampl i show i Fig 6 wher th lath ithre impingin packet (A B an C al appea clos tedge-on pattern take fro th individuapacket an fro th packe boundarie ar show i thcorrespondin selecte are diffraction SAD pattern (A BC D an diffractio pattern fro th regionA B an C indicat a singl zon axipattern take fro th packe boundar region D an E givtw (111 pattern slightl rotate abouwit respec t eac other Fig 7(b show anothe cubi(111 SA patter containin reflection whic agai suggessom sligh misorientation I previou studie thi wa(3 B 84attribute ' ' t smal angl boundarie betwee thadjacen lath i a packet However a detaile analysishow tha th reflection i thi diffractio patter comfro packet wit lath "end-on (lath 1 an "edge-on(lath 2 i th micrograph Henc regio 2 i no a "single 2Fig B imag show thre impingin packet A Ban C i a 0.3 alloy Th microdiffractio patternA t E take fro th correspondin region i thimage singl ( t C o tw superimpose , pattern ( an E (Allo 4)a' 2lath ha bee ascertaine tha th lath (1 hav a {111} an lath (2 a Similarl iFig lath A hav a lath B hav ahabi an lath C hav a Thi i a rotatioaroun a vecto norma t th plan o thᄀᄐ I superimpose SA pattern (Figs 6an E an Fig 7b) th rel-vector correspondin tdifferen packet ar the expecte t b separate b a 10.54 rotatio (refe t th analysi i Fig 7c) Thmeasure separatio (10° i th diffractio pattern agreequit wel wit thi predictio (se th SADP' ian ar fou differen {111 plane i thaus.tenit lattice Hence fou o mor variant o packetca operat i a singl austenit grai (som variant maals b repeate i differen part o thpondingly fou o mor packet ma fil u th austenitgrains i Fig 6 thre separat {111 variant oaustenit (namel (111) (ill) ar presen(correspondin t (Oil) (loi o martensiterespectively)Th larg strain (dilatationa an principa sheastrain create durin th austenit t martensit transformatio ' ' ' ' i steel ar accommodate b som meansuc a shearin (sli o twinning i martensite varioudegree o rotation o adjacen laths an deformatio oretaine austenit i th lat boundarie (t b discusse 2Fig (a B imag an (b th SA patter take froth encircle regio i (a) Not tha th lath ipacke (2 ar i end-o configuratio shortedimensions an "b" ar show a mode o th imag (a an give th analysi o thdiffractio patter (b (Allo 4Fig B imag (a an th portion o th microdiffractio pattern correspondin t th region indicateb (1 t (8) Notic th zig-za interface (Allo 2below . 2later i additio t stres relievin proces inheren ith differen transformatio variants Pig 8e i a uniquexampl wher a leas si packet ar s oriente witrespec t eac othe i suc a smal region A show ithi figure eve a singl lat ca correspon t a packe(e.g. (1)) Thi i confirme b microdiffractionpattern whic sho eithe a singl pattero superimpose pattern (2,3,6, an 8 rotate ^70(o 110° accordin t th sit o whic th electro probwa placed3.3.2 Orientation o Adjacen Lath i a Packe(i Twi relate laths Thi les frequentl observerelativ orientatio betwee adjacen lath i on i whicth lath o eithe sid o th boundar hav approximatelopposit orientations i.e. ar rotate + an -90 witrespec t thei commo axi an henc ar "twirelated. Th situatio i als observe i highe carboalloy an a exampl i show i Fig 9 Her i th centrapacke B th lath ar twi relate (dark-brighAl microdiffractio pattern fro th individua lath havnearl th sam typ o symmetries i.e. al pattern• (91A microdiffractio patter i forme b placin aelectro bea o certai siz (usuall 50 A o smalle idiameter o a regio o interest whil a SA patter iforme b placin a selecte are apertur o a regio anlimitin th diffractio comin fro tha certai are(usuall 0. vi o Fig (^ A sectio fro a austenit grai befor th transformatio haoccurre ( i electro bea Th sam grai i (a afteth martensiti transformation showin th arrangement o twin (T anmartensit (M produce fro o austenite (c A TE showin a packe B whic contain twi relate laths Insetar th SA pattern iᄐ fro th individua lath (1 throug (5)XB 824-349 2( throug 5 bu wit reflection havin differen intensities lath hav thei (110 face paralle t eacother th twi plan i (110 an th directioi [111 , whic mean lath ar i [111 an [111 zon axiorientations successively Wit th bea exactl o a zonaxis lath shoul appea t giv th sam contrast bufo a ver smal misorientatio (0.5 a i thi examplesuccessiv lath appea brigh an dar du t th phascontrast(ii Paralle laths Thi commonl observe situatiostimulate th interestin hypothesistha i a grou o paralle lath ever successiv lat ifurthe rotate aroun thi commo "n i a "Tr-rotation result (e.g. i th cas otwi relate lath n is o course two i.e. th rotatio complete b tw adjacen rotatio i no a physica on bu result a a consequenc o th principashea vecto assumin differen variant o austenite Thihypothesi wa advance a i minimize th strai create th nucleatio o individua martensit unit ove a relativel smal numbe o lath i a packet Fo examplewhe 4 o 5 larh gav al diffractio patterns iwa interprete t mea tha eac individua lat wa i a differen (z.a. orientation e.g.&#xIll0; ᄐ etc. aroun th commoᄀ (e.g. seFig 2Fig Th experimen t confir th hypothesi thath successiv lath ar relate geometricall b a rotatio aroun th coiiatio o thlath (a befor tiltin an (b afte tilting (cStereographi projectio showin th expecte change ith typ o th diffractio pattern (alon th pole 9degree t commo [110] afte tilting 3I th presen alloy th abov situatio i frequentlobserve an ne experiment wer designe t chec thhypothesi reviewe above Th experimen require orientin packe o interes i suc a manne al th lath lin u"edge-on" i the don i on directio perpendicula t commoᄀ Assumin th firs lat i i [111z.a orientation an th secon i i [111 z.a orientationan s on tiltin th counterclockwise aroun[110] b 4 degree bring lat (1) e.g. nealat (2 nea othe lath assum z.a orientatio accordingl (Fig experiment ar tedioubecaus durin tiltin throug hig angle an maintainin a smal are i th microdiffractio conditio i difficultWit magneti alloy th proble i eve wors becaus bendino th foi b th magneti flu ma revers durin tiltingNonetheless experiment wer carrie ou t obtaith require conditions e.g. smal foi dimension (prepare b windo technique t decreas th amoun o magnetiflu line throug th carbo alloy wher thlat boundarie ten t becom straighter th lath longean th dislocatio densit lowerTh concep o suc a experimen i show i Fig 11wher i (a an (b) th microdiffractio pattern take(wit electro prob siz o 40 ft) fro individua regionmarke (1 throug (12 ar given On ma not thapattern take (1 throug (4 shoᄐ symmetry an (5) 3"TILTINGFig (a Th analysi o th tiltin experimen (0.carbo pattern (1 t (12wer take fro th correspondin lath befor tiltingXB 817-703 3(6) (10) (11) (12 shoက symmetr wit slighmisalignments Afte tiltin th foi abou 30 aroun th[110] commo pattern fro th samindividua lath a befor agai gav simila pattern folath (1 throug (4) i.e.ᄰ an fo (5) (6) an(10 throug (12) i.e. aroun , Fig 11-b Thresul i tha th firs stac o lath ((1 throug (4)cluste abou th samᄐ pole sa [111] an no(4 92differenᄐ poles a ha bee propose earlier ' Analysi o th secon stac produce a simila resulalthoug unlik th 11 vs Il cas [100 canno btinguishe fro th [010 poleThu i i clea tha th lath i a stac ten tcluste aroun a particula pol wit onl sligh misalignmen i a directio perpendicula t a commo [110]„ Thmisorientation o th lath nea [111] an(an als nea [113] an [112 afte tilting wer measuret a accurac o 0.5° Th results plotte o a Kikuchma alon th (110 pair indicat tha th misorientatioi abou a directio i plan an varie 0 t 10 ieithe sens aroun tha directio (Fig finasituatio i als show o a [001 stereographi projectio(Fig whic show th pole correspondin t thdiffractio pattern observe befor an afte tiltinaroun th (110) trace 3Fig (b Th microdiffractio pattern afte tiltingPattern (7) (8 an (9 wer take alon th sam lath[(2) an (5) t assur th sligh misorientatioalon a particula lat whic i no mop tha 0.5°XB 817-702 3ANALYSI o th TILTIN EXPERIMENKIKUCH MAPN / y ••TI3/T1 II/ _ _ Jf~—t,.— -•U&*-^W® /InnnSTEREOGRAP IPROJECTIONHi-^-'«V ' \iu,-\ / I /"\ \ I I \ / \ / \ / I i X 7M .XB 824-922Fig (c Th analysi o th orientation o thadjacen lath o a Kikuch map (d Th analysi o a [001] projection 3(iii Abrup change i lat orientation e.g. froᄐက i a packet I wa note earlie tha thstack o lath withi a packe ten t cluste aroun oက „ Th questio stil remain whethe lath i a particula stac ar i onᄐ orientatio wit othestack i differenᄐ Furthe tiltinexperiment o severa group o lath i a particulapacke (i bot 0. an 0. wt carbo alloys wer carrieou t revea th exac chang i th orientatio betweeadjacen stacksA typica exampl o a packe wit paralle latboundarie whic contai retaine austenit i show i Fig12-a pattern fro individua lath (1throug (19 sho tha ther i a cycli patter o abrupchange i orientatio fro tက an the bac t . Th microdiffractio pattern indicat tha lathr(1 (11) (14) (15) an (18 ar nea a zonM axi (z.a. orientation whil th other singl an igroup ar clos t a z.a (FigAssumin tha th ove th whol numbeo lath i fixed th questio i whethe th lath fro batche o hav th sam o differenM indices whethe differen stack belon t th samparticula pol o not T chec thi point anothe experimen wa conducte o th sam regio o th a particula fixe i spac 3BEFORE TILTING Fig Th tiltin experiment (a Microdiffractiopattern (1 throug (19 wer take fro th regionshow i th B imag befor tiltin (Allo . 824-349 3AFTER TILTING mFig (b Microdiffractio pattern take fro thcorrespondin lath afte tilting 3th foi i tilte counterclockwis abou 35° Th microdiffractio pattern the take fro th correspondinarea fro th lath (1 throug (19 ar show i Fig 12-bI ca b see tha th lath previousl havinᄐ orientatio no ar nea zon axi orientation an lath i orientatio befor th tilt ar no nea a z.aTh result o th analysi o th preceedin orientatio change betwee lath ar summarize i Fig 12- f wher eac figur represent a portio o th [001cubi stereographi projection superimpose t a "Wul Net,nea 111 001 Il3 an 11 poles respectively Th Wul i oriente wit respec t th projectio an th pattern (i Fig 12- an b t permi analysi misorientatio wit hig accurac hal a degre o arc) Th poles therefore i Fig 12- througf respectively aligne wit respec t th pattern i (a (b (als refe t (9 whic show th relativ positiono th abov pole wit respec t [110 pole tha i thpole [001] [113] [Tl2 li o th trac o th(110 pattern obtaine (Fig 12-b afte tiltin th foi revea tha al stack correspon t thsam pole sa ther ar onl sligh misorientatio betwee adjacen lath withi th stac whil thstack themselve ar separate b th lath(s wit th[001] zon (i Fig 12- an d tha th angl 3XB 824-349Fig (c throug (f ar portion ographi projectio correspondin t th pole [Oil].[Il3 an (g Relativ position opole wit respec t commo [110 pole (h Th summaro th analysi o a [001 stereographi Kikuch ma o Fe 4betwee th lath belongin t a certai z.a (eithe [111o [001] chang fro fractio o a degre t 5 o mor (an thes are o course repeate betwee th lath i Fig 12- an interestint not i tha th misorientation ar clustere nea thtrac o (110 plan an tha th misorientation fal oeithe sid o th pole o tha trace Th overalanalysi i summarize i a [001 Stereographi Kikuch mai Fig 12- whic reveal th sens o tiltin throug th[110 Kikuch bandTh abrup chang betwee th pole [Tilt [H31 [112],. ma b explaine o th basi th orientatio relationship betwee th retaine fil an th adjacen martensit lath a discusse Sectio 3.3.4 (not tha i tha sectio i Fig 19- th pattern take fro th are show i Fig a o tw martensit patterns [Il3 an a superimpose austenit [112 smal chang ialignmen aroun th wil hel t minimiz strai create durin th y t a transformation(iv Observatio o th lath belongin t differenz.a withi th sam packet I i commonl observe thaparalle lath i 0. wt carbo alloy hav mor tha onpol syste an tha lath belongin t severa differenpole ar arrange sid b sid wit th commo , axiperpendicula t thei broa faces Suc a situatio i 4show i th SA patter i Fig 1 whic i take fro a regio withi a singl packe whic contain paralle lath(se th severa pattern correspondin tmartensit an on correspondin t austenit ar superimposed Fro th indexe schemati diffractio patter i13(b) wa foun tha th martensit pattern belon t[121] [311] [111 z.a an th austenitpatter belong t [121 z.a Not tha withi th accuraco SAD th pattern ar no i a z.a orientatio an artilte roughl i th directio perpendicula t [110direction i.e. withi (110 , plane tilte i bot sensefro th electro bea direction Th importan fac heri tha s man variant o lath ar presen withi thsam packe al havin thei [110 , directio i commonThi poin wa als checke b D imaging Whe th reflec belongin t differen individua z.a wer use foD analysis th lath belongin t thos correspondinvariant chang contrast However whe th commo g..reflectio wa use , al th variant change contrassimultaneousl includin th retaine austenit whos (111reflectio i superimpose wit (110 , Th retaine wa als reveale b D imagin usin a isolatereflectio suc a Withi th accurac o SA orientatio relation betwee austenit anmartensit plane an pole ma b state a follows Fig Brigh fiel (a an selecte cire diffractio patter (btake coverin man lath i (a) Th orientation ar sketche ic Dar fiel image ar show i d e an f fro reflection 1 2an 3 respectively 4(110)an [121][311][121][lll]Furthe discussio o thi subjec appear i th sectio oth orientatio relationshi analysi (Sectio ith Discussio (Sectio 3.4)(v Occurrenc o a "packet betwee paralle lathsI i frequentl observe tha i highe carbo alloy sompacket o paralle lath contai a occassiona lat thai slightl misoriente wit respec t thos o bot sidesi.e. th situatio o "lath (2 i Figur 14 Th microdiffractio pattern take (usin a smal selecte are nea th boundarie o lath (1 an (3 arisolate singl diffractio pattern (herth diffractio patter take fro th "lath (2 show twsuperimposeᄐ pattern slightl rotate wit respec teac othe abou th electro bea direction i.e. neaᄐ Th analysi show tha thi i no becaus thlath (1 an (2 o (2 an (3 hav lo angl boundariebetwee them bu i i du t th fac tha th "lath (2i indee compose o muc smalle martensit crystal (olaths whic ar viewe "end-on i th imag (Fig 1- aoppose t "edge-on configuration o th lath 1 3 4 4Fig A TE micrograp fro a regio showin severalath wit paralle boundaries Not tha th SA pattertake fro lat (2 consist of_tw superimpose patternbelongin to say [illjj an 11 (Not alsi patter (5 retaine austenit reflection ar alsrevealed (Allo . 796-821 4etc.) situatio i simila t th on mentione iSectio 3.3.1 (Als se an th crystal regio (2 assum differen austenit variano whic i paralle t {110}., tha th lath(1 an (3 (whic bot hav anothe austenitvariant) regio (2 shoul b a ne packeforme betwee th lath (1 an (3) I electrobea directio tw superimpose ar observe fro th crystal i regio (2 an fro th lat(1 (o (3)) whic ar rotate 70 wit respec t eacothe aroun (o ) i.e. a rotatio b a anglbetwee th {111 plane o austenit (her (111a i SA patter ma als b interprete a a 10 betwee th belongin tdifferen zone o martensit (her [111] an [lll]„se Fig 7 fo a schemati i alwayretaine austenit betwee th lath whic wa reveale bD imaging However on ma not tha retaine austenit ar ver wea an difficul t se excep i SA (5)3.3.3 Determinatio o Orientatio Relationshipbetwee Martensit an AusteniteOn importan aspec o th crystallograph o martensi i i th lattic orientatio relationship(OR betwee th paren an th produc phase whic identifth correspondenc betwee th plane an th direction i 4th tw phases I mos studie th investigatorfoun a singl O.R specifi t a particula allo typeFo example i Fe-Ni allo th Bai O.R ' waidentified Simiarl i Fe-Ni- martensite Greninge anTroian (G-T O.R wa found I man othe lo allo(1 3an mediu carbo steel eithe Kurdjumov-Sach (K-S ' ' 40,64,83,84,93 Nishivama (N' O.R. hav bee reported Thes author reachethei conclusion throug analysi o indirec measurementssuc a relatin know trace i martensit t th crystallographi feature o austenite.(93a recen observatio mad i a clas o mediu carbosteels t th one use i thi study crystallographi analyse wer performe i whic i wa show thabot K- an N- O.R. ar presen withi th sam regio oth specimen.*Actuall a smal amoun o austenit doe remai untransforme a thi film betwee th martensit lathwhic wa lon overlooke i lo carbo martensiti Fe-alloys.(15,33,39,53,56,80-87 tQta amoun o retaineaustenit i relativel lo (2- ' an detaileX-ra analysi ' i no possible However determinatioo th orientatio relationship betwee austenit anmartensit i possibl b mean o precis electro diffractio analysi i th TEM Fig (a BF (b typica SA patter an (c D usin gn02 reflectiorevealin th genera appearanc o th film ( N Alloy . XB 796-788 4A typica microstructur specifi th allosystem th (b) take— (8 89usin reflectio (heraustenit change contras th lat boundariea continuou films are diffractio pattertake fro coverin severa lath thaustenit films patter therbasicall thre superimpose pattern present namelo i . thcorrespondenc th th bstate[110y(111anY (2(111)I follow that identitie correspondt th K-earlier.(40,45,89,92,93I orde se th betweeO.R.s i commonl observeO.R. betwe an y i Thes a(110)[lll]an H tll2 4INTERPRETJV J, \ Lil31Di 100JCCIDN-IIQ [Tilniu' J K-Fig Indexin an th interpretatio o th frequentloccurrin 10 • SA pattern ThM M A pattern belongin t thos z.a appea t b fairlsymmetrical 5)[001]an fi fro [112Th correspondenc betwee th plane an th axe i botphase ar als show i detai o [0Q1 stereographiject i Fig 17(a an (b) On ma se fro thi figurtha th interrelationship betwee K- an N- i tha wheth crystal havin K- O.R. i rotate abou a commo[Oil , abou 5.26° on achieve th N- O.R Fig 1 alsreveal thi interrelationshi betwee th tw O.R.schematicall wher th plan o th pape i th cas oaustenit i {111 an i martensit [Oil , derive fro [211 i N- relatio whil ther i a betwee thes direction i th K- relationTher i als a thir kin commonl observe O.R.namel G-T i whic th relation betwee th planean th axe ar almos thos o betwee th N- an K-variants B makin us o Fig 1 thes relationma b summarize a follows(110 . withi abou 1 fro (111a Y [lll] fro [U0]an [oillTh interpretatio o th tripl / /diffractio pattern suc a th one show i an16 base o th accurac o th SA patterns th 5/OQ- v.'^uNANIil3| 12!:i!JN--V.''/;„•.:.^..HV, 5,? fr m _.4.7 '.,'• .. S57 ,Fig Stereographi projectio analysi o th moscommonl occurrin orientatio relationships namel K-(a an N- (b) (Indice o th paralle plane anth relation betwee th direction ar indicate oth right. 824 -�a--^-o-—-joBTEN;T ^fr' 0 ' 51OiARTENSITi -0 5 Fig Th relationshi betwee th K- an N- O.R.'s Plan o thpape fo austenit i fo martensit i th5.26 betwee th tw O.R.'s XB 818-1115 5symmetr o th pattern belongin t differen z.a. an thdeviatio o thes z.a wit respec t eac other ' Becaus o th spherica aberration th inaccurac i a SA(9 95patter ca b 5 ' ' an eve wit th Ryde an Pitsc(95procedur th accurac i stil no bette tha 2(extremel difficul i th presen cas becaus th analyserequire tw set o reflection belongin t tw nearlparalle zone an th intensit o th reflections witcomposit patterns a i ou case som reflection arsuperimpose whic increase th intensit o thosI addition th hig dislocatio densit bot ir y an as diffuse th Kikuchi pattern tha the canno b use tdetermin orientation t withi 1° Thi accurac ineede becaus onl a 2,5 angl distinguishe th K-S N-Wo G- O.R variantsTherefore i orde t unambiguousl determin thexac O.R. on ha t obtai singl isolate pattern fro austenit an th martensit crystal i eithe obot side o th boundary ' Thi i obviousl no a straightforwar experimen t perform give th fac thati additio t th hig dsfec density th austenit fil i(92ver thin th microdiffractio method whic utilize a smal (^40 A dia. an highl convergen electro bea tobtai diffractio pattern fro a are roughl equa t thprob size 5A numbe o experiment hav bee conducte b microdiffractio method A exampl i show i Fig 1 whicinvolve a regio givin th frequentl observe tripl / • diffractio pattern B usin a smal prob (convergenc angle 5 mrad i th TE mode thmicrodiffractio pattern wer taken fro th region i(a) show i (b) (c an (d) whic ar identifie a[111 [110 an [100 i , respectively Th foi wa thetilte abou th commo , // axe b approximatel 30 an th microdiffractio pattern wer agai takefro th correspondin regions Tiltin wa don t ensurth reproducibilit o th analyses Th resul o thexperimen i show plotte i a [001 stereographijectio i Fig thi analysis on obtain thfollowin relations(110)[lll]an [001 , [011 , (i opposit senseo (111)[ll3] ofan [112 , [112 , (i opposit senseI ca b see fro th foregoin discussio tha thesrelation ar clos t eithe K- o N- O.R.s Howeverwithi th reproducibilit o th tiltin experiments th 5Fig Microdiffractio experimen t determin th correc O.R betwee th martensit lath ( an 3 an austenit (2 i a 0. carbo alloy Micrograp (a show tharrangemen o austenit an martensit o bot sidesMicrodiffractio pattern (b) (c an (d wer take beforan (e) (f an (g wer take afte th controlle tilting(h i th SA patter take fro th regio an (i i thindexe schemati patter correspondin t 5Before Tilt After Tilt iVbrtesi'te-D u Austenite �6 • Vbrtensite£ A A Fig Stereographi analysi o th orientatio relationship i Fig 19 Not th scatte o th pole o thmicrodiffractio pattern wit respec t th electro beadirectio o a grea circl correspondin t commo[110]fd//[Hl] zone (Relativ misorientation o M an Mar correcte accordin t tha o A. „ „„a 825-988 5relation ar muc close t th G- O.R Henc i ma bconclude tha bot martensit lath ar i th G- O.Rwit th retaine austeniteTh result o man experiment involvin retaineaustenit an lath o eithe o bot side o th fil(sometime includin tiltin t anothe 2.a orientationshowe tha ther i indee a scatte o pole betwee K-an N-W Therefore orientatio relationship ca b anywher betwee K- an N-W bu mostl occu aroun G- O.R3.3.4 Habi Plan Analysis Habi plan analysi opacket o lath wit fairl straigh boundarie wa carrieou o severa differen alloys Th trac analyse werdon o stereographi projection bot befor an aftetiltin abou [110 , directio commo t al lath i a packet analysi give i Pig 21- show th tracnormal fo th boundarie whe th lath ar i differen orientation Eac pai o th line indicate th twextrem deviation fro th norma trace Th boundar trace fal int a specifi zon whic define thindice o th pol o th norma t th habi plane Thpresenc o retaine austenit i th sampl enable direc o th habi plan knowin th fac tha th(110 , i paralle t (111 a ensure b microdiffractioinformatio take fro austenit durin th tiltin 5Oltu XB 825-977('to},-Fig (a Analysi o th habi plan determinatio oa projection Th line represen thtw extrem deviation fro th norma trac fo eacorientatio befor an afte tilting (b Approximat shapo th lath an th film an correspondin crystallographiplane an directions 5experiments a sense therefore th habi plan analysi i don directl o austenite A show i Fig 21-awhic i a composit stereographi projectio belongin ta an retaine austenit (i K- habi plan iver clos (withi 5° t (111 . Fro th observation o man lat configuration i thTE a approximat shap o th individua martensit unitha bee determined Thes unit ar usuall see i th"edge-on (Figs 6 an 15) occasionall "endon 7 an 14 an almos neve "fac on bu sometime obliqu t th broa faces Thu th unit ar"laths parallellopipes wit curve "nd (seee.g. Fig 3 an retaine austenit film o varyin coverin th surfaces A show i Fig 21-b thbroa face o th lath ar paralle t (110 , (whic ara paralle t (111 ) th lon directio o th lath ar i , directio (/ t , an th othe orthogonadirectio (i th plan o th broa fac an perpendiculat th lon direction i , (whic i paralle t e3.43.4.1 Morpholog MartensiteA illustrate sectionstur carbo steel "latmartensite th thacrystal frocharacteristi lenticula "plates whic for carbo p i ui heigh 10-3 y i ar largeth lowe content prio austenit4grai wil transfor t 1 whicarrange a patter energresultin fro shap change th volumchang th arrang themselve group whic tak th 4graphi variantsTh lat shape martensit singl crystal hav curveend boundarie wit interlat continuou filmo retaine austenite contain hig leve accommodatio th create thdislocatio densit appear th conteno increase becaus th g magnitudth principa stresse whic musdislocatio densit achieve estimate b11 cm/c 0. wt alloy 6Th retaine austenit films presen a th boundariesar mostl continuou an hav width varyin betwee50-20 A (sometime extendin u t 50 coulno b an direc correlatio mad betwee austenit thicknes an th lat width Nonetheles i i observe thath film ar wide i lowe carbo alloys Retaineaustenit i als highl deforme an ha a defec densit1 3 increasin u t 1 cm/c i 0. wt C alloysExtensiv autotemperin occur i th alloys Th M an M temperature ar hig s tha carbo redistributioca occu durin an eve afte th completio o th transformatio t for carbo cluster bot withi th martensit an a th dislocation a th boundaries3.4.2Sinc transformatio fro austenit t martensit(1-4involve principa shea a wel a dilatationa stresse(volum increase a ne strai i create durin th transformation Thi ca b accommodate b man means suc ab slip twinning martensitedeformatio i retaine austenite formatio o packet o(99differen austenit variants an finall b varioulat arrangement withi a packet.Th minimizatio o th tota shea b individuamartensit unit ma b analyze schematicall a show iFig I a larg bloc o origina austenit transformint martensite ther wil b a larg strai create afte 6\ \ -^ • -I . J • • j L ! • XB 825-988Fig (a Analysi o th principa shea vecto Vi o a (110) stereographi projection 180 rotatio i Vi woulcreat n shea strai ove th numbe o laths e.g. betwee [Til an [111 orientations (b Shea proces o a bloc o origina austenite A int eithe a bloc omartensite M relate lath (case-2 (nretaine austenit a th t th martensitlath i differen orientatio a stack (case-3 witretaine austenit betwee them 6th transformatio a a resul o th shap change Thistrai ca b reduce i th origina bloc o austenit ibroke u int severa smalle unit upo transformatiowhic the assum differen crystallographi arrangementi th resultan structure Thi ma happe b a chang ith directio o th principa V confine t uni a show o a stereogTaphi projectio i Fig22-aTh simples situatio i whe th adjacen tw unitar twi related i.e. V i i V i i [IIIorientation The th ne shea ove th tw lath wil bzer a schematicall show i Fig 22-bno ver ofte thi situatio wa observe i th presen system Th twi relate lath wer wide an ha lo dislocatio densit an lacke th othemorphologica feature usuall see otherwis du t autotempering Thi ma mea tha the ma hav forme a thbeginnin o th transformatio whe ther wa a larg forc availabl an relativel les obstacle (suca boundaries dislocations wer on lat hanucleate an grown th stresse ahea o th boundar is larg i initiate th nucleatio o th nex lat i a wa tha th principa shea wil b i oppositdirectio (Fig 2 - 2n wa th ne shea wil zer ove th tw laths The ther i n retaineaustenit lef betwee th lath (e.g. Fig 9) 6Ther wil b mor nucleatio site availabl fo thmartensit crystal a th temperatur i decreased Whethes lath grow retaine austenit i lef betwee the(du t som stabilizatio mechanism involve a discussei ther i n direc effec o on lat tinitiat th nex i twi orientation Th stresse createb th formatio an th growt o th firs lat arpartl accommodate b thi austenit retaine behin thboundary Ther ar probabl smal residua stresse lef ma hel t initiat th formatio o th nex lathTh overal stresse created however b thos lathwithi th sam regio o a packe become s larg an ino totall accommodate b th retaine austenite There­fore abrup chang occur i th orientatio o th (aroun whic i commo t al neorientatio i suc tha th stresse create b thossevera lath ar decrease t som exten (Fig 22- an cas i lath i th ne orientatio an thaustenit ma assum eithe a ne orientatio relationshio a varian o th sam orientatio relationshi (e.g.K- o N-W)Th situatio whe th principa shea vecto i thfirs lat i iᄐ directio an i th nex on iM directio i desirabl a describe i Fig 22-a oThi i fac correspond t a twi varian o KurdjumovSach orientatio relationship Th earlie interpretatioabou th natur o th relativ orientatio o adjacenlath wa tha i a stac o lath iᄐ orientatioshea vecto i eac lat assume differen directios tha ove n numbe o lath th tota rotatio o thshea vecto aroun commo directio i completedTherefore th overal strai woul b zero Thi idealizesituatio rarel o neve happen i reality A th tiltinexperiment reveale tha th lath cluste aroun certaipole i a packe (a stacks an the hav sligh misorientatio betwee them (Th reaso fo thi misorientatiowil b explaine i a late paragraph.A simila configuration however occur betwee thpackets th syste overal tend t g t a loweenerg configuration th packet themselve arrang i suc wa t contribut t thi process I ha bee ascertaine b microdiffractio tha th packet ar actualldifferen Agai fro Fig 2 assuminsuccessiv packet hav {111} variants theth tota strai ove th whol austenit grai wil bminimizedTher ca b differen orientatio relationship betweeth individua martensit lath an th retaine austenitwit a {011}„//{111} commo boundarie i a packet Therar overal 1 variant o orientatio relationship possibl 6( fo K-S 3 fo N-W an 3 fo G-T withi a singl packewit a particula {111 habit Therefore ther ma b a numbe o situation possibl fo th relativ orientationo th adjacen lath wit respec t eac other Sivariant o K- orientatio relationshi witar show i Fig 23 Th possibl N-W K- an G-orientatio relationship give i Sectio 3.3.3. lea tth followin situations(i I ther i a 70 o 180 angl betwee twadjacen lath the the ar twi relate (e.g. th situa betwee 1-4 2- an 3- i Fig 23)(ii I ther i onl 10 angle the th adjacenlath ar variant o K- (o N-W (e.g. 1-5 3- an 2-6)Th sam i tru i th angl i 60 (1-2 1-3 4-5 4-62-3 5-6 o 50 (1-6 3-4 2-5 a show i Fig 23)(iii I th angl i 5 i eac direction the eaclat i i eithe K- o N- orientatio relatio (onvarian o each wit th austenite Thi i als tru ith angl i 55 o 65°(iv I ther i 2.5 angl the eac lat obey G-orientatio relationshi (tru als fo angle 57.5 52.57.5 62.5(v Differen combination o th abov case ma alsoccurMan o th abov case hav bee observe i thpresen allo system Diffractio analyse reveale tha 63 6 VARIANTS �!"M*Jk ' ini"M//ioiDoXBFig Analysi o six-variant o Kurdjumov-Sach O.Rwhe stereographi projection belongin t si variant ar superimpose ont eac othet revea th angula interrelationshi betwee them 6paralle lath o one tw o mor orientation i eacpacke ca b present Fo instanc i Fig 13 suc a situatio i show wher i th SA patter ther ar foumartensit zone an on austenit zon wit a commoᄀ Therefore ther ar fou differen orientation o martensit withi tha regio o a packet fiel analyse mad usin th reflection belongin t differen zon axe reveale contras change ithos lath belongin t th sam zon (a experimennecessar i suc comple configuration t distinguisbetwee differen shoul b als remindethat practice i i difficul t distinguis K-variant whic hav generate on N- (o G-T variant I necessar t hav a configuratio o th foi wher th electro bea i perpendicula t //commo direction I suc a case th cross-sectio o thpacke i th foi plan wil b paralle t th habi plan th lath an ther wil generall b onl on lat lyini tha crossection I addition a show i a exampl iFig tiltin th foi i th directio perpendicula tᄀ an takin microdiffractio informatio frobot austenit an martensit fo th correc identificatio th orientatio relationship i als necessary Theris alway discrepancie du t th lattic relaxation whic ma chang th loca orientation Th microdiffractio pattern take fro differen site withi a 6particula lat showe tha th misorientation scatte onlwithi a degree3.4.3 Formatio o LathsTh reaso wh s man orientatio relationship maexis withi a give packe ma b attribute t th increase variant availabl fo th lath durin th nucleation discusse i th previou section th systealway trie t minimiz th stresse an differen varianto th orientatio relationship partiall fulfil thi b decreasin th magnitud o th overalprincipa shea vector I ha bee ascertaine tha thlon direction o th lath ar i direction Havin(110)„//(111 habit then th austenit ha it M A directio paralle t th fil plane Bot o thesdirection ar close-packe direction i martensit an respectively Thereforeᄐ likel t bth eas growt directio i martensiteDurin cooling nucleatio probabl start a th straicenter (embryos i austenit (o a th boundaries jus th M temperature Th critica nucle then havin certai orientatio wit austenite star growin slowliᄀ fas i . Whe th growt i stoppedth austenit i stabilize ahea o th interfac becauso th larg strai buildup Adjacen lath nucleatb th ai o thi strai alread availabl nea 7® POSSIBL SEQUENC C 0 RACKET/ PACKET1Fig Possibl sequenc o growt o lath an subsequen formatxo o packet i a austenit grainXB 817-703 7th boundar an gro assumin a ne orientatio (o a varian o th austenit i the trappe betweethem Thi progresse unti al th packe i fille witparalle laths Simila proces als occur i othe packeto paralle lath whic ar themselve differen {111variant t finall fil th austenit grai (se Fig 2fo3.5 CONCLUSION(i Th microstructure o experimenta lo an mediucarbo lo allo steel consis o dislocated autotempere"lattr wit thi fil retaine austenit a thboundaries wher twi relate lath occur(ii Individua martensit crystal occu a laths0.1-0. width ^ vi heigh an 10-3 vi i length witstraigh paralle boundaries Stack o lath wit differen{111 transformatio variant for packets(iii Adjacen lath i a particula packe hav onᄀ i commo an ma b rotate abou thiaxi wit smal angle (0 0 10° o larg angle(� 20° lath iᄐ က ᄠ ᄰ orientation wit commo (110 ) Th angl orotatio ma eve b T i whic cas th lath ar twi Thi configuratio i attribute t th necessito strai minimizatio durin th successiv formatio(nucleation an growt o individua martensit crystals 7(iv Orientatio Relationship betwee austenit anmartensit al li betwee K- an N-W G- bein th mosfrequentl observed thi i perhap du t th greateavailabilit o variant durin nucleatio o th laths(v Th macroscopi habi plan ha bee identifie tb nea (111 (withi 5° whic i paralle t(vi Th lon direction o th lath hav beeidentifie t b paralle t , whic i paralle t an broa face paralle t (110 ,(vii I i hypothesize tha upo quenchin severalath nucleat a discret unit i differen part o thaustenite Thos whic nucleat firs establis th variant i.e. a particula b othe lath a the nucleat an gro on afte anothet for packets Retaine austenit i trappe a thboundaries relativ orientation betwee adjacenlaths orientatio relation betwee austenit anmartensite th packe orientations th structura dislocation bot i martensit an austenit an th rarmicrostructura twin i martensit ar al arrange tminimiz th overal strai (shea an dilatational createdurin th transformation 74THI FIL RETAINE AUSTENIT4.1 INTRODUCTIOPlat martensiti structure a encountere i higcarbo (o hig nitrogen)-hig allo steels whic usuallhav M temperature belo ambien hav bee studieextensively. ferrou martensitecontai retaine austenit a "blocky island fillin uth spac betwee th lenticula twinne martensiti plates(6,15-17.33,34 microstructurecrystallographi relationship " wit martensit ar documented Th existenc o thi phase whic amount(t mor tha 10 b volum i affecte b hea treatment ' 10,62,63,106-111 composition(9.10.30-33,81 stresstate.(107,115-118However previou investigator workin o thmorphologi an crystallographica aspect o lo allosteel wit o withou carbo di no identif retaineaustenit i th lat martensiti structure.80—87(On ma not tha i publishe paper o th studieinvolvin lat martensiti steel containin carbo on carecogniz thi film o retaine austenit i th micrographs. I wa no unti recentl tha retaine austenitha bee identifie a "thi films a th lat boundariei th mediu carbon-lo allo experimenta steel ' ' (simila compositio a th presen study havin M abov 7roo temperature Followin thi finding limite studieconcentrate o th effec o thi phas o th mechanicapropertie o thos steels, littlsystemati wor ha bee undertake t explai it14 45occurrence ' On o th objective o th presen researc wa texplai th existenc an behavio o austenit i thpresen serie o lo carbo steel i ligh o conceptdevelope t appl t hig carbo steels Th stabilizatio retaine austenit coul b accounte fo b partitionin alloyin elements especiall carbon durin th y t a a th M temperature o th presen alloyar wel abov roo temperatur {usuall abov 200°C . Sinc chemica partitionin i th M -M rang woul partlexplai th presenc o retaine austenit thi possibilitwa teste usin sophisticate technique t chec fopartitionin o alloyin element betwee martensit anaustenit an obtai a compositio profil acros thmicrostructure 74.2 IDENTIFICATIO AN TH OCCURRENC O RETAINEAUSTENITRetaine austenit i thes steel i i th for othi layer o film o th boundarie o th roughlparallelopipe martensit crystal (laths arrange ipackets ar usuall continuou an foun a packeboundarie an preaustenit grai boundarie athicknes o th film varie fro ver thi (^5 A t ver (40 A (bu usuall i th 150-20 A range dependino factor suc a th siz o th lath (large th laththicke th orientatio o th adjacen lathan th allo conten o th stee (increasin wit austenitstabilizer suc a M o Ni).Significan variation occu i th amoun o retaineaustenit withi a give specime eithe as-quenche o afte(89lo temperatur (e.g. 200°C aging I som regionther i n retaine austenit presen a th lat boundarieperhap becaus o th influenc o th adjacen lat orien(99tations example i Fig 9 retaine austenit wafoun i th packe B containin lath projecte edge-ohavin alternatel bright-dar contrast Thes lath ar"twi related an the ar sel accommodating I th samfigur i packet A an C surroundin th centra packe Bth lath ar no twi relate an indee detaile analysio thes packet confir th presenc o retaine austeniteI lo carbo alloys th lath ar mor regula i shap 7Fig Genera vie o th microstructur o 0. C containin allo (Allo 2 showin th paralle lat morphologwit retaine austenit (a BF an (b D images I locarbo alloy th lath hav mor regula shap an thickeretaine austenit film tha 0. C alloys 7an th parallelis o th boundarie i mor pronouncedTh retaine austenit film ar muc mor unifor throughou a particula packe (fo example se Fig 25)Retaine austenit film ar ofte ver heavil deformea ma b discerne fro th hig magnificatio micrographshow i Fig 2 whic wer take fro a regio containinexceptionall thic retaine austenit film (M5 A . Retaine austenit ha a dar contras i B imag wheoriente properl t a zon i als strai con arisin fro th hig densit o imperfectio du theav deformatio o austenit whic ha take place t th stresse create durin th y t amation i especiall noticeabl i D image (e.g.se Fig observe i retaine austenit filmgiv contras simila t dislocations stackin faul ribbonan ledges N defec analysi coul b don becaus o thlimite siz o th films However th dislocatio densit1 1 3 wa estimate t b aroun 1 -1 cm/c . Th tota amoun o retaine austenit i lo carbosteel i ver low i.e. 2- vol a determine b X-radiffractio technique ' Therefore considerablattentio i require t detec retaine austenit i th I i ofte difficul t identif retaine austeniti B image fro othe variou feature i th microstructur becaus o th lac o regula contrast Thereforecarefu electro diffractio an D analyse hav t b 7Pig Hig magnificatio micrograph o a regio containin thic retaine austenitfilm sho th contras chang du t defect fro heav deformation (a B an(b DF XB 823-232 7performe includin tiltin th foi t differen orientation t provid th bes contrastInterpretatio o diffractio pattern i als oftedifficul an sometime puzzlin sinc the usuall contaireflection fro severa feature i th microstructur suca martensit lath havin slightl differen orientationsfro autotempere carbides interna o multipl twin imartensite th retaine austenit itself an als frodoubl diffractio (forme b th elasticall diffracte fro th abov feature interactin t for extr Fig 27 severa o th mos commonlobserve diffractio pattern ar plotte schematically clarity tw patterns on fro a an th othefro y, ar indicated I reality i on use a selecteare aperture on ma obtai reflection fro man lath idifferen orientation withi th sam packet O courseal austenit wil b i th sam orientatio an contras lon a th observatio i containe withi th sampreaustenit grainTh mos commonl observe cas i a composit tripldiffractio patter cause b tw set o martensit lathsi.e. , an , an a thir patter du t thaustenite i.e. (Pig thi configuratio reflectio t b use fo th D analysi o retaine i (002 whic ha th larges angula separatio(^1 mRad fro th othe reflection i th pattern Onc 8o MARTENSITE • ALSTENSITE ••• i-. o * • w o� o '• - . ° . o . -'•' o,f- '• '• tP-csTt,'•* * * • » » • '' r * .» t& �• 0�a CIHICIOI)•COOpDQo S30 f 0 •-o .; o • ^% «• • • • m * •i » . . . * , .'o o i 6 i " diox"G .,J*-/f^.,(ri2)„ ^?. /**•T (fl3} li)Fig illustratio th commonlobserve superimpose diffractio pattern fro an y. XB 823-859 8th variou pattern ar identified thi reflectio ca bdistinguishe an shoul b use fo th D imagingI shoul als b note tha considerabl archin occuri th patter a a resul o th interna strain an mineorientatio difference o th laths ' Du t th shap(122factor, retaine austenit reflection ar usuallelongate i th direction Fo certai orientationo a particula regio (no exactl o th [110 zon axibu slightl tilted i.e. 1-2° toward streakinY ' coul b observe fro th austenit reflection i a directio perpendicula t th lon direction o th film (Fig28- an c)Th morpholog o retaine austenit i variou alloy(containin differen carbo an othe alloyin elements ishow i th B an P image i Fig 2 t revea thgenera appearanc o thi phas i th lat martensitistructure o thes alloys4.3 CHEMICA ANALYSI O RETAINE AUSTENIT4.3.1 Experimenta Techniques Sinc austenit coulb stabilize b carbo partitionin durin th transforma(45tio (chemica stabilization techniqu capabl oresol'.i.i C i retaine austenit unde th exceptionallsever condition o th presen problem mus b employe tmeasur th exten o carbo enrichment Th difficultieinvolve are 8Fig 28 (a Frequentl encountere diffractio pattern�•ail (b Du t th lon an thi shapo th austenit film "streaking occur i th SApatter iᄐ Th effec -i see whe th foi islightl tilte i th directio show b th arro i (c(her abou 1°) XB 824-350 8Fig Genera appearanc o retaine austenit idifferen allovs la) B an fb D (Allo 31 (a) B differen alloys (a B an (b D (Allo Ban (d D (Allo D (Allo 4) (f B an(g D (Allo 5) XB 823-232 8(i Th amoun o retaine austenit i lo (1-(ii Th retaine austenit i i th for o narrofilm (t A thick betwee th martensit lathsHenc th techniqu mus involv hig spatiaresolutions(iii Th C conten coul b n mor tha a fe at iretaine austenit an onl i thmatrixFo thes reasons lattic paramete measuremen bconventiona X-ra diffraction ' coul no b use a th o th diffractio peak (111 an (112 woulb extremel wea an broa an los i th backgroundAlso C i a ligh element direc technique suc a micro-analysi coul no b used sinc th wea C an K peak ar maske b th intens Brehmsstrahlun ath lowe en o th spectrumNonetheless sophisticate technique aravailabl t obtai a estimat o th C conten i locaregions Resolutio Lattic Fring Imagin (HRLFI Convergen Bea Electro Diffractio (CBED ' ca buse t measur th loca lattic paramete an relat i th carbo conten containe withi tha region Othemor direc technique ar Electro Energ Los Spectroscop(127(EELS an Fiel Io Microscop - Ato Prob Analysi(FIM-APA ' ' wher chemica compositio o a particularegio i determine b direc analysi o th chemica 8specie present Al fou o thes technique wer exploredbu th CBE an FIM-AP prove t b th mos useful Thresult obtaine b thes tw technique wil b discusse detai i th nex sections Followin immediatel i a brie discussio o th individua technique mentioneabovea Indirec Technique(i Hig Resolutio Lattic Imaging Measurement oth loca lattic parameter HRL suggestetha th carbo conten i austenit coul b betwee 0.41(451.04 i a 0.3 bul steel Thi resul wa deduce froth analysi o th fring spacing o austenit (usin thao martensit a reference b Lase Optica Diffractionmad o th negativ fil whic i take b simultaneouimagin o th lattic fringe fro (101 an (111 planesHowever hig uncertaintie exis i thi metho due e.g.t th amoun o defocussing loca thicknes variation(becaus o preferentia etchin o eithe martensit oaustenite) o strai i austenit phas highl th error an th uncertaintieinvolve i fring measurements(ii Convergen Bea Electro Diffraction Thi methodalthoug indirect give a accurac o 2/1000 i latticpari..iete measurement whe th experimenta requirement ar met Th CBE metho ha bee utilize successfulli curren investigatio (se sectio 4.3. 8b Technique(i Electro Energ Los Spectroscopy Direc quantitativ analysi o light elements suc a carbon ca bmad b thi method Th techniqu basicall involve thstud o th angula an energ distributio o hig energelectron whic hav suffere momentu change (energlosses durin th passag throug a "thin regio o a TEfoil resultin Electro Energ Los spectr i.e.transmitte signa intensit a a functio o th energ fo electron accepte throug th entranc sli contain informatio abou th typ an quantitie o th i th thi techniqu reveale it unde th sever condition i it applicatiot th existin problem Althoug considerabl effor wa o quantitativ analysi o th carbo conten i botmartensit an austenit phase wit th Berkele spectro(129meter, carbo (hydrocarbon contamination thic foieffect (multipl lo spatia an energresolution limite th carbo detectabilit t abou 2 wt%(130(ii Fiel Io Microscopy-Ato Prob Analysis FIM-APha prove t b a appropriat metho fo estimatin point poin carbo level i steel wit martensiti struc Thi i a uniqu metho whic ha higspatia resolutio (atomi resolution an mas independen(132-137sensitivity ' B thi method i wa possibl t 8analyz quantitativel th carbo an substitutiona alloyin elemen (suc a Cr Hn Ni Mo distributio bot imartensit an retaine austenit (se Sectio4.3.2 CONVERGIN BEA ELECTRO DIFFRACTION-AUSTENITLATTIC PARAMETE MEASUREMENa Experimenta BackgrounConvergin Bea Electio Diffractio (CBED patternar two-dimensiona map o th diffractio intensit a a functio o th inclinatio betwee th inciden electronan a particula crysta direction ' The arnormall compose o a serie o discs eac on correspondin t a differen Brag reflectio (Fig 30) Th intensitvariatio withi th disc carrie importan informatio th specime orientation thickness lattic parametean o th specimen ' Th CBE patter i obtaine a follows i th TEmode start b formin a imag o th specime undestandar conditio an the increas th firs condense t it maximu excitatio whil adjustin th seconcondense len t produc a focusse spo o eucentricalladjuste for a CBE pattern th objectivapertur i remove an th diffractio mod i selectedTh diffractio patter forme o th bac foca plan oth objectiv len i compose o disc o th 0 ordeLau zon an a circl compose o disc o th highe Lau zone (HOLZ dependin o th particula materia Fig CBE patter take dow t (111 zon afe crysta (her Ni showin th three-fol symmetr (a10 kV) Th pattern insid th disc forme b thdiffracte beam ar du t th HOL line an thi th backgroun i du t Kikuch lines S16 8use an zon axi chose (se Fig 3 fo sizi define b th angl o convergenc whic i directlproportiona t th diamete o th secon condense aperture th CBE patter i formed furthe adjustment mad t obtai a focusse pattern A 300-40 & diameteelectro prob siz ca b obtaine i th TE mod an les 10 & prob siz i possibl i th STE mod i a Phillip 40 microscopeRequirement fo obtainin a usefu CBE patter fro a particula material includ unifor thicknes an a straifre region I th presen case martensit ha a hig dislocatio density an austenit i heavil deformet accommodat th strain create durin martent;it growthHenc i wa quit difficul t obtai an CBE patter witshar HOL (Hig Orde Lau Zone line i it eve thougconsiderabl effor wa take t optimiz th microscop an parameters However a foi preparatio metho "dipping (se Sectio b use t preferentiall etc th martensit lath an produc austenitfilm extendin lik "fingers fro th martensit matri(Fig i Appendi 2) Retaine austenit polishe ithi wa i fre fro volum constraint an dislocationsChange i th lattic paramete o a crystallinmateria produc shift i th HOL lin position becausdeviation betwee th reciproca lattic point an thEwal spher ar altered Th smal angl approximatio 9T Bragg" Law fo a cubi materia o lattic paramete a isa Aewher A i th chang i th lin positio an A i thai th lattic parameter ' Accordin t th argumentpu forwar b Steeds accurat measurement (th limio accurac bein 2/10000 ca b mad o th change o thlattic paramete wit respec t a know standar wit thsam crysta structur an clos i lattic parameters Thshif i th positio o th HOL line wit lattic paramete i give bf = f[VPi- iT «rewher q represent th height o th triangle a thcenter o th disc forme b th HOL lines p th spacinbetwee th equivalen point i th centra an an o th20 disc (Fig 31) th radiu o th HOL rin(i.e. | I a furthe facto o 2 53 ""i introduce becaus thi radiu i equivalen t 2bTh experiment hav bee performe o a 0.17C/2.05Sstee (allo 3 whic ha exceptionall wid austenitstrip (se th inse i Fig A2-1 Appendi 2) Higpurit (99.99% N ( . = 3.523 A ha bee use a a reference resul obtaine wa cross-checke wit 99.999purit C (a = 3.614 fro thes wer mad 9Fig Pattern sho th detail insid th discforme b th forwar scattere bea i Ni retaineaustenite an C i (111 orientatio a lOOkV Not th descriptio o q an p use i Eqn (1)XB 816-573 9afte th annealin treatment. Th lattic paramete oretaine austenit whic wa calculate b thi metho i a numbe o experiment performe o differen retaineaustenit films ha bee foun t b abov C ca b foun fro th followin formul whicrelate th C conten t th lattic paramete o retaine. . (24,140-142austenitea 3 + x wt (2Usin thi formula a typica valu fo a averag C o conten i retaine austenit (e.g. fo a b calculate t b 1.1 ± 0.1 wt (o ±valu i fa abov th bul C concentratio (0.1 wt% oth alloy use fo thi study Th C conten foun i thiwa i a averag valu correspondin t a loca volum oretaine austenit whic i determine b th diamete oo th electro prob siz (fe hundre decreas an i th measurement du t hig voltaginstabilities th experiment wer complete withi onwee ( voltag instabilit o 5 volt i lOOk usualloccur i th Berkele Philip EM40 TEM) Not als iFig tha th retaine austenit CBE patter showexactl th sam symmetr a th C o Ni patterns i.e.trigona symmetr i (111 patter (se als Fig 30)Therefore i i conclude tha th lattic paramete o austenit i isotropic i.e. n chang i lattic 9fvlAxvofCaMSL.IEIttfLcFig CBE pattern take fro Ni retaine austenitean C a lOOk showin th highe orde Lau zones A.io th pattern ar simila t schemati patter (d) i.ethe hav 9paramete wit directio du probabl t th differenoctahedra site occupie b C whic coul occu i somSimila attempt wer mad t measur th lattic paramete o martensit an consequentl determin th C conteni thi phas (se th relatio a vs wt C i Appendi 2)However lowerin th C conten fro 0. t 0.0 wt tdecreas th dislocatio density changin th hea treatment produc martensit relativel fre o defects chang iacceleratin voltag t chang th radiu o th Ewald wer al unsuccessful Also differen camerlength wer used an smalle an brighte electro probewer trie (Philip 40 S wit FEG-AS wit d 10 A - aArizon Stat result wer unsuccessfu i HOL line i th CBP Ther i jus no enoug an strai fre volum t contribut t th formatio a patter suitabl fo lattic paramete measuremen (seAppendi 2 fo a representativ compare t fro retaine4.3.3 Measurement o Solut Partitionin intRetaine Austenit b Fiel Io Microscopy-Ato Analysia Experimenta BackgrounLin sensitivit o th FIM-A i independen os i ca b use fo analyse o bot substitutionaalloyin element (Cr Mn Ni etc. a wel a interstitia 9iojwroqiT ii rwimnim,n • y w . ^ ran 10c SC I IIPMil il -I' IjJJill3 3 4 4-U L ATOII W5 IWT6 1•rto"toi«tBH!I;S't?ctR1W|II1TIW',PJM CC M1 [i; t 4 s3 4 4ATmi was willl I I I I sFig Typica mas spectra obtaine b Ato ProbAnalyzer fro martensit (a an retaine austenit(b showin th characteristi (m/n peak i a 0.3allo a 200° tempere conditio (Allo 4)XB 8112-1273 9C a hig spatia resolution.th advantage o th technique typica mas spectr o thmartensit an th austenit phase fro a A Analysi arshow i an b fro allo No 4 Th spectr arplotte o a semilogarithmi scal t emphasiz th minopeak du t C Cr Mn an Mo Not tha th C spectru iparticularl comple wit peak a mass-to-charg ratio(m/n o 6 12 18 2 an 3 due respectively t th monomer an polycarbo ion C , C , o C , C , C an C,I shoul b note tha th presenc o molybdenu somewhacomplicate th evaluatio o th carbo concentration ThM +4 an + state hav m/ value aroun 2 an 1 whicoverla wit C an C*" peaks respectively possibl tha carbo an molybdenu coul b associate soli solution I thi wer th case the a C-M comple io coul possibl b produced th m/ valu o whici th triple charge stat woul b aroun 36 thu over wit C peak However i practice i i foun ' 144tha th M charg state whic woul giv ris t thesambiguitie occu extremel rarel (i fact non wa founi a analysi o 750 ion take fro th martensit phasese Tabl 2) I a C-M comple io i formed i ca onlb a ver smal extent sinc th nomina 3eve o M ca b i th + an + charg states Althoug nomina o C an M ar higher ther i n repor o C associatin wit thes alloyin element i solutio o 9a th surfac o th ti befor th fiel evaporationTherefore th C peak ar accepte t b correspondin tpur C specie an th analyse hav bee don accordingly(4 143However genera deconvolutio ' i alway necessarfo ever C pea a wel a fo othe alloyin elementsespeciall fo Mn C i usuall evaporate aan C (m/n 2 26. an 52 respectively an M idetecte a = shoul b note tha th+ +Mn,., lie betwee th Fe pea an th muc largeFe- Henc i i difficul t recogniz th M peai a mas spectr (se Fig 33) Therefore th distributioo M require specia attentio i th deconvolutio processbTh result o quantitativ analyse o C level i thmartensit matri an austenit film i a numbe o 0. wtC-specimen i th as-quenche an lo temperatur temperecondition ar give i Tabl 2 (not tha th result argive i atomi percent 0. wt C = 1. at C)I th 2M stee i th as-quenche condition th averag C conten i th martensit matri i ver low--les thaone-fourt o th bul concentratio at = 0.0wt%) tha o retaine austenit i mor tha twic thorigina concentratio o th allo at = 0. wt%On ver significan observatio i tha th C leve a tha'/y-interfac i mor tha 1 at� ( th a 200° fo 1 hou reduce th averag C 9Tabl 2 Th Analysi o C Concentration i Martensite, Retaine Austenit an aAlloNo1 Conditio 1 Phas | At# o Ion1 Analyze4 AQuenche0.2 ± 0.00.3 ± 0.00.2 ± 0.10.1 ± 0.1 ± 0.0nea a carbide3001801400.2 ± 0.2 ± 0.10.2 ± 0.1707050Averag0.2 ± 0.01310Ret- : 2.8 ± 0.660a'/Y-interfac11 1a' (ave.0.1 ± 0.0630Ret-y0.8 ± 0.60200° a : Tempere . 0.1 ± 0.02.2 ± 0.2a"/y-interface6-1a : 0.0 ± 0.20.1 ± 0.00.0 ± 0.00.1 ± 0.0161749700Ret- : 2.2 + 0.22.2 ± 0.3 ± 0.3 + 0.4343160150110A a : Quenche0.1 ± 0.0 + 0.00.2 ± 0.10.4 ± 0.20.2 ± 0.124028035010090AverageRet- : a0.2 i3.4 i 0.85 9 9 2820 9concentratio i martensit considerably whil i thretaine austenit i staye nearl th sam o decreaseinsignificantly Ther appear t b littl o n chang ith C conten a th a'/Y Th depletio o C i i martensit ma b du t C segregatio t dislocation an th formatio o cementite It shoul b note th C concentratio a th a'/Y-interfac i highe isom regions e.g. aroun ledge o dislocations givinris t correspondin hig peak i th C profil acros thinterface O th othe hand durin tempering diffusiot th interfac regio durin tempering occur an th o th C conten a th interfac become unifor dut th hig mobilit o C a thes sites Als give iTabl 2 i th analysi o C i 2N allo i th A.Q condition ca b discerne i thi table ther i agai a significan enrichmen o C i retaine austenit (averag at% an depletio o C i th martensit matri(^0.2 hig degre o C build-u a th a'/Y-interfac i agai obviou (^1 at%)Th C distributio i th martensit matri i als nouniform Ato Prob analysi showe tha thi i tru isample analyze bot i A.Q an 200 tempere conditionsA analysi o th martensit i show i Fig 34 Peak ith C profil wer frequentl observe durin sequentiaanalyse fro severa A' thic C-ric region (size othes area ca b foun b bac calculatio fro th numbe 10o ion region ar no identifiabl iFI image becaus the ar diffus an d no stan oufro th matrix However som peak wit muc highe C level (2. at% wer als observe i th specimen agaii th A.Q an especiall i th 200 tempere conditionsThes region contai fin carbide whic ca b discernei th correspondin FI image becaus o thei siz andar contrast Earlie investigatio b TE reporte tha i WBD imag o th dislocation i martensite smalirregularl disperse region attache t th dislocationsThes ma b C cluster forma a th dislocations e-carbide i bot o th alloy (2M an 2Ni wer als observei th A.Q conditio (Fig . carbide latetransfor t muc large cementit upo 200° tempering e-carbide wer stil observe i foil containin2 wt Ni eve a thi condition Therefore thi i a goocorrelatio betwee th FIM-A result observe her an thearlie TE observation o th martensit matrix ' Mor detaile observatio o th C distributio througth microstructur wa mad possibl b sequentia analysi bot martensit an retaine austenit regionsFig 3 an 3 sho C profile revealin segregatio a tha'/Y~i.Th cumulativ tota numbe o ion caught whic iplotte a th absciss o th graphs ma b take a a o distanc evaporate throug th specimen Fo 103Mf'-f•'$:* P Pfl-O100 200 300 fOO 500 600 700 BOO 900 3Total #of Ions Fig C profil acros th martensit matrix Not thpeak whic ma correspon t C cluster a th defect i matri !A sequentia analysis—200° tempere (Allo 4))Fig (a FI imag an (b schemati illustratio oth tip Retaine austenit i reveale a a dar circulaband (c C concentratio profil acros a an y. 10example martensit wit ti orientatioᄐ an a b 2 A-dia prob hole 10 ion caugh woul corresponroughl t a 8 A-dee laye o materia fro th ti o thspecimen Considerin th chang i th orientatio o thti (an als th crysta th ti radiu ocurvature thi figur ma b change t 5 t 8 A'sEve wit ver hig coolin rateᄐ (1stil appear t b considerabl partitionin o C iretaine austenit (Fig 34) Th sampl use i th analysiwa wir o compositio 3 wt Ni-0. wt Cr0.3 wt C (1.6 FI imag whic i take durinth evaporatio sequenc an it schemati drawin ar als i Fig 35- an b depictin th favorabl configuratio o th retaine austenit fil wit respec t th shoul b note tha th C leve iretaine austenit i no uniform an showe a averagconcentratio o 4 at% Wherea th averag concentratio imartensit appear t b muc les tha th bul value hig concentration o C a th a'/y-interface (8-10% wer observe eve fo sever quenchin conditionsI shoul b note her tha th pea a th interfac i nosharp rathe broa a i depend o th configuratio oretaine austenit fil whic ma distor th tru shap oth C profile (Thi importan poin wil b iterate latei thi section. A numbe o profile obtaine b sequentiaanalysi i bot 2M an 2N alloy unde thes condition 10produce similarly-shape C profiles O th average mortha 1 at C ha bee observe a th interfac (seTabl 2)Temperin th alloy a 200° fo on hou di nochang th shap o th C profil appreciably Th C levei y ha retaine abov 2 at i thi unusuall thic (50 Afil (Fig 36) However a decreas i th matri C conten apparen a wa als discusse before Thi i probabldu t C distributio int cluster a th dislocation ancarbide i th martensite A smal increas i th C conten a th a'/y-interface i als expecte du t th mobilit a. C I fact i som case localize C conten a th interfac rise abov 2 at whic migh bth indicatio tha carbide ar jus startin t nucleata thes sites Analyse o substitutiona alloyin element(Cr Ni Si Mo reveale n apparen chang i th leveo th alloyin element neithe i th A.Q no i th 200° condition Fig 3 show C an M profile i 2Mstee i th 200° tempere condition Excep fo insignifican localize variations ther i n deviatio i thaverag valu o th C an M fro th bul concentrationI i relevan t not her tha attempt t uetermin thconcentratio o alloyin element i martensit an retaine b X-ra microanalysi (EDX i th STE wa no J20-R |/6G-3/0,«^UVVf*?'2 J 4 5 7bfa # o Ions • Fig Substitutiona (Cr Mn an interstitia (C alloyin elemenpartitionin i martensit an austenit i Allo 4 Althoug n chang ith compositio o C an M level occur i th microstructure thC-leve i hig� ( at% i thi thic retaine austenit fil (MO A anrise t nea saturatio value a th interfac (sequentia analysis200° tempere condition 10Som retaine austenit film observe i th FI werver thi ( A an i thes case th C conten appearet b mor uniform I th exampl show i Fig 37 wherth A analysi wa don b selecte area ther wer npeak a th aVy-interfaces bu a singl pea correspondin t th fil itself I shoul b note her tharetaine austenit film (o bands ar distinguishe froan othe featur i th martensit matri {e.g. especiallcarbides b thei continuou dark-ban appearanc i conjunctio wit th chemica data Ther i n wa odistinguishin retaine austenit fro th crystallographo th band sinc i i usuall no wid enoug t recog an crystallographi poles However i i easil i a favorabl configuratio a show i35 an 3 whic wil b discusse i th nex paragraphTh FI image Fig 37-a show a upsid dow Y shape austenit fil situate betwee martensit laths configuratio i frequentl observe i TE micrographs i th optica microscop imag o th ti whicshow th genera appearanc o th ti a lo magnifications image a i schematicall draw i b wher thcircle represen th prob hol o th region selecte fo analysis Th number insid th circle ar th averagC content (i at% a thes sites Th sequenc o thstippin proces i schematicall show i (d wher eac i block correspond t th one i th circle i 10c m 2BEA-4ft-*/ijXci^•.#"^fc-*.'# ---"' %/"'''-"T*x\1 '•'*x\1-.•\'I1s'.'',''',-4I;N 1 Fig (a PI imag o th ti showin ver thi retaineaustenit (^5 K). (b Schemati illustratio o th imagi (a) circula region correspon t th prob hol an thnumber indicat at C foun a thos regions (c Opticamicrograp o th tip (d Th dat chang thevaporatio sequenc (th boxe correspon t are(e an (f ar TEM-B an D image o th tip (g Selecteare analysi acros th fil (lin AA i (b) (AQ-Allo 4)XB 810-1149 10(b) C profil draw fa acros th retaine austenitfil i show i (g wher ther appear t b a singlpeak I shoul b note tha th prob hol use i thiparticula exampl i ^3 & dia Henc th singl pea mab a resul o th siz o th effectiv apertur an thconfiguratio o th retaine austenit fil wit respec tth ti a wel a th extremel thi widt o th retaine fil whic i reveale b HVE micrograph (e an(f wher y appear a a brigh ban (i DF i thi sid take a ther appear tb abou 7 at C a th retaine austenit regionI shoul b emphasize agai tha th shap o th C profil largel depend o th configuratio o th retaineaustenit fil wit respec t th specime tip Ther arthre possibl suc configurations(i retaine austenit fil perpendicula t thspecime(ii retaine austenit fil i obliqu orientation(iii retaine austenit fil paralle t th specimeaxisThes configuration an thei correspondin FI image cab show a i Fig 38 Th mos favorabl conditio icertainl conditio I a th prob hol i simpl place oth specime ti an a sequentia analysi carrie ou tstri atom awa fro th ti laye b layer Sinc ther n compositiona variatio withi th are covere b th 10SIDE TOP Fig Possibl configuratio o retaine austenit filwit respec t th tip (a Sidevie (e.g. TE(b Projecte imag (e.g. FI Sequentiaanalysis Selecte are analysis I th obliquorientatio i Case-Il sequentia analysi i preferre(refe t Fig 40) 10effectiv apertur o th specime plane th spatiaresolutio o th analysi i basicall th distanc betweeth atomi layer austenite o martensite(dept configuratio give a analysiclos t th actua valu an i essentia fo analysi ointerface wher th variation i th chemistr d noexten beyon severa atomi planes Thi configuratio waalway favore durin th analysi i th curren investiga th analysi i Fig 37 Th FI imag 37- inearl i favorabl orientation Unde thes circumstances analysi o th phase (a an y an a'/ interfac arno subjec t interferenc fro eac othe a th prob alway lie entirel ove on o th phase o interfacan doe no pic u dat fro th adjacen regionsA prolonge searc fo suitabl region require thath specime tip b evaporate fo lon time (whicquentl ende u bluntin o flashin th tips o manspecimen exchange fo a favorabl retaine austenitconfiguration—a extremel time-consumin process Nonetheless configuration wer achieve afte lon trialsAnothe exampl i show i Fig 39- wher th plan o thretaine austenit fil i nearl bu no exactl perpendicula t th ti configuratio cause th retaineaustenit fil t swee obliquel acros a th fielevaporatio continue (Fig 39-b (compar Fig 39- an b wit I II an III) nFig FI micrograph sho th obliqu configuratio oretaine austenit i martensit matrix Not th regula i th image fro (a t (b whic correspond tth abov geometr reveale durin th evaporatio sequenci A sequentia analysi (refe t FigXB 810-1142 IlTh secon configuratio (Fig 38-11 i th moscommonl encountere cas wher th retaine austenit fili obliquel oriente a th specime tip Durin thanalysi o th interfac th phase o bot side o thinterfac an y) a wel a th a'/y-interface regio itsel contribut t th analysi i th ato probebecaus th effectiv apertur cover bot th matri anth interface Thi result i a lowe apparen concentratio tha actua an cause th broadenin o th pea ith concentratio profil a th interfac regio i a sequentia analysi (Fig 40) Not als that sinc thinterfac i approache obliquely th numbe o ion caughdoe no represen th tru distanc fro th interfaceTh thir configuratio i th mos favorabl foselecte are analysi (e.g. a i Fig 37 as th specimei move i th plan perpendicula t th ti axi andifferen region o th surfac chose fo analysiinterface martensit o austenit thicause problem i th determinatio o th concentratio a interface Becaus o th finit dimensio o th probhol (20-3 & dia effectiv aperture an o th natur oth io countin statistics th spatia resolutio (latera resolution deteriorates Therefore th actua resolutio i determine mainl b th siz o theffectiv apertur an th topograph o th ti a thinterfac region I thi case on ha t tak a averag 1112 3 4 5 T 1 1 1 rFig Schemati illustration o th chang i th tigeometr an th correspondin concentratio profile (e.g.fo Retaine austenit i perpendicula t th lon o th tip Thi i th mos desirabl configuratioan th concentratio profil represent th tru concentratio chang wit distanc D perpendicula t th a'/interface (b Retaine austenit i i obliqu orientatiowit respec t th lon axi o th tip Th concentratioa an nea th a'/ boundar i no th tru value bu aaverag valu whic depend o th angl&#xJ000; Henc thdistanc D i no th tru distanc bu a projectio o D 11o th ion counte withi th regio covere b thperipher o th aperture achievin onl a averag valuwithin 2 A area Therefore th measure concentratio o th plane o th interfac become muc lowe thath actua valu lowerin th pea heigh an give ris ta broade peak I th extrem cas show i Fig 37 anespeciall i th retaine austenit fil i ver thi(40-5 A—littl thicke tha th detail ith profile i.e pea a th interfac woul disappeaan onl on peak correspondin t th averag C conten oth interfac an retaine austenit film ma appear I particula example C atom ma hav bee uniformldistribute withi bot th interfac region an thretaine austenit becaus o it thinness4.4 DISCUSSION - MECHANISM O AUSTENIT STABILIZATIORetaine austenit i steel i normall associatewit th plat martensiti structure ' ' tha fori hig carbo steels I suc steel th microstructurconsist o "lenticular martensit crystal whic crisscros th prio austenit grain alon th fixe crystallographi orientation wit austenit remainin i "blocksbetwee th plates Numerou factor ar know t contri t th presenc o retaine austenit i thesstructure whic ca amoun t 5 t 20 b volumeFo example i ha bee establishe thatcompressiv stres wil decreas th M temperatur 11(especiall hydrostati pressure wherea a tensil streswil increas it Plasti deformatio o a fe percenusuall rise th martensit transformatio star temperatur transformation bularge amount o deformatio inhibit th transformation(15,106,108,117,147-149 ^ ^forme decreases ' ' Th temperatur o deformatio an th tim o holdin als affec th amoun oretaine austenit significantl a bot changin th microstructur o th microscopica leve (allowin C an N tdiffus int dislocations subboundarie ansmal coheren precipitate whic inhibi th transformatio, , _. .,.(15,107,147-149 .dislocatio movement an allowin lonrang chemica change t tak plac (partitionin o solutelements. ' ' Thi i importan wit respec tth favorabl change i th fre energ o th austenitwhic a thi "ne state (a lowe temperatures hadifferen chemica compositio tha i ha a hig tempera thi effec i ver sluggis bearin th tha th M temperatur i plat martensiti steel arver low Th rat o coolin als affect a shif i thM temperature th highe th coolin rat th higheth M wil be Th exten o thi chang depend o thallo compositio an previou austenitizin temperatur(an effec i als significan a i ha beeobserve tha whil almos complet transformatio occur i 11bul samples onl partia transformatio occur (wit mucaustenit retained i smal spherica particle (e.g.50 A dia. sain Ntransformation occu a al i smalle particle an theremai a austeniteA discusse a lengt i th precedin section othi thesis i th cas o lat martensite th shape sizedistribution amount an chemica constituent o th "thifilm retaine austenit ar differen fro tha i platmartensiti structures Th mechanism responsibl fo thexistenc o retaine austenit i lat ma::tensite wil bdiscusse i th ligh o th experimenta evidence i thfollowin sections4.4.1 Chemica StabilizationTh initiatio an progres o th y t a' transformatio ar bot controlle b th chemica an nonchemicaenergie o th system ' Th chemica fre energdepend largel o th chemica specie o th phases Jan variatio i th chemistr o th phase occur througdiffusio durin coolin the th cours o th transformatio i altered Thu partitionin o alloyin element(substitutional suc a Cr Ni Mn o interstitial^ suca C an N betwee th phase durin th transformatio wil th drivin forc require t continu th trans Thi decrease th M t lowevalue an henc th austenit ma b retaine a a 11metastabl phase Thi phenomeno i calle "chemicastabilization.I ha bee foun i thi stud tha althoug ther in noticeabl chang i th substitutiona allo conten oeithe austenit o martensite a appreciabl increas ith C conten o th austenit an a decreas i th martensit wer recorde (Figs 34 3 an 36) Fro thmetastabl extensio o th equilibriu diagra show iFig i ca b see tha th austenit ca accommodat 4 wt carbo a roo temperature I th currenstudy th analyse performe b A an CBE technique havreveale tha th carbo conten o th retaine austeniti mor tha twic th nomina conten i th bul samplesMos o th A dat indicate tha th C i y i aroun 2.at woul resul i a decreas i M temperatur oabou 150° (Fig 41) I shoul als b note th^ th Mtemperatur correspondin t th abov C leve i austeniti eve lowerTh amoun o carbo enrichmen i th retaineaustenit i determine b th exten o carbo redistributio b diffusio durin quenching Man critica variable a th coolin rat o th sampl (Appendi5 th diffusio coefficien o carbo i a an y (Appendi M (an M temperatur o th bul alloy.an th growt rat (especiall sidewis growth o martensit lath (Appendi 3) 11M— [Calculatedfor Ms0.0 0.2 04 0.6 0.8 1.0 1.2 1-4 1.6 IS 2.0 2.2 Wt.Fig Th M M vs C conten i steels (Not Ref1 i no Ref 3 i th text. Not tha M o retaineaustenit drop t muc lowe temperature whe C increasesXB 8112-1309 11Th coolin rat depend o th quenchin mediu ansiz an shap o th sample Coolin curve determine fothi plat sample ar plotte i Fig 41- fo oi quenching shoul b note tha th sample use fo thspectroscopi analysi wer muc smalle i size.A presente i Appendi 5 analyse wer mad odeterminatio o th effectiv diffusio distanc fodifferen shape sample coole i th differen mediumuse i thi investigation Tabl I i Appendi 5 show fo fas coolin rates i i possibl t achievth experimentall foun averag concentratio i y. Th critica tim availabl fo th carbo redistributio durin coolin depend largel o th diffusivit ocarbo i a an y. Calculatio o diffusio coefficientoto carbo i bot phases i.e. D an D i discusse iAppendi 4 A averag valu fo th diffusio coefficieno carbo i ferrit ( ) ha bee calculate fro th dati th literature equate t D Th diffusio coefficien i C an y a lo temperatur i thextrapolatio o th averag valu foun fro hig temperatur data " Th mos representativ value ar afollow (activatio energ i give iD 22^0 (1D = 0.01 exp( iSgSfi ,2 11Thes value ar use t calculat th effectiv diffusiodistances ca b travelleb a carbo ato i martensit durin coolin ( . valuear tabulate i Tabl I i Appendi 3) I orde fo thi t affec austenit stabilization carboatom mus b abl t mov t th a'/ interfac an o intth YTher i n quantitativ analysi mad o th determinatio o th growt rat o lat martensit i a systesimila t th presen case Som investigatorworke wit th bainit transformatio wher th bainitiferrit crystal ar simila i shap an siz t latmartensit crystals Tw separat growt rate werdetermined lengthwis an widewid growths Growt i th4 ° lon directio o th crystal i ver rapi (1 A/secwhil thickenin o th crystal i slo (averag rat ii marte(157,1581 A/sec an occur a a deceleratin rate Lat martensite gro lengthwis ver fas an sidewis slowl(a reveale i Ref 33 althoug n quantitativ determinatio wa mad o th growt rates A th lat slowlthicken considerabl carbo redistributio ca tak placeI fact a analysi ha bee mad o th estimatio o thgrowt rat o lath fro th know experimenta dat(Appendi 3) Th valu foun give a sidewis growt rat4 5 ° o abou d = 1 -1 A/sec Fro estimatio wamad o th availabl tim fo carbo t redistribut itsel 12i th martensite i.e. th effectiv diffusio distancei th M -M range a give i Appendi 5I i see fro th foregoin discussio tha durinth deceleratin growt o th lath durin coolin i thM -M range carbo atom redistribut themselve i a' an toward y wher the ar soluble However Dthi temperatur rang i no hig enoug fo carbo tdiffus i y throug th whol fil thickness Excep wher retaine austenit thicknes i ver smal (e.g.,40-5 A th concentratio acros th interfac is o thaverage th sam leve (refer e.g. t Fig 37)However i th mos commonl observe case wher th austenit i wide� {10 A carbo diffusin fro accumulate a th semicoheren a'/ interfac t verhig value (1 at o higher e.g. se Fig 3 an 36whic ha othe consequence i th stabilizatio proces i th nex section Consequently th overalconcentratio profil ma b represente schematicall ashow i Fig 4 wher th C leve i martensit gradualldecrease (wit shar peak correspondin t carbo clustera th dislocations toward th interface risin t higconcentration a th a'/ interfac an dropin i y wher averag concentratio i stil twic (o higher thatha o th nomina concentrations Althoug thi carboenrichmen wil provid som chemica stabilizatio i canno accoun fo th existenc o retaine austenit becaus M temperatur i no lo enoug (se Fig 41) Ther 30D\ .t /5 750 * -« dCA)Fig (a Th coolin curv fo oi quenche Allo 4 sample (1m(b Schemati diagra showin th approximat carbo profil acros thmartensit austenit microduple structur i th a quenche sample Sharpeak i th profil i martensit ar du t carbo clusterin a earlstag o carbid formationt 12ar othe effect whic als contribut t th tota stabilizatio a wil b discusse i th nex sections4.4.2 Therma StabilizationTh characteristi shap chang o th martensititransformatio implie continuit acros th interfac which[2 177consequently mus b substantiall coherent ' However complet coherenc ca exis onl unde th circumstanc tha th deformatio relatin an pai o uni celli the tw lattice an be i steels contain onprincipa strai whic i effectivel zero Sinc thiconditio i no me betwee austenit an martensit phasei steels th mismatc i correcte b periodi discontinuitie (e.g. whic ensur strain d no accumulat ove larg distances Thinterfac thu contain region o misfi a a networ odislocation whic becom mobil unde th applicatio( 2o suitabl chemica (o mechanical drivin force thes interfac dislocation ca b immobilize(36,105-107,147-149 interstitia solut ato segrega(178tio whic for "Cottrel atmospheres t slo dow oeve caus complet ceassatio o martensit crysta growthTherefore austenit trappe betwee tw suc interface iretaine a a metastabl phase thi phenomeno i calle"Therma stabilizatio o austenite.I plat martensiti steel th amoun o "blockyaustenit retaine ca b controlle b holdin th sampl 12a a temperatur abov o belo th H .Th amoun o retaine austenit increase a th temperatur an tim o hol i extended Thi i mainl t carbo segregatio t th dislocation a th t for "groupings whic oppos advancemen o interfac resultin i therma stabilization Similaphenomen ma b attribute t th cas o lat martensite thi fil retaine austenite Whe a latnucleates interfac advance quickl lengthwis anmove slowl sidewis b th movemen o dislocation ' ( 177(and/o ledges ' a th coolin i continued Th nebe lattic i straine b th trappe C whic diffusetoward th interfac an int th y t th exten tha tim temperatur permit Th carbo atom diffusin int th segregat a th dislocation presen a thes ther i dislocation-interstitia carbo interactio whic produce a ne forc whic oppose th interfacmovemen (se Appendi 6) Th stabilizatio phenomen mab explaine i detai a followsTh interactio o th elasti stres field arouninterstitia solut atom wit tha o semicoheren inter decrease th elasti strai energ whic furnishe a drivin forc fo th segregatio o thes solut atom tth interface' (simila t "strai aging hig temperature an lo dislocatio veloci velocit o interfac propagatio a i 12tisotherma martensit o i th presen cas o latmartensit (se Appendi 3) carbo ca diffus ove londistance unde th influenc o th strai fiel o thinterface Th degre o stabilizatio i dependen upoth concentratio o th anchorin atom a th interface stabilizatio i determine b th diffusiorat o carbo a a temperature whic changes wit a decreasin rate durin cooling Th bindin o C t th increase wit decreasin temperatur ' an s th effec o therma stabilizatio wil als increaswit decreasin temperature Durin th lat growth thicorrespond t th tim toward th en o th transformatio nea M_ maximu effec ma no b obtainea' becaus o decrease D a lo temperature consequentllimitin th C diffusionI ha bee foun i thi stud tha th carbo contena th interfac ma attai value (5-1 at%hig enoug t sto th interface A approximat analysii Appendi 6 determine th concentratio neede a th t preven it furthe advancement I shoul bnote tha th concentratio increas mainl occur durin whil th temperatur i stil hig enoug fo diffusion However th occurrenc o mor segrega o carbo a roo temperatur i unavoidable tha causin "Thermal-stabilization, furthe 12increase th "stability o retaine austenit a ambieno lowe temperature (eve a liqui nitroge4.4.3 Mechanica StabilizationStabilizatio o austenit b mechanica mean ha twcomponents plasti deformatio o y slow th progreso th y t a transformation (ii th residua stresselef withi th sampl affect th transformationDurin transformatio t martensite th parenaustenit lattic i highl deforme ahea o th transformatio fron an contain hig densit o imperfection suca dislocations faults ledges, etc (refe t an43) situatio result fro th accommodatio ostrain create durin th shea stres an dilatationa(3,4,115-117 I i mordifficul 1, transfor a distorte lattic t anothelattic b a shea transformation Henc th progreso th transformatio i controlle partl b th resistancoffere b th matri t th growt o th ne crystals(15,108,115,147,148 Q individualath o martensit ca b viewe a takin plac b thsystemati motio o th interfac consistin o array odislocation movemen o whic wil b impede b th o obstacles suc a othe dislocation i th. , (15,118,148,159 „ , _ , , paren phase Whe th plasti deformatiotake plac i y durin th formatio o a' mor disloca wil b generate an continuall increasin drivin 12Fig Hig magnificatio B (a an D (b micrographshow th contras du t hig dislocatio densit bot iaustenit an martensit (Allo 5) Othe defects suca "ledges ar als apparen acros th filmsXB 817-703 12forc mus b provide befor mor martensit coul formTherefore a th transformatio proceed i becomegressivel mor difficul t transfor th remainin y somo whic i ultimatel retaine a a thi metastabl filbetwee martensit crystalsI additio t thi direc obstructio t th movemeno th coheren interface ther i als a effec o carboredistributio i thi deforme austenite Thi happen a temperature withi th M -M rang wher interstitia segregate t th dislocation i austenit areveale b Ato Prob analysis Simila t deliberatelapplie "ausforming proces thi effectivelincrease th strengt o austenit whic i tur increaseth "mechanica stability. spontaneou o a wil b oppose a i mus baccompanie wit mor deformationSample quenche t ambien temperature fro thaustenitizin temperatur contai a hig degre o residua th cas o th austenit t martensit stresse o th orde o 5 ks resultin quenchin correspon t abou 10 o th stressecreate b th volum chang durin th fe t be transformation residua strain whic mus b accommodatema als effec th loca stabilizatio o austenite ' 116,147 124.4.4 SummaryTh overal stabilizatio o austenit i probabl dut som combinatio o al th stabilizatio mechanismcusse above I term o mechanica stabilizatio thinterna stresse create b th volum chang produc aenerg barrie whic mus b overcome i.e. M reducelocall b abou 5°C/ksi Thi i th mos importan effecan alway occur irrespectiv o rat o coolin o change th chemica species Th measure residua strai correspond t onl 5-10 o th tota expecte b th latticexpansion However th stres effec shoul b les i lo carbo alloy becaus AV a M i smallerTh reduce plasti deformatio i accommodate mor readil muc Jes Y i preserve tha i hig carbon-hig allostealsChemica stabilizatio i mor importan i lo carbosteel becaus th K i highe an mor diffusio ca occudurin th transformation Henc C buildu cause th presen a thi films t persis eve durinquenchin t liqui nitroge temperaturesTher i als a therma effec wher carbo segregatet a'/ interface durin growt pinnin th transformatio s tha th interfac i immobilized Interfacmotio i als oppose b th austenit deformatio whiceventuall lead t complet suppression Th remainin verhigM deforme austenit i retaine i th for o thifilm an trappe betwee th martensit crystals 124.5 MECHANISM O DECOMPOSITIO O RETAINE AUSTENIT4.5.1 Mechanica Instabilit (Effec o Deformatioan VolumTh effect o deliberatel induce plasti deformatioo th behavio o thi fil retaine austenit wer noinclude i thi researc program However som interestinstructura feature wer observe (probabl induce durinth foi preparation whic ar relate t mechanicainstabilit o th fe phaseI som instances instea o retaine austenitemartensit strip wer observe a th lat boundaries Aexampl (fro a i illustrate i Fig 4 whic a larg are containin smal twinne region idifferen orientation betwee martensit lath o eitheside o th SA patter (44-c reveale tha thtwin for o {112 plane o martensite Th D micrographtake usin a twi spot reveal th contras chang t th twi variant contributin t thiparticula reflection Measurement i th region containin edge-o twin showe tha the ar onl 3 A i thicknes(122whic give ris t lon streak (o spikes extendinfro th twi reciproca lattic points Not tha i thSA patter (Fig zon axi i ver nea t (111bcc reflections forme b th intersectio o th spike wit th Ewal sphere ar arrange suc tha 130_ ,w:n sc,Aei'-i �"* **'5a .'•O^-»i' II•"l/ d 3.It f ' ai'3.QiiX,.Fig (a B image— wid regio exhibitin twin ith for o a stri betwee th martensit laths (b Di fro th twi reflectio nea th matri(lll) SA patter fro th are covere b th matri anth twins (d Th analysi o th SA patter (Allo 1)XB 810-H49 13the li alon eithe 1112 o [321 reciproca latticdirections th twin ar o (112 o (211 planesSimila twin wer observe i a sampl containin0.1 wt C I th inse i Fig 4 th dar contrasregion ar expecte t b retaine austenit film extendin int th perforatio o th som o thesquit thic strip correspon t retaine austenit a confirme b CBE (refe t Sectio 4.3. o Appendi 3)However other suc a th on show i Fig 4 ar composeo smal twinne regions A hig magnificatio D micrograp i Fig 45- take b usin a twi reflectio (Fig reveal th contras chang correspondin t th twi crystals Bot th siz o th twinneregio ('v pm an th twin (5 A width ar large tha withi th bul o th sampl (Fig 44) Th SApatter (c show tw pattern correspondin t th matri an th twin whic ar forme o (112plane o martensit (FigTransformatio twin ar th commo feature i platmartensiti structures ' ' ' ' Accordin t thphenomenologica theor o th martensiti transformationtwinnin i essentia fo lattic invarian shea t produca lattic invarian plan fll2havin th habi whic originate fro th {110 plane oaustenite ar mos ofte observe althoug {110 twin(3 188hav als bee reported ' 13O O �•£"!Si .p* " m*. (a Twin forme i a pre-austenit stri i 017Figsteel lo magnificatio imag o th microstructur neath perforatio show i th inse i (a reveal dar stripfreel extendin int th hole Som o thes strip (likA ar retaine austenit (prove b CBED discusse iSectio som other (lik B ar transformeregion exhibitin twinne structur (BF) (b D micrograp(fro a twi reflection showin twi contras whic extendealon th strip (c Microdiffractio patter showin tharrangement o th reflection fro th (110 matrix (112 an doubl diffractio forme b th dinteractio o thtwi an matri reflections 13Observations simila t describe above wer mad b(189—192earlie investigator wher smal F precipitatei a meta matri (e.g. Cu wer transforme martensiticalleithe b deformatio b remova o th volumconstrain (e.g. nea th foi th presen casepresumabl th twinne region i an 4 transformefro th origina retaine austenit becaus o th locadeformatio durin handlin o th foi o throug relaxatio b th remova o th surroundin martensit matridurin thinning Sinc th C conten o th austenit iraise t nea 1 wt throug partitionin i woul onl plat martensiticall whic contain transforma twins I Fig 4 a exceptionall wid origina regio betwee th lath transforme int marten deformation whic contain severa variant o{112 twin whic hav ver smal siz an orien themselvei a configuratio t accommodat th stresses Retaineaustenit ma als transfor b removin th constrain(192cause b th surroundin martensite A show i Fig45 ther wa n additiona constrain aroun thstri (unlik th cas i Fig 44) ver lon an large wer fre t form 134.5.2 Therma Instability Decompositio o RetaineAustenit upo TemperingA discusse above durin coolin belo M th recedin austenit region becom increasingl resistan ttransformatio an smal quantitie persis t roo temperatur o below However durin subsequen temperin a higtemperature wil decompos thermally bactivate nucleatio an growth t cementit an ferriteWherea temperin i th interio o th lath follow thwel know clusterin an metastabl carbid sequence iappear tha Fe, nucleate directl o th austenite-martensit interface Ther wil b a definit incubatio periodetermine b th temperin temperatur unti th C concen alon th boundar i hig enoug t for a carbidembry an gro wit tim t attai th critica siz o a nuclei a sufficientl hig temperin temperature T , situatio i ver favorabl a fa a th chemica anvolum fre energie ar concerne (neglectin al thothe energie whic hav relativel smal critica nucle i formed C diffuse fro y violentlan literall pour int th ne cementit a th boundaryTh C deplete region i austenit probabl transfort ferrit wit ver lo C conten b a mechanis 13whic involve shea (simila t a massiv transformatio oy int a + Fe, a isotherma hol abov M i th uppe(192bainiti proces i repeate alon thct'/ boundar a th loca region i y wher C achieve a concentratio valu i Fe,C Hence island o Fe, for angro fro y Th eas growt directio mus b alon thfil rathe tha int th martensite Th particle groeve large tha th retaine austenit widt leaving ibetween th lo C austenit whic shear int a Hence thoveral transformatio i bainiti whic involve diffusioan redistributio o C i y an a' an th transformatioo remainin Y int a b th propagatio o a coheren a'/Yinterfac b a shea process th kinetic o whic dependo th C rejectio fro y I thi sense th reactio ilimite b carbo diffusio i y an a th a'/y-interface, . (120,121t for carbidesI orde t trea th decompositio o thi metastablaustenit a lowe temperin temperature withi th thermodynamica framework i i necessar t establis a/y/carbidphas equilibri whic ca b derive fro th calculatefree-energ compositio curve fo ferrite cementit an(15 193austenite Som earlie investigator ' hav studieth subjec i detai an arrive a a/y/Fe,metastabl phas equilibria Th metastabl Fe- equilibriudiagra (Fig 46 i draw b usin thei data ' t°°K 900 \ / TS+ft--SOO Yff 700 4 -600 ..900 soo -i^~^-^^r iff efCofivr OL+FethO. .-400 J i 'y kX*5.«.600 SOO \*y.. SOO !•• •300 Xx" .-100 1 ...,tX- I 2~~3 " 5 t 7 8~ trtl I2~l3 14 15 IB \7• 0% C.8 US 1.83 2-34 Z8? 1 j "Fig A th Fe- diagra showinc equilibriuline betwee (reproduce fro an 193)x a th solidus effect.XB 8110-1173 13I thi diagram x i th atomi fractio (at frac.o C i a a th a/( + Y phas boundaryx at frac o C i a a a/(a + Fe-C phas boundaryx at frac o C i y a a phas boundaryx at frac o C i Y a cm phas boundaryx an x at frac o C i Fe, a cm/(c + Y ancm cm 3 cm/(c + a phas boundaries respectively Fro thiinformatio on ma dra a carbo profilacros th ferrit (martensitic + austenit (retainedstructur (Fig 47) Her th averag carbo concentratioi th bul materia i take t b 0. wt (1. at%—horizonta broke line i Fig 47) Th concentration o tha'/Y-interfac an i Y ar th averag value fro th atoprob analysi result (therefor th upper dashezonta lin i Y i xOnc a nucle o Fe, concentratio i achieve i Yi wil star growin wit a rat determine b th C diffusio i Y an throug th a'/Y-interface Becaus o thfavorabl chemica potential th nucle wil for iretaine austenite e the o a defec withi th film oo th Y sid o th a'/Y-interface A show schematicalli Fig 48 C wil probabl diffus int th carbid mainl th austenit phase i whic C ha hig chemicapotential som C wil als diffus fro martensitea' wher D i ver high an a loca equilibriu wil bachieve betwee a an Fe, a th ne 13DCarb MartensiteRet-Aust oi 2.0-05-**ecu P ****r——A••^offtFig Schemati representatio o th C concentratiothroug a' carbides a'/ interfac an retaine austeniti th microstructureXB 8110-1173 13Martensite oC 6625 Oi.o 0.5 0.3 0O3 002 Fig Schemati representatio o th locaequilibriu nea th a'/ interfac jus afteth carbide star t for fro th decomposinretaine austenite 14Ther wil b C deplete region i y a th C diffuseint th newl forme carbid "islands alon th interfaceThes region wil the transfor int ferrit (a probablb a shea mechanism Hence th overal configuratio wilb a discontinuou "stringer o carbide alon th a'/a A actua exampl o suc a configuratio ishow i th hig magnificatio TE image i Fig 4 whicwer take fro a sampl containin 0.3 tempere a 400"fo on hourNot tha ther shoul b a loca equilibriu betwee y an Pe, (wit x =0. fro Fig 46) Tha is thJ ycdecompositio reactio wil continu unti C i f i 0. anceas there a whic poin ther wil b a ver smal amouno " lef a th interface Her w als assum tha theri n "ferrit bridge (a suggeste b Darke an Fisher(betwee y an Fe, whic ma b th reaso wh threactio y •* Fe, + a i observe rathe tha th reactio�Y--Fe, + a + y ( = 0.5) Th firs o thes reaction. yha mor drivin forc (large -v A ) i whic cas thremainin y wil transfor t a'Thi regim o th transformatio temperature corre­spond t a uppe bainiti regio i a stee wit a compositio o retaine austenit (Fig 50) Thi i whth origina as-quenche structur ha bee terme aix • -x. ,,(40,104,120,121,194"incomplet uppe bainitic." ' ' ' 14Fig B (a an D (b TE image revea th "stringerso discontinuou bu elongate carbide transforme froretaine austenit a a'/a' boundar i a tempere specime(Allo 4 tempere a 400° fo 1 800 700 GOO ^J)500 "400 300 200 I K Bulk \$///)i//////M May /H/ •3Cr-03C -—JCT-LOC i//j///\ft ,,B 'In 010 10 10' 10 tJO 10 10 y,BLFig stee a hypothetica stee 195rpv^\r"?r . 144.6 CONCLUSION(i Retaine austenit ha bee identifie i e numbeo lo an mediu carbo steel a continuou film (widtusuall betwee 5 t 20 A wit a tota amoun les tna5 b volume(ii Althoug n partitionin o substitutiona solutelement (suc a Cr Mn Mo Ni wa detected a consider amoun o carbo doe partitio int th austenit(2- at o an t th a'/y-interfac (aroun1 at o 2. carbo drop t verlo value i th martensit nea th boundary Th hig conten produce considerabl chemica stabilizatioo th austenite(iii Therma an mechanica stabilization appea tb responsibl fo th presenc o y I therma stabilizatio th movemen o th o'/ interfac i b th higdensit o carbo cluster a th interface Th retaineaustenit i mechanicall stabilize becaus o th accommodatio o th stresse create durin th y t amatio whic highl deform y ahea o th interfacinhibitin it furthe advance(iv Removin th volum constrain an recoverin thfilm fro thei defect ma caus th retaine austenit(containin hig carbon t decompos t plat ir.artensite 14(v Hetastabl austenit persist durin lo temperatur agin (i.e. decomposes however t a morpholog ver simila t "uppe bainite durin agin ahighe temperature (i.e. 145.1A COMPUTE PROGRA T PLO STEREOGRAPHI PROJECTIO ANCOMPOSIT STEREOGRAPHI PROJECTION FO MARTENSIT ANAUSTENIT HAVIN DIFFEREN ORIENTATIO RELATIONSHIPA existin main-fram compute progra wa modifie tplo stereographi projection (s-p fo an crysta systeb a PDP-1 minicomputer. Thi ma als provid a handmetho o indexin electro diffractio pattern (e.g. Ref94) orientatio relationshi analysi on ca superimpos composit s- fo eac phase Th angula relationshipbetwee th plane and/o th direction ca b discernefro thi b inspection Dat consist o th lattic parameter ( , b , c ) th angle betwee th principa direction (a 6 ,' indice o th centra pole tha o a polo th grea circl an th radiu o th grea circleThre ver commonl observe O.R. betwee a an y arplotte i Fig Al-1 I al o thes case ther i aleas on commo directio (o plane i bot phaseTh othe pol (o th grea circle ma b coinciden fobot phases i whic cas bot stereogram ar plotteregularly I non o th pole i eac phas ar coincidentthe th secon stereogra i simpl rotate abou th cente a appropriat amoun (e.g. fo G- O.R. th austenitstereogra i rotate 2.5 wit respec t martensite i.e. angl betwee (111 an (110 ,j•Progra ca b supplie upo reques b th author 14�x-s-i^-* ,, *-st!-V•/ / ^ ? t... 1 W-S ^ *— * -Stl-' * \ •p ^' "•„ ' -N// • * " -. \ . = C '•*.': ' . ' \ . • = ,''- *'; !..'"*,**,.*•,. I •:*9 / Fig Compute plotte sterecgraphi projectionfo austenit an martensit fo differen OrientatioRelationships 145.2VARIATIO O TH LATTIC PARAMETER O AUSTENIT ANMARTENSIT WIT ALLOYIN ADDITION AN A ATTEMP TMEASUR LATTIC PARAMETE O MARTENSIT B CBE TECHNIQUFo calculatin th lattic paramete o retaineaustenite th formula = 0.0 x wt Cwa used Thi formul ha bee adjuste s tha i accountfo th stat o retaine austenite becaus ther i a volum constrain i retaine austenit b th surroundinmartensit matrix However i ou cos retaine austenitstrip ar fre fro an kin o matri constraint (seFig instead th formul fo th equilibriu(2 140-142austenit lattic paramete ' shoul b used Thais = C (see e.g. Ref 141) Not alstha othe alloyin elements interstitials suc a N ansubstitutionals i steel als effec th lattic paramete(24i retaine austenit althoug slightly ' I th particula cas above th allo contain 2 wt S whos effec inegligible, becaus o th smal amoun (les tha 1 i 1000a 2 wt Si) Nonetheless th belo formula whic relat lattic paramete chang i equilibriu y a a functioo th alloyin element i steel ar give fo furtheinformation(i a 3.5780 + 0.0330 + 0.0220 + 0.0095M(±0.0110 (+0.00015 - 0.0002N + 0.006C + 0.0220*3.573 accordin t Ref 141 14(±0.00004)(10.0003 (±0.0034 + 0.0056A- 0.0004C + 0.0015C + 0.0031M + 0.0051N+ 0.0039T (±0.0004)(±0.0012)(±0.0009 + 0.0018+ 0.0018 (±0.0004)(±0.0003 (I wt%(ii a (±0.0021 = 3.577 + 0.0065 + 0.0056o (±0.00004)(±0.0003)(40.0011 + O.OOlOM (±0.0029(+0.0002 - 0.0002N + 0.0006C + 0.0028A- 0.0004C + 0.00014C (±0.00045)(±0.0002)(±0.0007+ 0.0053M + 0.0079H + 0.0032T (±0.0006)(±0.0027(±0.0010 + 0.0017 + 0.0057 (±0.0004)(±0.0012).(at%Man attempt hav bee trie t determin th latticparamete o th martensit matri wit differen carbo content o th heav dislocatio density residuastresses multipl scattering ther wa no enougvolum i martensit t contribut t for a CBE patternalthoug ver smal diamete ( A electro probes whicwer produce b a Fiel Emissio Gun wer used A examplo a martensit (111 CBE patter i show i Fig A2-laI th centra disc th paralle concentri fringe arextinctio contour du t th foi thickness Ther i nothe featur (e.g. HOL lines Kikuch lines whic woul b presen i a CBE pattern suc a th on i whic wa take fro retaine austenite Furthe wil b continue o thi aspec o th researc bexplorin ne method t prepar dislocatio fre matrix 14Th formula whic relat th lattic parameter ( an c omartensit t C conten ar give belo fo furthe usea (A = 2.866 - C (Ref 24 141c (A = 2.866 + C (Ref 24 141C (a M (a b usea referenc material Fig CBE pattern fo martensit (a an austenit (b i (111 zonaxi orientations Detail o th HOL line ar missin i patter (a dut heav dislocation i martensite Patter (b whic reveal HOL line watake fro th regio show i (c) Not i (c tha th austenit "fingersar fre fro surroundin matrix 155.3 APPEND1X-CRITICA DIFFUSIO TIM DURIN LAT GROWT AN THPREDICTIO O TH GROWT RAT O MARTENSIT LATHTh concentratio gradient determine b AP showetha whil C i deplete i a nea th boundar i iincrease i y t twic tha o th bul valu an rise tmuc highe value (5-1 at% a th a'/ interface Aapproximatio o th distanc tha C ha t trave t forthi concentratio profil ca b obtained Assumin ththicknes o th boundar id. t b 2 A austenit hal t b 8 & (Fig takin c = 1. at%th distance d , t b travelle b C t achiev th Cfile foun t b 60 & fro th two-dimensiona geometro th boundar regionThi C redistributio take plac durin quenchinbetwee M an M (33 an 260° fo 2 M alloy respectively . on ca mak a assumptio tha thquenchin i interrupte fo a tim perio o t a T (FigA3-lb) T equa t 290° frod = 2/ t (diffusio distanc travelle b C durin t x c c c awher D i th diffusio coefficien o C i a" Appendi4 a T on find t t b abou 0.0 Sec (Thi figur ia lowe boun fo t whic applie provide tha adequatdrivin forc exist fo C t diffus i a'.Thi i a perio durin whic th carbo concentratiogradien i forme an i a measur o th tim tha lat 15thicken t it fina dimension Hence takin thi a thavailabl tim fo th growth on ca fin a approximatvalu fo th sidewis growt rate Takin th averag latthicknes betwee 0. t 0. um th growt rat lie betwee1 t 1 A/sec Th growt rat predicte fo platmartensit i 10 bainiti ha bee measure t b 1 cm/sec (1Fro th stud i Ref 3 (th onl observation o thisubject b th roug measurement mad usin Fig 1 o thireferenc on ma predic th sidewis growt rat o4 o martensit lath t b abou 3 x 1 A/sec Thi valu liewithi th qualitativel predicte rang foun i th abovappendix [However i shoul b strongl mentione her th lat growt i no linea an proceed a a deceleratin rat a th temperatur decrease withi th M -Mrang (probabl exponentially) Thi i actuall th cas ibainit growth ' A T decreases D als decrease henc t ca b different However thidiscrepanc i no expecte t b ver large Th reaso t4 kee th sidewis growt rat i a larg rang t5 ° 1 A/sec. i partl becaus o thi discrepancy. 15to 0 H 1 \J o t=tK t=to6^•l 1 J .'•t=t(a4 T t-(bhtI T t c tI t c C&#x? 00;-*-^C(d XB 823-859Schemati illustratio o th possibl sequenclengthwis growt o lath dotredistribution (b Th ste a T thcoolin curv i a hypothetica interruptio (t whic iequivalen t th tim elapse durin coolin betwee MMf Approximat C profil simila t thos determine bFIM-APA (d Illustratio o th amoun o C i eac regionFig (ao sidewis anrepresen th C 155.4DIFFUSIO COEFFICIENT O CARBO I MARTENSIT AN AUSTENITA numbe o investigation i th literatur hav reporte th measurement o th diffusio coefficient ocarbo i ferrit ( ) Thes determination hav bee dontheoreticall a wel a experimentall usin differentechnique an employin statistica methods Th valuegive li withi th simila rang an th mos frequentl one ar a followsVD = 0.0 exp( ^0D = (3)D = 0.0 exp( 2^0D = ±i±l° (5»D 0.0 ,6)Th value o D give her ar mor o les i th temperatur rang betwee ambien temperature an ° holfo th compositio o C o abou 0. wt (n alloyinelemen value ar plotte i Fig A4-laTh bes fi (represente b th dotte lin i th figurecorrespond t D = £1, . shoul ba . note tha becaus o th tetragonalit o martensit D i :5a littl lowe tha D (becaus o th increas i activatioTh sam sor o analyse wer performe fo th diffuv sio o C i y, i.e. D' frequentl cite value ar afollowsDl = 0.0 exp( ii2M )D . exp(--23|P (8)«A recen theoretica analysi als state thaD = ^^ (9Thes value ar al fo hig temperatur� 700°Ci th austenit range Fo ou purpos o interes fo thcalculation i th tex th extrapolate value wer used th retaine austenite th bes fi o whic giveD - 0.01 exp( i^ 15I0I \ rft o' 1 ii-D\ \ u \ \ \ \ �." I.150 60 4020101 20 -51 1 12345H/ »Fig (a Redeterminatio of,th diffusio coefficient o carbo i martensit (D ) an austenit (D^(th number o th curve correspon t th eqn numberi th Line o th averag value o D° an D^XB 825-977 155.5 APPENDIX-EFFECTIV DIFFUSIO DISTANC FO CARBO DURIN THTRANSFORMATIO WITHI M - RANGC diffuse t ot'/ interfac durin quenching a wela t th dislocation withi laths Th assumptio i thath lat boundarie behav a infinit sink fo C Alsoth boundarie ar considere t b fixe i spac an al o complication resultin fro thei movemen durin quench ar ignored Th diffusio proble i reduce t onwit boundar condition wher C 0 (carbo concentratioa th interfac o a' effectiv diffusiodistanc i give b (se Ref 17 fo furtheM |ri 1 , ,V . .,) - 20°2 (172wher a therma diffusivit ( /sec) ^ ' hea transfe coefficienh , o L/ quenchin mediu . . . . . k = conductivit t00; = 1-5 c quenco th specimea r : Are t Volum rati o th specimeT : Temperatur o th specime (AustenitizintemperatureT : Temperatur o th quenchin mediumHer th valu o D(T i take fro th previou appendi(ne valu fo D ) Th effectiv diffusio distanc fovariou sampl geometrie fo ic brin an oi quenchin 15ca b calculate b numerica analysi o th integralTh limit o integra wer chose a M an M (no 25°sinc w ar onl intereste i th C diffusio withi H -Mrange som diffusio stil occur belosampl considere fo th analysi contain 0.2 wt C witM an M ° 250°C respectivelySampl * ' X = 2/ ' ^ A IA2.1 X 1 " 3. X 16688B IA3.4 x 10~ll 6.1 x83110(h/k) stil brin (IB•vl.0 agitate oi (AOSampl dimensions A 0. c x 0. c x 1. cB c x 1. cm 1. c 155.6 APPENDIX-INTERSTITIA ATOM-DISLOCATIO INTERACTIOSolut atom segregatin th reducelasti energ th region arounb determine th energ betweeinterstitia ato thth movemen a amoun(AF.). whethechemica drivin forc coul mov bpredicted a transformationth austenite/martensit interfac contain dislocatio net(1 "works Th forc opposinfac movemen becaus th build-u thdislocation Atmospheres" ma bdetermine fre energ transformatio M , AG^i 300-40 Cal/mole U9,24,78,111,184 carboa th ot'/ interfac i abou 1 at (fro th ato probdat i th text o abou 1 - 1 carbo atoms/c ; thseparatio betwee th dislocation a th interfac i abou1 A (th averag o th value give i 108 1591 2 178) densit o dislocations S then i 1 /c . Thes value correspon t abou 1 carbo ato pe dislocatio a th interfac plane Therefore th tota bindinenerg o carbo atom a th dislocation a th interface(AF abou 10 (give tha F 1 eV/atom 16dislocation fo carbo i chemica forcavailabl t mov th dislocation b on Burger vector1 1 2 A , i approximatel 1 - 1 eV/c . Notin tha Aan (4F. ar o opposit sig an hav comparabl magnitude a'/ interfac movemen i the prevente bpinnin o th dislocation b C atom a th interface 166. ACKNOWLEDGEMENTI appreciat an acknowledg th opportunitie createb Professo G Thoma wh ha supervise an supporte thproject involve i thi thesis I a gratefu t Or R MFishe whos invaluabl advic an interest wer stimulatingDee appreciatio an thank ar als extende t ProfessorP M Kell an K E Easterling an Drs B V NarasimhRao M Krive an J W Steed whos interests suggestions hel mad direc contribution t differen parto thi researchTh continuou mora suppor fro m parent an friendsChannin an Ulrich an fro Madelein Penton wh ha alskiudl type th manuscript ha bee valuable I thanOndre Krivane an Pete Re fo showin m "th rea worl microscopy.Ato Prob Analyse wer mad i collaboratio witDrs J Barnard G D W Smit (Oxfor KMille an S S Brenne (U.S Stee Co.)Professor an A G Evan criticall reviewean commente o th manuscriptI gratefull acknowledg th Scienc Scholarshi froTUBITAK-BAY (Turkey an appreciat th researc facilitiean technica staf provide b Lawrenc Berkele LaboratoryThi wor wa supporte b th Director Offic oEnerg Research Offic o Basi Energ Science Materia Divisio o th U.S Departmen o Energ unde No DE-AC03-76SF00098 167M Cohe an C M Wayman "Treatie i Metallurgy,J K Tie an J F ElliotJ W Christian "Physica Propertie o Martensit anBainite, Repor #93 Th Iro an Stee Inst(1965)G V Kurdjumo an A G Khachaturyan Act Met. Z3 (1975)Z Nishiyama "Symp o Mart Trans., Mem o th Inso Sci an Ind Res. Osak Univ. XXIM Cohen "Phas Transformation i Solids, RSmaluchowskiy J E Mayer an W A WeyWile an Son 17 588-660C M Wayman Metallography 8M Cohen G Olson an P C Clapp Proc ICOMA '79MI PresA L Roitbur an G V Kurdjumov Mat Sci an Eng.3C M Wayman "Int t Crystallograph o MartensitiTransformations, Macmilla Co. Ne YorZ Nishiyama "Martensiti Transformations, AcademiPress YorG K Bansa an A H Heuer I-Act Met. 2I (1972)409-417 II-ibid. 22_ (1974) 1281-1289G M Wolten Act Crystallogr. V 1613 C Subbaro H S Maiti an K K Srivastava PhysStat Sol. A 2114 J Altstette an C M Wayraan Act Met. JJ99215 M Kell an J Hutting JISI 1916 Bel an W S Owen JISI 2017 Thomas Met Trans. 2 R Marde an G Krauss Trans ASM S2_96919 Cohen E S Machlin an V G Paranjpe "Thermodynamic i Physica Metallurgy, AS20 S Machli an M Cohen Trans AIM489-50021 Kaufma an M Cohen Progr Meta Phys. T_165-24622 R Pat an M Cohen Act Met. r L Mage an R G Davies Act Met. 2J1031-104324 M Moye an G S Ansell Met Trans. 61785-179125 Olso an M Cohen Act Met. 2_26 R Troian an A B Greninger Meta Progress 5J(1946)27 S Lieberman Act Met. 628 C Frank Act Met. 1_ 1629 Osmond "Microscopi Analysi o Iron-Carbo Alloys,Bulleti de a Soc d'Encouragemen pou L'IndustriNational H30 M Kell an J Nutting Proc Roy Soc (A) 25(1960)31 Wayman Adv Mat. H HemaInterscience Ne York 147-16032 S Owen F J Schoen an G R Srinivasan Proc"Phas Transformations (ASM33 Kraus an A R Marder Met Trans. 22243-225734 Shimiz an Z Nishiyama Met Trans. 31055-106835 Mak an C M Wayman Act Met. 269336 J Schoen J L Nilles an W S Owen Met Trans.237 Raghava an M Cohen Met Trans. 2241838 Christian Proc ICOMA '79 Cambridge MI Pres(1979)39 Wakas an C M Wayman Act Met. 2101240 V Narasimh Ra an G Thomas Proc ICOMA '79Cambridge MI Pres41 K Da an G Thomas Trans ASM .6 1642 H Huan an G Thomas Met Trans. 2159443 Y Ko an G Thomas Met Trans. 853344 Thomas Battell Collogui o Fundamenta Aspect oStructura Allo Design R I Jaffe an B A Wilco(eds) Publ Co45 V Narasimh Ra an G Thomas Met Trans 11(1980)46 K Miller P A Beaven an G D w smith Surfacan Interfac Ana. 1, No 547 S Brenner Surfac Science 248 K Miller Ph.D Thesis Univ o Oxfor49 G Bain Trans AIME 750 Gilber an W S Ower Act Met. U±51 Maki S Shimooka T Arimoto an I Tamura TransJIM 152 S Ansell M J Carr an J R Strife Japa Insto Metals Proc 1s JI Symp o "Ne Aspect oMartensiti Transformations, Kobe Japa53 Tanak an K Shimizu Trans JIM ^ G Davie an C L Magee Met Trans. 21939-194755 F Meh an D M Va Winkle Societ Francais dMetallurgie L No 7 1656 J Baker P M Kelly an J Nutting "PhasTransformations, ASM Ch 257 G Paranjpe M Cohen M B Bever an C F FloeTrans AIM58 Bel an W S Owen Trans AIME 23±194959 G A Biswa an I Codd JISI .2060 R Marde an G Krauss Trans ASM 666061 C Low P R Howell an D V Edmonds Meta Sci.Sept62 J Harri an M Cohen Trans AIME ,1847063 P Maximov an E I Estrin Fiz Met Metallored±64 V Kurdjumo an G Sachs Eei* fu Phys. 6£325-34365 Nishiyama Sci Rep Tohok Univ. 22.66466 Wassermann Mitt K-W-I Eisenforsch149-15567 B Greninge an A R Troiano Metal Trans. 18(194968 S Wechsler D S Lieberman an T A Read TransAIM 1669 S Bowle an J K Mackenzie Act Met. 2I 130-137 II 138-147 III 224-23470 A Bilb an F C Frank Act Met. j40271 G Crocke an B A Bilby Act Met. ±678-58872 S Lieberman "Phas Transformations, ASCh 1-5873 Knap an U Dehlinger Act Met. 474 L Magee "Phas Transformations, AS 3115-15575 B Olso an M Cohen " Perspectiv o MartensitNucleation, Ann Rev Mat Sex 176 Zener Trans AIME 1JT Phy Rev.277 Beisswenge an E Scheil Arch Eisenhuttenwes 2J_ (1956)78 C Fisher J H Hollomon an D Turnbull MetalTrans.79 R Entwistle Met Trans. 280 Swa an G Krauss Met Trans. 7 J Irvin an F B Pickering JISI 198182 S Pascove an S V Radcliffe Trans AIME 24(1968)83 R Speic an P R Swann JISI 20 1684 M Chilton C J Barton an G R Speich JISI2085 D Bolto an E R Petty Meta Sci. 9166-17086 S Sarma J A Whiteman an J H Woodhead MetaSci.87 Davenport "Solid-Soli Phas Transformations, ASMPittsburgh 1980 H I Aaronso an C M Waymai Press88 V Narasimh Rao J Y Koo an G Thomas Proc33r EMS Conf. Claitor Publ Div89 V Narasimh Rao M Sarikaya an G Thomas Proc37t EMS Conf. Claitor Publ Div90 J H Cockayne Zeitsch Naturforsch452-46091 B Warren "Int t Analytica J HrenJ I Goldstein an D C Jo Pres(1979)92 V Narasimh Ra an G Thomas Proc ICOMA '77Kiev USS93 V Narasimh Rao Met Trans. 10 W Andrews D J Dyson an S R Keown "Interpretation o Electro Diffractio Patterns, HilgerLondo95 L Ryde an W Pitsch Phil Mag. 15446. 1696 Vo Heimendal W Bell an G Thomas J ApplPhys.97 Wakasa C M Wayman H Kubo an K ShimizuScriptaMet. 1±98 K D H Bhadeshi an D V Edmonds Proc ICOMA•79 MI Pres99 Sarikaya B V Narasimh Rao an G Thomas Proc38t EMS Conf. Claitor Publ Div100 M Kelly Privat Communicatio101 Sarikaya Proc 34t EMS Conf. Claitor PublDiv102 P Nemirovski an Yu R Nemirovskiy Fiz MetMetalloved. _3£ No 4103 V Eterashvili L M Utevskiy an M N SpasskiyFiz Metalloved. j48 No 4104 Thoma an M Sarikaya Proc "Solid-Soli PhasTransformations, H I Aaranso an C M Wayma (eds(1981 , press105 Sarikaya K E Easterling an G Thomas ProcEUREM Electro Microscopy Vol 1106. R Morga an T Ko Act Met. 1107 B G Yeo Trans AIME 22 R Kinsma an J C Shyne Act Met. 11527-1543109 P Klie an A R Tro.:ano Trans AIME 16175-185 17110 S Ansel1 S J Donache an R W Messier Jr.Met 2111 Zener Trans AIME 18J112 McMaho an G Thomas Proc 3r Int Conf Strengto Metal an Alloys 1, Institut o Metals Londo(1973)113 F Zackay E R Parker an W E Wood Proc 3rInt Strengt o Meta an Alloys 1, Instituto Metals Londo114 K 0 H Bhadeshi an D V Edmonds Meta Science1115 C Fiedler B L Averbach an M Cohen Trans ASM£116. A Kulin M Cohen an B L Averbach Trans AIM(JIM)117 R Pate an M Cohen Act Met. 11538118 H Richma an G F Boiling Met Trans. 22451-2462119 C Fisher Trans AIME 18J120 Thomas Met Trans. 9121 Sarikaya A K Jhingan an G Thomas Met Trans(1982) press122 B Hirsh R Howie R B Nicholson D W Pashleyan M J Whelan "E.M o Thi Crystals, ButterworthsLondo 17123 L Mill r Trans ASM 5 ibid. £(1968)124 Zaluzec "Int t Analytica J HrenJ I Goldstein an D C Jo Pres(1979)125 Sinclair R Gronsky an G Thomas Act Met. 2(1976 ,126 W Steeds "Intr t Analytica J HrenJ I Goldstein an D C Jo Pres(1979)127 C Joy ibid. 223-224128 V Narasimh Ra an G Thomas Proc 35t EMSConf. Publ Div129 G Self H K D H Bhadeshia an W M StoobsUltramicrcscopy 6130 L Krivanek Proc 38t EMS Conf. Claitor PublDiv131 Sarikay an 0 L Krivanek Unpublishe ResearchUniv o California Berkeley Sprin 198C132 S Brenner M K Miller an R M Fisher Proc28t IFES L Swanso an A Bel(1981)133 W Bach J Beyer an C A Vebraak Script Met.14 ,134 S Brenne an J T McKinney Surfac Sci. ^(1970) 17i 135 J Turne an J M Papazian Metal Sci. 281-86136 W Miiller "Method o Surfac Analysis, A WCzandern Amsterda137 J Wilkens G D W Smith an D A SmithMetallography 7138 M Jones G M Rackham an J W Steeds ProcRoy Soc (A) London 35139 B Pearsoi: "Handboo o Lattic Spacing anStructure o Metals, Pergamo Press Ne Yor656. 140 S Roberts Trans AIME 19J141 Rindley H Stuart an L Zwell Trans AIME 24(1969)142 Steve an A G Haynes JISI .18143 J Barnard G D W Smith M Sarikaya an GThomas Script Met. Jj144 K Miller P A Beaven an G D W Smith MetTrans. 12145 Sarikaya G Thomas J W Steeds S J Barnardan G D W Smith Prod "Solid-Soli Phas Transformations, H I Aaranso an C M WaymaPittsburgh Penn press 17146 D W Smith A J Garrat-Reed an J B VandeSande Conf o "Quantitativ Analysi wit HigSpatia Resolutions, UMIST/ManchesteSoc J W Arrowsmith Ltd. Bristol 238-249147 W McReynolds Jrnl Appl Phys. 2907148 Edmondso an T Ko Act Met. 2 Philiber an C Crussard C R Akad Sci. Paris24150 E Easterlin an P Swann Act Met. 1117-121151 L Magee Met Trans. 2152 C Maxwell A Goldberg an J C Shyne MetTrans.153 Kaufman S V Radcliffe an M Cohen "Decompositio o Austenit b Diffusiona Processes, V FZacka an H I Aaronso NeYor154 F Bunsha an R F Mehl J Met. 51255155 B G Yeo Trans ASM 5_156 R Speic an M Cohen Trans AIME .211050-1059157 R Speich "Decompositio o Austenit b Diffusiona Processes, V F Zacka an H I Aaronso Ne Yor 17158 F Goodeno an R F Hehemann Discussio t abovpaper159 J Schoe an W S Owen Met Trans. 22431-2441160 J Smithsells "Metal Referenc Book Vol II3r Ed. Butterworths Washingto161 G Homar "Diffusio i BC Materials, ASMGathinbur162 P Smith Trans AIME 22163 M Fishe an L S Darken U.S Stee Co.Monroeville Unpublishe Research164 E Lor an D N Beshers Act Met. 11659-1672165 Hillert Act166. Wells W Batz an R F Mehl Trans AIME 18(1950167 P Smith Act Met. 1168 K D H Bhadeshia Metal Sci. 1169 Crank "Mathematic o Diffusion, 2n ed.Clarendo Press Oxfor 4170 Scott Trans ASM 2171 Schumann "Metallurgica Engineering, 2, AddisonWesle172 A Grossman M Asimov an S F Urban "Hardenabilit o Allo Steels, ASM Clevelan146. 17173 R Speich Trans AIME 24174 F Hehemann "Phas Transformations, ASCh 397-432175 W Christian Met Trans. 13176 W Christian "Th Theor o Transformation iMetal an Alloys, 2n ed. Pergamo Press Oxfor(1975) 1177 B Olson Act Met. 27178 H Cottrel an B A Bilby Proc Phys Soc LXI(1948)179 G Ramachandra an C Dasarathy Act Met. 8 (1960)180 Edmonson Act Met. .181 R L Guimarae an J C Shyne Met Trans. 2 (1971)182 Harper Phys Rev. j!3183 S Ham Jrn App Phys. 6184 K D H Bhadeshia Metal Sci. Apri175-177185 J McEvily Jr. R H Bush F W Schaller anD J Schmatz Trans ASM� j186 Making Shapin an Treatin o Steels, H EMcGanno Stee Co. 9t ed187 Johar an G Thomas Trans ASM (1965)578 17188 Dechamp an L M Brown Act Met. Z1281-1291189 R Kinsman G Das an R F Hehemann Act Met.2190 R Kinsman Proc 28t EMS Conf. Claitor Publ• Div ,191 E Easterlin an H M Miekk-oja Act Met. L(1967)192 F..Heheman K R Kinsman an H I Aaronson MetTrans.193 S Darke an R: M Fisher "Decompositio oAustenit b Diffussiona Processes, V F Zacka anH I Aaronso Ne Yor249-294194 Sarikaya B G Steinberg an G Thomas MetTrans press19b o Isotherma Transformatio an CoolinTransformatio Diagrams, ASM Metal Park Ohi218