Steel Heat Treatment - Metallurgy And Technologies - 2006...

Steel Heat Treatment - Metallurgy And Technologies
Download Document
Showing page : 1 of 832
This preview has blurred sections. Sign up to view the full version! View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 2006 by Taylor & Francis Group, LLC. 2006 by Taylor & Francis Group, LLC. 2006 by Taylor & Francis Group, LLC. 2006 by Taylor & Francis Group, LLC.PrefaceThe first edition of the Steel Heat Treatment Handbook was initially released in 1997. Theobjective of that book was to provide the reader with well-referenced information on thesubjects covered with sufficient depth and breadth to serve as either an advanced undergraduate or graduate level text on heat treatment or as a continuing handbook reference forthe designer or practicing engineer. However, since the initial release of the first edition of theSteel Heat Treatment Handbook, there have been various advancements in the field thatneeded to be addressed to assure up-to-date coverage of the topic. This text, Steel HeatTreatment: Metallurgy and Technologies, is part of a revision of the earlier text. Some of thechapters in this text are updated revisions of the earlier book and others are completely newchapters or revisions. These chapters include:ChapterChapterChapterChapterChapterChapterChapterChapter1. Steel Nomenclature (Revision)2. Classification and Mechanisms of Steel Transformations (New Chapter)3. Fundamental Concepts in Steel Heat Treatment (Minor Revisions)4. Effects of Alloying Elements on the Heat Treatment of Steel (Minor Revisions)5. Hardenability (Minor Revisions)6. Steel Heat Treatment (Minor Revisions)7. Heat Treatment with Gaseous Atmospheres (Revision)8. Nitriding Techniques, Ferritic Nitrocarburizing, and Austenitic Nitrocarburizing Techniques and Methods (Revision)Chapter 9. Quenching and Quenching Technology (Revision)Chapter 10. Distortion of Heat-Treated Components (New Chapter)Chapter 11. Tool Steels (New Chapter)Chapter 12. Stainless Steel Heat Treatment (New Chapter)Chapter 13. Heat Treatment of Powder Metallurgy Steel Components (New Chapter)Approximately a third of the book is new and a third of the book is significantly revisedversus the first edition of the Steel Heat Treatment Handbook. This new text is current withrespect to heat treatment technology at this point at the beginning of the 21st century and isconsiderably broader in coverage but with the same depth and thoroughness that characterized the first edition.Unfortunately, my close friend, colleague and mentor, Dr. Maurice A.H. Howes, whohelped to bring the first edition of Steel Heat Treatment Handbook into fruition was unable toassist in the preparation of this second edition. However, I have endeavored to keep the sameconsistency and rigor of coverage as well as be true to the original vision that we had for thistext as a way of serving the heat treatment industry so that this book will be a value resourceto the reader in the future.George E. Totten, Ph.D., FASMPortland State UniversityPortland, Oregon 2006 by Taylor & Francis Group, LLC. 2006 by Taylor & Francis Group, LLC.EditorGeorge E. Totten, Ph.D. is president of G.E. Totten & Associates, LLC in Seattle, Washington and a visiting professor of materials science at Portland State University. He is coeditor ofa number of books including Steel Heat Treatment Handbook, Handbook of Aluminum,Handbook of Hydraulic Fluid Technology, Mechanical Tribology, and Surface Modificationand Mechanisms (all titles of CRC Press), as well as the author or coauthor of over 400technical papers, patents, and books on lubrication, hydraulics, and thermal processing. He isa Fellow of ASM International, SAE International, and the International Federation forHeat Treatment and Surface Engineering (IFHTSE), and a member of other professionalorganizations including ACS, ASME, and ASTM. He formerly served as president ofIFHTSE. He earned B.S. and M.S. degrees from Fairleigh Dickinson University, Teaneck,New Jersey and a Ph.D. degree from New York University, New York. 2006 by Taylor & Francis Group, LLC. 2006 by Taylor & Francis Group, LLC.ContributorsS.S. BabuEdison Welding InstituteColumbus, OhioRonald Lesley PlautUniversity of Sao PauloSao Paulo, BrazilElhachmi EssadiqiCANMET, Materials TechnologyLaboratoryOttawa, ON, CanadaDavid PyePye Metallurgical Consulting, Inc.Meadville, PennsylvaniaJohann GroschInstitut fuer WerkstofftechnikTechnische UniversitaetBerlin, Germany Bozidar LiscicFaculty of Mechanical Engineering andNaval ArchitectureUniversity of ZabrebZabreb, CroatiaGuoquan LiuBeijing University of Science and TechnologyBeijing, ChinaMichiharu NarazakiUtsunomiya UniversityUtsunomiya, JapanArnold R. NessBradley UniversityPeoria, IllinoisJoseph W. NewkirkUniversity of Missouri-Rolla,Rolla, MissouriAngelo Fernando PadilhaUniversity of Sao PauloSao Paulo, Brazil 2006 by Taylor & Francis Group, LLC.Paulo Rangel RiosFluminense Federal UniversityV. Redonda, BrazilAnil Kumar SinhaAKS AssociatesFort Wayne, IndianaAnton StichTechnical University of MunichMunich, GermanyAlexey V. SverdlinBradley UniversityPeoria, IllinoisHans M. TensiTechnical University of MunichMunich, GermanySanjay N. ThakurHazen Powder Parts, LLCHazen, ArkansasGeorge E. TottenPortland State UniversityPortland, OregonChengjian WuBeijing University of Science and TechnologyBeijing, China 2006 by Taylor & Francis Group, LLC.ContentsChapter 1Steel NomenclatureAnil Kumar Sinha, Chengjian Wu, and Guoquan LiuChapter 2Classification and Mechanisms of Steel TransformationS.S. BabuChapter 3Fundamental Concepts in Steel Heat TreatmentAlexey V. Sverdlin and Arnold R. NessChapter 4Effects of Alloying Elements on the Heat Treatment of SteelAlexey V. Sverdlin and Arnold R. NessChapter 5Hardenability Bozidar LiscicChapter 6Steel Heat Treatment Bozidar LiscicChapter 7Heat Treatment with Gaseous AtmospheresJohann GroschChapter 8Nitriding Techniques, Ferritic Nitrocarburizing, andAustenitic Nitrocarburizing Techniques and MethodsDavid PyeChapter 9Quenching and Quenching TechnologyHans M. Tensi, Anton Stich, and George E. TottenChapter 10Distortion of Heat-Treated ComponentsMichiharu Narazaki and George E. TottenChapter 11Tool SteelsElhachmi EssadiqiChapter 12Stainless Steel Heat TreatmentAngelo Fernando Padilha, Ronald Lesley Plaut, and Paulo Rangel Rios 2006 by Taylor & Francis Group, LLC.Chapter 13Heat Treatment of Powder Metallurgy Steel ComponentsJoseph W. Newkirk and Sanjay N. ThakurAppendicesAppendix 1 Common Conversion ConstantsAppendix 2 Temperature Conversion TableAppendix 3 Volume Conversion TableAppendix 4 Hardness Conversion Tables: Hardened Steel and Hard AlloysAppendix 5 Recommended MIL 6875 Specification Steel HeatTreatment ConditionsAppendix 6 Colors of Hardening and Tempering HeatsAppendix 7 Weight Tables for Steel Bars 2006 by Taylor & Francis Group, LLC.1Steel NomenclatureAnil Kumar Sinha, Chengjian Wu, and Guoquan LiuCONTENTS1.11.21.3Introduction .................................................................................................................. 2Effects of Alloying Elements......................................................................................... 21.2.1Carbon ........................................................................................................... 31.2.2Manganese ..................................................................................................... 31.2.3Silicon ............................................................................................................ 41.2.4Phosphorus .................................................................................................... 41.2.5Sulfur ............................................................................................................. 41.2.6Aluminum ...................................................................................................... 51.2.7Nitrogen ......................................................................................................... 51.2.8Chromium ...................................................................................................... 51.2.9Nickel ............................................................................................................. 51.2.10 Molybdenum.................................................................................................. 51.2.11 Tungsten ........................................................................................................ 61.2.12 Vanadium....................................................................................................... 61.2.13 Niobium and Tantalum ................................................................................. 61.2.14 Titanium......................................................................................................... 61.2.15 Rare Earth Metals ......................................................................................... 71.2.16 Cobalt ............................................................................................................ 71.2.17 Copper ........................................................................................................... 71.2.18 Boron ............................................................................................................. 71.2.19 Zirconium....................................................................................................... 81.2.20 Lead ............................................................................................................... 81.2.21 Tin.................................................................................................................. 81.2.22 Antimony ....................................................................................................... 81.2.23 Calcium .......................................................................................................... 8Classification of Steels .................................................................................................. 81.3.1 Types of Steels Based on Deoxidation Practice ............................................... Killed Steels ....................................................................................... Semikilled Steels............................................................................... Rimmed Steels ................................................................................. Capped Steels................................................................................... 111.3.2 Quality Descriptors and Classifications ......................................................... 111.3.3 Classification of Steel Based on Chemical Composition ............................... Carbon and CarbonManganese Steels ........................................... Low-Alloy Steels.............................................................................. High-Strength Low-Alloy Steels ...................................................... Tool Steels ....................................................................................... Stainless Steels ................................................................................. 33 2006 by Taylor & Francis Group, LLC. Maraging Steels ...............................................................................Designations for Steels................................................................................................1.4.1 SAE-AISI Designations ................................................................................. Carbon and Alloy Steels .................................................................. HSLA Steels..................................................................................... Formerly Listed SAE Steels.............................................................1.4.2 UNS Designations .........................................................................................1.5 Specifications for Steels...............................................................................................1.5.1 ASTM (ASME) Specifications.......................................................................1.5.2 AMS Specifications........................................................................................1.5.3 Military and Federal Specifications ...............................................................1.5.4 API Specifications..........................................................................................1.5.5 ANSI Specifications.......................................................................................1.5.6 AWS Specifications........................................................................................1.6 International Specifications and Designations............................................................1.6.1 ISO Designations ........................................................................................... The Designation for Steels with Yield Strength............................... The Designation for Steels with Chemical Composition .................1.6.2 GB Designations (State Standards of China) ................................................1.6.3 DIN Standards...............................................................................................1.6.4 JIS Standards .................................................................................................1.6.5 BS Standards..................................................................................................1.6.6 AFNOR Standards ........................................................................................References ...........................................................................................................................1.41.1444546464747475050515154666666666684858686868687INTRODUCTIONAccording to the ironcarbon phase diagram [13], all binary FeC alloys containing less thanabout 2.11 wt% carbon* are classified as steels, and all those containing higher carbon contentare termed cast iron. When alloying elements are added to obtain the desired properties, thecarbon content used to distinguish steels from cast iron would vary from 2.11 wt%.Steels are the most complex and widely used engineering materials because of (1)the abundance of iron in the Earths crust, (2) the high melting temperature of iron(15348C), (3) a range of mechanical properties, such as moderate (200300 MPa) yieldstrength with excellent ductility to in excess of 1400 MPa yield stress with fracture toughnessup to 100 MPa m2 , and (4) associated microstructures produced by solid-state phase transformations by varying the cooling rate from the austenitic condition [4].This chapter describes the effects of alloying elements on the properties and characteristicsof steels, reviews the various systems used to classify steels, and provides extensive tabulardata relating to the designation of steels.1.2EFFECTS OF ALLOYING ELEMENTSSteels contain alloying elements and impurities that must be associated with austenite, ferrite,and cementite. The combined effects of alloying elements and heat treatment produce anenormous variety of microstructures and properties. Given the limited scope of this chapter, it*This figure varies slightly depending on the source. It is commonly taken as 2.11 wt% [1] or 2.06 wt% [2], while it iscalculated thermodynamically as 2.14 wt% [3]. 2006 by Taylor & Francis Group, LLC.would be difficult to include a detailed survey of the effects of alloying elements on the ironcarbon equilibrium diagram, allotropic transformations, and forming of new phases. Thiscomplicated subject, which lies in the domain of ferrous physical metallurgy, has beenreviewed extensively in Chapter 2 of this handbook and elsewhere in the literature [4,5,812].In this section, the effects of various elements on steelmaking (deoxidation) practices andsteel characteristics will be briefly outlined. It should be noted that the effects of a singlealloying element on either practice or characteristics is modified by the influence of otherelements. The interaction of alloying elements must be considered [5].According to the effect on matrix, alloying elements can be divided into two categories:1. By expending the g-field, and encouraging the formation of austenite, such as Ni, Co,Mn, Cu, C, and N (these elements are called austenite stabilizers)2. By contracting the g-field, and encouraging the formation of ferrite, such as Si, Cr, W,Mo, P, Al, Sn, Sb, As, Zr, Nb, B, S, and Ce (these elements are called ferritestabilizers)Alloying elements can be divided into two categories according to the interaction withcarbon in steel:1. Carbide-forming elements, such as Mn, Cr, Mo, W, V, Nb, Ti, and Zr. They go intosolid solution in cementite at low concentrations. At higher concentrations, they formmore stable alloy carbides, though Mn only dissolves in cementite.2. Noncarbide-forming elements, such as Ni, Co, Cu, Si, P, and Al. They are free fromcarbide in steels, and normally found in the matrix [5,11,12].To simplify the discussion, the effects of various alloying elements listed below aresummarized separately.1.2.1 CARBONThe amount of carbon (C) required in the finished steel limits the type of steel that can be made.As the C content of rimmed steels increases, surface quality deteriorates. Killed steels in theapproximate range of 0.150.30% C may have poorer surface quality and require specialprocessing to attain surface quality comparable to steels with higher or lower C contents.Carbon has a moderate tendency for macrosegregation during solidification, and it isoften more significant than that of any other alloying elements. Carbon has a strong tendencyto segregate at the defects in steels (such as grain boundaries and dislocations). Carbideforming elements may interact with carbon and form alloy carbides. Carbon is the mainhardening element in all steels except the austenitic precipitation hardening (PH) stainlesssteels, managing steels, and interstitial-free (IF) steels. The strengthening effect of C in steelsconsists of solid solution strengthening and carbide dispersion strengthening. As the Ccontent in steel increases, strength increases, but ductility and weldability decrease [4,5].1.2.2 MANGANESEManganese (Mn) is present in virtually all steels in amounts of 0.30% or more [13]. Manganese is essentially a deoxidizer and a desulfurizer [14]. It has a lesser tendency for macrosegregation than any of the common elements. Steels above 0.60% Mn cannot be readilyrimmed. Manganese is beneficial to surface quality in all carbon ranges (with the exception ofextremely low-carbon rimmed steels) and reduction in the risk of red-shortness. Manganesefavorably affects forgeability and weldability. 2006 by Taylor & Francis Group, LLC.Manganese is a weak carbide former, only dissolving in cementite, and forms alloyingcementite in steels [5]. Manganese is an austenite former as a result of the open g-phase field.Large quantities (>2% Mn) result in an increased tendency toward cracking and distortionduring quenching [4,5,15]. The presence of alloying element Mn in steels enhances theimpurities such as P, Sn, Sb, and As segregating to grain boundaries and induces temperembrittlement [5].1.2.3SILICONSilicon (Si) is one of the principal deoxidizers used in steelmaking; therefore, silicon contentalso determines the type of steel produced. Killed carbon steels may contain Si up to amaximum of 0.60%. Semikilled steels may contain moderate amounts of Si. For example,in rimmed steel, the Si content is generally less than 0.10%.Silicon dissolves completely in ferrite, when silicon content is below 0.30%, increasing itsstrength without greatly decreasing ductility. Beyond 0.40% Si, a marked decrease in ductilityis noticed in plain carbon steels [4].If combined with Mn or Mo, silicon may produce greater hardenability of steels [5]. Dueto the addition of Si, stress corrosion can be eliminated in CrNi austenitic steels. In heattreated steels, Si is an important alloy element, and increases hardenability, wear resistance,elastic limit and yield strength, and scale resistance in heat-resistant steels [5,15]. Si is anoncarbide former, and free from cementite or carbides; it dissolves in martensite and retardsthe decomposition of alloying martensite up to 3008C.1.2.4PHOSPHORUSPhosphorus (P) segregates during solidification, but to a lesser extent than C and S. Phosphorus dissolves in ferrite and increases the strength of steels. As the amount of P increases,the ductility and impact toughness of steels decrease, and raises the cold-shortness [4,5].Phosphorus has a very strong tendency to segregate at the grain boundaries, and causesthe temper embrittlement of alloying steels, especially in Mn, Cr, MnSi, CrNi, and CrMnsteels. Phosphorus also increases the hardenability and retards the decomposition ofmartensite-like Si in steels [5]. High P content is often specified in low-carbon free-machiningsteels to improve machinability. In low-alloy structural steels containing ~0.1% C, P increasesstrength and atmospheric corrosion resistance. In austenitic CrNi steels, the addition of Pcan cause precipitation effects and an increase in yield points [15]. In strong oxidizing agent, Pcauses grain boundary corrosion in austenitic stainless steels after solid solution treatment asa result of the segregation of P at grain boundaries [5].1.2.5SULFURIncreased amounts of sulfur (S) can cause red- or hot-shortness due to the low-melting sulfideeutectics surrounding the grain in reticular fashion [15,16]. Sulfur has a detrimental effect ontransverse ductility, notch impact toughness, weldability, and surface quality (particularly inthe lower carbon and lower manganese steels), but has a slight effect on longitudinalmechanical properties.Sulfur has a very strong tendency to segregate at grain boundaries and causes reduction ofhot ductility in alloy steels. However, sulfur in the range of 0.080.33% is intentionally addedto free-machining steels for increased machinability [5,17] .Sulfur improves the fatigue life of bearing steels [18], because (1) the thermalcoefficient on MnS inclusion is higher than that of matrix, but the thermal coefficient ofoxide inclusions is lower than that of matrix, (2) MnS inclusions coat or cover oxides (such as 2006 by Taylor & Francis Group, LLC.alumina, silicate, and spinel), thereby reducing the tensile stresses in the surrounding matrix[5,10,19].1.2.6 ALUMINUMAluminum (Al) is widely used as a deoxidizer and a grain refiner [9]. As Al forms very hardnitrides with nitrogen, it is usually an alloying element in nitriding steels. It increases scalingresistance and is therefore often added to heat-resistant steels and alloys. In precipitationhardening stainless steels, Al can be used as an alloying element, causing precipitationhardening reaction. Aluminum is also used in maraging steels. Aluminum increases thecorrosion resistance in low-carbon corrosion-resisting steels. Of all the alloying elements, Alis one of the most effective elements in controlling grain growth prior to quenching.Aluminum has the drawback of a tendency to promote graphitization.1.2.7 NITROGENNitrogen (N) is one of the important elements in expanded g-field group. It can expand andstabilize the austenitic structure, and partly substitute Ni in austenitic steels. If the nitrideforming elements V, Nb, and Ti are added to high-strength low-alloy (HSLA) steels, finenitrides and carbonitrides will form during controlled rolling and controlled cooling. Nitrogen can be used as an alloying element in microalloying steels or austenitic stainless steels,causing precipitation or solid solution strengthening [5]. Nitrogen induces strain aging,quench aging, and blue brittleness in low-carbon steels.1.2.8 CHROMIUMChromium (Cr) is a medium carbide former. In the low Cr/C ratio range, only alloyed cementite(Fe,Cr)3 C forms. If the Cr/C ratio rises, chromium carbides (Cr,Fe)7 C3 or (Cr,Fe)23 C6 or both,would appear. Chromium increases hardenability, corrosion and oxidation resistance of steels,improves high-temperature strength and high-pressure hydrogenation properties, and enhancesabrasion resistance in high-carbon grades. Chromium carbides are hard and wear-resistant andincrease the edge-holding quality. Complex chromiumiron carbides slowly go into solution inaustenite; therefore, a longer time at temperature is necessary to allow solution to take placebefore quenching is accomplished [5,6,14]. Chromium is the most important alloying element insteels. The addition of Cr in steels enhances the impurities, such as P, Sn, Sb, and As, segregatingto grain boundaries and induces temper embrittlement.1.2.9 NICKELNickel (Ni) is a noncarbide-forming element in steels. As a result of the open g-phase field, Niis an austenite-forming element [5,11,15]. Nickel raises hardenability. In combination with Ni,Cr and Mo, it produce greater hardenability, impact toughness, and fatigue resistance insteels [5,10,11,18]. Nickel dissolving in ferrite improves toughness, decreases FATT50% (8C),even at the subzero temperatures [20]. Nickel raises the corrosion resistance of CrNiaustenitic stainless steels in nonoxidizing acid medium.1.2.10 MOLYBDENUMMolybdenum (Mo) is a pronounced carbide former. It dissolves slightly in cementite, whilemolybdenum carbides will form when the Mo content in steel is high enough. Molybdenumcan induce secondary hardening during the tempering of quenched steels and improves thecreep strength of low-alloy steels at elevated temperatures. 2006 by Taylor & Francis Group, LLC.The addition of Mo produces fine-grained steels, increases hardenability, and improvesfatigue strength. Alloy steels containing 0.200.40% Mo or V display a delayed temperembrittlement, but cannot eliminate it. Molybdenum increases corrosion resistance and isused to a great extent in high-alloy Cr ferritic stainless steels and with CrNi austeniticstainless steels. High Mo contents reduce the stainless steels susceptibility to pitting [5,15].Molybdenum has a very strong solid solution strengthening in austenitic alloys at elevatedtemperatures. Molybdenum is a very important alloying element for alloy steels.1.2.11 TUNGSTENTungsten (W) is a strong carbide former. The behavior of W is very similar to Mo in steels.Tungsten slightly dissolves in cementite. As the content of W increases in alloy steels, W formsvery hard, abrasion-resistant carbides, and can induce secondary hardening during thetempering of quenched steels. It promotes hot strength and red-hardness and thus cuttingability. It prevents grain growth at high temperature. W and Mo are the main alloyingelements in high-speed steels [5,13]. However, W and Mo impair scaling resistance.1.2.12 VANADIUMVanadium (V) is a very strong carbide former. Very small amounts of V dissolve in cementite.It dissolves in austenite, strongly increasing hardenability, but the undissolved vanadiumcarbides decrease hardenability [5]. Vanadium is a grain refiner, and imparts strength andtoughness. Fine vanadium carbides and nitrides give a strong dispersion hardening effect inmicroalloyed steels after controlled rolling and controlled cooling. Vanadium provides a verystrong secondary hardening effect on tempering, therefore it raises hot-hardness and thuscutting ability in high-speed steels. Vanadium increases fatigue strength and improves notchsensitivity.Vanadium increases wear resistance, edge-holding quality, and high-temperature strength.It is therefore used mainly as an additional alloying element in high-speed, hot-forging, andcreep-resistant steels. It promotes the weldability of heat-treatable steels. The presence of Vretards the rate of tempering embrittlement in Mo-bearing steels.1.2.13 NIOBIUMANDTANTALUMNiobium (Nb) and tantalum (Ta) are very strong carbide and nitride formers. Small amountsof Nb can form fine nitrides or carbonitrides and refine the grains, therefore increasing theyield strength of steels. Niobium is widely used in microalloying steels to obtain high strengthand good toughness through controlled rolling and controlled cooling practices. A 0.03% Nbin austenite can increase the yield strength of medium-carbon steel by 150 MPa. Niobiumcontaining nonquenched and tempered steels, including microalloyed medium-carbon steelsand low-carbon bainite (martensite) steels, offer a greatly improved combination of strengthand toughness. Niobium is a stabilizer in CrNi austenitic steels to eliminate intergranularcorrosion.1.2.14 TITANIUMTitanium (Ti) is a very strong carbide and nitride former. The effects of Ti are similar to thoseof Nb and V, but titanium carbides and nitrides are more stable than those of Nb and V. It iswidely used in austenitic stainless steels as a carbide former for stabilization to eliminateintergranular corrosion. By the addition of Ti, intermetallic compounds are formed inmaraging steels, causing age hardening. Titanium increases creep rupture strength throughformation of special nitrides and tends significantly to segregation and banding [15]. 2006 by Taylor & Francis Group, LLC.Ti, Nb, and V are effective grain inhibitors because their nitrides and carbides are quitestable and difficult to dissolve in austenite. If Ti, Nb, and V dissolve in austenite, thehardenability of alloy steels may increase strongly due to the presence of Mn and Cr in steels.Mn and Cr decrease the stability of Ti-, Nb-, and V-carbides in steels [5].1.2.15 RARE EARTH METALSRare earth metals (REMs) constitute the IIIB group of 17 elements in the periodic table. They arescandium (Sc) of the fourth period, yttrium (Y) of the fifth period, and the lanthanides of the sixthperiod, which include the elements, lanthanum (La), cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm), europium (Eu), gadolinium (Gd), terbium (Tb),dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm; Tu), ytterbium (Yb), and lutecium(or lutecium, Lu). Their chemical and physical properties are similar. They generally coexist andare difficult to separate in ore beneficiation and metal extraction so they are usually supplied as amixture and used in various mixture states in metallurgical industries. REMs are strong deoxidizers and desulfurizers, and they also react with the low-melting elements, such as antimony(Sb), tin (Sn), arsenic (As), and phosphorus (P), forming high-melting compounds and preventing them from causing the red-shortness and temper embrittlement [21,22]. The effects of REMon shape control and modification of inclusions would improve transversal plasticity andtoughness, hot ductility, fatigue strength, and machinability. REMs tend strongly to segregateat the grain boundaries and increase the hardenability of steels [21,23].1.2.16 COBALTCobalt (Co) is a noncarbide former in steels. It decreases hardenability of carbon steels, butby addition of Cr, it increases hardenability of CrMo alloy steels. Cobalt raises the martensitic transformation temperature of Ms (8C) and decreases the amount of retained austenite inalloy steels. Cobalt promotes the precipitation hardening [5]. It inhibits grain growth athigh temperature and significantly improves the retention of temper and high-temperaturestrength, resulting in an increase in tool life. The use of Co is generally restricted to high-speedsteels, hot-forming tool steels, maraging steels, and creep-resistant and high-temperaturematerials [13,15].1.2.17 COPPERCopper (Cu) addition has a moderate tendency to segregate. Above 0.30% Cu can causeprecipitation hardening. It increases hardenability. If Cu is present in appreciable amounts,it is detrimental to hot-working operations. It is detrimental to surface quality and exaggeratesthe surface defects inherent in resulfurized steels. However, Cu improves the atmosphericcorrosion resistance (when in excess of 0.20%) and the tensile properties in alloy and low-alloysteels, and reportedly helps the adhesion of paint [6,14]. In austenitic stainless steels, a Cucontent above 1% results in improved resistance to H2 SO4 and HCl and stress corrosion [5,15].1.2.18 BORONBoron (B), in very small amounts (0.00050.0035%), has a starting effect on the hardenabilityof steels due to the strong tendency to segregate at grain boundaries. The segregation of B insteels is a nonequilibrium segregation. It also improves the hardenability of other alloyingelements. It is used as a very economical substitute for some of the more expensive elements.The beneficial effects of B are only apparent with lower- and medium-carbon steels, with noreal increase in hardenability above 0.6% C [14]. The weldability of boron-alloyed steels isanother reason for their use. However, large amounts of B result in brittle, unworkable steels. 2006 by Taylor & Francis Group, LLC.1.2.19 ZIRCONIUMZirconium (Zr) is added to killed HSLA steels to obtain improvement in inclusion characteristics, particularly sulfide inclusions, where modifications of inclusion shape improveductility in transverse bending. It increases the life of heat-conducting materials. It is also astrong carbide former and produces a contracted austenite phase field [5,15].1.2.20 LEADLead (Pb) is sometimes added (in the range of 0.20.5%) to carbon and alloy steels throughmechanical dispersion during teeming to improve machinability.1.2.21 TINTin (Sn) in relatively small amounts is harmful to steels. It has a very strong tendency to segregateat grain boundaries and induces temper embrittlement in alloy steels. It has a detrimental effecton the surface quality of continuous cast billets containing small amounts of Cu [24]. Smallamounts of Sn and Cu also decrease the hot ductility of steels in the austenite ferrite region [25].1.2.22 ANTIMONYAntimony (Sb) has a strong tendency to segregate during the freezing process, and has adetrimental effect on the surface quality of continuous cast billets. It also has a very strongtendency to segregate at grain boundaries and cause temper embrittlement in alloy steels.1.2.23 CALCIUMCalcium (Ca) is a strong deoxidizer; silicocalcium is used usually in steelmaking. The combination of Ca, Al, and Si forms low-melting oxides in steelmaking, and improves machinability.1.3 CLASSIFICATION OF STEELSSteels can be classified by different systems depending on [4,6,8]:1. Compositions, such as carbon (or nonalloy), low-alloy, and alloy steels2. Manufacturing methods, such as converter, electric furnace, or electroslag remeltingmethods3. Application or main characteristic, such as structural, tool, stainless steel, or heatresistant steels4. Finishing methods, such as hot rolling, cold rolling, casting, or controlled rolling andcontrolled cooling5. Product shape, such as bar, plate, strip, tubing, or structural shape6. Oxidation practice employed, such as rimmed, killed, semikilled, and capped steels7. Microstructure, such as ferritic, pearlitic, martensitic, and austenitic (Figure 1.1)8. Required strength level, as specified in the American Society for Testing and Materials(ASTM) standards9. Heat treatment, such as annealing, quenching and tempering, air cooling (normalization), and thermomechanical processing10. Quality descriptors and classifications, such as forging quality and commercial qualityAmong the above classification systems, chemical composition is the most widely used basisfor designation and is given due emphasis in this chapter. Classification systems based onoxidation practice, application, and quality descriptors are also briefly discussed. 2006 by Taylor & Francis Group, LLC.Ferrous alloysClassification bycommercial nameor applicationClassificationby structureSteelAlloys withouteutectic(<2% C on FeCdiagram)Plain carbonsteelLow-carbonsteel(<0.2% C)FerriticMedium-carbonsteel(0.25% C)FerriticpearliticHigh-carbonsteel(>0.5% C)PearliticLow- and mediumalloy steel10% alloyingelementsMartensiticBainiticHigh-alloysteel>10% alloyingelementsCorrosionresistantHeatresistantWearresistantAusteniticPrecipitationhardenedAusteniticferriticDuplexstructureFIGURE 1.1 Classification of steels. (Courtesy of D.M. Stefanescu, University of Alabama, Tuscaloosa,AL. Slightly modified by the present authors.)1.3.1 TYPES OF STEELS BASED ON DEOXIDATION PRACTICESteels, when cast into ingots, can be classified into four types according to the deoxidationpractice or, alternatively, by the amount of gas evolved during solidification. These four typesare called killed, semikilled, capped, and rimmed steels [6,8]. Killed SteelsKilled steel is a type of steel from which there is practically no evolution of gas during solidification of the ingot after pouring, because of the complete deoxidation, and formation of pipe inthe upper central portion of the ingot, which is later cut off and discarded. All alloy steels, most 2006 by Taylor & Francis Group, LLC.low-alloy steels, and many carbon steels are usually killed. The continuous casting billets are alsokilled. The essential quality criterion is soundness [2628]. Killed steel is characterized by ahomogeneous structure and even distribution of chemical compositions and properties.Killed steel is produced by the use of a deoxidizer such as Al and a ferroalloy of Mn or Si;However, calcium silicide and other special deoxidizers are sometimes used. SteelsGas evolution is not completely suppressed by deoxidizing additions in semikilled steel,because it is partially deoxidized. There is a greater degree of gas evolution than in killedsteel, but less than in capped or rimmed steel. An ingot skin of considerable thickness isformed before the beginning of gas evolution. A correctly deoxidized semikilled steel ingotdoes not have a pipe but does have well-scattered large blow holes in the top-center half of theingot; however, the blow holes weld shut during rolling of the ingot. Semikilled steelsgenerally have a carbon content in the range of 0.150.30%. They find a wide range of usesin structural shapes, skelp, and pipe applications. The main features of semikilled steels areUJ variable degrees of uniformity in composition, which are intermediate between those ofkilled and rimmed steels and less segregation than rimmed steel, and (2) a pronouncedtendency for positive chemical segregation at the top center of the ingot (Figure 1.2). SteelsRimmed steel is characterized by a great degree of gas evolution during solidification in themold and a marked difference in chemical composition across the section and from the top tothe bottom of the ingot (Figure 1.2). These result in the formation of an outer ingot skin orrim of relatively pure iron and an inner liquid (core) portion of the ingot with higherconcentrations of alloying and residual elements, especially C, N, S, and P, having lowermelting temperature. The higher purity zone at the surface is preserved during rolling [28].Rimmed ingots are best suited for the manufacture of many products, such as plates, sheets,wires, tubes, and shapes, where good surface or ductility is required [28].The technology of producing rimmed steels limits the maximum content of C and Mn, andthe steel does not retain any significant amount of highly oxidizable elements such as Al, Si, orTi. Rimmed steels are cheaper than killed or semikilled steels for only a small addition ofdeoxidizer is required and is formed without top scrap.12KilledSemikilled345Capped678RimmedFIGURE 1.2 Eight typical conditions of commercial steel ingots, cast in identical bottle-top molds, inrelation to the degree of suppression of gas evolution. The dotted line denotes the height to which thesteel originally was poured in each ingot mold. Based on the carbon, and more significantly, the oxygencontent of the steel, the ingot structures range from that of a completely killed ingot (No. 1) to that of aviolently rimmed ingot (No. 8). (From W.D. Landford and H.E. McGannon, Eds., The Making,Shaping, and Treating of Steel, 10th ed., U.S. Steel, Pittsburgh, PA, 1985.) 2006 by Taylor & Francis Group, LLC. Capped SteelsCapped steel is a type of steel with characteristics similar to those of a rimmed steel but to adegree intermediate between that of rimmed and semikilled steels. Less deoxidizer is used toproduce a capped ingot than to produce a semikilled ingot [29]. This induces a controlledrimming action when the ingot is cast. The gas entrapped during solidification is excess of thatrequired to counteract normal shrinkage, resulting in a tendency for the steel to rise in the mold.Capping is a variation of rimmed steel practice. The capping operation confines the timeof gas evolution and prevents the formation of an excessive number of gas voids within theingot. The capped ingot process is usually applied to steels with carbon contents greater than0.15% that are used for sheet, strip, tin plate, skelp, wire, and bars.Mechanically capped steel is poured into bottle-top molds using a heavy cast iron cap toseal the top of the ingot and to stop the rimming action [29]. Chemically capped steel is cast inopen-top molds. The capping is accomplished by the addition of Al or ferrosilicon to the topof the ingot, causing the steel at the top surface to solidify rapidly. The top portion of theingot is cropped and discarded.1.3.2 QUALITY DESCRIPTORS AND CLASSIFICATIONSQuality descriptors are names applied to various steel products to indicate that a particularproduct possesses certain characteristics that make it especially well suited for specific applications or fabrication processes. The quality designations and descriptors for various carbon steelproducts and alloy steel plates are listed in Table 1.1. Forging quality and cold extrusion qualitydescriptors for carbon steels are self-explanatory. However, others are not explicit; for example,merchant quality hot-rolled carbon steel bars are made for noncritical applications requiringmodest strength and mild bending or forming but not requiring forging or heat-treating operations. The quality classification for one steel commodity is not necessarily extended to subsequent products made from the same commodity; for example, standard quality cold-finished barsare produced from special quality hot-rolled carbon steel bars. Alloy steel plate qualities aredescribed by structural, drawing, cold working, pressure vessel, and aircraft qualities [27].The various physical and mechanical characteristics indicated by a quality descriptor resultfrom the combined effects of several factors such as (1) the degree of internal soundness, (2) therelative uniformity of chemical composition, (3) the number, size, and distribution of nonmetallic inclusions, (4) the relative freedom from harmful surface imperfections, (5) extensivetesting during manufacture, (6) the size of the discard cropped from the ingot, and (7) hardenability requirements. Control of these factors during manufacture is essential to achieve millproducts with the desired characteristics. The degree of control over these and other relatedfactors is another segment of information conveyed by the quality descriptor.Some, but not all, of the basic quality descriptors may be modified by one or moreadditional requirements as may be appropriate, namely macroetch test, special discard,restricted chemical composition, maximum incidental (residual) alloying elements, austeniticgrain size, and special hardenability. These limitations could be applied forging quality alloysteel bars but not to merchant quality bars.Understanding the various quality descriptors is difficult because most of the prerequisitesfor qualifying steel for a specific descriptor are subjective. Only limitations on chemicalcomposition ranges, residual alloying elements, nonmetallic inclusion count, austenitic grainsize, and special hardenability are quantifiable. The subjective evaluation of the other attributesdepends on the experience and the skill of the individuals who make the evaluation. Althoughthe use of these subjective quality descriptors might appear impractical and imprecise, steelproducts made to meet the requirements of a specific quality descriptor can be relied upon tohave those characteristics necessary for that product to be used in the suggested application orfabrication operation [6]. 2006 by Taylor & Francis Group, LLC.TABLE 1.1Quality Descriptionsa of Carbon and Alloy SteelsCarbon SteelsSemifinished for forgingForging qualitySpecial hardenabilitySpecial internalsoundnessNonmetallic inclusionrequirementSpecial surfaceCarbon steel structuralsectionsStructural qualityCarbon steel platesRegular qualityStructural qualityCold-drawing qualityCold-pressing qualityCold-flanging qualityForging qualityPressure vessel qualityHot-rolled carbon steelbarsMerchant qualitySpecial qualitySpecial hardenabilitySpecial internalsoundnessNonmetallic inclusionrequirementSpecial surfaceScrapless nut qualityAxle shaft qualityCold extrusion qualityCold-heading and coldforging qualityCold-finished carbon steelbarsStandard qualitySpecial hardenabilitySpecial internalsoundnessNonmetallic inclusionrequirementSpecial surfaceCold-heading and coldforging qualityCold extrusion qualityHot-rolled sheetsCommercial qualityDrawing qualityDrawing quality specialkilledStructural qualityCold-rolled sheetsCommercial qualityDrawing qualityDrawing quality specialkilledStructural qualityPorcelain enameling sheetsCommercial qualityDrawing qualityDrawing quality specialkilledLong terne sheetsCommercial qualityDrawing qualityDrawing quality specialkilledStructural qualityGalvanized sheetsCommercial qualityDrawing qualityDrawing quality specialkilledLock-forming qualityElectrolytic zinc coatedsheetsCommercial qualityDrawing qualityDrawing quality specialkilledStructural qualityHot-rolled stripCommercial qualityDrawing qualityDrawing quality specialkilledStructural qualityCold-rolled stripSpecific qualitydescriptions are not 2006 by Taylor & Francis Group, LLC.Alloy SteelsMill productsSpecific qualitydescriptions are notapplicable to tin millproductsAlloy steel platesDrawing qualityPressure vessel qualityStructural qualityAircraft physical qualityCarbon steel wireIndustrial quality wireCold extrusion wiresHeading, forging, androll-threading wiresMechanical spring wiresUpholstery springconstruction wiresWelding wireHot-rolled alloy steel barsRegular qualityAircraft quality or steelsubject to magneticparticle inspectionAxle shaft qualityBearing qualityCold-heading qualitySpecial cold-headingqualityRifle barrel quality,gun quality, shell orA.P. shot qualityCarbon steel flut wireStitching wireStapling wireCarbon steel pipeStructural tubingLine pipeOil country tubular goodsSteel specialty tubularproductsPressure tubingMechanical tubingAircraft tubingHot-rolled carbon steelwire rodsIndustrial qualityRods formanufacture ofwire intended forelectric welded chainRods for heading,forging, and rollthreading wireRods for lock washerwireRods for scrapless nutwireRods for upholsteryspring wireRods for welding wireAlloy steel wireAircraft qualityBearing qualitySpecial surface qualityCold-finished alloy steelbarsRegular qualityAircraft quality orsteel subject tomagnetic particleinspectionAxle shaft qualityBearing shaft qualityCold-heading qualitySpecial cold-headingqualityRifle barrel quality,gun quality, shell orA.P. shot qualityLine pipeOil country tubular goodsSteel specialty tubulargoodsPressure tubingMechanical tubingStainless and heatresisting pipe,pressureTABLE 1.1 (Continued )Quality Descriptionsa of Carbon and Alloy SteelsCarbon Steelsprovided in coldrolled strip becausethis product is largelyproduced for specificand useAlloy Steelstubing, andmechanical tubingAircraft tubingpipeaIn the case of certain qualities, P and S are usually finished to lower limits than the specified maximum.Source: From H. Okamoto, CFe, in Binary Alloy Phase Diagrams, 2nd ed., T.B. Massalski, Ed., ASM International,Materials Park, OH, 1990, pp. 842848.1.3.3 CLASSIFICATIONOFSTEEL BASED ON CHEMICAL COMPOSITION1.3.3.1 Carbon and CarbonManganese SteelsIn addition to carbon, plain carbon steels contain the following other elements: Mn up to1.65%, S up to 0.05%, P up to 0.04%, Si up to 0.60%, and Cu up to 0.60%. The effects of eachof these elements in plain carbon steels have been summarized in Section 1.2.Carbon steel can be classified according to various deoxidation processes (see Section 1.3.1).Deoxidation practice and steelmaking process will have an effect on the characteristics andproperties of the steel (see Section 1.2). However, variations in C content have the greatest effecton mechanical properties, with C additions leading to increased hardness and strength. As such,carbon steels are generally grouped according to their C content. In general, carbon steels containup to 2% total alloying elements and can be subdivided into low-carbon, medium-carbon, highcarbon, and ultrahigh-carbon (UHC) steels; each of these designations is discussed below.As a group, carbon steels constitute the most frequently used steel. Table 1.2 lists variousgrades of standard carbon and low-alloy steels with the Society of Automotive Engineers andAmerican Iron and Steel Institute (SAE-AISI) designations. Table 1.3 shows some representative standard carbon steel compositions with SAE-AISI and the corresponding UnifiedNumbering System (UNS) designations [6,8,30].Low-carbon steels contain up to 0.25% C. The largest category of this class is flat-rolledproducts (sheet or strip), usually in the cold-rolled or subcritical annealed condition andusually with final temper-rolling treatment. The carbon content for high formability and highdrawability steels is very low (<0.10% C) with up to 0.40% Mn. These lower carbon steels areused in automobile body panels, tin plates, appliances, and wire products.The low-carbon steels (0.100.25% C) have increased strength and hardness and reducedformability compared to the lowest carbon group. They are designated as carburizing or casehardening steels [9]. Selection of these grades for carburizing applications depends on thenature of the part, the properties required, and the processing practices preferred. An increaseof carbon content of the base steel results in greater core hardness with a given quench.However, an increase in Mn increases the hardenability of both the core and the case.A typical application for carburized plain carbon steel is for parts with hard wear-resistantsurface but without any need for increased mechanical properties in the core, e.g., smallshafts, plunges, or highly loaded gearing [8]. Rolled structural steels in the form of plates and 2006 by Taylor & Francis Group, LLC.14 2006 by Taylor & Francis Group, LLC.TABLE 1.2SAE-AISI Designation System for Carbon and Low-Alloy SteelsNumerals andDigitsType of Steel and Nominal AlloyContent (%)Carbon steels10xxa . . . . . . . . . Plain carbon (Mn 1.00 max)11xx . . . . . . . . . Resulfurized12xx . . . . . . . . . Resulfurized and rephosphorized15xx . . . . . . . . . Plain carbon (max Mn range: 1.001.65)Manganese steels13xx . . . . . . . . . Mn 1.75Nickel steels23xx . . . . . . . . . Ni 3.5025xx . . . . . . . . . Ni 5.00Molybdenum steels40xx . . . . . . . . . Mo 0.20 and 0.2544xx . . . . . . . . . Mo 0.40 and 0.52Chromiummolybdenum steels41xx . . . . . . . . . Cr 0.50, 0.80, and 0.95; Mo 0.12, 0.20,0.25, and 0.30aType of Steel and NominalAlloy Content (%)Nickelchromiummolybdenum steels43xx . . . . . . . . . Ni 1.82; Cr 0.50 and 0.80;Mo 0.2543BVxxNi 1.82; Cr 0.50; Mo 0.12 and0.25; V 0.03 min47xx . . . . . . . . . Ni 1.05; Cr 0.45; Mo 0.20 and0.3581xx . . . . . . . . . Ni 0.30; Cr 0.40; Mo 0.1286xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.2087xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.2588xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.3593xx . . . . . . . . . Ni 3.25; Cr 1.20; Mo 0.1294xx . . . . . . . . . Ni 0.45; Cr 0.40; Mo 0.1297xx . . . . . . . . . Ni 0.55; Cr 0.20; Mo 0.2098xx . . . . . . . . . Ni 1.00; Cr 0.80; Mo 0.25Nickelmolybdenum steels46xx . . . . . . . . . Ni 0.85 and 1.82; Mo 0.20 and0.2548xx . . . . . . . . . Ni 3.50; Mo 0.25Chromium steels50xx . . . . . . . . . Cr 0.27, 0.40, 0.50, and 0.6551xx . . . . . . . . . Cr 0.80, 0.87, 0.92, 0.95, 1.00, and1.05Numerals and DigitsType of Steel and Nominal AlloyContent (%)Chromium (bearing) steels)50xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 0.5051xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 1.02 min C 1.0052xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 1.45Chromiumvanadium steels61xx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 0.60, 0.80, and 0.95;V 0.10 and 0.15 minTungstenchromium steel72xx . . . . . . . . . . . . . . . . . . . . . . . . . . . W 1.75; Cr 0.75Siliconmanganese steels92xx . . . . . . . . . . . . . . . . . . . . . . . . . . . Si 1.40 and 2.00; Mn0.65, 0.82, and 0.85;High-strength low-alloy steelsCr 0 and 0.659xxBoron steels. . . . . . . . . . . . . . . . . . . . . . . . . . . Various SAE gradesxxBxxLeaded steels. . . . . . . . . . . . . . . . . . . . . . . . . . . B denotes boron steelxxLxx. . . . . . . . . . . . . . . . . . . . . . . . . . . L denotes leaded steelThe xx in the last two digits of these designations indicates that the carbon content (in hundredths of a percent) is to be inserted.Source: From Courtesy of ASM International, Materials Park, OH. With permission.Steel Heat Treatment: Metallurgy and TechnologiesNickelchromium steels31xx . . . . . . . . . Ni 1.25; Cr 0.65 and 0.8032xx . . . . . . . . . Ni 1.75; Cr 1.0733xx . . . . . . . . . Ni 3.50; Cr 1.50 and 1.5734xx . . . . . . . . . Ni 3.00; Cr 0.77Numerals andDigitsTABLE 1.3Standard Carbon Steel Compositions with SAE-AISI and Corresponding UNS DesignationsPlain Carbon Steel (Nonresulfurized, 1.0% Mn Max)aCast or Heat Chemical Ranges and Limits (%)aUNSNumberSAE-AISINumberCMnP maxS maxG10060G10080G10090G10100G10120G10150G10160G10170G10180G10190G10200G10210G10220G10230G10250G10260G10300G10330G10350G10370G10380G10390G10400G10420G10430G10450G10490G10500G10550G10600G10640G10650G10700G10740G10750G10780G10800G10840G10850G10860G10900G109501006100810091010101210151016101710181019102010211022102310251026103010331035103710381039104010421043104510491050105510601064106510701074107510781080108410851086109010950.08 max0.10 max0.15 max0. max0.50 max0.60 max0.300.600.300.600.300.600.600.900.300.600.600.900.701.000.300.600.600.900.701.000.300.600.300.600.600.900.600.900.701.000.600.900.701.000.600.900.701.000.600.900.600.900.701.000.600.900.600.900.600.900.600.900.600.900.500.800.600.900.600.900.500.800.400.700.300.600.600.900.600.900.701.000.300.500.600.900.300.500.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.050Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.3 (Continued)Standard Carbon Steel Compositions with SAE-AISI and Corresponding UNS DesignationsFree-Cutting (Resulfurized) Carbon Steel CompositionsaCast or Heat Chemical Ranges and Limits (%)UNSNumberSAE-AISINumberCMnP maxSG11080G11100G11170G11180G11370G11390G11400G11410G11440G11460G11S10110811101117111811371139114011411144114611510. Resulfurized and Rephosphorized Carbon SteelsaUNSNumberGl2110G12120G12130G12150G12144Cast or Heat Chemical Ranges and Limits, %(a)SAE-AISINumberC maxPSPb0. . . . . . . . . .1212 . . . . . . . . .1213 . . . . . . . . .1215 . . . . . . . . .12L14b . . . . . . . . .Mn0.600.900.701.000.701.000.751.050.851. Nonresulfurized Carbon Steels (Over 1.0% Manganese)UNSNumberG15130G15220G15240G15260G15270G15360G15410G15480G15510G15520G15610G15660151315221524152615271536154115481551155215611566Cast or Heat Chemical Ranges and Limits, %SAE-AISINumber....................................C.................................................................................................................................................................................... maxS max1.101.401.101.401.351.651.101.401.201.501.201.501.351.651.101.400.851.151.201.500.751.050.851.150.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.0500.050Applicable to semifinished products for forging, hot-rolled and cold-finished bars, wire rods, and seamless tubing.It is not common practice to produce the 12xx series of steels to specified limits for silicon because of its adverse effecton machinability.bContains 0.150.35% lead; other steels listed here can be produced with similar amounts of lead.Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of AutomotiveEngineers, Warrendale, PA, pp. 1.011.189.a 2006 by Taylor & Francis Group, LLC.sections containing ~0.25% C, with up to 1.5% Mn and Al are used if improved toughness isrequired. When used for stampings, forgings, seamless tubes, and boilerplate, Al addition shouldbe avoided. An important type of this category is the low-carbon free-cutting steels containing upto 0.15% C and up to 1.2% Mn, a minimum of Si and up to 0.35% S with or without 0.30% Pb.These steels are suited for automotive mass production manufacturing methods [4].Medium-carbon steels containing 0.300.55% C and 0.601.65% Mn are used wherehigher mechanical properties are desired. They are usually hardened and strengthened byheat treatment or by cold work. Low-carbon and manganese steels in this group find wideapplications for certain types of cold-formed parts that need annealing, normalizing, orquenching and tempering treatment before use. The higher carbon grades are often colddrawn to specific mechanical properties for use without heat treatment for some applications.All of these steels can be used for forgings, and their selection is dependent on the sectionsize and the mechanical properties needed after heat treatment [8]. These grades, generallyproduced as killed steels, are used for a wide range of applications that include automobileparts for body, engines, suspensions, steering, engine torque converter, and transmission [31].Some Pb or S additions make them free-cutting grades, whereas Al addition produces grainrefinement and improved toughness. In general, steels containing 0.400.60% C are used asrails, railway wheels, tires, and axles.High-carbon steels containing 0.551.00% C and 0.300.90% Mn have more restrictedapplications than the medium-carbon steels because of higher production cost and poorformability (or ductility) and weldability. High-carbon steels find applications in the springindustry (as light and thicker plat springs, laminated springs, and heavier coiled springs), farmimplement industry (as plow beams, plowshares, scraper blades, discs, mowers, knives, andharrow teeth), and high-strength wires where improved wear characteristics and higherstrength than those attainable with lower carbon grades are needed.UHC steels are experimental plain carbon steels with 1.02.1% C (1532 vol% cementite)[3234]. Optimum superplastic elongation has been found at about 1.6% C content [9]. Thesesteels have the capability of emerging as important technological materials because theyexhibit superplasticity. The superplastic behavior of these materials is attributed to thestructure consisting of uniform distribution of very fine, spherical, discontinuous particles(0:1-- :5 mm diameter) in a very fine-grained ferrite matrix (0:5-- :0 mm diameter) that can be-1-2readily achieved by any of the four thermomechanical treatment routes described elsewhere [4]. Low-Alloy SteelsAlloy steels may be defined as those steels that owe their improved properties to the presenceof one or more special elements or to the presence of large proportions of elements such asMn and Si than are ordinarily present in carbon steels [26]. Alloy steels contain Mn, Si, or Cuin quantities greater than the maximum limits (e.g., 1.65% Mn, 0.60% Si, and 0.60% Cu) ofcarbon steels, or they contain special ranges or minimums of one or more alloying elements.However, in some countries Mn, Si, or Cu as an alloy element in low-alloy and alloy steels isonly greater than 1.00% Mn, 0.50% Si, or 0.10% Cu [7].The alloying elements increase the mechanical and fabrication properties. Broadly, alloysteels can be divided into (1) low-alloy steels containing less than 5 wt% total noncarbon alloyaddition, (2) medium-alloy steels containing 510 wt% total noncarbon alloy addition, and (3)high-alloy steels with more than 10 wt% total noncarbon alloy addition. Table 1.4 lists somelow-alloy steel compositions with SAE-AISI and corresponding UNS designations.Low-alloy steels constitute a group of steels that exhibit superior mechanical propertiescompared to plain carbon steels as the result of addition of such alloying elements as Ni, Cr,and Mo. For many low-alloy steels, the main function of the alloying elements is to increase 2006 by Taylor & Francis Group, LLC.18 2006 by Taylor & Francis Group, LLC.TABLE 1.4Low-Alloy Steel Compositions Applicable to Billets, Blooms, Slabs, and Hot-Rolled and Cold-Finished Bars (Slightly Wider Rangesof Compositions Apply to Plates)Ladle Chemical Composition Limits (%)aUNSNumberSAENumberCorrespondingAISI NumberCMnPG13300G13350G13400G13450133013351340134513301335134013450.280.330.330.380.380.430.430.481.601.901.601.901.601.901.601.900.0350.0350.0350.035G40230G40240G40270G40280G40320G40370G40420G40470402340244027402840324037404240474023402440274028403740470. Heat Treatment: Metallurgy and TechnologiesSi46154617462046264615462046260. max0. Nomenclature 2006 by Taylor & Francis Group, LLC.G46150G46170G46200G4626019Continued20 2006 by Taylor & Francis Group, LLC.TABLE 1.4 (Continued)Low-Alloy Steel Compositions Applicable to Billets, Blooms, Slabs, and Hot-Rolled and Cold-Finished Bars (Slightly Wider Rangesof Compositions Apply to Plates)Ladle Chemical Composition Limits (%)aSAENumberCorrespondingAISI NumberCMnPSSiNiCrMoVG61180G6150061186150611861500.160.210.480.530.500.700.700.900.0350.0350.0400.0400.150.350.150.350.500.700.801. minG81150G81451G86150G86170G86200G86220G86250G86270G86300G86370G86400811581B45c861586178620862286258627863086378640811581B458615861786208622862586278630863786400.130.180.430.480. Heat Treatment: Metallurgy and TechnologiesUNSNumber8642864586B45c865086558660G87200G8740087208740G88220864286450.400.450.430.480.430.480.480.530.510.590.560.640.751.000.751.000.751.000.751.000.751.000.751.000.0350.0350.0350.0350.0350.0350.0400.0400.0400.0400.0400.0400.150.350.150.350.150.350.150.350.150.350.150.350.400.700.400.700.400.700.400.700.400.700.400.700.400.600.400.600.400.600.400.600.400.600.400.600. Nomenclature 2006 by Taylor & Francis Group, LLC.G86420G86450G86451G86500G86550G866008655aSmall quantities of certain elements that are not specified or required may be found in alloy steels. These elements are to be considered as incidental and are acceptable to thefollowing maximum amount, copper to 0.35%, nickel to 0.25%, chromium to 0.20%, and molybdenum to 0.06%.bElectric furnace steel.cBoron content is 0.00050.003%.Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. hardenability in order to optimize the strength and toughness after heat treatment. Insome instances, however, alloying elements are used to reduce environmental degradationunder certain specified conditions.Low-alloy steels can be classified according to: (1) chemical composition such as nickelsteels, nickelchromium steels, molybdenum steels, chromiummolybdenum steels, and soforth, based on the principal alloying elements present and as described in Table 1.2, (2) heattreatment such as quenched and tempered, normalized and tempered, annealed and so on,and (3) weldability.Because of the large variety of chemical compositions possible and the fact that somesteels are employed in more than one heat-treated conditions some overlap exists among thelow-alloy steel classifications. However, these grades can be divided into four major groupssuch as (1) low-carbon quenched and tempered (QT) steels, (2) medium-carbon ultrahighstrength steels, (3) bearing steels, and (4) heat-resistant CrMo steels (see Table 1.5).Low-carbon QT steels (also called low-carbon martensitic steels) are characterized byrelatively high yield strength with a minimum yield strength of 690 MPa (100 ksi) and goodnotch toughness, ductility, corrosion resistance, or weldability. The chemical compositions oflow-carbon QT steels are listed in Table 1.5. These steels are not included in SAE-AISIclassification. However, they are covered by ASTM designations, and a few steels, such asHY-80 and HY-100, are included in military (MIL) specifications. The steels listed areprimarily available in the form of plate, sheet, bar, structural shape, or forged products.They are extensively used for a wide variety of applications such as pressure vessels, earthmoving, and mining equipment and as major members of large steel structures. They are alsoused for cold-headed and cold-forged parts as fasteners or pins and heat-treated to the desiredproperties [26].Medium-carbon ultrahigh-strength steels are structural steels with very high strength.These steels exhibit a minimum yield strength of 1380 MPa (200 ksi). Table 1.5 lists typicalcompositions such as SAE-AISI 4130, high-strength 4140, deeper hardening higher-strength4340, 300M (a modification of 4340 steel with increased Si content (1.6%) to raise the thirdtransformation temperature of tempering and prevent temper embrittlement of martensite)and Ladish D-6a and Ladish D-6ac steels (another modification of 4340 with grain refiner Vand higher C, Cr, and Mo contents, developed for aircraft and missile structural applications). Other less prominent steels that may be included in this family are SAE-AISI 6150 steel(a tough shock-resistant, shallow-hardening CrV steel with high fatigue and impact resistance in the heat-treated conditions) and 8640 steel (an oil-hardening steel exhibiting properties similar to those of 4340 steel) [35]. Product forms include billet, bar, rod, forgings, plate,sheet, tubing, and welding wire.These steels are used for gears, aircraft landing gear, airframe parts, pressure vessels, bolts, springs, screws, axles, studs, fasteners, machinery parts, connecting rods, crankshafts, piston rods, oil well drilling bits, high-pressure tubing, flanges, wrenches, sprockets,etc. [35].Bearing steels used for ball and roller bearing applications comprise low-carbon (0.100.20%C) case-hardened steels and high-carbon (~1% C) through-hardened or surface-inductionhardened steels (see Table 1.5). Many of these steels are covered by SAE-AISI designations.Chromiummolybdenum heat-resistant steels contain 0.59% Cr, 0.51.0% Mo, andusually less than 0.20% C. They are ordinarily supplied in the normalized and tempered,quenched and tempered, or annealed condition. CrMo steels are extensively used in oilrefineries, oil and gas industries, chemical industries, electric power generating stations andfossil fuel and nuclear power plants for piping, heat exchangers, superheater tubes, andpressure vessels. Various product shapes and corresponding ASTM specifications for thesesteels are provided in Table 1.6. Nominal chemical compositions are given in Table 1.7. 2006 by Taylor & Francis Group, LLC.TABLE 1.5Chemical Compositions for Typical Low-Alloy SteelsSteelCSiMnPSNiCrMoOtherLow-carbon quenched and tempered steelsA 514/A 517 grade A0.150.210.400.800.801.100.0350.040.500.800.180.28A 514/A 517 grade F0. 514/A 517 grade RA 533 type AA 533 type CHY-800. Znb0.0025 B0.030.08 V0.150.50 Cu0.00050.005 B0.030.08 V0.25 Cu0.03 V0.02 Ti0.25 Cu0.03 V0.02 TiMedium-carbon ultrahigh-strength steels41304340300MD-6a0.280.330.380.430.400.460.420.480.200.350.200.351.451.800.150.300.400.600.600.800.650.900.600.901.652.001.652.000.400.700.801.100.700.900.700.950.901. bearing steels4118512033100. bearing steels52100A 485 grade 1A 485 grade 30.981.100.901.050.951.100.150.300.450.750.150.350.250.450.951.250.650.900.0250.0250.0250.0250.0250.0250.250.251.301.600.901.201.101.500.100.200.30Steel Nomenclature 2006 by Taylor & Francis Group, LLC.Composition, wt%a0.35 Cu0.35 Cu0.05 V min0.050.10 VaSingle values represent the maximum allowable.Zirconium may be replaced by cerium. When cerium is added, the cerium/sulfur ratio should be approximately 1.5:1, based on heat analysis.Source: From Anon., ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140194.b23TABLE 1.6ASTM Specifications for ChromiumMolybdenum Steel Product FormsTypeTubesPipeCastingsPlateA 182-F2A 387-Gr21Cr1/2MoA 182-F12A 336-F12A 335-P2A 369-FP2A 426-CP2A 335-P12A 369-FP12A 387-Gr1211/4Cr1/2MoA 182-F11A 336-F11/F11AA 541-C11CA 182-F22/F22aA 336-F22/F22AA 541-C22C/22DA 182-F21A 336-F21/F21AA 217-WC6A 356-Gr6A 389-C23A 217-WC9A 356-Gr10A 387-Gr11A 387-Gr211Forgings1/2Cr /2Mo21/4Cr1Mo3Cr1Mo3Cr1MoV5Cr1/2Mo5Cr1/2MoSiA 182-F21bA 182-F5/F5aA 336-F5/F5AA 473-501/5025Cr1/2MoTi7Cr1/2MoA 182-F7A 473-501A9Cr1MoA 182-F9A 336-F9A 473-501BA 426-CP12A 199-T11A 200-T11A 213-T11A 199-T22A 200-T22A 213-T22A 199-T21A 200-T21A 213-T21A 199-T5A 200-T5A 213-T5A 213-T5bA 213-T5cA 199-T7A 200-T7A 213-T7A 199-T9A 200-T9A 213-T9A 335-P11A 369-FP11A 426-CP11A 335-P22A 369-FP22A 426-CP22A 335-P21A 369-FP21A 426-CP21A 335-P5A 369-FP5A 426-CP5A 335-P5bA 426-CP5bA 335-P5cA 335-P7A 369-FP7A 426-CP7A 335-P9A 369-FP9A 426-CP9A 217-C5A 387-Gr22A 542A 387-Gr5A 387-Gr7A 217-C12A 387-Gr9Source: From Anon., ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140194. Low-Alloy SteelsA general description of HSLA steel is as that containing: (1) low carbon (0.030.25%)content to obtain good toughness, formability, and weldability, (2) one or more of the strongcarbide-forming microalloying elements (MAEs) (e.g., V, Nb, or Ti), (3) a group of solidsolution strengthening elements (e.g., Mn up to 2.0% and Si), and (4) one or more of theadditional MAEs (e.g., Ca, Zr) and the rare earth elements, particularly Ce and La, for sulfideinclusion shape control and increasing toughness [4,5,21,22,36,37]. In many other HSLAsteels, small amounts of Ni, Cr, Cu, and particularly Mo are also present, which increaseatmospheric corrosion resistance and hardenability. A very fine ferrite grain structure in thefinal product produced by a combination of controlled rolling and controlled cooling with anoptimum utilization of microalloying additions, in HSLA steels, is an important factor insimultaneously increasing strength and toughness and decreasing the ductilebrittle transitiontemperature (to as low as 708C). Carbides (NbC, VC, TiC), nitrides (TiN, NbN, AlN), andcarbonitrides (e.g., V(C,N), Nb(C,N), (Nb,V) CN, (Nb,Ti) CN) are the dispersed second-phaseparticles that act as grain size refiners or dispersive strengthening phases in HSLA steels. 2006 by Taylor & Francis Group, LLC.Steel Nomenclature 2006 by Taylor & Francis Group, LLC.TABLE 1.7Nominal Chemical Compositions for Heat-Resistant ChromiumMolybdenum SteelsComposition (%)aType1/2Cr1/2Mo1Cr1/2Mo11/4Cr1/2Mo11/4Cr1/2Mo21/4Cr1Mo3Cr1Mo3Cr1MoVb5Cr1/2Mo7Cr1/2Mo9Cr1Mo9Cr1MoVcUNS DesignationCMnSPSiCrMoK12122K11562K11597K11592K21590K31545K31830K41545K61595K909410. 650.450.650.871.130.801.060.901.100.450.650.450.650.901.100.851.05aSingle values are maximums.Also contains 0.020.030% V, 0.0010.003% B, and 0.0150.035% Ti.cAlso contains 0.40% Ni, 0.180.25% V, 0.060.10% Nb, 0.030.07% N, and 0.04% Al.Source: From Anon., ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140194.b25HSLA steels are successfully used as ship, plate, bar, structural sections, and forged barproducts, and find applications in several diverse fields such as oil and gas pipelines; in theautomotive, agricultural, and pressure vessel industries, in offshore structures and platformsand in the constructions of crane, bridges, buildings, shipbuildings, railroad, tank cars, andpower transmission and TV towers [36]. Classification of HSLA SteelsSeveral special terms are used to describe various types of HSLA steels [3739]:1. Weathering steels: Steels containing ~0.1% C, 0.20.5% Cu, 0.51.0% Mn, 0.050.15%P, 0.150.90% Si, and sometimes containing Cr and Ni, exhibiting superior atmospheric corrosion resistance. Typical applications include railroad cars, bridges, andunpainted buildings.2. Control-rolled steels: Steels designated to develop a highly deformed austenite structureby hot rolling (according to a predetermined rolling schedule) that will transform to avery fine equiaxed ferrite structure on cooling.3. Pearlite-reduced steels: Steels strengthened by very fine-grained ferrite and precipitation hardening but with low carbon content, and therefore exhibiting little or nopearlite in the microstructure.4. Microalloyed steels: Conventional HSLA steels containing V, Ti, or Nb, as definedabove. They exhibit discontinuous yielding behavior.5. Acicular ferrite steels: Very low-carbon (typically 0.030.06%) steels with enoughhardenability (by Mn, Mo, Nb, and B additions) to transform on cooling to a veryfine, high-strength acicular ferrite structure rather than the usual polygonal ferritestructure. In addition to high strength and good toughness, these steels have continuous yielding behavior.6. Low-carbon bainite steels: Steels are strengthened by bainite, with very fine grains andprecipitations. They contain 0.020.2% C, 0.61.6% Mn, 0.30.6% Mo, and MAEs(such as V, Nb, Ti, and B), usually containing 0.40.7% Cr. The yield strength of thesesteels is higher than 490 MPa, with good toughness [5].7. Low-carbon martensite steels: Steels are strengthened by martensite with high hardenability (by addition of Mo, Mn, Cr, Nb, and B) and fine grains (by Nb addition).These steels contain 0.050.25% C, 1.52.0% Mn, 0.200.50 Mo, and MAEs (such asNb, Ti, V, and B). Some steels containing small amounts of Ni, Cr, and Cu, afterrolling or forging, and directly quenching and tempering attain a low-carbon martensite structure with high yield strength (7601100 MPa), high toughness (CVN 50130 J),and superior fatigue strength [5,40,41].8. Dual-phase steels: Steels comprising essentially fine dispersion of hard strong martensitebut sometimes also retained austenite or even bainite in a soft and fine-grained ferritematrix. The volume function of martensite is about 2030%. Steels are characterized bycontinuous yielding (i.e., no yield point elongation), low yield stress (the YS/UTS ratiobeing around 0.50), high UTS, superior formability, and rapid initial work-hardeningrate. Additionally, they possess greater resistance to onset of necking (i.e., plasticinstability) in the uniaxial sheet material forming process to provide large uniform strain[4245]. Table 1.8 lists HSLA steels according to chemical composition and minimummachining property requirements. 2006 by Taylor & Francis Group, LLC.TABLE 1.8Composition Ranges and Limits for SAE HSLA SteelsHeat Composition Limits (%)aSAEDesignationb942X945A945C945X950A950B950C950D950X955X960X965X970X980XC maxMn maxP max0. contents of sulfur and silicon for all grades: 0.050% S, 0.90% Si.Second and third digits of designation indicate minimum yield strength in ksi. Suffix X indicatesthat the steel contains niobium, vanadium, nitrogen, or other alloying elements. A second suffix Kindicates that the steel is produced fully killed using fine-grain practice; otherwise, the steel isproduced semikilled.Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, MaterialsSociety of Automotive Engineers, Warrendale, PA, pp. 1.011.189.b1.3.3.4 Tool SteelsA tool steel is any steel used to shape other metals by cutting, forming, machining, battering,or die casting or to shape and cut wood, paper, rock, or concrete. Hence tool steels aredesigned to have high hardness and durability under severe service conditions. They comprisea wide range from plain carbon steels with up to 1.2% C without appreciable amounts ofalloying elements to the highly alloyed steels in which alloying additions reach 50%. Althoughsome carbon tool steels and low-alloy tool steels have a wide range of carbon content, most ofthe higher alloy tool steels have a comparatively narrow carbon range. A mixed classificationsystem is used to classify tool steels based on the use, composition, special mechanicalproperties, or method of heat treatment.According to AISI specification, there are nine main groups of wrought tool steels. Table 1.9lists the compositions of these tool steels with corresponding designated symbols [46], which arediscussed herein.High-speed steels are used for applications requiring long life at relatively high operatingtemperatures such as for heavy cuts or high-speed machining. High-speed steels are the mostimportant alloy tool steels because of their very high hardness and good wear assistance in theheat-treated condition and their ability to retain high hardness and the elevated temperaturesoften encountered during the operation of the tool at high cutting speeds. This red- or hothardness property is an important feature of a high-speed steel [47,48]. 2006 by Taylor & Francis Group, LLC.28 2006 by Taylor & Francis Group, LLC.TABLE 1.9Composition Limits of Principal Types of Tool SteelsCompositiona (%)DesignationAlSIUNSCSiCrNiMoWVMolybdenum high-speed steelsM1T113010.780.88M2T113020.780.88; 0.951.05M3 class 1T113131.001.10M3 class 2T113231.15125M4T113041.251.40M7T113070.971.05M10T113100.840.94; 0.951.05M30T113300.750.85M33T113330.850.92M34T113340.850.92M35T113350.820.88M36T113360.800.90M41T113411.051.15M42T113421.051.15M43T113431.151.25M44T113441.101.20M46T113461.221.30M47T113471.051.15M48T113481.421.52M62T113621.251.350.150.400.150.400.150.400.150.400.150.400.150.400.100.400.150.400.150.400.150.400.150.400.150.400.200.600.150.400.200.400.200.400.200.400.150.400.150.400.150.400.200.500.200.450.200.450.200.450.200.450.200.550.200.450.200.450.150.500.200.450.200.450.200.450.150.500.150.650.150.650.300.550.400.650.200.450.150.400.150.403.504.003.754.503.754.503.754.503.754.753.504.003.754.503.504.253.504.003.504.003.754.503.754.503.754.503.504.253.504.254.004.753.704.203.504.003.504.003.504.000.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max8.209.204.505.504.756.504.756.504.255.508.209.207.758.507.759.009.0010.007.759.204.505.504.505.503.254.259.0010.007.508.506. high-speed steelsT1T12001T2T12002T4T12004T5T12005T6T12006T8T12008T15T120150.100.400.200.400.100.400.200.400.200.400.260.400.150.400.200.400.200.400.200.400.200.400.200.400.200.400.150.403.754.503.754.503.754.503.755.004.004.753.754.503.755.000.30 max0.30 max0.30 max0.30 max0.30 max0.30 max0.30 max1.00 max0.401.000.501.250.401.000.401.001.00 max17.2518.7517.5019.0017.5019.0017.5019.0018.5021.0013.2514.7511.7513.000.901.301.802.400.801.201.802.401.502.101.802.404.505.250.650.800.800.900.700.800.750.850.750.850.750.851.501.60Co4.505.507.758.757.758.754.505.507.758.754.755.757.758.757.758.7511.0012.257.808.804.755.258.0010.004.255.757.009.5011.0013.004.255.754.755.25Steel Heat Treatment: Metallurgy and TechnologiesMn0.200.600.200.603.754.503.504.300.30 max0.30 max3.904.754.004.900.751.500.801.251.652.25Chromium hot-work steelsH10T20810H11T20811H12T20812H13T20813H14T20814H19T208190.350.450.330.430.300.400.320.450.350.450.320.450.250.700.200.500.200.500.200.500.200.500.200.500.801.200.801.200.801.200.801.200.801.200.200.503.003.754.755.504.755.504.755.504.755.504.004.750.30 max0.30 max0.30 max0.30 max0.30 max0.30 max2. max0.801.201.752.20Tungsten hot-work steelsH21T20821H22T20822H23T20823H24T20824H25T20825H26T208260.260.360.300.400.250.350.420.530.220.320.450.55b0.150.400.150.400.150.400.150.400.150.400.150.400.150.500.150.400.150.600.150.400.150.400.150.403.003.751.753.7511.0012.752.503.503.754.503.754.500.30 max0.30 max0.30 max0.30 max0.30 max0.30 max8.5010.0010.0011.7511.0012.7514.0016.0014.0016.0017.2519.000.300.600.250.500.751.250.400.600.400.600.751.25Molybdenum hot-work steelsH42T208420.550.70b0.150.403.754.500.30 max4.505.505.506.751.752.20Air-hardening, medium-alloy, cold-work steelsA2T301020.951.05A3T301031.201.30A4T301040.951.05A6T301060.650.75A7T301072.002.85A8T301080.500.60A9T301090.450.55Al0T301101.251.50c1.00 max0.400.601.802.201.802.500.80 max0.50 max0.50 max1.602.100.50 max0.50 max0.50 max0.50 max0.50 max0.751.100.951.151.001.504.755.504.755.500.902.200.901.205.005.754.755.504.755.500.30 max0.30 max0.30 max0.30 max0.30 max0.30 max1.251.751.552.050.901.400.901.400.901.400.901.400.901.401.151.651.301.801.251.750.501.501.001.500.150.500.801.403.905.150.801.40High-carbon, high-chromium, cold-work steelsD2T304021.401.60D3T304032.002.35D4T304042.052.40D5T304051.401.60D7T304O72.152.500.60 max0.60 max0.60 max0.60 max0.60 max0.60 max0.60 max0.60 max0.60 max0.60 max11.0013.0011.0013.5011.0013.0011.0013.0011.5013.500.30 max0.30 max0.30 max0.30 max0.30 max0.701.200.701.200.701.200.701.201.00 max1.10 max1.00 max1.00 max1.00 max3.804.404.004.502.503.50Continued290.150.450.150.45Steel Nomenclature 2006 by Taylor & Francis Group, LLC.Intermediate high-speed steelsM50T113500.780.88M52T113520.850.9530 2006 by Taylor & Francis Group, LLC.TABLE 1.9 (Continued )Composition Limits of Principal Types of Tool SteelsCompositiona (%)DesignationAlSIUNSCSiCrNiMoWVCoOil-hardening cold-work steelsO1T315010.851.00O2T315020.850.95O6T315061.251.55cO7T315071.101.301.001.401.401.800.301.101.00 max0.50 max0.50 max0.551.500.60 max0.400.600.50 max0.30 max0.350.850.30 max0.30 max0.30 max0.30 max0.30 max0.200.300.30 max0.400.601.002.000.30 max0.30 max0.40 maxShock-resisting steelsS1T41901S2T41902S5T41905S6T41906S7T419070.100.400.300.500.601.001.201.500.200.900.151.200.901.201.752.252.002.500. max1.201.503.003.500.30 max0.30 max0.50 max0.300.600.201.350.300.501.301.801.503.000.150.300.50 max0.35 max0.200.400.200.30d0.100.900.250.800.50 max0.50 max0.701.200.601. max0.50 max0.100.300.200.30d0.400.550.400.550.500.650.400.500.450.55Low-alloy special-purpose tool steelsL2T612020.451.00bL6T612060.650.75Steel Heat Treatment: Metallurgy and TechnologiesMnSteel Nomenclature 2006 by Taylor & Francis Group, LLC.Low-carbon mold steelsP2T51602P3T51603P4T51604P5T51605P6T51606P20T51620P21T516210.10 max0.10 max0.12 max0.10 max0. tool steelsW1T723010.701.50eW2T723020.851.50eW5T723051. max0.100.400.40 max0.100.400.200.800.200.400.751.250.400.754. max0.104.501.001.500.35 max3.253.753.904.250.150.400.401.000.300.550. max0.15 max0.400.600.20 max0.20 max0.20 max0.10 max0.10 max0.10 max0.15 max0.15 max0.15 max0.10 max0.150.350.10 maxaAll steels except group W contain 0.25 max Cu, 0.03 max P, and 0.03 max S; group W steels contain 0.20 max Cu, 0.025 max P, and 0.025 max S. Where specified, sulfur may beincreased to 0.06 to 0.15% to improve machinability of group A, D, H, M, and T steels.bAvailable in several carbon ranges.cContains free graphite in the microstructure.dOptional.eSpecified carbon ranges are designated by suffix numbers.Source: From A.M. Bayer and L.R. Walton, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials, Park, OH, 1990, pp. 757779.31High-speed steels are grouped into molybdenum type M and tungsten type T. Type Mtool steels contain Mo, W, Cr, V, Mo, and C as the major alloying elements, while type T toolsteels contain W, Cr, V, Mo, Co, and C as the main alloying elements. In the United States,type M steels account for 95% of the high-speed steels produced. There is also a subgroupconsisting of intermediate high-speed steels in the M group. The most popular grades amongmolybdenum types are M1, M2, M4, M7, M10, and M42, while those among tungsten typesare T1 and T15.The main advantage of type M steels is their lower initial cost (approximately 40% cheaperthan that of similar type T steels), but they are more susceptible to decarburizing, therebynecessitating better temperature control than type T steels. By using salt baths and sometimessurface coatings, decarburization can be controlled. The mechanical properties of type M andtype T steels are similar except that type M steels have slightly greater toughness than type Tsteels at the same hardness level [4].Hot-work tool steels (AISI series) fall into three major groups: (1) chromium-base,types H1H19, (2) tungsten-base, types H20H39, and (3) molybdenum-base, typesH40H59. The distinction is based on the principal alloying additions; however, all classeshave medium carbon content and Cr content varying from 1.75 to 12.75%. Among thesesteels, H11, H12, H13 are produced in large quantities. These steels possess good redhardness and retain high hardness (~50 Rc) after prolonged exposures at 5005508C. Theyare used extensively for hot-work applications, which include parts for aluminum andmagnesium die casting and extrusion, plastic injection molding, and compression and transfermolds [47].Cold-work tool steels comprise three categories: (1) air-hardening, medium-alloy toolsteels (AISI A series), (2) high-chromium tool steels (AISI D series), and (3) oil-hardeningtool steels (AISI O series). AISI A series tool steels have high hardenability and harden readilyon air cooling. In the air-hardened and tempered condition, they are suitable for applicationswhere improved toughness and reasonably good abrasion resistance are required such as forforming, blanking, and drawing dies. The most popular grade is A2. AISI D series tool steelspossess excellent wear resistance and nondeforming properties, thereby making them veryuseful as cold-work die steels. They find applications in blanking and cold-forming dies,drawing and lamination dies, thread-rolling dies, shear and slitter blades, forming rolls, andso forth. Among these steels, D2 is by far the most popular grade [47]. AISI O series toolsteels are used for blanking, coining, drawing, and forming dies and punches, shear blades,gauges, and chuck jaws after oil quenching and tempering [47]. Among these grades, O1 is themost widely used.Shock-resisting tool steels (AISI S series) are used where repetitive impact stresses areencountered such as in hammers, chipping and cold chisels, rivet sets, punches, driver bits,stamps, and shear blades in quenched and tempered conditions. In these steels, high toughness is the major concern and hardness the secondary concern. Among these grades, S5 andS7 are perhaps the most widely used.Low-alloy special-purpose tool steels (AISI series) are similar in composition to the Wtype tool steels, except that the addition of Cr and other elements render greater hardenabilityand wear-resistance properties, type L6 and the low-carbon version of L2 are commonly usedfor a large number of machine parts.Mold steels (AISI P series) are mostly used in low-temperature die casting dies and inmolds for the injection or compression molding of plastics [46].Water-hardening tool steels (AISI W series): Among the three compositions listed, W1 is themost widely used as cutting tools, punches, dies, files, reamers, taps, drills, razors, woodworking tools, and surgical instruments in the quenched and tempered condition. 2006 by Taylor & Francis Group, LLC. Stainless SteelsStainless steels may be defined as complex alloy steels containing a minimum of 10.5% Crwith or without other elements to produce austenitic, ferritic, duplex (ferriticaustenitic),martensitic, and precipitation-hardening grades. AISI uses a three-digit code for stainless steels.Table 1.10 and Table 1.11 list the compositions of standard and nonstandard stainless steels,respectively, with the corresponding designated symbols, which are discussed below [49].Austenitic stainless steels constitute about 6570% of the total U.S. stainless steel production and have occupied a dominant position because of their higher corrosion resistancesuch as strength and toughness at both elevated and ambient temperatures, excellent cryogenic properties, esthetic appeal, and varying specific combination and properties that can beobtained by different compositions within the group [50].In general, austenitic stainless steels are FeCrNiC and FeCrMnNiN alloys containing 1626% Cr, 0.7519.0% Mn, 140% Ni, 0.030.35% C, and sufficient N to stabilizeaustenite at room and elevated temperatures. The 2xx series (CrMnNi) steels contain N,5.515.5% Mn, and up to 6% Ni, the 3xx (CrNi) types contain higher amounts of Ni and upto 2% Mn. Mo, Cu, and Si may be added to increase corrosion resistance. Ti and Nb may beadded to decrease the sensitivity of intergranular corrosion. The addition of Mo and N mayincrease halide-pitting resistance; Si and Cu may be added to increase resistance to stresscorrosion cracking. S and Se may be added to certain series to enhance machinability.Nitrogen is added to increase yield strength.Broadly, austenitic stainless steels can be classified into ten groups [4,51]. These classifications are not straightforward because of the overlapping effects.Ferrite stainless steels contain essentially 10.530% Cr with additions of Mn and Si andoccasionally Mo, Ni, Al, Ti, or Nb to confer particular characteristics. As they remain ferriticat room and elevated temperatures, they cannot be hardened by heat treatment. The ductilebrittle transition temperature of ferrite stainless steels is higher than room temperature; ifC N < 0:0015 wt%, the transition temperature can be kept well below the room temperature. These extra-low interstitial ferritic stainless steels have good ductility and toughness.The yield strength of ferritic stainless steels in the annealed condition is usually in the range275415 MPa (4060 ksi). They are used because of their good ductility, good resistance togeneral liquid corrosion, and high-temperature oxidation, resistance to pitting and stresscorrosion cracking, and generally lower cost than the austenitic grades [10]. As in the ferriticgrade, S and Se may be added to improve machinability.The standard ferrite stainless steels are types 405, 409, 429, 430, 430F, 430FSe, 434, 436,439, 444, and 446 (Table 1.10). In addition, high-quality ferrite stainless steels are typesE-Brite 26-1, MoNiT (25-4-4), AL29-4c, and AL29-4-2 (Table 1.11).Duplex stainless steels contain 1829% Cr, 2.58.5% Ni, and 14% Mo, up to 2.5% Mn, upto 2% Si, and up to 0.35% N. They possess a mixed structure of ferrite and austenite.The volume fractions of ferrite and austenite vary between 0.3 and 0.7 in a duplex structure.The ratio of the ferrite and austenitic determines the properties of duplex stainless steels. Theyield strength increases with increasing ferrite content. The ultimate tensile strength rises to amaximum at 7080% ferrite, then decreases as the ferrite goes to 100%. Compared to austeniticgrades, they can offer improved yield strength (about two to three times greater) and greaterresistance to stress corrosion cracking, but the deep drawability is less than austenitic grades.Compared to ferritic grades, they can provide improved toughness, formability, and weldability. The duplex stainless steels can be embrittled due to the formability a0 and s phases. Ingeneral, duplex stainless steels cannot be used in the temperature range from 300 to 9508C.Types AISI 329 and Carpenter 7-Mo and 7-Mo-Plus (UNS S32950) are the more popularduplex steels (Table 1.10 and Table 1.11). The new type SAF2507 contains ultralow carbon 2006 by Taylor & Francis Group, LLC.34 2006 by Taylor & Francis Group, LLC.TABLE 1.10Compositions of Standard Stainless SteelsTypeUNSDesignationAustenitic types201202205301302302B303303Se304304H304L304LN302Cu304N305308309309S310310S314316316F316H316L316LN316NComposition (%)aMnSiCrNiPSOtherS20100S20200S20500S30100S30200S30215S30300S30323S30400S30409S30403S30453S30430S30451S30500S30800S30900S30908S31000S31008S31400S31600S31620S31609S31603S316530. min0. min0.030.030.03S316510.082.001.0016. N0.25 N0.320.40 N0.6 Mob0.15 min Se0.100.16 N3.04.0 Cu0.100.16 N2.03.0 Mo1.752.5 Mo2.03.0 Mo2.03.0 Mo2.03.0 Mo;0.100.16N2.03.0 Mo;0.100.16 NSteel Heat Treatment: Metallurgy and TechnologiesCS31700S31703S32100S32109N08330S34700S34709S348000. types405409S40500S409000. min0.060.030.03439S430350.071.001.0017.019.00.500.040.03442444S44200S444000.20.0251. Mo3.04.0 Mo5 %C min Ti5 %C min Ti10 %C min Nb8 %C min1.0 max Nb0.2 Co; 10 %C min Nb;0.10 Ta0.2 Co; 8 %C min1.0 max Nb; 0.10 TaSteel Nomenclature 2006 by Taylor & Francis Group, LLC.317317L321321H330347347H3480.100.30 A16 %C min0.75 max Ti0.6 Mob0.15 min Se0.751.25 Mo0.751.25 Mo;5 %C min0.70 max Nb0.15 Al; 12 %C min1.10 Ti1.752.50 Mo;0.025 N; 0.2 4 (%C %N) min0.8 max (Ti Nb)0.25 NContinued35TypeUNSDesignation36 2006 by Taylor & Francis Group, LLC.TABLE 1.10 (Continued)Compositions of Standard Stainless SteelsComposition (%)aCMnSiCrNiPSOther0.201.000.7523.028.02.505. MoMartensitic types403410414416416Se420420F422S40300S41000S41400S41600S41623S42000S42020S422000. min0.15 min0. min0.060.030.15 min0.03431440A440B440CS43100S44002S44003S440040.200.600.750.750.950.951. Mob0.15 min Se0.6 Mob0.751.25 Mo;0.751.25 W;0.150.3 V0.75 Mo0.75 Mo0.75 MoPrecipitation-hardening typesPH13-8MoS138000.050.200.1012.2513. values are maximum values unless otherwise indicated.Optional.Source: From S.D. Washko and G. Aggen, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 841907.b2.02.5 Mo;0.901.35 Al;0.01 N2.54.5 Cu;0.150.45 Nb3.05.0 Cu;0.150.45 Nb0.751.5 AlSteel Heat Treatment: Metallurgy and TechnologiesDuplex (ferriticaustenitic) type329S32900Composition (%)bUNSDesignationCMaSiAustenitic stainless steelsGall-Tough203 EZ (XM-II)S20161S203000. 50 (XM-19)S209100. (XM-31)Cryogenic Tenelon(XM-14)Esshete 1250S214000.1214.516.0S21460S215000.120.15Type 216 (XM-17)S21600Type 216 L (XM-18)DesignationaCrNiPS4. 60Nitronic 40 (XM-10)21-6-9 LCNitronic 33 (18-3-Mn)Nitronic 32 (18-2-Mn)18-18 PlusS21800S21900S21904S24000S24100S282000. Plus X (XM-5)MVMAc304BldS30310S30415S304240. min0.030OtherSteel Nomenclature 2006 by Taylor & Francis Group, LLC.TABLE 1.11Compositions of Nonstandard Stainless Steels0.080.20 N0.5 Mo;1.752.25 Cu1.53.0 Mo;0.20.4 N;0.10.3 Nb;0.10.3 V0.35 N0.350.50 N0.0030.009 B;0.751.25 Nb;0.150.40 V2.03.0 Mo;0.250.50 N2.03.0 Mo;0.250.50 N0.080.18 N0.150.40 N0.150.40 N0.200.40 N0.200.45 N0.51.5 Mo; 0.51.5Cu; 0.40.6 N0.6 Mo0.15 N; 0.04 Ce0.10 N; 1.001.25 B37Continued38 2006 by Taylor & Francis Group, LLC.TABLE 1.11 (Continued)Compositions of Nonstandard Stainless SteelsComposition (%)bUNSDesignationC304 HN (XM-21)Cronifer 1815 LCSiRA 85 Hc253 MAS30452S30600S30615S308150.040.100.0180.200.050.l02.002.000.800.801. 309 S CbS309400.082.001.00Type 310 CbS310400.082.00254 SMOS312540.020Type 316 TiS31635Type 316 CbDesignationaMaSiCrNiSOther8.010.514.015.514.5010.012.00.0450.0200.0400.0300.0200.03022. 316 HQ0.0302.001.0016.0018.2510.0014.000.0300.015Type 317 LM17-14-4 LNS31725S317260. 317 LNType 370S31753S370000. N0.2 Mo1.0 Al0.140.20 N;0.030.08 Ce;1.0 Al10 %C min to1.10 max Nbl0 %C min to1.10 max Nb Ta6.006.50 Mo;0.501.00 Cu;0.1800.220 N5 %(C N) min to0.70 max Ti;2.03.0 Mo;0.10 N10 %C min to1.10 max Nb Ta; 2.03.0 Mo;0.10 N3.004.00 Cu;2.003.00 Mo4.05.0 Mo; 0.10 N4.05.0 Mo;0.100.20 N0.100.22 N1.52.5 Mo;Steel Heat Treatment: Metallurgy and TechnologiesPS38100S631980.080.280.352.000.751.501. 28N080280.022.001.0026.028.029.532.50.0200.015AL-6XAL-6XNN08366N083670.0350.0302. 332N088000.011.501.0019. 1925 hMoN089250.021.000.5024. Nomenclature 2006 by Taylor & Francis Group, LLC.18-18-2 (XM-15)19-9 DL0.10.4 Ti;0.005 N; 0.05 Co1.01.75 Mo;0.10.35 Ti;1.01.75 W;0.250.60 Nb2.03.0 Mo;3.04.0 Cu;8 %C min to1.00 max Nb3.505.00 Mo;0.501.50 Cu;0.150.35 Nb5.006.70 Mo;2.004.00 Cu3.04.0 Mo;0.61.4 Cu6.07.0 Mo6.007.00 Mo;0.180.25 N4.35.0 Mo;8 %C min to0.5 max Nb;0.5 Cu; 0.005 Pb;0.035 S0.150.60 Ti;0.150.60 Al4.05.0 Mo:1.02.0 Cu6.07.0 Mo;0.81.5 Cu;0.100.20 N40 2006 by Taylor & Francis Group, LLC.TABLE 1.11 (Continued )Compositions of Nonstandard Stainless SteelsComposition (%)bDesignationaUNSDesignationCronifer 2328CMaSiCrNiPSOther0.750.7522. Cu;0.40.7 Ti;2.53.0 Mo1.52.5 Mo5 %C min to0.75 max Ti0.3 9 (%C) minto 0.90 max Nb;0.10.5 Ti; 0.03 N0.751.5 Mo;0.050.2 Nb;0.015 N; 0.2 Cu3.54.5 Mo; 0.2 4 (%C %N) minto 0.8 max (Ti Nb); 0.035 N2.53.5 Mo; 0.2 4(%C %N) minto 0.8 max (Ti Nb); 0.035 N3.604.20 Mo;0.201.00 Ti Nband 6 (%C %N)min Ti Nb;0.045 N3.54.2 Mo; 0.15 Cu;0.02 N; 0.025 max(%C %N)Ferritic stainless steels18-2 FM (XM-34)Type 430 TiS18200S430360. min0.030Type 441S441000.031.001.0017.519.51.000.0400.040E-Brite 26-1S446270.010.400.4025.027.00.500.0200.020MoNiT (25-4-4)S446350.0251.000.7524.526. (SC-1)S446600.0251.001.0025. Heat Treatment: Metallurgy and Technologies0. IVSealmet 10.301. stainless steels44LNS312000.0302.001.0024.026.05.506.500.0450.030DP-3S312600.0301.000.7524.026.05.507.500.0300.0303RE602205S31500S318030.0300.0301. 50S324040. 255S325500.041.501.0024.027.04.506.500.040.037-Mo-PlUSS329500.032.000.6026.029.03.505.200.0350.010Martensitic stainless steelsType 410 SS41008Type 410 Cb (XM-30)S410400. A1; 0.4 Ti1.2 Al; 0.3 Ti2.754.25 Al; 0.6 Ti0.751.25 Al;0.650.75 Nb;0.30.5 Ti; 0.03 N4.755.25 Al;0.0050.035 Ce;0.03 N0.04 NSteel Nomenclature 2006 by Taylor & Francis Group, LLC.18 SR(c)12 SR(c)406408 Cb1.202.00 Mo;0.140.20 N2.503.50 Mo;0.200.80 Cu;0.100.30 N;0.100.50 W2.503.00 Mo2.503.50 Mo;0.080.20 N0.050.60 Mo;0.050.60 Cu;0.050.20 N2.03.0 Mo;1.02.0 Cu; 0.20 N2.004.00 Mo;1.502.50 Cu;0.100.25 N1.002.50 Mo;0.150.35 N0.050.20 Nb41Continued42 2006 by Taylor & Francis Group, LLC.TABLE 1.11 (Continued)Compositions of Nonstandard Stainless SteelsComposition (%)bDesignationaUNSDesignationCMaSiCrNiPSS41050S41500S416100. min0.030S41800S42010S420230. 440 FType 440 FSeS44020S440230.951.200.951. N0.51.0 Mo0.6 Mo2.53.5 W0.401.00 Mo0.15 min Se; 0.6 Zr;0.6 Cu2.53.0 Mo;0.20.3 V0.08 N0.15 min Se;0.60 MoSteel Heat Treatment: Metallurgy and TechnologiesE4CA6NM416 Plus X (XM-6)Type 418 (GreekAscolloy)TrimRiteType 420 FSeOtherSteel Nomenclature 2006 by Taylor & Francis Group, LLC.Precipitation-hardening stainless steelsPH14-4MoS148000.051.001.0013.7515.07.758.750.0150.010PH15-7Mo (Type 632)S157000.091.001.0014. (Type 633)S350000. (Type 634)S355000. 450 (XM-25)S450000.051.001.0014. 455 (XM-16)S455000.050.500.5011. Mo;0.751.50 Al2.03.0 Mo;0.751.5 Al2.53.25 Mo;0.070.13 N2.53.25 Mo;0.070.13 N1.251.75 Cu;0.51.0 Mo;8 %C min Nb1.52.5 Cu;0.81.4 Ti;0.10.5 Nb;0.5 MoaXM designations in this column are ASTM designations for the listed alloy.Single values are maximum values unless otherwise indicated.cNominal compositions.dUNS designation has not been specified. This designation appears in ASTM A 887 and merely indicates the form to be used.Source: From S.D. Washko and G. Aggen, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 841907.b43content (0.010.02% C), high Mo content (about 4% Mo), and high N content (about 0.3%N). This new type duplex steel gives the excellent resistance to pitting corrosion.Duplex stainless steels find applications as welded pipe products for handling wet and dryCO2 and sour gas and oil products in the petrochemical industry, as welded tubing for heatexchanges, for handling chloride-containing coolants, and for handling hot brines andorganic chemicals in the chemical, electric, and other industries [4].Martensitic stainless steels contain 11.518% Cr, 0.081.20% C, and other alloying elements less than 2 to 3%. They can be hardened and tempered to yield strength in the range of5501900 MPa (80275 ksi). The Cr content provides these steels with such high hardenabilitythat they can be air hardened even in large sections. If they are to be heat treated formaximum strength, the amount of d-ferrite should be minimized [10].The standard martensitic grades are types 403, 410, 414, 416, 416Se, 420, 422, 431, 440A,440B, and 440C (Table 1.10). They are used in manifold stud bolts, heat control shafts, steamvalves, Bourdon tubes, gun mounts, water pump parts, carburetor parts, wire cutter blades,garden shears, cutlery, paint spray nozzles, glass and plastic molds, bomb shackle parts, drivescrews, aircraft bolting, cable terminals, diesel engine pump parts, instrument parts, crankshaft counterweight pins, valve trim, ball bearings, and races.PH stainless steels are high-strength alloys with appreciable ductility and goodcorrosion resistance that are developed by a simple heat treatment comprising martensiteformation and low-temperature aging (or tempering) treatment; the latter heat treatmentstep may be applied after fabrication. PH stainless steels can have a matrix structure ofeither austenite or martensite. Alloy elements added to form precipitates are Mo, Cu, Al,Ti, Nb, and N. PH stainless steels may be divided into three broad groups: (1)martensitic type, (2) semiaustenitic type, and (3) austenitic type (Table 1.10 and Table1.11). A majority of these steels are classified by a three-digit number in the AISI 400series or by a five-digit UNS designation. However, most of them are better known bytheir trade names or their manufacturer. All steels are available in sheet, strip, plate, bar,and wire.Martensitic PH stainless steels (also called single-treatment alloys) are most widely usedand include 17-4PH (AISI 430 or UNS S17400), stainless W (AISI 635 or UNS S17400), 155PH (UNS S15500), PH13-8Mo (UNS S13800), and Custom 450 (UNS S45000). These steelshave a predominantly austenitic structure at the solution-annealing temperature, but theyundergo an austenitic-to-martensite transformation during cooling to room temperature.These steels can be readily welded [49].Semiaustenitic PH stainless steels (also called double-treatment alloys) were developed forincreased formability before the hardening treatment. Important alloys are 17-7PH (UNSS17700) and PH15-7Mo (UNS S15700). These alloys are completely austenite in theas-quenched condition after solution annealing (which displays good toughness and ductilityin the cold-forming operations), and eventually martensite can be obtained by conditioningtreatment or thermomechanical treatment. Ultrahigh strength can be obtained in these steelsby combinations of cold working and aging.Austenitic PH stainless steels possess austenitic structures in both the solution annealedand aged conditions. The most important steels in this class include A-286 (AISI 600 or UNSS66286), 17-10P, and 1417CuMo alloys. Of these grades, A-286 is the most extensively usedin the aerospace applications. SteelsMaraging steels are a specific class of carbon-free (or small amounts) ultrahigh-strengthsteels that derive their strength not from carbon but from precipitation of intermetallic 2006 by Taylor & Francis Group, LLC.TABLE 1.12Nominal Compositions of Commercial Maraging SteelsComposition (%)aGradeStandard grades18Ni(200)18Ni(250)18Ni(300)18Ni(350)18Ni(Cast)12-5-3(180)cNi181818181712Cobalt-free and low-cobalt bearing gradesCobalt-free 18Ni(200)18.5Cobalt-free 18Ni(250)18.5Low-cobalt 18Ni(250)18.5Cobalt-free 18Ni(300)18.5MoCoTiAlNb3. grades contain no more than 0.03% C.Some producers use a combination of 4.8% Mo and 1.4% Ti, nominal.cContains 5% Cr.Source: From K. Rohrbach and M. Schmidt, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park,OH, 1990, pp. 793800.bcompounds and martensitic transformation [5,5052]. The commonly available maragingsteels contain 1019% Ni, 018% Co, 314% Mo, 0.21.6% Ti, 0.10.2% Al, and someintermetallic compounds are Ni3 Ti, Ni3 Mo, Fe2 Mo, etc. Since these steels develop veryhigh strength by martensitic transformation and subsequent age-hardening, they are termedmaraging steels [53].There are four types of maraging steels, namely 200, 250, 300, and 350; the number refers tothe ultimate tensile strength in ksi (kpsi). The tensile strength is based on the Ti content, whichvaries between 0.2 and 1.85%. Table 1.12 lists the compositions of these grades [54]. In thesegrades, C content is maintained at a very low level (<0.03%); the sum of Si and Mn is lower(0.2%); and P and S contents are also very small (<0.005 and <0.008%, respectively) [4].Maraging steels have found applications where lightweight structures with ultrahighstrength and high toughness are essential and cost is not a major concern. Maraging steelshave been extensively used in two general types of applications:1. Aerospace and aircraft industry for critical components such as missile cases, load cells,helicopter flexible drive shafts, jet engine drive shafts, and landing gear2. Tool manufacturing industries for stub shafts, flexible drive shafts, splined shafts,springs, plastic molds, hot-forging dies, aluminum and zinc die casting dies, coldheading dies and cases, diesel fuel pump pins, router bits, clutch disks, gears in themachine tools, carbide die holders, autofrettage equipment, etc.1.4 DESIGNATIONS FOR STEELSA designation is the specific identification of each grade, type, or class of steel by a number,letter, symbol, name, or suitable combination thereof unique to a certain steel. It is used in a 2006 by Taylor & Francis Group, LLC.specific document as well as in a particular country. In the steel industries, these terms havevery specific uses: grade is used to describe chemical composition; type is used to denotedeoxidation practice; and class is used to indicate some other attributes such as tensilestrength level or surface quality [8].In ASTM specifications, however, these terms are used somewhat interchangeably.For example, in ASTM A 434, grade identifies chemical composition and class indicatestensile properties. In ASTM A 515, grade describes strength level; the maximum carboncontent allowed by the specification is dependent on both the plate thickness and the strengthlevel. In ASTM A 533, type indicates chemical analysis, while class denotes strength level. InASTM A 302, grade identifies requirements for both chemical composition and tensileproperties. ASTM A 514 and A 517 are specifications for high-strength quenched andtempered alloy steel plate for structural and pressure vessel applications, respectively; eachhas a number of grades for identifying the chemical composition that is capable of developingthe required mechanical properties. However, all grades of both designations have the samecomposition limits.By far the most widely used basis for classification and designation of steels is the chemicalcomposition. The most commonly used system of designating carbon and alloy steels in theUnited States is that of the AISI and SAE numerical designations. The UNS is also increasingly employed. Other designations used in the specialized fields include Aerospace MaterialsSpecification (AMS) and American Petroleum Institute (API) designation. These designationsystems are discussed below.1.4.1SAE-AISI DESIGNATIONSAs stated above, the SAE-AISI system is the most widely used designation for carbonand alloy steels. The SAE-AISI system is applied to semifinished forgings, hot-rolled andcold-finished bars, wire rod, seamless tubular goods, structural shapes, plates, sheet, strip,and welded tubing. Table 1.2 lists the SAE-AISI system of numerical designations for bothcarbon and low-alloy steels. and Alloy SteelsWith few exceptions, the SAE-AISI system uses a four-digit number to designate carbon andalloy steels that is specific for chemical composition ranges. Certain types of alloy steels aredesignated by five digits (numerals). Table 1.2 shows an abbreviated listing of four-digitdesignations of the SAE-AISI carbon and alloy steels. The first digit, 1, of this designationindicates a carbon steel; i.e., carbon steels comprise 1xxx groups in the SAE-AISI system andare subdivided into four series due to the variance in certain fundamental properties amongthem. Thus, the plain carbon steels comprise 10xx series (containing 1.00% Mn maximum);resulfurized carbon steels comprise the 11xx series; resulfurized and rephosphorized carbonsteels comprise the 12xx series; and nonresulfurized high-manganese (up to 1.65%) carbonsteels are produced for applications requiring good machinability.Carbon and alloy steel designations showing the letter B inserted between the second andthird digits indicate that the steel has 0.00050.003% boron. Likewise, the letter L insertedbetween the second and third digits indicates that the steel has 0.150.35% lead for enhancedmachinability. Sometimes the prefix M is used for merchant quality steels and the suffix His used to comply with specific hardenability requirements. In alloy steels, the prefix letter E isused to designate steels that are produced by the electric furnace process.The major alloying element in an alloy steel is indicated by the first two digits of thedesignation (Table 1.2). Thus, a first digit of 2 denotes a nickel steel; 3, a nickelchromium 2006 by Taylor & Francis Group, LLC.steel; 4, a molybdenum, chromiummolybdenum, nickelmolybdenum, or nickelchromiummolybdenum steel; 5, a chromium steel; 6, a chromiumvanadium steel; 7, a tungstenchromium steel; 8, a nickelchromiummolybdenum steel; and 9, a siliconmanganese steelor a nickelchromiummolybdenum steel. In the case of a simple alloy steel, the second digitrepresents the approximate percentage of the predominant alloying element. For example,2520 grade indicates a nickel steel of approximately 5% Ni (and 0.2% carbon).The last two digits of four-numeral designations and the last three digits of five-numeraldesignations indicate the approximate carbon content of the allowable carbon range inhundredths of a percent. For example, 1020 steel indicates a plain carbon steel with anapproximate mean of 0.20% carbon, varying within acceptable carbon limits of 0.18 and0.23%. Similarly, 4340 steels are NiCrMo steels and contain an approximate mean of 0.40%carbon, varying within an allowable carbon range of 0.380.43%, and 51100 steel is achromium steel with an approximate mean of 1.00% carbon, varying within an acceptablecarbon range of 0.981.10% [4,30,55].Potential standard steels are listed in SAE J1081 and Table 1.13. They are experimentalsteels to which no regular AISI-SAE designations have been assigned. The numbers consist ofthe prefix PS followed by a sequential number starting with 1. Some were developed tominimize the amount of nickel and others to enhance a particular attribute of a standardgrade of alloy steel [30]. HSLA SteelsSeveral grades of HSLA steels have been described in the SAE Recommended Practice J410.Their chemical composition and minimum mechanical property requirements are provided inTable 1.8 [30]. Formerly Listed SAE SteelsA number of grades of carbon and alloy steels have been excluded from the list of standardSAE steels because of their inadequate applications. A detailed list of formerly used SAEcarbon and alloy steels is given in SAE J1249, and producers of these steels should becontacted for their availability.1.4.2 UNS DESIGNATIONSThe UNS has been developed by the ASTM E 527, the SAE J1086, and several other technicalsocieties, trade associations, and U.S. government agencies [29]. A UNS number, which is adesignation of chemical composition and not a specification, is assigned to each chemicalcomposition of the standard carbon and alloy steel grades for which controlling limits havebeen established by the SAE-AISI [26,30,56].The UNS designation consists of a single-letter prefix followed by five numerals (digits).The letters denote the broad class of alloys; the numerals define specific alloys within thatclass. The prefix letter G signifies standard grades of carbon and alloy steels; the prefix letterH indicates standard grades that meet certain hardenability requirement limits (SAE-AISI Hsteels); the prefix T includes tool steels, wrought and cast; the prefix letter S relates to heatand corrosion-resistant steels (including stainless steel), valve steels, and iron-base superalloys; the prefix letter J is used for cast steels (except tool steels); the prefix letter K identifiesmiscellaneous steels and ferrous alloys; and the prefix W denotes welding filler metals (forexample, W00001W59999 series represent a wide variety of steel compositions) [56]. The firstfour digits of the UNS number usually correspond to the standard SAE-AISI designations,while the last digit (except zero) of the five-numeral series denotes some additional 2006 by Taylor & Francis Group, LLC.48 2006 by Taylor & Francis Group, LLC.TABLE 1.13SAE Potential Standard Steel CompositionsLadle Chemical Composition Limits (wt %)aSAE PS NumberMnP max0. max0.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.0400.040SiNi0.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.350.200.400.701.000.701.000.701.000.20 minCr0.250.400.400.600.400.600.400.600.400.600.400.600.400.600.400.600.450.650.450.650.450.650.450.650.20 min0.400.600.450.65MoB0. min0. Heat Treatment: Metallurgy and TechnologiesPS 10PS 15PS 16PS 17PS 18PS 19PS 20PS 21PS 24PS 30PS 31PS 32PS 33bPS 34PS 36CSteel Nomenclature 2006 by Taylor & Francis Group, LLC.PSPSPSPSPSPSPSPSPSPSPSPSPSPSPS3839405455565758596163646566c670.430.480.480.530.510.590. max0. max1.001.301.001.301.001.300.751.101.001.301.001.300.400.700.801.200.0350.0350.0350.0350.0350.0350.0400.0350.0350.0350.0350.0350.0350.0350.0350.0400.0400.0400.0400.0400.0400.150.350.0400.0400.0400.0400.0400.0400.0400.0400.150.350.150.350.150.350.150.350.150.350.150.351.00 max0.150.350.150.350.150.350.150.350.150.350.150.350.150.350.150.351.652.001.652.001.652.000.450.650.450.650.450.650.400.700.450.650.450.6517.0019.000.450.650.700.900.700.900.450.650.700.900.700.900.450.750.851. min0.650.800.650.801.752. PS steels may be supplied to a hardenability requirement.Supplied to a hardenability requirement of 15 HRC points within the range of 2343 HRC at J4 (4/16 in. distance from quenched end), subject to agreement between producer anduser.cPS 66 has a vanadium content of 0.100.15%.Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 1.011.189.b49composition requirements, such as boron, lead, or nonstandard chemical ranges. Table 1.3and Table 1.4 list the UNS numbers corresponding to SAE-AISI numbers for variousstandard carbon and alloy steels, respectively, with composition ranges.1.5SPECIFICATIONS FOR STEELSA specification is typically an acronym or abbreviation for a standards organization plus aspecific written statement of both technical and commercial requirements that a product mustsatisfy. It is a document that restrains or controls procurement and is issued by that standardsorganization. All material specifications contain general and specific information [57]. Anyreasonably adequate specification will furnish the information about the items stated below [6,8].The scope of the document may include product classification, required size range,condition, and any comments on product processing considered helpful to either the supplieror the user. An informative title and a statement of the required form may be employedinstead of a scope item.Chemical composition may be described, or it may be denoted by a well-known designation based on chemical composition. The SAE-AISI designations are normally used.A quality statement covering any appropriate quality descriptor and whatever additionalprerequisites might be necessary. It may also include the type of steel and the steelmakingprocesses allowed.Quantitative requirements recognize permissible composition ranges and all physical andmechanical properties necessary to characterize the material. Testing methods employed tocheck these properties should also be included or reference made to standard test methods.This section should only address those properties that are vital for the intended application.Additional requirements can cover surface preparation, special tolerances, and edge finishon flat-rolled products as well as special packaging, identification, and loading instructions.Engineering societies, trade associations, and institutes whose members make, specify, orpurchase steel products publish standard specifications; many of them are well recognizedand highly respected. Some of the notable specification-writing groups or standard organizations in the United States are listed below. It is clear from these names that a particularspecification-writing group is limited to its own specialized field.OrganizationAssociation of American RailroadsAmerican Bureau of ShipbuildingAerospace Materials Specification (of SAE)American National Standards InstituteAmerican Petroleum InstituteAmerican Railway Engineering AssociationAmerican Society of Mechanical EngineersAmerican Society for Testing and MaterialsAmerican Welding SocietySociety of Automotive Engineers1.5.1AcronymAARABSAMSANSIAPIAREAASMEASTMAWSSAEASTM (ASME) SPECIFICATIONSThe most widely used standard specifications for steel in the United States are those publishedby ASTM, many of which are complete specifications, usually adequate for procurement 2006 by Taylor & Francis Group, LLC.purposes. These specifications frequently apply to specific products, which are usuallyoriented toward the performance of the fabricated end product. They begin with theprefix ASTM, followed by letter A, identifying a ferrous material, then a numberindicating the actual specification, which may be followed by letters or numbers subdividingthe material by analysis. The AISI code is sometimes used for this purpose. Finally the yearof origin is mentioned. A letter T after this denotes a tentative specification. Generally,each specification includes a steel in a specific form or for a special purpose rather than byanalysis.ASTM specifications represent a consensus drawn from producers, specifiers, fabricators,and users of steel mill products. In many cases, the dimensions, tolerances, limits, andrestrictions in the ASTM specifications are the same as the corresponding items of thestandard practices in the AISI steel product manuals. Many of the ASTM specificationshave been adopted by the American Society of Mechanical Engineers (ASME) with slight orno modifications. ASME uses the prefix S with the ASTM specifications; for example, ASMESA 213 and ASTM A 213 are the same.Steel products can be distinguished by the ASTM specification number, which denotestheir method of production. Sometimes, citing the ASTM specification is not sufficient tocompletely identify a steel product. For example, A 434 is a specification used for heat-treated(hardened and tempered) alloy steel bars. To fully identify steel bars indicated by thisspecification, the grade/AISI-SAE designation and class (the required strength level) mustalso be quoted. The ASTM specification A 434 also covers, by reference, two standards fortest methods (A 370 for mechanical testing and E 112 for grain size determination) and A 29specifying general requirements for bar products.SAE-AISI designations for the chemical compositions of carbon and alloy steels aresometimes included in the ASTM specifications for bars, wires, and billets for forging.Some ASTM specifications for sheet products incorporate SAE-AISI designations for chemical composition. ASTM specifications for plates and structural shapes normally specifythe limits and ranges of chemical composition directly without the SAE-AISI designations.Table 1.14 incorporates a list of some ASTM specifications that include SAE-AISI designations for compositions of different steel grades.1.5.2 AMS SPECIFICATIONSAMS, published by SAE, are procurement documents, not design specifications. The majorityof the AMS pertain to materials intended for aerospace applications. These specificationsgenerally include mechanical property requirements and limits that are significantly moresevere than those for materials or steel grades with identical compositions but meant fornonaerospace applications. Their compliance will ensure procurement of a specific form andcondition or a specific material (or steel grade) or process. Table 1.15 and Table 1.16 show theAMS designations of carbon and alloy steels, respectively, indicating the chemical composition, title of specification (covering specific form, chemical composition, process, and condition), and equivalent UNS number, nearest proprietary or AISI-SAE grade, and similarMIL or federal (FED) specifications [58].1.5.3 MILITARYANDFEDERAL SPECIFICATIONSMIL specifications and standards are produced and adopted by the U.S. Department ofDefense. MIL specifications are used to define materials, products, and services. MILstandards provide procedures for design, manufacturing, and testing instead of giving onlya particular material description. MIL specifications begin with the prefix MIL, followed by a 2006 by Taylor & Francis Group, LLC.52 2006 by Taylor & Francis Group, LLC.TABLE 1.14ASTM Specifications That Cover SAE-AISI DesignationsA 29A 108A 295A 304A 322A 331A 434A 506A 507A 510A 534A 535A 544A 545A 546A 547A 548A 549A 575A 576Carbon steel wire rods and coarseround wireCarburizing steels for antifrictionbearingsSpecial quality ball and rollerbearing steelScrapless nut quality carbon steelwireCold-heading quality carbon steelwire for machine screwsCold-heading quality medium highcarbon steel wire forhexagon-head boltsCold-heading quality alloy steelwire for hexagon-head boltsCold-heading quality carbon steelwire for tapping or sheet metalscrewsCold-heading quality carbon steelwire for wood screwsMerchant quality hot-rolledcarbon steel barsSpecial quality hot-rolled carbonsteel barsA 646A 659A 682A 684A 689A 711A 713A 752A 827A 829A 830Premium quality alloy steel bloomsand billets for aircraft andaerospace forgingsCommercial quality hot-rolledcarbon steel sheet and stripCold-rolled spring quality carbonsteel strip, genericUntempered cold-rolledhigh-carbon steel stripCarbon and alloy steel bars forspringsCarbon and alloy steel blooms,billets, and slabs for forgingHigh-carbon spring steel wire forheat-treated componentsAlloy steel wire rods and coarseround wireCarbon steel plates for forging andsimilar applicationsStructural quality alloy steel platesStructural quality carbon steelplatesSource: From Anon., Carbon and alloy steels, SAE J411, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 2.012.04.Steel Heat Treatment: Metallurgy and TechnologiesA 505Carbon and alloy steel bars, hotrolled and cold finishedStandard quality cold-finishedcarbon steel barsHigh carbonchromium ball androller bearing steelAlloy steel bars havinghardenability requirementsHot-rolled alloy steel barsCold-finished alloy steel barsHot-rolled or cold-finishedquenched and tempered alloysteel barsHot-rolled and cold-rolled alloysteel sheet and stripRegular quality hot-rolled andcold-rolled alloy steel sheet andstripDrawing quality hot-rolled andcold-rolled alloy steel sheet andstripTABLE 1.15AMS Number, Title of Specification, and Equivalent UNS Number, Proprietary/AISI-SAEAlloy, and Similar Specification for Wrought Carbon SteelsAMS No.5100H5020CTitle of Specification5030F5031CBars, screw stock, free machining, cold drawnBars, forgings, and tubing, 1.5Mn 0.25Pb(0.320.39C), free cuttingBars, forgings, and tubing, 0.140.20C, free cuttingBars, forgings, and tubing, 1.5Mn (0.320.39C), freecuttingWire, welding, 1.05Cr 0.55Ni 1.0Mo 0.07V(0.260.32C), vacuum melted, environmentcontrolled packagingWire, welding, 1.05Cr 0.55Ni 1.0Mo 0.07V(0.340.40C), vacuum melted, environmentcontrolled packagingWire, welding, 0.78Cr l.8Ni 0.35Mo 0.20V(0.330.38C), vacuum melted, environmentcontrolled packagingWire, welding, 0.06 carbon maximumWelding electrodes, covered, steel, 0.070.15C5032E5036G5040J5022L5024F5027C5028B5029B5042J5044G5045F5046A5047D5050J5053G5060F5061D5062E5069E5070G5075E5077E5080HUNS No.AlloyG12120G113741212111.37G11170G1137011171137K24728D6ACK23725SimilarSpecificationD6ACK23577K00606W06013S6013Wire, 0.180.23C, annealedSheet and strip, aluminum coated, low carbonG102001020Sheet and strip, 0.15 carbon maximum, deep drawinggradeSheet and strip, 0.15 carbon maximum, forminggradeSheet and strip, 0.15 carbon maximum, half hardtemperSheet and strip, 0.25 carbon maximum, hard temperSheet, strip, and plate, annealedG101001010G101001010G101001010G10200G10200G10250G101001020102010251010G10100G10100G10150K00802K02508101010101015G10180G10220G10250101810221025MIL-T-5066G102501025MIL-T-5066G103501035Sheet and strip, 0.080.13C, Al killed, deep-forminggradeTubing, seamless, 0.15 carbon maximum, annealedTubing, welded, 0.13 carbon maximum, annealedBars, forgings, and tubing, 0.130.18CBars and wire, low carbonBars, forgings, tubing, sheet, strip, and plate, lowcarbonBars, forgings, and tubing, 0.150.20CBars and forgings, 0.180.23CTubing, seamless, 0.220.28C, cold drawn and stressrelievedTubing, welded, 0.220.28C, normalized or stressrelievedBars, forgings, and tubing, 0.310.38CFED-QQ-E-450,Type 6013FED-QQ-W-461MIL-S-4174,Type 1,Grade BMIL-S-7952Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.15 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Proprietary/AISI-SAEAlloy, and Similar Specification for Wrought Carbon SteelsAMS No.UNS No.AlloyG10350G10500G10800G10900G107001070G10740G10950G109501074109510955132GBars, 0.901.30CG10950SimilarSpecification10351050108010905120J5121G5122GTubing, seamless, 0.310.38C, stress relievedSheet, strip, and plate, 0.470.55C, annealedWire, carbon, spring temper, cold drawn, 0.750.88CWire, spring quality music wire, 0.701.00C, colddrawnWire, valve spring quality, 0.600.75C, hardened andtemperedStrip, 0.680.80CSheet and strip, 0.901.40CStrip, 0.90l.04C, hard temper10955082E5085D5110F5112J5115GTitle of SpecificationMIL-S-7947MIL-S-7947,hard temperSource: From Specification for Drill Pipe, API Specification 5D, 3rd ed., August 1, 1992, American Petroleum Institute,Washington, D.C.code letter that represents the first letter of the title for the item, followed by hyphen and thenthe serial numbers or digits. Some examples of MIL specifications for steels with corresponding AMS numbers, UNS numbers, and nearest proprietary or AISI-SAE grades are listed inTable 1.15 and Table 1.16.Federal (QQ) specifications are identical to the MIL, except that they are provided by theGeneral Services Administration (GSA) and are used by federal agencies as well as by MILestablishments when there are no separate MIL specifications available. Federal specifications begin with the prefix FED-QQ, followed by the letter and code numbers. Examples offederal specifications for steels with equivalent UNS numbers in parentheses are FED-QQ-S700 (C10300); FED-QQ-S-700 (C1085) (G10850); FED-QQ-S-763 (309) (S30900); and FEDQQ-S-766 (316L) (S31603) [56] (see Table 1.16).1.5.4 API SPECIFICATIONSThe API fosters the development standards, codes, and safe practices within the petroleumindustry. The API standard appears with the prefix API before the specification. For example,API Spec 5D covers all grades of seamless drill pipe (for use in drilling and producingoperations), process of manufacture, chemical composition and mechanical property requirements, testing and inspection methods, and requirements for dimensions, weights, and lengths[59]. API Spec 5L covers all grades of seamless and welded steel line pipe and requirements fordimensions, weight, lengths, strengths, threaded ends, plain ends, belled ends, and threadprotectors, and testing and inspection methods. This specification includes A25, A, B, X42,X46, X52, X56, X60, X65, X70, and X80 grades, and grades intermediate to grade X42 andhigher. It provides the standards for pipe suitable for use in conveying gas, water, and oil onboth the oil and natural gas industries [60]. API Spec 5LC covers seamless, centrifugal cast, andwelded corrosion-resistant alloy line pipe (austenitic stainless steels, martensitic stainless steels, 2006 by Taylor & Francis Group, LLC.TABLE 1.16AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6250H6255A6256A62576260L6263H6264G6265H6266G6267D6270L6272H6274L6275F6276FTitle of SpecificationUNS No.AlloyBars, forgings, and tubing, 1.5Cr3.5Ni (0.070.13C)Bars, forgings, and tubing, 1.1Si1.45Cr 1.0Mo 0.08A1 (0.160.22C), premium air quality,double vacuum meltedBars, forgings, and tubing, 1.0Cr3.0Ni 4.5Mo 0.08A1 0.38V(0.100.16C), premium airquality, double vacuum meltedBars, forgings, and tubing, 1.6Si0.82Cr 1.8Ni 0.40Mo 0.08V(0.400.44C), consumableElectrode vacuum remelted,normalized and temperedBars, forgings, and tubing,carburizing grade, 1.2Cr 3.2Ni0.12Mo (0.070.13C)Bars, forgings, and tubing,carburizing grade, 1.2Cr 3.2Ni0.12Mo (0.110.17C)Bars, forgings, and tubing,carburizing grade, 3.2Ni 1.2Cr0.12Mo (0.140.20C)Bars, forgings, and tubing, 1.2Cr3.25Ni (0.070.13C), vacuumconsumable electrode remeltedBars, forgings, and tubing, 0.50Cr1.82Ni 0.25Mo 0.003B 0.06V(0.080.13C)Bars, forgings, and tubing, 1.2Cr3.25Ni 0.12Mo (0.070.13C),electroslag remelted or vacuumremelted, consumableelectrodeBars, forgings, and tubing, 0.5Cr0.55Ni 0.20Mo (0.110.17C)Bars, forgings, and tubing, 0.50Cr0.55Ni 0.20Mo (0.150.20C)Bars, forgings, and tubing, 0.50Cr0.55Ni 0.20Mo (0.180.23C)Bars, forgings, and tubing, 0.40Cr0.45Ni 0.12Mo 0.002B (0.150.20C)Bars, forgings, and tubing, 0.50Cr0.55Ni 0.20Mo (0.180.23C),consumable electrode vacuummeltedK449103310K21940CBS 600K71350CBS 1000MG931069310G931509315K444149317G931069310K2102843BV12G931069310G861508615G861708617G862008620G9417194B17G86200SimilarSpecification8620MIL-S-7393,Composition IContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6277D6278A6280H6281G6282G6290F6292F6294F6299C6300C6302E6303E6304GMAM63046305BAlloySimilarSpecificationTitle of SpecificationUNS No.Bars, forgings, and tubing, 0.50Cr0.55Ni 0.20Mo (0.180.23C),vacuum arc or electroslagremeltedBars, forgings, and tubing, 4.1Cr3.4Ni 4.2Mo 1.2V (0.110.15C),premium aircraft quality forbearing applications, doublevacuum meltedBars, forgings, and rings, 0.50Cr0.55Ni 0.20Mo (0.280.33C)Tubing, mechanical, 0.50Cr 0.55Ni0.20Mo (0.280.33C)Tubing, mechanical, 0.50Cr 0.55Ni0.25Mo (0.330.38C)Bars and forgings, carburizinggrade, 1.8Ni 0.25Mo(0.110.17C)Bars and forgings, carburizinggrade, 1.8Ni 0.25Mo(0.140.20C)Bars and forgings, carburizinggrade, 1.8Ni 0.25Mo(0.170.22C)Bars, forgings, and tubing, 0.50Cr1.8Ni 0.25Mo (0.170.23C)Bars and forgings, 0.25Mo(0.350.40C)Bars, forgings, and tubing, lowalloy, heat resistant, 0.65Si1.25Cr 0.50Mo 0.25V(0.280.33C)Bars and forgings, low alloy, heatresistant, 0.65Si 1.25Cr 0.50Mo0.85V (0.250.30C)Bars, forgings, and tubing, lowalloy, heat resistant, 0.95Cr0.55Mo 0.30V (0.400.50C)G862008620G863008630G863008630G873508735G461504615G461704617G462004620H432004320HG403704037K2301517-22A(S)K2277017-22A(V)K1467517-22AMIL-S-24502Bars, forgings, and tubing, lowalloy, heat resistant, 0.95Cr0.55Mo 0.30V (0.400.50C)Bars, forgings, and tubing, lowalloy, heat resistant, 0.95Cr0.55Mo 0.30V (0.400.50C),vacuum arc remeltedK1467517-22AMIL-S-24502K146751722AMIL-S-6050MIL-S-7493,Composition4615MIL-S-7493Composition4617Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6308A6312E6317F6320J6321D6322K6323H6324E6325F6327G6328H6330E63316342H6348A6349B6350H6351E6352FAlloySimilarSpecificationTitle of SpecificationUNS No.Bars and forgings, 0.90Si 1.0Cr2.0Ni 3.2Mo 2.0Cu 0.10V(0.070.13C), vacuum arc orelectroslag remeltedBars, forgings, and tubing, 1.8Ni0.25Mo (0.380.43C)Bars and forgings, 1.8Ni 0.25Mo(0.380.43C), heat treated, 125ksi (862 MPa) tensile strengthBars, forgings, and rings, 0.50Cr0.55Ni 0.25Mo (0.330.38C)Bars, forgings, and tubing, 0.42Cr0.30Ni 0.12Mo 0.003B(0.380.43C)Bars, forgings, and rings, 0.50Cr0.55Ni 0.25Mo (0.380.43C)Tubing, mechanical, 0.50Cr 0.55Ni0.25Mo (0.380.43C)Bars, forgings, and tubing, 0.65Cr0.70Ni 0.25Mo (0.380.43C)Bars and forgings, 0.50Cr 0.55Ni0.25Mo (0.380.43C), heattreated, 105 ksi (724 MPa) tensilestrengthBars and forgings, 0.50Cr 0.55Ni0.25Mo (0.380.43C), heattreated, 125 ksi (862 MPa) tensilestrengthBars, forgings, and tubing, 0.50Cr0.55Ni 0.25Mo (0.480.53C)Bars, forgings, and tubing, 0.65Cr1.25Ni (0.330.38C)Wire, welding, 0.50Cr 0.55Ni0.20Mo (0.330.38C), vacuummelted, environment controlledpackagingBars, forgings, and tubing, 0.80Cr1.0Ni 0.25Mo (0.380.43C)Bars, 0.95Cr 0.20Mo (0.280.33C),normalizedBars, 0.95Cr 0.20Mo (0.380.43C),normalizedSheet, strip, and plate, 0.95Cr0.20Mo (0.280.33C)Sheet, strip, and plate, 0.95Cr0.20Mo (0.280.33C),spheroidizedSheet, strip, and plate, 0.95Cr0.20Mo (0.330.38C)K71040Pyrowear, alloy53K224404640K224004640G873508735K0381081B40G874008740G874008740K116408740 ModG87408740MIL-S-6049G87408740MIL-S-6049K135508750MIL-S-6049K22033G873508735G984009840G413004130MIL-S-6758G414004140MIL-S-5626G413004130MIL-S-18729G413004130G413504135Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6354D6356D6357G6358F6359F6360J6361C6362D6365H6370K6371H6372H6373C6374A63756378E6379A6381EAlloySimilarSpecificationTitle of SpecificationUNS No.Sheet, strip, and plate, 0.75Si0.62Cr 0.20Mo 0.10Zr(0.100.17C)Sheet, strip, and plate, 0.95Cr0.20Mo (0.300.35C)Sheet, strip, and plate, 0.50Cr0.55Ni 0.25Mo (0.330.38C)Sheet, strip, and plate, 0.50Cr0.55Ni 0.25Mo (0.380.43C)Sheet, strip, and plate, 0.80Cr1.8Ni 0.25Mo (0.380.43C)Tubing, seamless, 0.95Cr 0.20Mo(0.280.33C), normalized orstress relievedTubing, seamless round, 0.95Cr0.20Mo (0.280.33C), 125 ksi(860 MPa) tensile strengthTubing, seamless, 0.95Cr 0.20Mo(0.280.33C), 150 ksi (1034MPa) tensile strengthTubing, seamless, 0.95Cr 0.20Mo(0.330.38C), normalized orstress relievedBars, forgings, and rings, 0.95Cr0.20Mo (0.280.33C)Tubing, mechanical, 0.95Cr0.20Mo (0.280.33C)Tubing, mechanical, 0.95Cr0.20Mo (0.330.38C)Tubing, welded, 0.95Cr 0.20Mo(0.280.33C)Tubing, seam-free, round, 0.95Cr0.20Mo (0.280.33C), 95 ksi (655MPa) tensile strengthWire, welding, 0.50Cr 0.55Ni0.20Mo (0.180.23C), vacuummelted, environment controlledpackagingBars, 1.0Cr 0.20Mo 0.045Se(0.390.48C), die drawn, 130 ksi(896 MPa) yield strength, freemachiningBars, die drawn, 0.95Cr 0.20Mo0.05Te (0.400.53C), tempered,165 ksi (1140 MPa) yield strengthTubing, mechanical, 0.95Cr0.20Mo (0.380.43C)K11914NAX 9115-ACG413204132G873508735G874008740G434004340G413004130MIL-T-6736Condition NG413004130MIL-T-6736G413004130G413504135MIL-T-6736ConditionHTMIL-T-6735G413004130MIL-S-6758G413004130MIL-T-6736G413504135G413004130G413004130G862008620K115424142H ModK115464140 ModG414004140MIL-T-6736Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6382K6385E6386B6390C6395D6396B6406C6407E640864096411D6412J6413H6414F6415M MAM64156417DTitle of SpecificationUNS No.AlloyBars, forgings, and rings, 0.95Cr0.20Mo (0.380.43C)Sheet, strip, and plate, low alloy,heat resistant, 1.25Cr 0.50Mo0.65Si 0.25V (0.270.33C)Sheet and plate, heat treated, 90 ksiand 100 ksi yield strengthTubing, mechanical, 0.95Cr0.20Mo (0.380.43C)Sheet, strip, and plate, 0.95Cr0.20Mo (0.380.43C)Sheet, strip, and plate, 0.80Cr1.8Ni 0.25Mo (0.490.55C),annealedSheet, strip, and plate, 2.1Cr0.58Mo 1.6Si 0.05V (0.410.46C), annealedBars, forgings, and tubing, 1.2Cr2.0Ni 0.45Mo (0.270.33C)Bars and forgings, tool, hotwork,5.2Cr 1.5Mo 1.0V (0.350.45C),electroslag remelted (ESR) orconsumable electrode vacuumarc remelted (VAR), annealedBars, forgings, and tubing, 0.80Cr1.8Ni 0.25Mo (0.380.43C),special aircraft qualitycleanliness, normalized andtemperedBars, forgings, and tubing, 0.88Cr1.8Ni 0.42Mo 0.08V (0.280.33C), consumable electroderemeltedBars and forgings, 0.80Cr 1.8Ni0.25Mo (0.350.40C)Tubing, mechanical, 0.80Cr 1.8Ni0.25Mo (0.350.40C)Bars, forgings, and tubing, 0.80Cr1.8Ni 0.25Mo (0.380.43C),vacuum consumable electroderemeltedBars, forgings, and tubing, 0.80Cr1.8Ni 0.25Mo (0.380.43C)Bars, forgings, and tubing, 0.80Cr1.8Ni 0.25Mo (0.380.43C)Bars, forgings, and tubing, 0.82Cr1.8Ni 0.40Mo 1.6Si 0.08V(0.380.43C), consumableelectrode remeltedG414004140K230151722A/SK11856SimilarSpecificationG414004140G41400MIL-S-56264140K22950K34378X200K33020IIS-220T20813H-13G434004340K230804340 ModG433704337G433704337G434004340G434004340MIL-S-5000G434004340MIL-S-5000K44220300MMIL-S-5000Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6418G6419C6421C6422F6423D6424B64256426D6427H6428D6429D6430D6431J6432A6433DUNS No.Bars, forgings, tubing, and rings,0.30Cr 1.8Ni 0.40Mo 1.3Mn1.5Si (0.230.28C)Bars, forgings, and tubing, 0.82Cr1.8Ni 0.40Mo 0.08V 1.6Si (0.400.45C), consumable electrodevacuum remeltedBars, forgings, and tubing, 0.80Cr0.85Ni 0.20Mo 0.003B(0.350.40C)Bars, forgings, and tubing, 0.80Cr0.85Ni 0.20Mo 0.003B 0.04V(0.380.43C)Bars, forgings, and tubing, 0.92Cr0.75Ni 0.52Mo 0.003B 0.04V(0.400.46C)Bars, forgings, and tubing, 0.80Cr1.8Ni 0.25Mo (0.490.55C)Bars, forgings, and tubing, 0.30Cr1.8Ni 0.40Mo 1.4Mn 1.5Si(0.230.28C), consumablevacuum electrode remeltedBars, forging, and tubing, 1.0Cr0.58Mo 0.75Si (0.800.90C),consumable electrode meltedBars, forgings, and tubing, 0.88Cr1.8Ni 0.42Mo 0.08V(0.280.33C)Bars, forgings, and tubing, 0.80Cr1.8Ni 0.35Mo 0.20V(0.320.38C)Bars, forgings, tubing, and rings,0.78Cr 1.8Ni 0.35Mo 0.20V(0.330.38C), consumableelectrode vacuum meltedBars, forgings, tubing, and rings,0.78Cr 1.8Ni 0.35Mo 0.20V0.75Mn (0.320.38C)Bars, forgings, and tubing, 1.05Cr0.55Ni 1.0Mo 0.11V(0.450.50C), consumableelectrode vacuum meltedBars, forgings, and tubing, 1.05Cr0.55Ni 1.0Mo 0.12V(0.430.49C)Sheet, strip, and plate, 0.80Cr1.8Ni 0.35Mo 0.20V 0.75Mn(0.330.38C)K32550Hy-TufMIL-S-7108K44220300MMIL-S-8844AlloySimilarSpecificationTitle of Specification98B37 ModK1194098BV40 ModK2433698BV40 ModK22950K32550Hy-TufK1859752CBK230804330 ModK234774335 ModK335174335 ModK335174335 ModK24728D6K24728D6AK335174335 ModMIL-S-8949Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6434D6435C6436B6437D6438D6439B6440J6442E6443E6444H6445E6446C6447D6448F6449CAlloySimilarSpecificationTitle of SpecificationUNS No.Sheet, strip, and plate, 0.78Cr 1.8Ni0.35Mo 0.20V (0.330.38C)Sheet, strip, and plate, 0.78Cr 1.8Ni0.35Mo 0.20V (0.330.38C), vacuumconsumable electrode melted,annealedSheet, strip, and plate, low alloy, heatresistant, 0.65Si 1.25Cr 0.50Mo0.85V (0.250.30C), annealedSheet, strip, and plate, 5.0Cr 1.3Mo0.50V (0.380.43C)Sheet, strip, and plate, 1.05Cr 0.55Ni1.0Mo 0.12V (0.450.50C),consumable electrode vacuummeltedSheet, strip, and Plate, 1.05Cr 0.55Ni1.0Mo 0.12V (0.420.48C),consumable electrode vacuummelted, annealedBars, forgings, and tubing, 1.45Cr(0.981.10C), for bearingapplicationsBars and forgings, 0.50Cr(0.981.10C), for bearingapplicationsBars, forgings, and tubing, 1.0Cr(0.981.10C), consumable electrodevacuum meltedBars, forgings, and tubing, 1.45Cr(0.981.10C), premium aircraftquality, consumable electrodevacuum meltedBars, forgings, and tubing, 1.05Cr1.1Mn (0.921.02C), consumableelectrode vacuum meltedBars, forgings, and tubing, 1.0Cr(0.981.10C), electroslag remeltedBars, forgings, and tubing, 1.4Cr(0.981.10C), electroslag remeltedBars, forgings, and tubing, 0.95Cr0.22V (0.480.53C)Bars, forgings, and tubing, 1.0Cr(0.981.10C), for bearingapplicationsK335174335 ModK335174335 ModK227701722A(V)T20811H-HK24728D6K24728D6ACG5298652100G5098650100G5198651100G5298652100K2209751100 ModG5198651100G5298652100G615006150MIL-S-8503G5198651100MIL-S-7420MIL-S-8949MIL-S-7420Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6450F6451A6452A64536454B6455G6456A6457A6458F6459B6460D6461G6462F6463B6464EAlloySimilarSpecificationTitle of SpecificationUNS No.Wire, spring, 0.95Cr 0.22V(0.480.53C), annealed and colddrawnWire, spring, 1.4Si 0.65Cr(0.510.59C), oil temperedWire, welding, 0.95Cr 0.20Mo(0.380.43C), vacuum melted,environment controlledpackagingWire, welding, 0.30 Cr 1.8Ni0.40Mo (0.230.28C), vacuummelted, environment controlledpackagingSheet, strip, and plate, 1.8Ni 0.8Cr0.25Mo (0.380.43C),consumable electrode meltedSheet, strip, and plate, 0.95Cr0.22V (0.480.53C)Wire, welding, 0.8Cr 1.8Ni 0.25Mo(0.350.40C), vacuum melted,environment controlledpackagingWire, welding, 0.95Cr 0.20Mo(0.280.33C), vacuum melted,environment controlledpackagingWire, welding, 1.25Cr 0.50Mo0.30V 0.65Si (0.280.33C),vacuum melted, environmentcontrolled packagingWire welding, 1.0Cr 1.0Mo 0.12V(0.180.23C), vacuum inductionmeltedWire, welding, 0.62Cr 0.20Mo0.75Si 0.10Zr (0.100.17C)Wire, welding, 0.95Cr 0.20V(0.280.33C), vacuum melted,environment controlledpackagingWire, welding, 0.95Cr 0.20V(0.280.33C)Wire, welding, 18.5Ni 8.5Co 5.2Mo0.72Ti 0.10A1, vacuumenvironment controlledpackagingElectrodes, welding, covered,1.05Mo 0.20V (0.060.12C)G615006150G925409254G43406E4340K 32550Hy-TufG434004340G615006150MIL-S-187314340 ModMIL-R-5632,Type IIIK131474130MIL-R-5632,Type IK230151722A(S)MIL-R-5632,Type IIK22720K11365NAX-915-ACK131486130K131496130K93130Mar 300W1001310013 (AWS)MIL-E-6843,Class E-10013Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6465B6466D6467C6468B6469A6470J6471D6472C64736475F647664776485G6487G6488EAlloySimilarSpecificationTitle of SpecificationUNS No.Wire, welding, 2.0Cr 10Ni 8.0Co1.0Mo 0.02A1 0.06V(0.100.14C), vacuum melted,environment controlled pakagingWire, welding, corrosionresistant, 5.2Cr 0.55MoElectrode, welding, covered, 5Cr0.55MoWire, welding, 1.0Cr 3.8Co 0.45Mo0.08V (0.140.17C), vacuum melted,environment controlled packagingWire, welding, 1.75Mn 0.80Cr 2.8Ni0.85Mo (0.090.12C), vacuummelted, environment controlledpackagingBars, forgings, and tubing, nitridinggrade, 1.6Cr 0.35Mo 1.1A1 (0.380.43C)Bars, forgings, and tubing, nitridinggrade, 1.6Cr 0.35Mo 1.2A1 (0.380.43C), consumable electrodevacuum meltedBars and forgings, nitridinggrade, 1.6Cr 0.35Mo 1.1 A1 (0.380.43C), hardened and tempered, 112ksi (772 MPa) tensile strengthWire, welding, 0.88Cr 1.8Ni 1.6Co0.42Mo 0.08V (0.280.33C), vacuummelted, environment controlledpackagingBars, forgings, and tubing, nitridinggrade, 1.1Cr 3.5Ni 0.25Mo 1.25A1(0.210.26C)Bars, forgings, and tubing, 0.50Cr0.12Mo (0.891.01C), for bearingapplicationsBars, forgings, and tubing, 0.80Cr(0.901.03C), for bearingapplicationsBars and forgings, 5.0Cr 1.3Mo 0.50V(0.380.43C)Bars and forgings, 5.0Cr 1.3Mo 0.50V(0.380.43C), consumable electrodevacuum meltedBars and forgings, 5.0Cr 1.3Mo 0.5V(0.380.43C)K91971HY-180S50280Type 502W50210Type 502K91461HP 9-4-20K24065135 ModK24065135 ModK24065135 ModMIL-S-6709T20811H-11FED-QQ-T-570Class H-11T20811H-11T20811H-11MIL-S-6709K52355Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6490D6491A6501A6512C6514C6518A6519A6520B6521A6522A6523CUNS No.Bars, forgings, and tubing, 4.0Cr4.2Mo 1.0V (0.770.85C),premium aircraft quality forbearing applications,consumable electrode vacuummeltedBars, forgings, and tubing, 4.1Cr4.2Mo 1.0V (0.800.85C),premium aircraft quality forbearing applications, Doublevacuum meltedWire, welding, maraging steel,18Ni 8.0Co 4.9Mo 0.40Ti0.10A1, vacuum inductionmelted, environment controlledpackagingBars, forgings, tubing, and rings,18Ni 7.8Co 4.9Mo 0.40Ti0.10A1, consumable electrodemelted, annealedBars, forgings, tubing, and rings,Maraging, 18.5Ni 9Co 4.9Mo0.65Ti 0.10A1, consumableelectrode melted, annealedT11350M-50T11350M-50K92890Maraging 250K92890Maraging 250MIL-S-46850Type III,GradeK93120Maraging 300MIL-S-46850Type 300MIL-S-13881,Type II,Class IK92890Maraging 250K93120Maraging 300K92571AF-1410K91472HP 9420Sheet, strip, and plate, maraging,19Ni 3.0Mo 1.4Ti 0.10A1,double vacuum melted, solutionheat treatedBars, forgings, tubing, and rings,maraging, 19Ni 3.0Mo 1.4Ti0.10A1, double vacuum melted,annealedSheet, strip, and plate, maraging250, 18Ni 7.8Co 4.9Mo 0.40Ti0.10A1, consumable electrodemelted, Solution heat treatedSheet, strip, and plate, 18.5NI9.0Co 4.9Mo 0.65Ti 0.10A1,consumable electrode melted,solution heat treatedplate, 2.0Cr 10Ni 14Co 1.0Mo(0.130.17C), vacuum melted,normalized and overagedSheet, strip, and plate, 0.75Cr9.0Ni 4.5Co 1.0Mo 0.09V(0.170.23C), vacuumconsumable electrode melted,annealedAlloySimilarSpecificationTitle of SpecificationMIL-S-46850Grade 300Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.16 (Continued )AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietaryor AISI-SAE Grade, and Similar Specification for Wrought Alloy SteelsAMS No.6524C6525A6526C6527B652865296530H653265336535G6543B6544B6546D6550HTitle of SpecificationUNS No.AlloyWire, welding, 1.0Cr 7.5Ni 4.5Co1.0Mo 0.09V (0.290.34C),consumable electrode vacuummeltedBars, forgings, tubing, and rings,0.75Cr 9.0Ni 4.5Co 1.0Mo 0.09V(0.170.23C), consumableelectrode vacuum meltedBars, forgings, tubing, and rings,1.0Cr 7.5Ni 4.5Co 1.0Mo 0.09V(0.290.34C), consumableelectrode vacuum melted,annealedBars and forgings, 2.0Cr 10Ni14Co 1.0Mo (0.150.19C),vacuum melted, normalized andoveragedBars, 0.95Cr 0.20Mo (0.280.33C),special aircraft qualitycleanliness, normalizedBars, 0.95Cr 0.20Mo (0.380.43C),special aircraft qualitycleanliness, normalizedTubing, seamless, 0.50Ni 0.55Cr0.20Mo (0.280.33C)Bars and forgings, 3.1Cr 11.5Ni13.5Co 1.2Mo (0.210.25C),vacuum melted, annealedWire, welding, 2.0Cr 10Ni 14Co1.9Mo (0.130.17C), vacuummelted, environment controlledpackagingTubing, seamless, 0.50Cr 0.55Ni0.20Mo (0.280.33C)Bars and forgings, 2.0Cr 10Ni8.0Co 1.0Mo (0.100.14C),double vacuum melted, solutionheat treatedPlate, maraging, 2.0Cr 10Ni 8.0Co1.0Mo (0.100.14C), doublevacuum melted, heat treatedSheet, strip, and plate, 0.48Cr0.80Ni 4.0Co 0.48Mo 0.09V(0.240.30C), consumableelectrode melted, annealedTubing, Welded, 0.55Cr 0.50Ni0.20Mo (0.280.33C)K91313HP 9430K91472HP 9420K91283HP 9430K92571AF 1410G413004130G414004140G863008630K92580Aermet 100K92571AF 1410G863008630K92571SimilarSpecificationAF 1410K91970K91122HP 9425G863008630Source: From 1994 SAE AMS Index, Society of Automotive Engineers, Warrendale, PA. 2006 by Taylor & Francis Group, LLC.MIL-T-6734duplex stainless steels, and nickel-base alloys), dimensions, weights, process of manufacture,chemical and mechanical property requirements, and testing and inspection methods [61]. APISpec 5LD covers seamless, centrifugal cast, and welded clad steel line pipe and lined steel pipewith increased corrosion-resistant properties. The clad and lined steel line pipes are composedof a base metal outside and a corrosion resistent alloy (CRA) layer inside the pipe; the basematerial conforms to API Spec 5L, except as modified in the API Spec 5LC document. Thisspecification provides standards for pipe with improved corrosion resistance suitable for use inconveying gas, water, and oil in both the oil and natural gas industries [62].1.5.5 ANSI SPECIFICATIONSANSI standard begins with the prefix ANSI, followed by an alphanumeric code with anuppercase letter, subsequently followed by one to three digits and additional digits that areseparated by decimal points. ANSI standards can also have a standard developers acronymin the title. Examples are ANSI H35.2, ANSI A156.2, ANSI B18.2.3.6M, ANSI/ASME NQ21989, ANSI/API Spec 5CT-1992, ANSI/API Spec 5D-1992, ANSI/API Spec 5L-1992, andANSI/API Spec 5LC-1991 [57,6062].1.5.6 AWS SPECIFICATIONSAWS standards are used to support welding design, testing, quality assurance, and otherrelated joining functions. These standards begin with the prefix AWS followed by the letterand numerals with decimal point. Examples of AWS specifications with corresponding AISISAE or proprietary grade and UNS number in parentheses are AWS A5.1 (E6010, W06010),AWS A5.2 (RG65, WK00065), and AWS A5.5 (E9018-D3, W19118).1.6INTERNATIONAL SPECIFICATIONS AND DESIGNATIONSSince steelmaking technology is available worldwide, familiarity with international specifications and designations for steels is necessary. Table 1.17 cross-references SAE steels withthose of a selected group of international specifications and designations, which are describedin the following paragraphs. More elaborate information on cross-referencing is available inRefs. [7,15,57,63].1.6.1ISO DESIGNATIONSThe International Organization for Standardization (ISO) system has standard designationfor steel. Designation for Steels with Yield StrengthThe designations are preceded by the letter and followed by yield strength value (MPa). Theprefix of nonalloy structural steel is letter S, for example, S235. The prefix of nonalloyengineering steel is letter E, for example, E235. The numbers indicate the yield strength!235 MPa.The method of HSLA steels is equivalent to nonalloy engineering steels. The lower limit ofyield strength is 355690 MPa, for example, E355 , . . . , . . . , E690, where E355 and E690 aredifferent steel grades. 2006 by Taylor & Francis Group, LLC.TABLE 1.17Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)Carbon steels10051.0288, D5-21.0303, QSt32-21.0312, D5-11.0314, D6-21.0393, ED31.0394, ED41.1012, RFe12010061.0311, D7-11.0313, D8-21.0317, RSD41.0321, St231.0334, StW231.0335, StW241.0354, St14Cu31.0391, EK21.0392, EK41.1009, Ck710081.0010, D91.0318, St281.0320, St221.0322, USD81.0326, RSt281.0330, St2, St121.0333, St3, St131.0331, RoSt21.0332, StW221.0336, USt4, USt141.0337, RoSt41.0344, St12Cu31.0347, RRSt131.0357, USt281.0359, RRSt231.0375, FeinstblechT57, T61, T65,T701.0385, WeissblechT57, T61, T65,T701.0744, 6P101.0746, 6P201.1116, USD610101.0204, UQSt361.0301, C101.0328, USD101.0349, RSD9Japan (JIS)UnitedKingdom (BS)France(AFNOR NF)China(GB)ISO05F970 015A03970 030A04970 040A04970 050A04A35-564XC6FFG3445STKM11A(11A)1449 3CR1449 3CS1449 3HR1449 3HS1717 ERW1013606 261A35-551 XC10XC6XC6FF08F0808A1CC8XG4051 S10CG4051 S9Ck1449 40F30,43F35,46F40,50F45,A33-101 AF34CC10C1010F10C10CE10C11xContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)Japan (JIS)1.1121, Ck101.1122, Cq1010121.0439, RSD1310131.0036, USt37-21.0037, St37-21.0038, RSt37-21.0055, USt34-11.0057, RSt34-11.0116, St37-31.0218, RSt41-21.0219, St41-31.0307, StE210.71.0309, St35.41.0315, St37.81.0319, RRStE210.71.0356, TTSt351.04171.0457, StE240.71.0401, C151.1132, CQ151.1135, Ck16A11015G4051 S12CG4051 F15CkG4051 S15CUnitedKingdom (BS)60F55,68F62,75F70(available inHR, HS, CSconditions)1449 4HR,4HS, 4CR,4CS970 040A10(En2A,En2A/1,En2B)970 045A10,045M10(En32A)970 050A10970 060A10980 CEW11449 12HS,12CS1501 141-360970 040A12(En2A,En2A/1,En2B)970 050A12970 060A123059 3603061 3603603 360970 040A15970 050A15970 060A15France(AFNOR NF)China(GB)A33-101 AF37A35-551 XC12C12A35-551 XC12CC12XC1515ISOC15C15E4C15M2Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)10161017Fed. R. of Germany(DIN)1.1140, Cm151.1141, Ck151.11441.1148, Ck15A11.0419, RS144.21.0467, 15Mn31.0468, 15Mn3A11.1142, GS-Ck1610181.0453, C16.8101910201.0402, C221.0414, D20-21.0427, C22.31.0460, C22.81.1149, Cm221.1151, Ck22102110221.0432, C211.0469, 21Mn41.0482, 19Mn51.1133, 20Mn5,GS-20Mn51.1134, Ck19Japan (JIS)UnitedKingdom (BS)France(AFNOR NF)China(GB)ISO970 080A15,080M15970 173H16G4051 S17CG4051 S20CG4051 S20CK3059 4403606 440970 080A15,080M15970 170H15970 173H161449 17HS,17CS970 040A17970 050A17970 060A17970 080A17970 040A20970 050A20(En2C,En2D)970 060A20970 070M20970 080A203111 Type 9970 120M19970 170H2015MnA35-551 XC18A35-552 XC18A35-566 XC18A35-553XC18SA35-554XC18SA33-101 AF42C20A35-551 XC18A35-552 XC18A35-566 XC18A35-553 C20A35-553 XC18SA35-554 XC18SCC20A35-551 21B3A35-552 21B3A35-553 21B3A35-557 21B3A35-566 21B3A35-55120MB5A35-55220 MB5A35-55320MB5A35-55620MB5A35-55720MB5A35-56620MB5A35-566 20M520C20CC21K20MnContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)Japan (JIS)10231.1150, Ck22.81.1152, Cq22G4051 S22C10251.0406, C251.0415, D25-2,D26-21.1158, Ck251.1155, GS-Ck251.1156, GS-Ck24G4051 S25C10291.0562, 28Mn410301.0528, C301.0530, D30-21.1178, Ck301.1179, Cm301.1811, G-31Mn4G3445STKM15A(15A),STKM15C(15C)G4051 S28CG4051 S30C10351.0501, C351.0516, D35-21.1172, Cq351.1173, Ck341.1180, Cm351.1181, Ck35G4051 S35C10371.0520, 31Mn41.0561, 36Mn4G4051 S35C1038No internationalequivalents1.1190, Ck42A110261039UnitedKingdom (BS)1449 2HS,22CS970 040A22(En2C,En2D)970 050A22970 060A22970 080A22970 070M26970 080A25970 080A27970 060A27970 080A27(En5A)France(AFNOR NF)China(GB)ISOA35-552 XC25A35-566 XC2525C25C25E425MnA33-101 AF50CC28C30C25M21449 30HS,30CS970 060A30970 080A30(En5B)970 080M30(En5)1717 CDS105/106970 060A35970 080A32(En5C)970 080A35(En8A)980 CFS63111 type 10970 080M36970 170H36A35-552 XC32A35-553 XC3230C30C30A33-101 AF55A35-553 C35A35-553 XC38A35-554 XC38XC35XC38TSC3535C35C35E4C35M2970 060A40970 080A40(En8C)970 080M40(En8)40M5A35-552XC38H2A35-55338MB535Mn40MnContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)Japan (JIS)UnitedKingdom (BS)970 170H4110401.0511, C401.0541, D40-21.1186, Ck401.1189, Cm40G4051 S40C10421.0517, D45-2G4051 S43C10431.0558, GS-60.3G4051 S43C104410451.0517, D45-21.0503, C451.1184, Ck461.1191, Ck45,GS-Ck451.1192, Cq451.1194, Cq451.1201, Cm451.1193, Cf451.0503, C451.0519, 45MnAl1.1159, GS-46Mn4G4051 S45CG5111 SCC510461049G3445STKM17A(17A)G3445STKM17C(17C)12871449 40HS,40CS3146 Class 1Grade C3146 Class 8970 060A40970 080A40(En8C)970 080M40(En8)970 060A42970 080A42(En8D)970 060A42970 080A42(En8D)970 080M46970 060A47970 080A47970 080M463100 AW2970 080M46970 060A47970 080A47France(AFNOR NF)A35-55638MB5A35-557XC38H2XC42, XC42TSA33-101 AF60C40China(GB)40A35-552XC42H1A35-553 C40CC45XC42, XC42TSA35-552XC42H2A33-101 AF65A35-552XC48H1A35-553 XC45A35-554 XC48XC48TSC4545M4TSA35-552XC48H1A35-552XC48H2XC48TSA35-552XC48H1A35-554 XC48XC48TSISOC40C40E4C40M2C45C45E4C45M24545MnContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)1050105310551059106010641065Fed. R. of Germany(DIN)1.0540, C501.1202, D53-31.1206, Ck501.1210, Ck531.1213, Cf531.1219, Cf541.1241, Cm501.1210, Ck531.1213, Cf531.1219, Cf541.0518, D55-21.0535, C351.1202, D53-31.1203, Ck551.1209, Cm551.1210, Ck531.1213, Cf531.1219, Cf541.1220, D55-31.1820, C55W1.0609, D58-21.0610, D60-21.0611, D63-21.1212, D58-31.1222, D63-31.1228, D60-31.0601, C601.0642, 60Mn31.1221, Ck601.1223, Cm601.1740, C60W1.0611, D63-21.0612, D65-21.0613, D68-21.1222, D63-31.1236, D65-31.0627, C681.0640, 64Mn31.1230,FederstahldrahtFD1.12331.1240, 65Mn41.1250,FederstahldrahtVD1.1260, 66Mn4Japan (JIS)G4051 S50CG4051 S53CUnitedKingdom (BS)China(GB)A35-553 XC505052M4TSA35-553 XC5450Mn3100 AW3970 060A57970 070M55970 080A52(En43C)970 080A57A33-101 AF70A35-552XC55H1A35-552XC55H2A35-553 XC54XC55C5555970 060A62A35-553 XC601449 60HS1449 60CS970 060A57970 080A57A35-553 XC60970 060A62970 080A62(En43D)970 060A67G4051 S53CG4051 S53CG4051 S55CG4051 S58C1549 50HS1549 50CS970 060A52970 080A52(En43C)970 080M50(En43A)970 080A52(En43C)France(AFNOR NF)XC65C50C50E4C50M2C55C55E460C60C60E4C60M265ISOSL, SMTypeDC970 080A67(En43E)Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)106910701074107510781080Fed. R. of Germany(DIN)Japan (JIS)UnitedKingdom (BS)France(AFNOR NF)1.0615, D70-21.0617, D73-21.0627, C681.1232, D68-31.12371.1249, Cf701.1251, D70-31.1520, C70W11.1620, C70W21.0603, C671.0643, 70Mn31.1231, Ck67A35-553 XC68XC7070XC7070TypeDC1.0605, C751.0645, 76Mn31.0655, C741.1242, D73-31.0614, D75-21.0617, D73-21.0620, D78-21.1242, D73-31.1252, D78-31.1253, D75-31.0620, D78-21.0622, D80-21.0626, D83-21.1252, D78-31.1253, D75-31.1255, D80-31.1262, D83-31.1525, C80W11.1259, 80Mn41.1265 D85-2A35-553 XC75XC7075A35-553-XC75XC7075G4801 SUP310841.1830, C85W10851.0647, 85Mn31.1273, 90Mn41.1819, 90Mn41449 70HS,70CS970 060A72970 070A72(En42)970 080A72970 070A72(En42)970 080A72970 060A78XC801449 80HS,80CS970 060A78970 060A83970 070A78970 080A78970 080A83970 060A86970 080A86970 080A83XC80China(GB)ISO8085XC85Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)1086Fed. R. of Germany(DIN)Japan (JIS)11411.0616, C85, D85-21.0626, D83-21.0628, D88-21.1262, D83-31.1265, D85-31.1269, Ck851.1272, D88-31.1273, 90Mn41.1819, 90Mn41.1282, D95S31.0618, D95-21.1274, Ck 1011.1275, Ck1001.1282, D95S31.1291, MK971.1545, C105W11.1645, C105W2No internationalequivalentsG4804 SUM4211441.0727, 45S20G4804 SUM43114611511.0727, 45S201.0728, 60S201.0729, 70S2010901095114012L141.0736, 9SMn36France(AFNOR NF)China(GB)970 050A86A35-553 XC901449 95HS1449 95CS970 060A961449 95HS1449 95CS970 060A99T9AA35-553XC100T10A35-56245MF4Y40MnG4801 SUP4G4804 SUM23970 212A42(En8DM)970 216A42970 212A42(En8DM)970 212M44970 216M44970 225M44970 226M44970 212M44970 220M07(En1A)970 230M07970 240M07(En1A)970 240M07(En1B)85ISOResulfurized/rephosphorized carbon steels1211No internationalequivalents12121.0711, 9S20G4804 SUM211.0721, 10S201.1011, RFe160KG4804 SUM2212131.0715, 9SMn281.0736, 9SMn361.0740, 9SMn401215UnitedKingdom (BS)Type DC44SMn28A35-56245MF645MF410F212MF4S200A35-561 S250S250A35-561 S300Y15Pb44SMn289S2011SMnPb28Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)Alloy steels133013351.5069, 36Mn713401.5223, 42MnV713451.0625, StSch90C1.0912, 46Mn71.0913, 50Mn71.0915, 50MnV71.5085, 51Mn71.5225, 51MnV740231.5416, 20Mo340241.5416, 20Mo340271.5419, 22Mo4402840321.54114037404240474118413041351.2382, 43MnSiMo41.5412,GS-40MnMo4 31.5432, 42MnMo71.2382, 43MnSiMo41.5432, 42MnMo7No internationalequivalents1.7211, 23CrMoB41.7264, 20CrMo51.2330, 35CrMo41.7220, 34CrMo41.7220, GS34CrMo41.7226, 34CrMoS41.7231, 33CrMo4Japan (JIS)UnitedKingdom (BS)France(AFNOR NF)30Mn235Mn240Mn245Mn228Mn636Mn642Mn6970 605M30970 605A32970 605H32970 605M30970 605M36(En16)3111 Type 2/13111 Type 2/2970 605A37970 605H37970 708H20970 708M20G5111 SCMnM3G4052 SCM15HG4105 SCM21HG4052SCM418HG4105SCM418HG4105 SCM1G4105 SCM432G4105 SCM2G4105 SCM430G4106 SCM2G4054 SCM3HG4054SCM435HG4105 SCM1G4105 SCM432G4105 SCM3G4105 SCM4351717 CDS110970 708A30970 708A37970 708H37A35-55230CD4A35-55630CD4A35-55730CD435CD4A35-55235CD4A35-55335CD4A35-55635CD4A35-55734CD4China(GB)G20CrMoISO36Mo3E18CrMo418CrMo4E30CrMo30CrMoA35CrMo35CrMoV34CrMo434CrMo4E34CrMoS4Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)4137Fed. R. of Germany(DIN)1.7225, GS42CrMo4Carbonmanganese steels15131.0424,SchiffbaustahlCS:DS1.0479, 13Mn61.0496, 12Mn61.0513,Schiffbaustahl A321.0514,Schiffbaustahl B321.0515,Schiffbaustahl E321.05491.05791.0583,Schiffbaustahl A361.0584,Schiffbaustahl D361.0589,Schiffbaustahl E 361.05991.8941, QStE260N1.8945, QStE340N1.8950, QStE380N15221.0471, 21MnS151.0529, StE350-Z21.1120, GS-20Mn51.1138, GS-21 Mn51.1169, 20Mn61.8970, StE385.71.8972, StE415.71.89781.8979Japan (JIS)G4052 SCM4HG4052SCM440HG4105 SCM4G4106 SMn21UnitedKingdom (BS)France(AFNOR NF)China(GB)ISO3100 type 5970 708A37970 708H37970 709A3740CD442CD4A35-55238CD4A35-55738CD41449 40/30 HR1449 40/30 HS1449 40/30 CS1453 A22772 150M12970 125A15970 130M15970 130M15(En201)12M5A33-101AF50-SA35-501 E35-4A35-501 E36-2A35-501 E36-315Mn1503 2214601503 2234091503 2244903146 CLA2980 CFS7A35-55120MB5A35-552 20M5A35-556 20M5A35-55220MB5A35-55320MB5A35-55620MB5A35-55720MB5A35-56620MB520MnContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)15241.0499, 21Mn6A11.1133, 20Mn5, GS20Mn51.1160, 22Mn61526Japan (JIS)G4106 SMn21G5111 SCMn115271.0412, 27MnSi51.1161, 26Mn51.1165, 30Mn51.1165, GS-30Mn51.1170, 28Mn6G5111 SCMn215361.0561, 36Mn41.1165, 30Mn51.1165, GS-30Mn51.1166, 34Mn51.1167, 36Mn5,GS-36Mn51.1813, G-35Mn5G4052 SMn1HG4052 SMn433HG4106 SMn1G4106 SMn433G5111 SCMn2G5111 SCMn315411.0563, E1.0564, N-801.1127, 36Mn61.1168, GS-40Mn515481.1128, 46Mn51.1159, GS-46Mn41.0542, StSch801.0624, StSch90B1.1226, 52Mn51.0908, 60SiMn51.12331.1240, 65Mn41.1260, 66Mn7G4106 SMn2,SMn438G4052 SMn2H,SMn438HG4106 SMn3,SMn443G4052 SMn3H,SMn443HG5111 SCMn51551155215611566UnitedKingdom (BS)1456 Grade A970 150M19(En14A,En14B)970 175H23980 CDS9,CDS10970 120M281453 A31456 Grade B1,Grade B23100 A53100 A6970 150M28(En14A,En14B)10453100 A5, A6970 120M36(En15B)970 150M36(En15)970 135M44970 150M40France(AFNOR NF)China(GB)ISO20Mn2A35-56625MS525Mn30MnA35-552 32M5A35-55238MB5A35-55338MB5A35-55638MB5A35-55738MB540M545M5A35-552 40M635Mn24M4TS55M540Mn2SL, SM45MnSL, SM50Mn2SL, SMSL, SM60Mn65MnSL, SMContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. ofGermany (DIN)Resulfurized carbon steels11081.0700, U7S101.0702, U10S1011101.0703, R10S1011171118G4804 SUM12G4804 SUM11G4804 SUM311137Japan (JIS)11391.3563, 43CrMo41.7223, 41CrMo41.7225, 42CrMo41.7225,GS-42CrMo41.7227,42CrMoS44142G4804 SUM411.0726, 35S2041401.3563, 43CrMo41.7223, 41CrMo4G4052 SCM4HG4052SCM440HG4103 SNCM4G4105 SCM4G4105 SCM440UnitedKingdom (BS)970 210A15970 210M17(En32M)970 214A15970 214M15(En202)970 214M15(En201)970 212M36(En8M)970 216M36(En15AM)970 225M36970 212A37(En8BM)970 212M36(En8M)970 216M36(En15AM)970 225M363100 Type 54670 711M40970 708A40970 708A42(En19C)970 708H42970 708M40970 709A40970 709M40970 708A42(En19C)970 708H42970 709A42France(AFNOR NF)China (GB)A35-562 10F110S20Y2035MF4A35-56235MF6ISOY3535MF4A35-56235MF640CD4A35-55242CD4,42CDTSA35-55342CD4,42CDTSA35-55642CD4,42CDTSA35-55742CD4,42CDTS40CD4A35-55242CD4,42CDTSA35-55342CD4,42CDTSA35-55642CD4,42CDTSA35-55742CD4,42CDTS42CrMo42CrMo442CrMo4E42CrMoS442CrMo40CrMnMo42CrMo4Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)41451.2332, 47CrMo4414743201.2332, 47CrMo41.3565, 48CrMo41.7228, 50CrMo41.7228, GS50CrMo41.7230,50CrMoPb41.7328, 49CrMo41.3565, 48CrMo41.7238, 49CrMo41.7228, 50CrMo41.7228, GS50CrMo41.7230, 50CrMoPb41.7238, 49CrMo41.7229, 61CrMo41.7266,GS-58CrMnMo4GS-58CrMnMo44343401.6565, 40NiCrMo641504161E43404422442740CrNiMoA45CrNiMoVA1.6562, 40NiCrMo7345CrNiMo VA40CrNiMoA1.5419, 22MoNo internationalequivalentJapan (JIS)G4052 SCM5HG4052SCM445HG4105 SCM5,SCM445G4052 SCM5HG4052SCM445HG4105 SCM5,SCM445UnitedKingdom (BS)France(AFNOR NF)China(GB)ISO970 708H45A35-55345SCD6A35-55245SCD6970 708A47A35-55245SCD6A35-55345SCD6A35-57150SCD6A35-57150SCD6G4801 SUP133100 BW43146 CLA12Grade CG4103 SNCM23G4103SNCM420G4103SNCM420HG4103 SNCM8G4103SNCM439G4108 SNB23-15G4108 SNB24-1536CrNiMo64670 818M40970 2S.11950CrMo420NCD7A35-56518NCD4A35-56520NCD7970 2S.11923D5Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. ofGermany (DIN)Japan (JIS)46154617462046264718472048154817482050B4050B44504650B4650B5050605115No internationalequivalentNo internationalequivalentNo internationalequivalentNo internationalequivalent1.7003, 38Cr21.7023, 28CrS21.3561, 44Cr2No internationalequivalent1.7138,52MnCrB31.2101,62SiMnCr41.7131, 16MnCr5,GS-16MnCr51.7139,16MnCrS51.7142,16MnCrPb51.7160,16MnCrB5UnitedKingdom (BS)970 665A17970 665H17970 665M17(En34)970 665A19970 665H20970 665M20970 665A24(En35B)France(AFNOR NF)China (GB)ISO15ND82ND818NCD4G4052 SMnC3HG4052SMnC443HG4106 SMnC3G4106 SMnC443G5111 SCMnCr4A35-552 38C2A35-556 38C2A35-557 38C2A35-552 42C2A35-556 42C2A35-557 42C245C255C261SC7A35-552 60SC716MC5A35-55116MC515Cr15CrMnG4052 SCr21HG4052 SCr415HG4104 SCr21G4104 SCr415970 526M60(EnH)970 527A17970 527H17970 527M17Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)511751205130Fed. R. of Germany(DIN)1.3521, 17MnCr51.7016, 17Cr31.7131, 16MnCr5,GS-16MnCr51.7139,16MnCrS51.7142,16MnCrPb51.7168,18MnCrB51.2162, 21MnCr51.3523, 19MnCr51.7027, 20Cr41.7028, 20Cr5 41.7121,20CrMnS3 31.7146,20MnCrPb51.7147, GS20MnCr51.7149,20MnCrS51.8401, 30MnCrTi4Japan (JIS)UnitedKingdom (BS)18Cr4A35-55116MC5G4052 SCr22HG4052 SCr420HG4052 SMn21HG4052 SMn421HG4104 SCr22G4104 SCr420A35-55120MC5A35-55220MC520Cr20CrMn970 530A30(En18A)970 530H3028C430Cr970 530A32(En18B)970 530A36(En18C)970 530H323111 Type 3970 530A36 (En18C)970 530H36A35-552 32C4A35-553 32C4A35-556 32C4A35-557 32C4G4052 SCr2HG4052 SCr430HG4104 SCr2G4104 SCr430G4104 SCr3G4104 SCr43551321.7033, 34Cr41.7037, 34CrS451351.7034, 37Cr41.7038, 37CrS41.7043, 38Cr4G4052 SCr3HG4052 SCr435H51401.7035, 41Cr41.7039, 41CrS41.7045, 42Cr4G4052 SCr4HG4052 SCr440HG4104 SCr4G4104 SCr4403111 Type 3970 2S.117970 530A40 (En18D)970 530H40970 530M40France(AFNOR NF)38C4A35-552 38Cr4A35-553 38Cr4A35-556 38Cr4A35-557 38Cr4A35-552 42C4A35-557 42C4A35-556 42C4China(GB)ISO20Cr420MnCr520Cr4E20CrS435Cr34Cr434CrS440Cr38Cr41Cr441CrS441Cr4EContinued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. ofGermany (DIN)Japan (JIS)51471.7145, GS50CrMn4 451501.7145, GS50CrMn4 41.8404,60MnCrTi451551.7176, 55Cr351601.2125, 65MnCr451B60E501001.2018, 95Cr11.3501, 100Cr21.2057, 105Cr41.2109, 125CrSi51.2127,105MnCr41.3503, 105Cr41.2059, 120Cr51.2060, 105Cr51.2067, 100Cr61.3505, 100Cr61.3503, 105Cr41.3514, 101Cr61.3520,100CrMn6No internationalequivalents1.8159, GS-50CrV4E51100E5210061186150811581B4586158617G4801 SUP11G4801 SUP9G4801 SUP9AG4801 SUP10No internationalequivalentsNo internationalequivalentsUnitedKingdom (BS)3100 BW23100 BW33146 CLA12Grade A3146 CLA12Grade B3100 BW23100 BW33146 CLA12Grade A3146 CLA12Grade BFrance(AFNOR NF)ISO45Cr50Cr55CrMnA60CrMnA60CrMnBAGCr6GCr9970 534A99(En31)970 535A99(En31)GCr15970 735A 50(En47)970 S.204A35-552 50CV4970 527A60(En48)970 527H60970 805A17970 805H17970 805M17(En 361)50C 4China (GB)A35-571 55C3A35-565 100C250CrVA150CrV4A35-553 50CV4A35-57150CVA15NCD215NCD418NCD418NCD6Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)Fed. R. of Germany(DIN)86201.6522,20NiCrMo21.6523,21NiCrMo21.6526,21NiCrMoS21.6543,21NiCrMo2 286221.6541,23MnNiCrMo5 2Japan (JIS)UnitedKingdom (BS)G4052SNCM21HG4052SNCM220HG4103 SNCM21G4103SNCM2202772 806M20970 805A20970 805H20970 805M20(En362)2772 806M22970 805A22970 805H22970 805M228625970 805H25970 805M25862786308637No internationalequivalents1.6545, 30NiCrMo2 286401.6546, 40NiCrMo2 2970 945M38(En100)3111 Type 7,2S.147970 945A40(En 100C)8642France(AFNOR NF)China (GB)ISO20CrNiMoG20CrNiMo20NiCrMo220NiCrMo2E20NiCrMoS230NCD240NCD340NCD240NCD2TS40NCD3TS40NCD341CrNiMo2E18NCD420NCD2A35-55119NCDB2A35-55219NCDB2A35-55120NCD2A35-55320NCD2A35-56520NCD2A35-56620NCD223NCDB4A35-55623MNCD5A35-55623NCDB2A35-56622NCD225NCD4A35-55625MNCD6A35-56625MNDC6No internationalequivalentsNo internationalequivalentsNo internationalequivalents864586B45Continued 2006 by Taylor & Francis Group, LLC.TABLE 1.17 (Continued)Cross-Reference to SteelsUnitedStates(SAE)86508655866087208740882292549260Fed. R. ofGermany (DIN)No internationalequivalentsNo internationalequivalentsNo internationalequivalents1.6546,40NiCrMo2 2No internationalequivalentsE93101.6657,14NiCrMo1 3 4G10CrNi3Mo94B15Japan (JIS)UnitedKingdom (BS)3111 Type 7,2S.147China (GB)ISO970 805A60970 805H60France(AFNOR NF)G4801 SUP760S761S794B3016NCD1360Si2Mn60Si2MnA56SiCr759Si761SiCr7No internationalequivalentsNo internationalequivalentsNo internationalequivalents94B17970 250A58(En45A)970 250A61(En45A)970 832H13970 832M13(En36C)S.15740NCD240NCD2TS40NCD3TSSource: From Anon., Classification and designation of carbon and low-alloy steels, ASM Handbook, 10th ed., Vol. 1, ASMInternational, Materials Park, OH, 1990, pp. 140194; H. Lin, G. Lin, and Y. Ma, Eds., Designation and Trade Name Handbookof Steels Worldwide, mechanical Industry Press, Beijing, 1997. Designation for Steels with Chemical CompositionThe prefix of nonalloy heat-treatable steel is C, and followed by numerical value. Thenumerical value is 100 times of the average of carbon content, for example, C25 is the steelcontaining an approximate mean of 0.25% C. The additional suffix EX or MX indicated thequality steel or the high quality steel.The designation for structural alloy steels and spring steels are similar to that of DINsystems. For example, 36CrNiMo4 contains 0.36% C and alloy elements, such as Cr, Ni,and Mo, the number 4 is the product of multiplier of the amount of first alloy element(here Cr). 2006 by Taylor & Francis Group, LLC.The designation for bearing steels is indicated by a numeric code. Types 15 representhigh-carbon chromium-bearing steels (fully hardened hearing steels), Types 1016 arethe surface-hardened bearing steels, Types 2021 denote stainless-bearing steels, and Types3032 are the high-temperature bearing steels. Stainless steels are also indicated by a numericcode. For example, the ferritic stainless steels include Type 1Ti, 1, 2, 8, . . . , martensiticstainless steels include Type 3, 4, 5, . . . , austenitic stainless steels include Types 1024 andA2A4, etc.Heat-resistant steels are indicated with prefix letter H and followed by numeric code. Forexample, types H1H7 are ferritic steels and types H10H18 are austenitic steels.Nonalloy tool steels are preceded by the prefix letters TC and followed by a numeric code,which indicates 100 times of the average of the carbon content.The designation for alloy tool steels is equivalent to that of alloy structural steels.High-speed steels are preceded by the prefix letters HS and followed by a numeric code,which indicates the percentage content of alloy elements of W, Mo, V, and Co. For example,HS 2-9-1-8 indicates that steel contains 2% W, 9% Mo, 1% V, and 8% Co, respectively. Highspeed steel free from Mo uses numeric 0. If a high-speed steel were free from Co, then 0 wouldnot be added. For example, tungsten high-speed steel HS 18-0-1 contains 18% W, 0% Mo, 1%V, and 0% Co. Some steels of ISO designations are provided in Table GB DESIGNATIONS (STATE STANDARDSOFCHINA)The state standard of China for steels is called Guojia Biaozhun, abbreviated as GB. TheGB designations for nonalloy common steels and HSLA steels use the prefix letterQ, followed by the yield strength value (MPa). For example, Q235, Q345, Q390 denotenonalloy common steels and HSLA steels with their yield strength 235, 345, and 390 MPa,respectively.Nonalloy structural steels and alloy structural steels are represented by numeric codes,which are 100 times of the average carbon content. For example, numeric code 45 shows thesteel containing 0.45% C. Alloy elements in steel use the descriptive code with chemicalsymbols, and followed by its average content. As the average content is less than 1.5%, it isindicated only with the chemical symbol, for example, 34CrNi3Mo containing 0.300.40% C,0.701.10% Cr, 2.753.25% Ni, and 0.250.40% Mo.Nonalloy tool steels are represented by the prefix letter T, followed by numeric codes,which are ten times of the average carbon content. For example, T8 means the steel containsan average carbon content about 0.80%. When the average carbon content is more than 1.0%in alloy tool steels, the steel grade would not indicate the carbon content; but the averagecontent is less than 1.0%, it uses numeric code ten times carbon content. For example, CrMnsteel contains 1.301.50% C, 1.301.60% Cr, 0.450.75% Mn, and 9Mn2V steel contains 0.850.95% C, 1.702.00% Mn, 0.10 to 0.25% V. The descriptive method for the alloy element is thesame in alloy structural steels.Carbon content is not indicated in high-speed steels, and only uses the descriptive codewith chemical symbols and followed by alloy element content. For example, tungsten highspeed steel 18-4-1 (T1) is represented by W18Cr4V, and WMo high-speed steel 6-5-4-2 (M2)is indicated by W6Mo5Cr4V2.They are represented by a numeric code, which indicates ten times of the carbon contentand followed by chemical symbols of alloy elements with their content in stainless steels andheat-resistant steels, but microalloy elements only show the chemical symbols. For example,steel 9Cr18MoV contains 0.850.95% C, 1719% Cr, 0.01.3% Mo, 0.070.12% V. If thecarbon content is less than 0.03 or 0.08%, 00 or 0 would be used for the steel designations,respectively, such as 00Cr18Ni10 and 0Cr13. 2006 by Taylor & Francis Group, LLC.1.6.3 DIN STANDARDSDIN standards are developed by Deutsches Institut fuer Normung in Germany. All Germansteel standards and specifications are represented by the letters DIN and followed by analphanumeric or a numeric code. An uppercase letter sometimes precedes this code. Germandesignations are reported in one of the following two methods. One method uses the descriptivecode with chemical symbols and numbers in the designation. The second, called the Werkstoffnumber, uses numbers only, with a decimal after the first digit. There are four figures after thedecimal point, the first two of which are used to identify the alloy, and the last two the quantity.Most steels are covered by the significant figure 1, but some have no significant figure before thedecimal point. Examples of both methods are provided in Table 1.17, which cross-referencesDIN designations and indicates chemical composition for DIN steels. However, standards forheat-resistant steels are prefixed with the letter SEW (Stahl-Eisen-Werkstoff Blaetter, steel-ironmaterial sheets). Examples of DIN designations in both methods with equivalent UNS numbers in parentheses are as follows: DIN 40NiCrMo6 or DIN 1.6565 (G43400) is a NiCrMosteel that contains 0.350.45% C, 0.91.4% Cr, 0.50.7% Mn, 0.20.3% Mo, 1.41.7% Ni,0.035% S, and 0.150.35% Si; DIN 17200 1.1149 or DIN 17200 Cm22 (G10200) is a nonresulfurized carbon steel containing 0.170.24% C, 0.30.6% Mn, 0.035 max P, 0.020.035% S,and 0.4% max Si.1.6.4JIS STANDARDSJIS standards are developed by the Japanese Industrial Standards Committee (JISC) inTokyo, Japan. The specifications begin with the prefix JIS, followed by a letter G for carbonand low-alloy steels. This is followed by a space and series of numbers and letters indicatingthe particular steel. JIS designations are provided in Table 1.17. As examples of JIS designations with equivalent UNS-AISI numbers in parentheses, JIS G3445 STKM11A (G10080) is alow-carbon tube steel containing 0.12% C, 0.35% Si, 0.60% Mn, 0.04% P, and 0.04% S; JISG3445 STK 17A (G10490) is a medium-carbon nonresulfurized steel containing 0.450.55%C, 0.401.0% Mn, 0.04% P, 0.04% S, and 0.04% Si; JIS G4403 SKH2 (AISI T1 grade) is atungsten high-speed tool steel containing 0.730.83% C, 3.84.5% Cr, 0.4% Mn, 0.4% Si,0.81.2% V, and 1719% W; and JIS G4403 SKH59 (AISI M42 grade) is a molybdenumultrahard high-speed tool steel containing 11.15% C, 7.58.5% Co, 3.54.5% Cr, 0.4%max Mn, 910% Mo, 0.5% max Si, 0.91.4% V, 1.21.9% W, 0.25% max Ni, 0.03% maxP, 0.03% max S, and 0.25% Cu.1.6.5BS STANDARDSBS standards are developed by the British Standards Institute (BSI) in London, England. Theletters BS precede the standard numerical code, and, like JIS standards, each British designation covers a product form and an alloy code. Table 1.17 lists steels identified by Britishstandards. Some example of BS designations with equivalent AISI designations in parentheses are given: BS 970 708A30 (4130) is a CrMo low-alloy steel containing 0.280.33% C,0.91.2% Cr, 0.40.6% Mn, 0.150.25% Mo, 0.035% P, 0.04% S, and 0.10.35% Si; and BS970 304S15 (304) is a wrought austenitic stainless steel (sheet, strip, plate) containing 0.06% C,17.519% Cr, 0.52.0% Mn, 811% Ni, 0.05% P, 0.03% S, and 0.21.0% Si.1.6.6AFNOR STANDARDSAFNOR standards are developed by the Association Francaise de Normalisation in Paris,France. The AFNOR standards, which are given in Table 1.17, usually begin with the letters 2006 by Taylor & Francis Group, LLC.NF, followed by an alphanumeric code constituting an uppercase letter followed by a series ofdigits, which are subsequently followed by an alphanumeric sequence. For example, resulfurized (free-cutting) steel is listed in AFNOR NF A35562 standard or specification, and35MF6 designation (equivalent to SAE 1137) represents the steel bar containing 0.330.39% C, 1.301.70% Mn, 0.100.40% Si, 0.040% P, and 0.090.13% S; whereas 45MF4designation (equivalent to SAE 1146) contains 0.420.49% C, 0.81.1% Mn, 0.10.4% Si,0.04% max P, 0.090.13% S. Similarly, AFNOR NF A35-573 Z6CN 18.09 is a wrought (SAE304) stainless steel (sheet, strip, plate) and contains 0.07% C, 1719% Cr, 2% Mn, 810% Ni,0.04% P, 0.03% S, and 1% Si.REFERENCES1. J.R. Davis, Ed., Concise Metals Engineering Data Book, ASM International, Materials Park, OH,1997, p. 44.2. W.M. Garrison, Jr., Steels: classification, in Encyclopedia of Materials: Science and Technology,K.H.J. Buschow, R.W. Cahn, M.C. Flemings, B. Ilschner, E.J. Kramer, and S. Mahajan, Eds.,Elsevier, Amsterdam, 2001, pp. 88408843.3. H. Okamoto, CFe, in Binary Alloy Phase Diagrams, 2nd ed., T.B. Massalski, Ed., ASM International, Materials Park, OH, 1990, pp. 842848.4. A.K. Sinha, Ferrous Physical Metallurgy, Butterworths, London, 1989.5. S. Zhang and C. Wu, Ferrous Materials, Metallurgical Industry Press, Beijing, 1992.6. Anon., Classification and designation of carbon and low-alloy steels, ASM Handbook, 10th ed., Vol. 1,ASM International, Materials Park, OH, 1990, pp. 140194.7. H. Lin, G. Lin, and Y. Ma, Eds., Designation and Trade Name Handbook of Steels Worldwide,Mechanical Industry Press, Beijing, 1997.8. Anon., Carbon and alloy steels, SAE J411, 1993 SAE Handbook, Vol. 1, Materials Society ofAutomotive Engineers, Warrendale, PA, pp. G. Krauss, SteelsHeat Treatment and Processing Principles, ASM International, Materials Park,OH, 1990.10. W.C. Leslie, The Physical Metallurgy of Steels, McGraw-Hill, New York, 1981.11. E.C. Bain and H.W. Paxton, Alloying Elements in Steel, American Society for Metals, Cleveland,OH, 1966.12. R.W.K. Honeycombe, SteelsMicrostructure and Properties, Adward Arnold, London, 1982.13. H.E. Boyer, in Fundamentals of Ferrous Metallurgy, Course 11, Lesson 12, Materials EngineeringInstitute, ASM International, Materials Park, OH, 1981.14. R.B. Ross, Metallic Materials Specification Handbook, 4th ed., Chapman & Hall, London, 1992.15. C.W. Wegst, Stahlschlussel (Key to Steel ), Verlag Stahlschlussel Wegst GmbH, 1992.16. W.J. McG. Tegart and A. Gittins, in Sulfide Inclusions in Steel, J.T. Deabradillo and E. Snape, Eds.,American Society for Metals, Cleveland, OH, 1975, p. 198.17. C.W. Kovach, in Sulfide Inclusions in Steel, J.T. Deabradillo and E. Snape, Eds., American Societyfor Metals, Cleveland, OH, 1975, p. 459.18. C.M. Lyne and A. Kazak, Trans. ASM 61: 10 (1968).19. D. Brovoksbank and K.W. Andraws, JISI 206: 595 (1968).20. N.S. Stoloff, in Fracture Vol. VI, Fracture of Metals, H. Liebowitz, Ed., Academic Press, New York,1969.21. J. Yu, Z. Yu, and C. Wu, J. Metals 40(5): 2631 (1988).22. S. Zhang and C. Wu, in Proceedings of 3rd International Congress on Heat Treatment of Materials,Shanghai, 1983, pp. T.Y. Hsu (Zuyao Xu), ISIJ Int. 38: 11531164 (1998).24. M.J.U.T. Van Wijngaarden and G.P. Visagie, in Proceedings of the 79th Steelmaking Conference,Vol. 79, Pittsburgh Meeting, March 2427, pp. 627631.25. H. Matsuka, K. Osawa, M. Ono, and M. Ohmura, ISIJ Int. 37: 255262 (1997). 2006 by Taylor & Francis Group, LLC.26. W.D. Landford and H.E. McGannon, Eds., The Making, Shaping, and Treating of Steel, 10th ed.,U.S. Steel, Pittsburgh, PA, 1985.27. Steel Products Manual: Plates; Rolled Floor Plates: Carbon, High Strength Low Alloy, and AlloySteel, American Iron and Steel Institute, Washington, D.C., 1985.28. E. Hornbogen, in Physical Metallurgy, R.W. Cahn and P. Haasen, Eds., Elsevier, New York, 1983,pp. 10751138.29. J.D. Smith, in Fundamentals of Ferrous Metallurgy, Course 11, Lessons 2 and 5, Materials Engineering Institute, ASM International, Materials Park, OH, 1979.30. Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials, Society ofAutomotive Engineers, Warrendale, PA, pp. W.F. Smith, Structure and Properties of Engineering Alloys, McGraw-Hill, New York, 1981.32. O.D. Sherby, B. Walser, C.M. Young, and E.M. Cady, Scr. Metall. 9: 596 (1975).33. E.S. Kayali, H. Sunada, T. Oyama, J. Wadsworth, and O.D. Sherby, J. Mater. Sci. 14: 26882692 (1979).34. D.W. Kum, T. Oyama, O.D. Sherby, O.A. Ruano, and J. Wadsworth, Superplastic Forming,Conference Proceedings, 1984, American Society for Metals, Cleveland, OH, 1985, pp. 3242.35. T.A. Phillip, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990,pp. 430448.36. A.K. Sinha, in Production of Iron/Steel and High Quality Product Mix (Conf. Proc. 1991), ASMInternational, Materials Park, OH, 1992, pp. 195206.37. L.F. Porter and P.E. Repas, J. Metals 34(4): 1421 (1982).38. L.F. Porter, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford, 1986,pp. 21572162.39. H. Zhang, in Chinese Encyclopedia of Metallurgy, Metallic Materials, Metallurgical Industry Press,Beijing, 2001, pp. 152154.40. P. Charlier and L. Bacher, Proc. Seventh Int. Conf. Strength of Metals and Alloys, Vol. 2, Montreal,Canada, August 1216, 1986, Pergamon Press, Oxford, p. 1019.41. P. Charlier and A. Guimier, Proc. Int. Sem. Automotive Steels, Moscow, April 1921, 1988, p. 167.42. A.H. Nakagawa and G. Thomas, Metall. Trans. 16A: 831840 (1985).43. D.Z. Yang, E.L. Brown, D.K. Matlock, and G. Krauss, Metall. Trans. 16A: 15231526 (1985).44. R.C. Davies, Metall. Trans. 10A: 113118 (1979).45. A.R. Marder, Metall. Trans. 13A: 8592 (1982).46. A.M. Bayer and L.R. Walton, in ASM Handbook, 10th ed., Vol. 1, ASM International, MaterialsPark, OH, 1990, pp. 757779.47. M.G.H. Wells, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford,1986, pp. 51155120.48. R. Higgins, Engineering Metallurgy, 5th ed., Krieger, Malabar, FL, 1983.49. S.D. Washko and G. Aggen, in ASM Handbook, 10th ed., Vol. 1, ASM International, MaterialsPark, OH, 1990, pp. 841907.50. R.F. Decker, J.T. Each, and A.J. Goldman, Trans. ASM 55: 58 (1962).51. S. Floreen, Met. Rev. 13: 115128 (1968).52. S. Floreen, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford, 1986,pp. 51715177.53. T. Morrison, Metall. Mater. Technol. 8: 8085 (1976).54. K. Rohrbach and M. Schmidt, in ASM Handbook, 10th ed., Vol. 1, ASM International, MaterialsPark, OH, 1990, pp. 793800.55. Heat Treaters GuideStandard Practices and Procedures for Steel, ASM, Cleveland, OH, 1982.56. Metals and Alloys in the Unified Numbering System, 6th ed., Society of Automotive Engineers,Warrendale, PA, 1993.57. Worldwide Guide to Equivalent Irons and Steels, ASM International, Materials Park, OH, 1992.58. 1994 SAE AMS Index, Society of Automotive Engineers, Warrendale, PA.59. Specification for Drill Pipe, API Specification 5D, 3rd ed., August 1, 1992, American PetroleumInstitute, Washington, D.C.60. Specification for Line Pipe, API Specification 5L, 40th ed., November 1, 1992, American PetroleumInstitute, Washington, D.C. 2006 by Taylor & Francis Group, LLC.61. Specification for CRA Line Pipe, API Specification 5LC, 2nd ed., August 1, 1991, AmericanPetroleum Institute, Washington, D.C.62. 1994 Publications, Programs & Services, American Petroleum Institute, Washington, D.C.63. D.L. Potts and J.G. Gensure, International Metallic Materials Cross-Reference, Genium Publishing,New York, 1989. 2006 by Taylor & Francis Group, LLC.2Classification and Mechanismsof Steel Transformation*S.S. BabuCONTENTS2.12.22.3Introduction ................................................................................................................ 91Phase Transformation Mechanisms ............................................................................ 92Microstructure Evolution during Austenite Decomposition....................................... 962.3.1 Allotriomorphic Ferrite .................................................................................... 962.3.2 Widmanstatten Ferrite...................................................................................... 972.3.3 Bainite............................................................................................................... 992.3.4 Pearlite............................................................................................................. 1012.3.5 Martensite........................................................................................................ 1032.4 Microstructure Evolution during Reheating .............................................................. 1062.4.1 Tempered Martensite....................................................................................... 1072.4.1.1 Carbon Segregation and Aging of Martensite .................................. 1072.4.1.2 First Stage of Tempering................................................................... 1072.4.1.3 Second Stage of Tempering............................................................... 1072.4.1.4 Third Stage of Tempering ................................................................. 1082.4.1.5 Fourth Stage of Tempering ............................................................... 1082.4.2 Austenite Formation ....................................................................................... 1092.5 Summary of Steel Microstructure Evolution ............................................................. 1092.6 Prediction of Microstructure Evolution during Heat Treatment ............................... 1102.6.1 Calculation of Multicomponent Multiphase Diagrams................................... 1122.6.2 Calculation of Diffusion-Controlled Growth .................................................. 1132.7 Summary.................................................................................................................... 1162.8 Acknowledgments ...................................................................................................... 118References .......................................................................................................................... 1182.1 INTRODUCTIONThe microstructure of steels consists of a spatial arrangement of crystalline aggregates ofdifferent phases. The size, shape, distribution, composition, and crystal structure of thesephases essentially control the final properties of any given steel, including hardness, strength,ductility, impact toughness, and creep strength. Steel is the most versatile alloy among all theindustrial alloys. It exhibits a diverse range of microstructures that possess different combinations of strengths and toughnesses. In a majority of the steels, this versatility is made possible* The submitted manuscript has been authored by a contractor of the U.S. Government under contract DE-AC0500OR22725. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproducethe published form of this contribution, or allow others to do so, for U.S. Government purposes. 2006 by Taylor & Francis Group, modifying the decomposition of a high-temperature d-ferrite (body-centered cubic [bcc]crystal structure) to high-temperature austenite phase (face-centered cubic [fcc] crystal structure) and then decomposition of austenite to a low-temperature a-ferrite (bcc) phase bychanging the composition and cooling rate. In low-alloy steels, the most important phasechange is from austenite to a-ferrite. For example, for a given composition of low-alloy steel,a ferritepearlite microstructure can be obtained by slow cooling. With an increase in coolingrate, Widmantstatten or upper bainite microstructure can be obtained from austenite. With afurther increase in the cooling rate, a hard martensite microstructure can be obtained. Pearlitecontains a lamellar aggregate of ferrite and cementite phases, upper bainite contains ferriteplatelets separated by austenite or carbides, and martensite contains carbon-supersaturatedferrite platelets with a high density of dislocations or twin boundaries.Another feature of steels is that all of the above microstructures are far from equilibrium(i.e., the equilibrium microstructure at room temperature is a mixture of ferrite and graphite).The ferritepearlite microstructure, which contains ferrite orthorhombic cementite lamella,is the closest to the equilibrium of the above structures. In contrast, the martensite microstructure has a body-centered tetragonal (bct) crystal structure with supersaturated carbon and is thefarthest from equilibrium. Therefore, tempering martensite at high temperature can be used toattain intermediate-phase mixtures that are closer to final equilibrium, thus providing anothermethodology to control the microstructure and properties of the steel. The methods of controlling steel microstructure through alloy modification and heat treatment have been followedby metallurgists and this will continue to be the goal of material scientists. The focus of thischapter is to extend this approach to new classes of steels without trial and error experimentation to heat treatment based on the mechanisms of phase transformations.Over the decades, mechanisms of phase transformations that occur in steel have beenillustrated in a simple FeC metastable diagram (Figure 2.1) that describes the stability regionsfor ferrite, austenite, and cementite structures. On cooling from the liquid region, the first phaseto solidify is d-ferrite. With continued cooling, the d-ferrite transforms to g-austenite. Withfurther cooling, the austenite transforms to a-ferrite and cementite. Basic and applied researchin the past has led to a fundamental understanding of these structural changes and theirrelationship to microstructure evolution in alloys ranging from simple FeC systems to complex FeCX steels (where X stands for many different substitutional alloying elements,including manganese, nickel, chromium, silicon, and molybdenum). The relationships betweencrystal structure changes, the interface structure, the phase morphologies (i.e., ferrite orbainite), and the distribution of alloying elements between phases are understood. In addition,computational thermodynamic and kinetic tools have been developed to the extent that it ispossible to predict transformation kinetics quantitatively as a function of steel composition. Inthis chapter, a framework to classify transformation mechanisms is introduced. Then, the steelmicrostructures are classified and discussed based on this framework. Finally, a brief introduction is given to the application of computational thermodynamics and kinetics to describephase transformations in low-alloy steels.2.2PHASE TRANSFORMATION MECHANISMSThe various phase transformations that occur in steels can be classified according to athermodynamic basis, a microstructural basis, and a mechanistic basis [1]:1. The thermodynamic basis classifies the phase transformation based on the derivativesof the Gibbs free energy (G) of the system with temperature change at a constantpressure. If the first derivative shows a discontinuity at a transformation temperature,it is called first-order transformation. For example, the transformation of a solid to a 2006 by Taylor & Francis Group, LLC.bccLiquid-ferrite1600 +L1400+L+fcc1200Temperature (8C)-austenitebcc1000-ferrite800 ++600-ferrite+-cementite4002000. percentage carbon1.0FIGURE 2.1 Illustration of body-centered cubic (bcc) ferrite and face-centered cubic (fcc) austenite with acalculated metastable FeC binary diagram showing the phase stability of austenite, ferrite, and cementite.liquid of pure metal at melting temperature is a first-order transformation. Similarly,in other systems, with continued differentiation of G, and the phase transformation isclassified as the nth order if the discontinuity shows up after n differentiations(dn G=dT n )P . An example of a second-order transition is the ordering of the bcc ferritephase in an FeAl system.2. The microstructural basis relates to the origin of a product phase from a parent phase. Ifthe product phase forms everywhere within a sample without the need for any nucleation,it is classified as a homogeneous transformation. If the product phase forms as a smallentity with a sharp interface and if it grows into the product phase, it is classified as aheterogeneous transformation through a nucleation and growth process. Microstructural evolution in steel is predominantly a heterogeneous transformation.3. The mechanistic basis relates to the way the crystal structure change is achieved duringtransformation. The heterogeneous transformation in a material can occur by threemechanisms [2]: (a) athermal growth through glissile interfaces (e.g., austenite tomartensite formation), (b) thermally activated growth (e.g., pearlite formation fromaustenite), and (c) growth controlled by heat transport (e.g., solidification). Details ofeach subclassification of growth mechanisms have been discussed by Christian [2] andare shown in Figure 2.2.In summary, the transformation in metals and alloys can be classified based on thermodynamic, microstructural, or growth mechanisms. The heterogeneous transformation can be 2006 by Taylor & Francis Group, LLC.94 2006 by Taylor & Francis Group, LLC.HeterogeneoustransformationsAthermal growth(glissile interfaces)CoherentinterfaceSemi-coherentinterfaceMartensite(steels)MechanicaltwinningLow-angleboundaryNo long-rangetransport(interface controlled)Polymorphictransitionsgrowth from vapourRecrystallizationgrain growthOrderdisorderchangesLong-rangetransportDiscontinousreactionsdiffusion +interface controlled?ContinuousreactionsInterfacecontrolledCrystal growthfrom vapourSolidificationmeltingDiffusioncontrolledPrecipitationdissolutionEutectoidalreactionDiscontinuousprecipitationPrecipitationdissolutionFIGURE 2.2 Classification of heterogeneous transformation based on mechanisms of growth. Most relevant phase transformation mechanisms for steel heattreatment are encircled. (Adapted from J.W. Christian, The Theory of Transformations in Metals and Alloys, 2nd ed., Part 1, Pergamon Press, Oxford, 1981, p. 9.)Steel Heat Treatment: Metallurgy and TechnologiesCoherentmartensiteGrowth controlledby heat transportThermally activatedgrowthcontrolled by glissile interface motion, thermally activated interface movement, or heattransport. The relevant mechanism for steel microstructure evolution during heat treatmentis a heterogeneous phase transformation involving glissile interface (displacive) or long-rangetransport (reconstructive).The reconstructive transformation involves mixing of atoms on either side of the parent andproduct phase interface through diffusion. The displacive transformation involves coordinatedatom motion in the parent phase to change the parent crystal structure to the product crystalstructure. These reconstructive and displacive transformation mechanisms in steels are discussed with a schematic illustration of austenite decomposition to ferrite or martensite as shownin Figure 2.3 [3]. Let us assume that the austenite fcc lattice is represented by the square motifshown in the upper left part of Figure 2.3 bounded by a rectangle abcd. Various substitutional atoms such as silicon, manganese, chromium, and molybdenum are schematicallydescribed by different symbols. Initial arrangements of some atoms are denoted by numbers 1to 6. During the decomposition of austenite to ferrite or martensite, a crystal structure changetakes place at the interface between austenite and the product phase (represented by line ef).The schematic illustration of reconstructive transformation is shown in the bottom of the figure,and the bcc lattice of ferrite is represented by a diamond motif. This crystal structure changefrom fcc to bcc occurs at the ef interface and is accompanied by somewhat random hopping ofatoms across the interface. Because of this hopping of the atoms across the interface, the originalatomic arrangements (implied by the change in the position of 16 numbered atoms) are lost. Asa result, the transformation involves the reconstruction of an austenite lattice into a ferritelattice, which leads to no macroscopic shape change of the crystal and mixing or separation ofatoms on either side of the austenite and ferrite that would lead to a change in the composition ofthe phase. However, there is a distinct change in volume. Based on the above facts, the ratetransformation is expected to be controlled by long-range diffusion of substitutional atoms inboth the ferrite and austenite lattices. In practice, the transformations in steels are complicatedby the presence of carbon, an interstitial element, which is not shown in the Figure 2.3. Thegrowth rate of ferrite (which has low solubility of carbon) is controlled by diffusion of bothcarbon in austenite and substitutional atoms in ferrite and austenite. Under reconstructivetransformation, diffusion of all atoms occurs during nucleation and growth and is usuallysluggish below 6008C [4].AusteniteAusteniteadf612345cade1bInterface6234Displacive transformation5cbAtomic correspondenceIPS shape change with asignificant shear componentDiffusionlessMartensiteAustenite1d2a5ef4InterfaceReconstructive transformation3No atomic correspondenceNo shape change with shear componentPossible composition change6bcFerriteFIGURE 2.3 Schematic illustration of crystal structure change due to reconstructive and displacivetransformations. (Adapted from H.K.D.H. Bhadeshia, Worked Examples in the Geometry of Crystals,The Institute of Materials, London, 1987.) 2006 by Taylor & Francis Group, LLC.The schematic illustration of martensite formation is shown in the upper right side ofFigure 2.3, and the bct lattice is represented by a diamond motif. The main difference in thiscase, compared to the fcc to bcc transformation, is that the crystal structure change from fccto bct lattice is brought about by coordinated atom movement at the interface ef, or in otherwords, a shear deformation of the original fcc lattice. As a result, the original atomiccorrespondences within the parent phase are maintained in the product phase. For example,the row of atoms 16 is just displaced laterally parallel to interface ef, and the atomicsequence remains the same. This mode of transformation involving a shear deformation iscalled invariant plane strain (IPS). IPS leads to a macroscopic shape change, as indicated inthe schematic illustration. This macroscopic change (from abcd to aebcfd) manifestsitself as a surface relief on the polished surface of the sample after the decomposition ofaustenite to martensite. For more crystallographic details, the reader is referred to theliterature [57]. The displacive transformation is also associated with a large strain energy.Therefore, to sustain this reaction, the chemical thermodynamic driving force must be large,and as a result, the martensite transformation occurs only at low temperatures where themagnitude of chemical free energy increases above the strain energy requirement. The largestrain energy also induces the lenticular or plate-shaped region of martensite to minimize thestrain energy. This mode of transformation also implies that both interstitial and substitutional atoms are trapped within the martensite and that there is no change in compositionbetween the austenite and martensite phases (i.e., it is a diffusionless mode of transformation). As a result, the growth rate of martensite is usually athermal. The Widmanstatten andbainitic ferrite transformations occur at higher temperatures than the martensite transformation. They exhibit displacive transformation for a crystal structure change and varyingdegrees of diffusion of interstitial carbon during nucleation and growth.In the following sections, the steel microstructure evolution during cooling (austenitedecomposition) and heating (tempering and austenite formation) will be described based onreconstructive and displacive transformation mechanisms.2.3 MICROSTRUCTURE EVOLUTION DURING AUSTENITE DECOMPOSITIONThere have been numerous reviews of austenite decomposition over the past five decades[810]. The ferrite morphologies, which form during austenite decomposition, were originallyclassified by Dube [11,12]. A change in morphology from one to another was found to occuras the austenite decomposition temperature was lowered. The common ferrite morphologiesare grain boundary allotriomorphic ferrite, idiomorphic ferrite, Widmanstatten ferrite, andintragranular ferrite [13,14]. Other microstructures that form at lower transformation temperatures are pearlite, bainite, and martensite. The evolution of each one of the abovemicrostructure is discussed below.2.3.1ALLOTRIOMORPHIC FERRITEDecomposition of austenite to allotriomorphic ferrite occurs over a wide range of temperatures below the austenite to austenite ferrite phase boundary (see Figure 2.1). This section isnot intended to be a comprehensive review of this subject, but rather a short summary ofextensive literature. The reader can consult the classic reviews by Aaronson and otherresearchers for details [10,13].Allotriomorphic ferrite usually nucleates along the austeniteaustenite grain boundary. Itfirst grows laterally along the boundary and then can proceed perpendicularly into theaustenite grain. The nucleation and growth involve a reconstructive mode of crystal structurechange that leads to an absence of any macroscopic shape change, and only a volume change 2006 by Taylor & Francis Group, observed. During nucleation, the allotriomorphic ferrite crystals exhibit a preferred orientation relationship with one of the austenite grains (g1). The orientation relationship is usuallyof the KurdjumovSachs (KS) type:{1 1 1}g =={1 1 0}aandh1 " 0ig ==h1 " 1ia :11The KS relationship indicates that the close-packed planes of austenite and ferrite areparallel to each other and that the close-packed directions of austenite and ferrite are parallelto each other. On some occasions, the ferrite may deviate slightly from having close-packedparallel directions, as shown by the NishiyamaWasserman (NW) orientation relationship:{1 1 1}g =={1 1 0}aand"h1 1 2ig ==h1 " 0ia :1The KS or NW orientations allow for the existence of a semicoherent boundary betweenaustenite and ferrite and thereby minimize the surface energy required to nucleate the ferriteat the austenite boundary. A small free energy is sufficient to satisfy the KS and NWorientations with one of the austenite grains for the nucleation of a ferrite, and then theferrite can grow into the adjacent austenite grain (g2 ) with no specific orientation relationship.On the latter side of the interface, which shows the random orientation, a disorderedboundary will exist and rapidly grow into the austenite. An example of such ferrite growthin an FeCSiMn steel is shown in Figure 2.4. The optical micrograph (Figure 2.4a) showsthe allotriomorphic ferrite grains along the prior austenite grain boundary (marked byarrow). The remaining microstructure is bainitic ferrite. A transmission electron microscopyimage of this sample (see Figure 2.4b) shows an allotriomorphic ferrite grain situated alongthe boundary (marked as B) of two austenite grains (g1 and g2 ). Electron diffraction analysisof the above structure indicated that the allotriomorphic ferrite had a KS orientationrelationship with a g1 grain and that there was no special orientation relationship with g2 .Further detailed observation also showed small steps in the interface along the protuberanceson the g1 side of the allotriomorphic ferrite. However, the interface on the g2 side showeddisordered boundary with no specific structure [15]. The rate of growth on the g2 side is morerapid than that on the g1 side. This apparent difference in growth rate was shown by classic insitu transmission electron microscopy experiments by Purdy [16]. The observed semicoherentboundary on only one side of the austenite grain also plays a critical role in the developmentof secondary Widmanstatten ferrite, which is described in the following section.2.3.2 WIDMANSTATTEN FERRITEWidmanstatten ferrite describes a ferrite morphology in the form of side plates or laths andgrows into austenite grains with a KS orientation relationship. The lath ferrite, which formsfrom the austenite grain boundary, is referred to as primary Widmanstatten ferrite. Theferrite that nucleates on the preexisting allotriomorphic ferrite is referred to as secondaryWidmanstatten ferrite [14]. Nucleation of secondary Widmanstatten ferrite occurs mostly onthe semicoherent interphase boundary between the ferrite and the austenite. A typicalmicrostructure of secondary Widmanstatten ferrite in a steel weld is shown in Figure 2.5a.A schematic illustration of such a microstructure evolution is shown in Figure 2.5b. As thesteel cools from high temperature, the allotriomorphic ferrite forms with a KSNW relationship with austenite grain g1 and has a semicoherent interface boundary. With further cooling,secondary Widmanstatten ferrite nucleates on the g1 side of the allotriomorphic ferrite andgrows into the austenite grain. This nucleation and growth process leads to apparent sawtooth morphology. 2006 by Taylor & Francis Group, LLC.a20 mbB211 mFIGURE 2.4 (a) Optical micrograph showing the presence of allotriomorphic ferrite (marked by arrow)and bainite microstructure in an FeCSiMn steel. (b) Transmission electron micrograph showing thepresence of allotriomorphic ferrite along an austenite grain boundary (g1 and g2 ) marked by b.The orientation relationship between allotriomorphic ferrite and the g1 grain was close to a KSNWorientation relationship and was in a random orientation relationship with the g2 grain.Two growth mechanisms of Widmanstatten ferrite have been proposed. In the firstmechanism, the growth is attributed to lateral movement of semicoherent interfaces bysmall steps (ledges) in the interface [17,18]. In the second mechanism, the Widmanstattenferrite may grow through a displacive transformation mechanism [19]. Many experimentalresults are in agreement with the second mechanism. Thin, wedge-shaped Widmanstattenferrite is produced due to cooperative, back-to-back growth of two ferrite crystallographicvariants. The Widmanstatten ferrite plates grow into untransformed austenite by extensionalong their length. Since Widmanstatten ferrite forms at temperatures well below that forallotriomorphic ferrite, growth occurs by a paraequilibrium (PE) mode (i.e., the ratio of the 2006 by Taylor & Francis Group, LLC.aAllotriomorphicferriteWidmanstttenferrite20 mOriginalallotriomorphicferritebWidmanstttenferritePrior austenitegrain boundary12FIGURE 2.5 (a) Optical micrograph showing secondary Widmanstatten ferrite microstructure in an FeCMn steel weld obtained through continuous cooling. The growth of these plates occurs on only oneside of the allotriomorphic ferrite. (b) Schematic illustration showing possible mechanism for theformation of Widmanstatten ferrite microstructure on only one side of the original allotriomorphicferrite.iron concentration to the substitutional atom concentration is frozen in both the parent andthe product phases). The rate of growth is controlled only by carbon diffusion in the austeniteahead of the plate. It is possible to estimate the growth rate of Widmanstatten ferrite withdiffusion-controlled growth models of Bhadeshia et al. [20] and Trivedi [21,22] by consideringonly the carbon diffusion.2.3.3 BAINITEThe growth of bainitic ferrite has been discussed in the literature based on either reconstructive or displacive transformation mechanisms. In the reconstructive definition, the growth ofbainite is the product of diffusional, noncooperative, competitive ledgewise growth of ferriteand cementite into austenite during eutectoid decomposition with cementite appearing in anonlamellar form [23]. In this definition, the transformation kinetics is related to the rate of 2006 by Taylor & Francis Group, LLC.ledge movement at the interface and is controlled by carbon diffusion. In the displacivedefinition, a subunit of ferrite forms from the austenite with complete supersaturation ofcarbon through a displacive transformation involving IPS deformation. Carbon diffusion toaustenite occurs through a post-transformation event. The overall transformation kinetics isthen related to the nucleation of this ferrite subunit. The bainitic subunits are expected toshow surface relief, no substitutional partitioning, an incomplete reaction phenomenon, and aKSNW orientation relationship [24]. They are also expected to show compliance with anexternally applied elastic stress on the austenite similar to martensite [25]. The reader canconsult the recent discussions by Hillert [26] for current research in bainite transformationmechanisms. In this section, the bainitic transformation mechanisms are discussed from thedisplacive mode of transformation point of view [4].The bainite microstructure consists of aggregates of ferrite plates separated by thin filmsof austenite, martensite, or cementite. These aggregates of plates are called sheaves [4].A typical bainitic sheaf in an FeCrC steel is shown in Figure 2.6a. This microstructurewas attained by austenitization followed by isothermal transformation below the bainitic starttemperature Bs [25]. The small ferrite plates that make up the sheaves are often referred to assubunits, and often they are related to each other through a specific crystallographic orientation. These subunits are normally plate-shaped and under certain conditions may exhibit alath structure. The presence of subunits within the sheaf can also be seen by transmissionelectron microscopy. Transmission electron microstructures of an FeCSiMn steel (seeFigure 2.6b) and an FeCrC steel (see Figure 2.6c) show the subunit ferrite plates within aab1 m50 mBainitic sheafdc1 mAustenite/carbideSubunitAustenite grain boundaryFIGURE 2.6 (a) Optical micrograph from an FeCrC steel showing a bainitic sheaf structure.(b) Transmission electron micrograph from an FeCSiMn steel showing many sheaves of bainiticferrite. (c) Another transmission electron micrograph of a bainitic sheaf structure from an FeCrCsteel. (d) Schematic illustration of the growth mechanism by subunit nucleation and growth. (Adaptedfrom H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., The Institute of Materials, London, 2001.) 2006 by Taylor & Francis Group, LLC.bainitic sheaf. The first ferrite subunit that nucleates on the austeniteaustenite grain boundary grows to a particular size, and successive plates form to produce the microstructureshown in Figure 2.6d. The plates are separated by untransformed austenite, which cantransform to martensite on cooling to low temperature. In certain conditions, this austenitecan decompose into cementite, giving rise to a classical ferrite cementite mixture of bainiticmicrostructure.The bainitic steels can occur in two different forms, upper bainite and lower bainite.The microstructures shown in Figure 2.6 are examples of the upper bainite. The upper bainiteforms at temperatures higher than the martensite start (Ms ) temperature. Because of the IPSthat accompanies the bainitic transformation, a large number of dislocations are observedwithin the austenite and the ferrite. These dislocations limit the growth of each subunit to acertain size.Due to the displacive mode of transformation, the substitutional and interstitial atoms areconfigurationally frozen and there is no redistribution of these elements between austenite andbainitic ferrite during transformation. However, after the bainite reaches a certain size, thecarbon mobility is sufficient to redistribute to austenite while substitutional atoms remainfrozen in place. This carbon partitioning from the bainite to austenite enriches the austenitewith the progress of bainitic transformation. The successive nucleation of subunits from thiscarbon-enriched austenite continues to occur. The successive spatial alignment of nucleatedsubunits occurs due to an autocatalytic mechanism that leads to a typical sheaf structure. Whenthe carbon enrichment of austenite increases above a critical level, the displacive mode of bainiteformation cannot be sustained (the free energy of ferrite is equal to that of austenite) and thereaction stops (i.e., an incomplete reaction). This residual austenite may remain untransformedin the form of films between the ferrite subunits or may transform to martensite on cooling,depending upon the carbon concentration. In certain steels containing strong carbide formerssuch as chromium, this austenite film may decompose to a mixture of ferrite and carbide.In certain steels, most of the carbides are present within the bainitic than in between bainitesubunits and are referred to collectively as lower bainite. Typically, lower bainite forms in highcarbon steels, below the temperature at which upper bainite forms and above Ms . Although, thelower bainite microstructure is similar to that of tempered martensite, the main difference isthe occurrence of only one crystallographic variant* of the carbide in lower bainite. When thesecarbides are cementite, the major axis of the cementite is aligned at ~608 to the long axis of theferrite plates. These cementite plates appear to form on {1 1 2}a ferrite planes. In certain steels,lower bainitic ferrite has e-carbides rather than cementite. It is noteworthy that both upper andlower bainites form from austenite through a displacive transformation. The only difference isthat, in the case of upper bainite, the carbon from the bainitic subunit partitions into austenitebefore the carbide precipitates within the subunit. In the case of lower bainite, carbide precipitation occurs within the ferrite before all the carbon partitions completely from the ferrite. Thiscompetition between carbon escape from supersaturated ferrite plate and precipitation of carbidewithin the ferrite plates is schematically shown in Figure 2.7 [4].2.3.4 PEARLITEA typical pearlite microstructure has many colonies of a lamellar mixture of ferrite andcementite (see Figure 2.8a). Under an optical microscope, each colony of pearlite (schematically shown in Figure 2.8b) may appear to be made up of many alternating crystals of ferriteand cementite; however, they are mostly interpenetrating single crystals of ferrite and cementite in three dimensions. The ferrite and cementite within the pearlite colony show preferred* In contrast, tempered martensite shows many crystallographic variants of carbides. 2006 by Taylor & Francis Group, LLC.Carbon supersaturated plateCarbon diffusion intoaustenite and carbideprecipitation in ferriteCarbondiffusion intoausteniteCarbon precipitationfrom austeniteUpper bainite(high temperature)Lower bainite(low temperature)FIGURE 2.7 Schematic illustration of upper and lower bainite formation mechanism. The dark regionsbetween the plates represent carbides that form in the residual austenite. The dark lines representcarbides that form within the ferrite plates. (Adapted from H.K.D.H. Bhadeshia, Bainite in Steels,2nd ed., The Institute of Materials, London, 2001.)a500310 mAustenite grain boundarybFerritePearlite colonyCementiteFIGURE 2.8 (a) A typical optical microstructure of a pearlite in an FeCMn steel. Several distinctcolonies are marked with arrows. (b) Schematic illustration of pearlite colony growth. 2006 by Taylor & Francis Group, LLC.crystallographic relationships. The most common orientation relationships are PitschPetchrelationship [14]:513(0 0 1)cementite == " 2 " ferrite ; (0 1 0 )cementite 2 3 from 1 1 " ferrite ; (1 0 0 )cementite2 3 from [1 3 1 ]ferrite ,and the Bagaryatski relationship,1 0 0 cementite ==[0 1 "]ferrite ; [0 1 0 ]cementite ==[" 1 1]ferrite ; (0 0 1)cementite ==(2 1 1)ferrite :11The pearlite grows into austenite by cooperative growth of ferrite and cementite. Neither theferrite nor the pearlite shows any preferred crystallographic orientation with the austenite intowhich they are growing. The colony interface with austenite is an incoherent high-energyinterface. As a result, the pearlite that nucleates on preexisting allotriomorphic ferrite alwayschooses the ferrite side (high-energy interface) that has no crystallographic relationship with anaustenite grain. In contrast, Widmanstatten ferrite and bainite always nucleate on low-energyinterfaces (see Section 2.3.2 and Section 2.3.3). There is a well-known relationship betweenpearlite colonies and transformation conditions. The spacing of lamellae decreases with adecrease in the transformation temperature. It is possible to determine the lamellar spacing byequating the increase in interfacial energy to a decrease in energy due to transformation. Thepearlite transformation occurs very close to thermodynamic equilibrium (i.e., partitioning ofalloying elements occurs under local equilibrium), and often occurs at higher temperaturesand exhibit sluggish growth rates in comparison to bainite. Therefore, it is possible to use stateof-the-art computational thermodynamic and kinetic tools to predict the growth rate of pearlite.2.3.5 MARTENSITEMartensite forms from austenite through a displacive transformation. This section brieflyoutlines the morphology, nucleation, and growth of martensite and the crystallographicaspects of the martensite transformation. This section is not a comprehensive review of thissubject; rather, it is a short summary of extensive existing literature [7].The martensite transformation occurs athermally below Ms , i.e., the extent of transformation is proportional to undercooling below the Ms and does not depend on the time spent belowthe Ms . The transformation from austenite to martensite proceeds as the temperature is reducedbelow Ms until the martensite finish temperature (Mf ) is reached, at which point 100% martensite is expected. However, if the Mf temperature is below room temperature, then someaustenite may be retained if only cooled to room temperature. The martensite transformation isalso sensitive to external and internal stresses. The martensite plates are related to austenitethrough a KS orientation relationship and show very preferred habit planes. The habit plane ofmartensite in low-carbon steel is parallel to {1 1 1}austenite and in high-carbon steel it is parallelto {2 2 5}austenite . Because the martensite transformation occurs through a shear mechanismand without diffusion, the morphology of martensite is mostly lath-, lenticular-, or plate-like.Experimental evidence has shown that the martensite formation (bct crystal structure)from austenite (fcc crystal structure) is accompanied by IPS. However, the bct structurecannot be obtained crystallographically by just one IPS. This anomaly, schematically illustrated in Figure 2.9, was elucidated by Bowles and Mackenzie [5] and Wechsler et al. [6]. If anaustenite crystal structure is represented by a shape bounded by ABCD in (I), on applying ashear deformation (IPS deformation), the observed martensite shape is attained as shown inII. However, this leads to the wrong crystal structure. Nevertheless, another homogeneousshear (III) can be applied that leads to correct the crystal structure (bct). However, because the 2006 by Taylor & Francis Group, LLC.IIIDIIIAInvariantplanestrainAusteniteAObservedshape(wrongstructure)CBHomogeneousshearDDMartensite(wrongshape)CBCBALatticeinvariantdeformationAADDTwin boundaryIVCBBTwinned martensiteCSlipped martensiteFIGURE 2.9 Schematic illustration of the phenomenological theory of martensite formation (bcc or bctstructure) from austenite (fcc structure), showing the intermediate stages before the fcc structure changesto a bct structure through invariant plane strain, homogeneous shear, and lattice-invariant deformation.The model agrees with the experimentally observed orientation relationship as well as with the macroscopic shape of the martensite plate.shape attained in (III) does not match the observed shape change during martensite transformation, the shape attained after (III) needs to be reset to the shape at (II) by slip and twin latticeinvariant deformations. Therefore, in the next step, an inhomogeneous lattice invariant deformation produces a slipped or twinned martensite that matches the observed shape. Insummary, the change in crystal structure from austenite to martensite is achieved throughtwo IPSs and an inhomogeneous lattice invariant strain. Because this is a phenomenologicaltheory, it does not predict the sequences of these deformations. Rather, it defines a method bywhich the austenite crystal can be transformed to a martensite crystal. Another importantfeature of martensite transformation in carbon containing steels is that the tetragonal distortionof bct structure (a b 6 c) increases with carbon content, given by the following relation [14]:c 1 0:045 (wt% C):aAn increase in tetragonality also leads to the hardening of martensite. Martensite thatforms in low-carbon steel (<0.5 wt% C) shows mainly dislocations, whereas martensite thatforms from high-carbon steels shows twins. A typical example of dislocated martensite formedfrom 0.05 wt% C steel is shown in Figure 2.10. The martensite was attained even with thepresence of a low concentration of carbon by rapid quench conditions attained throughresistance spot welding. Transmission electron microscopy shows lath martensite with a highdislocation density. 2006 by Taylor & Francis Group, LLC.a50 mb2 mFIGURE 2.10 (a) Optical microstructure of lath martensite observed in a low-carbon steel obtained byrapid cooling in a weld. (b) Transmission electron micrograph of the same microstructure exhibitingparallel and long lath martensite plates with high dislocation density. 2006 by Taylor & Francis Group, LLC.The kinetics of martensite is often discussed in terms of nucleation and growth that arerelated to the thermodynamic free energy of austenite and martensite, defect density, andglissile interfaces between austenite and martensite. The Ms temperature is closely linked tothermodynamic parameter T0 (i.e., the temperature at which the Gibbs free energy ofaustenite equals that of the martensite with the same composition). However, a certainamount of undercooling is needed before martensite forms from austenite. This undercooling0is defined by the driving force of austenite to martensite DGg!a and by the amount ofstrain energy (1250 J=mol) that needs to be accounted for because of the growth of martensite. It has been shown that the strain energy needed for martensite formation is independentof the carbon concentration. It is now possible to use thermodynamic models to calculate theMs temperature* by using similar principles for any multicomponent steel [27]. The next stepis to describe the nucleation rate of martensite in austenite. Application of homogeneousnucleation theory principles clearly shows that martensite nucleation cannot be describedbased on pure atomic vibrations due to thermal energy. As a result, heterogeneous nucleationtheories have been developed based on the presence of preexisting embryos of martensite [7].These embryos are dislocation loops in certain crystallographic orientations and are hypothesized to grow by nucleation of new loops. Recently, Olson showed that the first step in thecreation of these dislocation loops is the formation of stacking faults in the closely packedplanes of austenite [7]. Once the nuclei of martensite is established, the next step is the growthof these crystals. Because the growth of the martensite is modeled as the glide of paralleldislocations, the rate of growth is rapid. Research has shown three types of martensite growthexist: athermal, burst, and isothermal transformations [7,14]. In athermal growth, martensiteformation is only a function of undercooling below the Ms temperature and not the time spentat each temperature. This transformation starts exactly at Ms temperature. This mode oftransformation also exhibits a strong dependence on thermal cooling conditions. If thequenching of austenite to a low temperature was interrupted and aged at a temperature inbetween the Ms and Mf temperatures, then further martensite formation would not occuruntil further undercooling is instituted. This austenite stabilization is related to the carbonsegregation to nucleation sites for martensite. Burst transformation involves the formation ofa certain fraction of martensite (ranging from a few percent to 50%) from austenite below Mstemperature, instead of a gradual increase as in the case of athermal growth. The time intervalfor a burst of martensite transformation is about 1 ms. Burst transformation is attributed toan extreme form of autocatalysis (i.e., the formation of one martensite plate triggers theformation of another). Isothermal transformation involves an increase in the martensitefraction with holding time at a given temperature. This mode of transformation is also relatedto autocatalytic events. The isothermal cooling transformation diagrams (or time temperaturetransformation [TTT] diagrams) for isothermal martensite formation show a typical C-curvebehavior, indicating that both thermal and athermal characters are present. Athermal orburst martensite formation is observed in steels; isothermal martensite formation is observedonly in iron alloys that do not contain carbon.2.4MICROSTRUCTURE EVOLUTION DURING REHEATINGThe previous section focused on microstructure evolution due to austenite decompositionthat occurs during cooling from the austenitization temperature. In most heat treatments,microstructure evolution during reheating is also important. For example, the tempering ofmartensite to impart toughness is achieved by reheating the martensite to a high temperature.* Software to calculate Ms temperature can be downloaded freely from 2006 by Taylor & Francis Group, LLC.There is a need to understand the various phase transformations that occur during temperingto achieve an optimum combination of strength and toughness. Austenite formation itself isalso important. For example, in the case of induction heat treatment, the depth of hardeningis related to the depth of austenite formation. In this section, various phase transformationsthat occur during tempering and austenite formation are highlighted.2.4.1 TEMPERED MARTENSITEIn principle, various physical processes that lead to tempering of martensite start in as soon asmartensite is heated to about room temperature [7]. These physical processes are associatedwith different temperature ranges. The transformation mechanisms in these stages are predominantly reconstructive modes with short- or long-range transport of atoms. However, incertain cases, decomposition may be possible by just the movement of carbon atoms with onlya small distortion of the martensite lattice through displacive transformation [28,29]. Carbon Segregation and Aging of MartensiteDuring aging of martensite in some alloys, a coherent spinodal decomposition (i.e., modulated martensite structure with high- and low-carbon regions) may occur up to 1008C [30].This spinodal decomposition may also lead to formation of transition carbides such as Fe4 Cand Fe16 C2 . The carbon atoms may segregate to dislocations or may diffuse to interlathretained austenite. Ohmori and Tamura have postulated that carbon segregation to defectssuch as dislocations may be the overwhelming factor for the observation of carbon-rich andcarbon-depleted regions in the martensite [31]. The evidence for carbon clustering or spinodaldecomposition has been obtained indirectly through electron diffraction or atom probe fieldion microscopy [28,32]. First Stage of TemperingIn the first stage of tempering (100 to 2008C), e-carbide forms from martensite. The composition of this carbide is close to Fe2:4 C. In the case of alloyed steels, the iron atoms may bereplaced by other elements. The e-carbide has a close-packed hexagonal structure and occurs asnarrow laths or rods on cube planes of the martensite with Jacks orientation relationship [7]:112(1 0 1)a0 == 1 0 " 1 e ; (0 1 1)a0 ==(0 0 0 1)e ; [1 1 "]a0 ==[1 " 1 0]e :The nucleation of this transition carbide is related to the modulated structure that formedduring the low-temperature aging step or even the carbon clustering along the dislocations.There is some evidence that the growth of this carbide may show some displacive characteristics instead of a reconstructive transformation mechanism [33]. After the precipitation of ecarbide in stage I, the martensite is still supersaturated with carbon to certain extent andwould undergo further decomposition on heating to higher temperatures. Second Stage of TemperingIn the temperature range of 200 to 3508C, the retained austenite in the steel decomposes intoferrite and cementite. This decomposition was detected successfully by x-ray diffraction anddilatometric and specific volume measurements [7]. The kinetics of this decomposition arerelated to carbon diffusion in austenite. The untransformed austenite may undergo transformation on application of strain and thus affect the toughness of the steel. However, thefraction of retailed austenite is usually low in steels containing less than 0.2 wt% C. The 2006 by Taylor & Francis Group, LLC.formation of stable carbides that is typical to third stage of tempering may overlap in thesecond stage of tempering. Stage of TemperingIn the temperature range of 250 to 7508C, cementite precipitates within the martensite. Thecomposition of the cementite is Fe3 C. In alloyed steels, it is referred as M3 C, where Mcorresponds to substitutional alloying additions (e.g., Cr, Mn) in addition to Fe. Cementitehas an orthorhombic crystal structure and usually occurs as Widmanstatten plates. Anexample of tempered martensite in 300-M steel is shown in Figure 2.11a. The orientationrelationship between ferrite and cementite is of the Bagaryatski type:11 0 0cementite ==[0 1 "]ferrite ;[0 1 0]cementite ==[" 1 1]ferrite ; (0 0 1)cementite ==(2 1 1)ferrite :1The habit planes of cementite can be parallel to either {0 1 1}a or {1 1 2}a of ferrite. Thenucleation of cementite may occur at e-carbide and may grow by dissolution of the e-carbide. Inhigh-carbon steels, the cementite precipitates along the twin boundaries of martensite.Other sites for nucleation of cementite are the prior austenite grain boundaries or interlathboundaries. With the formation of cementite, most of the carbon in martensite is removedfrom solid solution. As a result, the tetragonality of bct structure is lost. Early stages ofcementite growth occur only by carbon diffusion with no significant partitioning of substitutional alloying elements [32,33]. However, with extended tempering, redistribution of alloyingelements also occurs between ferrite and cementite [32]. In addition, the plate-like cementiteparticles may coarsen and spheroidize with extended tempering. At this stage, the recovery andrecrystallization of martensite laths may also be initiated. In high-carbon martensite, higherorder carbides such as M5 C2 (x-carbide) can also form. It has been shown that higher ordercarbides are actually polytypes of the basic trigonal prism basis of cementite structure [34]. Stage of TemperingTempering at higher temperatures (>7008C) leads to the precipitation of more equilibriumalloy carbides such as M7 C3 and M23 C6 . In steels containing Cr, Mo, V, and Ti, thesecarbides are associated with hardening of the steel that is called secondary hardening. Theprecipitation of these carbides also leads to the dissolution of cementite. An example of alloycarbide formation by tempering at 6008C in 300-M steel is shown in Figure 2.11b. At thisa0.5 mb0.5 mFIGURE 2.11 Transmission electron micrographs of quenched and tempered 300-M steel samples(a) after tempering at 3008C for 2 min, showing cementite plates in a martensite lath and (b) aftertempering at 6008C for 1 min, showing alloy carbides. 2006 by Taylor & Francis Group, LLC.stage, the recrystallization of martensite lath is more complete, and there is a tendency for theformation of equiaxed grains and extensive grain growth.2.4.2 AUSTENITE FORMATIONThe early work on austenite formation by Robert and Mehl [35] focused on the nucleation ofaustenite from a ferritepearlite mixture. Various researchers have since shown the complexity of austenite formation from a two-phase mixture of ferrite and cementite and have madeattempts at modeling austenite formation as a function of composition and microstructure[3641]. On heating steel, with a spheroidized ferrite cementite mixture, the austenite phasenucleates at the ferritecementite boundary. With further heating, the austenite phase consumes the cementite and then grows into ferrite through diffusion-controlled growth. In apearlitic microstructure, the austenite may nucleate in the cementite and grow into the colonyby dissolving both ferrite and cementite. A typical micrograph of austenite growth into apearlite colony is shown in Figure 2.12a. A schematic illustration of the growth is shown inFigure 2.12b. Recent work has shown that it is possible to model the above phenomenon withcomputational tools [42]. It is possible to construct TTT diagrams for the austenite growth forany steel to evaluate the effect of the initial microstructure. These diagrams do not show aC-curve behavior because both the driving force for austenite formation and the diffusivityincrease with temperature. This results in monotonic acceleration of austenite formation asthe temperature increases. The rate of austenite formation also increases with the presence ofresidual austenite in the microstructure, as demonstrated by Yang and Bhadeshia, whostudied the austenite growth kinetics in a bainitic microstructure [43,44]. The microstructurecontained residual austenite films, and there was no requirement for nucleation of austeniteand the austenite films grew with an increase in the temperature. After the completion ofaustenite formation, continued heating leads to grain growth of austenite. The grain growth isalso affected by the presence of fine carbonitride precipitates. With the presence of theseprecipitates the grain boundaries are pinned and therefore, grain growth characteristicsare sluggish. However, on heating above the dissolution temperature of these precipitates,the austenite grain may coarsen at an accelerated rate [45,46]. Most of the austenite formationfrom ferrite occurs by a diffusion-controlled reconstructive transformation mechanism.However, at rapid rates, the transformation of ferrite to austenite may occur by interfacecontrolled or by displacive transformation [47].2.5 SUMMARY OF STEEL MICROSTRUCTURE EVOLUTIONAn overview of all microstructure evolution through reconstructive or displacive transformation mechanisms during heating and cooling of steel can be classified as shown in Figure 2.13[4]. Reconstructive transformation involves substitutional diffusion. The reconstruction of aparent lattice into a product lattice occurs through a noncoordinated motion of atoms acrossthe interface. The growth mostly occurs by nucleation and growth of product phases.Reconstructive transformations are slow below 6008C. The formation of allotriomorphicferrite, idiomorphic ferrite, massive ferrite, pearlite, carbide, and austenite all belong to thecategory of reconstructive transformation. Displacive transformation involves coordinatedatom, causing a change from a parent crystal to a product crystal. This change is achieved byIPS deformation with a large shear component, leading to a plate or lath shape. During thistransformation, the substitutional atoms do not diffuse. However, displacive transformationsoccurring at high temperatures (above Ms ) may involve varying amounts of interstitial carbondiffusion during nucleation and growth. In the case of Widmanstatten ferrite formation, bothnucleation and growth involve carbon diffusion. In contrast, in bainitic ferrite formation, 2006 by Taylor & Francis Group, LLC.aAllotriomorphicferritePearlite colonyAustenite/pearliteboundaryAllotriomorphicferrite10003 5 mPearlitecolonybAusteniteFIGURE 2.12 (a) Optical micrograph of a steel sample with ferrite pearlite microstructure heated toan intercritical temperature for a short time, showing the formation of austenite in the pearlite colony.(b) Schematic illustration of the austenite growth mechanism that dissolves the pearlite colonies andeventually the allotriomorphic ferrite to attain an equilibrium austenite fraction.carbon diffusion occurs only during nucleation. In the case of martensite, the carbon diffusion does not occur during nucleation and growth. Recently, attempts have been made to usetheoretical formulations to describe this varying carbon supersaturation during displacivetransformation within a coupled diffusional-displacive transformation framework [4850].2.6PREDICTION OF MICROSTRUCTURE EVOLUTION DURINGHEAT TREATMENTThe final properties of heat-treated parts depend on the microstructure that evolves duringthe heat-treating process. The microstructure evolution is in turn controlled by the complex 2006 by Taylor & Francis Group, LLC.ReconstructiveDisplaciveDiffusion of all atomsduring nucleation andgrowthSluggish below about850 KInvariant-plane strain shapedeformation with large shearcomponentNo iron or substitutional solutediffusion and thin plate shapeAllotriomorphicferriteIdiomorphicferriteMassiveferriteNo change in bulkcompositionPearliteCooperative growth offerrite and cementiteWidmanstttenferriteCarbon diffusion duringparaequilibrium nucleationand growthBainite and acicularferriteCarbon diffusion duringparaequilibrium nucleationNo diffusion during growthMartensiteDiffusionless nucleation andgrowthAustenite formationFull austenitization and intercritical annealingTempering reactionCementite and alloy precipitation, laves phaseprecipitation and coarseningFIGURE 2.13 Classification of steel microstructure evolution during heating and cooling based on themechanism of transformation. (Adapted from H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., theInstitute of Materials, London, 2001.)thermal cycles and the composition of the alloy. Over a period of many decades,heat-treatment processes have been developed through extensive experimentation and characterization of particular alloy compositions and final properties. This methodology, inconjunction with an extensive experimental database, often achieves the required properties.However, this approach has the following limitations: (1) it is rigid and cannot be changedeasily to meet the ever-increasing demand for optimization of cost and quality and to meet therequirements of strict codes and standards and (2) it cannot be extended to newly developedsteels. In this regard, tools that can predict the evolution of microstructure as a function ofcomposition and heat-treatment history will be useful. These predictive tools should becapable of addressing the effect of alloy composition on the stability of various phases. Inaddition, the effect of timetemperature histories on the growth and dissolution of thesephases must be addressed. For example, given the phase diagram information for a steel,these tools must be capable of predicting the continuous heating and cooling transformationdiagrams as a function of steel composition (see Figure 2.14). The figure on the left-hand sideshows an iron-rich corner of the FeC phase diagram showing the phase stability at differenttemperature. The schematic figure in the middle shows the initiation of austenite formationfrom ferrite as a function of heating curves. The schematic figure in the right shows theinitiation of ferrite formation from austenite as a function of different cooling rate. With the 2006 by Taylor & Francis Group, LLC.Temperature ( C)1000AusteniteCHT( + Fe3C to )CoolingFerrite +austenite800CCT( to )600Heating400Ferrite + cementite2000. 110Weight percentage carbon2424100101Time (s)102 101100101Time (s)102FIGURE 2.14 Schematic illustration, showing the importance of continuous heating transformation(CHT) and continuous cooling transformation (CCT) diagrams with reference to an FeC metastablebinary diagram.framework of microstructural evolution presented in the earlier sections and state-of-the-artcomputational and kinetic models presented in the following sections, it is possible to predictmicrostructural evolution during heat treatment of steels as a function of composition andthermal history.2.6.1 CALCULATIONOFMULTICOMPONENT MULTIPHASE DIAGRAMSThe stability of a phase is governed by its free energy, which is a function of temperature andits constitution. A generic description of free energy of a solid-solution phase, f, Gf , is givenby the following equation [51]:Gf Gf Gf mix Gf mix ,oidealexcess(2:1)where Gf is the free energy contribution from pure components in that phase, Gf mix is theoidealcontribution from ideal mixing, and Gf mix is the contribution due to nonideal interexcessactions between the components.Many thermodynamic models describe various phases (see Ref. [15]). With the descriptionof this Gf for all phases that can form in a given alloy, it is possible to estimate equilibriumfractions of each phase and their constitution at a given temperature. This estimation isperformed by minimization of the free-energy curves of the various phases. This procedurealso allows for determination of the tie line, which is schematically shown in Figure 2.15. Thetie line is represented by a common tangent and is mathematically represented by thefollowing equation:ma mb ; ma mb ,ABBA(2:2)where mf is the chemical potential of element i in the f-phase. The chemical potentialsiare obtained from the free-energy expressions given in Equation 2.1 with the followingequation: 2006 by Taylor & Francis Group, LLC.-phaseFree energy (J/mol)-phasexBmAmBmBxBTie linemAABConcentration of BFIGURE 2.15 Schematic illustration of the methodology used to calculate phase stability in a binaryalloy: free energy of an a-phase and a b-phase and the common-tangent construction for the descriptionof the tie line and chemical potential of elements A and Gf xfBA G f; mf Gf 1 xf:BBx fxfBBGf(2:3)Equation 2.2 and Equation 2.3 can be extended to multicomponent systems by invoking theequality of the chemical potentials of all the components. The following equation determinesthe governing equilibrium between a- and g- phases in the FeCrNiC alloy system:mg ma ; mg ma ; mg ma ; mg ma :CFeCrNiFeCCrNi(2:4)Using the relations in Equation 2.4 and mathematical methods, a phase equilibria betweenthe g- and a-phases as a function of Cr, Ni, and C concentrations can be calculated. Thismethodology has proven to be useful for describing the phase stability in multicomponentalloys such as aluminum alloys, stainless steels, low-alloy steels, and Nibase alloys [5257]. Forexample, a simple FeC binary diagram is compared with an (Fe,Cr,Mo)C quasibinary phasediagram in Figure 2.16. The diagram shows that a simple ferrite austenite cementite phaseequilibrium is modified to ferrite austenite cementite M23 C6 M6 C M3 C2 MC dueto the addition of Cr and Mo to the FeC systems. ThermoCalc software was used to constructthese diagrams with solid solution database [58].2.6.2 CALCULATIONOFDIFFUSION-CONTROLLED GROWTHAlthough the phase stability calculation allows the estimation of the equilibrium microstructure at a particular temperature, microstructure control in most heat-treatment processesrelies on the kinetics of product-phase formation from the parent phase. For example, in dualphase low-alloy steels, it is important to understand the kinetics of austenite formation tocontrol the mixture of ferrite and martensite. In addition to the kinetics of transformation,there is a need to understand the equilibration of the nonequilibrium microstructure that isformed during processing. Both phenomena can be described by diffusion-controlled growthmodels by coupling thermodynamic models with diffusion-controlled growth calculations.The methodology is presented below. 2006 by Taylor & Francis Group, LLC.Ferrite +austeniteM23C61000Temperature ( C)AusteniteAusteniteFerrite +austenite800Ferrite +M23C6600Ferrite +M23C6 +M6C400Ferrite +M23C6 +M6C + LavesFerrite + cementiteFerrite +M23C6 +M3C2 + MC2000. percentage carbon0. percentage carbonFIGURE 2.16 Comparison of a calculated FeC binary-phase diagram with an FeCrMoC quasibinary diagram, showing the complex stability between ferrite, austenite, and other carbides.The formation of a product phase with a different composition from the parent phaseinvolves diffusion of partitioning elements. If local equilibrium exists between the parent andproduct phases at the interface, the interfacial concentrations can be given by the tie linesdrawn in the phase diagram. Given this condition, it is possible to describe the movement ofthis interface as a function of temperature and time by solving Ficks second law and bymaintaining a mass balance at the interface. The governing equation for a-phase formation ing-phase in one-dimensional geometry is given in the following example [59,60]:m,aCIm,g CImm(dl =dt) Dm dCg =dx Dm dCa =dx :ga(2:5)m ,aIn the above equation, (dl =dt) is the rate of interface movement or velocity. The terms CIm,gand CI are the interface concentrations of element m in the a- and g-phases. The terms Dmamand Dm are the diffusivities of element m in a- and g-phases. The terms dCg =dx and mg dCa =dx are the concentration gradients* of element m in a- and g-phases at the interface.The conditions at the interface are schematically shown in Figure 2.17. In a multicomponentalloy, Equation 2.5 must be satisfied for all elements, and a unique velocity of the interface isobtained. This restriction leads to the selection of the interface compositions as determined bythe tie lines that may shift with time and may not pass through the nominal alloy composition. Therefore, these calculations must adjust local equilibrium conditions while solving thediffusion equations and require close coupling with thermodynamic models. This is themethodology implemented in the DicTra software [61] and other published works.An example calculation is presented below. The final microstructure of a stainless steelweld often contains ferrite and austenite at room temperature. This microstructure is far fromequilibrium. However, given a high-temperature heat treatment, equilibration of this microstructure will occur. During this heat treatment, the ferrite may grow or dissolve, dependingupon the alloy composition, the initial state, and the heat treatment temperature. Examplesare shown in Figure 2.18. In case A, the initial austenite composition was Fe15%Cr20%Ni* Ideally, the chemical potential gradient needs to be used instead of concentration gradient. 2006 by Taylor & Francis Group, LLC.ConcentrationCIm,CIm,dldtDistanceFIGURE 2.17 Schematic illustration of interface concentrations and diffusion in both a- and b- phasesto describe diffusion-controlled transformations.(wt%), and the initial ferrite composition was Fe20%Cr10%Ni (wt%). The initial volumepercentage of ferrite was 9%. The calculations predicted rapid dissolution of ferrite if this weldwas aged at 13008C. In case B, the initial austenite composition was Fe21%Cr11%Ni (wt%),and the initial ferrite composition was Fe30%Cr4.5%Ni (wt%). The initial volume percentageof ferrite was again set to 9%. The calculations predicted that the volume percentage of ferritewould increase from 9 to 25% if the weld was aged at 13008C. In addition, the calculationsshowed different stages where the kinetics of this transformation were controlled by eithersolute diffusion in ferrite or solute diffusion in austenite. This condition is due to the largedifferences in the diffusivity of Cr and Ni in the ferrite and austenite phases. The abovemethodology has been applied to continuous cooling and heating conditions. Such calculationsallow for the design of postweld heat treatment of stainless steel welds.However, the local equilibrium assumption used in the above model leads to someparadoxes for simulating growth below ~6008C. At temperatures below 6008C, the diffusionprofile of substitutional atoms (see Figure 2.17) may be extremely steep, with a width that issmaller than interatomic distances. Hultgren and other authors have addressed this issue [6265]. It is hypothesized that under these conditions, the concentration ratio of substitutionalatoms to iron atoms will remain configurationally frozen and that interface growth will becontrolled purely by carbon (interstitial) diffusion. The carbon concentration at the interfacewill now be determined by a phase-boundary calculation similar to the FeC phase diagram,with a constraint that the C activity will be modified by the configurationally frozen substitutional atoms. This mode of transformation is called PE transformation. PE appears tocorrelate well with the accelerated ferrite growth observed under large undercooling in mostheat treatment and welding conditions. In steels, PE growth rates are always higher than259% Ferrite91% AusteniteFerrite %2015105Case ACase B00.010.1110Time (s)1001000FIGURE 2.18 Variation of ferrite percentage while aging at 1573 K for two initial conditions: (A)corresponds to Fe15Cr20Ni (wt%) austenite and Fe20Cr10Ni (wt%) ferrite; (B) corresponds to Fe21Cr11Ni (wt%) austenite. (The ferrite composition was Fe30Cr4.5Ni [wt%] ferrite.) 2006 by Taylor & Francis Group, LLC.105Thickening rate (m s0.5)42106421074ParaequilibriumLocal equilibrium2108600650700750Temperature (8C)800FIGURE 2.19 Calculated parabolic thickening rate of ferrite growth into austenite in an Fe0.1C2.25Cr1Mo steel under (.) local equilibrium and (*) paraequilibrium.growth rates calculated by local equilibrium models. This discrepancy is demonstrated by acomparison of the parabolic thickening rates of ferrite into austenite in an FeCrMoC steelat different temperatures (see Figure 2.19). It is now possible to use models to predict TTTand continuous cooling transformation (CCT) diagrams with commercial software that areavailable on the Internet and from the commercial organizations [6668]. Typical TTT andCCT diagrams calculated with the JMatPro software are shown in Figure 2.20a andFigure 2.20b, respectively. A TTT and CCT diagram for 0.1% transformation to ferritecalculated by using an Internet tool is shown in Figure 2.20c. The methods of calculationfor each tool are different and are constantly refined. As a result, these results must be used asguidance rather than as accurate predictions. However, with the advent of new models [69]and detailed thermodynamic and diffusion data, an accurate description of transformation insteels should be possible. The final goal of these predictive approaches is to couple thesemicrostructure models with thermal and structural models to describe the overall response ofsteel structures to heat treatment as envisioned by Kirkaldy [70] (see Figure 2.21). In thisintegrated model, it would be possible to describe transient thermal fluctuations in a steel partduring heat treatment in a furnace and during quenching or cooling. In addition, during suchthermal responses, it should be possible to describe the thermal stress evolution based on theconstitutive relations of different phases. By coupling all the above parameters with amicrostructure model that describes the transformation kinetics as a function of temperature,time, and stress, it is possible to evaluate the properties of steel parts subjected to differentheat treatments as a function of steel composition.2.7 SUMMARYIn this chapter, phase transformations in alloys are categorized in terms of different thermodynamic, microstructural, and mechanistic bases. The predominant transformation mechanisms in steels (i.e., reconstructive and displacive mechanisms) are explained. The importantmicrostructures observed during steel heat treatment, including allotriomorphic ferrite, Widmanstatten ferrite, bainite, pearlite, martensite, and carbide formation during tempering andaustenite formation during reheating, are described based on this framework. Finally, computational thermodynamic and kinetic methodologies that are available to predict the microstructure evolution in steels are highlighted. 2006 by Taylor & Francis Group, LLC.(a)Composition (wt%):Fe: 97.93Mn: 1.45Si: 0.25C: 0.37Transitions:Pearlite: 710.68 CBainite: 565.04 CFerrite: 769.61 CMartensite: 337.77 CTTT HSLAS800Temperature (8C)700600500Ferrite (0.1%)Pearlite (0.1%)Bainite (0.1%)Pearlite (99.9%)Bainite (99.9%)4003000.1100010Time (s)Grain size: 6.0 ASTMAustenitization: 819.61 CComposition (wt%):Fe: 97.93Mn: 1.45Si: 0.25C: 0.37Transitions:Pearlite: 710.69 CBainite: 565.04 CFerrite: 769.62 CMartensite: 337.78 CCCT HSLAS(b)900Temperature (8C)800700600Ferrite (0.1%)Pearlite (0.1%)Bainite (0.1%)Pearlite (99.9%)Bainite (99.9%)100.0 C/s10.0 C/s1.0 C/s0.1 C/s5004003000.1101000Time (s)Grain size: 6.0 ASTMAustenitization: 819.61 C(c)TTT1100CCTTemperature (K)1000900800Ba700MaORNL600101100101102103104Time (s)FIGURE 2.20 Calculated (a) time temperature transformation (TTT) diagram and (b) continuouscooling transformation (CCT) diagram for a steel using JMatPro1 software for an Fe0.37C0.25Si1.45Mn steel. (c) TTT and CCT transformation diagram is calculated using the tool found on theInternet at http:= 2006 by Taylor & Francis Group, LLC.Chemical compositionAustenitizing conditionAustenite grain sizeThermal boundary conditionsHeat transfer coefficientsThermophysical propertiesTransformationbehaviorTransienttemperaturefieldCreep deformationionatrm t ))fo,rens f (T ctutra =ruof ion , stat(Tathe rm = fsfo iesan ertpro.pTrnt(Heat generation due todeformation)Thermal stressteComponent geometryFE - meshLaatM[transformationMechanicalresponsedispl. s, e = f (x,y,z,t )TransformationkineticsPhase diagramTTT diagramT = f (x,y,z,t )Microstructural= f (s )]s curvedevelopment= f (T, structure)Transformation strain phases = f (x,y,z,T,t )Transformation plasticityMechanical boundary cond.loading, material propertiesHardness distributionFIGURE 2.21 Modeling phase transformations in steels, in response to thermomechanical conditions.(Adapted from J.S. Kirklady, Scand. J. Metall., 20, 1991.)2.8ACKNOWLEDGMENTSPart of the results presented in this paper is based on the research sponsored by the Divisionof Materials Sciences and Engineering and Assistant Secretary for Energy Efficiency andRenewable Energy, Office of Industrial Technologies, Advanced Industrial Materials Program, U.S. Department of Energy, under contract DE-AC05-00OR22725 with UT-Battelle,LLC. The author also wishes to thank Prof. H.K.D.H. Bhadeshia, Dr. J.M. Vitek, and Dr.S.A. David for help and guidance in the research related to different topics discussed in thischapter. Finally, the author would like to thank Drs. M.K. Miller and Dr. J.M. Vitek ofORNL for helpful suggestions and review of the manuscript.REFERENCES1. J. Agren, Scr. Mater., 46, 2002, 8938.2. J.W. Christian, The Theory of Transformations in Metals and Alloys, 2nd ed., Part 1, PergamonPress, Oxford, 1981, p. 9.3. H.K.D.H. Bhadeshia, Worked Examples in the Geometry of Crystals, The Institute of Materials,London, 1987.4. H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., The Institute of Materials, London, 2001.5. J.S. Bowles and J.K. Mackenzie, Acta Metall., 2, 1954, 129.6. M.S. Wechsler, D.S. Lieberman, and T.A. Read, Trans. AIMME, 197, 1953, 1503.7. MartensiteA Tribute to Morris Cohen, G.B. Olson and W.S. Owen, Eds., ASM International,Materials Park, OH, 1992.8. C. Zener, Trans. AIMME, 167, 1946, 55095.9. A. Hultgren, Trans. ASM, 39, 1947, 9151005.10. H.I. Aaronson, Decomposition of Austenite by Diffusional Processes, Interscience, New York, 1962. 2006 by Taylor & Francis Group, LLC.11. C.A. Dube, Ph.D. Thesis, Carnegie Institute of Technology, Pittsburgh, PA, 1948.12. C.A. Dube, H.I. Aaronson, and R.F. Mehl, Rev. de Met., 55, 1958, 201.13. W.T. Reynolds, Jr., M. Enomoto, and H.I. Aaronson, The Proeutectoid Reaction, in The Proceedings of an International Conference on Phase Transformations in Ferrous Alloys, A.R. Marder andJ.I. Goldstein, Eds., 1988.14. R.W.K. Honeycombe and H.K.D.H. Bhadeshia, Steels, Microstructure, and Properties, 2nd ed.,Edward Arnold, London, 1995.15. S.S. Babu and H.K.D.H. Bhadeshia, Mater. Sci. Eng., A141, 1990, 209.16. G.R. Purdy, Acta Metall., 26, 1978, 47786.17. H.I. Aaronson, Metall. Trans. A, 24A, 1993, 24176.18. G. Spanos, W.T. Reynolds, Jr., and R.A. Vandermeer, Metall. Trans. A, 22A, 1991, 136780.19. H.K.D.H. Bhadeshia, Acta Metall., 29, 1981, 1117.20. H.K.D.H. Bhadeshia, L.-E. Svensson, and B. Gretoft, Acta Metall., 33, 1985, 127183.21. R. Trivedi, Metall. Trans., 1, 1970, 921.22. R. Trivedi, Acta Metall., 18, 1970, 287.23. H.I. Aaronson, W.T. Reynolds, Jr., G.J. Shiflet, and G. Spanos, Metall. Trans. A, 21A, 1990,13480.24. H.K.D.H. Bhadeshia and J.W. Christain, Metall. Trans. A, 21A, 1990, 76797.25. S.S. Babu and H.K.D.H. Bhadeshia, Mater. Sci. Eng., A156, 1992, 19.26. M. Hillert, Scr. Mater., 47, 2002, 1378.27. H.K.D.H. Bhadeshia, Met. Sci., 16, 1982, 15965.28. K.A. Taylor, L. Chang, G.B. Olson, G.D.W. Smith, M. Cohen, and J.B. Vander Sande, Metall.Trans. A, 20A, 1989, 274965.29. B.P.J. Sandvik, Metall. Trans. A, 13A, 1982, 789800.30. K.A. Taylor, L. Chang, G.B. Olson, G.D.W. Smith, M. Cohen, and J.B. Vander Sande, Metall.Trans. A, 20A, 1989, 271737.31. Y. Ohmori and I. Tamura, Metall. Trans. A, 23A, 1992, 214758.32. S.S. Babu, K. Hono, and T. Sakurai, Metall. Trans. A, 25A, 1994, 499508.33. R.C. Thomson and M.K. Miller, Acta Mater., 46, 1998, 220313.34. S. Nakagura, T. Suzuki, and M. Kusunoki, Trans. Jpn. Inst. Met., 22, 1981, 699.35. G.A. Robert and R.F. Mehl, ASM Trans., 31, 1943, 61350.36. R.R. Judd and H.W. Paxton, Trans. AIME, 242, 1968, 20615.37. G. Molinder, Acta Metall., 4, 1956, 56571.38. M. Hillert, K. Nilsson, and L.E. Trondahl, JISI, 209, 1971, 4966.39. T. Akbay, R.C. Reed, and C. Atkinson, Acta Metall. Mater., 42, 1994, 146980.40. C. Garcia De Andres, F.G. Caballero, C. Capdevilla, and H.K.D.H. Bhadeshia, Scr. Mater., 39,1998, 7916.41. F.G. Caballero, C. Capdevilla, and C. Garcia De Andres, Mater. Sci. Tech., 17, 2000, 11148.42. A. Jacot and M. Rappaz, Acta Mater., 47, 1999, 164551.43. J.R. Yang, C.Y. Huang, Mater. Chem. Phys., 35, 1993, 16875.44. J.R. Yang and H.K.D.H. Bhadeshia, Mater. Sci. Eng. A, 131, 1991, 99113.45. M. Chapa, S.F. Medina, V. Lopez, and B. Fernandez, ISIJ Int., 42, 2002, 128896.46. T. Gladman, Heat Treat. Met., 21, 1994, 114.47. W.J. Kaluba, R. Taillard, and J. Foct, Acta Mater., 46, 1998, 591727.48. G.B. Olson, H.K.D.H. Bhadeshia, and M. Cohen, Acta Metall., 37, 1989, 3819.49. G.B. Olson, H.K.D.H. Bhadeshia, and M. Cohen, Metall. Trans. A, 21A, 1990, 8059.50. S.A. Mujahid and H.K.D.H. Bhadeshia, J. Phys. D, 34, 2001, 257380.51. N. Saunders and A.P. Miodownik, CALPHAD-Calculation of Phase DiagramsA ComprehensiveGuide, Pergamon, Elsevier Science, Amsterdam, 1998.52. S.S. Babu, S.A. David, J.M. Vitek, and M.K. Miller, Metall. Mater. Trans. A, 27A, 1996, 76374.53. S.S. Babu, S.A. David, and M.K. Miller, Appl. Surf. Sci., 94=95, 1996, 28087.54. S.A. David, J.M. Vitek, S.S. Babu, L. Boatner, and R.W. Reed, Sci. Tech. Weld. Join., 2, 1996, 7988.55. P. Hofer, M.K. Miller, S.S. Babu, S.A. David, and H. Cerjak, Metall. Mater. Trans. A, 31A, 2000,975984. 2006 by Taylor & Francis Group, LLC. Miller, S.S. Babu, and M.G. Burke, Mater. Sci. Eng., A270, 1999, 148.S.S. Babu, S.A. David, J.M. Vitek, and R.W. Reed, Sci. Tech. Weld. Join., 6, 2001, 3140.B. Sundman, B. Jansson, and J.O. Andersson, CALPHAD, 9, 1985, 15390.J.M. Vitek, S.A. Vitek, and S.A. David, Metall. Mater. Trans. A, 26A, 1995, 200725.H.K.D.H. Bhadeshia, Prog. Mater. Sci., 29, 1985, 32186.J. Agren, ISIJ Int., 32, 1992, 2926.A. Hultgren, Jerkontorets Ann., 135, 1951, 403.J.B. Gilmour, G.R. Purdy, and J.S. Kirkaldy, Metall. Trans., 3, 1972, 321322.G. Ghosh and G.B. Olson, Acta Mater., 50, 2002, 200919.E. Kozeschnik, J. Phase Equilibria, 21, 2000,, Modeling Microstructure Development in Welds, Internet online computational tool, Oak Ridge National Laboratory, Oak Ridge, Tenn.68. JMatPro, J. Odqvist and M. Hillert, Acta Mater., 50, 2002, 321125.70. J.S. Kirkaldy, Scand. J. Metall., 20, 1991, 5061. 2006 by Taylor & Francis Group, LLC.3Fundamental Concepts in SteelHeat TreatmentAlexey V. Sverdlin and Arnold R. NessCONTENTS3. ............................................................................................................... 122Crystal Structure and Phases ..................................................................................... 1223.2.1 Crystal Structure of Pure Iron........................................................................ 1223.2.2 IronCarbon Equilibrium Diagram ............................................................... 1233.2.2.1 Metastable FeFe3C Equilibrium Diagram...................................... 1233.2.2.2 Stable FeC Equilibrium Diagram................................................... 1243.2.3 Effect of Carbon............................................................................................. 1253.2.4 Critical (Transformation) Temperatures ........................................................ 127Structural Transformations in Steel ........................................................................... 1283.3.1 AustenitePearlite Transformation................................................................. 1283.3.2 Structure of Pearlite........................................................................................ 1293.3.3 Transformation of Austenite in Hypo- and Hypereutectoid Steels ................ 1303.3.4 Martensite Transformation ............................................................................ 1313.3.5 Morphology of Ferrous Martensites .............................................................. 1333.3.6 Bainite Transformation .................................................................................. 1343.3.7 Morphology of the Bainite Transformation................................................... 1353.3.8 Tempering....................................................................................................... 135Kinetics of Austenite Transformation ....................................................................... 1373.4.1 Isothermal Transformation Diagrams............................................................ 1373.4.2 Continuous-Cooling Transformation Diagrams ............................................ 1383.4.2.1 Transformations That Take Place under Continuous Coolingof Eutectoid Steels ............................................................................ 1393.4.2.2 Transformations of Austenite on Cooling in theMartensite Range ............................................................................. 1413.4.3 Derivation of the Continuous-Cooling Transformation Diagramfrom the Isothermal Transformation Diagram .............................................. 1423.4.4 Continuous-Cooling Transformation Diagram as a Functionof the Bar Diameter........................................................................................ 1433.4.5 Definition of Hardenability ............................................................................ 145Grain Size .................................................................................................................. 1463.5.1 Structure of Grain Boundaries ....................................................................... 1463.5.1.1 Structural Models............................................................................. 1473.5.2 Determination of Grain Size .......................................................................... 1493.5.3 Austenite Grain Size Effect and Grain Size Control ...................................... 1503.5.4 Grain Size Refinement.................................................................................... 152Strengthening Mechanism in Steel ............................................................................. 153 2006 by Taylor & Francis Group, LLC. Solution Strengthening .......................................................................... 153Grain Size Refinement.................................................................................... 154Dispersion Strengthening ............................................................................... 154Work Hardening (Dislocation Strengthening)................................................ 158Thermal Treatment of Steels .......................................................................... 1593.6.5.1 Annealing ......................................................................................... 1593.6.5.2 Quenching (Strengthening Treatment) ............................................. 1623.6.5.3 Tempering......................................................................................... 163References .......................................................................................................................... 163Further Reading ................................................................................................................. 1633.1INTRODUCTIONThe purpose of heat treatment is to cause desired changes in the metallurgical structure and thus inthe properties of metal parts. Heat treatment can affect the properties of most metals and alloys,but ferrous alloys, principally steels, undergo the most dramatic increases in properties, andtherefore structural changes in ironcarbon alloys are considered in this chapter. In general, themost stable steel structures are produced when a steel is heated to the high-temperature austeniticstate (to be defined later) and slowly cooled under near-equilibrium conditions. This type oftreatment, often referred to as annealing or normalizing, produces a structure that has a low levelof residual stresses locked within the part, and the structures can be predicted from an equilibriumdiagram. However, the properties that interest heat treaters the most are those exhibiting highstrength and hardness, usually accompanied by high levels of residual stresses. These are metastable structures produced by nonequilibrium cooling or quenching from the austenitic state.Most of this chapter discusses equilibrium and nonequilibrium structures, their properties,and the tools that we have at our disposal to predict different types of phase formations and theirproperties. It is essential that heat treaters have a clear understanding of the structures that canbe produced in steel under different treatment conditions that they can apply in their equipment. STRUCTURE AND PHASESCRYSTAL STRUCTUREOFPURE IRONIron in the solid state is known in two allotropic states. (Allotropy is the phenomenon ofan element having different crystal lattices depending on the particular temperature andpressure.) Starting from low temperatures and up to 9108C (16708F), iron possesses a bodycentered cubic (bcc) lattice and is called a-iron (a-Fe). At 9108C a-iron crystals turn intog-iron crystals possessing a face-centered cubic (fcc) lattice. The g crystals retain stability upto temperature of 14008C (25008F). Above this temperature they again acquire a bcc latticeand are usually called d crystals. The d crystals differ from a crystals only in the temperatureregion of their existence. Iron has the following lattice constants: 0.286 nm for bcc lattices(a-Fe, d-Fe) and 0.364 nm for fcc lattices (g-Fe).At low temperatures, a-Fe exhibits a strongly ferromagnetic character. When it is heatedto about 7708C (14188F), ferromagnetism vanishes. In accordance with the latest findings,this is because the lattice loses its ferromagnetic spin ordering. The state of iron above 7708Cis called b-Fe. The lattice of paramagnetic b crystals is identical to the lattice of a crystals.The points at which one allotropic form of iron transforms to another are conventionallysymbolized by the letter A with subscripts indicating the ordinal number of the transformation. The subscripts 0 and 1 signify transformations that are absent in pure iron but areobserved in carbon alloys of iron. The subscript 2 denotes a magnetic transformation of thea-phase, while the subscripts 3 and 4 stand for transformation of a to g and g to d. 2006 by Taylor & Francis Group, LLC.Temperature, C160015391400 -FeAc41400120010009008007006001528-FeAr4-Fe-Fe910906Ar3Ac3770Ar2-FeAc2-FeTimeFIGURE 3.1 Heating and cooling curves for pure iron.In going from one form to another, iron is capable of undercooling. This causes adifference in the position of transformation points on heating and cooling. The differencedepends on the cooling rate (Figure 3.1) and is termed hysteresis. The letters c and rindicate whether the transformation is due to heating or cooling.A change in the density of a-Fe as it transforms to g-Fe results in an abrupt change in thevolume of the material. Sometimes this gives rise to stresses that exceed the elastic limit andlead to failure. The density of g-Fe is about 4% higher than that of a-Fe.3.2.2 IRONCARBON EQUILIBRIUM DIAGRAMThe structure of ironcarbon alloys can contain either pure carbon (graphite) or a chemicalcompound (cementite) as the carbon-enriched component. Cementite is present even in relatively slowly cooled alloys: a long holding at elevated temperatures is required to decomposecementite to iron and graphite. For this reason the ironcarbon diagram is usually treated asthe ironiron carbide diagram. The former is stable, whereas the latter is metastable.The ironcarbon diagram is shown in Figure 3.2. Dashed lines stand for the stable FeCdiagram, and solid lines denote the metastable FeFe3C diagram. Metastable FeFe3C Equilibrium DiagramAs shown in Figure 3.2, the lattices of allotropic forms of iron (d, g, and a) serve as sites offormation of d, g, and a solid solutions of carbon in iron (the same symbols are adopted forthe designation of solid solutions).When carbon-depleted alloys crystallize, crystals of the d solid solution precipitate at theliquidus AB and solidus AH. The d solid solution has a bcc lattice. At the maximum carbontemperature of 14908C (27148F), the d solution contains 0.1% C (point H). At 14908C aperitectic reaction takes place between the saturated d solution and the liquid containing 0.5%C (point B). As a result, the g solid solution of carbon in g-Fe is formed. It contains 0.18% C(point I).If the carbon content is higher than 0.5%, the g solid solution crystallizes directly from theliquid (at the liquidus BC and solidus IE). At 11308C (20668F) the limiting solubility ofcarbon in g-Fe is close to 2.0% C (point E). Decreasing the temperature from 11308C(20668F) leads to lowering the carbon solubility in g-iron at the line ES. At 7238C (13338F)the solubility is 0.8% C (point S). The line ES corresponds to precipitation of iron carbidefrom the g solution.As the carbon content is raised, the temperature at which the g lattice transforms to thea lattice lowers, and the transformation takes place over the temperature interval corresponding to the curves GS and GP. 2006 by Taylor & Francis Group, LLC.AL1400 + L BH+1300 NDD+L1200Fe3C + L(1.98)C(1135 )FTemperature, 8CEE2.011001130FC4.31000900800700 + Fe3CG9108+P 768P(0.69)S738KS 0.80723K0.025500210200A0 + Fe3C100012345Carbon, %FIGURE 3.2 Ironcarbon diagram.The a-phase precipitation curve GS intersects the iron carbide precipitation curve ES. Thepoint S is a eutectoid point with the coordinates 7238C (13338F) and 0.80% C. At this point asaturated a solution and Fe3C precipitate simultaneously form the eutectoid concentrationg solution.The lattice of the a solid solution is identical to the lattice of the d solid solution. At theeutectoid temperature of 7238C (13338F) the a solid solution contains 0.02% C (point P).Further cooling leads to lowering of the carbon solubility in a-Fe, and at room temperature itequals a small fraction of a percent (point D).When the carbon content is 2.04.3%, crystallization starts with precipitation of theg solution at the line BC. An increase in the carbon content to over 4.3% causes precipitationof iron carbide at the line CD.Precipitation of the surplus primary phase in all alloys containing over 2.0% C is followedby a eutectic crystallization of the g solution and iron carbide at point C, whose coordinatesare 11308C (20668F) and 4.3% C.The line Ao is associated with a magnetic transformation, that is, a transition from theferromagnetic to the paramagnetic state.Table 3.1 describes structural components of the ironcarbon system. Stable FeC Equilibrium DiagramGiven very low rates of cooling, carbon (graphite) can crystallize directly from the liquid. Inthis case, a eutectic mixture of austenite and graphite is formed instead of the eutectic ofaustenite and cementite. As is seen in Figure 3.2, the dashed lines symbolizing the irongraphite system are at higher temperatures than the lines of the ironcementite system. Thistestifies to the greater stability and closeness to a full equilibrium of the irongraphite system. 2006 by Taylor & Francis Group, LLC.TABLE 3.1Components of the IronCarbon SystemPhase or Mixture of PhasesNameSolid solution of carbon in a (d) ironSolid solution of carbon in g-ironIron carbide (Fe3C)Eutectoid mixture of carbon solid solution in g-iron with iron carbideEutectoid mixture of carbon solid solution in a-iron with iron carbideFerriteAusteniteCementiteLedeburitePearliteThe conclusion is also supported by the fact that heating of high-carbon alloys with a largeamount of cementite leads to its decomposition: Fe3C ! 3Fe C.At intermediate rates of cooling, part of the alloy can crystallize according to the graphitesystem and the other part according to the cementite system.Phase equilibrium lines in the diagrams of both the systems can be displaced depending onparticular cooling rates. A most pronounced displacement can be observed for the lines ofprecipitation of the carbon solid solution in g-Fe (austenite). For this reason the diagramholds completely true only with respect to the alloys that are subjected to a relatively slowcooling rate.3.2.3 EFFECTOFCARBONA maximum solubility of carbon in a-Fe is observed at 7218C (13308F) and is equal to0.018% C. Subject to quenching, carbon can remain in the a solid solution, but soonprecipitation of phases commences, by an aging mechanism.In the a solid solution, carbon can form either (1) a homogeneous solution, a staticallyuniform interstitial distribution (a rare case), or (2) an inhomogeneous solution, with theformation of clusters at places where the crystal lattice structure is disturbed (grain boundaries, dislocations). The latter is the most probable state of the solid solution. The clustersthus formed represent an obstacle to movement of dislocations during plastic deformationand are responsible for an inhomogeneous development of the deformation at the onset ofplastic flow.To analyze the influence of the carbon content on ironcarbon alloys, every structuralcomponent should be characterized. Slowly cooled alloys comprise ferrite and cementite orferrite and graphite. Ferrite is plastic. In the annealed state, ferrite has large elongation (about40%), is soft (Brinell hardness is 65130 depending on the crystal dimension), and is stronglyferromagnetic up to 7708C (14188F). At 7238C (1338F), 0.22% C dissolves in ferrite, but atroom temperature only thousandths of a percent of carbon is left in the solution.Cementite is brittle and exhibits great hardness (the Brinell hardness is about 800); it is weaklymagnetic up to 2108C (4108F), is a poor conductor of electricity and heat, and has a complicatedrhombic lattice. Usually a distinction is made between primary cementite, which crystallizesfrom the liquid at the line CD; secondary cementite, which precipitates from the g solution atthe line ES; and tertiary cementite, which precipitates from the a solution at the line PQ.Graphite is soft. It is a poor conductor of current but transfers heat well. Graphite doesnot melt even at temperatures of 300035008C (543063308F). It possesses a hexagonal latticewith the axis relationc> 2:a 2006 by Taylor & Francis Group, LLC.FIGURE 3.3 Steel microstructure of ferrite and tertiary cementite at grain boundaries, 500.Austenite is soft (but is harder than ferrite) and ductile. Elongation of austenite is 4050%.It has lower conductivity of heat and electricity than ferrite, and is paramagnetic. Austenitepossesses an fcc lattice.The structure of the steel containing 00.02% C comprises ferrite and tertiary cementite(Figure 3.3). A further increase in the carbon content leads to the appearance of a newstructural componenta eutectoid of ferrite and cementite (pearlite). Pearlite appears first asseparate inclusions between ferrite grains and then, at 0.8% C, occupies the entire volume.Pearlite represents a two-phase mixture, which usually has a lamellar structure (Figure 3.4).As the carbon content of steel is raised to over 0.8%, secondary cementite is formed alongwith pearlite. The secondary cementite is shaped as needles (Figure 3.5). The amount ofcementite increases as the carbon content is increased. At 2% C it occupies 18% of the field ofvision of the microscope. A eutectic mixture appears when the carbon content exceeds 2%. Inrapidly cooled steels, not all the surplus phase (ferrite or cementite) has time to precipitatebefore a eutectoid is formed.Alloys with 3.6% C contain ledeburite (a eutectic mixture of carbon solid solution in g-Feand iron carbide). An electron microscopic image of the carbides is shown in Figure 3.6. Thealloys would be more properly classified with hypoeutectic white cast irons.FIGURE 3.4 Steel microstructure of pearlite, 500. 2006 by Taylor & Francis Group, LLC.FIGURE 3.5 Steel microstructure of secondary cementite (needles) and pearlite, 500.3.2.4 CRITICAL (TRANSFORMATION) TEMPERATURESCarbon has a pronounced effect on transformations of iron in the solid state. The positions ofthe lines GS and NI in the ironcarbon equilibrium diagram show that an increase in thecarbon content leads to lowering of the point A3 and raising of the point A4 with respect totheir counterparts depicted in Figure 3.2 for pure carbon. So carbon extends the temperaturerange of the d-phase.When a eutectoid (pearlite) is formed, heating and cooling curves exhibit a stop, which isdesignated as the point A1 (Ac1 on heating and Ar1 on cooling). This phenomenon takes placeat 0.9% C (point S in the FeC diagram). Precipitation of ferrite in hypoeutectoid steels (oncrossing the line GOS) shows up in heating and cooling curves as an inflection symbolized asthe point A3. The point corresponds to the g ! a transformation in pure iron. Precipitationof cementite (crossing of the line ES), which precedes the eutectoid precipitation, is seen in thecooling curve as a weak inflection designated as the point Acm (Ac,cm on heating and Ar,cm oncooling). Addition of carbon has little influence on the magnetic transformation temperature(point A2). Therefore, the line MO corresponds to the magnetic transformation in alloyswith a low-carbon content. In alloys containing greater amounts of carbon, this transformation occurs at the line GOS, which corresponds to the onset of ferrite precipitation. Ifthe carbon content is higher than the one corresponding to point S, then the magnetictransformation coincides with the temperature A1.FIGURE 3.6 Steel microstructure of electron microscopic image of iron carbides, 3000. 2006 by Taylor & Francis Group, LLC.Cementite undergoes a magnetic transformation. Whatever the carbon content, the transformation takes place at a temperature of 2102208C (4104308F). It occurs without amarked hysteresis, as does the magnetic transformation of pure iron at point A2.3.3STRUCTURAL TRANSFORMATIONS IN STEELWhen a steel part is hardened, it is heated to a high temperature in order to convert the entirestructure to the austenite phase. As discussed earlier, austenite is a single-phase structure ofiron and carbon stable at high temperatures. If the steel were cooled slowly, the austenite wouldtransform to pearlite, which is the equilibrium phase at room temperature. A pearlitic structureis an annealed structure and is relatively soft with low physical properties. If the steel is cooledvery rapidly, a very hard and strong structure called martensite forms that is a metastablephase of carbon dissolved in iron. It may be tempered to produce lower hardness structuresthat are less brittle. Intermediate cooling rates will produce other structures referred to asbainites, although this type of structure is only produced in quantity in an alloy steel. Eutectoidcarbon steels produce predominantly martensite or pearlite, depending on the cooling rate.3.3.1AUSTENITEPEARLITE TRANSFORMATIONTransformation of the fcc lattice of austenite to the bcc lattice of ferrite is hampered due tothe presence of dissolved carbon in austenite. The austenite lattice has enough spaceto accommodate carbon atoms at the centers of unit cells. The bcc lattice of ferrite has nosuch space. For this reason the solubility of carbon is lowered considerably on transition fromaustenite to ferrite. During the b ! a transformation, almost all carbon precipitates fromthe austenite lattice. In accordance with the metastable FeFe3C diagram, it precipitates as ironcarbide (cementite). This transformation can be described by three interconnected processes:1. Transformation of the g-Fe lattice to the a-Fe lattice2. Precipitation of carbon as the carbide Fe3C (cementite)3. Coagulation of the carbidesAt the temperature of point A1 processes 1 and 2 proceed almost simultaneously, with theformation of a lamellar mixture of ferrite and cementite.Atoms of dissolved carbon are distributed randomly in the lattice. For this reasoncementite nucleates in carbon-rich regions and ferrite in carbon-depleted regions that havelittle if any carbon. Such a redistribution of carbon is realized through diffusion and dependson temperature and time.When hypoeutectoid steels containing less than 0.8% C are subjected to slow cooling, thetransformation starts with the formation of ferrite at grain boundaries. The boundaries act asferrite crystallization centers. Carbon is forced inside the crystallite. As ferrite precipitates, aconcentration necessary for the ferrite formation is achieved in central volumes.When hypereutectoid steels (carbon content less than 0.8%) are subjected to slow cooling,on crossing the line ES cementite starts precipitating at grain boundaries. Here the grainboundaries also serve as crystallization sites.The carbon diffusion rate in the lattices of g- and a-Fe decreases rapidly as the temperature is lowered, since the diffusion coefficient depends on temperature asQ=RTD D0:Presenting an appropriate cooling rate, undercooling can be enhanced to such an extent thatformation of pearlite becomes impossible. 2006 by Taylor & Francis Group, LLC.In the range of low temperatures, the transformation mechanism and the character of theformed structure depend solely on the temperature at which the transformation takes place.Considering the degree of undercooling, three transformation temperature ranges are distinguished:(1) the pearlite range, (2) the intermediate range, and (3) the martensite range. A continuoustransition from one transformation mechanism to another can take place over these temperatureranges. The processes strongly depend on the content of alloying elements, especially of carbon, insteel. They can commence by a more rapid mechanism and end by a slower one.In the pearlite range, the transformation is characterized by the simultaneous formation ofa mixture of ferrite and carbide. Free ferrite or carbides can precipitate at the austenite grainboundaries. Here the formation and growth of both phases are controlled by diffusionprocesses (diffusion crystallization). Diffusion of iron and other alloying elements alsoplays a significant part. The structure fineness is enhanced as the temperature is lowered,until a longer time is required for diffusion crystallization of ferrite and carbides.3.3.2 STRUCTUREOFPEARLITEA mechanical mixture of ferrite and carbide plates is formed on transformation in the pearliterange. The rate at which nuclei of pearlite crystallization are formed depends on supersaturation of austenite with carbide, which increases as the temperature is lowered. The rate alsodepends on the diffusion rate, which decreases with temperature. The growth of pearliteislets depends in the main on the diffusion rate of carbon and iron atoms. The other decisivefactors are the degree of supersaturation and the free energy advantage during the ferriteformation. Pearlite islets grow not only through the formation of new plates but also by wayof further growth of old plates in all directions. Carbide plates grow faster than ferrite plates.The process can start, however, with the formation of ferrite nuclei. Multiple alternations ofnucleation of ferrite and cementite plates and branching of the plates of both phases lead tothe formation of plane-parallel and fan-shaped pearlite plates.Pearlite nuclei appear predominantly in the lattice regions with crystal structure defects:grain boundaries, insoluble carbides, or nonmetal inclusions such as sulfides. A very significant characteristic of pearlite is the plate-to-plate spacing. Strength properties of steelimprove with a decrease in that spacing.The formation rate of cementite and ferrite crystallization centers in the pearlite rangeaccelerates as the temperature is lowered. The plate-to-plate spacing decreases, and the finenessof the structure increases. In the eutectoid steel, the pearlite transformation takes place oncooling to 6007008C (110013008F). In this case, the plate-to-plate spacing equals 0.51 mm.Precipitation of austenite over the temperature interval of 6506008C (120011008F) providesthe plate-to-plate distance of 0.40.2 mm. In this case, the eutectoid is finer pearlite. Whenaustenite precipitates over the temperature interval of 6005008C (11009308F), an extremelyfine eutectoid mixture is formed, where the plate-to-plate spacing equals ~0.1 mm.An important characteristic that influences the properties of steel is the dimension of thepearlite colony. A decrease in the colony dimension is accompanied by a growth of the impactstrength and decrease of brittleness. The critical brittleness temperature depends on thepearlite morphology as1Tcr f pdwhere d is the pearlite colony dimension. Thus a relatively high strength pearlite is formed in thecase of the breaking of ferrite and cementite plates, forming a high density of dislocations insidethe ferrite. 2006 by Taylor & Francis Group, LLC.A better fracture strength of pearlite is achieved through spheroidization of cementiteparticles. The spheroidization can be facilitated by deformation of pearlite, subsequentheating, and holding at a temperature near Ac1. Another method providing relatively highstrength and ductility of pearlite consists in deformation during pearlite transformation. Thisleads to the formation of a polygonal structure and spheroidization of cementite.The yield stress of the ferritepearlite mixture depends on the properties of ferrite andpearlite in an additive manner:s2 fa sa (1 fa )sp ,0where fa is the volume fraction of ferrite, sa is the yield stress of pearlite, and sp is the yieldstress of pearlite.3.3.3 TRANSFORMATIONOFAUSTENITEINHYPO-ANDHYPEREUTECTOID STEELSThe transformation of austenite in eutectoid composition steels was considered above.In hypo- and hypereutectoid steels, the pearlite transformation should be preceded byprecipitation of excess phasesferrite and secondary cementite (see the FeC equilibriumdiagram in Figure 3.2).The relative amount of the structurally free excess phase depends on the degree ofaustenite undercooling. The amount of excess ferrite or cementite decreases with an increasein the cooling rate. Given a sufficient degree of undercooling, the formation of an excessphase as an independent structural component can be avoided.When a hypoeutectoid steel containing a small amount of eutectoid austenite is subjectedto slow cooling, eutectoid ferrite grows on the grains of excess ferrite and eutectoid cementiteis left as structurally free interlayers at grain boundaries. In a hypereutectoid steel, theeutectoid can also be subject to structural degeneration. Cementite, which is formed as aresult of the eutectoid precipitation under a very low cooling below the point A1 (above~7008C or 13008F), is deposited on secondary cementite. Areas of structurally free ferrite arefound alongside.This eutectoid transformation, which is accompanied by separation of the phases, isreferred to as abnormal. In normal eutectoid transformation, ferrite and cementite growcooperatively in the form of colonies with a regular alternation of the two phases. In the caseof abnormal transformation, a coarse mixture of ferrite and cementite does not have acharacteristic eutectoid structure. During a eutectoid transformation the mechanism canchange from abnormal to normal. Therefore, with a rapid cooling and a correspondinglygreat undercooling of austenite, the abnormal transformation can be suppressed altogether.Consider the forms and structure of excess ferrite in hypoeutectoid steels. The ferrite isfound in two forms: compact equiaxial grains and oriented Widmannstatten plates (Figure3.7). Compact precipitates of hypoeutectoid ferrite appear predominantly at austenite grainboundaries, whereas Widmannstatten plates are formed inside grains. The Widmannstattenferrite is observed only in steels with less than 0.4% C and rather coarse grains of austenite. Asthe dimensions of austenite grains decrease, the share of ferrite in the form equiaxial grainsgrows. The Widmannstatten ferrite is formed over the temperature interval from A3 (508C or908F) to 600 to 5508C (111210228F). With an increase in the carbon content of steel, theshare of the Widmannstatten ferrite in the structure lowers.It is assumed that the Widmannstatten ferrite is formed owing to a shear g ! arearrangement of the lattice, which is accompanied by an ordered interrelated movement ofatoms. Equiaxial grains of ferrite grow by a normal diffusive rearrangement of the lattice witha disordered transition of atoms across the g=a boundary. 2006 by Taylor & Francis Group, LLC.FIGURE 3.7 Structure of excess ferrite in hypereutectoid steel, 500.One of the methods used to strengthen steels consists in providing a structure withhypoeutectoid ferrite containing dispersed carbide precipitates. To produce such a structure,the steel should be heated until special carbides dissolve in austenite and then cooled rapidlyso as to preclude the usual precipitation of carbide directly from austenite before hypoeutectoid ferrite starts forming.3.3.4 MARTENSITE TRANSFORMATIONThe martensite transformation takes place on quick cooling of the high-temperature phase,a process that is referred to as quenching. The most characteristic features of the martensitetransformation in carbon steels are as follows:Temperature, C1. The martensite transformation is realized on rapid cooling of steel from a temperatureabove A1 in, for example, water. In this case, diffusive precipitation of austenite to amixture of two phases (ferrite and carbide) is suppressed. The concentration of carbon inmartensite corresponds to that in austenite. The main difference between the martensitetransformation and the pearlite transformation is that the former is diffusionless.2. Transformation of austenite to martensite starts from the martensite start temperature(Ms). Whereas the pearlite start temperature lowers with an increase in the cooling rate,the martensite start temperature depends little if at all on the cooling rate. Martensite isformed over a certain temperature interval. The particular temperature is determinedby the carbon content of the steel (Figure 3.8).3. Termination of cooling over the temperature interval MsMf suspends formation ofmartensite. This feature distinguishes the martensite transformation from the pearlite800600400Ms200Mf0200Fe 0.40.8 1.21.6CQuantity, %FIGURE 3.8 Martensite start Ms and finish Mf temperatures versus carbon content. 2006 by Taylor & Francis Group, LLC.transformation. In the latter case, transformation continues to the end at a constanttemperature below the point A1, and the final result is a complete disappearance ofaustenite given a sufficient isothermal holding time. With the martensite transformation, a certain amount of retained austenite is left.4. As distinct from the pearlite transformation, the martensite transformation has noincubation period. A certain amount of martensite is formed instantaneously below thetemperature Ms.5. On cooling below Ms, the amount of martensite increases rapidly owing to the quickformation of new plates. The initially formed plates do not grow with time. This featurealso distinguishes the martensite transformation from its pearlite counterpart; in thelatter case new colonies nucleate and old colonies continue growing.6. The martensite lattice is regularly oriented relative to the austenite lattice. A certainorientation relationship exists between the lattices. With the pearlite transformation,lattices of the phases comprising the eutectoid mixture exhibit a random orientationwith respect to the starting austenite grain.The temperature Ms characterizes an alloy of a certain composition that has beensubjected to a particular pretreatment. In a given steel, the martensite transformation startsat the same temperature whatever the cooling rate. That temperature depends on the alloycomposition and decreases greatly as the carbon content of the steel is raised (see Figure 3.8).Part of the carbon enters carbides, which coexist with austenite. The carbides dissolve inaustenite if the quenching temperature is elevated. Consequently, the carbon concentration ofaustenite increases and the Ms point lowers.The martensite formation is characterized by a shear mechanism of the austenite latticerearrangement. The martensitic (shear) mechanism of phase transformation is distinguishedby an ordered interrelated movement of atoms to distances shorter than the interatomicspacing, and the atoms do not exchange places. An atom in the initial phase preserves itsneighbors in the martensite phase. This is the main feature specific to a shear rearrangementof the lattice.This character of the lattice rearrangement provides coherence of the boundary betweenthe old and new phases. Coherence, or an elastic conjugation of lattices at the boundary between martensite and the initial phase, ensures a very fast movement of the boundary toward thematrix even at low temperatures. The atoms move cooperatively to distances shorter thanthe interatomic spacing; hence the growth of the martensite crystal.As the martensite crystal grows, an elastic strain accumulates at the coherence boundary.On reaching the yield stress, coherence is disturbed. Atoms become disordered at the boundarybetween the martensite crystal and the starting matrix. Slipping movement of the boundary isrendered impossible. Hence, growth of the crystal by the martensitic mechanism is terminated, and subsequently the crystal can grow by diffusion only. But the martensite transformation takes place at low temperatures, where the diffusion rate is very small. Therefore, aftercoherence is broken, little if any growth of the martensite crystal is observed.The polymorphous transformation of solid solutions by the martensitic mechanism ischaracterized by the absence of diffusive redistribution of the components. We consider theconditions necessary for the martensitic mechanism by which the high-temperature phasetransforms to the low-temperature phase in the following discussion. The martensite transformation is impossible at a small undercooling. This is explained by the fact that in the caseof a disordered rearrangement of the lattice, elastic deformation is determined by changes inthe volume only, whereas with the martensite transformation, it additionally depends oncoherence of the lattices of the initial and martensite crystals. As the degree of undercooling isincreased, the disordered rearrangement rate of the lattice increases, achieves a maximum, 2006 by Taylor & Francis Group, LLC.and then drops. When g-Fe is undercooled to 911 to 7508C (167013808F), the normalg ! a transformation is realized, while below 7508C (13808F) the martensite g ! atransformation takes place. To realize the martensitic mechanism of polymorphous transformation in iron, samples should be strongly overheated in the g range and then cooled veryquickly to suppress development of the normal transformation.3.3.5 MORPHOLOGYOFFERROUS MARTENSITESConsider the crystallogeometry of the rearrangement of the fcc lattice of austenite to the bcctetragonal lattice of martensite, which is similar to the bcc lattice of a-Fe.The austenite lattice transforms into the martensite lattice through the Bain deformation.The deformation consists in compression of the tetragonal cell of austenite along the c-axisand a simultaneous increase in dimensions along the a-axis. The degree of the tetragonaldistortion of the martensite lattice, c=a, grows directly as the carbon concentration of martensite. The martensite lattice retains tetragonality at room temperature.The orientation relationship of the initial and martensite phases has been established.Three basic orientation relationships are known for austenite and martensite lattices in ironalloys: those due to Kurdyumov and Zacks, Nishiyama, and Treninger and Trojano.The KurdyumovZacks relationship: (1 1 1)A k(1 0 1)M; [1 1 1 0]Ak[1 1 1]M.The Nishiyama relationship: (1 1 1)A k(1 0 1)M; [1 2 1]Ak[1 0 1]M.The TreningerTrojano relationship is intermediate between the first two relationships.Several hypotheses are available as to the character of martensite nucleation. Most ofthem suggest a heterogeneous nucleation at special defect sites in the starting matrix. It wasshown experimentally that the sites do not include grain and subgrain boundaries, as these arenot places of preferable nucleation of martensite. They might be stacking fault arising in theg-phase during splitting of dislocations. According to other hypotheses, the sites includespecial configuration dislocation pile-ups or separate dislocations, which are the sources offields of internal stresses. This decreases the work on critical nucleus formation.By morphology, martensite can be divided into two basic types: plate and massivemartensite. They are different in shape, mutual arrangement of crystals, substructure, andhabit plane. Plate (needle) martensite is found most frequently in high-carbon steels andcarbon-free iron alloys. Martensite crystals are shaped as thin lenticular plates (Figure 3.9).Neighboring plates are not parallel to one another.FIGURE 3.9 Martensite plates and retained austenite (dark) in quenched steel, 1000. 2006 by Taylor & Francis Group, LLC.Plates that appear first pass throughout the unit, dividing it into separate parts. But theycannot cross the matrix grain boundary. Therefore, the plate dimension is limited by thedimension of the austenite grain. New martensite plates are formed in austenite sections.Here the plate dimension is limited to the dimension of the section (see Figure 3.9). If theaustenite grain is small, martensite plates are so fine that the needle structure of martensite cannot be seen in it microsection specimens. Such martensite is called structurelessmartensite, and it is most desirable.Massive (lath) martensite can be observed in low- and medium-carbon steels. Crystals ofthis type of martensite are shaped as interconnected plates having approximately the sameorientation. The habit plane of laths is close to the {1 1 1}A plane. Plates of massivemartensite are separated with low-angle boundaries. An electron microscopic image ofmassive martensite is given in Figure 3.10. As is seen, a package of plates is the mainstructural component. Several martensite packages can be formed in an austenite grain.3.3.6BAINITE TRANSFORMATIONThe bainite transformation is intermediate between pearlite and martensite transformations.The kinetics of this transformation and the structures formed exhibit features of bothdiffusive pearlite transformation and diffusionless martensite transformation.A mixture of the a-phase (ferrite) and carbide is formed as a result of the bainitetransformation. The mixture is called bainite. The bainite transformation mechanism involvesg ! a rearrangement of the lattice, redistribution of carbon, and precipitation of carbide.Most researchers are of the opinion that ferrite precipitates from austenite by the martensitic mechanism. This is attested to by the presence of retained austenite in alloyed steels,a similarity in the structure of lower bainite and martensite, and the resemblance of upperbainite to low-carbon martensite.Closeness of the bainite transformation to its pearlite and martensite counterparts can beexplained as follows. The diffusive movement of atoms of the basic component, iron, isalmost completely suppressed over the bainite transformation range. Then the g ! aformation of ferrite is difficult because pearlite precipitation is suppressed. However, carbondiffusion is rather active and causes precipitation of carbides.Over the intermediate range the g-phase crystals are formed through coherent growthsimilarly to martensite plates. But the a-phase plates are formed slowly rather than instantaneously. This is due to the fact that over the intermediate temperature range the a-phase canFIGURE 3.10 Electron microscopic image of lath martensite, 20,000. 2006 by Taylor & Francis Group, LLC.precipitate only from the carbon-depleted g-phase. Thus the growth rate of the a-phasecrystals depends on the carbon diffusive removal rate. In this case, the martensite startpoint Ms in austenite rises and the martensite g ! a transformation takes place at temperatures above the temperature Ms typical of the steel with a given composition.At the instant of martensite transformation the carbon concentration remains unchanged.Only the crystal lattice is altered and a supersaturated a solution is formed. Carbide precipitates after g ! a transformation.3.3.7 MORPHOLOGYOF THEBAINITE TRANSFORMATIONA distinction is drawn between upper and lower bainite, which are formed in the upper andlower parts of the intermediate temperature range. The conventional boundary between thebainites is close to 3508C (6608F). Upper bainite has a feathery structure, whereas lowerbainite exhibits an acicular morphology, which is close to that of martensite. The difference inthe structures of upper and lower bainites is attributed to a different mobility of carbon in theupper and lower parts of the bainite temperature range.An electron microscopic analysis showed that the a-phase substructure of upper bainiteresembles the substructure of massive martensite in low-carbon steels, while the a-phasestructure of lower bainite approximates the structure of martensite in high-carbon steels. Inupper bainite, carbide particles can precipitate both at lath boundaries and inside laths. Thisfact suggests that here carbides precipitate directly from austenite. In lower bainite, carbide isfound inside the a-phase. This means that carbide is formed during precipitation of asupersaturated solid solution of carbon in the a-phase. Both upper and lower bainites exhibita high density of dislocations inside the a-phase.Cementite is the carbide phase in upper bainite, and e-carbide in lower bainite. As theholding time is increased, e-carbide turns into cementite. The austenite grain dimensions haveno effect on the martensite transformation kinetics.3.3.8 TEMPERINGThe main processes that take place during tempering are precipitation and recrystallization ofmartensite. Quenched steel has a metastable structure. If subjected to heating, the structurebecomes closer to equilibrium. The character of the processes that occur during tempering isdetermined by three major features of quenched steel: strong supersaturation of the martensite solid solution, high density of crystal lattice defects (dislocations, low- and large-angleboundaries, twin interlayers), and the presence of retained austenite.The main process taking place during tempering of steels is precipitation of martensiteaccompanied by formation of carbides. Depending on the temperature and duration oftempering, the martensite precipitation may involve three stages: preprecipitation, precipitation of intermediate metastable carbides, and precipitation and coagulation of cementite.Retained austenite can precipitate simultaneously.Owing to a high density of dislocations in martensite, its substructure is similar to thesubstructure of a work-hardened (deformed) metal. Hence, polygonization and recrystallization can develop during tempering.When carbon steels are tempered, supersaturation of the g0 solution in austenite increaseswith an increase in the carbon content of steel. This leads to lowering of the temperature Msand transition from massive martensite to plate martensite. The amount of retained austenitealso increases.Carbon segregation represents the first structural changes that take place during tempering of carbon steels. The segregated carbon can nucleate heterogeneously at lattice defects or 2006 by Taylor & Francis Group, LLC.homogeneously in the matrix. The heterogeneous nucleation of the segregated carbon occurseither during quenching or immediately after it.Flat homogeneous clusters of carbon atoms that are not connected with lattice defects areformed at tempering temperatures below 1008C (2128F). Their formation is due to considerable displacements of iron atoms and the appearance of elastic distortions. As the temperingtemperature is increased, the clusters become larger and their composition is close to Fe4C.This process depends on carbon diffusion. Metastable e-carbide (Fe2C) is formed above1008C (2128F). It possesses a hexagonal lattice and appears directly from carbon clusterswhen the carbon concentration is increased. Metastable e-carbide can also precipitate directlyfrom the a solution. At low temperatures e-carbide precipitates as very fine (10100 nm)plates or rods (Figure 3.11). With an increase in tempering temperature or time, e-carbideparticles become coarser. This carbide precipitates in steels containing a minimum of 0.2% C.In steels having a high Ms temperature, i.e., in all structural steels, partial precipitation ofmartensite accompanied by deposition of excess carbide is accomplished during quench coolingin the martensite range. Then self-tempering of these steels occurs during their quenching.Cementite, Fe3C, is formed at a temperature above 2508C (4828F). Two mechanisms ofcementite nucleation have been known. First, it precipitates directly from a supersaturateda solid solution. Cementite particles grow at the expense of the dissolution of less stablecarbides. Second, cementite appears as a result of transformation of the intermediate carbidelattice to the Fe3C lattice.The final stage of the carbide formation during tempering is coagulation and spheroidization of carbide. These processes develop intensively starting from 350 to 4008C (6607508F).Above 6008C (11128F), all cementite particles have a spherical shape and undergo coagulation only.A considerable part of the tempering process is devoted to the precipitation of retainedaustenite accompanied by deposition of carbides. Precipitation occurs over the temperatureinterval 2003008C (4005708F). During tempering, retained austenite transforms into lowerbainite.A decrease in the carbon concentration of the a-phase during carbide formation causeschanges in the phase structure. Martensite precipitation can conventionally be divided intotwo stages. The first stage of precipitation is realized below 1508C (3008F). At these temperatures, the mobility of carbon atoms is sufficient for the formation of carbide plates.However, it is insufficient for the carbide plates to grow by diffusion of carbon from the areasof unprecipitated martensite with a high-carbon concentration. This results in a nonuniformFIGURE 3.11 Electron microscopic image of the e-carbide, 50,000. 2006 by Taylor & Francis Group, LLC.content of carbon in different areas of the martensite and consequently inhomogeneity ofmartensite with respect to its tetragonality. In areas with precipitated carbide, tetragonality islower than in unprecipitated areas. Two solid solutions with different carbon concentrationscoexist. For this reason the precipitation is referred to as a two-phase precipitation. The twophase precipitation of martensite results from the deposition of new carbide particles in areascontaining martensite with the initial carbon concentration. Carbide particles do not grow atthis stage.At the second stage of martensite precipitation (1503008C; 3005708F) the a solution isdepleted of carbon owing to diffusive growth of carbide particles. But the process proceedsvery slowly. Therefore, the precipitation kinetics are characterized by a rapid depletion of thea solution in carbon (the timespan decreases as the annealing temperature is increased).Subsequently, depletion of the solid solution in carbon stops. At 3008C (5708F) about0.1% C is left in the a solution. Above this temperature no difference between the lattice ofthe a solution and that of the a-Fe is detected. Below 3008C the degree of tetragonality(c=a > 1) is still measurable. Above 4008C (7508F) the a solution becomes completely free ofexcess carbon and transformation of martensite to ferrite is finished.As mentioned earlier, plates (needles) of quenched martensite have a high density ofdislocations, which is comparable to the density of the deformed material. However, recrystallization centers and their development to recrystallized grains have not been observed. Thisis because carbide particles pin dislocations and large-angle boundaries. It is only above6008C (11128F), when the density of the particles decreases owing to their coagulation, thatrecrystallization growth of grains takes place at the expense of migration of large-angleboundaries. Therewith the morphological features of lath martensite vanish. These processesare hampered in high-carbon steels in comparison with low-carbon alloys, because the densityof carbides is greater in high-carbon steels. The acicular structure is retained up to thetempering temperature of about 6508C.The structural changes that occur during tempering cause alteration of steel properties.They depend on the tempering temperature and time. Hardness lessens as the temperingtemperature is raised.3.4 KINETICS OF AUSTENITE TRANSFORMATION3.4.1 ISOTHERMAL TRANSFORMATION DIAGRAMSTo understand the kinetics of transformations to austenite, it is important to follow theprocess at a constant temperature. To this end, a diagram was constructed that characterizesthe isothermal process of austenite precipitation. In this diagram, the transformation time isthe abscissa on the logarithmic scale and the temperature is plotted on the ordinate (seeFigure 3.12). From this diagram, the incubation period (left-hand curve) can be determinedand also the time required for completion of the process (right-hand curve). The instant analloy passes the points A3 and A1 during quenching is usually taken as the zero time reference.The time required to achieve the temperature of the quenching medium is often neglected.The start and finish of the transformation are difficult to determine from the transformationcurve behavior at the initial and final sections of the curve. Therefore, the lines of theisothermal transformation (IT) diagram usually correspond to a certain final volume thatunderwent transformation, e.g., 3 and 97% for the transformation start and finish, respectively. The volume value is usually not shown in the diagram.In addition to the above-mentioned curves, the diagram often contains intermediatecurves that correspond to certain values of the transformed volume, e.g., 10, 50, or 90%.A decrease in the transformation rate causes displacement of the transformation start and 2006 by Taylor & Francis Group, LLC.Temperature of quenching 885 CA1800A1700Temperature, C600500400300200Martensite10000.511010260121034min8153011041056024681624TimeFIGURE 3.12 Isothermal transformation diagram.finish curves to the right, i.e., toward greater duration. This phenomenon can be observed ifthe quenching heating temperature rises as a result of a decrease in the number of foreigninclusions, enlargement of austenite grains, etc.An increase in the transformation rate leads to displacement of the IT curves to the left.This phenomenon can be accounted for by a decrease in the quenching heating temperature,the presence of carbides or foreign inclusions, and refinement of the austenite grain. For agiven steel the temperature that corresponds to a maximum transformation rate (the so-callednose of the sigmoid curve) does not, as a rule, change significantly.3.4.2 CONTINUOUS-COOLING TRANSFORMATION DIAGRAMSContinuous-cooling transformation (CCT) diagrams consider the transformation kinetics ofa eutectoid steel. The major transformation that takes place during annealing cooling of steelis a eutectoid precipitation of austenite into a mixture of ferrite and carbide. The eutectoidtransformation kinetics are given by IT diagrams of austenite (Figure 3.13) at a temperatureof 7278C (13408F). The structure obtained after tempering below 3008C (5728F) is calledtempered martensite. An acicular structure is observed after tempering at 3004508C (5728428F). Tempering over the temperature interval of 4506008C (84211128F) exhibits apronounced dot structure. The structure obtained after tempering below 3008C (5728F) iscalled tempered martensite. Austenite is in a thermodynamically stable equilibrium with theferritecementite mixture. Stability of undercooled austenite is defined by a period of timeduring which the appearance of precipitation products in the diagram cannot be registered byconventional methods (see Figure 3.13). It is equal to the distance from the y-axis to the lefthand curve. The degree of austenite undercooling is the main factor that determines the steelmicrostructure. The necessary degree of undercooling is provided by either continuous cooling or isothermal treatment. The diagram in Figure 3.13 shows the entire range of structuresformed in a eutectoid steel depending on particular undercooling conditions. 2006 by Taylor & Francis Group, LLC.ginolCoLines of starting and finishingof transformationAiAinPearlitemoltO3 + NAU + F + Cof KNben Plt bathTemperature, 8CeacrnfuHold in furnaceg ing in salinCooCoolinAu450 C Upper bainiteaNO 3230 C250 CLower bainiteMartensiteM101001000Time, logFIGURE 3.13 Isothermal transformation diagram for a eutectoid composition steel. A, stable austenite;Au, undercooled austenite; F, ferrite; C, carbide. Transformations That Take Place under Continuous Cooling of Eutectoid SteelsAs mentioned above, in hypoeutectoid steels the formation of pearlite is preceded by precipitation of hypoeutectoid ferrite. With a decrease in the transformation temperature and a risein the degree of undercooling, precipitation of hypoeutectoid ferrite is suppressed. Theamount of pearlite increases and the carbon content becomes less than that in pearlite ofthe eutectoid steel. In the region of the maximum transformation rate, the two curves merge.Thus, a purely pearlitic structure is formed in steel with 0.4% C (Figure 3.14). In steelscontaining greater amounts of carbon, the precipitation of ferrite cannot be suppressedeven if the carbon content decreases. Ferrite precipitation precedes the formation of pearliteA r3 Upper critical rateTemperature, 8CA r2A r1Lower critical rateArMartensite point (A r )Rate of coolingFIGURE 3.14 Schematic diagram showing changes in location of the critical points depending on theparticular cooling rate. 2006 by Taylor & Francis Group, LLC.even at a maximum transformation rate, but the amount of ferrite will be less than that wasformed at smaller undercooling.These propositions are valid for the precipitation of cementite in hypereutectoid steels, butit can be suppressed even at relatively small undercooling. In this case, the carbon content ofpearlite becomes higher than that in the eutectoid steel.As a result of suppression of the hypoeutectoid ferrite precipitation under continuouscooling from the region of the g solid solution, the point Ar3 lowers much faster than thepoint Ar1 as the cooling rate is increased. Given a certain cooling rate, both points merge into0one point A 2 (see Figure 3.14), which corresponds to the formation of a fine plate structure ofthe pearlite type free of ferrite.Under continuous cooling the transformation process can also be pictured as diagrams intemperaturetime coordinates (Figure 3.15). Hence the behavior of cooling curves should beanalyzed to obtain characteristics of the transformation processes. In this diagram, the ferriteand pearlite start lines are shifted toward longer periods of time compared to the IT diagramof Figure 3.13. This is due to an increase in the temperature interval necessary for preparingthe transformation processes in the austenite lattice. As a result, only part of the incubationperiod, which is required for the IT to start, is effective. In this case, the incubation period isthe mean of the effective lengths of time corresponding to different periods of time in thegiven range. This proposition can be used to calculate the behavior of the transformation startline in the pearlite range from the IT diagram. The reverse calculation is also possible.Similar to the pearlite range, in the bainite temperature range, the precipitation of undercooled austenite starts after a certain incubation period. Resemblance of the bainite and pearlitetransformation kinetics consists not only in the presence of an incubation period but also in thecharacter of the volume increase during isothermal soaking: the fraction of the transformedvolume of austenite increases first with acceleration and then with deceleration. At the same time,as in the case of the martensite transformation, retained austenite does not disappear completelyduring the bainite transformation. Every point in the bainite finish curve corresponds to a certainamount of retained austenite. Similar to the pearlite transformation, the bainite transformationcan take place both during isothermal soaking and under continuous cooling (see Figure 3.15).Austenite that has not been transformed over the bainite range turns partially into martensite1000900AcTemperature, C800700600A75P125004003AcMs300F35 3535 6513565652065B7 10 3030M200100580104627 2310220103220104105106Time, sFIGURE 3.15 Continuous-cooling transformation (CCT) diagram. A, austenite range; F, ferrite range;P, pearlite range; B, bainite range; M, martensite range. Shown in circles is hardness HV or RC.Numerals at the curves denote the relative amounts of structural components. 2006 by Taylor & Francis Group, LLC.Temperature, C8007006005004003002002131101102103Log time (t )4104FIGURE 3.16 Diagram of isothermal precipitation of austenite in steel with 0.43% C and 3% C. Curve 1,pearlite formation start; curve 2, pearlite formation finish; curve 3, bainite formation start; curve 4,bainite formation finish.when the steel is cooled to room temperature. Because after the bainite transformation austeniteis inhomogeneous with respect to the carbon content, martensite is formed predominantly incarbon-enriched regions.In the case of high-alloy steels, isothermal curves can be separated by a temperatureinterval in which undercooled austenite is highly stable. In this interval, pearlite precipitationdoes not take place for many hours, while undercooling is insufficiently great for the bainitetransformation (Figure 3.16). In carbon steels, the bainite transformation proceeds concurrently with the pearlite transformation. Products of the pearlite transformation dominate athigher temperatures, and those of the bainite transformation at lower temperatures. Transformations of Austenite on Cooling in the Martensite RangeThe martensite component in the steel structure appears when the cooling rate achieves acertain value. The minimum cooling rate at which the martensite component is formed iscalled the lower critical rate of cooling. The rate at which transformations by the pearlite andbainite mechanisms are suppressed completely is referred to as the upper critical rate ofcooling (quenching). If the conditions of austenite formation (austenitization temperatureand the holding time at this temperature) and the cooling conditions (cooling rate shouldexceed the upper critical rate) are constant, the location of the martensite point Ms dependsonly on the content of carbon and alloying elements in the steel.If the cooling rate is high, the formation rate of separate needles of martensite is also high,and transformation of austenite to martensite commences on reaching Ms. It continues onsubsequent cooling to lower temperatures. As the temperature of the quenching medium islowered, the amount of formed martensite rises first rapidly and then slowly. With an increasein the quenching heating temperature (austenitization temperature), the transformation alsoshifts toward lower temperatures (Figure 3.17) as more of the alloying elements are taken intosolution. A certain amount of martensite may be formed during isothermal holding, but it isnot high in carbon steels. Retained austenite is stabilized during isothermal holding. As aresult, more martensite is formed during subsequent cooling. Formation of martensite stopsat the point Mf.Figure 3.18 shows a relationship between some factors that influence the stabilization ofmartensite. As is seen, if continuous cooling is stopped at the temperature Th1 and a holdingtime is allowed at this temperature, the formation of martensite starts after passing through acertain temperature interval rather than immediately when cooling is resumed. Subject tocooling below the point M 0 s, further formation of martensite takes place. If holding is realizedat a lower temperature, Th2, the effect of stabilization is enhanced, because further formationof martensite commences at the temperature Ms2 after passing through a greater temperature 2006 by Taylor & Francis Group, LLC.Volume fraction of martensite10084580609254010402002050100150Temperature, C200250FIGURE 3.17 Curves showing variation of the relative amount of martensite as a function of thetransformation in steel with 1.1% C and 2.8% Cr for different homogenization temperatures.interval (curve 3, Figure 3.18). The effect of stabilization increases with the amount ofmartensite in the structure or, the amount of martensite being equal, with temperature.Joining of the points M0 s determined after holding at different temperatures yields a curvethat intersects the curve that corresponds to the relative amount of martensite formed undercontinuous cooling. The point of intersection of the curves, ss, means that stabilization ofaustenite is impossible at a higher temperature.3.4.3DERIVATION OF THE CONTINUOUS-COOLING TRANSFORMATION DIAGRAMTHE ISOTHERMAL TRANSFORMATION DIAGRAMFROMQuantity of martensiteWhen solving practical problems involved in thermal treatment of steel, it is often necessaryto know how the continuous cooling rate affects the structure formed as a result of austenitetransformation. To this end, attempts were made to establish the relationship between thetransformation kinetics of austenite under isothermal conditions (IT diagram) and undercontinuous cooling (CCT diagram). The attempts started from the concept of additivity of thetransformation processes at different temperatures. It was assumed that holding of undercooled austenite at a preset temperature is part of the incubation period. It was found,however, that calculated and experimental data coincide satisfactorily only if the pearlitetransformation is continuous.1234Ms2Th2 Ms1Th1dsMsTemperature, 8CFIGURE 3.18 Curves showing the relative amount of martensite as a function of the austenite stabilization temperature. 1, Under continuous cooling; 2, after isothermal holding at Th1; 3, after isothermalholding at Th; 4, M0 s curve, ts is the limiting stabilization temperature. 2006 by Taylor & Francis Group, LLC.If the pearlite transformation is preceded by precipitation of eutectoid pearlite or thepearlite and bainite transformations occur concurrently, calculated data are at a discrepancywith the experimental data. It was found that the discrepancy is due to the following factors:1. Holding of austenite during the time accounting for fractions of the incubation periodcauses acceleration of the subsequent intermediate transformation at the expense ofpreparatory processes.2. Precipitation of hypoeutectoid ferrite alters the austenite composition. This delays thesubsequent intermediate transformation.3. Partial transformation of austenite over the intermediate range decreases the rate of thesaid transformation at lower temperatures and facilitates an increase in retainedaustenite. This is due to a redistribution of carbon and enrichment of the nontransformed part of austenite in carbon.4. A change in the cooling rate over the martensite range affects stabilization of austenitein different ways.For this reason, special methods of constructing thermokinetic transformation diagrams ofaustenite subject to continuous cooling were elaborated for noneutectoid steels (see Figure3.15). From these diagrams it is possible to determine the critical rate of quenching cooling orcontinuous cooling that is necessary to complete a particular stage of austenite precipitation.3.4.4 CONTINUOUS-COOLING TRANSFORMATION DIAGRAMOF THE BAR DIAMETERAS AFUNCTIONWhen steel is subjected to martensitic hardening, it should be cooled from the quenchingtemperature so that on undercooling to a temperature below the Ms point austenite has notime to precipitate and form a ferritecarbide mixture. To achieve this, the cooling rate shouldbe less than the critical value. The critical cooling rate is the minimum rate at which austenitedoes not precipitate to a ferritecarbide mixture. Of course, the cooling rate of steel productsis nonuniform over their cross section. It can be higher than the critical rate on the surfaceand lower than the critical rate at the center.The critical cooling rate at different points of a product can be directly determined froman IT diagram (Figure 3.19). In the first approximation, it is given by the slope of the tangentto the C curve that denotes the austenite precipitation onset. This method gives a value that is(a)(b)Vcoolingt12eUnquenching fieldQuenching fieldCenterSurfacTemperature, 8CVcriticalSurface1r2rCenterTime,tFIGURE 3.19 Diagram showing distribution of the cooling rate over the cross section of a sample andthe corresponding IT curve. 2006 by Taylor & Francis Group, LLC.800Temperature, C70025600% Ferrite403535235006520CBA% Pearlite60655DE40010 40 401303005 % Bainite50%95% Martensite2001002352 48 46570130 27 252102321 20171031010410510660s1(a)248 15 30min60124 6 8 16 24hTime124 7 10Days60AHardness, HRC5040BC30ED201000510(b)15202530354045505560657075Distance from cooling surface, MMFIGURE 3.20 (a) Timetransformation temperature diagram for continuous cooling of steel containing0.38% C compared with (b) the process of the samples cooling during face quenching. Numerals at thebottom of the curves denote hardness (RC) after cooling to room temperature.about 1.5 times the true critical rate. The cooling rate can be determined more accurately ifone uses thermokinetic diagrams (Figure 3.20). Intercepts of the cooling curves with the linesof the thermokinetic diagrams show the start and finish temperatures of the correspondingtransformation.From the transformation diagram it is possible to determine, for example, the rate thatwill provide 50% martensite in the structure or the rates at which the entire transformationoccurs in the pearlite range, i.e., hardening is excluded altogether. Because the data on thecritical hardening rate depend on cooling time and should be associated with a particulartemperature (at which direct measurements of the hardening rate are practically impossible),it is appropriate to specify the cooling time for a specific temperature interval, for example,from the point A3 to 5008C (9328F). Point A3 in the diagram is the time reference. Then it ispossible to straightforwardly determine the critical cooling time K: Km for fully martensitichardening; Kf for initial appearance of ferrite; Kp for full transformation in the pearlite range.Since the cooling time (see Figure 3.20) and the progress of the subsequent cooling of the 2006 by Taylor & Francis Group, LLC.sample during end-face hardening are known, the outcome of hardening can be determinedfrom the transformation diagram. It should be remembered that a transformation diagram isvalid only for particular conditions of melting and homogenization. Deviations in thecomposition or grain dimensions cause changes in the trend of thermodynamic curves. Thisis explained by the fact that an increase in the homogenization temperature and time and,consequently, enlargement of the grains enhance the stability of austenite. Conversely,refinement of grains lowers the critical cooling rate, because stability of austenite decreaseswith an increase in the extent of grain boundaries.3.4.5 DEFINITIONOFHARDENABILITYHardness, HRCThe depth of the hardened zone is termed hardenability. This is one of the most importantcharacteristics of steel. Since the cooling rate is nonuniform along the cross section of asample (see Figure 3.19), austenite can pass into martensite in surface layers only, while at thecenter of the sample austenite undergoes the pearlite transformation. In the first place,hardenability depends on the critical cooling rate. An examination of the temperature curves(see Figure 3.20) plotted for different areas of the sample shows that the cooling rate of thecore of a large-diameter product is lower than the critical value and therefore the core is notmartensitically hardened. Martensite is present in the surface layer only.After hardening treatment, a bulky part with a large cross section may exhibit the entirerange of structures: a smooth transition from martensite near the surface through troostitemartensite and troostite to pearlite at the center.The geometry of samples can influence the character of the cooling curves. However,given the same surface-to-volume ratio, the curves coincide in general. The greatest changes inthe cooling rate are incurred by the diameter of samples.Considering what has been said above, to achieve a through hardening of bulky productsor full martensitic hardening to the core of a product, one has to provide the critical hardening rate along the entire cross section of the product. IT and CCT diagrams can be used todetermine this rate. The diagrams were plotted for different grades of steel, taking intoaccount the progress of cooling in different sections and in different hardening media.Note that hardenability depends on the steel composition, specifically on the carboncontent. Hardenability of each grade of steel is presented as a hardenability band (Figure3.21). These diagrams have been plotted for almost all existing grades of steel. They show howto achieve hardening of a product made of a particular steel.6560555045403530252015100510 15 20 25 30 35 40 45Distance from cooling surface, mmFIGURE 3.21 Steel hardenability band. 2006 by Taylor & Francis Group, LLC.50Hardenability of steel is also characterized by transformation timetemperature curves(IT curves). The more the curve is shifted to the right along the abscissa axis, the greater is thehardenability of the steel. This is explained by the fact that the rightward shift of the IT curveis due to better stability of austenite.An improvement in the stability of undercooled austenite and hence an increase in the criticalhardening rate lead to a greater depth of hardening. Then hardenability depends on all thefactors that improve the stability of undercooled austenite. For example, the stability of austenitecan be raised by alloying steel with chromium and tungsten. These elements lower the austeniteprecipitation rate and can make a steel an air-hardening one. Steel with a usual (commercial)content of impurities is hardened to a strength ten times that of a pure ironcarbon alloy.Elevation of the hardening temperature favors an increase in the hardening depth thanksto the homogenization of austenite and enlargement of austenite grains. Refinement of grainsimpairs hardenability as grain boundaries affect the stability of austenite.The hardening depth also depends on the hardening medium used. The greater theintensity of cooling, the greater the depth of hardening. Besides, the hardening depth dependson the cross-sectional diameter of the products. The critical diameter is that of the greatestcross section that lends itself to through hardening in a given hardening medium. The criticaldiameter is different for different hardening media and characterizes the hardenability provided by a particular method only.Hardenability has an effect on the mechanical properties of steel. In the case of throughhardening, the properties do not differ along the cross section of a product. Otherwise theydecrease from the surface to the center. Let us analyze the influence of hardenability on theproperties of steels that were tempered after hardening. A high temperature favors equalizationof hardness along the cross section. However, the structure of weakly hardenable steels remainsinhomogeneous; a grain structure will appear on the surface, where martensite is formed duringquenching, while a lamellar structure will remain at the center. A grain structure will be presentalong the entire cross section of a through-hardening steel. This determines the character ofchanges in the properties of steels with different hardenability. The properties that are independent of the cementite form (yield stress, specific elongation, impact strength) will differ.A decrease in ss and ak is observed at the center of nonthrough-hardening steels, while in athrough-hardening steel these quantities remain unchanged along the cross section.The properties of tempered steels (fracture stress, yield stress, impact strength, reductionof area) are impaired if ferrite precipitates during quenching. The mechanical properties of aproduct depend on its cross-sectional area. To obtain the best mechanical properties in thetempered state, a grain structure should be provided along the entire cross section; i.e.,through hardenability should be ensured in the quenched state. SIZESTRUCTUREOFGRAIN BOUNDARIESWhen analyzing any processes or properties associated with grain boundaries, it is necessaryto know the structure of the material. The overwhelming majority of structural materials arepolycrystalline. They comprise a set of grains separated by boundaries. The grain boundary isone of the basic structural elements in polycrystalline materials.The grain boundary represents an interface between two differently oriented crystals. Thisis the region of crystal imperfection. It is capable of moving and adsorbing impurities. Theboundary has a high diffusive permeability.In polycrystalline materials, the boundaries determine the kinetics of many processes. Forexample, movement of grain boundaries controls the process of recrystallization. A high 2006 by Taylor & Francis Group, LLC.FIGURE 3.22 Electron microscopic image of (a) low-angle boundary and (b) large-angle boundary,50,000.diffusive permeability of grain boundaries determines the kinetics of diffusion-dependentprocesses at moderate temperatures. Grain boundaries adsorb impurities. Embrittlement ofmetal material is connected with enrichment of grain boundaries in impurities.Grain boundaries may conventionally be divided into two large groups: low-angle andlarge-angle boundaries. Low-angle boundaries (or subgrain boundaries with an angle of lessthan 108) represent networks or walls of dislocations. The structure of large-angle boundariesis much more complicated. Figure 3.22 shows both types of grain boundaries.The progress in understanding the structure of grain boundaries is connected withelaboration of the models describing the observed microscopic properties of the boundaries. Structural ModelsThe pioneering structural model is the model of an amorphous boundary. It allows anexplanation of the value of the surface tension G3 and the grain boundary slip. In terms ofthis model, it was assumed that the usual boundary with a large angle has random regions ofincontingency similar to a liquid. The width of the regions does not exceed three atomicdiameters. In later models, amorphous portions of the boundaries were added with crystallineportions. According to Mott and coworkers [1,2], G3 represents portions of good and poorcontingency. In the opinion of Smoluchowski [3,4], even when the boundary angle exceeds158, dislocations combine themselves into groups and form incontingency regions separatedby undistorted areas. If the misorientation angle is greater than 358, then G3 is a solid regionof incontingency. Geisler and Hill [5] and Hargreaves [6] described the grain boundary in 2006 by Taylor & Francis Group, LLC.terms of the model of a transition lattice. According to this model, a certain system in thearrangement of atoms exists in G3. The arrangement corresponds to a minimum energypossible under given conditions.At certain mutual orientations of neighboring grains (special orientations), a superlattice,which is common for both grains, may appear. The superlattice sites will be atoms that arecommon to the crystal lattices of both grains. The boundaries lying in close-packed planes ofsuch superlattices will be most favorable with respect to energy. If the misorientation angle issmall, the coincidence is upset.Coinciding atoms are present in the boundary plane for some discrete values of the grainmisorientation angles. Boundaries that meet the conditions required for the coinciding atomsto appear are called partial contingency (or special) boundaries.Direct experimental studies of G3 are scarce. Microscopy studies and transmission electronmicroscopy have shown that the transition zone occupies two to three interatomic spacings.The zone is saturated with defects like grain boundary dislocations, steps, and microfacets.Particular grain boundary characteristics are closely connected with the way in whichgrain boundaries are formed. A grain structure, and correspondingly G3, can be formed as aresult of crystallization from the liquid state, phase transformations in the solid state, orrecrystallization annealing of a deformed material during deformation.Only conjectures can be made as to the formation of grain boundaries during crystallization. Under real conditions of crystallizations the growth of crystals often exhibits an orientedrather than chaotic character. Correspondingly, the spectrum of boundaries in a cast materialshould differ from the random distribution.In the case of recrystallization in the solid state, i.e., polymorphous transformations ofmetals and alloys, the new phase has certain orientation relationships with the initial phase.Obviously, when transformations within a single grain of the matrix phase are completed, theformed boundaries should have strictly defined and crystallographically determined misorientations rather than random orientations. Many boundaries that appear during apolymorphous transformation are close to special boundaries of coincident sites. Experimental studies into misorientations of the crystals formed during phase transformations inchromiumnickel steel and titanium alloy showed that misorientations at real boundariesagree with theoretical ones (what is meant here is the calculation of crystallographicallydetermined misorientations for the fcchcp and hcpfcc, hcpbcc and bcchcp, and fccbcctransformations). Then the crystallogeometry of the boundaries resulting from polymorphoustransformations is controlled by orientation relationships of the phases formed.In the case of recrystallization processes, the grain structure depends on the stage ofrecrystallization at which annealing was stopped. During primary recrystallization the formation of the structure starts with the appearance of nuclei, that is, dislocation-free portionsof the matrix. They are surrounded by large-angle G3. The proposed models of nucleationassume that nuclei of new grains are formed near the initial G3 owing to a rearrangement ofintergrain lattice dislocations. However, it has been established recently that new grains withlarge-angle boundaries can be formed without participation of intergrain dislocationsbut rather during splitting of initial boundaries. This process can be accounted for inthe following way. After plastic deformation the grain boundaries are in a nonequilibriumstate owing to the trapping of lattice dislocations. During annealing the grain boundarystructure regains the equilibrium state at the expense of splitting of the boundaries. Splittingof the boundaries during recrystallization is caused by lowering of the total energy of thegrain system because high-energy boundaries are replaced by low-energy ones. Here mutualmisorientations depend on misorientation of the nuclei in the deformed matrix.At subsequent states of recrystallization the grains become coarser owing to migration ofthe boundaries. One would expect that the average statistical trend of the process should be 2006 by Taylor & Francis Group, LLC.toward formation of low-energy special boundaries. However, the available experimentaldata are contradictory. This fact suggests that in addition to the tendency to a thermodynamic equilibrium, kinetic factors (different mobility of the boundaries, their pinning byimpurities and precipitates) play an important role in the process of structure formationduring annealing. It was found, for example, that at the stage of collecting recrystallization,random boundaries dominate in iron and molybdenum alloys. However, in ultrapure aluminum the fraction of these boundaries decreases with an increase in the recrystallizationtemperature and time. In contrast, in commercially pure aluminum the fraction of specialboundaries decreases.The state of grain boundaries in a material depends not only on their misorientation butalso on the content of lattice defects. For this reason the boundaries of recrystallization nucleiare not in equilibrium; they are formed in the regions of the deformed matrix with an excessdensity of dislocations of like sign. Rearrangement of the dislocations within the nucleusboundaries is not complete. A nonequilibrium state of the boundaries is also preserved duringtheir migration through the deformed matrix as the matrix absorbs lattice dislocations. Thenthe boundaries are nonequilibrium in ultrafine grain materials formed at the early stage ofrecrystallization. The degree of boundary nonequilibrium decreases at later stages of recrystallization during collecting growth of the grains.Grain boundaries of the deformation origin can be divided into two groups: grainboundaries formed at a low-tempered (<0.30.4Tmelt) deformation and those formed underdeformation at high temperatures.At low temperatures new boundaries are formed at relatively large degrees of deformation. First they represent broken boundaries, which appear nonuniformly in separate grainsof a polycrystal. A continuous network is formed in the areas adjacent to the initial boundaries. It is only under large deformations that a network of these boundaries covers the wholevolume. On the average, two out of three boundaries are large-angle ones. In this case, thefraction of special boundaries is small.When subject to deformation at high temperatures, the formation of grain boundaries is dueto development of recrystallization processes directly during deformation. This phenomenon iscalled a dynamic recrystallization. The grains formed during a dynamic recrystallization arelarge-angle ones. Data on crystallogeometrical parameters of these boundaries are very scarce.From the above discussion it appears that, depending on their origin, grain boundarieshave different structures and therefore possess different properties. The properties of polycrystalline materials are largely determined by the extent of these structural components,which is controlled by the grain size.3.5.2 DETERMINATIONOFGRAIN SIZEThe size of the grain that is formed under a given treatment is determined from microsectionsafter their etching. For carbon and alloyed steels the following reagent is used: 15 mlHNO3 100 ml ethyl or methyl alcohol. Austenitic steel is etched in a copper sulfate chloridesolution containing 10 g copper sulfate, 50 ml hydrochloric acid, and 50 ml water. Whencarbon low-alloy steels are etched, the reagents turn pearlite dark and make visible the ferritegrain boundaries, the martensite structure, and tempering products. The etching rate riseswith the amount of nitric acid. The etching time is from several seconds to a minute. Etchingof austenitic steel reveals the austenite structure and the austenite grain boundaries.Carburization is also used to establish the austenite grain boundaries. In this case, samplesare heated to 9308C (17008F) in a carburizing medium (e.g., a mixture of 40% BaCO3 and60% charcoal), cooled, and etched. 2006 by Taylor & Francis Group, LLC.In addition, an oxidation method is used according to which microsections are heated invacuum to a temperature 20308C (35558F) higher than the quenching temperature and aresoaked for 3 h. Subsequently air is fed to the furnace for 3060 s, and the samples are cooledin water. Before quenching it is recommended to heat samples in a borax melt at 9309508C(170017508F) for 3040 s and then cool them in water. After these treatments microsectionsare polished and etched in a 15% solution of hydrochloric acid in ethyl alcohol. Grainboundaries are seen as the oxide network.Apart from this, use is made of the method of etching austenite grain boundaries, themethod of the network of ferrite (for steels with a carbon content of up to 0.6%) or cementite(for hypereutectoid steels), and the method of the pearlite network for steels that are closer incomposition to eutectoid steels.The grain size is determined by comparing the observed microstructure at a 100 magnification with standard scales (the scales are elaborated so that at a magnification of 100 thegrain number N corresponds to the formula n 8 2n, with n the number of grains per 1 mm2of the microsection area) or by counting the number of grains per unit area of the microsection,or by calculating the mean nominal diameter of the grains or their number per cubic millimeter.The number of grains (at least 50) is counted on the focusing screen of the microscope orfrom a photomicrograph within the area bounded by a circle 79.8 mm in diameter. At 100magnification this value corresponds to a microsection area of 0.5 mm2. The total number ofgrains is calculated from the formula m100 m 0.5m1, where m is the number of grainsinside the circle and m1 is the number of grains intersected by the circle. The number ofgrains per mm2 of the microsection is M 2m100. If a magnification other than 100 power isused, M 2(g=100)2mg (g the magnification power used and mg the number of grains countedat this magnification power). The mean number of grains (Mmean) is calculated using threecharacteristic areas.The mean area (Smean) and diameter (dmean) of the grains are calculated using the formulaSmean 1=Mmean and dmean 1(Mmean )1=2 :The values of equiaxial grains are characterized by the mean nominal diameter, which isdetermined on the focusing screen of the microscope or from a photomicrograph. For thispurpose several arbitrary straight lines are drawn so that every line intersects at least tengrains. The number of intersections on the length of all the lines is counted. Finally, the meandiameter of the grains is calculated.Statistical methods are used, and bar charts are plotted to obtain quantitative characteristics of the structure, particularly grain dimensions. The mean diameter of grains is calculated using the distribution curve.It is possible to calculate the mean area of grains (Smean) from the formula usedto determine dmean if one assumes that the grain is spherical in shape (x pD2=4):Smean k2 (pD2=4) m=m. Then the number of grains (N ) per mm2 is found from theformula N 1=Smean.3.5.3AUSTENITE GRAIN SIZE EFFECTANDGRAIN SIZE CONTROLThe austenite grain boundary structure that is produced on heating above the critical points isimportant because the austenite transformation products formed during cooling (martensite,pearlite, etc.) appear inside austenite crystals. A coarse austenite grain determines a coarseplate structure of martensite during quenching or a coarse cellular network of ferrite (cementite) precipitates at the boundary of the initial austenite grains during annealing or normalization. The pearlite structure is also the coarser, the larger the pearlite grain. 2006 by Taylor & Francis Group, LLC.As is known, a coarse-grain structure of steel (ferritepearlite, martensite, etc.) is characterized by lower mechanical properties. For this reason a fine-grain structure of steel ispreferable in practice. Then the primary task is to produce fine-grain austenite. Since austenite appears during heating of a ferritecarbide mixture, growth centers of the austenite phaseare very numerous, and initially austenite grains are extremely small, on the order of1020 mm. But with an increase in the heating temperature or holding time in the austeniterange, the grains begin to grow intensively.Two types of steels exist: hereditarily coarse-grained steels and hereditarily fine-grainedsteels. This difference is due to the grain growth kinetics with an increase in temperature. Inhereditarily coarse-grained steels a grain gradually and rather uniformly becomes larger as thetemperature is raised above Ac3. In hereditarily fine-grained steels, fine grains are preservedup to about 9508C (17508F). On transition through the coarsening temperature, separategrains start growing intensively and variations in grain size arise. Near 110012008C (200022008F), grains of hereditarily fine-grained steels may be even larger than those of hereditarilycoarse-grained steels.Such differences in the growth of grains in steels are explained by the differences innumber and state of disperse nonmetal inclusions such as, above else, aluminum nitrides,certain carbides, and oxides. These articles retard movement of grain boundaries untiltemperatures are reached at which the particles dissolve in austenite. The barrier effect ofthe particles diminishes nonuniformly, which leads to variations in grain size.A standard test can be used to distinguish between the steel classes. If a noticeable growthof austenite grains is not observed for 8 h after carburization at 9258C (17008F), the steel isassumed to be a hereditarily fine-grained one. Extrapure steels, those produced with aminimum amount of foreign impurities, nitrogen and oxygen, are distinguished by a rapidgrowth of grains above the critical point Ac3.In the case of the usual commercial steels, a grain 2025 mm in size corresponds tostandard heating for quenching, normalization, or annealing. As the temperature is elevatedto 120012508C (220022508F), the grain size reaches 0.1 mm, and in large forgings andwelds, grains of several millimeters in size occur. In ingots and castings, grains can be as largeas several centimeters.If a steel is heavily alloyed with elements that stabilize austenite, the austenite structure isfixed during cooling to or below room temperature and the steel grain is equal to the initialaustenite grain. If austenite passes to pearlite, then, for example, for a hypereutectoid steelone should take into account the size of the pearlite colony, which is characterized by thesame crystallographic orientation of ferrite and cementite plates. A pearlite colony usuallydiffers in size from an austenite grain. Several pearlite colonies are formed in every grain. Soan austenite grain is broken into several grains. This is also true of the ferritepearlitestructure of a hypoeutectoid steel. But in the latter case a network of excess ferrite is formedat grain boundaries. This suggests a connection between a grain of a thermally treated steeland the initial austenite grain.When steel undergoes quenching, a large number of martensite crystals appear in everyaustenite grain. They are connected with the initial austenite grain by certain orientationrelationships. For this reason a correlation is easily seen between the initial austenite grainand grains of the quenched steel. Refinement of the initial grain under heating above Ac3results in refinement of grains in the quenched steel. Then it is possible to correct a coarsegrained structure by heating to the austenite state.However, correction of a coarse-grained austenite or bainite structure may be complicatedby structural inheritance. When crystallographically ordered structures of bainite or martensiteare heated, austenite can also be formed, under certain conditions, in a crystallographicallyordered way. Therefore, under heating above Ac3 the austenite grain is equal in size to a coarse 2006 by Taylor & Francis Group, LLC.grain of steel. In this case, refinement of the crystal structure as a result of phase recrystallizationduring the a ! g transformation does not take place. After the a ! g transformation thestructure of the initial austenite is restored. Both the grain size and its crystallographic orientation are reestablished. However, the restored austenite is structurally unstable. If the temperature or holding time is increased, the austenite structure changes. But the grain is refinedrather than becoming larger as is normally the case. The degree of grain refinement is different,but the structure changes completely above a certain temperature in the range of the stable gphase. The austenite structure is altered within the temperature interval where phase transformations do not occur. Therefore, this phenomenon is attributed to a spontaneous recrystallization of austenite. It is caused by the g ! a ! g transformation hardening.The primary recrystallization of austenite, which is due to the transformation hardening,is followed, with a further increase in temperature, by collective recrystallization. The grainbecomes still coarser. Note once again the unusual character of this two-stage process, whichincludes first a reestablishment of the initial austenite grain and then a refinement of the grainwith temperature.Plastic deformation inhibits the structural inheritance. This is due to the appearance ofglobular austenite in deformed steel that is subjected to either rapid or slow heating. Besides,deformation intensifies the austenite recrystallization in the a state.If a hypoeutectoid steel undergoes sufficiently slow heating, austenite is often formed assame-oriented sections. As the temperature is raised, excess ferrite dissolves in these sections.When ferrite dissolves completely, newly formed grains of austenite fully duplicate the initialaustenite grains. With an increase in the heating rate, sections of austenite with a differentorientation appear. If isothermally heated these sections grow larger and absorb the restoredaustenite and excess ferrite. The greater the number of such sections formed at a faster rate ofheating, the finer the austenite grain.Formation of these sections cannot be explained in terms of the austenite recrystallizationbecause their growth stops as soon as the excess ferrite dissolves completely. They appear in asomewhat overheated and therefore nonequilibrium ferriteaustenite structure. Note ananomalous dependence of the point Ac3 on the heating rate. Increasing the heating rate ofthe steel allows completion of austenitization at lower temperature.3.5.4GRAIN SIZE REFINEMENTTchernov [7] was the first to show that it was possible to refine a coarse-grained structure in1868. Since that time the procedure has been widely used for treatment of steel products. Thegrain refinement, which takes place on heating steels above the temperature Ac3, is related to atransition to the austenite state through nucleation of numerous centers of the austenitephase. Development of these centers leads to formation of a relatively fine-grained structure.Above Ac3, the cross-sectional size of the grain is 1030 mm. Initially the grain size isindependent of the grain of the starting structure; it can be very fine irrespective of whetherthe starting structure of the steel was fine or coarse. A fine-grained structure of the restoredaustenite provides a fine-grained structure of cooled steel whatever structural components areformedpearlite, bainite, or martensite. This is due to the fact that all the transformationproducts nucleate within each separate grain of austenite.Excess phases (ferrite in hypoeutectoid steels and cementite in hypereutectoid steels)precipitate at boundaries of small austenite grains, and the pearlite transformation is accompanied by the appearance of smaller pearlite colonies. Fine austenite grains determine theformation of fine-needle martensite. This underlies the grain refinement effect that is associatedwith hearing above Ac3. Heating the steel above Ac3 during full annealing, normalization,or quenching is followed by recrystallization. Given an initially coarse-grained structure, 2006 by Taylor & Francis Group, LLC.recrystallization results in refinement of grains at a heating temperature corresponding to Ac3.If the heating temperature is much higher than Ac3, the grain is enlarged again, and the expectedcorrection of the structure during the g ! a transformation does not take place. Refinement ofcrystallites is especially pronounced when transformation to the austenite state starts in manycenters inside the initial structure. The formed centers should have a random orientation, whichis not connected with the orientation of the a-phase in the initial structure. Normally suchcenters are sufficiently great in number that the grain size does not exceed 1530 mm.Breaking of an austenite grain into pearlite colonies, each of which can be considered anindependent grain, also represents refinement of steel during pearlite precipitation of austenite.3.6 STRENGTHENING MECHANISM IN STEEL3.6.1 SOLID SOLUTION STRENGTHENINGSolid solution strengthening is a phenomenon that occurs when the number of impurity atomsin the lattice of the basic element is so small that they are incapable of forming both stable andmetastable precipitation phases under any thermal treatment conditions. Nevertheless theimpurity atoms favor improvement of mechanical properties. This can be accounted for bythe following. The presence of impurity atoms in the matrix lattice leads to distortion of thelattice because of the difference in size between the atomic radii of the impurity and the basiccomponent. This in turn leads to the appearance of elastic deformation fields, which retardmovement of dislocations in slip planes under the action of applied stresses. In addition, theimpurity atoms can inhibit movement of dislocations by forming impurity atmospheres aroundthem. Both of the above factors play a leading role in solid solution strengthening.Consider the influence of carbon, which is statistically uniformly distributed in the latticeof the a-iron, on the structure and properties of a-iron. Solubility of carbon in a-iron is muchlower than in the g-iron. It forms interstitial solid solutions with both irons. However,whereas the g-iron lattice has sufficiently large pores for implantation of carbon atoms, thecubic lattice of the a-iron suffers, upon introduction of carbon atoms, a tetragonal distortionsimilar to the one of the martensite lattice, except that in the former case the distortion ismuch smaller. In addition, implantation of carbon atoms causes the entire lattice of the a-ironto expand somewhat. For example, at a carbon content of 0.015% the lattice constantincreases at room temperature by 0.025c.From the above discussion it is evident that carbon affects the properties of the a-phase.Indeed, a dependence of the yield stress on the carbon concentration in the solid a solutionwas detected. The yield stress rises most dramatically with an increase in the carbon concentration from 107 to 104103%. The influence that carbon exerts on plastic deformationresistance of the a-phase is due to both its strong interaction with dislocations and pinning ofthe dislocations and elastic deformations arising as a result of the tetragonal distortion of thea-phase lattice after implantation of carbon atoms.What is more, the presence of carbon in lattices of different structural components formedduring thermal treatment of steel also leads to changes in their mechanical characteristics. Forexample, the location of implanted carbon atoms predominantly in one of the sublattices ofinterstitial sites during the martensite formation brings about additional tetragonal distortions of the martensite crystal lattice. This enhances plastic deformation resistance owing tothe interaction between the stress fields around carbon atoms and those at dislocations.The flow stress grows linearly or in proportion to the square root of the percent carbonwith an increase in the carbon content. This is accompanied by impairment of the plasticcharacteristics of the steel and lowering of the fracture stress. For example, if the carboncontent is raised from 0.25% to 0.4% in a steel with 5% Cr, after quenching and low tempering 2006 by Taylor & Francis Group, LLC.the tensile strength increases from 1600 to 2000 MPa, the fracture stress of samples with apurpose-produced crack decreases from 1300 to1000 MPa, and the impact strength dropsfrom 0.3 to 0.04 J m2.The influence of carbon dissolved in the a-phase on the mechanical properties of steel isalso observed in the case of the ferritepearlite transformation. The factors responsible forthis phenomenon were analyzed above. Both in the homogeneous a-phase and the ferritepearlite mixture, the yield stress rises most sharply when the carbon concentration of ferrite israised from 107 to 104103%.A direct examination of the crystal structure of the a-phase formed over the temperatureinterval of 2503008C (4825728F) during the intermediate (bainite) transformation alsorevealed a tetragonal structure with the c=a axis ratio equal to 1.006 and 1.008 at carboncontents of 1 and 1.2%, respectively. This attests to dissolution of part of the carbon in thea-phase and suggests that the solid solution strengthening of the phase is one of the factorsproviding the high strength properties of intermediate transformation products.3.6.2GRAIN SIZE REFINEMENTIn Section 3.5.4 the possibility of refining steel grains by phase recrystallization under heatingto a temperature above Ac3 was considered. Although austenite passes to other phases duringcooling, its grain size represents an important characteristic of steel. This is due to the factthat all structural components are formed within each separate crystal. The smaller theaustenite grains, the finer the network of excess ferrite at their boundaries and the smallerthe pearlite colonies and martensite crystals. Therefore, a fine grain corresponds to a finecrystal fracture of steel and vice versa at the temperatures where austenite has alreadyprecipitated. Impact strength is especially sensitive to the austenite grain size, and it decreaseswith grain enlargement. A decrease in the dimensions of pearlite colonies inside the initialaustenite grain favors a rise in impact strength also.Although the grain size has a considerable effect on impact strength, its influence is small ifany on the statistical characteristics of mechanical properties such as hardness, fracture stress,yield stress, and specific elongation. Only the actual grain size affects steel properties, theinherited size has no effect. However, the technological process of heat treatment is determinedby the inherited grain. For example, a hereditarily fine-grained steel may be deformed at ahigher temperature with the assurance that the coarse-grained structure will not occur.3.6.3DISPERSION STRENGTHENINGIn the majority of metal alloys, precipitation of supersaturated solid solutions formed duringquenching is followed by precipitation of disperse particles enriched in atoms of the alloyingcomponents. It was found that the strength (hardness) of the alloys increases with theprecipitation of these particles. The increment in the value of these characteristics increasesas the dispersion and volume fraction of the particles increase. This phenomenon has beenreferred to as dispersion strengthening.Precipitation of supersaturated solid solutions occurs during heating (aging) of quenchedalloys. The study of precipitation processes is ultimately aimed at elaboration of the mostefficient methods for strengthening aging materials. In a general case, strengthening resultsfrom an increase in resistance to the movement of dislocations in a crystal when obstacles(barriers) of any type are formed. In aging alloys, dislocations meet regions (ordered atomicclusters [GP zones] or different-structure precipitate particles) that retard their movement.The character of interaction between moving dislocations and precipitates of the secondphase can be different depending on the phase morphology and structure. 2006 by Taylor & Francis Group, LLC.The total effect of aging on the strength properties of alloys is determined by (1) thestrength of the precipitates formed, (2) the volume fraction of precipitates, (3) the degree ofprecipitate dispersion, (4) morphology, structure, and type of binding with the matrix, and (4)test temperature.When a solid solution of carbon in a-iron is cooled below point A1, carbon shouldprecipitate as cementite with lowering of the carbon solubility and a decrease in temperature.This process is realized under sufficiently slow cooling, which is accompanied by diffusionprocesses, leading to the formation of cementite.In the case of abrupt cooling, e.g., water quenching, carbon has no time to precipitate.A supersaturated a solid solution appears. At room temperature the retained amount of carboncan correspond to its maximum solubility of 0.018%. During subsequent storage at roomtemperature (natural aging) carbon tends to precipitate from the solid solution. Carbonenriched regions appear predominantly in defective sections of the matrix. Precipitation ofcarbon from a supersaturated solid solution during natural aging results in improvement of itsstrength characteristics and hardness. However, plastic characteristicsreduction of area,specific elongation, and impact strengthare impaired (see Figure 3.23). A clearly pronouncedyield stress appears after a long natural aging. Hardness may increase by 50% over that of the asquenched state. The phenomenon of dispersion strengthening is observed.As the heating temperature is increased (artificial aging), dispersion strengthening accelerates. At 508C (1228F) the precipitation rate of carbon from the a-iron increases to such anextent that in several hours of aging it reaches the value obtained after several days of naturalaging (Figure 3.24). This is due to intensification of diffusion processes with an increase intemperature. As the temperature is elevated further, precipitation of the supersaturated a solidsolution proceeds still faster.The total process of carbon precipitation from the supersaturated solid solution in a-ironcomprises several consecutive processes. Mechanical characteristics and hardness are notsensitive to structural changes that take place during the aging of alloys. Sharp changes inproperties indicate alterations in the structural state of the steel.2101907017060HBsHB15050d,y %60s280b skG mm1304030y4020201000d48121620Time, daysFIGURE 3.23 Curves showing strengthening of fcc crystals. 2006 by Taylor & Francis Group, LLC.2428HB20019050 C180170160100 C15014020 C130200 C120150 C110 300 C04812(a)20 C16202413(b)Time, h6912 15 18Time, daysFIGURE 3.24 The aging temperature dependence of hardness of carbon steel.A maximum change in mechanical properties during precipitation is achieved only ifexcess crystals in a highly disperse state precipitate. Subsequent coagulation of the crystalsleads to degradation of the properties (Figure 3.24 and Figure 3.25).As the temperature is raised above 1008C (2128F), carbides begin to homogeneouslyprecipitate directly from the solid solution. The precipitating phase has the carbide latticebelow 2008C and the cementite lattice above 3008C (5728F). A transition from one phase tothe other is realized over the temperature interval of 2003008C (3925728F). The onset oftransition from atomic clusters near dislocations to precipitation of the e-carbide remains tobe ascertained. The temperature and time during which the e-carbide crystals precipitate fromthe inhomogeneous solid solution depend on the degree of the solution supersaturation andconcentration of vacancies.Coagulation of the e-carbide crystals lowers the increment in hardness, fracture stress, andyield stress as the effect of breaking the slip dislocations diminishes. Above 2008C (3928F),10040 C2220751825Degree of precipitation5016010065 C2075501825160100ss14130 C207550182516014Time, hFIGURE 3.25 Variation of the yield stress as a result of carbon precipitation from the a-iron lattice atdifferent temperatures. 2006 by Taylor & Francis Group, LLC.where precipitation of the particles is detected even by metallographic methods, hardnessstops increasing.If a naturally aged sample of steel is heated at a temperature of 1002008C (2123928F),a decrease in hardness can be observed. This is due to the phenomenon of recovery wherethe phase nuclei that were formed at room temperature dissolve on heating to highertemperatures.The influence of different solubilities of carbon in a-iron on the properties of the alloy(dispersion strengthening) during low-temperature aging is pronounced in steels with a verylow content of carbon. In steels containing over 0.4% C, the effects considered above areobscured by the influence of cementite particles formed during the pearlite transformation.Besides, nucleation of the precipitating phase can be inhibited owing to migration of carbonto the cementiteferrite interfaces. As a result, the amount of carbon concentrated at latticedefects decreases.Cold plastic deformation greatly accelerates precipitation of a supersaturated solid solution. This is due both to an increase in the density of dislocations, which are preferable sites ofheterogeneous nucleation of precipitates, and to an increase in the concentration of vacancies,which facilitates the diffusion of carbon to clusters. The phenomenon has also been observedin other aging alloys. Mechanical properties change during aging after cold deformation inthe same way as after quenching. That is, the yield stress, the fracture stress, and hardness arealtered. With an increase in aging time, specific elongation and reduction of area decrease andthe tendency to brittle fracture is enhanced. The rate of change is greater than in a quenchedalloy. What is more, the character of the changes is different. Whereas in the case of agingafter quenching, hardness reaches a maximum and then drops, after cold deformationhardness does not decrease with the aging time (Figure 3.26). As the aging temperature israised, the maximum hardness of a quenched alloy lowers, while after cold deformationhardness is independent of the aging temperature. This is explained by the fact that aconsiderable amount of carbon is concentrated near dislocations. Few if any clusters nucleatein the matrix homogeneously. Consequently, clusters cannot grow at the expense of otherclusters, i.e., they cannot coagulate.As the solubility of carbon in g-iron is also susceptible to changes, one can also expect theeffect of dispersion strengthening. However, the g-phase is not fixed during quenching butundergoes martensite transformation. For this reason an additional amount of carbontransferred to the solid solution at the line ES will just enhance the precipitation of martensiteand retained austenite during tempering of steel. Still, an increase in hardness as a result ofcarbide precipitation is observed in purely austenitic steels.Hardness HB9085100 C8040 C7570100 C650.1110Time of aging, h100500FIGURE 3.26 Aging after quenching from 7208C (13288F) (- - - -) and after 10% cold deformation ()of cast steel. 2006 by Taylor & Francis Group, LLC.3.6.4 WORK HARDENING (DISLOCATION STRENGTHENING)An important method used to strengthen steels is deformation strengthening. Strengtheningachieved with crystal deformation can be judged from the shape of stressstrain curves. Theactual shape of these curves largely depends on the crystal lattice type of the metal, its purity,and thermal treatment.In the case of cubic lattice metals, strengthening curves are parabolic, whereas forhexagonal lattice metals a nearly linear dependence is observed between the stress and thestrain. This fact suggests that plastic deformation strengthening is determined mainly bythe interaction of dislocations and is associated with the structural changes that impede themovement of dislocations. Metals with a hexagonal lattice are less prone to deformationstrengthening than cubic lattice metals because the hexagonal lattice has fewer easy slipsystems. In cubic lattice metals, the slip proceeds in several intersecting planes and directions.Examinations of fcc crystals showed that their strengthening curve is complicated(Figure 3.27). Three stages can be distinguished in this curve.Stage I is characterized by easy slip. It depends on the orientation of the crystal relative toexternal forces and on the presence of impurities. This stage is characterized by a lineardependence of strain stresses on the strain at a small work-hardening rate. Dislocations slip inprimary systems.The work-hardening rate is much greater at stage II than at stage I. Dislocations move inintersecting slip planes and, on colliding, form additional obstacles to their movement. Thisstate is most extensive in the stressstrain curve. The ratio between the work-hardening rateand the shear modulus (or any other elastic constant) is almost independent of the appliedstress and temperature. It depends little on the crystal orientation and presence of impurities.For most fcc metals the ratio between the work-hardening rate and the shear modulus isabout 4 103.A cellular structure is formed at stage II. Cells 13 mm in diameter are practically free ofdislocations (Figure 3.28). Groups of like dislocations represent subboundaries of the cells.During their movement, dislocations overcome stress fields of different groups. The formation of obstacles that inhibit propagation of the shear in slip planes and cause a high degree ofstrengthening at stage II leads to a nonuniform distribution of the strain over the crystalvolume.At stage III, changes are possible in the distribution of dislocations. They can either getaround obstacles that retard their movement at stage II or interact with dislocations. As aresult, the work-hardening rate lowers compared to that observed at stage II. At this stage att1t 2 (t 2 > t 1)t3t2t0e2FIGURE 3.27 Curves showing strength of fcc crystals. 2006 by Taylor & Francis Group, LLC.e3eFIGURE 3.28 Electron microscopic image of a cellular structure, 50,000.partial relaxation of stresses may occur owing to the appearance of the secondary slip system.The diminishment of distortion may have the result that deformation continues in theprimary system, which gets rid of a certain number of dislocations passing to the system.A characteristic feature of deformation at stage III is the development of a cross-slip representing the main mechanism by which dislocations bypass the obstacles formed at stage II.3.6.5 THERMAL TREATMENT OF STEELSThere are three basic types of thermal treatment of steels: annealing, quenching, and tempering. AnnealingAnnealing has the following forms: (a) diffusion annealing; (b) softening; (c) phaserecrystallization annealing or full annealing (normalization, high-temperature or coarsegrain annealing, pearlitization); and (d) stress relief annealing and recrystallization annealing. Diffusion AnnealingThe goal of diffusion annealing is to eliminate, insofar as possible, inhomogeneities of thechemical composition, in particular liquation inhomogeneities, which appear during crystallization of alloys. This annealing is usually carried out in the range of the g solid solution at atemperature of 110013008C (201223728F). Diffusion annealing can be used primarily tosmooth out a difference in the content of alloying elements, the difference being due to theintercrystal liquation. This shows up as smearing of dendrites with an increase in temperatureand heating time. Differences in microhardness are eliminated simultaneously. The overallhardness of the alloy decreases because liquation regions possessing high hardness areremoved. Some average hardness is obtained. The success of diffusion annealing largelydepends on the steel purity and liquation. This type of annealing is usually used to improveproperties of medium-purity steels. SofteningSoftening is used to produce the structure of globular pearlite. This structure is very soft andreadily lends itself to deformation during drawing, cold rolling, etc. Steels with a low-carboncontent become too soft after this annealing treatment. The globular pearlite structure isfavorable in steels with a carbon concentration of more than 0.5%. Another goal of softeningis to produce a uniform fine structure with finely dispersed carbon after quenching. 2006 by Taylor & Francis Group, LLC.The simplest method of softening consists in holding for many hours at a temperatureslightly above Ac1. In this case, martensite that is left from the previous treatment is removedand the work hardening caused by deformation (e.g., forging) is eliminated.Carbide plates of pearlite fully coagulate only after a long annealing time. As fine-platepearlite transforms more easily to globular pearlite, it is recommended that normalization (seeSection be carried out before softening treatment. Cooling after softening can bedone in air starting from 6008C (11128F).Refinement of the structure subjected to softening is achieved only above the point A1. Inpractical applications, this type of annealing represents an intermediate treatment, and therefore no strict requirements are imposed on the mechanical properties of annealed materials. Recrystallization Annealing (Normalization, High-Temperature Annealingor Coarse-Grain Annealing, Pearlitization)A twofold g ! a transformation, which takes place during phase recrystallization annealing,leads to the appearance of a fine-grained uniform structure differing completely from theinitial structure. Refinement of the grain during normalization results in the disappearance ofthe Widmannstatten and coarse-grained cast structures, which have poor mechanical properties. Inhomogeneity of the structure in the deformed state is eliminated.The closer the annealing temperature is to Ac3 and the shorter the holding time at thistemperature, the finer the grain. The mechanisms of this phenomenon are analyzed in detail inSection 3.5.3. Refinement of the grain structure is also facilitated if the heating rate to theannealing temperature and the cooling rate from this temperature are increased.In the case of normalization, cooling is done in air. Here it is important to allow fordifferent rates of cooling along the cross section of large-diameter products. The arisingthermal stresses are removed by stress relief annealing or high-temperature tempering. Toobtain a fine-grained structure, rapid cooling is realized only over the transformation temperature interval.The normalization heating temperature should not be much higher than the transformation point; otherwise the grain may be too coarse (overheating). An excessively long holdingtime will have the same result.The optimal heating temperature is determined by the carbon content. For steels with acarbon concentration of up to 0.9%, it is 20308C higher than Ac3. For eutectoid steels,heating between Ac1 and Acm suffices. In the case of low- and medium-carbon steels, the bestresults are obtained if ferrite and plate pearlite are formed during subsequent heat treatment. Stress Relief Annealing and Recrystallization AnnealingDislocation pile-ups and crystal lattice distortions arising in cold-deformed metals may resultin the appearance of macroscopic stresses (stresses of the first kind). Usually these stresses arevery high. Changes in properties that occur under cold deformation can be rectified duringsubsequent heating. The greater the degree of deformation, the lower the heating temperature. Depending on the temperature and time of annealing, various structural changes takeplace in a cold-deformed material. The changes are divided into recovery and recrystallizationprocesses.Recovery is a totality of any spontaneous processes of variation in the density anddistribution of defects before the onset of recrystallization. If recovery proceeds without theformation and migration of subgrain boundaries inside the recrystallized grains, it is calledrestoring. If subgrain boundaries are formed and migrate inside the crystallites, recovery isreferred to as polygonization.Restoring does not include an incubation period. Properties start changing right at thebeginning of annealing. Restoring is accompanied by a redistribution of point defects whose 2006 by Taylor & Francis Group, LLC.concentration decreases subsequently from excess concentration to the equilibrium concentration. Simultaneously, dislocations are redistributed and unlike-sign dislocations are annihilated. The total density of dislocations decreases during restoring. Restoring is realized at atemperature below 0.3Tmelt.The main process that takes place during polygonization is the redistribution of dislocations accompanied by formation of walls. A dislocation wall does not have long-range stressfields, and therefore the wall formation process is energetically favorable. A wall composed oflike-sign dislocations represents a low-angle boundary separating neighboring subgrains witha small misorientation of the lattices. As the annealing time and temperature are increased,the subgrains tend to become coarser. They may be as large as ~10 mm. However, thesubgrains grow within the old-deformed grains.In iron, polygonization starts at 2008C (3928F) (block boundaries appear in the structure).Clearly delineated boundaries of the blocks are retained up to 8508C (15628F) and persisteven after long holding at this temperature.Starting from a certain annealing temperature, the structure changes drastically. Newrather equilibrium grains are observed along with extended deformed grains. They differ fromthe grains of the deformed matrix by having a more perfect internal structure. While thedensity of dislocations in a strongly deformed matrix is 10111012 cm2, after recrystallizationit lowers to 106108 cm2. As distinct from the polygonized structure, recrystallized grainsare separated from the matrix with large-angle boundaries.The formation and growth of grains with a more perfect structure that are surrounded bylarge-angle boundaries at the expense of initially deformed grains of the same phase is calledprimary recrystallization. Recrystallization begins with an incubation period. The recrystallization rate first increases from zero to a maximum and then decreases owing to an everrising number of new grains in contact with one another.Inclusions of insoluble impurities (carbides, nitrides) lower the tendency to growth ofrecrystallized grains. This is especially important in the case of ferritic steels, which are proneto grain growth. Another phase may precipitate during recrystallization in alloys that weresubjected to a strong cold deformation.Sometimes the intensive growth of individual crystals can be observed after a strongdeformation and long holding (for several days) at temperatures close to the melting point.This phenomenon is called secondary or collective recrystallization.The carbon content of steel affects the polygonization and recrystallization kinetics. Withan increase in the carbon content, polygonization slows down or shifts toward higher temperatures. Given a large initial grain size, recrystallization commences the earlier, the greater thedegree of deformation. At a given degree of deformation, higher the recrystallization temperature, the coarser the initial grain. After recrystallization an initially coarse-grained structuregives a larger grain than a fine-grained structure does. In ironcarbon alloys, coarse grains areformed until the appearance of new grains associated with a polymorphous transformation.Under critical conditions of recrystallization the grain size decreases with an increase inthe carbon content. This is due to lowering of the point A3 and narrowing of the recrystallization temperature range. Besides, the number of g-phase crystals formed between Ac1 andAc3 increases. They impede the growth of the a-phase grains at temperatures above Ac1.Carbides also retard growth of the grains.As recrystallization proceeds, strengthening lowers. A fine-grained material possesses animproved long-time strength at lower temperature, while a coarse-grained material exhibitsthis property at higher temperatures. A required size can be obtained by a proper choice of thedeformation and recrystallization conditions. In the case of steels, where no transformationstake place (pure ferritic or austenitic steels) this combination of technological operations is theonly opportunity to influence the grain size. 2006 by Taylor & Francis Group, LLC. (Strengthening Treatment)Quenching refers to cooling from the temperature range of the solid solution at such a ratethat transformations in the primary and bainite ranges are suppressed and martensite isformed. In this state, steels are characterized by the greatest hardness. A distinction ismade between (a) normal quenching, which is used mainly for treatment of medium- andhigh-carbon steels and (b) quenching after a thermochemical treatment (carburization, hightemperature cyaniding), which is used for low-carbon steels. Normal QuenchingTo provide a required cooling rate during quenching, various cooling media and methods areemployed. Water, oil, or air can serve as the cooling medium. Many alloyed steels, which arecharacterized by a high stability of austenite, are subjected to step quenching. With thismethod of quenching, the temperature drop is less than in the case of direct cooling to roomtemperature and consequently quenching stresses are less.A certain amount of austenite is retained during quenching even in steels with a relativelysmall content of carbon. For this reason it is impossible to impart the maximum hardness to aproduct. Since austenite is stable at room temperature and passes to martensite at lowertemperatures, steels undergo a subzero treatment. Under this treatment quenching is continued and steels with a high content of retained austenite are immersed in liquid air orquenching mixtures whose temperature is below room temperature.For surface quenching (if it is necessary to harden only the surface layer to a preset depth),special quenching heating regimes are used. The surface of the product is fully heated, whilethe core is cold and remains unquenched on subsequent rapid cooling. The selection of steelfor surface quenching must be governed by the sensitivity of the metal to quick heating andcooling. For this reason the carbon concentration is limited to 0.7%. Otherwise cracks areformed.Among the main quenching defects are excessive holding and overheating. They show up asenlargement of martensite needles and coarse-grain fracture. This leads to a high brittleness ofquenched products and the formation of cracks. Cracks often form at the boundaries of initialaustenite grains. A low quenching temperature or too short a holding time at the giventemperature causes incomplete quenching. In this case, a quenched metal is insufficiently hard. Thermochemical TreatmentCarburization is associated with surface saturation of steel with carbon and nitrogen. Theseelements quickly dissolve in iron by the interstitial method and are capable of rapid diffusionto a considerable depth. Products made of low-carbon (up to 0.25%) steels are subject tocarburization. Carburization is carried out at 9009508C (165017508F) and sometimes 100010508C (180019008F). Gas carburization is used mostly, under which steel is heated in theatmosphere generated from natural gas (containing predominantly CH4) or from liquidhydrocarbons (kerosene, gasoline, etc.). Carburization is aimed at enrichment of the surfacelayer in carbon. The required strengthening of the surface layer is achieved by quenching,which is performed after carburization. The specific volume of the quenched carburized layeris greater than the specific volume of the core, and therefore considerable compressionstresses arise in the layer. This enhances the fatigue strength of products.Cyaniding is the saturation of the surface of products with carbon and nitrogen in acyanide-containing salt bath. The carbonnitrogen ratio in the diffusion layer is controlled bychanging the mediums composition and the processing temperature. Advantages of cyaniding over carburization consist in a shorter process time and improved wear and corrosionresistance (owing to the presence of nitrogen in the surface layer). 2006 by Taylor & Francis Group, LLC. TemperingThe main purpose of tempering is to provide a disperse structure at a preset degree of cooling.In the case of low-carbon steels, quenching serves as tempering; even if not subjected to hightemperature tempering, the steel has a high viscosity and a relatively great strength.When certain steels are quenched in oil, a structure is formed even during transformation inthe bainite range that is more disperse than the one formed after cooling in air. But the mostdisperse distribution of carbides and the most favorable properties are obtained after martensite tempering. The structure dispersion has the greatest effect on the yield stress. An improvement of the fracture stress and yield stress and an increase in the fracture stressyield stress ratiomay be taken as a measure of the tempering efficiency. The tempering efficiency depends on thecross-sectional area and on the content of carbon and alloying elements in the steels.Although to achieve a thorough quenching the critical quenching rate has to be exceededover the entire cross section, full tempering does not require this procedure. For example, in aquenched steel that has martensite in the surface zone and pearlite in the core, the hardness ofthe core sometimes may be higher than that of the surface zone after tempering. This isespecially the case during a short tempering when precipitation of carbides from martensiteproceeds faster than the coagulation of pearlite plates.Tempering of hypoeutectoid steels, which do not contain free ferrite, provides a uniformimproved structure. In the presence of ferrite precipitates, the fracture stressyield stress ratiodecreases and the impact strength is smaller than in the surface zone. Therefore, in selectingthe content of carbon and alloying elements and particular conditions of austenitization andcooling, the size of the product to be tempered must be considered. For tempering to yieldadequate properties, it often suffices to suppress the formation of ferrite during continuouscooling. Only when a very high fracture stress is required an abrupt cooling is used fortempering. In this case, susceptibility to full tempering can be improved by raising thequenching temperature and thus enlarging the austenitic grain size.REFERENCES1. Mott, Imperfections in Nearly Perfect Crystals, John Wiley & Sons, New York, 1952, p. 173.N.F. Mott and F.R.N. Nabarro, Proc. Phys. Soc. 52:8 (1940).R. Smoluchowski, Physica 15:179 (1949).R. Smoluchowski, Phase Transformations in Solids, John Wiley & Sons, New York, 1952, p. 173.A.H. Geisler and J.K. Hill, Acta Cryst. 11:238 (1948).M.E. Hargreaves, Acta Cryst. 4:301 (1951).D.K. Tchernov, Metals Science, Metallurizdat, Moscow, 1950 (in Russian).FURTHER READINGM.V. Belous, V.T. Cherepin, and M.A. Vasiliev, Transformations During Tempering of Steel, Metallurgiya, Moscow, 1973 (in Russian).M.L. Bernshtein and A.G.M. Richshtadt (Eds.), Physical Metallurgy and Thermal Treatment of SteelsHandbook, 3rd ed., Vols. 13, Metallurgiya, Moscow, 1983 (in Russian).M.E. Blanter, Phase Transformations During Thermal Treatment of Steel, Metallurgiya, Moscow, 1962(in Russian).M.E. Blanter, Physical Metallurgy and Thermal Treatment, Mashinostroyeniye, Moscow, 1963 (inRussian).V.A. Delle, Structural Alloy Steel, Metallurgiya, Moscow, 1959 (in Russian).M.I. Goldshtein, S.V. Grachev, and Yu.G. Veksler, Special Steels, Metallurgiya, Moscow, 1985 (inRussian). 2006 by Taylor & Francis Group, LLC.E. Gudreman, Special Steels, Vols. 1 and 2, Metallurgiya, Moscow, 1959 (in Russian).A.P. Gulyaev, Physical Metallurgy, Metallurgiya, Moscow, 1976 (in Russian).A.P. Gulyaev, Pure Steel, Metallurgiya, Moscow, 1975 (in Russian).H.K. Hardy and T.J. Heal, Prog. Met. Phys. 5:143 (1954).G.A. Kaschenko, Fundamentals of Physical Metallurgy, Metallurgiya, Moscow, 1964 (in Russian).G.V. Kurdyumov, L.M. Utevski, and R.I. Entin, Transformations in Iron and Steel, Nauka, Moscow,1977 (in Russian).V.S. Meskin, Fundamentals of Steel Alloying, Metallurgiya, Moscow, 1964 (in Russian).N.F. Mott, Proc. Phys. Soc. B 64:729 (1951).N.F. Mott, Phil. Mag. 8(1):568 (1956).I.I. Novikov, Theory of Thermal Treatment of Metals, Metallurgiya, Moscow, 1986 (in Russian).A.A. Popov, Phase Transformations in Metal Alloys, Metallurgiya, Moscow, 1963 (in Russian).M.I. Vinograd and G.P. Gromova, Inclusions in Alloy Steels and Alloys, Metallurgiya, Moscow, 1972 (inRussian).R. Zimmerman and K. Gunter, Metallurgy and Materials ScienceHandbook, Metallurgiya, Moscow,1982 (in Russian). 2006 by Taylor & Francis Group, LLC.4Effects of Alloying Elements onthe Heat Treatment of SteelAlexey V. Sverdlin and Arnold R. NessCONTENTS4. of Alloying Elements on Heat Treatment Processingof IronCarbon Alloys............................................................................................... 1664.1.1 g- and a-Phase Regions.................................................................................. 1664.1.2 Eutectoid Composition and Temperature ...................................................... 1694.1.3 Distribution of Alloying Elements ................................................................. 1714.1.4 Alloy Carbides................................................................................................ 172Effect of Alloying Elements on Austenite Transformations ...................................... 1744.2.1 Influence of Alloying on Ferrite and Pearlite Interaction .............................. 1754.2.2 Effect on Martensite Transformation............................................................. 1774.2.3 Retained Austenite ......................................................................................... 1794.2.4 Effect on Bainite Transformation................................................................... 1814.2.5 Transformation Diagrams for Alloy Steels .................................................... 183Hardening Capacity and Hardenability of Alloy Steel .............................................. 1854.3.1 Hardness and Carbon Content....................................................................... 1854.3.2 Microstructure Criterion for Hardening Capacity ......................................... 1874.3.3 Effect of Grain Size and Chemical Composition ........................................... 1894.3.4 Boron Hardening Mechanism ........................................................................ 1934.3.5 Austenitizing Conditions Affecting Hardenability ......................................... 195Tempering of Alloy Steels.......................................................................................... 1964.4.1 Structural Changes on Tempering.................................................................. 1964.4.2 Effect of Alloying Elements............................................................................ 1974.4.3 Transformations of Retained Austenite (Secondary Tempering) ................... 1984.4.4 TimeTemperature Relationships in Tempering ............................................ 1994.4.5 Estimation of Hardness after Tempering ....................................................... 1994.4.6 Effect of Tempering on Mechanical Properties .............................................. 2004.4.7 Embrittlement during Tempering ................................................................... 201Heat Treatment of Special Category Steels................................................................ 2014.5.1 High-Strength Steels ....................................................................................... 2014.5.2 Boron Steels.................................................................................................... 2024.5.3 Ultrahigh-Strength Steels ............................................................................... 2024.5.4 Martensitic Stainless Steels............................................................................. 2044.5.5 Precipitation-Hardening Steels ....................................................................... 2054.5.5.1 Structural Steels................................................................................ 2054.5.5.2 Spring Steels ..................................................................................... 2064.5.5.3 Tool Steels ........................................................................................ 2074.5.5.4 Heat-Resistant Alloys....................................................................... 2084.5.6 Transformation-Induced Plasticity Steels ....................................................... 208 2006 by Taylor & Francis Group, LLC.4.5.7Tool Steels ...................................................................................................... 2094.5.7.1 Carbon Tool Steels ........................................................................... 2094.5.7.2 Alloy Tool Steels .............................................................................. 2094.5.7.3 Die Steels .......................................................................................... 2104.5.7.4 High-Speed Steels ............................................................................. 210Further Reading ................................................................................................................. 2114.1 EFFECTS OF ALLOYING ELEMENTS ON HEAT TREATMENT PROCESSINGOF IRONCARBON ALLOYSA steel that contains, in addition to iron and up to 2% carbon, specially introduced chemicalelements not found in a usual carbon steel is called an alloy steel. Chemical elementspurposely added into steel are termed alloying elements. Steels may contain various numbersof alloying elements, and accordingly they are classified as ternary steels, which have, alongwith Fe and C, one specially introduced alloying element; quaternary steels, which containtwo additional alloying elements, and so on. Alloying elements impart a wide variety ofmicrostructures to steel after heat treatment that gives scope for a wide range of properties.The following elements, arranged in descending order of their application in practice, areusually used for alloying of steel: Cr, Ni, Mn, Si, W, Mo, V, Co, Ti, Al, Cu, Nb, Zr, B, N, and Be.The alloying elements interact with iron, carbon, and other elements in the steel, resulting inchanges in the mechanical, chemical, and physical properties of the steel. Improvement of theproperties of steel in accordance with its designated purpose is the main goal of alloying. The levelto which the properties of steel are changed by alloying depends on the amount of alloyingelements introduced and the character of their interaction with the main elements of the steel, i.e.,Fe and C. That is why an analysis of the influence of alloying elements on the properties of steelshould begin with consideration of the relationship between particular alloying elements, and Feand C. What should be considered is the effect of alloying elements on the critical points of ironand steel, and also the distribution of the alloying elements in the steel.4.1.1g-ANDa-PHASE REGIONSThe position of the critical points A3 and A4 and the location of the eutectoid temperature A1are of great significance because they determine the lowest temperature to which a steelshould be heated for quenching, annealing, or normalization as well as the temperaturesof the maxima in the cure showing the precipitation rate of undercooled austenite. Theprocesses that take place at the critical temperatures in steels are associated with the Feg Fea transformations and dissociation of carbides.Different alloying elements have different effects on the position of the critical points A3and A4. The alloying elements are accordingly divided into two large groups, each in turnbroken down into two subgroups.Addition of the elements from the first group is followed by lowering of the critical pointA3 and a simultaneous rise of the point A4. This effect is shown schematically in Figure 4.1and is most vividly pictured in FeMn and FeNi equilibriums. It is seen that with an increasein the content of the alloying element, the g-phase region broadens considerably, and startingfrom a certain concentration the alloys are found in the state of the g-solid solution until theymelt. This shift of the critical points is brought about by such elements as Ni, Co, Mn, Pt, Pd,Rh, and Ir (Ni group).The other subgroup of the first group includes elements that in general have a limitedsolubility in iron. Given a certain concentration of these elements in iron alloys, chemicalcompounds are formed and eutectic or eutectoid transformations are observed. In other 2006 by Taylor & Francis Group, LLC.FeC systemScheme1,5281,600Temperature, C1,4001,400A41,2001,000906A38006004002000 10 20 30 40 50 60 70 80 90 100Alloying element, %(a)(b)Ni, %FIGURE 4.1 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with extendedg-phase range and unlimited solubility.words, heterogeneous regions appear in diagrams of the iron-alloying elements system. Theheterogeneous regions limit the g-phase occurrence range. This type of phase diagram ofalloys is exemplified in Figure 4.2. As is seen, with an increase in the concentration of thealloying element in the alloy, the critical point A3 lowers and A4 rises. As a result, the rangeof g-solid solutions widens. But then, owing to the formation of heterogeneous regions, theg-phase narrows and, finally, vanishes. Equilibrium diagrams of this type (exhibiting first awide range of the g-phase and then a narrowing of the phase caused by the appearance ofheterogeneous regions) are found for N, C, Cu, Zn, Au, Re, etc.As distinct from the elements of the first group, elements entering the second groupelevate the point A3 and lower the point A4 as their content in the alloy is raised. This leadsinitially to narrowing and then to a complete closing of the region of the g-solid solution asshown schematically in Figure 4.3. This shift of the critical points of alloys is induced by suchelements as Cr, Mo, W, Si, T, Al, and Be (Cr group). These elements can be placed in the firstsubgroup of the second group of alloying elements. The second subgroup includes elementsFeC systemSchemeTemperature, C152914001600150014001300120011001000900800700600A4906A3A10123Carbon, %010 20 30 40Alloying element, %(a)(b)Fe3C, %FIGURE 4.2 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with extendedg-phase range and limited solubility. 2006 by Taylor & Francis Group, LLC.Temperature, 8CScheme15291400 A4FeCr system18001600140012001000800600400200906A3Alloying element, %(a)(b)0 10 20 3040 50 60 70 80 90 100Cr, %FIGURE 4.3 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with closedg-phase range.whose introduction causes the appearance of other phases in the equilibrium diagrams beforethe g-phase range is closed. It follows from Figure 4.4 that in this case the narrow range of theg-phase is limited by adjacent heterogeneous regions. Equilibrium diagrams of this type (witha narrow range of the g-phase and its limitation by an adjacent heterogeneous region) are dueto Zr, Ta, Nb, Ce, etc.The above-described division of alloying elements into two large groups can be applied toternary and more complex systems. The first basic ternary diagram is obtained when iron isalloyed with two elements, each leading to broadening of the g-phase range in binary ironalloys. Such alloys can be exemplified by FeCoNi, FeCoMn, and FeNiMn.The second basic diagram covers iron alloys with two elements, which close the g-phaserange. An example of these alloys is the FeCrMo system, but it includes, along with regionsof solid solutions, intermetallic compounds that are formed at high concentrations.The third basic type of equilibrium diagram applies to a ternary system where one of theelements widens the g-phase range and the other element closes it. An example is the FeCrNisystem, which is important in technical terms. Thus even ternary systems may include purelyferritic (a-phase) and purely austenitic (g-phase) alloys as well as alloys possessing a multiphase structure.SchemeFeTa system18001600Temperature C15291400140012001000906800(a)Alloying element, %(b)600010203040Ta, %FIGURE 4.4 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with narrow g-phaserange limited by adjacent heterogeneous region. 2006 by Taylor & Francis Group, LLC.4.1.2 EUTECTOID COMPOSITIONTEMPERATUREANDThe aforementioned division of alloying elements into groups according to their influence onallotropic transformations in alloys of the iron-alloying element system allows one to predictto some extent the effect of the elements on the critical points of carbon steel. For example,considering the diagram lines that correspond to the transition of Fe from one allotropic formto another, it can be expected that the elements extending the g-phase range (Ni group) willlower the a ! g iron transition point Ac3, while the elements narrowing the g-phase range (Crgroup) will elevate that point.A similar effect of the elements is observed, to a certain extent, in the pearlite transformation Ac1 as in this case, too, an allotropic change of iron takes place: Fea transforms to Feg.Figure 4.5 illustrates the influence of the most important alloying elements on the position ofthe critical point Ac1. As is seen, the elements narrowing the g-phase range do raise the criticalpoint Ac1, while the elements broadening the g-phase range lower it.It should be noted that in the case of Cr group elements one observes a known relationshipbetween the limiting concentration necessary to close the g-phase range in iron-alloying elementalloys and the degree of elevation of the point Ac1. The lower the concentration of the element atwhich the g-phase range is closed, the more abrupt the rise of the critical point Ac1.If a steel simultaneously contains two or more alloying elements that influence its criticalpoints in the same direction, the critical points usually lower or elevate to a greater extentthan would be the case if only one of the elements exerted its influence. But here the resultcannot be presented by a simple dependence. If a carbon steel contains alloying elements thathave an opposite effect on the position of the critical points during heating, the influence ofthe elements shows up differently depending on their quantitative ratio. Table 4.1 givesvalues of the critical points during heating and cooling for some multialloy steels. As isseen, rather high Ac1 and Ac3 points are characteristic of chromiumsilicon and chromiumsiliconmolybdenum steels of the heat-resistant type, high-chromium steels of the stainlesstype, chromiummolybdenumvanadium steels, and others. The data of Table 4.1 are alsointeresting in that they show a simultaneous effect of the most significant alloying elementson the critical points Ar3 and Ar1 under constant-rate cooling. In particular, a very sharplowering of these points in multialloy steels is caused by molybdenum. Molybdenumis responsible for a drastic drop of the critical points under cooling in steels thatcontain chromium and nickel at the same time. This last fact is especially important forstructural steel.Eutectoid temperature, 8C1300Ti1200MoSi11001000WCr900800700600Ni50002Mn4 6 8 10 12 14 16 18Alloying elements, %FIGURE 4.5 Effect of alloying elements on the eutectoid transformation temperature Ac1. 2006 by Taylor & Francis Group, LLC.170 2006 by Taylor & Francis Group, LLC.TABLE 4.1Position of Critical Points during Heating and Cooling of Some Multialloy SteelsCritical Pointsb (8C)Chemical Compositiona (%)HeatingCoolingCMnSiCrNiMoVWAc1Ac3Ar3Ar11234567891011121314151617c0. content of Mn and Si is standard if not specified otherwise. Residual Cr and Ni are also present in all steels.The critical points were determined by the dilatometric method; the cooling rate ~28C=s (~48F=s).cNo. 17 also has 0.25% Al.bSteel Heat Treatment: Metallurgy and TechnologiesNo.Eutectoid carboncontent, %0.800.600.400.200CrMo SiWMnTi2Ni4681012141618Alloying elements, %FIGURE 4.6 Effect of alloying elements on the concentration of carbon in eutectoid.The effect of alloying elements manifests itself in a shift of the critical points with respectnot only to temperature but also concentration. Figure 4.6 illustrates how the content ofalloying elements in steel affects the carbon concentration at the eutectoid point. As can beseen from the figure, all the alloying elements shift the eutectoid point to the left, i.e., towardlowering of the carbon concentration, and consequently decrease the carbon content of alloypearlite. In analogy to the shift of the eutectoid point to the left, the addition of most alloyingelements in steel is followed by a leftward displacement of the point E in the FeC equilibriumdiagram, which determines the solubility limit of carbon in austenite. The point E is shiftedmost by Cr, Si, W, Mo, V, and Ti, which are arranged here in ascending order oftheir influence. All these elements narrow the g-phase range in alloys of the iron-alloyingelement system.If a carbon steel contains a certain amount of an alloying elements, point E is displaced tothe left to such an extent that even at a carbon concentration of several tenths of a percent thesteel structure may have ledeburite, which is present in pure ironcarbon steels only when thecarbon content is over 1.7%.It is of interest to note that the more strongly an element shifts the points E and S, thelower the element concentration at which it closes the g-phase range in the iron-alloyingelement diagram. Therefore a leftward shift of the points E and S can be considered as thetendency of a specific alloying element to narrow the g-phase (austenite) range.Therefore the introduction of alloying elements into a carbon steel is accompanied by ashift of the equilibrium critical points with respect to both temperature and carbon concentration. The greater the shift, the larger the amount of the elements introduced.4.1.3 DISTRIBUTIONOFALLOYING ELEMENTSIn commercial alloy steels, which are multicomponent systems, alloying elements can befound (1) in the free state; (2) as intermetallic compounds with iron or with each other;(3) as oxides, sulfides, and other nonmetal inclusions; (4) in the carbide phase as a solution incementite or in the form of independent compounds with carbon (special carbides); or (5) as asolution in iron.As to the character of their distribution in steel, alloying elements may be divided into twogroups:1. Elements that do not form carbides in steel, such as Ni, Si, Co, Al, Cu, and N2. Elements that form stable carbides in steel, such as Cr, Mn, Mo, W, V, Ti, Zr, and NbThe law determining the manner in which elements of the first group are distributed in steelis very simple. These elements do not form chemical compounds with iron and carbon,and consequently the only possible form in which they can be present in steel is in solid 2006 by Taylor & Francis Group, with iron. The only exceptions are Cu and N. Copper dissolves in a-iron at normaltemperatures in amounts of up to 1.0%. If the Cu content exceeds 7%, iron will containcopper in the free state as metal inclusions. Similar behavior is typical of the alloying elementsthat do not dissolve in solid iron at all (e.g., Pb or Ag). Nitrogen also has a limited solubilityin ferrite. When the N content is higher than 0.015%, nitrogen is found in steel in the form ofchemical compounds with iron or some alloying elements (V, Al, Ti, Cr). These chemicalcompounds are called nitrides.Most alloying elements can form intermetallic compounds with iron and with each other.But these compounds are formed only at concentrations of the alloying elements, which arenot used in the usual commercial steels. Therefore it can be assumed that the commonquantity-produced steels do not have intermetallic compounds of alloying elements withiron or with each other. Intermetallic compounds are formed in high-alloy steels, a fact thatis of great significance for these steels.Alloying elements, whose affinity for oxygen is greater than that of iron, are capable offorming oxides and other nonmetal compounds. When added at the very end of the steelmelting process, such elements (e.g., Al, Si, V, Ti) deoxidize steel by taking oxygen from iron.The deoxidizing reaction yields Al2O3, TiO2, V2O5, and other oxides. Owing to the fact thatalloying elements that are deoxidizers are introduced at the final stages of the steel meltingprocess, the majority of oxides have no time to coagulate or to pass to slag, and as a resultthey are retained in the solid steel as fine nonmetal inclusions. In addition to a great affinityfor oxygen, some alloying elements have a greater affinity for sulfur than iron does, and uponintroduction into steel, they form sulfides.Compared to noncarbide-forming elements, alloying elements that form stable carbidesin steel exhibit a much more complicated distribution. They can be found in the form ofchemical compounds with carbon and iron or be present in the solid solution. The distribution of these elements depends on the carbon content of steel and the concurrent presence ofother carbide-forming elements. If a steel contains a relatively small amount of carbon and agreat quantity of an alloying element, then obviously carbon will be bound to carbides beforethe carbide-forming elements are completely used. For this reason excess carbide-formingelements will be found in the solid solution. If a steel has a large amount of carbon and littleof the alloying elements, the latter will be present in steel mainly as carbides. Carbideformation is treated in detail in the next section.Note in conclusion that most alloying elements, except C, N, O, B, and metalloidsstanding far from iron in the periodic table, dissolve in great amount in iron. The elementsstanding to the left of iron in the periodic table are distributed between iron (base) andcarbides; those to the right of iron (Co, Ni, Cu, etc.) form solutions with iron only and do notenter into carbides. Thus one can state that alloying elements dissolve predominantly intobasic phases (ferrite, austenite, cementite) of ironcarbon alloys or form special carbides.4.1.4ALLOY CARBIDESCarbides are formed in steels only by iron and metals that stand to the left of iron in theperiodic table: Mn, Cr, W, V, Zr, Nb, Ti. Here the elements are arranged in accordance withtheir affinity for carbon. The elements at the left end of the row form relatively unstablecarbides that dissociate readily on heating. In contrast, the elements at the right end of therow form extremely stable carbides that dissociate at temperatures much higher than thecritical points of steel.Similar to iron, the above-mentioned carbide-forming elements refer to the elements oftransition groups but possess a less perfect d-electron band. The further left a carbide-formingelement stands in the periodic table, the less perfect is its d-band. 2006 by Taylor & Francis Group, LLC.There is reason to believe that during carbide formation carbon donates its valence electronsto fill the d-electron band of the metal atom, while valence electrons of the metal form a metalbond, which determines the metallic properties of carbides. At the same time numerous experiments show that the more to the left an element stands in the periodic table, the less perfect is its delectron band and the more stable is the carbide. From these facts, it is possible to formulate thegeneral principles of carbide formation in steels: only metals whose d-electron band is filled lessthan that of iron are carbide-forming elements. Their activity as carbide-forming elements isgreater and the stability of the carbide phases formed is the higher, the less perfect is the d-band ofthe metal atom. This principle allows specifying conditions of carbide formation in steels in thepresence of several carbide-forming elements, the sequence of dissolution of various carbides inaustenite, and other factors that are important for the theory of alloying, manufacturing practice,and application of alloy steels. The formation activity and stability of carbides in alloy steels willincrease in going from Mn and Cr to Mo, W, V, Ti, and other elements whose d-bands are lessperfect than those of Mn and Cr. This means that if a steel contains, e.g., both chromium andvanadium, one should expect vanadium carbides to form first (under equilibrium conditions).If the atomic radius of carbon is taken equal to 0.079 nm, it is easy to calculate that for allcarbide-forming elements except Fe, Mn, and Cr, the ratio of atomic radii of carbon andmetal is less than 0.59. It is known that if the ratio of atomic radii of a transition group metaland a metalloid with a small atomic radius (C, N, H) is less than 0.59, special types ofcompounds called interstitial phases can be formed.The carbon=metal ratio of most carbide-forming alloying elements is lower than 0.59, andtherefore the elements and carbon are capable of forming interstitial phases. It was found thatthe following carbide compounds may be formed in steels:Carbides of Group IFe3CMn3CCr23C6, Cr7C3Fe3Mo3CFe3W3CCarbides of Group IIMo2CW2C, WCVCTiCNbCTaC, Ta2CZrCHowever, the above carbides are not found in steels in pure form. Carbides of all alloying elementscontain iron in solution, and if other alloying elements are present, they include these elementstoo. Thus, in a chromiummanganese steel the carbide (Cr, Mn, Fe)23C6, which contains iron andmanganese in the solution, is formed instead of the pure chromium carbide Cr23C6.Owing to the fact that carbides with the same chemical formula mutually dissolve, in thepresence of titanium and niobium, for example, rather than two kinds of carbides forming, a singlecarbide will be formed that includes titanium and niobium on equal terms. For this reason possiblevariants of carbide formation are fewer, and actually we have only six kinds of carbides in steels:Carbides of Group IM3CM23C6M7C3M6C 2006 by Taylor & Francis Group, LLC.Carbides of Group IIMCM2Cwhere M denotes a sum of carbide-forming (metal) elements.The carbides placed in group I possess a complicated crystal structure; an example iscementite. A specific structural feature of the carbides of group II as interstitial phases is asimple crystal lattice. They usually crystallize with a great carbon deficiency.It is worth noting that interstitial phases dissolve poorly in austenite and may not passinto solid solution even at high temperatures. This distinguishes interstitial phases from thecarbides of group I, which readily dissolve in austenite on heating. All carbide phases have ahigh melting temperature and high hardness. Interstitial phases surpass the carbides of groupI in this respect.4.2 EFFECT OF ALLOYING ELEMENTS ON AUSTENITE TRANSFORMATIONSThe overwhelming majority of alloy steels are used after heating to the austenite stage,quenching, and subsequent annealing. During quenching and annealing, austenite transformswith three types of transformations possible: pearlite transformation (often called diffusivetransformation, precipitation to the ferritecarbide mixture, or stage I transformation),intermediate transformation (bainite or stage II transformation), and martensite transformation (stage III transformation). The precipitation stability of undercooled austenite is characterized by the diagrams of isothermal and thermokinetic austenite transformation.The isothermal diagrams characterize the precipitation kinetics of austenite at constanttemperature of undercooling. Such diagrams are useful for comparative evaluation of different steels and also for clarifying the influence of alloying and other factors (heating temperature, grain size, plastic deformation, and so on) on the precipitation kinetics of undercooledaustenite.Thermokinetic diagrams characterize the precipitation kinetics of austenite under continuous cooling. These diagrams are less illustrative but have great practical importance,because when subjected to thermal treatment, austenite precipitates under continuoustemperature variation rather than under isothermal conditions. Under continuous cooling,transformations occur at a lower temperature and take a longer time than in the isothermalcase. Alloying elements have considerable influence on the kinetics and mechanism of allthree types of transformations of undercooled austenite: pearlite, bainite, and martensitetransformations. However, these elements influence austenite precipitation in differentways.Alloying elements that dissolve only in ferrite and cementite without the formation ofspecial carbides exert just a quantitative effect on the transformation processes (Figure 4.7).Cobalt speeds up a transformation but the majority of elements, including Ni, Si, Cu, Al, etc.,slow it down.Carbide-forming elements produce both quantitative and qualitative changes in thekinetics of isothermal transformations. Thus, the alloying elements forming group I carbidesinfluence the austenite precipitation rate differently at different temperatures:At 13009308F (7005008C) (pearlite formation), they slow the transformation.At 9307508F (5004008C), they dramatically slow the transformation.At 7505708F (4003008C) (bainite formation), they speed up the transformation.Thus, steels alloyed with carbide-forming elements (Cr, Mo, Mn, W, V, etc.) have twomaxima of the austenite isothermal precipitation rate separated by a region of relative 2006 by Taylor & Francis Group, LLC.A1TemperatureA1Cr, W, V, MosteelNi, Mn, SisteelCarbonsteelCarbonsteelMMTime(a)(b)TimeFIGURE 4.7 Diagrams of isothermal austenite precipitation. (a) Carbon steel and steel alloyed withnoncarbide-forming elements; (b) carbon steel and steel alloyed with carbide-forming elements.stability of undercooled austenite (Figure 4.7). The isothermal precipitation of austenite hastwo clearly defined intervals of transformation: (1) to a lamellar structure (pearlite transformation) and (2) to a needle-like structure (bainite transformation). At temperatures lowerthan those indicated above and given greater degrees of undercooling as in the g ! atransformation start temperature, the martensite transformation develops in alloy steels. Asa result, a supersaturated a-iron-based solid solution is formed.The remainder of this section considers in more detail the influence of alloying elements onthe mechanism and kinetics of austenite precipitation for all three types of transformations.4.2.1 INFLUENCEOFALLOYINGONFERRITE AND PEARLITE INTERACTIONThe most important practical feature of alloying elements is their capacity to decrease theaustenite precipitation rate in the region of the pearlite transformation, which shows up as arightward shift of the line in the isothermal austenite precipitation diagram. This favors adeeper hardening and undercooling of austenite up to the range of martensite transformationunder slow cooling such as air cooling.In alloy steels, the pearlite transformation consists of a polymorphous g ! a transformation and diffusion redistribution of carbon and alloying elements. As a result, specialcarbides and a ferritecement mixture (pearlite) are formed. Particular alloying elementsand their amounts in the initial g-solid solution determine the rates of the individual stepsof pearlite transformation and consequently its kinetics as a whole.The polymorphous g ! a transformation in iron under small undercooling of austenite(near the temperature of stage I) proceeds by means of disordered displacement of atoms, asdistinct from the martensite transformation (under greater undercooling), which proceedsthrough ordered shear. As mentioned above, all alloying elements dissolved in austenite,except Co, slow the pearlite transformation and shift the top section of the isothermalaustenite precipitation curve to the right.The nature of the increase in stability of undercooled austenite under the influence ofalloying elements is rather complicated. Whereas in carbon steels the pearlite transformationis associated with the g ! a rearrangement of the lattice and diffusion redistribution of carbon, 2006 by Taylor & Francis Group, alloy steels these processes can be supplemented with the formation of special carbides anddiffusion redistribution of alloying elements dissolved differently in ferrite and carbide.Not only do austenite-dissolving elements have small diffusion coefficients of their ownwhich are sometimes several orders of magnitude smaller than that of carbon, but also someof them (e.g., Mo, W) slow the diffusion of carbon in the g lattice. Besides, some of theelements (e.g., Cr, Ni) retard the g ! a rearrangement, which is part of the pearlitetransformation. Depending on the steel composition and degree of undercooling, the decisiverole may belong to one of the above-mentioned factors.The formation of carbides during pearlite transformation in steel results from a redistribution of carbon and alloying elements between the phases that are formed: ferrite andcarbides. In the presence of dissolved strong carbide-forming elements (Nb, V, Cr, etc.),special carbides are formed in undercooled austenite before the g ! a transformation begins,in excess ferrite (in eutectoid and hypereutectoid steels this stage is absent) and in eutectoidferrite (pearlite). Every stage yields special carbides whose type depends on the austenitecomposition.If a steel contains carbide-forming elements (V, Nb, Ti, Zr) that pass into solid solutionduring austenitization, then a carbide of one type, MeC (VC, NbC, TiC, ZrC), is formed at allthe stages.The scheme of austenite precipitation and the carbide formation process during pearlitetransformation in steels with V, Nb, Ti, and Zr are as follows:g ! MeCA g0ase geMeCf aee Fe3 C ase ! MeCe aeeHere g is the initial undercooled austenite; MeCA is the carbide precipitated in austenite; g0 isaustenite after carbide precipitation; ase is excess ferrite supersaturated with a carbide-forming element and carbon; aee is equilibrium excess ferrite; MeCf is the carbide precipitated inexcess ferrite; ge is austenite of eutectoid composition; Fe3C is the eutectoid cementite(pearlite); ase is the supersaturated eutectoid ferrite (pearlite); aee is the equilibrium eutectoidferrite; MeCe is the carbide precipitated in the eutectoid ferrite.The formation of carbide (MeCA) in undercooled austenite before the g ! a transformation starts is due to the fact that solubility of the carbide-forming element and carbon inaustenite decreases with decreasing temperature. As is seen from the scheme, after thepolymorphous g ! a transformation the ferrite (both excess and eutectoid) is first supersaturated with the carbide-forming element and carbon, then a carbide is formed from ferrite,and subsequently the state of ferrite approximates the equilibrium condition. This processlasts for a few seconds.In steels containing other carbide-forming alloying elements (Cr, Mo, W), the carbideformation process is much more complicated. Depending on their content in austenite, theseelements can form several types of carbides: alloy cementite (Fe, Cr)3C and special carbides(Fe, Cr)7C3 and (Fe, Cr)23C6 in Cr steels and carbides (Fe, Mo)23C6, MoC, Mo2C, and (Fe,Mo)6C in steels with Mo (W forms analogous carbides).Noncarbide-forming elements (Ni, Co, Si, etc.) do not participate directly in carbideformation. As a rule their amount in cementite equals their average concentration in steel.These elements can indirectly influence the thermodynamic activity of other elements, i.e., theprocess of their redistribution during carbide formation. As mentioned above, the process of 2006 by Taylor & Francis Group, LLC.carbide formation is limited by the mobility of the carbide-forming elements. With a decrease intemperature their diffusion mobility diminishes and special carbides are not formed below 4005008C (7509308F). At lower temperatures the intermediate (bainite) transformation takesplace, and at higher undercooling rates martensite (diffusionless) transformation occurs.4.2.2 EFFECTONMARTENSITE TRANSFORMATIONAs in carbon steels, the martensite transformation in alloy steels takes place under rapidcooling from temperatures higher than the equilibrium temperature of the g ! a transformation (A1). At the martensite transformation temperature both the diffusion movement ofmetal atoms of iron and alloying elements and that of metalloid atoms of carbon and nitrogenare suppressed. For this reason the martensite transformation in steels proceeds by a diffusionless mechanism.The martensite transformation can take place in carbon-containing alloy steels, noncarbon-containing alloy steels, and binary iron-alloying element alloys. The martensite transformation usually leads to formation of a supersaturated a-iron-based solid solution. Incarbon-containing steels, the solid solution is supersaturated mainly with carbon, and innoncarbon-containing alloy steels, with alloying elements. The content of carbon and alloyingelements in martensite is the same as that in the initial austenite.The transformation of austenite into martensite during cooling starts at a certain temperature called Ms. This temperature is independent of the cooling rate over a very wide rangeof cooling rates.The martensite transformation kinetics of most carbon and structural and tool alloy steels isathermal in character. The athermal martensite transformation is characterized by a smoothincrease in the amount of martensite as the temperature is lowered continuously in the martensiteinterval MsMf, where Mf is the martensite finish temperature. As a rule, this transformationtakes place in steels with the martensite point Ms higher than room temperature.A version of athermal martensite transformation is explosive martensite transformation,where a certain quantity of martensite is formed instantly at or a little below the temperature Ms.This transformation is observed in alloys with the martensite point below room temperature.The position of the martensite point also determines the microstructure and substructureof the martensitic-quenched steel. At temperatures Ms below room temperature, lamellar(plate) martensite is formed in quenched ironcarbon and alloy steels. Crystals of thismartensite are shaped as fine lenticular plates. In steels with the martensite point Ms higherthan room temperature, lath martensite is formed during quenching. Crystals of this martensite have the form of approximately equally oriented thin plates, which are combined intomore or less equiaxial packets. The substructures of needle and lath martensite are qualitatively different.From what has been said above, it might be assumed that the martensite transformationkinetics, the morphological type of martensite, the substructure of martensitic-quenchedsteels, and other phenomena are connected to a great extent with the martensite-starttemperature Ms. Thus the influence of the elements on martensite transformation is determined primarily by their influence on the position of the martensite point Ms. Of practicalimportance is also the martensite finish temperature Mf.Experiments concerned with the influence of alloying elements on the position of themartensite point show that Co and Al elevate the martensite start temperature, Si has little ifany effect, and all the other elements decrease Ms (Figure 4.8). 2006 by Taylor & Francis Group, LLC.700Temperature, C600500400300Ms2001000Mf10020000.4(a) content, %AI3000.76 % CCoTemperature M s, 8C2500.9% C200Si150Mn1% C100CuNiVCr50Mn00(b)123456Amount of alloying element, %FIGURE 4.8 The influence of the content of (a) carbon and (b) alloying elements at 1% C on themartensite point position.The quantitative influence of alloying elements is approximately as follows (per 1 wt% ofthe alloying element):ElementMnCrNiVSiMoCuCoAlShift of point Ms (8C)4581356326473054002545713122218 (8C)32 (8C)These data are given for carbon steels containing 0.91.0% C. For a wider range of C content,the quantitative influence of the elements can be different. In particular, it was establishedthat the smaller the C content, the weaker is the influence of the alloying elements on theposition of point Ms. The martensite start temperature of medium-carbon alloy steels can beestimated using empirical formulaMH (8C) 520 320(% C) 50(% Mn) 30(% Cr) 20[%(Ni Mo)] 5[%(Cu Si)],where % C, % Mn, etc. are the contents of the corresponding elements in weight percent. 2006 by Taylor & Francis Group, LLC.The results of calculations by this formula for steels containing 0.20.8% C are in goodagreement with the experimental data. However, for multialloy steels this formula does notalways yield reliable data because if a steel contains several alloying elements it is impossibleto determine their combined effect on the martensite point by simple summation. Thus, forexample, Mn lowers the point Ms to a greater extent than Ni, but in a steel with a high Crcontent its effect is weaker than that of Ni.The martensite point Ms is affected mostly by C dissolved in austenite (Figure 4.8). Thetransformation finish temperature Mf intensively decreases too, as the C content is increasedup to 1% [to 1008C (2128F)] and remains constant at higher amounts of carbon.The reason carbon and alloying elements influence the position of the martensite point ismainly a change in the relative thermodynamic stability of g- and a-phases of iron, becausethe martensite transformation itself is a g ! a transformation.4.2.3 RETAINED AUSTENITEA characteristic feature of the martensite transformation in steels, whatever its character(athermal, explosive, or fully isothermal), is that transformation of austenite to martensite isnever complete. Figure 4.9 shows the amount of martensite formed when the temperature isdecreased continuously in the martensite range MsMf (martensite curve) for the athermaltype of martensite transformation. The transformation starts at the point Ms, and the amountof martensite increases with decrease in temperature. The end of the transformation corresponds to the temperature Mf. At this temperature a certain amount of austenite is still left(retained austenite, A, %). Cooling below Ms does not lead to further transformation or lowerthe amount of retained austenite.Investigations show that martensite curves of different steels, both carbon and steels andsteels alloyed with different elements and in different amounts, exhibit approximately thesame behavior. Then, if the martensite finish temperature is below room temperature,the amount of retained austenite should, in a general case, be higher, the lower the martensite point Ms. Strictly speaking, the amount of retained austenite depends on the martensitetemperature range, i.e., on the MsMf temperature difference; it increases as the rangenarrows. But the martensite range itself depends on the position of the martensite point Ms:the range narrows as the point Ms lowers.100A, %Martensite content, %9080504030200 2006 by Taylor & Francis Group, LLC.Mf6010FIGURE 4.9 Martensite curve.A, %70Ms20Temperature, C100100MnCrHeating11508C90Retained austenite, %805070gtinC12CraHe6050Ni40Cu30W20VCo10AI021345678Alloying elements, %FIGURE 4.10 The influence of alloying elements on the amount of retained austenite in quenched steel(1% C).Thus the influence of alloying elements on the amount of retained austenite formed during thequenching of steels should qualitatively and, to a great extent, quantitatively correspond to theirinfluence on the position of the martensite start point Ms. Available experimental data show thatalloying elements that lower and raise the martensite start point increase and decrease the amountof retained austenite, respectively (Figure 4.10). Besides, a certain sequence in the arrangement ofthe elements is observed from the point of view of their quantitative influence. In particular, thelargest amount of retained austenite in accordance with their influence on the position of themartensite point is due to Mn, Cr, Ni, etc. As illustrated in Figure 4.11, the influence of theseelements on the martensite range follows the same sequence.Martensite range, 8C4000.6% C300NiCrNi200Mn1.0% CCr100Mn0246Alloying element content, %8FIGURE 4.11 The influence of alloying elements on the martensite range. 2006 by Taylor & Francis Group, LLC.In alloy steels, the martensite point Ms lowers most and the martensite range narrowsunder the influence of carbon. Therefore the influence of carbon on the amount of retainedaustenite is much stronger than that of alloying elements. An increase in the C content ofchromiumnickel steel from 0.4 to 0.6% increases the amount of retained austenite to ~8.5%after quenching; an increase in the Ni content of the same steel from 1 to 4% brings theamount of retained austenite to ~6% only. The fact that carbon promotes the greatestretention of austenite during quenching is especially unfavorable for low-alloy tool steels.In multiple alloy steels a given element favors the formation of a greater amount ofretained austenite than the law of summation suggests. However, in multiple alloy steelstoo, the relationship between lowering of the martensite point under the influence of a givenelement and an increase in the amount of retained austenite caused by the same elementpersists in the main.In addition to the content of carbon and alloying elements, other factors can influence theamount of retained austenite formed during quenching of steel. The most important of theseis the rate of cooling below the martensite point Ms and the quenching temperature.The steel cooling rate has no influence on the position of the martensite point, but itaffects the martensite transformation process in a certain way. A little below the point Ms,slower cooling enhances the transformation of austenite to martensite. The ability of austeniteto isothermally formate martensite at temperatures a little lower than the point Ms is realizedhere. At temperatures close to the martensite finish temperature Mf but within the intervalMsMf, when a rather significant amount of martensite has been formed already, accelerationof cooling favors a more complete transformation. Here a phenomenon called the stabilization of austenite comes into play. Holding in the region of the martensite finish temperaturemakes retained austenite less prone to subsequent transformation. With slow cooling theaustenite stabilizing processes have time to near completion and the transformation proceedsmore slowly. Austenite stabilization is associated with stress relaxation. The longer theholding time, the greater the stress relaxation and the greater the degree of the metal coolingneeded to accumulate stresses required for the martensite transformation to continue.The quenching temperature can influence largely, either directly or indirectly, the amountof retained austenite. Its direct effect can be connected with thermal stresses facilitating thetransformation of austenite. An indirect effect of the quenching temperature is associatedwith enrichment of intercrystallite boundaries of austenite in carbon and alloying elementsand, primarily, with the transfer of carbides, ferrite, and other phases to the solution. If a steelis heated to a temperature falling within the interval between the critical point Ac1 and thetemperature of full dissolution of ferrite or carbides, the heating temperature will determinethe content of carbon and alloying elements in austenite. If carbides dissolve above Ac1, thenthe amount of retained austenite will increase with quenching temperature. If the quenchingtemperature is elevated above Ac1 and excess ferrite dissolves (with resulting decrease in theaustenite concentration), then the martensite point will occupy the lowest position when asteel is quenched from temperatures slightly higher than the point Ac1. Correspondingly, theamount of retained austenite must be the largest at these temperatures and must subsequentlydecrease until the temperature of full ferrite dissolution is reached.4.2.4 EFFECTONBAINITE TRANSFORMATIONThe bainite transformation (stage II transformation) takes place in carbon steels under theprecipitation curve of undercooled austenite (C curve) in the interval of approximately 5002508C (9304808F). This is called the intermediate transformation. It occurs in between thepearlite and martensite transformations. The kinetics of this transformation and the structures produced are similar to those observed during the diffusion pearlite or diffusionless 2006 by Taylor & Francis Group, LLC.800Temperature, C70026001500400330042001101102103104Time, sFIGURE 4.12 The austenite precipitation diagram of alloy steels with separate C curves of pearlite andbainite transformations. 1, Start of pearlite formation; 2, finish of pearlite formation; 3, start of bainiteformation; 4, finish of bainite formation.martensite transformation. A mixture of the a-phase (ferrite) and carbide is formed as a resultof the bainite transformation, the mixture referred to as bainite.The kinetics of the intermediate transformation is characterized by a number of peculiarities, such as an incubation period; in the bainite temperature range, precipitation of undercooled austenite begins with a certain time delay. The temperature of the maximumtransformation rate (minimum incubation period) depends mainly on the chemical positionof the steel.For alloy steels, C curves of pearlite and bainite transformations can be separated by atemperature interval of a highly stable undercooled austenite where pearlite does not precipitatefor many hours, while undercooling is insufficient for the bainite transformation (Figure 4.12).Alloying elements affect the kinetics of the intermediate transformation, although to alesser degree than in the case of the pearlite transformation. In some alloy steels, theisothermal transformation is retarded over the entire range of the intermediate transformation, whereas in other steels it is inhibited only at temperatures in the upper part of thatrange. In steels alloyed with 2% Si or Cr, the transformation of austenite stops even at thelowest temperatures of the intermediate transformation. When steel is alloyed with Ni or Mn,the transformation is retarded only at high temperatures of the intermediate transformation,whereas at lower temperatures austenite transforms almost completely.Many alloying elements produce a marked effect of the duration of the incubation period,the temperature of minimum stability of austenite, and the maximum transformation rate inthe intermediate range. Figure 4.13 shows the influence of some alloying elements on theseparameters for high-carbon steels with 1.0% C. As is seen, Mn and Cr strongly influence thekinetics of the intermediate transformation, increasing the duration of the incubation periodand lowering the temperature of minimum stability of austenite and the maximum transformation rate. At the same time, alloying with Mo and W, which markedly delays thepearlite transformation, does not have a pronounced effect on the kinetics of the intermediatetransformation.The intermediate transformation in alloy steels consists of a diffusion redistribution ofcarbon in austenite, diffusionless g ! a transformation, and formation of carbides, namely 2006 by Taylor & Francis Group, LLC.1041% CrMn1% Cr0.3% Mn, s1031% CrMo1021% CrW1011(a)123456440tms, 8C4001% CrW3801% Cr0.3% Mn3201% CrMo1% CrMn2801(b)23456101% Cr0.3% Mnnmax, %/s1.01% CrW1% CrMo0.10.011% CrMn0.001(c)1234567Alloying elements, % (by mass)FIGURE 4.13 The influence of Cr, Mo, W, and Mn (a) on the bainite period t at a minimum stability ofaustenite; (b) on the temperature of minimum stability tms, and (c) on the maximum transformation ratevmax in the intermediate range.e-carbide (a type of Fe carbide) and cementite. Owing to the low diffusion mobility of metallicalloying elements, which are substitutional impurities, special carbides are not formed duringthe intermediate transformation. The content of alloying elements in the e-carbide andcementite of bainite is the same as in the initial austenite. Alloying elements do not undergoredistribution during the bainite transformation.4.2.5 TRANSFORMATION DIAGRAMSFORALLOY STEELSThe kinetics of austenite transformation, i.e., the form of the precipitation diagram, dependson a variety of factors, primarily on the chemical composition of austenite. Depending on thealloying of a steel, it is possible to distinguish six basic versions of the diagram of isothermalprecipitation of austenite (see Figure 4.14). In carbon steels and some low-alloy steels containing basically noncarbide-forming elements such as Ni, Si, and Cu, the isothermal precipitation is characterized by C-shaped curves with one maximum (Figure 4.14a). The pearlite andintermediate stages are not separated. When these steels are subjected to continuous cooling,three types of structuresmartensite, martensite and a ferritecarbide mixture, and only aferritecarbide mixturecan be formed depending on the cooling rate. 2006 by Taylor & Francis Group, LLC.Temperature, CA1700A1A113425004Mf300MsMsMs25%Mf100(a)Temperature, C2350050%75%A132Ms25%(c)A14300Mf75%(b)A170050%25Mf25%75%Ms50%1001 10 102 103 104 105(d)1 10 102 103 104 105Time, s(e)1 10 102 103 104 105(f)FIGURE 4.14 Basic versions of precipitation diagrams of undercooled austenite. (a) Carbon and lowalloy steels containing no carbide-forming elements; (b) alloy steels (up to 0.40.35% C) containingcarbide-forming elements; (c) steels alloyed with Cr, Ni, Mo, and W and having a low content of carbon(up to 0.20.25% C); (d) alloy steels containing carbide-forming elements (over 0.40.5% C); (e) highalloy steels with a high content of Cr; (f) high-alloy austenitic steels. 1, Transformation start; 2,transformation finish; 3, start of formation of a ferritecarbon mixture; 4, start of formation of theintermediate transformation products; 5, start of carbide precipitation.In the case of alloy steels containing carbide-forming elements such as Cr, MO, W, and V(Figure 4.14b and d), the precipitation diagrams have two clearly separated ranges of pearliteand intermediate transformations. Each of the ranges is characterized by its own C-shapedcurves. When the carbon content of structural steels is up to 0.40.5%, the stage I transformation is shifted to the right relative to the stage II transformation (Figure 4.14b); if the carboncontent is higher, stage I is found to the left of stage II (Figure 4.14d).Chromiumnickelmolybdenum and chromiumnickeltungsten steels containing 0.150.25% carbon (Figure 4.14c) are characterized by a rather high stability of undercooledaustenite in the pearlite range and a low stability of undercooled austenite in the bainiterange. As a consequence, stage I is absent from the austenite precipitation diagram.In high-alloy chromium steels, the intermediate transformation may be strongly inhibitedand shifted to the martensite temperature range. For this reason the austenite precipitation diagrams have only pearlite transformation and no intermediate transformation(Figure 4.14e).In steels of the austenitic class (high-alloy steels), the martensite start temperature is belowroom temperature and stages I and II precipitation practically do not take place owing to ahigh content of Cr, Ni, Mn, and C (Figure 4.14f). Thanks to the high content of carbon in theaustenite of these steels, excess special carbides may be formed on undercooling.It is worth noting that the aforementioned distinction of the diagrams is conventional toa certain measure as they do not cover a great variety of isothermal and thermokineticprecipitation diagrams of supersaturate austenite. 2006 by Taylor & Francis Group, LLC.4.3 HARDENING CAPACITY AND HARDENABILITY OF ALLOY STEELAs noted in Section 4.2, at great rates of cooling when the cooling curves do not touch theregion of isothermal transformation even at inflection points where austenite is least stable,the latter is undercooled to the martensite range (below the point Ms) and steels is fullymartensitically hardened. Martensitic transformation of austenite results in a supersaturatedsolid solution of carbon in a-Fe; the higher the carbon content of the austenite, the moresupersaturated the solution. Compared with other austenitic transformation products (pearlite and upper and lower bainite), martensite possesses the greatest hardness and gives veryhard steels.The ability of a steel to increase in hardness during quenching is called its hardenability orhardening capacity. The hardening capacity is characterized by the maximum hardness thatcan be obtained on the surface of a given steel product by quenching. To achieve maximumhardness it is necessary to observe basic conditions: the rate of cooling should be equal to orhigher than the critical rate at which quenching gives martensite alone (inevitability with someretained austenite, of course, but without bainite); all carbon at the quenching temperatureshould be in the solid solution in austenite (the quenching temperature should be above thecritical points Ac1 and Ac3 by 30508C (801208F) for hypereutectoid and hypoeutectoidsteels, respectively).Alongside the notion of hardening capacity, broad use is made in practice of the notion ofhardenability, though these two characteristics depend on different factors and are achievedin different ways. The hardening capacity of a steel is determined by the factors affecting thehardness of martensite, while its hardenability is determined by those affecting the quantityof the martensite obtained and the hardness penetration depth. Upon quenching, steel canfeature high hardening capacity and low hardenability at the same time. Such a steel wouldcorrespond to the schematic curve 1 in Figure 4.15. If for a workpiece of the same diameter Dcooled under the same conditions, the distribution of hardness over the cross section ischaracterized by curve 2; such a steel possesses medium or poor hardening capacity butgood hardenability. Finally, steel that corresponds to curve 3 would possess high hardeningcapacity and high hardenability.4.3.1 HARDNESSANDCARBON CONTENTThe hardening capacity of a steel whose general characteristic could be maximum hardnessdepends mainly on the carbon content and, to a lesser extent, on the amount of alloyingelements and austenite grain size. Increasing the carbon content of martensite increases its2Hardness31DFIGURE 4.15 Distribution of hardness over the cross section of workpiece for three steels differing inhardenability and hardening capacity. 2006 by Taylor & Francis Group, LLC.hardness (Figure 4.7). Note that the hardness of a quenched steel and the hardness ofmartensite crystals are not the same thing because quenched steel contains retained austenite.The hardness of a steel quenched from austenite temperatures passes its maximum at a carbonconcentration of 0.80.9% C and then decreases due to an increase in the volume fraction ofsoft retained austenite (Figure 4.16). For the above carbon content of steel, the martensitepoint Ms drops significantly, which leads to an increase in the proportion of retained austenitein quenched steel. Steel with 1.9% C quenched from a temperature higher than Ast has thesame hardness as quenched steel with 0.1% C. If hypereutectoid steels are quenched from atemperature of Ac1 (20308C) (70908F), as is common practice, all hypereutectoid steelswould have practically the same austenite composition at the same quenching temperatureand level of hardness (Figure 4.16, curve b).Another important feature of the dependence of steel hardness on carbon content is thatan increase in the carbon content to ~0.6% results in a most dramatic rise in the maximumhardness; then the curve becomes less steep. This is probably associated with the very natureof high martensite hardness in steel.The martensite transformation of austenite results in a supersaturated solid solution ofcarbon in a-Fe. An increase in the carbon content of martensite weakens, rather thanstrengthens, the interatomic bonds. This is due to an increase in the distance between ironatoms brought about by implanted carbon atoms. Carbon nevertheless increases the hardnessof martensite, which is explained primarily by the fact that carbon atoms implanted into thea-Fe lattice impede the slip of dislocations in martensite (the so-called solid-solution strengthening mechanism).During quenching or during the aging of quenched steel, carbon atoms in martensitecrystals surround dislocations (atmospheres around dislocations), thus pinning them. Thisleads to a general increase in plastic deformation resistance despite the fact that carbonweakens interatomic bonding in the martensite lattice. In steels with a high martensite startpoint Ms such as carbon steels containing less than 0.5% C [Ms > 3008C (5708F)], quenchcooling over the martensite range is characterized by the most favorable conditions for partialprecipitation of martensite with the release of disperse carbide particles. Moreover, in allsteels hardened at normal rates, carbon has time to segregate as the steel cools above the pointMs. The carbon segregates of austenite are inherited by martensite, and since the latter isalready supersaturated with carbon, these segregates become nucleation sites of carbideparticles. This is in agreement with the fact that at very high cooling rates the hardness of70cbHardness, HRC6050a40302000., %FIGURE 4.16 Hardness of carbon and alloy steels depending on the carbon content and quenchingtemperature. (a) Quenching above Ac3; (b) quenching above Ac1 (7708C); (c) microhardness of martensite. 2006 by Taylor & Francis Group, LLC.FIGURE 4.17 Microstructure of plate martensite. Light shading retained austenite 50,000.martensite crystals is two thirds of that obtained at normal cooling rates. High hardness ofmartensite may also be due to the fact that carbon makes a noticeable contribution tocovalent bonding whose main property is high plastic deformation resistance.Owing to the above strengthening mechanisms, carbon has such a strong strengtheningeffect on martensite that the hardness of a quenched steel does not depend on the concentration of alloying elements dissolved in the martensite by the substitutional mechanism, but isdetermined by the concentration of carbon.To conclude, extreme strengthening of steels during martensitic hardening is due to theformation of a carbon-supersaturated a-solution, an increase in the density of dislocationsduring the martensite transformation, the formation of carbon atom atmospheres arounddislocations, and precipitation of disperse carbide particles from the a-solution.4.3.2 MICROSTRUCTURE CRITERION FOR HARDENING CAPACITYStudies of the structure of hardened carbon steels and carbon-free iron-based alloys revealedtwo main morphological types of martensite: plate and lath. These two types of martensitediffer in the shape and arrangement of crystals, substructure, and habit plane.Plate martensite (which is also called needle type, low temperature, and twinned) is a wellknown classical type of martensite that is most pronounced in quenched high-carbon ironalloys with a high concentration of the second elementfor instance, FeNi alloys with a Nicontent higher than 28%. Martensite crystals are shaped as thin lenticular plates. Such a shapecorresponds to the minimum energy of elastic distortions when martensite is formed in theaustenite matrix; it is similar to the shape of mechanical twins (Figure 4.17).Neighboring plates of martensite are commonly not parallel to each other and formframe-like ensembles. Plates that are formed first (near the point Ms) extend through theentire length of the austenite grain, dividing it into sections. A martensite plate cannot,however, cross the boundary of the matrix phase; therefore the maximum size of the martensite plate is limited by the size of the austenite grains. As the temperature is lowered, newmartensite plates are formed in the austenite sections, the size of the plates limited by the sizeof the matrix sections. In the course of transformation the austenite grain splits into stillsmaller sections, in which smaller and smaller martensite plates are formed. In the case of asmall austenite grain caused by, for instance, a small overheating of steel above Ac3, themartensite plates are so small that the needle-type pattern cannot be observed in the microsection and the martensite is usually called structureless. It is this type of martensite that ismost desirable. After quenching, martensite retains some austenite between its plates at roomtemperature (Figure 4.18). 2006 by Taylor & Francis Group, LLC.FIGURE 4.18 Microstructure of lath martensite. 50,000.Lath martensite (which is also called massive, high-temperature, or nontwinned) is awidespread morphological type that can be observed in quenched low-carbon and mediumcarbon steels, the majority of structural alloy steels, and comparatively low-alloy noncarboncontaining iron alloysfor instance, FeNi alloys with a Ni concentration of less than 28%.Crystals of lath martensite are shaped like thin plates of about the same orientation, adjacentto each other and forming more or less equiaxial laths.The plate width within the lath is about the same everywhere, ranging from severalmicrometers to fractions of a micrometer (commonly 0.10.2 mm); i.e., it can reach or evenexceed the resolution limit of a light microscope. Inside the martensite laths there areinterlayers of retained austenite 2050 nm thick. One austenite grain can contain severalmartensite laths.The formation of lath martensite has all the main features specific to the martensitetransformation, including the formation of a relief on a polished surface. Transmissionelectron microscopy reveals a rather complicated fine structure of martensite crystals withmany dislocations and twins in many iron alloys.The substructure of the plate martensite shows an average zone of elevated etchability,also called a midrib, even under a light microscope. Electron microscopy has shown that themidrib is an area with a dense arrangement of parallel fine twin interlayers. The twinningplane in martensite of iron-based alloys is commonly {1 1 2}M. Depending on the compositionof the alloy and martensite formation conditions, the thickness of the twinned interlayers mayform several tenths of a nanometer to several tens of nanometers. On both sides of the centraltwinned zone there are peripheral areas of martensite plates that contain dislocations ofrelatively low density (1091010 cm2).The substructure of lath martensite is qualitatively different from the substructure of platemartensite in that there is no zone of fine twin interlayers. It is a complex dislocationstructure characterized by high-density dislocation pileups with densities on the order of10111012 cm2, i.e., the same as in a metal subjected to strong cold deformation. The lathsof lath martensite often consist of elongated slightly misoriented subgrains. Twin interlayerscan occur in lath martensite, but their density is much lower than in the midrib in platemartensite, while many of the laths do not contain twins at all.The substructure of retained austenite differs from that of the initial austenite by a higherdensity of imperfections occurring under local plastic deformation due to martensite crystals.Flat dislocation pileups, dislocation tangles, and stacking faults may be observed in austenitearound martensite crystals. 2006 by Taylor & Francis Group, LLC.At present it is believed that the decisive role in the formation of plate martensite belongsto accommodating (complementary) twinning deformation, whereas for the lath type this roleis played by slip. As temperature is decreased, resistance to slip increases at a higher rate thanthat of resistance to twinning; therefore, the martensite transformation at low and hightemperatures results in twinned and lath martensite, respectively. In alloys a decrease intemperature causes the morphology of martensite to change from the plate to lath type.The composition of iron-based alloys has a substantial effect on the martensite morphology.Shown below is the effect of some alloy compositions on the formation of plate and lathmartensite; the numerals show the second component content in percent.SystemFeCFeNiFeC20.60.62.0Lath type, MPlate, MFeN0.70.72.529293410Chromiumnickel, manganese, chromiummanganese, and other steel alloys with lowenergy stacking faults contain hexagonal e-martensite with plates in parallel to planes {1 1 1}gk {0 1 1}e (Figure 4.19). Some alloy steels have a mixture of e- and a-martensites.In conclusion it should be noted that the structure of a metal or alloy that has undergonemartensite transformation features many more imperfections than after a disordered rearrangement of its crystal lattice: the more the developed grain boundaries and subboundaries, the greater the density of dislocations and twin interlayers.4.3.3 EFFECTOFGRAIN SIZEANDCHEMICAL COMPOSITIONAs noted in Section 4.3.1, the hardening capacity of a steel, i.e., its ability to undergomartensitic hardening, depends mainly on its carbon content and to a lesser extent on itscontent of alloying elements and the size of austenite grains. At the same time these twofactorsgrain size and chemical composition of the steel (or austenite, to be more exact)can produce a substantial effect on hardenability, i.e., the depth to which the martensite zonecan penetrate. It is reasonable, therefore, to consider the effects of grain size and chemicalcomposition of austenite on hardening capacity and hardenability separately.FIGURE 4.19 Microstructure of e-martensite. 50,000. 2006 by Taylor & Francis Group, LLC.In the case of austenitemartensite transformation, martensite plates develop inside theaustenite grain, extending from one side to the other. If the steel is considerably overheatedabove the critical point Ac3, then coarse grains of austenite are formed; this results in largermartensite plates than normal. In quenched steel the volume fraction of retained austenite ishigher if its structure is dominated by large martensite plates. Although the hardness ofmartensite is practically independent of the plate size, an increase in the total soft austenitecontent of quenched steel leads to a decrease in its maximum hardness, i.e., impairs itshardening capacity. Moreover, the plastic properties of steel, particularly its toughness, alsodeteriorate with the coarsening of the structure. It is therefore advisable to obtain a fineneedle structure after quenching if a set of good mechanical properties is needed; this may beachieved with a fine-grain austenite structure, which is produced at small overheating of steelat temperatures higher than Ac3.Alloying elements can have direct or indirect effects on the hardening capacity of steel.Indirect effects are associated primarily with hardenability. Since the majority of the elementstend to shift the isothermal austenite precipitation curves to the right and hence decrease thecritical rate of cooling, it is easier to obtain maximum hardness for an alloy steel than for anordinary carbon steel. Specifically, the presence of alloying elements facilitates the achievement of maximum hardness by cooling the steel in more lenient media (in oil, for instance)when the mass of the workpiece to be hardened is comparatively large.A second indirect effect is related to the carbide-forming elements. If hardening is to beapplied to an alloy steel containing elements that form stable carbides by heating below thecarbide dissolution point, then the carbon content of the major martensite mass will be lowerthan the total carbon content of the steel. As a result, the maximum achievable hardness forthis carbon content will not be attained because the hardening capacity of the steel willdeteriorate.It should be noted that high-carbon steel, such as tool grade steel, features high hardnesseven if it is quenched from a temperature somewhat lower than the carbide dissolution point,despite the fact that part of the carbon stays outside the solution. The martensite hardnessalso decreases. This decrease is small and is due mainly to the fact that hardness versus carboncurves for carbon steels and alloy steels are rather smooth at carbon concentrations higherthan 0.6% (Figure 4.16); it is compensated for by a considerably lower amount of retainedaustenite and the high hardness of the carbides themselves.The situation is different for structural steels containing less than 0.4% C. The maximumhardness curve is so steep (Figure 4.16) that even a small decrease in the concentration ofcarbon in martensite in an alloy steel due to incomplete dissolution of special carbides wouldlead to a considerable reduction in the hardness of the martensite. The steepness of the curveover the range of 0.10.4% C shows that from the viewpoint of hardening capacity it isessential to heed even the smallest fluctuations in the carbon content, specifically fluctuationswithin the quality limits, up to individual ingots. Therefore the carbon content limits forspecific structural steels should be as narrow as possible, although this may encounter sometechnological difficulties.The indirect effect of alloying components on hardenability is not great. It is thereforepossible to construct a general curve for the dependence of maximum hardness on the carboncontent for carbon and alloy steels (Figure 4.16). This is understandable because martensite ofan alloy steel is a combined solid solution in which atoms of the alloying elements replace ironatoms in the lattice, while carbon atoms are implanted into this lattice. Carbon atomsintroduced into the a-Fe lattice impede the slip of dislocations in martensite and therebyincrease its hardness.A certain increase in the hardness of martensite due to alloying elements can be expectedonly because of the strengthening of a-Fe during quenching. The possibility of quench 2006 by Taylor & Francis Group, LLC.240Cr220Hardness, Hv200Ni180MnNb160TiSi140VWAI120Co1008000. elements in -Fe, %FIGURE 4.20 Effect of dissolved alloying elements on hardness of ferrite after quenching from 12008Cin water.strengthening an alloyed ferrite has been studied in detail. From Figure 4.20, one can judgeroughly how different elements increase the hardness of ferrite with ~0.01% C upon quenching from 12008C in water. Alloying elements causing substantial strengthening of ferriteshould also increase the hardness of martensite in a quenched steel, though this increase inhardness should be comparatively low.During the quenching of steel products, the rate of cooling is the greatest for the surface,decreasing steadily toward the center of the section. Evidently, the depth of the hardened zone(hardenability) will be determined by the critical rate of quenching; thus, hardenability willincrease with a decrease in the critical rate of quenching. This rate, in turn, depends on theresistance of austenite to precipitation at temperatures higher than the martensite point Ms.The farther to the right the lines in the isothermal austenite precipitation diagram, the lowerthe critical rate of quenching and the higher the hardenability of the steel products. Thus, thefactors that affect the stability of undercooled austenite will affect the hardenability as well.The main factors that produce a decisive effect on the hardenability of steel are (1) thechemical composition of the steel (composition of austenite, to be more exact); (2) austenitegrain size; and (3) the homogeneity of austenite. Under otherwise equal conditions, coarseaustenite grains improve the hardening capacity of steel. This circumstance is connected withthe extent of grain boundaries; the extent is less, the coarser the grain. Since nucleation centersare formed primarily along the austenite grain boundaries during austenite precipitationabove the point Ms, it is always easier to undercool austenite with coarse grains, therebyincreasing hardenability.To estimate hardenability, in practice, use is made of the quantity called the criticaldiameter. The critical diameter (Dcr) is the maximum diameter of a bar permitting throughhardening for a given cooling medium. To avoid putting hardenability in dependence on the 2006 by Taylor & Francis Group, LLC.Critical harding rate, C/s1,400I1,2001,000800600400II2000 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8C, %FIGURE 4.21 Curves showing the effect of carbon on the critical rates of quenching. I, low-heatingtemperatures; II, high-heating temperatures.method of cooling and type of coolant, use is made of still another notion, the ideal criticaldiameter (D1), which is the diameter of a maximum section allowing through hardening in anideal cooling liquid that is absorbing heat at an infinitely great rate. It was established that thegrain size and quantity D21 have an approximately linear inverse dependence on one another.The chemical composition of a steel has the strongest impact on its hardenability. Thisis due primarily to the fact that carbon and alloying elements affect the critical rate ofquenching.Figure 4.21 shows the effect of carbon on the critical rate of quenching for carbon steel. Itcan be seen that the minimum rate of quenching is observed in steels that are close toeutectoid with respect to carbon content (curve I). A decrease in the carbon content of steelbelow 0.4% leads to a sharp increase in the critical rate to the extent that at a certain minimumcarbon content martensitic hardening becomes virtually impossible. An increase in the criticalrate of hypereutectoid steel with an increase in the carbon content is explained by the presenceof cementite nuclei facilitating the austenite precipitation. Hence, the trend of curve I isrelated to the incomplete hardening of hypereutectoid steels. If completely hardened (sufficient holding at a temperature higher than Ast), an increase in the carbon content leads to acontinuous decrease in the critical rate of quenching (curve II), with a resulting rise inhardenability.The effect of alloying elements on hardenability can be estimated by the degree of increaseor decrease in stability of undercooled austenite in the pearlite and intermediate ranges. Withthe exception of cobalt, all alloying elements dissolved in austenite impede its precipitation,decrease the critical rate of quenching, and improve hardenability. To this end, broad use ismade of such additives such as Mn, Ni, Cr, and Mo. Particularly effect is complex alloyingwhereby a combination of elements enhances their individual useful effects on hardenability.Figure 4.22 shows the effect of third-element alloying on the hardenability of an ironnickelsteel. It can be seen that Mn, Cr, and Mo additives improve hardenability to a considerableextent.The improving effect of alloying on hardenability is used in two ways. First, alloyingensures through hardening across sections inaccessible for carbon steels. Second, in the caseof small-section products, replacing carbon steel with an alloy steel permits less radicalcooling regimes. Small-diameter carbon steel products can be hardened by quenching inwater. This, however, may result in impermissible residual stresses, deformations, and cracks,particularly in products of a complicated shape. If an alloy steel is used, quenching in watercan be replaced with softer hardening in emulsion, oil, or even air. 2006 by Taylor & Francis Group, LLC.Ideal critical diameter D, mmWith % Mn175With % Cr11.5With % Mo0.250.51501250.511000.1750.750000.3250012301230123Ni, %FIGURE 4.22 The effect of manganese, chromium, and molybdenum on the hardenability of steels withdifferent nickel contents.4.3.4 BORON HARDENING MECHANISMIt has long been observed that small additions of certain elements, e.g., titanium, aluminum,vanadium, zirconium, and boron, can considerably improve the hardening properties of steel.The most effective in this respect is boron. Hardenability of carbon and low-alloy steelsincreases considerably upon introduction of boron in amounts of thousandths of a percent. Afurther increase in the boron content does not produce any further improvement in hardenability. The improving effect of boron is noticeable only where steel has been preliminarilywell deoxidized and denitrified, because boron has good affinity for oxygen and nitrogen.Therefore, before introducing boron into steel it is necessary to add aluminum, titanium, orzirconium. Figure 4.23 shows hardenability curves for a low-carbon steel without boron, withboron added, with boron and vanadium, and with boron, vanadium, and titanium. It can beseen that the extent of the martensite range for all the steels with boron added is greater thanthat in steels without boron; however, it is practically the same for steels with severaladditives. At the same time, the extent of the half-martensite structure range is much greaterif the steel contains other additives along with boron.Hardness, HRC7046035040123020024681012Length from end, mm1416FIGURE 4.23 Curves of hardenability for steels with 0.440.43% C and various small amounts ofadditives. 1, Without boron; 2, with an addition of boron; 3, with boron and vanadium added; 4,with boron, vanadium, and titanium added. 2006 by Taylor & Francis Group, LLC.The effect of boron on hardenability decreases with an increase in the carbon content. Ifthe carbon content exceeds 0.9%, boron does not have any measurable effect on hardenability. Boron alloying for the purpose of improving hardenability is therefore useful only forlow-carbon steels of various applications. It does not have any effect on the hardenability oftool steels or on high-carbon carburized layers. The above relationship between the carboncontent and the effect of boron on the hardenability of steel is due to the fact that the twoelements have the same effect on austenite precipitation. Boron increases the length of theaustenite precipitation incubation period, thereby decreasing the critical temperature ofquenching. Like carbon, boron facilitates the enlargement of austenite grains under heating.These two factors have a positive effect on the hardenability of steel. Therefore, in highcarbon steels the effect of small doses of boron is practically negligible.It should be also noted that the positive effect of boron on hardenability is full only if thequenching temperature is sufficiently high [8509008C (156016508F)]. Various suggestionshave been put forward with respect to the mechanism by which boron affects the hardenability of steel. Quite probably, this is a very special mechanism because boron can produceits effect at very low concentrations above, above which no further effect on hardenability isobserved. Some researchers believe that boron increases hardenability just because it facilitates the increase in the size of austenite grains under heating. Although boron does tend toincrease the grain size, it produces the same hardenability-improving effect in steels with smallgrains. Others explain the fact that boron increases the stability of austenite and consequentlyimproves hardenability by the fact that it increases the coefficient of surface tension at theaustenitenew-phase nucleus interface. Therefore more energy is required for the formationof a nucleus of a critical size capable of growth. Because of this stability of austenite increases.An increase in hardenability following the addition of small doses of boron is more frequentlyexplained by the fact that this element is surface-active in austenite. There is experimentalevidence that boron is segregated at the boundaries of austenite grains and dissolves ininsignificant amounts in this layer. Since boron forms an interstitial solid solution withiron, interaction between its atoms and iron atoms must, evidently, be the same as in carbon,which leads to a decrease in the difference in free energy between the g- and a-phases. Thisimpedes the formation of new-phase nuclei of critical size, which are formed primarily at theaustenite grain boundaries. The resistance of austenite to precipitation improves, therebyincreasing the hardenability of the steel. The cessation of the boron effect on hardenabilitywith increasing boron content is due to its low limiting solubility in g-iron at a giventemperature [about 0.003% at 10008C (18328F)]. As soon as the boron solubility limit at thegrain interface is achieved, any further increase in its total content leads to the formation ofironboron compounds such as Fe2B within austenite grain boundaries and to the distribution of boron over the bulk of the grain. As centers of crystallization, chemical compoundparticles cause an earlier onset of austenite precipitation, which results in lower hardenability.At the same time, an earlier onset of austenite precipitation at grain boundaries is compensated for by a delay in the formation of critical new-phase nuclei in the bulk of the graincaused by boron due to an increase in its content up to the solubility limit for austenite.Therefore, alloying with boron in amounts exceeding thousandths of a percent does not haveany effect on hardenability and even may impair it.Based on this viewpoint, one can explain some other specific effects of boron on thehardenability of steel that were noted above. Thus, boron increases the duration of theaustenite precipitation incubation period only, the duration determined by the formation ofcritical nuclei at grain boundaries but not affecting the length of austenite precipitation. Fromthis viewpoint it is also clear that boron, like carbon, facilitates enlargement of austenitegrains under heating. An increase in the quenching temperature first improves hardenabilityowing to an increase in the concentration of boron at the austenite grain boundaries to its 2006 by Taylor & Francis Group, LLC.solubility limit at these boundaries. A further increase in the quenching temperature andenrichment of the grain boundaries in boron can lead to the formation of boronironcompounds. The hardenability of the steel will not increase further and may even decrease.Finally, the reduction of the boron effect by carbon can be explained by the fact that boronand carbon have virtually the same effect on hardenability. In high-carbon steels, the effect ofboron becomes practically negligible owing to its poor solubility in iron.4.3.5 AUSTENITIZING CONDITIONS AFFECTING HARDENABILITYThe austenite condition prior to quenching (chemical composition, grain size, homogeneity ofaustenite) has the decisive effect on the hardening capacity and, especially, hardenability ofsteel. Other factors are secondary or derive from the basic three. These factors, in turn, aredetermined by the carbon content, type, and amount of alloying elements at the time ofquenching, the quenching temperature (austenitizing temperature), and the holding time at agiven temperature.Austenitizing of a heated alloy steel consists of a polymorphous a ! g transformation,the dissolution of cementite, special carbides, nitrides, and intermetallics in austenite, andrecrystallization of the austenite grains.To improve the hardening capacity and hardenability of steel, the austenitizing conditionsshould be such as to ensure that a maximum amount of carbon passes from the ferritecarbide mixture to the solution and, at the same time, no marked growth of grains occurs as aresult of overheating, as this would lead to a high brittleness and the formation of quenchingcracks. The quenching temperature should be maintained as constant as possible, and theholding time should be just enough to ensure uniform heating of the workpiece and dissolution of carbides. For their complete dissolution in austenite, coarse-plate and coarse-graincarbides need more time than thin-plate and fine-grain ones. Steels alloyed with elementsforming special carbides should be heated to a temperature considerably exceeding Ac3. Smallcarbides available in the structure impede enlargement of grains and the nuclei of the newphase facilitate transformation of austenite in the pearlite range and increase the critical rateof quenching, thus decreasing the hardenability of the steel. As the quenching temperatureand time are increased, the critical rate of quenching decreases and, accordingly, hardenability rises, because carbides and other inclusions playing the role of new-phase nucleidissolve most.The degree of austenite homogeneity and dispersion of local carbon pileups (which can actas nuclei during transformation in the pearlite range) can have a strong effect on hardenability and hardening capacity. When the quenching temperature is increased, carbidesdissolve together with other minute, sometimes hardly measurable, quantities of inclusionssuch as nitrides and sulfides, which can also serve as nuclei during transformation.Finally, when the quenching temperature and holding time are increased, enlargement ofaustenite grains has its effect on the process of transformation. Since the pearlite transformation begins at grain boundaries, an increase in the austenite grain size causes a decrease inthe critical rate of quenching and hardenability improves.Nearly all of the alloying elements impede the growth of austenite grains. The exception ismanganese, which adds to the growth of grains. The strongest growth retardants are V, Ti,Al, Zr, W, Mo, and Cr; Ni and Si produce a weaker retarding effect. The main cause of thisretarding effect is believed to be the formation of low-soluble carbides, nitrides, and otherphases, which may serve as barriers for the growth of austenite grains. Such active carbideforming elements as Ti, Zr, and V impede growth more strongly than Cr, W, and Mo do,because the carbides of the former elements are more stable and less soluble in austenite.Experimental studies on the solubility of V, Nb, Ti, and Al carbides and nitrides in austenite 2006 by Taylor & Francis Group, LLC.NbC0.1% C 0.4% CZrC% Zr0.1% CTiC% Ti0.4% C% NbContent of metal in austenite, %%VVC0.1% C0.1% C0.8% C0.4% C1.2% C0.4% C1.2% C1.2% C0.8% C0.8% C0.8% C1.2% C800 900 10001000 12001000 12001000 1200Temperature, CFIGURE 4.24 Solubility of carbides in austenite at various temperatures depending on the carboncontent (shown as % C on the curves) for (a) vanadium; (b) niobium; (c) titanium; (d) that the carbides of these elements (Ti in particular) are more soluble than nitrides.Titanium nitrides are virtually insoluble in austenite no matter what the temperature is.Niobium and aluminum nitrides also have poor solubility in austenite. Carbon has a greateffect on the solubility of carbides. Figure 4.24 shows relevant data on V, Nb, Ti, and Zrcarbides. An increase in the temperature of carbide solubility in austenite with a rise in thecarbon content is due to the greater activity of carbon at its higher concentrations in a solidsolution and higher thermodynamic activity. It should be noted that C, N, and Al are notbound to carbides or nitrides, but found in the solid solution of austenite facilitate the growthof austenite grains. The elements B, Mn, and Si also favor the growth of grains. Therefore,addition of these elements into steel improves its hardenability.Different heats of steels of the same quality may considerably differ in their tendencytoward the growth of austenite grains because they contain different amounts of low-solubledisperse particles of carbides, nitrides, and other phases, which are barriers to the growth ofaustenite grains. The distribution and size of these particles depend both on steelmakingconditions and preliminary heat treatment. Thus, the tendency of steel to grain size growthunder heating depends on, in addition to its composition, the metallurgical quality andprocess, i.e., its history preceding the thermal treatment.Liquidation also has a considerable effect on hardenability. In order to obtain homogeneous austenite in steel exhibiting liquation, it is necessary to keep the steel at the quenchingtemperature for a sufficiently long time. This refers to cat steel where liquation is the highestand also to forged and rolled steels. Longer quenching times increase hardenability owing tothe elimination of residual liquation and fluctuation in homogeneity of austenite.Note in conclusion that hardening capacity and hardenability are not important bythemselves in practical applications. They are important if they can improve overall properties of steels in accordance with practical needs. OF ALLOY STEELSSTRUCTURAL CHANGESONTEMPERINGStructural changes on tempering were considered in detail in Chapter 3. Therefore, thissection briefly considers only the most characteristics effects.Tempering is a thermal martensitic treatment of quenched steels. The basic process thattakes place during tempering is martensite precipitation. The first structural change during 2006 by Taylor & Francis Group, LLC.tempering is carbon segregation at dislocations. The second stage of tempering is precipitation of intermediate e-carbide with a hexagonal lattice, which forms under heating above1008C (2128F). During the third stage, cementite precipitates above ~2508C (~4808F). At thefinal stage of tempering above 3508C (6608F), cementite particles coagulate and spheroidize.Consider the changes that take place in martensite and austenite structures at differentstages of tempering. During the first stage, beginning at 808C (1758F) and up to 1708C(3308F), the c parameter of the martensite lattice decreases. The ratio c=a becomes close tounity. Tetragonal martensite transforming to a cubic form is called tempered martensite. Thedecrease of tetragonality is connected with precipitation of carbon from the solution.Heating of steels over 2008C (3908F) and up to 3008C (5708F) activates transformation ofretained austenite to a heterogeneous mixture composed of a supersaturated a-solution andthe Fe3C carbide. This means that retained austenite transforms to tempered martensite,Feg (C) ! Fea (C) Fe3 CBy the end of transformation (~3008C; 5708F), the retained austenite contains about0.150.20% C.Heating above 3008C (5708F) leads to a further precipitation of carbon and the relaxationof internal stresses arising from previous transformations. Complete precipitation of carbonwas found to occur at 4008C (7508F). A further increase in temperature leads only tocoagulation of ferrite and cementite particles.During tempering, cementite acquires a globular form when a ferritecementite mixtureis formed from martensite. The different form of cementite in the ferritecementite mixturedetermines the difference in properties.4.4.2 EFFECTOFALLOYING ELEMENTSThe influence of alloying elements on transformations during tempering depends on whetherthey dissolve in ferrite and cementite or form special carbides. The diffusion mobility of atomsof alloying elements dissolved in a-Fe by the substitutional method is many orders ofmagnitude lower than the diffusion mobility of carbon atoms dissolved by the interstitialmethod. So at a temperature below 4008C (7508F) no diffusion redistribution of alloyingelements in the matrix takes place. First the e-carbide and then cementite precipitate from thea-solid solution. The concentration of alloying elements in them is the same as in martensite.Atoms of alloying elements in the e-carbide and cementite lattice formed below 4008C (7508F)partly replace iron atoms. Complex carbides such as (Fe, Cr)3C and (Fe, V)3C are formed.The first stage of transformations in martensite (formation of tempered martensite) ata temperature below 1508C (3008F) is affected little by alloying elements. At this stage oftempering, nucleation of carbide particles depends basically on supersaturation of thea-solution with carbon.The second stage of martensite precipitation is strongly influenced by a number ofalloying elements. They retard the growth of carbide particles, and consequently supersaturation of the a-solution with carbon is preserved. Thus the state of tempered martensite isretained up to temperatures of 4505008C (8409308F). Additions of Cr, W, Mo, V, Co, andSi bring about this effect.A delay in martensite precipitation can be explained by two factors. First, one of thealloying elements lowers the rate of carbon diffusion in the a-solution. Second, the otherelements can increase the strength of interatomic bonds in the a-solution lattice. This willprevent the atoms from crossing the a-solution carbide interface. Both factors impedeprecipitation of martensite. 2006 by Taylor & Francis Group, LLC.Alloying elements affect carbide transformation under tempering under 4508C (8408F)when their diffusion movement becomes possible. In this case special carbides are formed.With an increase in the tempering temperature, intermediate metastable carbides stabilize.For example, when molybdenum and tungsten steels are tempered, Me2C (Mo2C and W2C) isformed first, then Me23C6 appears, and finally Me6C emerges. The sequence of their formation can be written asFe3 C ! Me2 C Me23 C6 ! Me6 CAlloying elements affect the coagulation rate of carbide particles. Nickel accelerates thecoagulation rate while chromium, molybdenum, vanadium, and other elements slow itdown. Owing to the low diffusion rate of alloying elements, the coagulation of specialcarbides proceeds slowly. Even alloyed cementite (Fe, Cr)C3 coagulates much more slowerthan Fe3C in a carbon steel.Additions of alloying elements slow down recrystallization and polygonization. Atoms ofthese elements form impurity atmospheres near dislocations and prevent their movementduring polygonization. Disperse particles of special carbides retard movement of large-angleboundaries during polygonization.4.4.3TRANSFORMATIONSOFRETAINED AUSTENITE (SECONDARY TEMPERING)Alloying elements have the greatest influence on the martensite transformation temperature.This affects the amount of retained austenite in alloy steel. Some elements (e.g., cobalt) raise thepoint Ms, thus decreasing the amount of retained austenite. Others (e.g., silicon) have noinfluence on Ms. However, the majority of elements decrease the martensite point and increasethe amount of retained austenite in quenched steel. Up to 60% of retained austenite is left inhigh-carbon steels during quenching and 1015% in a large number of structural alloy steels.During tempering of carbon and low-alloy steels, retained austenite transforms over thetemperature interval of 2302808C (4405408F) or at lower temperatures if the holding time isextended. Alloying elements, especially Cr and Si, inhibit that transformation, shifting it tohigher temperatures and longer tempering time. The transformation kinetics of retainedaustenite during tempering is similar to those of undercooled austenite. Steels with two clearlydistinguished transformation ranges (pearlite and bainite) also exhibit two regions of fasttransformation of retained austenite during tempering that are separated by a zone of highstability of retained austenite.When alloy steels are tempered at 5006008C (93011108F), in many cases the transformation of retained austenite is not complete. The retained austenite that did not precipitate atthese tempering temperatures transforms during cooling from those temperatures (secondaryquenching). The phenomenon is most pronounced in high-speed and high-chromium steels.The secondary martensite transformation during cooling after tempering is caused by thedepletion of austenite in carbon and alloy elements in the course of tempering. As a result, thetemperature of retained austenite Ms during cooling is increased.Secondary quenching (double tempering) is also observed in structural steels. It takesplace if the primary quenching is accompanied by a partial intermediate transformationleading to an increase in the carbon content of austenite. During tempering at 5005508C(93010208F), retained austenite with a high content of carbon yields carbides intensively,and the martensite transformation temperature Ms increases. As a result, the secondarymartensite transformation takes place during cooling after tempering.In high-alloy steels, for example high-speed steels, even a very long tempering athigh temperatures does not completely eliminate the retained austenite. To obtain a full 2006 by Taylor & Francis Group, LLC.transformation, it is necessary to perform double tempering. Double tempering favorsadditional precipitation of special carbides and decreases the degree of austenite alloying.This causes another increase in the transformation temperature Ms. Sometimes multipletempering is required to realize the most complete transformation of retained austenite.4.4.4 TIMETEMPERATURE RELATIONSHIPS IN TEMPERINGThe kinetics of structural transformations during tempering is described by temperaturetimecurves similar to the curves shown in Figure 4.25. After quenching from 9008C (16508F),steels with 0.7% C, 1% Cr, and 3% Ni contain 30% of retained austenite. When plotting thecurve and diagram, the nontransformed austenite (30%) was taken to be 100%. It is foundthat at 6008C (11108F) 5% of primary austenite transforms in 7 min and 5% of retainedaustenite transforms in 30 s. However, after 1015% of the retained austenite is transformed,the transformation rate of retained austenite becomes smaller than that of the initial austenite. The transformation of retained austenite is not complete. It is inhibited on reaching 45%at 5008C (9308F) and 60% at 5508C (10208F). The retained austenite that does not precipitateimmediately at these tempering temperatures transforms during cooling after tempering. Thetransformation rate of retained austenite in the intermediate range is much higher than that ofthe initial austenite; 25% of initial austenite transforms at 3008C (5708F) in 75 min and thesame amount of retained austenite in 15 s; 75% of the initial austenite transforms in 220 minand 75% of the retained austenite in 9.5 min. In a typical structural steel (0.37% C, 1% Cr, 1%Mn, 1% Si), 5% of the initial austenite transforms in 19 min at 6008C (11108F) or in 5 min at4008C (7508F); 5% of retained austenite at the same temperature transforms in a few seconds.Another specific feature of retained austenite transformation of this steel in the intermediate range is the lowering of the transformation limit. For example, at 3508C (6608F), 70%of the initial austenite and only 40% of retained austenite transform. This difference decreaseswith increasing temperatures.4.4.5 ESTIMATIONOFHARDNESSAFTERTEMPERINGHardness decreases noticeably when alloy steels and addition-free steels are subjected totempering at 5006008C (93011108F). This decrease is due to the precipitation of martensiteand coagulation of cementite. However, when the tempering temperature is higher, the1260025%5005752552575t, 8C 4003002557520010012351020 30 50 1235smTimeFIGURE 4.25 Timetemperature transformation curves. 2006 by Taylor & Francis Group, LLC.1020 30123h510hardness of steels with additions of titanium, molybdenum, vanadium, or tungsten increases.This phenomenon is called secondary hardening.Secondary hardening is caused by the formation of clusters of atoms of alloying elementsand carbon (a maximum hardness often corresponds to the clusters) and the replacement ofrelatively coarse particles of cementite by much more disperse precipitates of special carbides(TiC, VC, Mo2C, W2C). When these particles coagulate, hardness decreases. Particles ofMe6C are rather coarse and do not add to strengthening.The chromium additive causes a small secondary hardening. This is connected with arapid coagulation of the Cr7C3 carbide at 5508C (10208F) as opposed to Mo2C and especiallyW2C. During secondary hardening an increase in the yield stress is accompanied by anincrease in toughness owing to dissolution of coarse cementite particles.4.4.6EFFECTOFTEMPERINGONMECHANICAL PROPERTIESThe manner in which structural changes that take place during tempering affect the propertiesof steels depends on the particular tempering conditions. The general tendency of changes inmechanical properties of carbon steels during tempering is that as the tempering temperatureis elevated, the strength parameters sB and s0.2 (fracture stress and yield stress) decrease,while the elasticity parameters d and c (percent elongation and percent reduction of area) areimproved. However, these properties change nonmonotonically, and the variation dependson the tempering temperature intervals.Low-temperature tempering (1202508C; 2504808F) is used for treatment of highstrength structural and tool steels. Medium-temperature tempering (3504508C; 6608408F)is applied mainly to spring steels to achieve high elasticity. High-temperature tempering (4506508C; 84012008F) is widely used for products made of structural steels combining arelatively high strength with resistance to dynamic loads.Alloying of high-strength steels preserves high-strength characteristics up to 4008C(7508F). In steels containing additions of chromium, nickel, tungsten, and aluminum it ispossible to obtain a very favorable combination of strength (sB, s0.2), ductility (d, c), andimpact strength under low-temperature tempering (1602008C; 3203908F). In low-carbonmartensitic steels containing chromium, manganese, nickel, and molybdenum, the tensilestrength remains unchanged up to 4005008C (7509308F). In steels with secondary hardening (e.g., in steels with 0.26% C, 5% Cr, 1% Mo, 1.2% V, and 1.4% Si), strength and impactstrength increase under high-temperature tempering.These data suggest that mechanical properties of every type of steel exhibit certain specificfeatures that vary with the tempering temperature. These features are determined by theinfluence of alloying elements on the kinetics of phase transformations: change of themartensite point Ms, stabilization of retained austenite, and carbide formation.The influence of structural evolution on properties during tempering can be most fullyunderstood through the example of a maraging (martensite-aging) alloy containing 0.020.03% C and also Co, Mo, Ti, and Al. Alloying with cobalt increases the temperature Ms andprovides 100% martensite after cooling. Tensile strength reaches 10001100 MPa. The subsequent tempering (aging) at temperatures of 4505008C (8409308F) results in considerablestrengthening. Thus, sB can reach 19002100 MPa, s0.2 18002000 MPa, and d 810%.Such high-strength properties are due to the segregation of impurity atoms during aging(initial stages) and then to Ni3Ti, Ni3Mo, Fe2Mo, etc. phases coherently bound with thematrix. The size of particles is approximately 100 nm. Coagulation of the precipitates withan increase in temperature leads to lowering of the strength characteristics and increasing ofthe ductility. 2006 by Taylor & Francis Group, LLC.4.4.7 EMBRITTLEMENTDURINGTEMPERINGWhen carbon and alloy steels are tempered over the temperature interval of 2504008C (4807508F), a dramatic drop in impact strength is observed. If the steel is subjected to a highertemperature tempering and then tempering is repeated at 2504008C, the brittle state is notrecovered. Therefore this phenomenon has been called irreversible tempering brittleness. Suchtempering brittleness is typical of almost all carbon steels and alloys. High-temperature mechanical treatment and refinement of grains weaken this type of brittleness. In high-purity steels itdoes not occur at all. The embrittlement may be caused by nonuniform precipitation ofmartensite at the second stage of tempering. This structure has a lower resistance to dynamicloads. This effect is enhanced when the initial grain boundaries of austenite get saturated withimpurities under quenching heating. The alloying elements, which retard the second stage ofmartensite precipitation, shift the interval of irreversible brittleness toward higher temperatures.Another drop in impact strength is found at tempering temperatures of 4506008C (84011108F). A very significant feature of embrittlement is its reversibility under high-temperaturetempering. If a steel that has undergone tempering embrittlement is heated to a temperatureabove 6008C (11108F) and then cooled rapidly, its impact strength is restored. Therefore suchbrittleness is termed reversible.In the state of reversible tempering embrittlement, steel possesses a structure that consistsof ferrite and carbide. When subjected to impact tests, fracture occurs mainly along theboundaries of the initial austenite grains.Embrittlement over a certain temperature interval is typical not only of martensiticallyhardened steels. It also shows up, although to a lesser degree, in steels with the bainitestructure and is least pronounced in steels with the pearlite structure. Additions of chromium,nickel, and manganese facilitate tempering embrittlement. Small additions of molybdenum(not more than 0.20.3%) weaken tempering embrittlement. The presence of Sb, P, Sn, and Asin industrial steels makes these steels most susceptible to tempering embrittlement.4.5 HEAT TREATMENT OF SPECIAL CATEGORY STEELS4.5.1 HIGH-STRENGTH STEELSLow-alloy steels are most often used as construction materials. The combination of highstrength and ductility with high resistance to destruction is of particular importance for steels.The mechanical properties of such steels can be improved after hot rolling or normalizationand after quenching with tempering. Alloying makes it possible to perfect the properties ofsteels without using quenching with tempering because1. Properties of ferrite are changed when alloying elements are dissolved in it (solidsolution strengthening).2. Disperse strengthening phases precipitate in the process of cooling after hot rolling ornormalization.3. Steel grains and microstructure components become finer, and changes occur in themorphology and location of structural components.The overall content of alloying elements in low-alloy steels does not exceed 2.5%.In accordance with carbon content and principles of strengthening they can be divided intothree groups:1. Low-carbon steels (0.110.22% C) used in the hot-rolled or normalized states. Thermaltreatment of such steels (quenching and tempering) only slightly improves theirstrength characteristics. 2006 by Taylor & Francis Group, LLC.2. Low-carbon steels (0.050.18% C) strengthened by disperse precipitation of carbidesand carbonitrides are used in the normalized or hot-rolled states.3. Medium-carbon steels (0.250.50% C). The required level of properties is achieved insuch steels by quenching and high tempering.The main alloying element in these steels is manganese. Additional alloying of manganesesteels with Mo, Nb, and V results in formation of needle ferrite. Owing to their needlestructure, these steels combine high strength, high viscosity, and cold resistance. Therefore,the steels with 0.05% C, 2% Mn, 0.4% Mo, 0.010.02% Nb, or 0.07% V exhibit the followingstrength characteristics: sB 650750 MPa, s0.2 630 MPa, d 33%. These steels are moreoften used in the normalized state and seldom in the hot-rolled state.Low-alloy steels with improved strength characteristics include steels containing mainlymanganese and silicon. Tensile strength sB in these steels is more than 600 MPa and can be ashigh as 1800 MPa. Their ductility and viscosity depend on their carbon content and on thetypes of treatment. High-strength low-alloy steels with sB 1800 MPa are used in the hotrolled or cold-worked state. However, in this case, they are characterized by low impactstrength. In the normalized state their strength decreases to 800 MPa, and after quenchingand tempering to 600 MPa, the impact strength increasing simultaneously.4.5.2BORON STEELSAlloying of austenitic steels with rather high amounts of boron results in disperse hardening.The maximum hardness of such steels is attained upon quenching from 12308C (22508F) andtempering at 8008C (14708F). The strength and yield limits the increase simultaneously. At thesame time the viscosity of such steels decreases more than in the usual austenite steels. Theprecipitation of borides, because of the high temperature of disperse hardening (8008C;14708F), has a beneficial effect on the properties of refractory alloys (chromium, chromiumnickel, chromiumnickelcobalt alloys).At test temperatures up to 7008C (12908F) but still below the temperature of borideprecipitation, the refractory characteristics of steels appreciably improve even at a very lowboron content.In low-carbon steels with 1630% Cr and 6.530% Ni, the effect of disperse hardeningassociated with the presence of boron was not observed. But boron binds the elementsstabilizing austenite, thus favoring the formation of martensite. Carbon steels with0.20.3% C, on the other hand, are hardened considerably owing to the formation ofboridecarbide precipitates.Addition of boron to cemented steels improves their hardenability and increases thestrength of the core. Boron somewhat accelerates carburizing, but its influence on the caselessens with increasing carbon content. The greatest effect of boron was observed in steelswith 0.70.8% C.The influence of boron is enhanced as the quenching temperature is raised. However, thesensitivity of steel to overheating also increases. Therefore, boron steels usually contain smallquantities of titanium and vanadium, which have a favorable effect on the structure of steelswhen they are heated to high temperatures.4.5.3ULTRAHIGH-STRENGTH STEELSLow-carbon steels with a martensite structure have been developed recently that, upon cooling in air, undergo subsequent dispersion hardening at 4005008C (7509308F). The tensilestrength of such steels is in the range 22002500 MPa. As a rule, they are alloyed with 1218%Ni, up to 10% Cr, 35% Mo, and 0.61.0% Ti. These martensite-aging steels are distinguished 2006 by Taylor & Francis Group, having a low temperature of brittle fracture, very low sensitivity to cracks, and highstrength characteristics.The strengthening of martensite-aging alloys is a result of three processes: strengthening ofsubstitutional solid solution in the course of alloying, strengthening brought about by themartensite g ! a transformation, and strengthening connected with different stages of thesolid solution precipitation accompanied by formation of segregates and disperse particles ofmetastable and stable phases, the main contribution made by the second and third processes.As a consequence of the martensite g ! a transformation (during cooling in air), a finesubstructure with a high density of dislocations is formed. The particles of intermetallics50100 A in size are found at the stage of maximum strengthening. These particles arecoherently connected with the matrix.The high resistance of martensite-aging alloys to brittle fracture is determined by the highviscosity of the matrixthe low-carbon martensite alloyed with Ni and Co, which enhancethe mobility of dislocations. In addition, a high density of dislocations in martensite isresponsible for high dispersion and homogeneous distribution of phases precipitatedduring aging.Alloying with Mo suppresses the precipitation of particles of strengthening phases at thegrain boundaries and prevents intergrain brittle fracture. Alloying with Co increases themartensite start temperature Ms and ensures a 100% martensite structure. At the sametime, alloying with cobalt reduces the solubility of molybdenum and fosters dispersionhardening.At low-carbon content and moderate cooling rates, the martensite structure in martensiteaging alloys is obtained through relatively high degrees of alloying. At 1018% Ni, the pointMs lowers so significantly that the g transformation can be realized only according to themartensite mechanism.Compared with the high-strength manganese steels considered in Section 4.5.1, themartensite-aging steels are distinguished by a greater degree of alloying of the g-solidsolution. This promotes almost complete transformation of austenite to martensite. A widerange of alloying elements ensures a stronger solid-solution strengthening and increases thevolume fraction of disperse particles precipitated during phase aging. The above-mentionedthree factors are responsible for considerable enhancement of strength characteristics of lowalloy steels.The high-strength state of alloys can be obtained by using various external means to affecttheir structure. The most advantageous of these is low-temperature thermomechanical treatment (LTMT), which consists of deformation of the undercooled austenite in the region ofhigh stability and subsequent quenching. Undercooling of austenite is used to achievedeformation below the temperature of its recrystallization. Such treatment allows the attainment of advanced mechanical properties. The results gained at LTMT can be achieved bysuch factors as the composition of the steel, the temperature of austenitization, the rate ofcooling to the deformation temperature, the temperature of deformation and holding time atthis temperature, the degree and rate of deformation, the rate of cooling to room temperature,and final tempering conditions. The most important are the composition of steel, the temperature, and the degree of steel deformation.Deformation of the undercooled austenite should be completed prior to the beginning ofthe bainite transformation. In conformity with this, the steels undergoing LTMT shouldcontain austenite-stabilizing elements. LTMT strengthening is usually employed for highalloy steels with 17% Cr, 15% Ni, 0.5% V, 2.5% Mo, and 2% Si and sometimes withother additions as well. The strengthening of steels in LTMT depends on their carboncontent. The strengthening effect of carbon is more pronounced in LTMT than in conventional quenching. 2006 by Taylor & Francis Group, LLC.With an increasing amount of deformation the yield stress of steel increases continuously.When the thickness of a billet decreases by 1% in LTMT, by rolling, the tensile strength isincreased by 7 + 2 MPa.A decrease in deformation temperature results in a more intensive strengthening of steel inLTMT. The strength characteristics of steel after a small (up to 30%) deformation are lowerand less sensitive to changes in the deformation temperature than at high degrees of compression.At deformations of up to 2030%, LTMT leads to a sharp drop in ductility; with a rise inthe degree of compression above this value, the ductility begins to increase. There is a criticaldegree of deformation in LTMT above which the ductility of steel is sufficient. As thetemperature of deformation increases, the ductility also increases.In LTMT, the steel should be tempered as after the usual quenching. It is the opinion ofthe majority of researchers that the strengthening effect of LTMT is retained up to 3504008C(6607508F). If steel is alloyed with Mo, V, or W, the strengthening effect of LTMT persistsup to 5008C (9308F). The tensile strength of steel alloyed with tungsten is 2600 MPa afterquenching at 3508C (6608F) and 2450 MPa after quenching at 5008C (9308F). The rate ofcooling after deformation affects the properties of steel undergoing LTMT only if nonmartensitic structures are formed at insufficiently strong cooling.The study of the fine structure of alloys subjected to low-temperature thermal treatmenthas allowed us to explain the appearance of superhigh strength properties at rather satisfactory degrees of ductility by two structural factors: considerable reduction of size of martensitecrystals and changes in their morphology. This can be attributed to the emergence of acellular structure during deformation of undercooled austenite. The sites of dislocationpileups in austenite remain the sites where dislocations accumulate in martensite after thetransformation. Upon LTMT deformation, the fragmentation of austenite crystals results inthe fragmentation of the martensite structure. Individual fragments measuring fractions of amicrometer mutually disoriented through 10158 are joined with each other by dense dislocation pileups. These fragments, in turn, consist of 100200 A fragments disoriented relative toeach other through angles greater than 18.Thus, one of the possible mechanisms of strengthening in LTMT is connected with thecreation of a high density of structural imperfections in austenite as a result of deformationand the inheritance by martensite of the dislocation structure of the work-hardened austenite.This mechanism provides the most comprehensive explanation for the high strength ofmartensite obtained with LTMT.4.5.4MARTENSITIC STAINLESS STEELSPure iron and low-alloy steels are not resistant to corrosion in the atmosphere, water, or manyother media. The resistance of steel to corrosion can be enhanced by alloying it with variouselements. High strength of such steels is achieved primarily by quenching to obtain themartensite structure and through its subsequent aging.In martensitic stainless steels, the amount of martensite necessary for strengthening isformed after high-temperature heating and subsequent cooling to room temperature at arelatively small content of alloying components. The majority of alloying additions improvethe resistance of martensite by lowering the point Ms. The possibilities for anticorrosionalloying of martensitic steels are limited.In austeniticmartensitic steels (transition class), quenching does not lead to the completetransformation of austenite to martensite because of the low position of the point Ms.Consequently, no considerable increase in strength occurs. The degree of the g ! a transformation in these steels can be increased by means of (1) deep freezing treatment to 2006 by Taylor & Francis Group, LLC.temperatures below Mg; (2) plastic deformation below Ms; and (3) heating in the region of themost intensive precipitation of alloyed carbides from austenite (7007508C; 129013808F);when the matrix is depleted in alloying elements, the resistance of austenite decreases.The austeniticmartensitic steels admit a high degree of alloying and therefore affordmore possibilities for achieving total corrosion resistance and high strength. Such alloyingelements as copper, tungsten, nickel, molybdenum, silicon, and chromium lower the martensite point at direct g ! a transformation. The intensity of the influence of one or anotherelement depends on their combination.Cold plastic deformation initiates the martensite transformation. The less stable theaustenite and the lower the deformation temperature, the quicker the transformation. Thestrength of austeniticmartensitic and martensitic stainless steels increases with decreasingdeformation temperature.If after treatment these steels contain 7090% martensite, their yield stress can amountto 7001000 MPa and their tensile strength to 11001400 MPa. Further improvement ofstrength is achieved by aging the martensite. This enhancement of strength is attributed tosegregation of the GuinierPreston regions type.The effect of martensitic aging is observed when steels are alloyed with titanium, beryllium, aluminum, manganese, zirconium, niobium, copper, or certain other elements. Depending on the alloying elements, intermetallic phases of the types A3B (Ni3Ti, Ni3Al, Ni3Mn,Ni3Be), A2B [Fe2Mo(Fe, Ni, Co)2], or AB (NiTi, NiAl, NiMn) precipitate during aging.Greater strength values can be achieved if the deformation of steel upon quenchingproceeds below the temperature Mc1 under rather high compression. On the one hand, thisaccelerates the martensite transformation, and on the other hand, aging takes place in themartensite strengthened by deformation (sometimes also in the presence of the deformationstrengthened austenite). After complete thermal treatment, steels have the following characteristics: s0.2 8301200 MPa, sB 12001300 MPa.The conditions of quenching are set with allowance for complete dissolution of carbidessubject to the absence of excessive grain growth. Deep freezing treatment after quenchingensures a more complete transformation of austenite to martensite. The amount of martensitecan be as high as 7090%.The conditions of aging should provide the required set of mechanical properties andcorrosion resistance. The maximum strength values are attained as a result of aging in thetemperature range 4505008C (8409308F). At the same time, the best corrosion resistance isattained at the lower aging temperature range of 3502808C (6605408F) (high total corrosionresistance is obtained at the stage preceding precipitation of strengthening phases).4.5.5 PRECIPITATION-HARDENING STEELSAs is known, steels are classified into structural, spring, tool, and heat-resistant alloy steelsin conformity with their application. This section considers the behavior of precipitationhardening alloys in each of these groups. Structural SteelsLow-carbon manganese steels (0.10.2% C) containing 1.31.7% Mn, 0.100.20% V, about0.1% Ti, and ~0.05% Al can be classified as steels strengthened with disperse precipitates.Such compositions favor the formation of disperse precipitates. Such compositions favor theformation of disperse precipitates of vanadium and titanium carbonitrides or aluminumnitrides. These disperse precipitates can improve not only the strength of the steel but also,owing to grain refinement, its viscosity and cold resistance. A number of industrial alloys with 2006 by Taylor & Francis Group, LLC.carbonitride strengthening have been developed. Usually, these steels are used in the normalized state. Their properties are determined by the degree of dissolution of strengtheningphases in the process of heating.Another group of low-alloy precipitation-hardening steels includes low-pearlite steels,which contain up to 0.1% C and up to 2% Mn as well as vanadium (~0.1%), niobium(~0.06%), and sometimes molybdenum (~0.150.3%). Aluminum (up to 0.05%) can also bepresent in these steels. The properties of the steels under consideration are formed in theprocess of rolling during precipitation of disperse particles of the strengthening phase andgrain refinement. The conditions of rolling should ensure maximum dissolution of components that subsequently cause the formation of disperse particles. These particles strengthenferrite, which leads to grain refinement. The rolling temperature of low-pearlite steels dependson their composition and the strength and viscosity requirements. A high-heating temperature (~12008C; 21908F) ensures more complete dissolution of vanadium and niobium. Thiscontributes to the strengthening effect during precipitation of phases containing these elements. However, owing to the grain growth, the strength of these alloys is lower than in the caseof heating at 105011008C (192020108F).The temperature at the end of rolling of low-pearlite steels is usually reduced to 7008008C(130014758F). This is due to (1) a decrease in the austenite grain size; (2) an increase in thedegree of dispersion of the strengthening phase and, hence, enhancement of the hardeningeffect; and (3) displacement of the g ! a transformation to the region of lower temperatures,which results in a finer ferrite grain. SteelsAlloys based on FeNi, FeNiCr, CoNiCr, NiCr, and other systems, predominantly withtitanium and aluminum or niobium additions, are used for spring steels strengthened byprecipitation hardening. The particles of strengthening phases in these alloys precipitateduring aging (tempering). Additional improvement of the strength properties of these alloyscan be achieved through plastic deformation between quenching and aging. In this case theprecipitation of the supersaturated solid solution may proceed according to a discontinuousmechanism. If the discontinuous precipitation cells completely occupy each grain (which ispossible for a very fine grain structure), a very strong strengthening of alloys takes place. Inthe process of aging, additional refinement of the initial grain occurs during discontinuousprecipitation.Strengthening is observed in alloyed martensite-aging steels under developing of disperseparticles of precipitating phases. A large number of steels differing in composition and properties are used in the industry. In addition to 0.40.8% C, they contain at least two of suchalloying elements as Si, Cr, V, Mo, Mn, more rarely Ni and W. Isothermal quenching withsubsequent tempering is advantageous for these steels, especially for those containing silicon.The maximum elastic limit in alloyed steels is attained with tempering at 3003508C (5706608F). These tempering conditions correspond to the conditions of a sufficiently completeprecipitation of austenite accompanied by preservation of a high density of dislocations, sincedisperse particles of carbides hammer the redistribution and annihilation of dislocations. Inaddition, the carbide particles increase the resistance to low-plastic deformation.For carbon steels, the amount of carbides can be increased and the martensite point can belowered owing to the higher carbon content. This brings about a significant improvement ofthe strength characteristics of such steels, the highest properties achieved when strongcarbide-forming elements (e.g., vanadium) enter into their composition.Martensite-aging alloys containing nickel and titanium possess the best set of properties.Such alloys are quenched from 870 to 11508C (160021008F) depending on their titanium 2006 by Taylor & Francis Group, LLC.content; the greater the titanium content, the higher the quenching temperature. The finer thegrain, the higher the properties of these steels. Fine grain can be attained either by multipleg ! a transformations or through deformation and recrystallization processes. To reduce thequantity of austenite retained upon quenching, a deep freezing treatment (708C; 948F)is employed. Then comes aging at 4508C (8408F) for 6 h, during which NiTi or Ni3Tiphases precipitate.Austenitic steels are also strengthened as a result of precipitation hardening. They maycontain chromium, nickel, titanium, or molybdenum. Upon quenching, these alloys have thestructure of the g-solid solution with chromium, titanium, and titanium carbonitride inclusions. The properties of aged alloys depend on the quenching temperature, which determinesthe degree of supersaturation of the solid solution, and on the cooling rate, which should be ashigh as possible. Aging may proceed by the discontinuous or continuous mechanism. Tool SteelsTool steels strengthened by precipitation-hardening tempering on the basis of the initialmartensite structure are used to manufacture dies for cold deformation of steels. As a resultof tempering, the hardness and strength characteristics of steels are enhanced when strengthening phases (carbides) precipitate from martensite. The retained austenite, a phase with lowhardness, transforms to martensite. These processes increase the yield stress under compression but reduce viscosity.Precipitation-hardening strengthening is also characteristic of heat-resistant steels. Thestructure of these steels represents a martensite matrix with particles of the strengtheningphasescarbides or intermetallicsprecipitating during tempering.The main principles of heat treatment of precipitation-hardening tool steels are nowconsidered in great detail. The basic operations of heat treatment are annealing, quenching,and tempering.The annealing heating temperature is chosen a little higher than A1. It is 7607808C(140014358F) for carbon steels, 7808108C (143514908F) for alloy steels, and 8308708C(152516008F) for high-alloy chromium steels, with 23 h as the holding time .Quenching of tool steels is aimed at obtaining martensite with a high concentration ofcarbon and alloying components with retained fine-grain structure. That is why quenching iscarried out at temperatures corresponding to complete dissolution of the basic carbides inaustenite. These temperatures, however, should not be conducive to austenite grain growth.Usually, the quenching temperature corresponds to the temperature of heating; it is alittle higher than A1 for steels in which the main carbide phase is cementite; up to 100010608C (183219408F) for steels with a chromium-based carbide phase of types Me7C3 andMe23C6, and 108011008C (197520108F) for steels with greater carbide content of thetype Me23C6.At tempering, the assigned level of properties is achieved by changing the structure of thequenched steel. Heating a quenched steel during tempering to 1502008C (3003908F) causesthe precipitation of small e-carbide plates from martensite and reduces the carbon concentration. Such tempering only slightly impairs the steels hardness but significantlyimproves its strength and viscosity. Heating to 2502808C (4805358F) during temperingnoticeably decreases the carbon concentration in martensite and enhances the strength andviscosity characteristics of the steel. This tempering permits almost complete removal of theretained austenite.An appreciable increase in the hardness of steels results from the precipitation of a largenumber of small carbide particles (intermetallics of alloying elements of the types Me2C,Me23C6, MeC, and Me7Me6) from martensite. 2006 by Taylor & Francis Group, LLC. AlloysA vast group of heat-resistant alloys consists of austenitic steels strengthened with carbidesand intermetallics. To increase heat resistance, elements that strengthen the solid solution andinduce precipitation hardening are introduced into ironnickel-based alloys. These elementsinclude Cr, Mo, W, Nb, V, Ti, and Al. To acquire high heat resistance, the alloys undergodouble quenching. The purpose of the first quenching is to obtain grains of a certain size andto transform the excess g0 -phase [intermetallic phases g0 -Ni3(Al, Ti, B)] to a solid solution.Quenching is followed by cooling in air. The g0 -phase partially precipitates in this process.The aim of the second quenching (10508C; 19208F) with subsequent aging is to obtaindisperse precipitates of the g0 -phase 200500 A in diameter. As a result of these quenchingprocedures, the strengthened alloy contains a certain number of larger precipitates along withfine inclusions. This structure ensures high strength and the necessary margin of ductility.4.5.6TRANSFORMATION-INDUCED PLASTICITY STEELSThere are numerous examples for improving the plasticity of load-bearing samples under theinfluence of phase transformations of the diffusion and shear types. The term transformationinduced plasticity was proposed to denote an improvement of plasticity under martensitetransformation. High plasticity of steels below the critical point was called subcritical superplasticity by A.P. Gulyaev.The Transformation-induced plasticity (TRIP) effect appears under the action of high stressesthat exceed the yield stress of austenite. In the segment where localization of flow sets in, martensitedeformation occurs. This segment is stronger than austenite, and because of this the flow extendsto the neighboring segments of the sample. Thus, the quasiequilibrium flow in TRIP steels is due tohigh deformation strengthening. The index of the flow stress rate sensitivity remains low.TRIP is observed at fixed test temperatures. In the case of isothermal transformations, thevolume that undergoes such a transformation reaches a certain level and does not increasefurther. Therefore, the greatest overall effects of plasticity improvement are observed at cyclictemperature changes that lead to multiple occurrences of the phase transformation.TRIP is found when the temperature of the sample is changed (within limits exceeding thetemperature range of the phase transformation), with a constant load. The load applied isusually lower than the yield stress of any of the phases involved in the transformation.In each temperature-changing cycle, the value of deformation is in tenths of a percent.With a large number of cycles it may amount to several hundred percent. The deformationvalue in one cycle is directly proportional to the applied stress. An increase in the appliedstress above a certain limit disrupts the linear dependence of deformation (per cycle) on stress.This is caused by transition from the plastic deformation to the usual deformation atcomparatively high stresses. As the volumetric effect of phase transformation increases, thedeformation value per cycle increases.TRIP can be observed in metals with any grain size, among them coarse-grain metals,and at any temperature, including low temperatures. For example, after 150 temperaturechanging cycles in the range of 2046488C (40012008F), a sample of Fe15.4% Ni alloy withan initial grain size of 150 mm became 160% longer, with no neck formed, owing to thereversible martensite transformation under load.The deformation mechanisms typical of TRIP have not been clearly established because ofdifficulties in using direct structural methods during phase transformation when the structureof a sample changes constantly. Among the proposed hypotheses the following may be quoted:1. Accelerated transfer of dislocations owing to an excess of vacancies formed duringvolumetric changes 2006 by Taylor & Francis Group, LLC.2. Weakening of bonding forces between atoms at the interface at the moment of transformation3. Changes in the form associated with realization of particular orientations of the formedmartensite4. Summation over the phase and applied stress, which determines the plastic deformationof the weaker phase (ferrite in the case of a!g transformation)5. Formation of ultrafine grain in the course of phase transformation4.5.7 TOOL STEELSTool steels can be classified into four groups according to their application: (1) steels forcutting tools used in mild conditions; (2) steels for cutting tools used in severe conditions; (3)measuring tools; and (4) die steels.Steels for cutting tools must have high hardness exceeding 60Rc. Therefore, such toolsteels contain a minimum of 0.6% C. The main requirement imposed on steels used in severeconditions (high-speed steels) is stable hardness under long heating. All tool steels fall into fourcategories: (1) carbon tool steels, (2) alloy tool steels, (3) die steels, and (4) high-speed steels. Carbon Tool SteelsCarbon tool steels contain 0.600.74% C, 0.250.35% Mn, and 0.30% Si. The quenchingtemperature of these steels is chosen in conformity with the FeC equilibrium diagram. Thetetrahedral structure of martensite and internal stresses in quenched steels bring aboutconsiderable brittleness. That is why tempering after quenching is an obligatory operation.The tempering temperature is determined by the required working hardness of the tools.Usually it ranges between 180 and 2408C (350 and 4658F).Of great importance in terms of machinability is the structure of annealed steels. Steelswith the structure of lamellar pearlite are difficult to machine. Therefore, with the help ofannealing at a temperature slightly above Ac1, easily worked steels with the structure ofglobular pearlite are obtained. As a rule, carbon steels are quenched in water. Because of this,tools made of such steels have a soft unannealed core and are less brittle than tools made ofthrough-hardened steels. Alloy Tool SteelsCompared with carbon steels, alloy tool steels possess greater hardenability and wear resistance. This is achieved by the introduction of small quantities of alloying elements, predominantly chromium. For chromium steels, it is imperative that quenching be accompanied bysubsequent tempering. If it is necessary to preserve hardness at the level of the quenched state,the tempering temperature should not exceed 1501708C (3003408F).In all cases where quenching should be accompanied by minimum deformation during thepearlite ! martensite transformation (pearlite is the initial structure in this process), lowdeformation tool steels are used. Such steels can be obtained by alloying with elements thatincrease the amount of retained austenite in the quenched state, namely, chromium andmanganese. These steels contain about 12% Cr and ~1.5% C. The formation of a largeamount of carbides (Cr,Fe)7C3 significantly improves their wear resistance.These high-chromium steels belong to the ledeburitic class. In the cast state, the initialcarbides form the eutectic ledeburite. In forging, the eutectic breaks down and the structure ofthe steels consists of sorbite-forming pearlite with inclusions of excess carbides. When heatedfor quenching, the carbides dissolve in austenite. The highest hardness of the steel is achievedupon quenching at ~10508C (~19208F). 2006 by Taylor & Francis Group, LLC.To obtain high hardness, the steel is quenched in oil. The retained austenite precipitates inthe process of cold treatment and tempering. Owing to the greater stability of martensitecompared to other steels, the tempering temperature is increased to 2002208C (3904308F). SteelsDies operating in the cold state need high-hardness steels. The steel to be used in hot pressingshould have low sensitivity to local heating. Different grades of steelsfrom carbon tocomplex alloy steelsare used in the production of dies. Carbon steel is used for diesoperating under mild conditions and alloy steel for dies operating under severe conditions.Carbon steel contains 0.61% C. Alloy steel includes 0.30.7% C, Cr, Si, and sometimes Ni.Dies made of alloy steels are usually quenched from ~8508C (15608F) in oil with subsequent tempering at 5005508C (93010208F). Hardness of the steel amounts to 350400HB.Dies for cold pressing are quenched in the temperature range of 86010508C (158019208F)(depending on the steel grade) in oil with subsequent tempering at 2003008C (3905708F).Depending on the tempering temperature and the steel grade, the steel hardness is withinRc 5662. SteelsHigh-speed steels must not only possess high hardness in the hot state, but also be able toretain it during long heating (red hardness). To preserve hardness during heating, it isnecessary to hamper the process of carbide coagulation. For this purpose, special carbidesshould be formed. Such carbides can be produced if the steel is alloyed with 3% Cr. Thespecial carbide Cr7C3 coagulates at high temperatures to a lesser degree than cementite.Noticeable precipitation and coagulation of special Cr, Mo, W, and V carbides occur attemperatures over 5008C (9308F).All high-speed steels are rated in the ledeburite class and in the cast state have thestructure of white hypoeutectic cast iron. As a result of forging, the structure of the highspeed steel changes and the eutectic is broken down into individual carbides. In the annealedstate, three types of carbides are observed: coarse primary carbides, smaller secondarycarbides, and fine-grain carbides entering into the composition of upper bainite. Ferrite,which is found in upper bainite, also contains some alloying impurities.Heating of the high-speed steel to the point Ac1 (8008508C; 147015608F) is not accompanied by structural changes. Above this point, the eutectoid transforms to austenite, the secondarycarbides dissolve in the austenite, and it is saturated with carbon and alloying elements.Solubility of carbides depends on how long the steel is held at the quenching temperature.With an increase in the holding time, there is more complete dissolution of carbidesin austenite.Carbon and alloying elements contained in austenite lower the martensite point andincrease the content of retained austenite. At quenching temperatures above 10008C(18328F) the martensite point decreases to 08C (328F) or lower. This peculiarity is takenadvantage of in the heat treatment of tools made of high-speed steels.In the process of steel tempering, the following structural changes take place. Heating to1002008C (2123908F) causes a small compression, since the tetragonal martensite transforms to the cubic modification. At 3004008C (5707508F), hardness deteriorates owing to adecrease in the work hardening of retained austenite. At 5006008C (93011108F), finelydisperse carbides precipitate from austenite. Cooling the steel from these temperatures bringsabout the secondary formation of martensite: depleted austenite transforms to martensite inlarger quantities. 2006 by Taylor & Francis Group, LLC.The higher the tempering temperature or the longer the tempering time, the greater theamount of retained austenite transformed to martensite. Complete transformation of austenite can be attained by multiple tempering. The microstructure of the quenched and temperedsteel should consist of finely dispersed martensite and carbides.The quenching temperature of high-speed steel should be as high as possible, but at thesame time it should not allow intensive grain growth (126012808C; 230023408F). Duringquenching the steel may be cooled comparatively slowly owing to a low-critical rate ofquenching (in air or oil). Tempering is an obligatory operation and is usually realized at5605808C (104010758F) for 3 h. To obtain still better properties, two- or threefold tempering is used, with the holding time at each stage at least for 1 h.FURTHER READINGBelous, M.V., Cherepin, V.T., and Vasiliev, M.A., Transformations During Tempering of Steel, Metallurgiya, Moscow, 1973.Bernshtein, M.L. and Richshtadt, A.G. (Eds.), Physical Metallurgy and Thermal Treatment of Steels,Handbook, Vols. I, II, and III, 3rd issue, Metallurgiya, Moscow, 1983.Blanter, M.E., Phase Transformations during Thermal Treatment of Steel, Metallurgiya, Moscow, 1962.Blanter, M.E., Physical Metallurgy and Thermal Treatment, Mashinostroyeniye, Moscow, 1963.Delle, V.A., Structural Alloy Steel, Metallurgiya, Moscow, 1959.Goldshtein, M.I., Grachev, S. V., and Veksler, Yu. G., Special Steels, Metallurgiya, Moscow, 1985.Gudreman, E., Special Steels, Vols. I and II, Metallurgiya, Moscow, 1959.Gulyaev, A.P., Physical Metallurgy, Metallurgiya, Moscow, 1976.Gulyaev, A.P., Pure Steel, Metallurgiya, Moscow, 1975.Kaschenko, G.A., Fundamentals of Physical Metallurgy, Metallurgiya, Moscow, 1964.Kurdyumov, G.V., Utevski, L.M., and Entin, R.I., Transformations in Iron and Steel, Nauka,Moscow, 1977.Meskin, V.S., Fundamentals of Steel Alloying, Metallurgiya, Moscow, 1964.Novikov, I.I., Theory of Thermal Treatment of Metals, Metallurgiya, Moscow, 1986.Popov, A.V., Phase Transformations of Metal Alloys, Metallurgiya, Moscow, 1963.Vinograd, M.I. and Gromova, G.P., Inclusions in Alloy Steels and Alloys, Metallurgiya, Moscow, 1972.Zimmerman, R. and Gunter, K., Metallurgy and Materials Science Handbook, Metallurgiya,Moscow, 1982. 2006 by Taylor & Francis Group, LLC.5Hardenabilityc idar Lis icBozCONTENTS5.15.25.3Definition of Hardenability ....................................................................................... 213Factors Influencing Depth of Hardening................................................................... 215Determination of Hardenability................................................................................. 2175.3.1 Grossmanns Hardenability Concept.............................................................. 2175.3.1.1 Hardenability in High-Carbon Steels.............................................. 2205.3.2 Jominy End-Quench Hardenability Test ........................................................ 2285.3.2.1 Hardenability Test Methods for Shallow-Hardening Steels ........... 2305.3.2.2 Hardenability Test Methods for Air-Hardening Steels................... 2335.3.3 Hardenability Bands ....................................................................................... 2375.4 Calculation of Jominy Curves from Chemical Composition ..................................... 2405.4.1 Hyperbolic Secant Method for Predicting Jominy Hardenability .................. 2435.4.2 Computer Calculation of Jominy Hardenability ............................................ 2475.5 Application of Hardenability Concept for Prediction of Hardness after Quenching..... 2495.5.1 Lamont Method ............................................................................................. 2535.5.2 Steel Selection Based on Hardenability .......................................................... 2565.5.3 Computer-Aided Steel Selection Based on Hardenability .............................. 2575.6 Hardenability in Heat Treatment Practice ................................................................. 2645.6.1 Hardenability of Carburized Steels................................................................. 2645.6.2 Hardenability of Surface Layers When Short-Time Heating MethodsAre Used......................................................................................................... 2665.6.3 Effect of Delayed Quenching on the Hardness Distribution .......................... 2675.6.4 A Computer-Aided Method to Predict the Hardness Distribution afterQuenching Based on Jominy Hardenability Curves ....................................... 2685.6.4.1 Selection of Optimum Quenching Conditions ................................ 273References .......................................................................................................................... 2755.1 DEFINITION OF HARDENABILITYHardenability, in general, is defined as the ability of a ferrous material to acquire hardnessafter austenitization and quenching. This general definition comprises two subdefinitions: theability to reach a certain hardness level (German: Aufhartbarkeit) and the hardness distribution within a cross section (German: Einhartbarkeit).The ability to reach a certain hardness level is associated with the highest attainablehardness. It depends first of all on the carbon content of the material and more specifically onthe amount of carbon dissolved in the austenite after the austenitizing treatment, because onlythis amount of carbon takes part in the austenite-to-martensite transformation and has relevantinfluence on the hardness of martensite. Figure 5.1 shows the approximate relationship betweenthe hardness of the structure and its carbon content for different percentages of martensite [1]. 2006 by Taylor & Francis Group, LLC.80MartensiteHardness, HRC6099.9%90%H max50%40200HRC99.9 = 35 + 50 . %CHRC90 = 30 + 50 . %CHRC50 = 23 + 50 . %C00. contentFIGURE 5.1 Approximate relationship between hardness in HRC and carbon content for differentpercentages of martensite. (From G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser, Munich, 1987, p. 1012.)The hardness distribution within a cross section is associated with the change of hardnessfrom the surface of a specified cross section toward the core after quenching under specifiedconditions. It depends on carbon content and the amount of alloying elements dissolved inthe austenite during the austenitizing treatment. It may also be influenced by the austenitegrain size. Figure 5.2 shows the hardness distributions within the cross sections of bars of100 mm diameter after quenching three different kinds of steel [2].In spite of quenching the W1 steel in water (i.e., the more severe quenching) and the othertwo grades in oil, the W1 steel has the lowest hardenability because it does not containalloying elements. The highest hardenability in this case is that of the D2 steel, which has thegreatest amount of alloying elements.70AISI D2Hardness, HRC6050AISI 0140AISI W1302000101/2203011/2Depth below surface14050 mm2 in.FIGURE 5.2 Hardness distributions within cross sections of bars of 100 mm diameter for three differentkinds of steel, after quenching. Steel W1 was water-quenched; the rest were oil-quenched. (FromK.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984, p. 145.) 2006 by Taylor & Francis Group, LLC.When a steel has high hardenability it achieves a high hardness throughout the entireheavy section (as D2 in Figure 5.2) even when it is quenched in a milder quenchant (oil).When a steel has low hardenability its hardness decreases rapidly below the surface (as W1 inFigure 5.2), even when it is quenched in the more severe quenchant (water).According to their ability to reach a certain hardness level, shallow-hardening highcarbon steels may reach higher maximum hardness than alloyed steels of high hardenabilitywhile at the same time achieving much lower hardness values across a cross section. This canbe best compared by using Jominy hardenability curves (see Section 5.3.2). Hardenability isan inherent property of the material itself, whereas hardness distribution after quenching(depth or hardening) is a state that depends on other factors as well.5.2 FACTORS INFLUENCING DEPTH OF HARDENINGDepth of hardening is usually defined as the distance below the surface at which a certainhardness level (e.g., 50 HRC) has been attained after quenching. Sometimes it is defined as thedistance below the surface within which the martensite content has reached a certain minimum percentage.As a consequence of the austenite-to-martensite transformation, the depth of hardeningdepends on the following factors:1. Shape and size of the cross section2. Hardenability of the material3. Quenching conditionsQuenching conditions include not only the specific quenchant with its inherent chemicaland physical properties, but also important process parameters such as bath temperature andagitation rate.The cross-sectional shape has a remarkable influence on heat extraction during quenchingand consequently on the resulting hardening depth. Bars of rectangular cross sections alwaysachieve less depth of hardening than round bars of the same cross-sectional size. Figure 5.3 isa diagram that can be used to convert square and rectangular cross sections to equivalentcircular cross sections. For example, a 38-mm square and a 25 100-mm rectangular crosssection are each equivalent to a 40-mm diameter circular cross section; a 60 100-mmrectangular cross section is equivalent to an 80-mm diameter circle [2].The influence of cross-sectional size when quenching the same grade of steel under thesame quenching conditions is shown in Figure 5.4A. Steeper hardness decreases from surfaceto core and substantially lower core hardness values result from quenching a larger crosssection.Figure 5.4B shows the influence of hardenability and quenching conditions by comparingan unalloyed (shallow-hardening) steel to an alloyed steel of high hardenability when each isquenched in (a) water or (b) oil. The critical cooling rate (ncrit) of the unalloyed steel is higherthan the critical cooling rate of the alloyed steel. Only those points on the cross section thathave been cooled at a higher cooling rate than ncrit could transform to martensite and attainhigh hardness. With unalloyed steel this can be achieved up to some depth only by quenchingin water (curve a); oil quenching (curve b) provides essentially no hardness increase. Withalloyed steel, quenching in water (because of the high cooling rate of water) produces acooling rate greater than ncrit even in the core, resulting in through-hardening. Oil quenching(curve b) provides, in this case, cooling rates higher than ncrit within quite a large depth ofhardening. Only the core region remains unchanged. 2006 by Taylor & Francis Group, mm2402508 200mm in.210200180250 1024019018017071602302202102001901801701601501401301201101009080706050403020161601506Thicknessmm f24023022022014014051301201101001204 10090808037060605024040301201620002014060280310041205140160618020078220 240 260 280910119876Diameter954321300 mm12 in.BreadthFIGURE 5.3 Correlation between rectangular cross sections and their equivalent round sections,according to ISO. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London,1984, p. 145.)4020(B)UnalloyedsteelAlloyedsteelncritaabbncrit0Surf. CoreHardness, HRC10 mm10 mmMartensitecontent93%90%50%60Cooling rateHardness, HRC(A)6040Martensitecontent99%90%50%aabb200Surf.CoreFIGURE 5.4 Influence of (A) cross-sectional size and (B) hardenability and quenching conditions onthe depth of hardening. (a) Water quenching; (b) oil quenching, ncrit, critical cooling rate. (FromG. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser, Munich,1987, p. 1012.) 2006 by Taylor & Francis Group, LLC.5.3 DETERMINATION OF HARDENABILITY5.3.1 GROSSMANNS HARDENABILITY CONCEPTGrossmanns method of testing hardenability [3] uses a number of cylindrical steel bars ofdifferent diameters hardened in a given quenching medium. After sectioning each bar atmidlength and examining it metallographically, the bar that has 50% martensite at its center isselected, and the diameter of this bar is designated as the critical diameter (Dcrit). Thehardness value corresponding to 50% martensite will be determined exactly at the center ofthe bar of Dcrit. Other bars with diameters smaller than Dcrit have more than 50% martensitein the center of the cross section and correspondingly higher hardness, while bars havingdiameters larger than Dcrit attain 50% martensite only up to a certain depth as shown inFigure 5.5. The critical diameter Dcrit is valid for the quenching medium in which the barshave been quenched. If one varies the quenching medium, a different critical diameter will beobtained for the same steel.To identify a quenching medium and its condition, Grossmann introduced the quenchingintensity (severity) factor H. The H values for oil, water, and brine under various rates ofagitation are given in Table 5.1[4]. From this table, the large influence of the agitation rate onthe quenching intensity is evident.To determine the hardenability of a steel independently of the quenching medium,Grossmann introduced the ideal critical diameter DI, which is defined as the diameter of agiven steel that would produce 50% martensite at the center when quenched in a bath ofquenching intensity H 1. Here, H 1 indicates a hypothetical quenching intensity thatreduces the surface temperature of the heated steel to the bath temperature in zero time.Grossmann and his coworkers also constructed a chart, shown in Figure 5.6, that allows theconversion of any value of critical diameter Dcrit for a given H value to the correspondingvalue for the ideal critical diameter (DI) of the steel in question [2].For example, after quenching in still water (H 1.0), a round bar constructed of steel Ahas a critical diameter (Dcrit) of 28 mm according to Figure 5.6. This corresponds to an idealcritical diameter (DI) of 48 mm. Another round bar, constructed of steel B, after quenching inoil (H 0.4), has a critical diameter (Dcrit) of 20 mm. Converting this value, using Figure 5.6,provides an ideal critical diameter (DI) of 52 mm. Thus, steel B has a higher hardenabilitythan steel A. This indicates that DI is a measure of steel hardenability that is independent ofthe quenching medium.Hardness, HRC6040HRCcrit 50% MD crit200f80f60f50f40FIGURE 5.5 Determination of the critical diameter Dcrit according to Grossmann. (From G. Spur (Ed.),Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser, Munich, 1987, p. 1012.) 2006 by Taylor & Francis Group, LLC.TABLE 5.1Grossmann Quenching Intensity Factor HH Value (in.21)Method of QuenchingOilBrine0.250.300.300.350.350.400.400.500.500.800.801.10No agitationMild agitationModerate agitationGood agitationStrong agitationViolent agitationWater1. 08048320400.4032Steel A0.2024Steel B160.1080Quenching intensity H80 120 160 200 240 280Ideal critical diameter D I, mm2.400Quenching intensity H1600Critical diameter Dcrit, mm0.0.10105. .00Critical diameter Dcrit, mm2405. Metals Handbook, 8th ed., Vol. 2, American Society for Metals, Cleveland, OH, 1964, p. 18.0.0108162432404856Ideal critical diameter D I, mm6472FIGURE 5.6 The chart for converting the values of the critical diameter Dcrit into the ideal criticaldiameter DI, or vice versa, for any given quenching intensity H, according to Grossmann and coworkers.(From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984, p. 145.) 2006 by Taylor & Francis Group, LLC.0.400.3810.0Grain sizeASTM0.360.349.59.040.328.58.060.2676.50.2486. , mm50.28DI , in.0.304.50.160 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9Carbon content, %FIGURE 5.7 The ideal critical diameter (DI) as a function of the carbon content and austenite grain sizefor plain carbon steels, according to Grossmann. (From K.E. Thelning, Steel and its Heat Treatment,2nd ed., Butterworths, London, 1984, p. 145.)If DI is known for a particular steel, Figure 5.6 will provide the critical diameter of thatsteel for various quenching media. For low- and medium-alloy steels, hardenability asdetermined by DI may be calculated from the chemical composition after accounting foraustenite grain size. First, the basic hardenability of the steel as a function of carbon contentand austenite grain size is calculated from Figure 5.7 according to the weight percent of eachelement present. For example: if a steel has an austenite grain size of American Society forTesting and Materials (ASTM) 7 and the chemical composition C 0.25%, Si 0.3%, Mn 0.7%,Cr 1.1%, Mo 0.2%, then the basic value of hardenability from Figure 5.7 (in inches) isDI 0.17. The total hardenability of this steel isDI 0:17 1:2 3:3 3:4 1:6 3:7 in:(5:1)For these calculations, it is presumed that the total amount of each element is in solution atthe austenitizing temperature. Therefore the diagram in Figure 5.8 is applicable for carboncontents above 0.8% C only if all of the carbides are in solution during austenitizing. This isnot the case, because conventional hardening temperatures for these steels are below thetemperatures necessary for complete dissolution of the carbides. Therefore, decreases in thebasic hardenability are to be expected for steels containing more than 0.8% C, compared tovalues in the diagram. Later investigations by other authors produced similar diagrams thataccount for this decrease in the basic hardenability that is to be expected for steels with morethan 0.8% C, compared to the values shown in Figure 5.8 [6]. Although values of DIcalculated as above are only approximate, they are useful for comparing the hardenabilityof two different grades of steel.The most serious objection to Grossmanns hardenability concept is the belief that theactual quenching intensity during the entire quenching process can be described by a single Hvalue. It is well known that the heat transfer coefficient at the interface between the metal 2006 by Taylor & Francis Group, LLC.Multiplying factorMultiplying factor3.88.4Mn3.47.6Cr3.0Ni6.82.66.0Mo2.25.2Si1.84.4Mn(continued)1.40.8 1.2 1.6 content, %FIGURE 5.8 Multiplying factors for different alloying elements when calculating hardenability asDI value, according to AISI. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984, p. 145.)surface and the surrounding quenchant changes dramatically during different stages of thequenching process for a vaporizable fluid.Another difficulty is the determination of the H value for a cross-sectional size otherthan the one experimentally measured. In fact, H values depend on cross-sectional size [7].Figure 5.9 shows the influence of steel temperature and diameter on H values for an 18Cr8Niround bar quenched in water from 8458C [7]. It is evident that the H value determined in thisway passed through a maximum with respect to terminal temperatures. It is also evident thatH values at the centers of round bars decreased with increasing diameter.Values of the quenching intensity factor H do not account for specific quenchantand quenching characteristics such as composition, oil viscosity, or the temperature of thequenching bath. Table of H values do not specify the agitation rate of the quenchant eitheruniformly or precisely; that is, the uniformity throughout the quench tank with respect tomass flow or fluid turbulence is unknown. Therefore, it may be assumed that the tabulated Hvalues available in the literature are determined under the same quenching conditions. Thisassumption, unfortunately, is rarely justified.In view of these objections, Siebert et al. [8] state: It is evident that there cannot be asingle H-value for a given quenching bath, and the size of the part should be taken intoaccount when assigning an H-value to any given quenching bath. in High-Carbon SteelsThe hardenability effect of carbon and alloying elements in high-carbon steels and the caseregions of carburized steels differ from those in low- and medium-carbon steels and areinfluenced significantly by the austenitizing temperature and prior microstructure (normalized or spheroidize-annealed). Using Grossmanns method for characterizing hardenabilityin terms of the ideal critical diameter DI, multiplying factors for the hardenability effects ofMn, Si, Cr, Ni, Mo, and Al were successfully derived [9] for carbon levels ranging from 0.75to 1.10% C in single-alloy and multiple-alloy steels quenched at different austenitizingtemperatures from 800 to 9308C. These austenitizing temperatures encompass the hardeningtemperatures of hypereutectoid tool steels, 1.10% C bearing steels, and the case regions of 2006 by Taylor & Francis Group, LLC.Temperature, C3005007003.2H value, in.12.8A 1/2-in. (13-mm) roundB 1-in. (25-mm) roundC 1-1/2-in. (38-mm) roundD 2-1/4-in. (57-mm) roundE 3-in. (76-mm) roundWater temperature 60 F (16 C)Center couplesA2. value, mm11000.051.20.04D0.80.03E0., FFIGURE 5.9 Change of the H value with temperature and size of the round bar. Calculated from coolingcurves measured at the center of bars made of 18Cr8Ni steel quenched in water from 8458C, accordingto Carney and Janulionis. (From D.J. Carney and A.D. Janulionis, Trans. ASM 43:480496, 1951.)carburized steels. All of these steels, when quenched, normally contain an excess of undissolved carbides, which means that the quantity of carbon and alloying elements in solutioncould vary with the prior microstructure and the austenitizing conditions. The hardenabilityof these steels is influenced by the carbide size, shape, and distribution in the prior microstructure and by austenitizing temperature and time. Grain size exhibits a lesser effect becausehardenability does not vary greatly from ASTM 6 to 9 when excess carbides are present.As a rule, homogenous high-carbon alloy steels are usually spheroidize-annealed formachining prior to hardening. Carburizing steel grades are either normalized, i.e., air-cooled,or quenched in oil directly from the carburizing temperature before reheating for hardening.So different case microstructures (from martensite to lamellar pearlite) may be present, all ofwhich transform to austenite rather easily during reheating for hardening. During quenching,however, the undissolved carbides will nucleate pearlite prematurely and act to reduce hardenability.In spheroidize-annealed steel, the carbides are present as large spheroids, which are muchmore difficult to dissolve when the steel is heated for hardening. Therefore the amount ofalloy and carbon dissolved is less when one starts with a spheroidized rather than a normalized or quenched microstructure. Nevertheless, it has been demonstrated that a spheroidizedprior microstructure actually yields higher hardenability than a prior normalized microstructure, at least for austenitizing temperatures up to approximately 8558C. This effect occursbecause larger carbides are not as efficient nuclei for early pearlite formation upon cooling asfine and lamellar carbides and the nuclei are present in lower numbers. With either priormicrostructure, if strict control is maintained over austenitizing temperature and time, thesolution of carbon and alloy can be reproduced with sufficient consistency to permit the 2006 by Taylor & Francis Group, LLC.Indicated hardenability D I7650% Martensite5495% Martensite321099.9% Martensite123456Hardenability D I, 50% martensite7FIGURE 5.10 Average relationships among hardenability values (expressed as DI) in terms of 50, 95,and 99.9% martensite microstructures. (From Metals Handbook, ASM International, Cleveland,OH, 1948, p. 499.)derivation of multiplying factors. For all calculations, it was important to establish whetherpearlite or bainite would limit hardenability because the effects of some elements on thesereactions and on hardenability differ widely.The multiplying factors were calculated according to a structure criterion of DI to 90%martensite plus retained austenite (or 10% of nonmartensitic transformation) and withreference to a base composition containing 1.0% C and 0.25% of each of the elements Mn,Si, Cr, and Ni, with 0% Mo to ensure that the first transformation product would not bebainite. The 50% martensite hardenability criterion (usually used when calculating DI) wasselected by Grossmann because this structure in medium-carbon steels corresponds to aninflection in the hardness distribution curve. The 50% martensite structure also results inmarked contrast in etching between the hardened and unhardened areas and in the fractureappearance of these areas in a simple fracture test. For many applications, however, it may benecessary to through-harden to a higher level of martensite to obtain optimum properties oftempered martensite in the core.In these instances, D1 values based on 90, 95, or 99.9% martensite must be used indetermining the hardenability requirements. These D1 values can be either experimentallydetermined or estimated from the calculated 50% martensite values using the relationshipsshown in Figure 5.10, which were developed for medium-carbon low-alloy steels [10]. A curvefor converting the D1 value for the normalized structure to the DI value of the spheroidizeannealed structure as shown in Figure 5.11 is also available. New multiplying factors for D1values were obtained from the measured Jominy curves using the conversion curve modifiedby Carney shown in Figure 5.12.The measured DI values were plotted against the percent content of various elements inthe steel. These curves were then used to adjust the DI value of the steels whose residualcontent did not conform to the base composition. Once the DI value of each analysis wasadjusted for residuals, the final step was to derive the multiplying factors for each elementfrom the quotient of the steels D and that of the base as follows:IfMn where DI is the initial reference value. 2006 by Taylor & Francis Group, LLC.D at x % MnIDI(5:2)D I, Normalized prior structure, in.43212345D I, Annealed prior structure, in.6FIGURE 5.11 Correlation between hardenability based on normalized and spheroidize-annealed priorstructures in alloyed 1.0% C steels. (From C.F. Jatczak, Metall. Trans. 4:22672277, 1973.)Excellent agreement was obtained between the case hardenability results of carburizedsteels assessed at 1.0% carbon level and the basic hardenability of the 1.0% C steels whenquenched from the normalized prior structure. It was thus confirmed that all multiplyingfactors obtained with prior normalized 1.0% C steels could be used to calculate the hardenability of all carburizing grades that are reheated for hardening following carburizing.Jatczak and Girardi [11] determined the difference in multiplying factors for prior normalized and prior spheroidize-annealed structures as shown in Figure 5.13 and Figure 5.14.The influence of austenitizing temperature on the specific hardenability effect is evident. Themultiplying factors shown in Figure 5.15 through Figure 5.18 were principally determined incompositions where only single-alloy additions were made and that were generally pearlitic ininitial transformation behavior. Consequently, these multiplying factors may be applied to7060DI50408030702060322.010246404856Sixteenths3.0in.644.08 10 12 14 16 18 20 22 24 26 28 30 32Distance from end-quenched endsixteenths. 1.2 1.4 1.6 1.8Distance from end-quenched endin.2.0FIGURE 5.12 Relationship between Jominy distance and DI. (From C.F. Jatczak, Metall. Trans.4:22672277, 1973.) 2006 by Taylor & Francis Group, LLC.Normalized prior structureBase : DI 1.4261.00MoMultiplying factor50.9040.80Carbon factor30.70Si-Multi-alloy steels2N0.751.001.251.50MnCr-Carburizing steelsCrSi-Single-alloy steelsNi%C1Mn, Cr, Si000.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25Percent elementFIGURE 5.13 Multiplying factors for calculation hardenability of high-carbon steels of prior normalizedstructure. (From C.F. Jatczak and D.J. Girardi, Multiplying Factors for the Calculation of Hardenability ofHypereutectoid Steels Hardened from 17008F, Climax Molybdenum Company, Ann Arbor, MI, 1958.)the calculation of hardenability of all single-alloy high-carbon compositions and to thosemultialloyed compositions that remain pearlitic when quenched from these austenitizingconditions. This involves all analyses containing less than 0.15% Mo and less than 2% totalof Ni plus Mn and also less than 2% Mn, Cr, or Ni when they are present individually. Ofcourse, all of the factors given in Figure 5.15 through Figure 5.18 also apply to the calculationof case hardenability of similar carburizing steels that are rehardened from these temperaturesfollowing air cooling or integral quenching.Annealed prior structure6Base : DI 1.421.00Multiplying factor50.9040.80Carbon factorMo30.701.000.75Si-Multi-alloy steels2Mn1.25%C1.50Si-Single-alloy steelsCrNiNi1Mn, Cr, Si000.250.500.751.00 1.25 1.50Percent element1.752.002.25FIGURE 5.14 Multiplying factors for calculation of hardenability of high-carbon steels of priorspheroidize-annealed structure. (From C.F. Jatczak and D.J. Girardi, Multiplying Factors for theCalculation of Hardenability of Hypereutectoid Steels Hardened from 17008F, Climax MolybdenumCompany, Ann Arbor, MI, 1958.) 2006 by Taylor & Francis Group, LLC.1.000.9045Multiplying factor0.801700670.701525 157580.6014750.500.400.200.300.400.500.60 0.70 0.80Percent carbon0.901.001.10FIGURE 5.15 Multiplying factors for carbon at each austenitizing condition. Data plotted on the lefthand side are data from Kramer for medium-carbon steels with grain size variation from ASTM 4 toASTM 8. (From C.F. Jatczak, Metall. Trans. 4:22672277, 1973.)3.5KramerManganese3.0170015752.5Multiplying factor15252.014751.51.0ChromiumKramer17002.517002.0152515751.514751.000.250.500.751.001.25Percent element1.501.752.00FIGURE 5.16 Effect of austenitizing temperature on multiplying factors for Mn and Cr at high-carbonlevels (Kramer data for medium-carbon steels). (From C.F. Jatczak, Metall. Trans. 4:22672277, 1973.) 2006 by Taylor & Francis Group, LLC.1700MolybdenumMultiplying factor5.04.0Kramer15753.0147515252. molybdenum1.251.50FIGURE 5.17 Effect of austenitizing temperature on multiplying factors for Mo at high carbon levels.(From C.F. Jatczak, Metall. Trans. 4:22672277, 1973.)Aluminum2.0147517001.5Kramer1.01700 MultialloyMultiplying factorSilicon2.517002.0Kramer14751525 Multialloy15751.5147515751.0Nickel2.0Kramer17001.5147515751.000.250.500.751.001.25Percent element1.501.752.00FIGURE 5.18 Effect of austenitizing temperature on multiplying factors for Si, Ni, and Al athigh-carbon levels. (Arrow on Al curve denotes maximum percentage studies by Kramer.) (FromC.F. Jatczak, Metall. Trans. 4:22672277, 1973.) 2006 by Taylor & Francis Group, LLC.For steels containing more Mo, Ni, Mn, or Cr than the above percentages, the measuredhardenability will always be higher than calculated with the single-alloy multiplying factorsbecause these steels are bainitic rather than pearlitic and also because synergistic hardenability effects have been found to occur between certain elements when present together. Thelatter effect was specifically noted between Ni and Mn, especially in steels made bainitic bythe addition of 0.15% or more Mo and that also contained more than 1.0% Ni.The presence of synergistic effects precluded the use of individual multiplying factors forMn and Ni, as the independence of alloying element effects is implicit in the Grossmannmultiplying factor approach. This difficulty, however, was successfully surmounted by computing combined Ni and Mn factors as shown in Figure 5.19.The factors from Figure 5.15 through Figure 5.18 can also be used for high-carbon steelsthat are spheroidize-annealed prior to hardening. However, the calculated DI value must beconverted to the annealed DI value at the abscissa on Figure 5.11. The accuracy of hardenability prediction using the new factors has been found to be within +10% at DI values ashigh as 660 mm (26.0 in.).50.8080406030 Mn.7040302020% Nickel30. Mn40Combinded Ni x Mn multiplying factor2010.8014758F (8008C)804030 Mn60.7040302020. Mn% Nickel3040201015258F (8308C).80805030 Mn6040402020% Nickel30.7040. Mn302015758F (8558C) nickelFIGURE 5.19 Combined multiplying factor for Ni and Mn in bainitic high-carbon steels quenched from800 to 8558C, to be used in place of individual factors when composition contains more than 1.0% Niand 0.15% Mo. (From C.F. Jatczak, Metall. Trans. 4:22672277, 1973.) 2006 by Taylor & Francis Group, LLC.5.3.2 JOMINY END-QUENCH HARDENABILITY TESTThe end-quench hardenability test developed by Jominy and Boegehold [12] is commonlyreferred to as the Jominy test. It is used worldwide, described in many national standards, andavailable as an international standard [13]. This test has the following significant advantages:1. It characterizes the hardenability of a steel from a single specimen, allowing a widerange of cooling rates during a single test.2. It is reasonably reproducible.The steel test specimen (25 mm diameter 100 mm) is heated to the appropriate austenitizing temperature and soaked for 30 min. It is then quickly transferred to the supportingfixture (Jominy apparatus) and quenched from the lower end by spraying with a jet of waterunder specified conditions as illustrated in Figure 5.20. The cooling rate is the highest at theend where the water jet impinges on the specimen and decreases from the quenched end,producing a variety of microstructures and hardnesses as a function of distance from thequenched end. After quenching, two parallel flats, approximately 0.45 mm below surface, areground on opposite sides of the specimen and hardness values (usually HRC) are measured at1=16 in. intervals from the quenched end and plotted as the Jominy hardenability curve (seeFigure 5.21). When the distance is measured in millimeters, the hardness values are taken atevery 2 mm from the quenched end for at least a total distance of 20 or 40 mm, depending onthe steepness of the hardenability curve, and then every 10 mm. On the upper margin of theJominy hardenability diagram, approximate cooling rates at 7008C may be plotted at severaldistances from the quenched end.1/2 in.(12.7 mm)1/8 in.(3.2 mm)11/8 in. (29 mm)45811/32 in. (26.2 mm)Unimpededwater jet4 in.(102 mm)1-in. (25.4-mm)round specimen2-1/2 in.(64 mm)Water at 75 5F(24 2.8C)1/2-in. (12.7-mm) i.d.orifice1/2 in.(13 mm)From quick-openingvalveFIGURE 5.20 Jominy specimen and its quenching conditions for end-quench hardenability test. 2006 by Taylor & Francis Group, LLC.270 70 18 5.6 K/s489" 124" 32.3" 10" F/s1/16 4/168/1616/16Cooling ratesDistance from quenched end, in.60Hardness, HRC50403020101.0002.03.0 in.75 mm2550Distance from quenched endFIGURE 5.21 Measuring hardness on the Jominy specimen and plotting the Jominy hardenabilitycurve. (From G. Krauss, Steels Heat Treatment and Processing Principles, ASM International, MetalsPark, OH, 1990.)Figure 5.22 shows Jominy hardenability curves for different unalloyed and low-alloyedgrades of steel. This figure illustrates the influence of carbon content on the ability to reach acertain hardness level and the influence of alloying elements on the hardness distributionexpressed as hardness values along the length of the Jominy specimen. For example, DINCk45, an unalloyed steel, has a carbon content of 0.45% C and exhibits a higher maximumhardness (see the value at 0 distance from the quenched end) than DIN 30CrMoV9 steel,60Hardness, HRC50CrV430CrMoV94050CrMo442MnV737MnSi520Ck4500406020Distance from quenched end, mm80FIGURE 5.22 Jominy hardenability curves (average values) for selected grades of steel (designationsaccording to German DIN standard). (From G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2,Warmebehandeln, Carl Hanser, Munich, 1987, p. 1012.) 2006 by Taylor & Francis Group, LLC.Distance from quenched end, mm102030407050Hardness, HRC60504030201004812162024Distance from quenched end,1/16 in.2832FIGURE 5.23 Reproductibility of the end-quench hardenability test. Hardenability range (hatchedarea between curves) based on tests by nine laboratories on a single heat of SAE 4068 steel. (FromC.A. Siebert, D.V. Doane, and D.H. Breen, The Hardenability of Steels, ASM International, Cleveland,OH, 1997.)which has only 0.30% C. However, the latter steel is alloyed with Cr, Mo, and V and shows ahigher hardenability by exhibiting higher hardness values along the length of the specimen.The Jominy end-quench test is used mostly for low-alloy steels for carburizing (corehardenability) and for structural steels, which are typically through-hardened in oils andtempered. The Jominy end-quench test is suitable for all steels except those of very low or veryhigh hardenability, i.e., D1 < 1.0 in. or D1 > 6.0 in. [8]. The standard Jominy end-quench testcannot be used for highly alloyed air-hardened steels. These steels harden not only by heatextraction through the quenched end but also by heat extraction by the surrounding air. Thiseffect increases with increasing distance from the quenched end.The reproducibility of the standard Jominy end-quench test was extensively investigated,and deviations from the standard procedure were determined. Figure 5.23 shows the results ofan end-quench hardenability test performed by nine laboratories on a single heat of SAE 4068steel [8]. Generally, quite good reproducibility was achieved, although the maximum difference may be 812 HRC up to a distance of 10 mm from the quenched end depending on theslope of the curve. Several authors who have investigated the effect of deviations from thestandard test procedure have concluded that the most important factors to be closelycontrolled are austenitization temperature and time, grinding of the flats of the test bar,prevention of grinding burns, and accuracy of the measured distance from the quenched end.Other variables such as water temperature, orifice diameter, free water-jet height, and transfertime from the furnace to the quenching fixture are not as critical. Test Methods for Shallow-Hardening SteelsIf the hardenability of shallow-hardening steels is measured by the Jominy end-quench test,the critical part of the Jominy curve is from the quenched end to a distance of about 1=2 in.Because of the high critical cooling rates required for shallow-hardening steels, the hardnessdecreases rapidly for every incremental increase in Jominy distance. Therefore the standardJominy specimen with hardness readings taken at every 1=16 in. (1.59 mm) cannot describeprecisely the hardness trend (or hardenability). To overcome this difficulty it may be helpful 2006 by Taylor & Francis Group, (1) modify the hardness survey when using standard Jominy specimens or (2) use special Lspecimens. Hardness Survey Modification for Shallow-Hardening SteelsThe essential elements of this procedure, described in ASTM A255, are as follows:1. The procedure in preparing the specimen before making hardness measurements is thesame as for standard Jominy specimens.2. An anvil that provides a means of very accurately measuring the distance from thequenched end is essential.3. Hardness values are obtained from 1=16 to 1=2 in. (1.5912.7 mm) from the quenchedend at intervals of 1=32 in. (0.79 mm). Beyond 1=2 in., hardness values are obtained at5=8, 3=4, 7=8, and 1 in. (15.88, 19.05, 22.23, and 25.4 mm) from the quenched end. Forreadings within the first 1=2 in. from the quenched end, two hardness traverses aremade, both with readings 1=16 in. apart: one starting at 1=16 in. and completed at 1=2in. from the quenched end, and the other starting at 3=32 in. (2.38 mm) and completedat 15=32 in. (11.91 mm) from the quenched end.4. Only two flats 1808 apart need be ground if the mechanical fixture has a grooved bedthat will accommodate the indentations of the flat surveyed first. The second hardnesstraverse is made after turning the bar over. If the fixture does not have such a groovedbed, two pairs of flats should be ground, the flats of each pair being 1808 apart. Thetwo hardness surveys are made on adjacent flats.5. For plotting test results, the standard form for plotting hardenability curves should beused. The Use of Special L SpecimensTo increase the cooling rate within the critical region when testing shallow-hardening steels,an L specimen, as shown in Figure 5.24, may be used. The test procedure is standard exceptthat the stream of water rises to a free height of 100+5 mm (instead of the 63.55 mm with astandard specimen) above the orifice, without the specimen in position.(a)(b)f32f25f32100 0.597 0.52525501097 0.5100 0.5f5f125f125f20f20f25f25FIGURE 5.24 L specimens for Jominy hardenability testing of shallow-hardening steels. All dimensionsin millimeters. 2006 by Taylor & Francis Group, LLC.h1 h2SSh3h4Hardness, HRCh5ABCDEh6 h 7FGCK116AADiameter, 1 in.S = average surface hardnessh1, h2, h3, etc. = average hardness at depths indicatedC= Average center hardnessThen Area of A = s + h1 1/162Area of B = h1 + h21/162Total area = 2(A + B + C + D + E + F + G + K )SC= 1/2 2 + h1 + h2 + h3 + h4 + h5 + h6 + h 7 + 2()FIGURE 5.25 Estimation of area according to SAC method. (From Metals Handbook, 9th ed., Vol. 1,ASM International, Metals Park, OH, 1978, pp. 473474.) [15] The SAC Hardenability TestThe SAC hardenability test is another hardenability test for shallow-hardening steels, otherthan carbon tool steels, that will not through-harden in sizes larger than 25.4 mm (1 in.) indiameter. The acronym SAC denotes surface area center and is illustrated in Figure 5.25.The specimen is 25.4 mm (1 in.) in diameter and 140 mm (5.5 in.) long. After normalizing atthe specified temperature of 1 h and cooling in air, it is austenitized by being held attemperature for 30 min and quenched in water at 24+58C, where it is allowed to remainuntil the temperature is uniform throughout the specimen.After the specimen has been quenched, a cylinder 25.4 mm (1 in.) in length is cut from its middle.The cut faces of the cylinder are carefully ground parallel to remove any burning or temperingthat might result from cutting and to ensure parallel flat surfaces for hardness measuring.First HRC hardness is measured at four points at 908 to each other on the surface. Theaverage of these readings then becomes the surface reading. Next, a series of HRC readingsare taken on the cross section in steps of 1=16 in. (1.59 mm) from the surface to the center ofthe specimen. From these readings, a quantitative value can be computed and designated by acode known as the SAC number.The SAC code consists of a set of three two-digit numbers indicating (1) the surface hardness,(2) the total Rockwell (HRC)-inch area, and (3) the center hardness. For instance, SAC 60-54-43indicates a surface hardness of 60 HRC, a total Rockwell-inch area of 54, and a center hardnessof 43 HRC. The computation of the total Rockwell-inch area is shown in Figure Hot Brine Hardenability TestFor steels of very low hardenability, another test has been developed [15] that involvesquenching several specimens 2.5 mm (0.1 in.) thick and 25 mm (1.0 in.) square in hot brineat controlled temperatures (and controlled quench severity), and determining the hardnessand percent martensite of each specimen. The brine temperature for 90% martensite structureexpressed as an equivalent diameter of a water-quenched cylinder is used as the hardenability 2006 by Taylor & Francis Group, LLC.criterion. Although somewhat complex, this is a precise and reproducible method for experimentally determining the hardenability of shallow-hardening steels. By testing several steelsusing this method, a linear regression equation has been derived for estimating hardenabilityfrom chemical composition and grain size that expresses the relative contribution of carbonand alloying elements by additive terms instead of multiplicative factors. Hardenability Test Methods for Air-Hardening SteelsWhen a standard Jominy specimen is used, the cooling rate at a distance of 80 mm from thequenched end (essentially the opposite end of the specimen) is approximately 0.7 K=s. Thehardenability of all steel grades with a critical cooling rate greater than 0.7 K=s can bedetermined by the standard Jominy end-quench hardenability test as a sufficient decrease inhardness will be obtained from increasing amounts of nonmartensite transformation products(bainite, pearlite, ferrite). However, for steels with a critical cooling rate lower than 0.7 K=sthere will be no substantial change in the hardness curve because martensite will be obtainedat every distance along the Jominy specimen. This is the case with air-hardening steels. Tocope with this situation and enable the use of the Jominy test for air-hardening steels, themass of the upper part of the Jominy specimen should be increased [16] by using a stainlesssteel cap as shown in Figure 5.26. In this way, cooling rates of the upper part of the specimenare decreased below the critical cooling rate of the steel itself.The complete device consists of the conical cap with a hole through which the specimen can befixed with the cap. When austenitizing, a leg is installed on the lower end of the specimen as shownin Figure 5.26 to equalize heating so that the same austenitizing conditions exist along the entire testspecimen. The total heating time is 40 min plus 20 min holding time at the austenitizing temperature. Before quenching the specimen according to the standard Jominy test procedure (togetherwith the cap), the leg should be removed. Figure 5.27 illustrates cooling rates when quenchinga standard Jominy specimen and a modified specimen with added cap. This diagram illustratesthe relationship between the cooling times from the austenitizing temperature to 5008C and thedistance from the quenched end of the specimen for different austenitizing temperatures.Figure 5.27 shows that at an austenitizing temperature of 8008C up to a distance of 20 mmfrom the quenched end, the cooling time curves for the standard specimen and the modified70 f f66All dimensionsin mm274858 ff426547.587Cap8Leg26 ff3032 f f4345 fFIGURE 5.26 Modification of the standard Jominy test by the addition of a cap to the specimen fortesting the hardenability of air-hardening steels. (From A. Rose and L. Rademacher, Stahl Eisen76(23):15701573, 1956 [in German].) 2006 by Taylor & Francis Group, LLC.Jominy distance from the quenched end, mm110100Modified Jominyspecimen (added cap)Standard Jominyspecimen90Austenitizingtemperature: 8008C808008C9008C 10008C 11008C7060504030201000400500600100200300Cooling time from austenitizing temp. to 5008C, sFIGURE 5.27 Cooling times between austenitizing temperature and 5008C for the standard Jominyspecimen and for a specimen modified by adding a cap. (From A. Rose and L. Rademacher, Stahl Eisen76(23):15701573, 1956 [in German].)specimen have the same path and thus the same cooling rate. At distances beyond approximately 20 mm, the cooling time curve for the modified specimen exhibits increasingly slowercooling rates relative to the standard specimen. By adding the cap, the cooling time is nearlydoubled, or the cooling rate is approximately half that exhibited by the unmodified test piece.Figure 5.28 shows two Jominy hardenability curves, one obtained with the standardspecimen and the other with the modified specimen, for the hot-working tool steel DIN45CrMoV67 (0.43% C, 1.3% Cr, 0.7% Mo, 0.23% V). Up to 20 mm from the quenched end,both curves are nearly equivalent. At greater distances, the retarded cooling exhibited by themodified specimen causes the decrease in hardness to start at 23 mm from the quenched end,while the decrease in hardness for the standard specimen begins at approximately 45 mm.The full advantage of the test with modified specimens for an air-hardening steel can beseen only if a quenched Jominy specimen is tempered at a temperature that will result in asecondary hardening effect. Figure 5.29 illustrates this for the tool steel DIN 45CrVMoW5870Austenitizing temp. 970 C60Hardness, HRC5040Standard Jominy specimenModified Jominy specimen(added cap)3020Depth of the ground flat 1 mm10005060708010203040Jominy distance from the quenched end, mmFIGURE 5.28 Jominy hardenability curves of grade DIN 45CrMoV67 steel for a standard specimen andfor a specimen modified by adding a cap. (From A. Rose and L. Rademacher, Stahl Eisen 76(23):15701573, 1956 [in German].) 2006 by Taylor & Francis Group, LLC.70Austenitizing temperature 1100 C60Hardness, HRC5040Not temperedTempered at:300 C550 C3020Modified Jominy specimen (added cap)Depth of the ground flat 1 mm100010203040506070Jominy distance from the quenched end, mm80FIGURE 5.29 Jominy hardenability curves of grade DIN 45CrVMoW58 steel after quenching (solidcurve) and after quenching and tempering (dashed curves) for a specimen modified by adding a cap.(From A. Rose and L. Rademacher, Stahl Eisen 76(23):15701573, 1956 [in German].)(0.39% C, 1.5% Cr, 0.5% Mo, 0.7% V, 0.55% W). After tempering at 3008C, the hardness near thequenched end decreases. Within this region martensitic structure is predominant. At about 25 mmfrom the quenched end the hardness curve after tempering becomes equal to the hardness curveafter quenching. After tempering to 5508C, however, the hardness is even more decreased up to adistance of 17 mm from the quenched end, and for greater distances a hardness increase up toabout 4 HRC units can be seen as a result of the secondary hardening effect. This increase inhardness can be detected only when the modified Jominy test is conducted.Another approach for measuring and recording the hardenability of air-hardening steelsis the Timken Bearing Company Air hardenability Test [17]. This is a modification of theair-hardenability testing procedure devised by Post et al. [18].Two partially threaded test bars of the dimensions shown in Figure 5.30 are screwed into acylindrical bar 6 in. in diameter by 15 in. long, leaving 4 in. of each test bar exposed. The totalsetup is heated to the desired hardening temperature for 4 h. The actual time at temperature is45 min for the embedded bar sections and 3 h for the sections extending outside the largecylinder. The test bar is then cooled in still air. The large cylindrical bar restricts the cooling ofthe exposed section of each test bar, producing numerous cooling conditions along the bar length.4 in.1 in.8 Thread61/2 in.11/8 in.1.0 in. Diameter6 in. Diameter875 in. Diameter800 in. Diameter10 in.1 in.15 in.FIGURE 5.30 Timken Roller Bearing Company air hardenability test setup. Two test specimens withshort threaded sections as illustrated are fixed in a large cylindrical bar. (From C.F. Jatczak, Trans.ASM 58:195209, 1965.) 2006 by Taylor & Francis Group, LLC.The various positions along the air-hardenability bar, from the exposed end to theopposite end (each test bar is 10 in. long), cover cooling rates ranging from 1.2 to 0.28F=s.The hardenability curves for six high-temperature structural and hot-work die steels areshown in Figure 5.31. The actual cooling rates corresponding to each bar position areshown. Each bar position is equated in this figure to other section sizes and shapes producing equivalent cooling rates and hardnesses at the section centers when quenched in air. Toprevent confusion, equivalent cooling rates produced in other media such as oil are not plotted in this chart. However, position 20 on the air-hardenability bar corresponds to the centerof a 13-in. diameter bar cooled in still oil and even larger cylindrical bars cooled in water.CoWSiCrNiMoVNorm.temp. 8FQuenchtemp. 8F0.61--0.671.300.180.470.26Ann17500.541.06-0.531.26-0.520.27"17500.380.40--0.854.870.111.340.60"1850A1150.390.52--0.855.12-5.100.68"1850Lapelloy182870.311.07--0.27 11.35 0.432.850.24"1900HTS-1100A1170.440.42-1.700.511.481.01"19006 in.312 in.6 in.312 in.0.83126 in.36 in.1.02251/2 in.312 in.2 in.312 in.3 in.312 in.2 in.312 in.3 in.312 in.4 in.312 in.322.01.51.23 in.33 in.4 in.34 in.5 in.35 in.21/2 in.3221/2 in.48633.072.5Coolingratein /Fmin2.42 in.32 in.Size round withsame still aircooling rateEquiv.A/Vratio1.3913Halmo5 in.312 in.1288751/2 in.312 in.H-117 in.37 in.0.31Interface0.29A1200.85104205 in.312 in."+Co13Mn4 in.312 in.C161722 ASCode11/2 in.312 in.Heat no.Size roundwith sameas quenchedhardness11/2 in.312 in. 11/4 in.312 in.TypeRockwell C hardness scale65605550454035302502468101214161820Distance in 1/2 in. units from large end of air hardenability barFIGURE 5.31 Chemistry and air-hardenability test results for various CrMoV steels. (From C.F.Jatczak, Trans. ASM 58:195209, 1965.) 2006 by Taylor & Francis Group, LLC.651340H60Limits for steel made tochemical specificationsHardness, HRC55504540StandardH- band353025200246810 12 14 16 18 20 22 24Distance from quenched end, 1/16 in.26283032FIGURE 5.32 Hardenability band for SAE 1340H steel.5.3.3 HARDENABILITY BANDSBecause of differences in chemical composition between different heats of the same grade ofsteel, so-called hardenability bands have been developed using the Jominy end-quench test.According to American designation, the hardenability band for each steel grade is marked bythe letter H following the composition code. Figure 5.32 shows such a hardenability band for1340H steel. The upper curve of the band represents the maximum hardness values, corresponding to the upper composition limits of the main elements, and the lower curve representsthe minimum hardness values, corresponding to the lower limit of the composition ranges.Hardenability bands are useful for both the steel supplier and the customer. Today themajority of steels are purchased according to hardenability bands. Suppliers guarantee that 93or 95% of all mill heats made to chemical specification will also be within the hardenabilityband. The H bands were derived from end-quench data from a large number of heats of aspecified composition range by excluding the upper and lower 3.5% of the data points. Steelsmay be purchased either to specified composition ranges or to hardenability limits defined byH bands. In the latter case, the suffix H is added to the conventional grade designation, forexample 4140H, and a wider composition range is allowed. The difference in hardenabilitybetween an H steel and the same steel made to chemical specifications is illustrated inFigure 5.32. These differences are not the same for all grades.High-volume production of hardened critical parts should have close tolerance of the depthof hardening. The customer may require, at additional cost, only those heats of a steel gradethat satisfy, for example, the upper third of the hardenability band. As shown in Figure 5.33,the SAE recommended specifications are: means-different ways of specifications.A minimum and a maximum hardness value at any desired Jominy distance. For example,J30-- 56 10=16 in: (A---A, Figure 5:33)-(5:3)If thin sections are to be hardened and high hardness values are expected, the selected Jominydistance should be closer to the quenched end. For thick sections, greater Jominy distancesare important.. The minimum and maximum distance from the quenched end where a desired hardnessvalue occurs. For example, 2006 by Taylor & Francis Group, LLC.70Hardness, HRC60ACD50BB40C30AD200481216202428Distance from quenched surface, 1/16 in.32FIGURE 5.33 Different ways of specifying hardenability limits according to SAE.J45 7=16 14=16 in: (B--B, Figure 5.33).Two maximum hardness values at two desired Jominy distances. For example,J52 12=16 in: ( max );.(5:4)J38 16=16 in: (max)(5:5)Two minimum hardness values at two desired Jominy distances. For example,J52 6=16 in: ( min );J28 12=16 in: (min)(5:6)Minimum hardenability is significant for thick sections to be hardened; maximum hardenability is usually related to thin sections because of their tendency to distort or crack,especially when made from higher carbon steels.If a structurevolume fraction diagram (see Figure 5.34) for the same steel is available, theeffective depth of hardening, which is defined by a given martensite content, may be determined from the maximum and minimum hardenability curves of the band. The structurevolume fraction diagram can also be used for the preparation of the transformation diagramwhen limits of the hardenability of a steel are determined. If the structurevolume fractiondiagram is not available, the limit values of hardness or the effective depth of hardening canbe estimated form the hardenability band using the diagram shown in Figure 5.35. Hardnessdepends on the carbon content of steel and the percentage of martensite after quenching.Figure 5.36. shows the hardenability band of the steel DIN 37MnSi5; the carbon content mayvary from a minimum of 0.31% to a maximum of 0.39%.The tolerance in the depth of hardening up to 50% martensite between a heat havingmaximum hardenability and a heat with minimum hardenability can be determined from thefollowing examples. For Cmin 0.31% and 50% martensite, a hardness of 38 HRC can bedetermined from Figure 5.35. This hardness corresponds to the lower curve of the hardenability band and found at a distance of 4 mm from the quenched end. For Cmax 0.39%and 50% martensite, a hardness of 42 HRC can be determined from Figure 5.35. Thishardness corresponds to the upper curve of the hardenability band and is found at 20 mmfrom the quenched end.In this example, the Jominy hardenability (measured up to 50% martensite) for this steelvaries between 4 and 20 mm. Using conversion charts, differences in the depth of hardeningfor any given diameter of round bars quenched under the same conditions can be determined. 2006 by Taylor & Francis Group, LLC.Hardness, HRC6050403020100P75FBStructure proportion, %50Ms250100P75B5025MsF010203004050Distance from quenched end of the Jominy specimen, mmFIGURE 5.34 Hardenability band and structurevolume fraction diagram of SAE 5140 steel. F ferrite, P pearlite, B bainite, Ms martensite. (From B. Liscic, H.M. Tensi, and W. Luty,Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.)70Martensite99.9%95908050%Hardness, HRC605040CNi30MnSiCrSiCrNiMoMoCrMoCrMaximum hardness after Burns,Moore and ArcherHardness at different percentagesof martensite after Hodge andOrehoski2010CrNi00.10.20.3 0.4 0.5 0.6Carbon content, wt % 5.35 Achievable hardness depending on the carbon content and percentage of martensite in the structure. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, SpringerVerlag, Berlin, 1992.) 2006 by Taylor & Francis Group, LLC.60Max. hardness difference32 HRC at J = 10 mmHardness, HRCGradientof hardness25 HRCminat J = 7.5 mm22 HRC/5 mm5047 HRCminat J = 2.5 mm50% Martensite40302037 Mn Si 5100102030405060Distance from quenched end, J, mm38 HRCmin at 4 mm42 HRCmax at 20 mm(Cmin = 0.31%;(Cmax = 0.39%;at 38 HRCmin)at 42 HRCmax)50% martensite50% martensiteHardenability: J(50 M) = 420 mmC 3139; J 420FIGURE 5.36 Hardenability band of DIN 37MnSi5 steel and the way technologically important information can be obtained. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology ofQuenching, Springer-Verlag, Berlin, 1992.)Effective depth of hardening is not the only information that can be derived from thehardenability band. Characteristic features of every hardenability band provide informationon the material-dependent spread of hardenability designated the maximum hardness difference as shown in Figure 5.36. The hardness difference at the same distance from the quenchedend, i.e., at the same cooling rate, can be taken as a measure of material-dependent deviations.Another important technological point that can be derived from the hardenability band is thehardness gradient. In Figure 5.36, this is illustrated by the minimum hardenability curve forthe steel in question where there is a high gradient of hardness (22 HRC for only 5 mmdifference in the Jominy distance). High hardness gradients indicate high sensitivity to coolingrate variation.5.4CALCULATION OF JOMINY CURVES FROM CHEMICAL COMPOSITIONThe first calculations of Jominy curves based on the chemical composition of steels wereperformed in the United States in 1943 [21,22]. Later, Just [23], using regression analysis offictitious Jominy curves from SAE hardenability bands and Jominy curves of actual heatsfrom the USS Atlas (USA) and MPI-Atlas (Germany), derived expressions for calculating thehardness at different distances (E) from the quenched end of the Jominy specimen. It wasfound that the influence of carbon depends on other alloying elements and also on the coolingrate, i.e., with distance from the quenched end (Jominy distance).Carbon starts at a Jominy distance of 0 with a multiplying factor of 50, while otheralloying elements have the factor 0 at this distance. This implies that the hardness at a Jominydistance of 0 is governed solely by the carbon content. The influence of other alloying elementsgenerally increases from 0 to values of their respective factors up to a Jominy distance of about10 mm. Beyond this distance, their influence is essentially constant. Near the quenched end the 2006 by Taylor & Francis Group, LLC.influence of carbon prevails, while the influence of other alloying elements remains essentiallyconstant beyond a Jominy distance of about 10 mm. This led Just to propose a single expressionfor the whole test specimen, except for distances shorter than 6 mm:ppJ680 95 C 0:00276E 2 C 20Cr 38Mo 14Mn 5:5Ni 6:1Si 39Vp 96P 0:81K 12:28 E 0:898E 13HRC(5:7)where J is the Jominy hardness (HRC), E the Jominy distance (mm), K the ASTM grain size,and the element symbols represent weight percentage of each.In Equation 5.7, all alloying elements are adjusted to weight percent, and it is valid withinthe following limits of alloying elements: C < 0.6%; Cr < 2%; Mn < 2%; Ni < 4%; Mo < 0.5%;V < 0.2%. Calculation of hardness at the quenched end (Jominy distance 0), using theequation for the maximum attainable hardness with 100% martensite, ispHmax 60 C 20 HRC,C < 0:6%(5:8)Although Equation 5.7 was derived for use up to a distance of 80 mm from the quenched endof the Jominy specimen, other authors argue that beyond a Jominy distance of 65 mm thecontinuous decrease in cooling rate at the Jominy test cannot be ensured even for low-alloysteels because of the cooling effect of surrounding air. Therefore, newer calculation methodsrarely go beyond a Jominy distance of 40 mm.Just [23] found that a better fit for existing mutual correlations can be achieved byformulas that are valid for groups of similar steels. He also found that multiplying hardenability factors for Cr, Mn, and Ni have lower values for case-hardening steels thanfor structural steels for hardening and tempering. Therefore, separate formulas for casehardening steels were derived:ppJ640 (case-hardening steels) 74 C 14Cr 5:4Ni 29Mo 16Mn 16:8 E 1:386E 7HRC(5:9)and for steels for hardening and tempering,pJ640 (steels for hardening and tempering) 102 C 22Cr 21Mn 7Ni 33Mop 15:47 E 1:102E 16HRC(5:10)In Europe, five German steel producers in a VDEh working group jointly developed formulasthat adequately define the hardenability from different production heats [24]. The goal was toreplace various existing formulas that were used individually.Data for some case-hardening steels and some low-alloy structural steels for hardening andtempering have been compiled, and guidelines for the calculation and evaluation of formulasfor additional families of steel have been established. This work accounts for influential factorsfrom the steel melting process and for possible deviations in the Jominy test itself. Multiplelinear regression methods using measured hardness values for Jominy tests and actual chemicalcompositions were also included in the analyses. The number of Jominy curves of a family ofsteel grades necessary to establish usable formulas should be at least equal to the square of thetotal number of chemical elements used for the calculation. Approximately 200 curves weresuggested. To obtain usable equations, all Jominy curves for steel grades that had similartransformation characteristics (i.e., similar continuous cooling transformation [CCT] diagram) 2006 by Taylor & Francis Group, LLC.TABLE 5.2Regression Coefficients for the Calculation of Jominy Hardness Values for Structural Steelsfor Hardening and Tempering Alloyed with about 1% CrJominyDistance(mm)1.53579111315202530Regression CoefficientsConstantCSiMn29.9626.7515.247.8227.2939.3442.6142.4941.7241.9444.6357.9158.6664.0481.1094.70100.7895.8588.6978.3472.2972.742.293.7610.8619.2722.0121.2520.5420.8217.5718.6219.123.772.164.8710.2414.7016.0617.7520.1820.7321.42SCr41.8573.7937.7665.8181.412.8612.2921.0224.8225.3926.4625.3323.8524.0824.39MoNi6.6638.3152.6354.9147.167.517.6910.75CuN2.652.5930.4138.9726.9535.9927.57Al83.3359.87115.50176.82144.074.568.587.979.08.899.969.649.71bben, H. Rohloff, P. Schu V. Schuler,ler, and H.J. Wieland,Source: R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. LuHarterei Tech. Mitt. 47(3):183188, 1992.when hardened were used. Therefore, precise equations for the calculation of Jominy hardnessvalues were derived only for steel grades of similar composition [24].The regression coefficients for a set of equations to calculate the hardness valuesat different Jominy distances from 1.5 to 30 mm from the quenched end are provided inTable 5.2. The chemistry of the steels used for this study is summarized in Table 5.3. Theregression coefficients in Table 5.2 do not have the same meaning as the hardenability factorsin Equation 5.7, Equation 5.9, and Equation 5.10; therefore, there is no restriction on thecalculation of Jominy hardness values at less than 6 mm from the quenched end. Because theregression coefficients used in this method of calculation are not hardenability factors, careshould be taken when deriving structural properties from them.The precision of the calculation was determined by comparing the measured and calculated hardness values and establishing the residual scatter, which is shown in Figure 5.37. TheTABLE 5.3Limiting Values of Chemical Composition of Structural Steels for Hardening and TemperingAlloyed with about 1% CraContent (%)CMin.Max.MeansSiMnPSCrMoNiAlCuN0.220.470.350.060.020.360.220.070.590.970.760.070.0050.0370.0130.0050.0030.0380.0230.0080.801. in calculations with regression coefficients of Table 5.2.bben, H. Rohloff, P. Schuler, V. Schu and H.J. Wieland,ler,Source: R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. LuHarterei Tech. Mitt. 47(3):183188, 1992. 2006 by Taylor & Francis Group, LLC.Steel 41Cr4 (DIN)600.92.1s= 2n2502.5s = 2.94 HRC1.1Hardness, HRC40 = calculated hardness measured hardness3.630600.70.1s = 7.45 HRC507.410.540 = calculated hardness measured hardness3001020Distance from the quenched end, mm1.430FIGURE 5.37 Comparison between measured (O) and calculated (.) hardness values for a melt withadequate consistency (top) and with inadequate consistency (bottom). (From R. Caspari, H. Gulden,K. Krieger, D. Lepper, A. Lubben, H. Rohloff, P. Schuler, V. Schuler, and H.J. Wieland, Harterei Tech.Mitt. 47(3):183188, 1992.)upper curve for a heat of DIN 41Cr4 steel, having a residual scatter of s 2.94 HRC, showsan adequate consistency, while the lower curve for another heat of the same steel, with aresidual scatter of s 7.45 HRC, shows inadequate consistency. Such checks were repeatedfor every Jominy distance and for every heat of the respective steel family. During this processit was found that the residual scatter depends on the distance from the quenched end and thatcalculated Jominy curves do not show the same precision (compared to measured curves) atall Jominy distances. For different steel grades with different transformation characteristics,the residual scatter varies with Jominy distance, as shown in Figure 5.38. In spite of theresidual scatter of the calculated results, it was concluded that properly calibrated predictorsoffer a strong advantage over testing in routine applications [25].When judging the precision of a calculation of Jominy hardness values, hardenabilitypredictors are expected to accurately predict (+1 HRC) the observed hardness values fromthe chemical composition. However, experimental reproducibility of a hardness value at afixed Jominy distance near the inflection point of the curve can be 812 HRC (see Figure 5.23for J10mm). Therefore it was concluded that a properly calibrated hardenability formula willalways anticipate the results of a purchasers check test at every hardness point better than anactual Jominy test [25].5.4.1 HYPERBOLIC SECANT METHOD FOR PREDICTING JOMINY HARDENABILITYAnother method for predicting Jominy end-quench hardenability from composition is basedon the four-parameter hyperbolic secant curve-fitting technique [26]. In this method, it is 2006 by Taylor & Francis Group, LLC.Hardness dissipation, HRC432Cr family of steelsCrMo family of steelsMnCr family of steelsC family of steels100301020Distance from the quenched end, mm40FIGURE 5.38 Residual scatter between measured and calculated hardness values versus distance to thequenched end, for different steel grade families. (From R. Caspari, H. Gulden, K. Krieger, D. Lepper,A. Lubben, H. Rohloff, P. Schuler, V. Schuler, and H.J. Wieland, Harterei Tech. Mitt. 47(3):183188,1992.)assumed that the Jominy curve shape can be characterized by a four-parameter hyperbolicsecant (sech) function (SECH).The SECH curve-fitting technique utilizes the equationDHx A B{sech[C (x 1)D ] 1}(5:11)DHx (A B) B{sech[C (x 1)D ]}(5:12)or alternativelyIH = ADH(P )B = IH DHDH(P )DH()x=PJominy Position, xFIGURE 5.39 Schematic showing the relationships between the hyperbolic secant coefficients A and Band Jominy curve characteristics. (From W.E. Jominy and A.L. Boegehold, Trans. ASM 26:574, 1938.) 2006 by Taylor & Francis Group, LLC.where the hyperbolic secant function for any y value issechy 2ey e y(5:13)where x is the Jominy distance from the quenched end, in 1/16 in., DHx the hardness at theJominy distance x, and A, B, C, D are the four parameters, which can be set such that DHxconforms closely to an experimental end-quench hardenability curve. The relationshipbetween parameters A and B and a hypothetical Jominy curve is illustrated in Figure 5.39.The parameter A denotes the upper asymptotic or initial hardness (IH) at the quenchedend. The parameter B corresponds to the difference between the upper and lower asymptotichardness values, respectively (DH1). This means that for a constant value of A, increasing thevalue of B will decrease the lower asymptotic hardness, as shown in Figure 5.40a.60A = 50C = 0.05D=2504030B = 2020(a)B = 10B = 30B variation1060Hardness, HRC50(b)A = 50B = 20C = 0.0540D = 2.030D = 0.5D = 3.520D variation at High C value1060A = 50B = 20C = 0.0550D = 0.540D = 2.0D = 3.53020D variation at Low C value(c)10004812 16 20 24 28 32 36Distance from quenched end of specimen in 1/16 in.FIGURE 5.40 Effect of SECH parameter variation on Jominy curve shape. (From W.E. Jominy andA.L. Boegehold, Trans. ASM 26:574, 1938.) 2006 by Taylor & Francis Group, LLC.6056C0.252Mn0.68Si0.32Ni1.59Cr0.51Mo0.45Grainsize848Hardness, HRC4440363228ID 527324A = 46.83B = 21.11C = 0.1859D = 0.9713201612048121620242832Distance from quenched end of specimen in 1/16 in.36FIGURE 5.41 Experimental end-quench hardenability data and best-fit hyperbolic secant function.(From W.E. Jominy and A.L. Boegehold, Trans. ASM 26:574, 1938.)The parameters C and D control the position of, and the slope at, the inflection point inthe calculated Jominy curve. If the A, B, and C parameters are constant, lowering the value ofparameter D will cause the inflection point to occur at greater Jominy distances, as shown inFigure 5.40b and Figure 5.40c. A similar result will be obtained if parameters A, B, and D arekept constant, and parameter C is shown by comparing Figure 5.40b and Figure 5.40c. Infact, it may be appropriate to set C and D to a constant value characteristic of a grade of steelsand describe the effects of compositional variations within the grade by establishing correlations with the other three parameters.It should be noted that some Jominy curves cannot be well described by a generalexpression such as Equation 5.11 or Equation 5.12. For example, if a significant amount ofcarbide precipitation were to occur in the bainite or pearlite cooling regime, a hump in theJominy curve might be observed that could not be calculated.To calculate the values of the four SECH parameters for each experimentally obtainedJominy curve, the minimum requirement is a data set from which the predictive equations willbe developed. This data set should contain compositions of each steel grade (or heat), withassociated values of Jominy hardness at different end-quench distances, as determined by theexperiment. Other metallurgical or processing variables such as grain size or austenitizingtemperature can also be included. The data set must be carefully selected; the best predictionswill be obtained when the regression data set is both very large and homogeneously distributed over the range of factors for which hardenability predictions will be desired.A linearnonlinear regression analysis program using least squares was used to calculateseparate values of the four parameters for each experimental Jominy curve in the regressiondata set by minimizing the differences between the empirical and analytical hardness curves,i.e., obtaining the best fit.Figure 5.41 provides experimental end-quench hardenability data and best fit hyperbolicsecant function for one steel in a data set that contained 40 carburizing steel compositions.Excellent fits were obtained for all 40 cases in the regression data set. Once the four 2006 by Taylor & Francis Group, LLC.TABLE 5.4Multiple Regression Coefficients for Backward-Elimination Regression AnalysisDependent Variable (SECH Parameter)BAInd. Var.C*C*C(Constant)CDCoeff.Ind. Var.Coeff.Ind. Var.Coeff.Ind. Var.Coeff.481.2703141.44362Cr*MoMn*SiGSNi*Ni*Ni(Constant)28.1776461.554991.716741.3535260.23736Cr*MoMn*SiNi*Ni*NiC*C*C*33.574790.799501.042080.0487114.852490.92535Cr*MoMn*SiNi*Ni*NiC*C*C(Constant)1.196951.976240.0926733.574790.26580parameters A, B, C, and D have been obtained for each heat as described above, four separateequations with these parameters as dependent variables are constructed using multipleregression analysis by means of a statistical analysis computer package.Table 5.4 provides multiple regression coefficients obtained with the backward elimination regression analysis of the above-mentioned 40 cases. In this elimination process, 31variables were arbitrarily defined for possible selection as independent variables in themultiple regression analysis. The list of these variables consisted of all seven single-elementand grain size terms, the seven squares and seven cubes of the single element and grainsize terms, and all 10 possible two-way element interaction terms that did not include carbonor grain size.Based on the multiple regression coefficients from Table 5.4, the following four equationsfor SECH parameters were developed for the regression data set of 40 carburizing steels:A 481C3 41:4(5:14)B 28:7CrMo 61:6MnSi 1:72GS 1:35Ni3 60:2(5:15)C 0:8CrMo 1:04MnSi 0:05Ni3 14:9C3 0:93(5:16)D 1:2CrMo 1:98MnSi 0:09Ni3 33:6C3 0:27(5:17)where an element name denotes percentage of that element in the steel and GS denotes grainsize. Equation 5.14 through Equation 5.17 are valid for steel compositions in the range of0.150.25% C, 0.451.1% Mn, 0.220.35% Si, 01.86% Ni, 01.03% Cr, and 00.76% Mo,with ASTM grain sizes (GS) between 5 and 9.After the four parameters are calculated, they are substituted into Equation 5.11 or Equation5.12 to calculate distance hardness (DH) at each Jominy distance x of interest. To validate thismethod, the Jominy curves were predicted for an independently determined data set of 24 heats,and this prediction was compared with those obtained by other two methods (AMAX [27] andJust [28] prediction methods). The SECH predictions were not as accurate as distance hardnesspredictions based on the two methods developed earlier because of the limited size and sparselypopulated sections (not homogeneously distributed) of the initial data set.5.4.2 COMPUTER CALCULATIONOF JOMINYHARDENABILITYThe application of computer technology has greatly enhanced the precision of these calculations. Commercial software is available for the calculation of Jominy hardness. For example, 2006 by Taylor & Francis Group, LLC.(a) 60(b) 60Processed J1 47.8Processed J32 18.9Inflection point 4.6HRC at inflection point 36.2Slope at inflection point 4.745Hardness, RcHardness, RcMeasuredProcessed3015453015481216202428324Jominy depth, 1/16 in.8121620242832Jominy depth, 1/16 in.FIGURE 5.42 Outputs from Minitech Predictor data processing program for best fit to measuredJominy data. (a) Initial trial; (b) final trial. (From J.S. Kirkaldy and S.E. Feldman, J. Heat. Treat.7:5764, 1989.)the Minitech Predictor [25] is based on the initial generation of an inflection point on theJominy curve. Figure 5.42 shows a typical output of the Minitech Predictor operating in thedata processing mode. Input values are Jominy hardness values, chemical composition, andestimated grain size.The Minitech program generates a predicted Jominy curve (Jn) and a predicted inflectionpoint distance from quenched end x and displays a comparison of the predicted andexperimentally obtained curves as shown in Figure 5.42a. A weighting pattern Jn is accessedthat specifies a weight of 1.5 for all distances from n 1 to n 2x and a weight of 0.75 forn > 2x to n 32 mm (or any limit of the data). Using an effective carbon content and grainsize as adjustable parameters, the theoretical curve is then iterated about J and x tonminimize the weighted root mean square deviation of the calculated curve from the experimental curve. The final best-fit calculated curve is plotted along with the main processed dataas shown in Figure 5.42b.Jominy distance (mm) Hardness (HV) is the Vickers pyramid hardness.Calculated Jominy hardness curves are used to replace Jominy testing by equivalentpredictions for those steel grades (e.g., very shallow-hardening steels) that it is difficult orimpossible to test. Although the accurate prediction of hardenability is important, it is moreimportant for the steel manufacturer to be able to refine the calculations during the steelmaking process. For example, the steel user indicates the desired Jominy curve by specifying threepoints within H band for SAE 862OH as shown in Figure 5.43 [29].Using these data, the steel mill will first compare the customers specification against twomain criteria: 2006 by Taylor & Francis Group, LLC.(a)(b)500(SAE 8620H)(SAE 8620H)500Customer demandCustomer demand400300300200Hardness, HV40020001020300102030Distance from the quenched end, mmFIGURE 5.43 (a) Customers specification of hardenability within an H band for SAE 8620H. (b) Jominycurve for finished heat. (From T. Lund, Carburizing Steels: Hardenability Prediction and HardenabilityControl in Steel-Making, SKF Steel, Technical Report 3, 1984.)1. That the hardenability desired is within limits for the steel grade in question2. That the specified points fall on a Jominy curve permissible within the analysis range forSAE 862OH, i.e., the specified points must provide a physically possible Jominy curveWhen the actual heat of steel is ready for production, the computer program will automatically select the values for alloy additions and initiate the required control procedures.The samples taken during melting and refining are used to compute the necessary chemicaladjustments. The computer program is linked directly to the ferroalloy selection and dispensing system. By successive adjustments, the heat is refined to a chemical composition thatmeets the required hardenability specification within the chemical composition limits for thesteel grade in question.The use of calculated Jominy curves for steel manufacturing process control is illustratedin the following example. Quality control analysis found that the steel heat should have amanganese value of 0.85%. During subsequent alloying, the analysis found 0.88% Mn. Thisoverrun in Mn was automatically compensated for by the computer program, which adjustedhardenability by decreasing the final chromium content slightly. The resulting heat hadthe measured Jominy curve shown in Figure 5.43b. In this case, the produced steel doesnot deviate from the required specification by more than +10 HV at any Jominy distancebelow 19 mm.5.5 APPLICATION OF HARDENABILITY CONCEPT FOR PREDICTIONOF HARDNESS AFTER QUENCHINGJominy curves are the preferred methods for the characterization of steel. They are used tocompare the hardenability of different heats of the same steel grade as a quality controlmethod in steel production and to compare the hardenability of different steel grades whenselecting steel for a certain application. In the latter case, Jominy curves are used to predictthe depth of hardening, i.e., to predict the expected hardness distribution obtained afterhardening parts of different cross-sectional dimensions after quenching under various conditions. Such predictions are generally based on the assumption that rates of cooling prevailing 2006 by Taylor & Francis Group, LLC.Cooling rate at 700 C (1300 F)270 170 110 70 43490 305 195 125 772342183314 12 10 9 7.8 6.9 6.1 5.626 21.4 18 16.3 14 12.4 11 of bar, mmC/sF/s5125Three-quarter radiusSurfaceHalf-radiusCenter75350225Diameter of bar, in.(a)31561Quenched in water at 60 m/min (200 ft/min)00024681012141618204. distance from quenched end, 1/16 in.Cooling rate at 700 C (1300 F)270 170 110 70 43490 305 195 125 77(b)31562342183314 12 10 9 7.8 6.9 6.1 5.626 21.4 18 16.3 14 12.4 11 10.0C/sF/s5125Three-quarter radiusHalf-radius4753SurfaceCenter50225Diameter of bar, in.Diameter of bar, mm1001Quenched in oil at 60 m/min (200 ft /min)0002468101214161820Equivalent distance from quenched end, 1/16 in.FIGURE 5.44 Correlation of equivalent cooling rates at different distances from the quenched end ofthe Jominy specimen and at different locations on the cross section of round bars of different diameters,quenched in (a) water agitated at 1 m=s and in (b) oil agitated at 1 m=s. (From Metals Handbook, 9th ed.,Vol. 1, ASM International, Metals Park, OH, 1978, p. 492.)at different distance from the quenched end of the Jominy specimen may be compared withthe cooling rates prevailing at different locations on the cross sections of bars of differentdiameters. If the cooling rates are equal, it is assumed that equivalent microstructure andhardness can be expected after quenching.The diagrams shown in Figure 5.44 have been developed for this purpose. These diagramsprovide a correlation of equivalent cooling rates at different distances from the quenched endof the Jominy specimen and at different locations (center, half-radius, three-quarter radius,surface) on the cross section of round bars of different diameters. They are valid for thespecified quenching conditions only. Figure 5.44a is valid only for quenching in water at anagitation rate of 1 m=s, and the diagram in Figure 5.44b is valid only for quenching in oil atan agitation rate of 1 m=s. 2006 by Taylor & Francis Group, mm2 50f 100 mm (4 i n.)45Distance below surface of bar40f 75 mm (3 i n.)135301f 50 mm (2 i n.)25f 38 mm (11/2 i n.)2015f 25 mm (1 i n.)1/210f 125 mm (1/2 i n.)500051015 202530 35 4011Distance from the quenched end4550 mm2 in.FIGURE 5.45 Relationship between cooling rates at different Jominy distances and cooling ratesat different points below the surface of round bars of different diameters quenched in moderatelyagitated oil. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London,1984, p. 145.)Another diagram showing the relation between cooling rates at different Jominy distancesand cooling rates at different distances below the surface of round bars of different diameters,taken from the ASTM standard, is shown in Figure 5.45. From this diagram, the samecooling rate found at a Jominy distance of 14 mm prevails at a point 2 mm below the surfaceof a 75-mm diameter bar, at 10 mm below the surface of 50-mm diameter bar, and at thecenter of a 38-mm diameter bar when all the bars are quenched in moderately agitated oil.Using this diagram, it is possible to construct the hardness distribution curve across thesection after hardening. This type of diagram is also valid for only the specified quenchingconditions.To correlate the hardness at different Jominy distances and the hardness at the center ofround bars of different diameters that are quenched in various quenchants under differentquenching conditions, the critical diameter (Dcrit), the ideal critical diameter (DI), andGrossmanns quenching intensity factor H must be used. The theoretical background ofthis approach is provided by Grossmann et al. [5], who calculated the half-temperaturetime (the time necessary to cool to the temperature halfway between the austenitizingtemperature and the temperature of the quenchant). To correlate Dcrit and H, Asimowet al. [31], in collaboration with Jominy, defined the half-temperature time characteristicsfor the Jominy specimen also. These half-temperature times were used to establish the relationship between the Jominy distance and ideal critical diameter DI, as shown in Figure 5.46. Ifthe microstructure of this steel is determined as a function of Jominy distance, the ideal criticaldiameter can be determined directly from the curve at that distance where 50% martensite isobserved as shown in Figure 5.46. The same principle holds for Dcrit when different quenchingconditions characterized by the quenching intensity factor H are involved. Figure 5.47 showsthe relationship between the diameter of round bars (Dcrit and DI) and the distance from thequenched end of the Jominy specimen for the same hardness (of 50% martensite) at the center ofthe cross section after quenching under various conditions [31]. 2006 by Taylor & Francis Group, LLC.Distance from quenched end, in.0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.07160Ideal critical diameter D I, in.Ideal critical diameter D I, mm061401205100480603240120030401020Distance from quenched end, mm050FIGURE 5.46 Relationship between the distance from the quenched end of the Jominy specimen andthe ideal critical diameter. (From M. Asimov, W.F. Craig, and M.A. Grossmann, SAE Trans.49(1):283292, 1941.)The application of the Figure 5.47 diagram can be explained for the two steel gradesshown in Figure 5.48. The hardness at 50% martensite for the unalloyed steel grade Ck45(0.45% C) is 45 HRC, while for the low-alloy grade 50CrMo4 steel (0.5% C) the hardness is 48HRC. The lower part of the diagram depicts two H curves taken from the diagram in Figure5.47. One is for vigorously agitated brine (H 5.0), and the other for moderately agitated oil(H 0.4). From both diagrams in Figure 5.48, it is seen that quenching the grade 50CrMo4steel in vigorously agitated brine provides a hardness of 48 HRC in the center of the crosssection of a round bar of 110-mm diameter. Quenching the same steel in moderately agitatedoil provides this hardness at the core of round bars of only 70-mm diameter. The unalloyedgrade Ck45 steel, having lower hardenability when quenched in vigorously agitated brine,provides a hardness of 45 HRC in the center of a 30-mm diameter bar. Quenching inmoderately agitated oil provides this hardness in the center of a round bar of only 10 mmdiameter.Diameter D crit or D I, mm200a150H5210.80.40.210050bc0.0200102030405060Distance from the quenched end, mmFIGURE 5.47 Relationship between the round bar diameter and the distance from the quenched end ofthe Jominy specimen, giving the hardness in the center of the cross section after quenching underdifferent quenching conditions, a, water; b, oil; c, air. (From M. Asimov, W.F. Craig, and M.A.Grossmann, SAE Trans. 49(1):283292, 1941.) 2006 by Taylor & Francis Group, LLC.Hardness, HRCDiameter, mm6050CrM044020Ck4548 or 45 HRCHardness at50 % martensite0150H = 5.0100H = 0.4500010 20 30 40 5060Distance from quenched end, mmFIGURE 5.48 Determining the critical diameter of round bars (i.e., the hardness of 50% martensite atthe center) from the Jominy hardenability curves of two steel grades quenched in vigorously agitatedbrine (H 5.0) and in moderately agitated oil (H 0.4). (Steel grade designation according to DIN.)(From G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser,Munich, 1987, p. 1012.)5.5.1 LAMONT METHODThe diagram shown in Figure 5.47 permits the prediction of hardness only at the center ofround bars. Lamont [32] developed diagrams relating the cooling rate at a given Jominydistance to that at a given fractional depth in a bar of given radius that has been subjected to agiven Grossmann quenching intensity (H) factor. Analytical expressions have been developedfor the Lamont transformation of the data to the appropriate Jominy distance J:J J (D, r=R, H )(5:18)where D is the diameter of the bar, r=R the fractional position in the bar (r=R 0 at thecenter; r=R 1 at the surface), and H the Grossmann quenching intensity factor. Theseexpressions [33] are valid for any value of H from 0.2 to 10 and for bar diameters up to200 mm (8 in.).Lamont developed diagrams for the following points and fractional depths on the crosssection of round bars: r=R 0 (center), r=R 0.1, r=R 0.2, . . . , r=R 0.5 (half-radius),r=R 0.6, . . . , r=R 1.0 (surface). Each of these diagrams is always used in connection withJominy hardenability curve for the relevant steel. Figure 5.49 through Figure 5.51 show theLamont diagram for r=R 0 (center of the cross section), r=R 0.5, and r=R 0.8, respectively.The Lamont method can be used for four purposes:1. To determine the maximum diameter of the bar that will achieve a particular hardnessat a specified location on the cross section when quenched under specified conditions.For example, if the Jominy hardenability curve of the relevant steel grade shows ahardness of 55 HRC at a Jominy distance of 10 mm, then the maximum diameter of thebar that will achieve this hardness at half-radius when quenched in oil with H 0.35will be 28 mm. This result is obtained by using the diagram in Figure 5.50 for r=R 0.5and taking the vertical line at a Jominy distance of 10 mm to the intersection with thecurve for H 0.35, giving the value of 28 mm on the ordinate. 2006 by Taylor & Francis Group, LLC.01015203037.54050 mm1.0in.Bar diameter, mm1401205. =0R6.0r5.00.50R10080605040200.354. intensity H1601.0001111Distance from the quenched end, in.2FIGURE 5.49 Relation between distance from the quenched end of Jominy specimen and bar diameterfor the ratio r=R 0, i.e., the center of the cross section, for different quenching intensities. (From J.L.Lamont, Iron Age 152:6470, 1943.)2. To determine the hardness at a specified location when the diameter of the bar, thequenching intensity H, and the steel grade are known. For example, if a 120-mmdiameter bar is quenched in still water (H 1.0), the hardness at the center (r=R 0)will be determined at a distance of 37.5 mm from the quenched end on the Jominy curveof the relevant steel grade (see Figure 5.49).220200Bar diameter180160in. 010.01020304050 mmr = 0.5R9. intensity Hmm2403.00.350.26040282002.01.0011Distance from the quenched end, in.2FIGURE 5.50 Relation between distance from the quenched end of Jominy specimen and bar diameterfor the ratio r=R 0.5, i.e., 50% from the center, for different quenching intensities. (From J.L. Lamont,Iron Age 152:6470, 1943.) 2006 by Taylor & Francis Group, LLC.mm240220in. mm0.6521.56.01.0Bar diameter401400. intensity H18010 15 0.8RrR1/2111/2Distance from the quenched end, in.2FIGURE 5.51 Relation between distance from the quenched end of Jominy specimen and bar diameterfor the ratio r=R 0.8, i.e., 80% from the center, for different quenching intensities. (From J.L. Lamont,Iron Age 152:6470, 1943.)3. To select adequate quenching conditions when the steel grade, the bar diameter, andthe location on the cross section where a particular hardness should be attained areknown. For example, a hardness of 50 HRC, which corresponds to the distance of15 mm from the quenched end on the Jominy curve of the relevant steel grade, shouldbe attained at the center of a 50-mm diameter bar. The appropriate H factor can befound by using Figure 5.49. In this case, the horizontal line for a 50-mm diameter andthe vertical line for a 15-mm Jominy distance intersect at the point that corresponds toH 0.5. This indicates that the quenching should be done in oil with good agitation.If the required hardness should be attained only up to a certain depth below the surface,the fractional depth on the cross section must be first established to select the appropriatetransformation diagram. For example, if 50 HRC hardness, which corresponds to a distanceof 15 mm from the quenched end on the Jominy curve of the relevant steel grade, should beattained at 7.6 mm below the surface of a 76-mm diameter bar, thenr 38 7:6 0: 8R38(5:19)This calculation indicates that the diagram for r=R 0.8 (Figure 5.51) should be used. In thiscase, the horizontal line for 76 mm diameter intersects the vertical line for 15-mm Jominydistance on the interpolated curve H 0.6. This indicates that quenching should be performed in oil with strong agitation (see Table 5.1).4. To predict the hardness along the radius of round bars of different diameters whenthe bar diameter and steel grade and its Jominy curve and quenching intensity H are 2006 by Taylor & Francis Group, LLC.known. For this calculation, diagrams for every ratio r=R from the center to the surfaceshould be used. The following procedure should be repeated with every diagram. At thepoint where the horizontal line (indicating the bar diameter in question) intersectsthe relevant H curve, the vertical line gives the corresponding distance from thequenched end on the Jominy curve from which the corresponding hardness can beread and plotted at the corresponding fractional depth. Because some simplifyingassumptions are made when using Lamont diagrams, hardness predictions are approximate. Experience has shown that for small cross sections and for the surface of largediameter bars, the actual hardness is usually higher than predicted.5.5.2STEEL SELECTION BASEDONHARDENABILITYThe selection of a steel grade (and heat) for a part to be heat-treated depends on thehardenability that will yield the required hardness at the specified point of the cross sectionafter quenching under known conditions. Because Jominy hardenability curves and hardenability bands are used as the basis of the selection, the method described here is confined tothose steel grades with known hardenability bands or Jominy curves. This is true first of allfor structural steels for hardening and tempering and also for steels for case hardening (todetermine core hardenability).If the diameter of a shaft and the bending fatigue stresses it must be able to undergo areknown, engineering analysis will yield the minimum hardness at a particular point on thecross section that must be achieved by hardening and tempering. Engineering analysis mayshow that distortion minimization requires a less severe quenchant, e.g., oil. Adequatetoughness after tempering (because the part may also be subject to impact loading) mayrequire a tempering temperature of, e.g., 5008C.The steps in the steel selection process are as follows:Step 1. Determine the necessary minimum hardness after quenching that will satisfy therequired hardness after tempering. This is done by using a diagram such as the one shown inFigure 5.52. For example, if a hardness of 35 HRC is required after hardening and thentempering at 5008C at the critical cross-sectional diameter, the minimum hardness afterquenching must be 45 HRC.As-quenched hardness, HRC6055Tempered600 C/60 min50454035Tempered500 C/ 60 min30251520 25 30 35 40Tempered hardness, HRC45FIGURE 5.52 Correlation between the hardness after tempering and the hardness after quenching forstructural steels (according to DIN 17200). 2006 by Taylor & Francis Group, LLC.Hardness, HRC70%C0. of martensite %90100FIGURE 5.53 Correlation between as-quenched hardness, carbon content, and percent martensite(according to Hodge and Orehovski). (From Metals Handbook, 9th ed., Vol. 1, ASM International,Metals Park, OH, 1978, pp. 473474, p. 481.)Alternatively, if the carbon content of the steel and the percentage of as-quenched martensite at the critical point of the cross section is known, then by using a diagram that correlateshardness with percent carbon content and as-quenched martensite content (see Figure 5.53),the as-quenched hardness may then be determined. If 80% martensite is desired at a criticalposition of the cross section and the steel has 0.37% C, a hardness of 45 HRC can be expected.Figure 5.53 can also be used to determine the necessary carbon content of the steel when aparticular percentage of martensite and a particular hardness after quenching are required.Step 2. Determine whether a certain steel grade (or heat) will provide the required asquenched hardness at a critical point of the cross section. For example, assume that a shaft is45 mm in diameter and that the critical point on the cross section (which was determined fromengineering analysis of resultant stresses) is three fourths of the radius. To determine if aparticular steel grade, e.g., AISI 4140H, will satisfy the requirement of 45 HRC at (3=4)Rafter oil quenching, the diagram shown in Figure 5.54a should be used. This diagramcorrelates cooling rates along the Jominy end-quench specimen and at four characteristiclocations (critical points) on the cross section of a round bar when quenched in oil at 1 m=sagitation rate (see the introduction to Section 5.5 and Figure 5.44). Figure 5.54a shows thatat (3=4)R the shaft having a diameter of 45 mm will exhibit the hardness that correspondsto the hardness at a distance of 6.5=16 in. (13=32 in.) from the quenched end of theJominy specimen.Step 3. Determine whether the steel grade represented by its hardenability band (ora certain heat represented by its Jominy hardenability curve) at the specified distancefrom the quenched end exhibits the required hardness. As indicated in Figure 5.54b, theminimum hardenability curve for AISI 4140H will give a hardness of 49 HRC. Thismeans that AISI 4140H has, in every case, enough hardenability for use in the shaftexample above.This graphical method for steel selection based on hardenability, published in 1952 byWeinmann and coworkers, can be used as an approximation. Its limitation is that the diagramshown in Figure 5.54a provides no information on the quality of the quenching oil and itstemperature. Such diagrams should actually be prepared experimentally for the exact conditions that will be encountered in the quenching bath in the workshop; the approximation willbe valid only for that bath.5.5.3 COMPUTER-AIDED STEEL SELECTION BASED ON HARDENABILITYAs in other fields, computer technology has made it possible to improve the steel selectionprocess, making it quicker, more intuitive, and even more precise. One example, using a 2006 by Taylor & Francis Group, LLC.Diameter of bar, mm(a) 125Three-quarter radius Half-radius100Surface755045Center25Quenched in oil at 1 m/s00246810 12 14 16Distance from quenched end, 1/16 in.182049 HRCminimumhardenabilityof 4140H meetsthe requirementHardness, HRC(b) 704140H605040300246810 12 14 16Distance from quenched end, 1/16 in.1820FIGURE 5.54 Selecting a steel of adequate hardenability. (a) equivalent cooling rates (and hardnessafter quenching) for characteristic points on a round bars cross section and along the Jominy endquench specimen. (b) Hardenability band of AISI 4140H steel. (From Metals Handbook, 9th ed., Vol. 1,ASM International, Metals Park, OH, 1978, pp. 473474, p. 493.)software package developed at the University of Zagreb [35], is based on a computer file ofexperimentally determined hardenability bands of steels used in the heat-treating shop. Themethod is valid for round bars of 2090-mm diameters. The formulas used for calculation ofequidistant locations on the Jominy curve, described in Ref. [23], were established throughregression analysis for this range of diameters.The essential feature of this method is the calculation of points on the optimum Jominyhardenability curve for the calculated steel. Calculations are based on the required asquenched hardness on the surface of the bar and at one of the critical points of its crosssection [(3=4)R, (1=2)R, (1=4)R, or center]. The input data for the computer-aided selectionprocess are the following:.....Diameter of the bar (D mm)Surface hardness HRCHardness at a critical point HRCQuenching intensity factor I (I equals the Grossmann quenching intensity factor H asgiven in Table 5.1)Minimum percentage of martensite required at the critical pointThe first step is to calculate the equidistant locations from the quenched end on the Jominycurve (or Jominy hardenability band). These equidistant locations are the points on theJominy curve that yield the required as-quenched hardness. The calculations are performedas follows [23]: 2006 by Taylor & Francis Group, LLC.On the surface:Es D0:7185:11I 1:28(5:20)At (3=4)R:E3=4R D1:058:62I 0:668(5:21)E1=2R D1:169:45I 0:51(5:22)E1=4R D1:147:7I 0:44(5:23)At (1=2)R:At (1=4)R:At the center:Ec D1:188:29I 0:44(5:24)[Note: The calculated E values are in millimeters.]After calculating the equidistant locations for the surface of the bar (Es) and for one ofthe critical points (Ecrit), using the hardenability band of the relevant steel, the hardness valuesachievable with the Jominy curve of the lowest hardenability (Hlow) and the hardnessvalues achievable with the Jominy curve of the highest hardenability (Hhigh) for both Esand Ecrit locations are then determined as shown in Figure 5.55.The degree of hardening S is defined as the ratio of the measured hardness after quenching(at a specified point of the cross section) to the maximum hardness that can be achieved withthe steel in question:SHHmax(5:25)H highsH highHRCcritH lowsH lows0EsE critJominy distance, mmFIGURE 5.55 Determination of minimum and maximum hardness for equidistant location Es and Ecritfrom a relevant hardenability band. (After T. Filetin, Strojarstvo 24(2):7581, 1982 [in Croatian].) 2006 by Taylor & Francis Group, LLC.TABLE 5.5Correlation between Degree of Hardening S andPercentage of Martensite in As-Quenched StructurePercent MartensiteDegree of Hardening S5060607070808085859090959597971000.700.740.740.760.760.780.780.810.810.860.860.910.910.950.951.00Source: T. Filetin and J. Galinec, Software programme for steelselection based on hardenability, Faculty of Mechanical Engineering,University of Zagreb, 1994.It can be easily calculated for the equidistant location Ecrit on the upper and lower curves ofthe hardenability band, taking the value for Hmax from the relevant Jominy curve at distance0 from the quenched end (J 0). In this way, two distinct values of the degree of hardening,Supper and Slower, are calculated. Each corresponds to a certain percentage of martensite in theas-quenched structure as shown in Table 5.5.It is also possible to determine whether the required percentage of martensite can beachieved by either Jominy curve of the hardenability band. Instead of providing the percentage of martensite in the as-quenched structure as input data, the value of S (degree ofhardening) may be given. For statically stressed parts, S < 0.7; for less dynamically stressedparts, 0.7 < S < 0.86; and for highly dynamically stressed parts, 0.86 < S < 1.0. In this way, adirect comparison of the required S value with values calculated for both Jominy curves at theEcrit location can be performed. There are three possibilities in this comparison:1. The value of S required is even lower than the S value calculated for the lower curve ofthe hardenability band (Slower). In this case all heats of this steel will satisfy therequirement. The steel actually has higher hardenability than required.2. The value of S required is even higher than the S value calculated for the upper curve ofthe hardenability band (Supper). In this case, none of the heats of this steel can satisfythe requirement. This steel must not be selected because its hardenability is too low forthe case in question.3. The value of required degree of hardening (S) is somewhere between the values for thedegree of hardening achievable with the upper and lower curves of the hardenabilityband (Supper and Slower, respectively).In the third case, the position of the S required, designated as X, is calculated accordingto the formula:X 2006 by Taylor & Francis Group, LLC.S SlowerSupper Slowerwhere X is the distance from the lower curve of the hardenability band on the ordinate Ecritto the actual position of S required, which should be on the optimum Jominy curve. Thiscalculation divides the hardenability band into three zones:The lower third, X 0.33The middle third, 0.33 < XThe upper third, 0.66 < X0.66All heats of a steel grade where the Jominy curves pass through the zone in which therequired S point is situated can be selected as heats of adequate hardenability. This zone isindicated in a graphical presentation of the method. Once the distance X is known, theoptimum Jominy hardenability curve can be drawn. The only requirement is that for everydistance from the quenched end the same calculated ratio (X) that indicates the same positionof the Jominy curve relative to the lower and upper hardenability curves of the hardenabilityband is maintained.The following example illustrates the use of this method in selecting a steel grade forhardening and tempering.A 40-mm diameter shaft after hardening and tempering should exhibit a surface hardnessof Hs 28 HRC and a core hardness of Hc 26 HRC. The part is exposed to high dynamicstresses. Quenching should be performed in agitated oil.The first step is to enter the input data and select the critical point on the cross section (inthis case the core) as shown in Figure 5.56. Next, the required percentage of martensite at thecritical point after quenching (in this case 95%, because of high dynamic stresses) and thequenching intensity I (in this case 0.5, corresponding to the Grossmann value H ) are selected.The computer program repeats the above-described calculations for every steel grade forwhich the hardenability band is stored in the file and presents the results on the screen asshown in Figure 5.57. This is a list of all stored steel grades regarding suitability for theapplication being calculated. Acceptable steel grades, suitable from the upper, middle, orlower third of the hardenability band, and unacceptable steel grades with excessively highhardenability are determined.Selection of steel in hardened and tempered conditionDiameter, mm (090):40Critical point onthe cross-section:CoreRequired value:<1> 3/4R<2> 1/2R<3> 1/4R<4> Core<1> Hardness, HRC (2050)<2> Tensile strength, N/mm2(7501650)Hardness, HRC On the surface: At the critical point:28 H s26 H cFIGURE 5.56 Input data for computer program. (From T. Filetin and J. Galinec, Software programme forsteel selection based on hardenability, Faculty of Mechanical Engineering, University of Zagreb, 1994.) 2006 by Taylor & Francis Group, LLC.Results of steel selectionJUSC4181C4730C4731C4781C4732C4782C4733C4738C4734AISI4130E413241404150Not suitableNot suitableSuitable heats from upper third of bandSuitable heats from upper third of bandSuitable heats from middle third of bandSuitable heats from middle third of bandSuitable heats from middle third of bandToo high hardenabilityToo high hardenabilityFIGURE 5.57 List of computer results. (From T. Filetin and J. Galinec, Software programme for steelselection based on hardenability, Faculty of Mechanical Engineering, University of Zagreb, 1994.)For each suitable steel grade, a graphical presentation as shown in Figure 5.58 can beobtained. This gives the optimum Jominy hardenability curve for the case required andindicates the desired zone of the hardenability band.In addition, the necessary tempering temperature can be calculated according to the formula:Ttemp 917vu8uln Hcrit68tHtemp(5:26)Swhere Ttemp is the absolute tempering temperature (K) (valid for 4008C < Ttemp < 6608C), Hcritthe hardness after quenching at the critical point HRC (taken from the optimum Jominycurve at the distance for the critical point), Htemp the required hardness after tempering at thecritical point HRC, and S the degree of hardening (ratio between hardness on the optimumJominy curve at the distance Ecrit and at the distance E 0).70Hardness, HRC60504030201005 E s 10 E crit152025303540Jominy distance, mmFIGURE 5.58 Graphical presentation of the optimum Jominy hardenability curve. (From T. Filetin andJ. Galinec, Software programme for steel selection based on hardenability, Faculty of MechanicalEngineering, University of Zagreb, 1994.) 2006 by Taylor & Francis Group, LLC.Tensile strength (Rm, N=mm2 ) is also calculated at the relevant points using the formula:Rm 0:426H 2 586:5 [N=mm2 ](5:27)where H is the corresponding hardness value in HRC. Knowing the tensile strength (Rm),other mechanical properties are calculated according to the formulas:Yield strength:Rp 0:2 (0:8 0:1S )Rm 170S 200 [N=mm2 ](5:28)A5 0:46 (0:0004 0:00012S )Rm [%](5:29)Z 0:96 (0:00062 0:00029S )Rm [%](5:30)Rd (0:25 0:45Z )Rm [N=mm2 ](5:31)KU [460 (0:59 0:29S)Rm ](0:7) [J](5:32)Elongation:Contraction:Bending fatigue strength:Impact energy (toughness):For every steel grade (and required zone of the hardenability band) that has been foundsuitable, the mechanical properties for the surface and for the critical cross section point canbe calculated. The computer output is shown in Figure 5.59.Compared to the previous steel selection processes, these computer-aided calculationshave the following advantages:(AISI 4140)Heats from the middle third of the bandCalculated tempering temperature: 643 CMechanical propertiesYield strength: Rp0.2, N/mm2Tensile strength: Rm, N/mm2Bending fatigue strength: Rd, N/mm2Elongation: A5, %Contraction: Z, %Impact engergy: KU, JSurface7939204992065125Criticalpoint7358744742165123FIGURE 5.59 Computer display of calculated mechanical properties. (From T. Filetin and J. Galinec,Software programme for steel selection based on hardenability, Faculty of Mechanical Engineering,University of Zagreb, 1994.) 2006 by Taylor & Francis Group, LLC.1. Whereas the previously described graphical method is valid for only one specifiedquenching condition for which the relevant diagram has been plotted, the computeraided method allows great flexibility in choosing concrete quenching conditions.2. Selection of the optimum hardenability to satisfy the requirements is much moreprecise.3. Calculations of the exact tempering temperature and all mechanical properties aftertempering at the critical point that give much more information and facilitate the steelselection are possible. IN HEAT TREATMENT PRACTICEHARDENABILITYOFCARBURIZED STEELSCarburized parts are primarily used in applications where there are high surface stresses.Failures generally originate in the surface layers where the service stresses are most severe.Therefore, high case strength and high endurance limits are critical factors. High casehardness improves the fatigue durability. Historically, it was thought that core hardenabilitywas required for the selection of carburizing steels and heat treatment of carburized parts andthat core hardenability would ensure adequate case hardenability. Equal additions of carbon,however, do not have the same effect on the hardenability of all steel compositions; thereforethe historical view of core hardenability may not be correct. In fact, hardenability of both caseand core is essential for proper selection of the optimum steel grade and the heat treatment ofcarburized parts.It is now also known that the method of quenching after carburizing, i.e., direct quenchingor reheat and quench, influences case hardenability. The case hardenability of carburized steelis determined by using the Jominy end-quench test.Standard Jominy specimens are carburized in a carburizing medium with a high Cpotential for sufficient time to obtain a carburized layer of the desired depth. In addition tothe Jominy specimens, two bars of the same steel and heat, the same surface finish, and thesame dimensions (25 mm diameter) are also carburized under identical conditions. These barsare used to plot the carbon gradient curve shown in Figure 5.60a, which is produced bychemical analysis of chips obtained from machining of the carburized layer at different layerthicknesses. In this way, as shown in Figure 5.60a, the following carbon contents were foundas a function of case depth:1.2Measured carboncontent curvecarburized at 925 Cfor 4.5 h1.0d1 = 0.20.7% Cd4 = 0.57%C0.81.0% C0.8% C0.6d3 = 0.450.40.2(a)d10d2d3d2 = 0.32 0.9% Cd40. from surface, mm1.2(b)FIGURE 5.60 (a) Measured carbon gradient curve after gas carburizing at 9258C for 4.5 h. (b) Grinding of the carburized Jominy specimen. (From T. Filetin and B. Liscic, Strojarstvo 18(4):197200, 1976 [inCroatian].) 2006 by Taylor & Francis Group, LLC.1.0% C at 0.2 mm depth (distance from the surface of the bar)d10.9% C at 0.32 mm depthd20.8% C at 0.45 mm depthd30.7% C at 0.57 mm depthd4One of the carburized Jominy specimens should be end-quenched in the standard wayusing the Jominy apparatus directly from the carburizing temperature (direct quenching), andthe other should be first cooled to room temperature and then reheated and quenched from atemperature that is usually much lower than the carburizing temperature (reheat and quench).After quenching, all Jominy specimens should be ground on four sides of the perimeter to thedepths, d1, d2, d3, and d4, as shown in Figure 5.60b. Hardness is measured in the standard way oneach of the ground surfaces, and the corresponding Jominy curves are plotted. Figure 5.61aDirect quenching from 925 C9006005550050HV800HRC7000.9 671.0 650.70.8% C 60454003000.17200403530252010000 2 4 6 8 10 121416 20(a)24 283236404448Jominy distance, mmIndirect quenching from 820 C900670.7657000.9%C60HV60055500HRC0.8800501.040045403530252030020010000 2 4 6 8 10 121416 20(b)24 283236404448Jominy distance, mmFIGURE 5.61 Jominy case hardenability curves of carburized DIN 16MnCr5 steel (a) after directquenching from 9258C and (b) after reheating followed by quenching from 8208C. (From T. Filetin and B. Liscic, Strojarstvo 18(4):197200, 1976 [in Croatian].) 2006 by Taylor & Francis Group, LLC.provides an example of Jominy hardenability curves for the carburizing steel grade DIN16MnCr5 (0.17% C, 0.25% Si, 1.04% Mn, 1.39% Cr). The carbon contents in the case were 1.0,0.9, 0.8, and 0.7% C, and the core carbon content was 0.17% C after direct quenching from thecarburizing temperature, 9258C. Figure 5.61b provides Jominy curves for the same carburizedcase after indirect quenching (reheated to 8208C). From both diagrams of Figure 5.61 thefollowing conclusions can be drawn:1. The hardenability of the core is substantially different from the hardenability within thecarburized case.2. The best hardenability of the carburized case is found for this steel grade at 0.9% C withdirect quenching and at 0.8% C with indirect quenching (reheat and quench).Consequently, the carburizing process should be controlled so that after carburizing asurface carbon content of 0.9% is obtained for direct quenching and one of 0.8% for indirectquenching.5.6.2 HARDENABILITYOFSURFACE LAYERS WHEN SHORT-TIME HEATING METHODS ARE USEDWhen short-time (zero time) heating processes for surface hardening are used, e.g., flamehardening, induction hardening, or laser hardening, the same metallurgical reactions occur asin conventional hardening except that the heating processes cycle must be much shorter thanthat of conventional hardening. Heating time for these proceses vary by one to three orders ofmagnitude; approximately 100 s for flame hardening, 10 s or less for induction hardening, and1 s or less for laser hardening. This means that the heating rates are very high. Problemsassociated with these high heating rates are twofold.1. The transformation from the bcc lattice of the a-iron to the fcc lattice of the g-iron doesnot occur between normal temperatures Ac1 and Ac3 as in conventional hardeningbecause the high heating rate produces nonequilibrium systems. The Ac1 and Ac3temperatures are displaced to higher temperatures as shown in Figure 5.62. Althoughan austenitizing temperature may be sufficiently high to form austenite under slowheating conditions (conventional hardening), the same temperature level may not besufficient to even initiate austenization under high heating rates [37]. Therefore, substantially higher austenitizing temperatures are used with flame, induction, and laserhardening (especially the latter) than for conventional hardening of the same steel.2. For quench hardening, the austenitization must dissolve and uniformly distribute thecarbon of the carbides in the steel. This is a time-dependent diffusion process (sometimescalled homogenization), even at the high temperatures used in short-time heatingmethods. At very high heating rates, there is insufficient time for diffusion of carbonatoms from positions of higher concentrations near carbides to the positions of lowerconcentrations (areas that originated from practically carbon-free ferrite). This diffusiondepends on the path length of carbon atoms and therefore is dependent on the distribution of carbon in the starting structure. Coarse pearlitic structures, spheroidized structures, and (particularly) nodular cast iron with a high content of free ferrite areundesirable in this regard. Tempered martensite, having small and finely dispersedcarbides, provides the shortest paths for carbon diffusion and is therefore most desirable.Figure 5.62a illustrates a time temperature transformation diagram for continuousheating at different heating rates when austenitizing an unalloyed steel with 0.7% C witha starting structure of ferrite and lamellar pearlite. Figure 5.62b shows a similar diagram for 2006 by Taylor & Francis Group, LLC.900Austenite880Temperature, C860840820Austenite + carbide800780Au760ste740720(a)+ ferr it e + c a rAc3bid eAc1Ferrite + pearite7006800.1n it e110102Time, s900103104Austenite880Temperature, C860Acm840820Austenite + carbide800780Au760740720(b)n it eAc1e+ f e r rit e + c a r b i d eAc1bTempered martensite700680101ste100101102Time, s103104105FIGURE 5.62 Time temperature transformation diagram for continuing heating with different heatingrates, when austenitizing an unalloyed steel with 0.7% C. (a) Starting structure, ferrite and lamellar pearlite;(b) starting structure, tempered martensite. (From A. Rose, The austenitizing process when rapid heatingmethods are involved, Der Peddinghaus Erfahrungsaustausch, Gevelsberg, 1957, pp. 1319 [in German].)a starting structure of tempered martensite. A comparison of the two diagrams illustrates theinfluence of starting structure on the austenitizing process. Whereas for the ferritepearlitestarting structure at maximum heating rate the upper transformation temperature Ac3 is8658C, for the starting microstructure of tempered martensite, the Ac3 temperature is 8358C.This means that the austenite from a starting structure of tempered martensite has a betterhardenability than the austenite of a pearliteferrite starting structure. The practical consequence of this is that prior to surface hardening by any short-time heating process, if the steelis in the hardened and tempered condition, maximum hardened case depths are possible. Ifthe annealed material has a coarse lamellar structure, or even worse, globular carbides,minimum hardening depths are to be expected.5.6.3 EFFECTOFDELAYED QUENCHINGON THEHARDNESS DISTRIBUTIONDelayed quenching processes have been known for a long time. Delayed quenching meansthat austenitized parts are first cooled slowly and then after a specified time they are quenchedat a much faster cooling rate. Delayed quenching is actually a quenching process in which adiscontinuous change in cooling rate occurs. In some circumstances, depending on steel 2006 by Taylor & Francis Group, LLC.AISI 4140Batch No.7345613111617155055121450HRCHRC5514545R3/4R 1/2R 1/4R01/4R 1/2R 3/4RR50 mm DiameterFIGURE 5.63 Measured hardness distribution on the cross section of 50 mm diameter 200 mm bars made of AISI 4140 steel quenched according to conditions given in Table 5.6. (From B. Liscic, S. Svaic,and T. Filetin, Workshop designed system for quenching intensity evaluation and calculation of heattransfer data. ASM Quenching and Distortion Control, Proceedings of First International ConfererenceOn Quenching and Control of Distortion, Chicago, IL, 2225 Sept. 1992, pp. 1726.)hardenability and section size, the hardness distribution in the cross section after delayedquenching does not have a normal trend (normally the hardness decreases continuously fromthe surface toward the core) but instead exhibits an inverse trend (the hardness increasesfrom the surface toward the core). This inverse hardness distribution is a consequence of thediscontinuous change in the cooling rate and related to the incubation period (at differentpoints in the cross section) before changing the cooling rate. This process has been explainedtheoretically by Shimizu and Tamura [40,41] in Figure 5.63.In every experiment, the delay in quenching was measured as the time from immersion tothe moment when maximum heat flux density on the surface (tqmax) occurred. As shown inFigure 5.63 and Table 5.6 for AISI 4140 steel with a section 50 mm in diameter, when thedelay in quenching (due to high concentration of the PAG polymer solution and corresponding thick film around the heated parts) was more than 15 s (tqmax > 15 s), a completely inverseor inverse to normal hardness distribution was obtained. In experiments where tqmax was lessthan 15 s, a normal hardness distribution resulted.Besides the inherent hardenability of a steel, delayed quenching may substantially increasethe depth of hardening and may compensate for lower hardenability of the steel [39].Interestingly, none of the available software programs for predicting as-quenched hardnesssimulates the inverse hardness distribution because they do not account for the length of theincubation period before the discontinuous change in cooling rate at different points in thecross section.5.6.4 A COMPUTER-AIDED METHOD TO PREDICT THE HARDNESS DISTRIBUTIONAFTER QUENCHING BASED ON JOMINY HARDENABILITY CURVESThe objective here is to describe one method of computer-aided calculation of hardnessdistribution. This method, developed at the University of Zagreb [44], is based on the Jominy 2006 by Taylor & Francis Group, LLC.TABLE 5.6Time from Immersion (tqmax) until Maximum Heat Flux Density under VariousQuenching Conditions for AISI 4140 Bars (50 mm Diameter 3 200 mm)aFigure 63 Curve No.111121314151617Quenching Conditionstqmax (s)Mineral oil at 208C, without agitationPolymer solution (PAG) 5%; 408C; 0.8 m=sPolymer solution (PAG) 15%; 408C; 0.8 m=sPolymer solution (PAG) 25%; 408C; 0.8 m=sPolymer solution (PAG) 20%; 358C; 1 m=sPolymer solution (PAG) 10%; 358C; 1 m=sPolymer solution (PAG) 5%; 358C; 1 m=sPolymer solution (PAG) 20%; 358C; 1 m=s1416337030121347aSee Figure 63. Source: B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensityevaluation and calculation of heat transfer data. ASM Quenching and Distortion Control,Proceedings of First International Confererence On Quenching and Control of Distortion,Chicago, IL, 2225 Sept. 1992, pp. 1726.hardenability curves. Jominy hardenability data for steel grades of interest are stored in adatabank. In this method, calculations are valid for cylindrical bars 2090 mm in diameter.Figure 5.64 shows the flow diagram of the program, and Figure 5.65 is a schematic of thestep-by-step procedure:Step 1. Specify the steel grade and quenching conditions.Step 2. Harden a test specimen (50 mm diameter 200 mm) of the same steel grade byquenching it under specific conditions.Step 3. Measure the hardness (HRC) on the specimens cross section in the middle of thelength.Step 4. Store in the file the hardness values for five characteristic points on the specimenscross section (surface, (3=4)R, (1=2)R, (1=4)R, and center). If the databank already containsthe hardness values for steel and quenching conditions obtained by previous measurements,then eliminate steps 2 and 3 and retrieve these values from the file.Step 5. From the stored Jominy hardenability data, determine the equidistant points onthe Jominy curve (Es, E3=4R, E1=2R, E1=4R, Ec) that have the same hardness values as thosemeasured at the characteristic points on the specimens cross section.Step 6. Calculate the hypothetical quenching intensity I at each of the mentioned characteristic points by the following regression equations, based on the specimens diameter Dspecand on known E values:"D0:718specIs 5:11Es"I3=4R 2006 by Taylor & Francis Group, LLC.#0:78D1:05spec8:62E3=4R(5:33)#1:495(5:34)StartInput: Steel grade,quenching conditionsDatabasestoreddata into files: Jominy hardenabilitySearch in databasefilesAdditionalexperimentsyes Quenching intensityrecorded as functions:T = f (t ), q = f (t ), q = f (Ts) Measuring ofquenchingintensitynoAll dataavailable? Hardness distributionon the test specimenscross section Hardening oftest specimenInput parameters: Steel grade; quenching mediumand conditions; Jominy hardenability data;hardness on the test specimens cross sectionReading of the corresponding Jominydistances (Ei )Calculation of the "hypothetical quenching intensity"within the test specimens cross sectionIi = f (Dsp, Ei)Input of the actualdiameter, DCalculation of Jominy distancescorresponding to the diameter, D (E'i = f (D, li ))Reading of the hardness data from the Jominyhardenability curve for Jominy distancescorresponding to: E'S, E'3R/4, E'R/2, E'R/4, E'CResults obtained: Hardnessdata in five points on thecross section of the bar.Hardness curve, graphicallyyesAnotherdiameter?noyesAnother steelgrade and/orquenchingconditionsnoStopFIGURE 5.64 Flowchart of computer-aided prediction of hardness distribution on cross section of quenched round bars. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching,Springer-Verlag, Berlin, 1992.)"I1=2RD1:16spec9:45E1=2R"I1=4R 2006 by Taylor & Francis Group, LLC.D1:14spec7:7E1=4R#1:495(5:35)#2:27(5:36)Hardenability 2006 by Taylor & Francis Group, LLC.HRCHRCHRCActual diameterDHsMeasured H3R/4hardness HJominy curve for therelevant steel gradeR /2HR /4HC120H'sCR/4 R/2 3R/4PredictedhardnessdistributionSH'cTest specimen50-mm diameterEsEcE3R/4 ER/2 ER/4E'sDistance fromquenched endE'c Distance fromE' quenched endE'3R /4 E'R /2R /4lxStep 1 to step 4Step 5Step 6Step 8C R /2SR /4 3 R /4Step 9 FIGURE 5.65 Stepwise scheme of the process of prediction of hardness distribution after quenching. (From B. Liscic, H.M. Tensi, and W. Luty, Theory andTechnology of Quenching, Springer-Verlag, Berlin, 1992.)271"Ic D1:18spec#2:278:29Ec(5:37)Equation 5.33 through Equation 5.37 combine the equidistant points on the Jominy curve,the specimens diameter, and the quenching intensity and were derived from the regressionanalysis of a series of CraftsLamont diagrams [22]. This analysis is based on Justs relationships [42] for the surface and the center of a cylinder:Ei AD B1I B2(5:38)where Ei is the corresponding equidistant point on the Jominy curve, A, B1, B2 the regressioncoefficients, D the bar diameter, and I the quenching intensity (H according to Grossmann)Step 7. Enter the actual bar diameter D for which the predicted hardness distributionis desired.Step 8. Calculate the equidistant Jominy distances E , E 3=4R, E =2R, E =4R, E c thats11correspond to the actual bar diameter D and the previously calculated hypothetical quenchingintensities Is Ic using the formulas:0Es D0:7185:11I 1:28(5:39)0E3=4R D1:058:62I 0:668(5:40)0E1=2R D1:169:45I 0:51(5:41)D1:147:7I 0:44(5:42)0E1=4R 0Ec D1:188:29I 0:44(5:43)Step 9. Read the hardness values H s, H =4R, H 1=2R, H =4R, and H from the relevant31cJominy curve associated with the calculated Jominy distances and plot the hardness distribution curve over the cross section of the chosen actual diameter D.Figure 5.66 provides an example of computer-aided prediction of hardness distributionfor 30-and 70-mm diameter bars made of AISI 4140 steel quenched in a mineral oil at 208Cwithout agitation. Experimental validation using three different steel grades, four differentbar diameters, and four different quenching conditions was performed, and a comparison topredicted results is shown in Figure 5.67. In some cases, the precision of the hardnessdistribution prediction was determined using the GerberWyss method [43]. From examples2, 3, 5, and 6 of Figure 5.67 it can be seen that the computer-aided prediction provides a betterfit to the experimentally obtained results than the GerberWyss method. 2006 by Taylor & Francis Group, LLC.7060Jominy curveHRC5040302010f 30f 70 mm0102030Jominy distance, mm40Prediction of hardness distributionInput data:Steel grade: C. 4732 (SAE 4140H) ; B. NO. 43111Quenching conditions: oil-UTO-2;20 C; Om/sDiameter for hardening, mm: 30Results of computer aided prediction:Calculated hardness:Diameter = 30mmSurface, hrc .......3/4 Radius ...............1/2 Radius ...............1/4 Radius ...............Center ....................= 55.3= 54.3= 53= 51.5= 51.1Graphic presentation (yes = 1, no = 0)Another diameter (yes = 1, no = 0)Diameter for hardening, mm: 70Diameter = 70Surface, hrc .......3/4 Radius ...............1/2 Radius ...............1/4 Radius ...............Center ....................= 53.1= 46.4= 40.7= 39.6= 39Graphic presentation (yes = 1, no = 0)Another diameter (yes = 1, no = 0)FIGURE 5.66 An example of computer-aided prediction of hardness distribution for quenched round bars of 30 and 70 mm diameter, steel grade SAE 4140H. (From B. Liscic, H.M. Tensi, and W. Luty,Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.) Selection of Optimum Quenching ConditionsThe use of above relationship and stored data permits the selection of optimum quenchingcondition when a certain hardness value is required at a specified point on a bar cross sectionof known diameter and steel grade. Figure 5.68 illustrates an example where an as-quenched 2006 by Taylor & Francis Group, LLC.651425Hardness, HRC605575045368403530123456789095101520 05101520Distance from the surface, mmSteel grade: SAE 6150 H, Oil UTO 2, 20 C, 1m/s,Steel grade: SAE 6150 H, Oil UTO 2, 20 C, 1m/s,Steel grade: SAE 4135 H, Oil UTO 2, 20 C, 1m/s,Steel grade: SAE 6150 H, Water,20 C, 1m/s,Steel grade: SAE 6150 H, Oil UTO 2, 20 C, 1m/s,Steel grade: SAE 4135 H, Water,20 C, 1m/s,Steel grade: SAE 4140 H, Oil UTO 2, 20 C, 0 m/s,Steel grade: SAE 4140 H, Oil UTO 2, 20 C, 0 m/s,Steel grade: SAE 4140 H, Mineral oil, 20 C, 1.67m/s,253035D = 40 mmD = 30 mmD = 40 mmD = 40 mmD = 40 mmD = 70 mmD = 30 mmD = 80 mmD = 80 mm40Obtained byexperimentComputer-aidedpredictionPrediction accordingto GerberWyss methodFIGURE 5.67 Comparison of the hardness distribution on round bar cross sections of different diameters and different steel grades, measured after experiments and obtained by computer-aided prediction as well as by prediction according to the GerberWyss method. (From B. Liscic, H.M. Tensi, andW. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.)Measured hardness onQuenching conditions:the test specimen1 Blended mineral oil: 20 C: 1 m/s2 Blended mineral oil: 20 C: 1.6 m/s603 Blended mineral oil: 70 C: 1.0 m/s 504 Salt both-AS-140: 200 C: 0.6 m/s HRCActual diameter60HRC 4055Jominy curve for the steel grade: 55C.4732 (SAE 414OH) B. No. 89960Hq = 5150Hq502482H'qHardnesstolerance451234454040353530S3/4RC053010 15 20 25 30 35 40S3/4RE3/4R E'3/4RDistance from quenched end, mmC FIGURE 5.68 An example of computer-aided selection of quenching conditions (From B. Liscic, H.M.Tensi, and W. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.) 2006 by Taylor & Francis Group, LLC.hardness of 51 HRC (Hq) at (3=4)R of a 40-mm diameter bar made of SAE 4140H steel isrequired. Using the stored hardenability curve for this steel, the equivalent Jominy distanceE3=4R yielding the same hardness can be found. Using E3=4R and the actual diameter D,hypothetical quenching intensity factor I3=4R can be calculated according to Equation 5.34.That equation also applies to the test specimen of 50-mm diameter and can be written asI3=4R7:05E3=4R!1:495(5:44)By substituting the calculated value of I3=4R and D 50 mm, the equivalent Jominy distanceE3=4R corresponding to (3=4)R of the specimens cross section, can be calculated:0E3=4R 10:85I3=4R0:668(5:45)For calculated E =4R, the hardness of 48 HRC can be read off from the Jominy curve as3shown in Figure 5.68. This means that the same quenching condition needed to produce ahardness value Hq 51 HRC at (3=4)R of a 40-mm diameter bar will yield a hardness Hs of48 HRC at (3=4)R of the 50-mm diameter standard specimen.The next step is to search all stored hardness distribution curves of test specimens made ofthe same steel grade for the specific quenching condition by which the nearest hardness Hqhas been obtained (tolerance is +2 HRC). As shown in Figure 5.68, the required hardnessmay be obtained by quenching in four different conditions, but the best-suited are conditions1 and 2.The special advantage of computer-aided calculations, particularly the specific methoddescribed, is that users can establish their own databanks dealing with steel grades of interestand take into account (by using hardened test specimens) the actual quenching conditionsthat prevail in a batch of parts using their own quenching facilities.REFERENCES1. G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser, Munich,1987, p. 1012.2. K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984, p. 145.3. M.A. Grossmann, M. Asimov, and S.F. Urban, Hardenability of Alloy Steels, ASM International,Cleveland, OH, 1939.4. Metals Handbook, 8th ed., Vol. 2, American Society for Metals, Cleveland, OH, 1964, p. 18.5. M.A. Grossmann, M. Asimov, and S.F. Urban, Hardenability its relation to quenching and somequantitative data, Hardenability of Alloy Steels, ASM International, Cleveland, OH, 1939.6. A. Moser and A. Legat, Determining hardenability from composition, Harterei Tech. Mitt.24(2):100105 (1969) (in German).7. D.J. Carney and A.D. Janulionis, An examination of the quenching constant H, Trans. ASM43:480496 (1951).8. C.A. Siebert, D.V. Doane, and D.H. Breen, The Hardenability of Steels, ASM International,Cleveland, OH, 1997.9. C.F. Jatczak, Hardenability in high carbon steel, Metall. Trans. 4:22672277 (1973).10. Metals Handbook, ASM International, Cleveland, OH, 1948, p. 499.11. C.F. Jatczak and D.J. Girardi, Multiplying Factors for the Calculation of Hardenability of Hypereutectoid Steels Hardened from 17008F, Climax Molybdenum Company, Ann Arbor, MI, 1958.12. W.E. Jominy and A.L. Boegehold, Trans. ASM 26:574 (1938). 2006 by Taylor & Francis Group, LLC. procedures are described in ASTM A 255, SAE J 406, and ISO=R 642 (1967).Metals Handbook, 9th ed., Vol. 1, ASM International, Metals Park, OH, 1978, pp. 473474.R.A. Grange, Estimating the hardenability of carbon steels, Metall. Trans. 4:2231 (1973).A. Rose and L. Rademacher, Weiterentwicklung des Stirnabschreckversuches zur Prufung derHarbarkeit von tiefer einhartenden Stahlen, Stahl Eisen 76(23):15701573 (1956) (in German).C.F. Jatczak, Effect of microstructure and cooling rate on secondary hardening of CrMoV steels,Trans. ASM 58:195209 (1965).C.B. Post, M.C. Fetzer, and W.H. Fenstermacher, Air hardenability of steel, Trans. ASM 35:85 (1945).G. Krauss, Steels Heat Treatment and Processing Principles, ASM International, Metals Park, OH, 1990. B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.J. Field, Calculation of Jominy end-quench curve from analysis, Met. Prog. 1943:402.W. Crafts and J.I. Lamont, Hardenability and Steel Selection, Pitman, London, 1949.E. Just, Formel der Hartbarkeit, Harterei Tech. Mitt. 23(2):85100 (1968).R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. Lubben, H. Rohloff, P. Schuler, V. Schuler, andH.J. Wieland, Errechnung der Hartbarkeit im Stirnabschreckversuch bei Einsatz und Vergutungsstahlen, Harterei Tech. Mitt. 47(3):183188 (1992).J.S. Kirkaldy and S.E. Feldman, Optimization of steel hardenability control, J. Heat. Treat. 7:5764(1989).J.M. Tartaglia, G.T. Eldis, and J.J. Geissler, Hyperbolic secant method for predicting Jominyhardenability; an example using 0.2 CNiCrMo steels, J. Heat. Treat. 4(4):352364 (1986).J.M. Tartgalia and G.T. Eldis, Metall. Trans. 15A(6):11731183 (1984).E. Just, Met. Prog. 96(5):8788 (1969).T. Lund, Carburizing Steels: Hardenability Prediction and Hardenability Control in Steel-Making,SKF Steel, Technical Report 3, 1984.Metals Handbook, 9th ed., Vol. 1, ASM International, Metals Park, OH, 1978, p. 492.M. Asimov, W.F. Craig, and M.A. Grossmann, Correlation between Jominy test and quenchedround bars, SAE Trans. 49(1):283292 (1941).J.L. Lamont, How to estimate hardening depth in bars, Iron Age 152:6470 (1943).D.V. Doane and J.S. Kirkaldy (Eds.), Hardenability Concepts with Applications to Steel, Proceedings of a Symposium, Chicago, Oct. 2426, 1977, The Metallurgical Society of AIME, 1978.T. Filetin, A method of selecting hardenable steels based on hardenability, Strojarstvo 24(2): 7581(1982) (in Croatian).T. Filetin and J. Galinec, Software programme for steel selection based on hardenability, Faculty ofMechanical Engineering, University of Zagreb, 1994. T. Filetin and B. Liscic, Determining hardenability of carburizing steels, Strojarstvo 18(4):197200(1976) (in Croatian).ASM Handbook, 9th ed., Vol. 4, Heat Treating, ASM International, Metals Park, OH, 1991, p. 287.A. Rose, The austenitizing process when rapid heating methods are involved, Der PeddinghausErfahrungsaustausch, Gevelsberg, 1957, pp. 1319 (in German). B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation andcalculation of heat transfer data. ASM Quenching and Distortion Control, Proceedings of FirstInternational Confererence On Quenching and Control of Distortion, Chicago, IL, 2225 Sept.1992, pp. 1726.N. Shimizu and I. Tamura, Effect of discontinuous change in cooling rate during continuouscooling on pearlitic transformation behavior of steel, Trans. ISIJ 17:469476 (1977).N. Shimizu and I. Tamura, An examination of the relation between quench-hardening behavior ofsteel and cooling curve in oil, Trans. ISIJ 18:445450 (1978).E. Just, Hardening and temperinginfluencing steel by hardening, VDI Ber. 256:125140 (1976) (inGerman).W. Gerber and U. Wyss, Hardenability and ability for hardening and tempering of steels, Von RollMitt. 7(2=3):1349 (1948) (in German). B. Liscic and T. Filetin, Computer-aided evaluation of quenching intensity and prediction ofhardness distribution, J. Heat. Treat. 5(2):115124 (1988). 2006 by Taylor & Francis Group, LLC.6Steel Heat Treatmentc idar Lis icBozCONTENTS6.16.26.3Fundamentals of Heat Treatment.............................................................................. 2786.1.1 Heat Transfer .................................................................................................. 2786.1.2 Lattice Defects................................................................................................. 2856.1.3 Application of TTT (IT) and CCT Diagrams ................................................. 2876.1.3.1 Isothermal Transformation Diagram ................................................ 2876.1.3.2 Continuous Cooling Transformation Diagram................................. 2886.1.3.3 Heat Treatment Processes for Which an IT or CCTDiagram May Be Used ..................................................................... 2926.1.3.4 Using the CCT Diagram to Predict Structural Constituentsand Hardness upon Quenching Real Workpieces ............................. 2936.1.3.5 Special Cases and Limitations in the Use of CCT Diagrams............ 2986.1.4 Oxidation......................................................................................................... 3016.1.4.1 Scaling of Steel .................................................................................. 3046.1.5 Decarburization ............................................................................................... 3066.1.5.1 The Effect of Alloying Elements on Decarburization ....................... 3086.1.5.2 Definitions and Measurement of Decarburization............................ 3096.1.6 Residual Stresses, Dimensional Changes, and Distortion ............................... 3126.1.6.1 Thermal Stresses in the Case of Ideal Linear-ElasticDeformation Behavior ...................................................................... 3146.1.6.2 Transformational Stresses ................................................................. 3156.1.6.3 Residual Stresses when Quenching Cylinders with RealElasticPlastic Deformation Behavior............................................... 3176.1.6.4 Dimensional Changes and Distortion during Hardeningand Tempering .................................................................................. 324Annealing Processes ................................................................................................... 3306.2.1 Stress-Relief Annealing.................................................................................... 3306.2.2 Normalizing..................................................................................................... 3346.2.3 Isothermal Annealing ...................................................................................... 3396.2.4 Soft Annealing (Spheroidizing Annealing) ...................................................... 3446.2.5 Recrystallization Annealing............................................................................. 3496.2.5.1 Grain Recovery ................................................................................. 3516.2.5.2 Polygonization................................................................................... 3516.2.5.3 Recrystallization and Grain Growth................................................. 352Hardening by Formation of Martensite .................................................................... 3556.3.1 Austenitizing.................................................................................................... 3556.3.1.1 Metallurgical Aspects of Austenitizing.............................................. 3556.3.1.2 Technological Aspects of Austenitizing ............................................ 364 2006 by Taylor & Francis Group, LLC.6.3.2 Quenching Intensity Measurement and Evaluation Based onHeat Flux Density ........................................................................................... 3746.3.3 Retained Austenite and Cryogenic Treatment................................................. 3846.3.3.1 Transforming the Retained Austenite ............................................... 3876.4 Hardening and Tempering of Structural Steels.......................................................... 3906.4.1 Mechanical Properties Required...................................................................... 3906.4.2 Technology of the Hardening and Tempering Process.................................... 3976.4.3 Computer-Aided Determination of Process Parameters.................................. 4026.5 Austempering ............................................................................................................. 407References .......................................................................................................................... 4136.1 FUNDAMENTALS OF HEAT TREATMENT6.1.1 HEAT TRANSFERHeat treatment operations require a direct or an indirect supply of energy into the treatedworkpieces and its subsequent removal to effect the heating and the cooling, respectively, ofthese pieces. Because this chapter deals only with heat treatment operations involving thewhole volume of treated workpieces, let us consider only the relevant heat transfer problems,not taking into account other heating methods connected to surface heat treatment operations. As an example, Figure 6.1 shows the temperature distribution on the cross section of aplate during heating (Figure 6.1a) and during cooling (Figure 6.1b).In heat treatment operations, when heating or cooling the treated workpieces, nonstationary temperature fields develop in which the temperature distribution changes with time.Through the surface F of the plate of thickness s (Figure 6.1), the heat flux Q is supplied(during heating) or extracted (during cooling):QdQdT lF, x 0, . . . , s=2dtdx(6:1)where T is temperature (K); t is time (s); l is heat conductivity (W/(m K)); F is surface area(m2); and dT/dx is the temperature gradient (K/m).FFt = t2 > t1Tut = t0t = t1 > t0t = t2 > t1t = t1 > t0t = t0QQQQTuXXS(a)S(b)FIGURE 6.1 Temperature distribution on the cross section of a plate (a) during heating and (b) duringcooling. t0, Beginning of temperature change; Tu, surrounding temperature; Q, heat flux; s, thickness ofthe plate. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.From the other side, based on the first law of thermodynamics,dQ r dx FCp zdT(6:2)where r is the density of the workpiece (kg/m3) and Cp the specific heat capacity of theworkpiece at constant pressure (J/(kg K)).From Equation 6.1 it follows thatd dQd2 Td dQ lF 2 dx dtdxdt dx(6:3)From Equation 6.2 it follows thatd dQdT rFCpdt dxdt(6:4)and for the one-dimensional heat flux, the time-dependent temperature distribution inside theworkpiece isdTl d2 Td2 Ta 22dtrCp dxdx(6:5)where a l/rCp (m2/s) is the temperature diffusivity.In the case of three-dimensional heat flux, Equation 6.5 reads!dTd2 T d2 T d2 T 2 a Tadtdx2 dy2dz(6:6)whered2d2d2 2 2dx2 dydzis the Laplace operator.Equation 6.5 and Equation 6.6 are the temperature conduction equations in which thetemperature diffusivity represents the amount of the time-dependent temperature change of aworkpiece because of nonstationary heat conduction.A heat flux dQ flowing through a surface of area F is, according to Fouriers heatconduction law, proportional to the temperature gradient at the relevant position:dQ ldTF dtdx(6:7)or, expressed as heat flux density per unit time (s) and unit surface (m2),q ldT l grad T [W=m2 ]dx(6:8)Equation 6.8 clearly shows that the temperature gradient is the driving force of the heat flux.The heat conductivity (l), as a proportional factor in this heat conduction equation, represents the influence of the materials properties on the heat transport. Table 6.1 gives approximate values for heat conductivity l (in W/(m K)) for selected materials.In the above equations, the heat conductivity l is assumed to be a constant value, but inreality it depends on temperature. Figure 6.2 shows the temperature dependence of heat 2006 by Taylor & Francis Group, LLC.TABLE 6.1Approximate Values for Heat Conductivity l, W/(m K)MetalsCopperAluminumBrassGray ironSteelLiquidsWaterOilStiff inorganic materialsChamotteBrickConcreteMineral wool0. (2020008C)H2 (2020008C)3501702308011558400.0260.110.180.75Source: From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, CarlHanser, Munich, 1987.conductivity for groups of steel. As can be seen, the biggest differences in heat conductivityamong different steel grades are at room temperature. Whereas for unalloyed steels the heatconductivity decreases with increasing temperature, for high-alloy steels it slightly increaseswith increasing temperature. At about 9008C (16528F), the value of l is almost the same forall steel grades. The specific heat capacity Cp depends also on temperature.The transport of thermal energy through a solid body, described by the heat conductionequation (Equation 6.8), extends naturally beyond the body surface; i.e., heat transfer takesplace between the body and its environment. This heat transfer is expressed as the amount ofheat exchanged between the surface of the body and its environment per unit surface area andper unit time. According to Newtons law of cooling, the amount of heat exchanged between abody and its environment depends on the difference between the body surface temperatureand the temperature of its environment. The relevant heat flux density isqdQ a (TK TU )dF dtfor T K > T U(6:9)where TK is the body surface temperature, TU is the temperature of the environment, and a isthe heat transfer coefficient, W/(m2 K).80Heat conductivity l, W/m Ka60b40cd2000200400600800Temperature T, 8C1000FIGURE 6.2 Temperature dependence of the heat conductivity l for selected steel groups. (a) Pure iron;(b) unalloyed steels; (c) low-alloy steels; (d) high-alloy steels. (From G. Spur and T. Stoferle (Eds.),Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.The actual conditions of the heat transfer in each case are represented by the relevant heattransfer coefficient a, which depends on1. shape and cross-sectional size of the bodyThe position of the body (standing or lying)The surface condition of the bodyThe physical properties of the bodys materialThe physical properties of the surrounding fluid (density, specific heat capacity,dynamic viscosity)6. The agitation (flow) rate of the surrounding fluidDuring every heating or cooling process, the temperature difference between the body surfaceand its environment becomes smaller with time, i.e., the exchanged heat quantity becomessmaller. The heat transfer coefficient a is therefore not a constant but varies with the bodysurface temperature.Gases, liquids, and vacuums are the environments in which heat transfer occurs duringheat treatment operations. Heat can be transferred by three different heat transfer mechanisms: heat conduction, heat convection, and heat radiation.Heat conduction (in fluids) is the heat transfer that occurs in a nonagitated liquid orgaseous medium between directly adjoining particles owing to a temperature gradient.Heat convection is directly connected to the movement (flow or agitation) of the heatcarrying fluid, because it is realized through movement of the fluid particles from one place toanother. Therefore heat convection is possible only in liquids and gases. The amount of heattransferred by heat convection in a gas also depends on the number of particles in the gas.Because this number depends on gas pressure, heat convection is proportional to gas pressure.If the only cause of particle movement is the difference in density caused by the temperaturedifference, the movement is called free or natural convection. When the movement of particlesof the fluid is caused by an outside force, the movement is called forced convection. Generally,free and forced convections take place simultaneously. The amount of free convectioncontributing to the heat transfer depends on the temperature gradient within the fluid, and thecontribution of the forced convection depends on the flow velocity, i.e., on the agitation rate.When an air stream passes toward a cylinder, the convective heat transfer coefficient aKcan be calculated, according to Eckstein [2], by using the formulaaK (4:64 3:49 103 DT )v 0:61[W=(m2 K)]D 0:39(6:10)where D is the diameter (m); DT is the temperature difference between air and cylindersurface; and v is the air velocity (m/s).The third heat transfer mechanism is heat radiation. Solid bodies, liquids, and gases canall transfer heat in the form of radiation. This kind of heat transfer does not depend on anyheat transfer carrier; therefore, it can take place in vacuum also. Heat radiation is in the formof electromagnetic waves whose length is in the range of 0.3500 mm. When radiation strikesthe surface of a body, part of it will be absorbed, part of it will be reflected, and the rest maypass through the body. Every body emits radiation corresponding to its temperature. Thebody that, at a certain temperature, emits or absorbs the largest amount of radiation is calleda blackbody. All other bodies emit or absorb less radiation than the blackbody. The ratio ofthe radiation of a body to that of a blackbody is called the emission-relation coefficient .The total heat flux density emitted by radiation from a body can be calculated accordingto the StefanBoltzmann lawq sT 4 2006 by Taylor & Francis Group, LLC.(6:11)where is the emission-relation coefficient, < 1.0; s is the StefanBoltzmann constant,s 5.67 108 W/(m2 K4); and T is the absolute temperature (K).If two bodies mutually exchange radiant heat, then not only is the warmer body emittingheat to the colder one, but the colder body is also emitting heat to the warmer body, so thatthe transferred heat consists of the difference of the amounts of heat absorbed by the twobodies. The total heat transferred by radiation from one body having a surface area F1 toanother solid body of any surface area can be calculated according to44Q 1;2 sF1 (T1 T2 )(6:12)where 1,2 is the emission-relation coefficient, which depends on the emission-relation coefficients of both bodies, their surface relation, and their mutual position in space; T1 is absolutetemperature of the emitting body; and T2 is absolute temperature of the absorbing body.In industrial furnaces, heat is transferred substantially by simultaneous heat convectionand heat radiation. The heat transferred by heat conduction (in fluids) is negligible comparedto the heat transferred by convection and radiation. When calculating the total heat transferred by both mechanisms, it is appropriate to express the heat transferred by radiation witha formula similar to Newtons lawq a (T1 T2 )(6:13)The heat transfer coefficient for radiation a can be calculated by combining Equation 6.12and Equation 6.13:a 1 , 2 s44T1 T2T1 T2(6:14)The total heat transfer coefficient is thena a k a(6:15)where ak is the heat transfer coefficient for convection.Table 6.2 gives, according to Eckstein [2], the average values of the heat transfer coefficient for cooling or quenching in liquid or gaseous media. This complex heat transfercoefficient depends in each case on many specific factors, discussed earlier, but also dependsstrongly on the surface temperature of the workpiece. It is a temperature-dependent andlocation-dependent value that changes during heat transfer as the body surface temperatureequalizes to the environments temperature. According to Eckstein [2], the complex heattransfer coefficient can increase 30 to 50 times between 50 and 15008C (122 and 27328F).At temperatures below 3008C (5728C), heat transfer by convection is predominant. Withincreasing temperature, heat transfer by radiation becomes more important, and at about8008C (14728F) it reaches 80% of the total heat transfer.Especially in operations that employ immersion quenching in liquids, where two-phaseheat transfer takes place with high heat flux densities under nonstationary conditions, theheat transfer coefficient value changes very much. Therefore nowadays when heat transfercalculations are carried out by computer a temperature-dependent function of the heattransfer coefficient instead of an average value should be used. One practical way to obtainthis function in each actual case is to measure the surface temperature of an adequatelyinstrumented probe (cylinder or plate of adequate dimensions) placed appropriately in thequenching tank and, from the measured surface temperature vs. time history, calculate thecorresponding heat flux density and heat transfer coefficient vs. temperature functions. 2006 by Taylor & Francis Group, LLC.TABLE 6.2Average Values of the Heat Transfer Coefficient a When Coolingor Quenching in Liquid or Gaseous Mediaa [W/(m2 K)]MediumFurnaceStill airMoving airCompressed airAirwater mixtureHardening oilLiquid leadWater1530407052058012003500Source: From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEBDeutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.Figure 6.3 shows, as an example, the heat transfer coefficient vs. surface temperature forquenching a stainless steel cylinder of 50-mm diameter 200 mm in still oil of 208C (688F) [3].If there is no adequate thermocouple to measure the surface temperature of the probe, thetemperature can be measured near the surface and, by using the inverse heat conductionmethod and an adequate mathematical procedure finite element method (FEM), the surfacetemperature of the probe can be calculated.To explain the dependence between the heat transfer conditions and the temperature fieldsin solid bodies, let us consider the heating of a plate of thickness s (see Figure 6.4). At thebeginning of heating (t 0), the plate has a temperature TK 0 and is suddenly transferred instanding position into a furnace, where the environmental temperature is TU. Equal amountsof heat are transferred from both sides of the plate. Because boundary conditions of the third2000Heat transfer coefficient a, W/m2K1800160014001200100080060040020000100200300400500600700800900oTemperature, CFIGURE 6.3 Heat transfer coefficient vs. surface temperature when quenching a stainless steel cylinderof 50-mm diameter 200 mm in still oil of 208C, calculated from the measured surface temperaturetime history. (From B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensityevaluation and calculation of heat transfer data, Proceedings of the First International Conference onQuenching and Control of Distortion, Chicago, IL, September 1992, pp. 1726.) 2006 by Taylor & Francis Group, LLC.la2laTTU1t3t2t1x20 s x12x0 < t1 < t2 < t3TK (x, t = 0) = 0+x1 + s2+xFIGURE 6.4 Change of the temperature distribution with time when heating a plate of thickness s,depending on different heat transfer conditions expressed by the ratio l/a. (From H.J. Eckstein (Ed.),Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie,Leipzig, 1987.)kind exist, the ratio between the heat conductivity and the heat transfer coefficient (l/a) gives apoint at temperature TU outside the plate. The straight line connecting this point and therelevant surface temperature is the tangent on the temperature distribution curve at the bodysurface. As time progresses, both the surface temperature and the temperature in the middle ofthe plate increase. The temperature gradient inside the body is different for different l/a ratiosand changes over time. If the heat conductivity l of the bodys material is small or the heattransfer between the environment and the body surface is large, the ratio l/a is small and heataccumulates in the surface region because the amount of heat transferred is greater than theamount transported by conduction into the bodys interior. The smaller the ratio l/a, the fasterthe surface temperature equalizes to the temperature of the environment. The relevant changesof temperature at points x1 and x2 and the value and the change in the temperature gradientover time are also greater. This can be seen in Figure 6.5 when comparing the curves for (l/a)1small to (l/a)2 big. If the heat conductivity l is big or the heat transfer coefficient a is small, i.e.,TUTemperatureTemperature gradient123Curve 4: T (x2) = f[(l/a)2, t ]465Curve 1: T (x1) = f[(l/a)1, t ]Curve 2: T (x1) = f[(l/a)2, t ]Curve 3: T (x2) = f[(l/a)1, t ]Curve 5: dTdxx1dTCurve 6: dxx1= f[(l/a)1, t ]= f[(l/a)2, t ]TimeFIGURE 6.5 Change of the plate temperatures at points x1 and x2 and temperature gradients at pointx1 in the cross section of the plate shown in Figure 6.4 when different heat transfer conditions (l/a)1 and(l/a)2, respectively, exist. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nded., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) 2006 by Taylor & Francis Group, LLC.the heat is transported from the surface to the core of the body faster than it is transferredfrom the environment to the bodys surface, the l/a ratio becomes big, and the temperature inthe interior of the body increases relatively faster than the surface temperature.Other factors that should be taken into account when analyzing such heat transferproblems are the shape and cross-sectional size of the body. As to the influence of differentshapes of workpieces it should be borne in mind that at constant heat transfer conditions andconstant thermal properties of the material and equal temperature of the environment, thetemperature change with time depends on the surface-to-volume ratio of the body. Thegreater the ratio is, the greater is the temperature change over time.6.1.2 LATTICE DEFECTSGenerally the lattice of a metal crystal contains imperfections, i.e., aberrations from a perfectatomic arrangement. These imperfections may be divided, from the geometrical standpoint,into the following categories:Zero-dimensional or point imperfectionsOne-dimensional or linearTwo-dimensional or superficialThree-dimensional or spatialThe most important lattice defects that occur in metals are shown schematically in Figure 6.6.Figure 6.6a shows a plane consisting of equal regular atoms a of which the spatial lattice isbuilt, with four different types of lattice point defects. At position b one atom is missing; thisdefect is called a vacancy. Atom c occupies a place between the regular places in the latticeand it is called an interlattice atom; in this case c is the same kind of atom as the regularatoms. In position d, a strange atom (of an alloying element) with a larger diameter has takenthe place of a regular atom; therefore it is called a substitutional atom. A practical example ofthis is a manganese atom dissolved in iron. In position e, a strange atom with a much smallerdiameter than the regular atoms of the lattice is inserted between regular atoms in a positionthat is not occupied by regular atoms. It is called an interstitial atom. A practical example ofthis is a carbon atom dissolved in iron. Both substitutional and interstitial atoms cause localdeformations and microstresses of the crystal lattice.Figure 6.6b shows at f a linear lattice defect. A row of atoms in the outlined atomic planeterminates at this point. If we imagine the outlined atomic plane as a section through a crystalthat stretches perpendicular to the plane shown, then the row of atoms terminating at fbecomes a half-plane of atoms that has been inserted between regular planes of atoms andends inside the crystal. The boundary line of the inserted half-plane of atoms that stretchesthrough the greater lattice region, perpendicular to the plane shown, is a linear lattice defectcalled an edge dislocation. Every edge dislocation is connected with characteristic deformations and microstresses of the lattice.The lattice defects g and h are superficial defects. The line gg represents schematically alow-angle grain boundary that consists of edge dislocations arranged regularly one under theother. The inserted half-plane of each edge dislocation terminates at the associated atomshown in black. The dashed area is a section through a low-angle grain boundary betweenneighboring parts of the crystal lattice that are inclined at a low angle to each other. The linehh represents a twinning boundary. It is characterized by the fact that the atoms on bothsides of the boundary are symmetrically distributed, and therefore neighboring parts of thecrystal lattice are completely equal, looking like twins in a mirror.Figure 6.6c shows at i a superficial imperfection (in the outlined plane) where a group ofatoms is missing. This zone of missing atoms could have developed by way of an accumulationof vacancies. It can be stretched to other planes of atoms perpendicular to the outlined one. Theimperfection k is a more or less irregular distribution of atoms between two neighboring parts of 2006 by Taylor & Francis Group, LLC.abgc(b)(a)hfdghe(c)klmi(d)nopqFIGURE 6.6 Lattice defects. (a) Lattice point defects; (b) linear and superficial lattice defects; (c)superficial lattice defects; (d) spatial lattice defects. a, Regular lattice atom; b, vacancy; c, interlatticeatom; d, substitutional atom; e, interstitial atom; f, edge dislocation; g, low-angle grain boundary; h,twinning boundary; i, vacancy zone; k, high-angle grain boundary; l, strange atoms zone; m, phaseboundary; n, precipitate; o, inclusion; p, microcrack; q, micropore. (From G. Spur and T. Stoferle (Eds.),Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)the crystal lattice with big differences in orientation, which interrupts the continuity of thelattice. It is called a high-angle grain boundary, or simply a grain boundary. The superficialimperfection at l is a section through a zone of strange atoms that stretches in two dimensionsperpendicular to the plane shown. The boundary plane m between two different lattices is calleda phase boundary, which is also a two-dimensional lattice defect.Figure 6.6d shows schematically the characteristic three-dimensional lattice defects. Inmany metal alloys, within the lattice of grains under specific thermodynamic conditions, newlattice regions with changed structure are formed. Such a lattice defect shown at n is called aprecipitate. The spatial imperfection at o is called an inclusion. Such inclusions, whichdevelop unfailingly during the production of alloys, are nonmetallic or intermetallic compounds. Like precipitates, inclusions have their own structure and phase and are separated bya phase boundary from the surrounding lattice. Microcrack is denoted by p, a spatialimperfection that is created by three edge dislocations that came to a phase boundary andformed a hollow among the three half-planes of the lattice. The hollow stretches perpendicular or at a slope to the plane shown. At q a sphere-like hollow inside the crystals lattice is 2006 by Taylor & Francis Group, LLC.shown; this is called a micropore. Such defects can originate from the accumulation of eithervacancies or gases.Of all the lattice defects discussed above, vacancies and edge dislocations are especiallyimportant in the heat treatment of metals. Vacancies enable neighboring atoms or substitutional atoms of alloying elements to change their positions and thus enable diffusion processes.The diffusion of interstitial atoms (e.g., a carburizing process) is possible without vacancies.Dislocations can move, increase in number, and accumulate. By lowering the share force (as aconsequence of the intermittent movement of the atoms), compared to the case of a perfect ironcrystal (whisker), dislocations facilitate the plastic deformation of the material.6.1.3 APPLICATIONOFTTT (IT)ANDCCT DIAGRAMSTimetemperaturetransformation diagrams for isothermal transformation (IT diagrams)and for continuous cooling transformation (CCT diagrams) are used to predict the microstructure and hardness after a heat treatment process or to specify the heat treatment processthat will yield the desired microstructure and hardness. The use of the either type of diagramrequires that the user be acquainted with its specific features, possibilities, and limitations. Isothermal Transformation DiagramFigure 6.7 shows an IT diagram of the low-alloy steel DIN 50CrV4. The regions of transformation of the structural phases ferrite (F), pearlite (P), and bainite (B) as positioned inthe timetemperature diagram (the abscissa of which is always in logarithmic scale) are validonly under conditions of fast quenching from the austenitizing temperature to the chosentransformation temperature and subsequent holding at that temperature. This is the way%C %Si %Mn %P%S %Cr %Cu %NiMethod ofmeltingMc QuaidEhnb.S-M.%V4Temperature, C0.43 0.41 0.82 0.041 0.015 1.22 0.14 0.04 0.11Austenitizing temperature = 880 CAC3 (0.25 /min)800AC1 (0.25 /min)Ferrite-startPearlite-start70022 Pearlite-finish27FP 35391%60099%50042Bainite-startB400394040MS99%4630051Bainite-finish200HardnessHRC10001101s60 102121038 15 30 60min1Time104105424 6 8 16 24hFIGURE 6.7 Isothermal transformation (IT) diagram of DIN 50CrV4 steel. (From A. Rose andW. Strassburg, Archiv. Eisenhuttenwes. 24(11/12):505514, 1953 [in German].) 2006 by Taylor & Francis Group, LLC.1.00.8M0.60.40.2005001000Time, s15002000FIGURE 6.8 Relation between the amount of transformed structure (M) and time in IT diagrams, according to Equation 6.16. (From H.P. Hougardy, Harterei-Tech. Mitt. 33(2):6370, 1978 [in German].)the diagram itself was developed. Therefore the IT diagram may be read only along theisotherms. The beginning and end of transformation of ferrite, pearlite, and bainite inisothermal processes take place according to the functionM 1 exp ( bt n )(6:16)where M is the fraction of phase transformed; t is time (s); and b 2 109 and n 3.Because, as shown in Figure 6.8, this function starts and ends very flat, the actualbeginning and end of transformation are difficult to determine exactly. Therefore, an agreement is reached, according to which the curve marking the beginning of transformationdenotes 1% of relevant phase originated, and the curve marking the end of transformation denotes 99% of the austenite transformed.Only the formation of martensite takes place without diffusion, instantly, depending onlyon the temperature below the Ms point. Hougardy [5] gave the following formula (valid forstructural steels for hardening and tempering) for this transformation:Ma 1 0:929 exp[ 0:976 102 (Ms T )1:07 ](6:17)where Ma is the amount of martensite, Ms is the martensite start temperature, and T is atemperature below Ms.Some IT diagrams, when read along the isotherms, enable the user to determine thepercentages of phases transformed and the hardness achieved. Figure 6.9, for example,shows that when the DIN 41Cr4 steel (austenitized at 8408C (15448F) with 5-min holdingtime) is fast quenched to 6508C (12008F) and held at this temperature, after 12 s ferrite startsto form. After 30 s the formation of pearlite begins. After 160 s the transformation iscompleted with 5 vol% of ferrite and 95 vol% of pearlite formed. The hardness achieved isabout 20 HRC. If a specimen of this steel is quenched to 3008C (5728F), instantly, 50% (v/v)of martensite will be formed.The accuracy of an IT diagram with respect to the position of isotherms can generally betaken as +108C (508F), and with respect to the time ordinates, as +10%. Cooling Transformation DiagramFigure 6.10 shows the CCT diagram of the same heat (as Figure 6.7) of the low-alloy DIN50CrV4 steel. 2006 by Taylor & Francis Group, LLC.C0.44ChemicalcompositionMn0.80Si0.22P0.0301000S0.023Cr1.04Cu0.17Mo0.04Ni0.26V<0.01Austenitizing temp. 840 CHolding time 5 min900Ac3800Ac1Temperature, C700FA2095P600317032500B38400 Ms30050%90%M2001000110102103104105106Time, sFIGURE 6.9 Isothermal transformation (IT) diagram of DIN 41Cr4 steel. (From H.P. Hougardy,Harterei-Tech. Mitt. 33(2):6370, 1978 [in German].)Bez. %C %Si %Mn %P %S %Cr %Ni %Cu %V10.43 0.41 0.82 0.041 0.015 1.22 0.04 0.14 0.11Method ofmeltingb.S.-M.Austenitizing temp. = 8808C2522FTemperature, 8C6004Grain size (ASTM) = 10 11800700MC QuoidEhn15 20108058511 30P25757678% Ferrite Ac3 (0.258 min)% Pearlite Ac1500400.. = Hardness HRCBMS3005 10 20 30 4030 3050200% Bainite% Martensite901005801011s57 57 53 52 46 4147 33 31 2560 102121034 8 15 30 60min1214201044 68h105106162412 3 4 6 10DaysFIGURE 6.10 Continuous cooling transformation (CCT) diagram of DIN 50CrV4 steel. (From A. Roseand W. Strassburg, Archiv. Eisenhuttenwes. 24(11/12):505514, 1953 [in German].) 2006 by Taylor & Francis Group, LLC.When comparing the curves for the start of transformation in CCT and IT diagrams forthe same heat and steel grade (Figure 6.7 and Figure 6.10), we found that in the CCT diagramthe curves are slightly shifted to longer times and lower temperatures. For example, in the ITdiagram of Figure 6.7, the shortest time to start the transformation of ferrite is 16 s at 6508C(12008F) and the corresponding time for bainite is 9 s at 4808C (9008F). In the CCTdiagram of Figure 6.10, however, the shortest transformation start time for ferrite is 32 sat 6208C (11508F) and the corresponding time for bainite is 20 s at 3808C (7168F). Thisindicates that in CCT processes the transformation starts later than in IT processes. The basisof this phenomenon is related to the incubation time and can be found in a hypothesis ofScheil [6].It should also be noted that with higher austenitizing temperature the curves denoting thestart of transformation of a particular phase can be shifted to longer times. Figure 6.11 showsthe CCT diagram of DIN 16MnCr5 steel after austenitizing at 8708C (16008F) (a), and afteraustenitizing at 10508C (19228F) (b). In the latter case, the regions of ferrite and pearlite areshifted to longer times. It is necessary, therefore, when using a CCT diagram, to ascertain thatthe austenitizing temperature used to develop the CCT diagram corresponds to the austenitizing temperature of the parts treated.A CCT diagram is developed in the following way. Many small specimens (e.g., 4-mmdiameter 2 mm for high cooling rates, and 4.5-mm diameter 15 mm for medium andlow cooling rates) are austenitized and cooled within a dilatometer with different coolingrates. Start and finish of transformation of relevant phases with each cooling curve arerecorded and these points are connected to obtain the regions of transformation for therelevant phases (see Figure 6.10). Therefore, a CCT diagram can be read only in the way inwhich it was developed, i.e., along the cooling curves. As can be seen from Figure 6.10, a singlephase structure occurs only in cases of very high cooling rates (martensite) and very slowcooling rate (pearlite). In all other cooling regimes a mixture of more structural phases results.How much of each phase such a mixture contains can be read in percentage from the numbersalong the cooling curve (usually marked in CCT diagrams of German origin). The numbers atthe end of each cooling curve denote the relevant hardness after quenching (usually in HRC(two-digit numbers) or in HV (three-digit numbers)). For example, as shown in Figure 6.10 forgrade DIN 50CrV4 steel, if cooling proceeds at the rate marked with , a mixture of 10%ferrite, 30% pearlite, 30% bainite, and (the rest) 30% martensite will result at room temperature, and the hardness after quenching will be 47 HRC. It should be noted that the part of thearea (region) of a phase that the cooling curve intersects is by no means a measure of theamount of transformed phase.Sometimes a CCT diagram can be supplemented with a diagram showing portions of eachstructural phase and hardness after quenching; see the lower part of Figure 6.12. The abscissaof this diagram denoting time enables the cooling time to 5008C (9328F) to be determined forevery cooling curve. To determine the portions of structural phases and hardness afterquenching, one should follow the relevant cooling curve until its intersection with the5008C (9328F) isotherm and from this point down along the vertical line read the phaseportions and hardness after quenching. For example, for cooling curve C, which intersects the5008C (9328F) isotherm at 135 s, the readings are 4% ferrite, 7% pearlite, 78% bainite, and11% martensite and a hardness of 34 HRC.It should be noted that every CCT diagram is exactly valid only for the heat of a steel thatwas used for its construction. The influence of different heats (having slightly differentcompositions) of the same grade of steel on the position of transformation curves in therelevant CCT diagram, as an example, is shown in Figure 6.13. 2006 by Taylor & Francis Group, LLC.Chemicalcomposition vol%CSiMnPSAl, ges.CrMoNiV0. temp. 8708C(Holding time 10 min) heated in 3 min900Ac3800Ac1Temperature, 8C700FA26002051BMs5606560353002572M2007100188423(a)2823 2750040065 72356550 60P62570412181268 260 229 207 182 16530501000Austenitizing temp. 10508C(Holding time 10 min) heated in 3 min900Ac3Temperature, 8C800Ac1F700A1600521250 5030P65352 10500400BMs601060 706865553003535M200100283425(b)00.71sTime422340101012632762902206250 195423103105104101001000minFIGURE 6.11 CCT diagrams of DIN 16MnCr5 steel (a) when austenitizing temperature is 8708C and(b) when austenitizing temperature is 10508C. (From F. Wever and A. Rose (Eds.), Atals zur Warmebehandlung der Stahle, Vols. I and II, Verlag Stahleisen, Dusseldorf, 1954/56/58.) 2006 by Taylor & Francis Group, LLC.C0.44ChemicalcompositionSi0.22MnPSCr0.80 0.030 0.023 1.041000Cu0.17Mo0.04Austenitizing temp. 8408CHolding time 8 min900Ac3800DTemperature, 8C700NiV0.26 <0.01ACE4600a8 1872500b82 c606266Ac140383450d400e316 1730058 78402005860010080362852 44 39 34201816MP60406040FB2020HRC00110102103104Hardness, HRCPortion of structure, %100105Time, sFIGURE 6.12 CCT diagram of 41Cr4 steel (top) with the diagram at bottom showing portions of eachstructural phase and hardness after quenching depending on the cooling time to 5008C. (From H.P.Hougardy, Harterei-Tech. Mitt. 33(2):6370, 1978 [in German].)As for IT diagrams, the accuracy of a CCT diagram, according to Hougardy [5], withrespect to the position of isotherms is +108C (508F) and with respect to time ordinates +10%of the relevant time. Treatment Processes for Which an IT or CCT Diagram May Be UsedTaking into account what was explained above about how IT and CCT diagrams can be read,Figure 6.14 shows the isothermal heat treatment processes for which only IT diagrams may beused. The first is isothermal annealing to obtain a coarse ferriticpearlitic structure, for bettermachinability (Figure 6.14a). In this case, the IT diagram gives the crucial information, theoptimum temperature at which annealing should take place to achieve the full transformationin the shortest possible time.The second process is isothermal transformation to bainite, i.e., the austempering process(Figure 6.14b). In this case, the IT diagram is used first of all to ascertain that the steel inquestion is applicable for this process, i.e., has enough hardenability (which means that itsstart of transformation curves are not too close to the ordinate). If this condition is fulfilled,the diagram enables the heat treater to select the appropriate temperature according to thehardness desired and read the minimum time needed at this temperature to achieve the fulltransformation to bainite. 2006 by Taylor & Francis Group, LLC.900800Heat 1 0.44% C and 1.20% CrHeat 2 0.41% C and 1.06% CrTemperature, 8C7006005004003002001000110102Time, s103104FIGURE 6.13 Influence of the difference in composition between two heats of DIN 41Cr4 steel on theposition of transformation curves in the relevant CCT diagram. (From H.P. Hougardy, Harterei-Tech.Mitt. 33(2):6370, 1978 [in German].)The third process is the martempering process (Figure 6.14c), an interrupted quenching ina hot bath to obtain the martensite structure with minimum stress and distortion. Theapplicability of a steel for the martempering process may be checked in the same way asabove. In this case the diagram gives information about the necessary temperature of the hotbath and the maximum time the parts can be immersed in it (in order to obtain onlymartensite) before they are taken out to be cooled in air.Figure 6.15 shows, as an example, the only three cases of continuous cooling for whichonly a CCT diagram may be used. The first case (Figure 6.15a) is direct quenching to obtainfull martensitic structure. In this case the diagram enables the user to determine the criticalcooling rate for the steel in question. The second case (Figure 6.15b) is a continuous slowcooling process, e.g., cooling in air after normalizing annealing. In this case the diagramenables the user to select the cooling rate required to yield the desired hardness of the ferriticpearlitic structure at room temperature. The percentage of ferrite and pearlite can be read asdescribed above if the diagram allows it. The third case (Figure 6.15c) represents anycontinuous cooling regime that results in more than two structural phases. In any of thesecases the diagram enables the user to determine the portion of each phase and the hardnessafter quenching. Using the CCT Diagram to Predict Structural Constituents and Hardnessupon Quenching Real WorkpiecesEach CCT diagram describes only those transformations of the structure that occur alongthe cooling curves of specimens used for its construction. This means that a CCT diagramis valid only for the cooling conditions under which it was constructed. The cooling law for thespecimens of small diameter and small volume that were used in constructing theCCT diagram can, according to Rose and Strassburg [4], be described by the exponentialfunction 2006 by Taylor & Francis Group, LLC.Ac3Ac1BMlog tAc3Ac1ArFPBMslog tAc3Ac1AFPterSurfaceTemperature, CMsCen(c)8007006005004003002001000PAceSurfa(b)9008007006005004003002001000FteCenTemperature, CTemperature, C800700600500400300200100(a) 0BMslog tFIGURE 6.14 Isothermal heat treatment processes for which only IT diagrams may be used. (a)Isothermal annealing; (b) austempering; (c) martempering.T T0 et(6:18)where T0 is the austenitizing temperature, a the heat transfer coefficient, and t the time.The exactness of the predictions of structural constituents (phases) and hardness valuesupon quenching depends on the extent to which the cooling law at different points in the crosssection of real workpieces corresponds to the cooling curves of specimens drawn in the CCTdiagram. Experimental work [4] using round bars of 50-mm diameter 300 mm with thermocouples placed 1, 5, 10, 15, and 25 mm below the surface showed that the cooling curves indifferent points of a round bars cross section correspond in form to the cooling curves in CCTdiagrams to the extent that the structural transformation, i.e., the resulting structuralconstituents and hardness values, can be compared. Figure 6.16 shows how a hardnessdistribution can be predicted by using this correspondence.If the temperaturetimescales of the measured cooling curves and the CCT diagram arethe same (as in Figure 6.16a and Figure 6.16b), then by using a transparent sheet of paper the 2006 by Taylor & Francis Group, LLC.Temperature, CAc3Ac1AV crit8007006005004003002001000(a)FPMsBMlog tAc3Temperature, C8007006005004003002001000(b)Ac1AFPBMsM220HV180HVlog tTemperature, CAc3(c)8007006005004003002001000Ac1A10F3MsBP70M30 HRClog tFIGURE 6.15 Heat treatment processes with continuous cooling for which only CCT diagrams may beused. (a) Direct quenching to obtain full martensitic structure; (b) slow cooling to obtain a ferritepearlite structure of required hardness; (c) continuous cooling regime where a mixed structure isobtained.measured cooling curves on CCT diagrams of different steel grades can be superimposed, andin this way steel grade can be selected that will develop the required structure and hardness atthe desired point of the cross section. The accuracy of such prediction from a CCT diagramdecreases as the radius of the workpieces cross section increases. According to Peter andHassdenteufel [8], sufficiently exact predictions are possible using CCT diagrams for roundbars up to 100 mm in diameter when quenching in oil and up to 150 mm in diameter whenquenching in water.It appears that the main problem in the practical use of CCT diagrams for predictionof structural constituents and hardness upon quenching is to establish exactly the coolingcurve for the specified point on the workpieces cross section. This can be done either bycalculation (if symmetrical parts and one-dimensional heat flow are involved and the boundary conditions are known) or experimentally (for asymmetric parts) by measuring the 2006 by Taylor & Francis Group, LLC.600IIIIIIIV6004000IIIIIIIVHardness, HRCTemperature, C8001 mm5 mm10 mm15 mm25 mmBelow surface20040200(c)0110(a)102Time, s103Core1020300Distance from surface, mm104Temperature, C800% Ferrite4020 FP% Pearlite156020 6523l lllll lV20Bl10 4030 5% Bainite5095% MartensiteHRCA6000400200MsM570110(b)52 46 36 27102Time, s103104FIGURE 6.16 Prediction of hardness distribution on a round bar cross section. (a) Cooling curvesmeasured at different points below surface, as indicated. (b) CCT diagram of the relevant steel withsuperimposed cooling curves from (a). (c) Hardness distribution on the bars cross section uponquenching, obtained by reading the hardness values from (b).temperaturetime history with a thermoelement. The correspondence between cooling curvesof real workpieces and cooling curves drawn on CCT diagrams also enables the reverse, todraw conclusions about the cooling history (curve) at a specified point of the cross section of aworkpiece of any shape and size based on metallographic analysis of the structure andmeasured hardness upon quenching.When CCT diagrams (of American origin) are used, the manner of predicting structuralconstituents and hardness is slightly different. For example, in Figure 6.17, instead of dilatometric cooling curves, cooling curves taken at different distances from the quenched end ofthe Jominy test specimen are superimposed.If one follows one of these cooling curves, e.g., the one for 19.1 mm (3/4 in.) from thequenched end (the heavier line in the diagram), one can read that after 25 s of cooling, theJominy specimen made of AISI 3140 steel at this distance from the quenched end starts todevelop ferrite, after 30 s, pearlite; and after 45 s bainite. After 90 s of cooling 50% of theaustenite is already transformed. After 140 s, when the temperature at this point has fallen to3158C (5998F), formation of martensite begins.The corresponding Jominy curve at the bottom of Figure 6.17a shows that this coolingcurve (at this Jominy distance) with the steel in question will yield a hardness of 48 HRC. Tocorrelate this hardness to different points of the round bars cross sections of differentdiameters, an auxiliary diagram (valid in this case only for quenching in moderately agitatedoil) such as that shown in Figure 6.17b should be used. From this diagram one can see that thesame hardness of 48 HRC can be met, after quenching in moderately agitated oil, 9 mm belowthe surface of a round bar of 75-mm diameter. 2006 by Taylor & Francis Group, LLC.Temperature871 16000.41 C, 0.86 Mn, 0.26 Si,1.28 Ni, 0.71 Cr760 14006.41/8 3/161/1612.7 19.33/81/225.43/41/4mm11/2in.F1A649 12001%3P1 12F+ A10%F+P+A4538 1000250%F+P+B+A427 80056A375%Ms316 60093 20012TransformedM+A austenite ferrite pearlite bainite martensite204 4001%1/1651010%2050Cooling time, s50%75%1001/4 3/81/2 3/42001/8 3/16500100075SHardness rockwell, C655545352515504081216202428323640in.166.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 mm(a)Distance from quenched end, 1/16 in.Distance belowsurface of barin. mm2 50 100 mm(4 in.)4511/24075 mm(3 in.)35301 50 mm(2 in.)25 38 mm(11/2 in.)20151/210525 mm(1 in.)125 mm(1/2 in.)000(b)510 151/22025 30 35 4011/21Jominy distance4550 mm2 in.FIGURE 6.17 (a) CCT diagram and Jominy hardenability curve for AISI 3410. (From Met. Prog.,October 1963, p. 134.) (b) Chart showing relationship between rate of cooling at different Jominydistances and rate of cooling in moderately agitated oil of round bars of 12.5100-mm diameter. (FromK.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) 2006 by Taylor & Francis Group, LLC.There also exist CCT diagrams of another type, developed by Atkins [9]; an exampleis given in Figure 6.18. These diagrams were developed by cooling and quenching roundbars of different diameters in air, oil, and water, recording their cooling curves in the centerof the bar, and later simulating these cooling curves in a dilatometric test in order to identifythe transformation temperatures, microstructures, and hardness. These diagrams thereforerefer only to the center of a bar. Instead of a timescale on the abscissa, these diagrams havethree parallel scales, denoting bar diameters cooled in air, quenched in oil, and quenched inwater. A scale of cooling rates (usually at 7008C (12928F)) in 8C/min is added.These diagrams are to be read only along vertical lines (from top to bottom), denotingdifferent cooling rates. For example, to determine the microstructure developed and resultinghardness in the center of a 10-mm bar of the steel in question when cooling it in air, one takesthe vertical line at 10-mm diameter on the scale for air cool (see Figure 6.18), starts in theaustenite region and proceeds downward. Transformation in this case (unalloyed steel gradewith 0.38% C) starts at 7008C (12928F) with the formation of ferrite, continuing to nearly 50%transformation at 6408C (11848F) when pearlite begins to form. At 5808C (10768F), a trace ofbainite is indicated before transformation is complete.If oil quenching of a 10-mm bar is now considered, the 10-mm position should be locatedon the oil-quenched bar diameter scale in Figure 6.18. Again starting in the top region andfollowing the vertical line down, it is seen that in this case bainite is the first phase to formfrom austenite at 5608C (10408F). At 3308C (6268F), after about 40% transformation, theremaining austenite transforms to martensite until the reaction is complete at 1508C (3008F).Similarly, the center of a water-quenched 10-mm diameter bar will transform to martensitestarting at 3608C (6808F) and finishing at 1508C (3008F).Relevant hardness values after quenching (and in some cases after tempering to differentspecified temperatures) can be read following the same vertical line further down into thehardness after transformation diagram.An examination of the left-hand side of the upper diagram in Figure 6.18 for the steel inquestion shows that martensite will form on air cooling with bars up to 0.18 mm in diameter, onoil quenching up to 8 mm in diameter, and on water quenching up to 13 mm in diameter.A special feature of this type of CCT diagram is that the hardenability of the steel can beassessed at a glance. Figure 6.19a is a CCT diagram for a very low hardenability steelpreviously rolled and austenitized at 9508C (12428F). It shows early transformation to ferriteand pearlite (even with oil and water quenching of smallest diameters). Figure 6.19b shows asimilar diagram for a high-hardenability steel previously rolled, softened at 6008C (11128F),and austenitized at 8308C (15268F). In this case the austenite changed predominately tomartensite and bainite over a wide range of bar diameters and quenching rates. Diagramsof this type representing 172 steel grades have been published in the British Steel Corporation(BSC) atlas [9]. Cases and Limitations in the Use of CCT DiagramsWhen dealing with carburized steels, one should be aware that, because of the big differencein carbon content between the core (%0.2%) and the case (%0.8%), the CCT diagram for thecase will be totally different from the one for the core of the same steel, as shown in Figure6.20 and Figure 6.21. The increased carbon content in the case increased the hardenability andcaused the pearlite and bainite regions to be shifted to much longer times. The ferrite regiondisappeared, and the Ms point was lowered. Cooling at the same rate results in differentportions of structural constituents and substantially different hardness values. 2006 by Taylor & Francis Group, LLC.Austenitized at 860 Cprevious treatment rolledAnalysis, wt%C900SiMn0.380.200.70SCrMoWaterquenchAc3NiAlNbVAir coolOilquench800700P0.020 0.020StartA10%50%90%Ac1FFinishP600BC500400300M2001001000500200100502010521C per minCooling rate at 750 C00.2mm 0.1Bardiameter50.5101012520201020100505010050150 200150 20010030030020050010002000 mmAir500500mm Oilmm Water800700Hardness aftertransformationHardenability bandBS 970 080 136600500HV400300As cooled60T 500 C 1 hT 600 C 1 hT 700 C 1 h50HRC40302020010100FIGURE 6.18 CCT diagram for rolled steel austenitized at 8608C. (From M. Atkins, Atlas of ContinuousTransformation Diagrams for Engineering Steels, British Steel Corporation, BSC Billet, Bar and RodProduct, Sheffield, U.K., 1977.) 2006 by Taylor & Francis Group, LLC.Another limitation in the use of CCT diagrams concerns cooling regimes with discontinuous change in cooling rate, as for example a delayed quenching in air followed by water or oilquenching. The left-hand part of Figure 6.22a shows the start of transformation as in theconventional CCT diagram for the steel in question. The right-hand part (Figure 6.22b) ofthis diagram holds for air cooling to approximately Ac1, followed by water quenching(a delayed quenching process). It shows a significant displacement of the ferrite and bainiteregions to longer times. Such a cooling mode enhances hardenability and results inhigher hardnesses than expected from the conventional CCT diagram for the same steel.The effect of discontinuous change in cooling rate is based on nucleation and on incubationtime before the change in cooling rate occurs and is theoretically explained by Shimizu andTamura [11].Austenitized at 9508Cprevious treatment rolled sofened 600CAnalysis, wt %CSi0.06900MnPSCrMoNiAlNbV0.30Ac3StartA10%50%80090%700FAc1FinishFPB600C500M400300200100500100020010050Cooling rate at 8508C20105218C per min0mm 0.1Bardiameter(a)0.20.55110252010205050100100 150 20030020050010002000 mmAir500mm Oil102050100150200 300500mm WaterFIGURE 6.19 Examples of (a) a low-hardenability and (b) a high-hardenability steel as depicted inCCT diagrams. (According to M. Atkins, Atlas of Continuous Transformation Diagrams for EngineeringSteels, British Steel Corporation, BSC Billet, Bar and Rod Product, Sheffield, U.K., 1977.) 2006 by Taylor & Francis Group, LLC.Austenitized at 830 C previous treatment rolled, softened 6008C 1 hAnalysis, wt %CSiMn0.400.250.60CrMoNi0.020 0.020 0.65PS0.552.55AlNbV900800Ac3A700 Ac1StartFC60050010%40050%90%300 MsBFinish200M100100050020010050Cooling rate at 700 C2010521C per min00.2mm 0.1Bardiameter(b)50.51012205501010020150 2005030010020050010002000 mmAir500mm Oil102050100150 200300500mm WaterFIGURE 6.19 (Continued)6.1.4 OXIDATIONOxidation takes place as an undesirable accompanying phenomenon during every heattreatment of metals in a noninert atmosphere. Chemical reactions that occur during theoxidation of a metal are generally expressed by the formula22x!Me O2 y Mex Oyy(6:19)where x and y denote integer numbers.The oxidation process proceeds at a set temperature spontaneously from left to right(according to formula 6.19). During this process the free enthalpy of the reaction products(GR) becomes smaller than the enthalpy of the original materials (GA), i.e., the difference canbe expressed as 2006 by Taylor & Francis Group, LLC.Chemicalcomposition %C0.22Si0.30Mn0.66PS0.018 0.011AlBCr0.049 <0.0005 0.561000Cu0.18Mo0.44N0.020Austenitizing temp. 10508CHolding time 15 min80071Temperature, 8CFA716918Ac3Ac1eAc1b712929P46Ni0.15360026B28Ms32048400M707969350200a)50000.1485142530034010b)270240102215103145180 165 155104105106Time, sFIGURE 6.20 The CCT diagram for the core (0.22% C) of DIN 20MoCr5 steel. (From H.J. Eckstein(Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)cChemicalcomposition %Si0.300.88Mn0.66P0.018S0.011AlBCr0.049 <0.0005 0.561000Cu0.18Mo0.44N0.020Ni0.15Austenitizing temp. 830 CHolding time 15 minAccm800Ac1aAc1bTemperature, C100100KA+KP60038 90562 70B4002002399580MsMRA30RA300 0.11RA3089088010RA3089510230 25 RA3890 885 870 420103390365104330 245220105106Time, sFIGURE 6.21 The CCT diagram for the case (0.88% C) of DIN 20MoCr5 steel. (From H.J. Eckstein(Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) 2006 by Taylor & Francis Group, LLC.900Air coolingrates, C/s800Temperature,8CAC11%1%7006000.6F+A5000.8B+F+AMs40030031.433 19 741 23 1756(a)(b)200Water quenching110100Time, s1000FIGURE 6.22 Cooling curves and CCT diagrams for a steel containing 0.20% C, 0.78% Mn, 0.60% Cr,0.52% Ni, and 0.16% Mo after austenitizing at 9008C. (a) Conventional CCT diagram and (b) variousair cooling rates to approximately Ac1 followed by water quenching. (From E.A. Loria, Met. Technol.,1977, pp. 490492.)DG0 GR GAfor GA > GR(6:20)If DG0 > 0, reaction 6.19 will take place from right to left, i.e., the metal oxide will be reduced.When the oxygen pressure (pO2) equals 1 bar, DG0 is called the free standard creatingenthalpy. Figure 6.23 shows the temperature dependence of the free standard creating enthalpy(DG0) for oxidation reactions of some metals.If (pO2) at temperature T differs from 1 bar, then the characteristic change of the freecreating enthalpy may be calculated as follows:DG DG0 RT ln pO2(6:21)where R is the universal gas constant and T is absolute temperature.As can be seen from Figure 6.23 for all of the metals represented except silver, the valuesof DG0 are negative with an increasing trend at higher temperatures. In the case of silver,DG0 0 at 1908C (3248F). At this temperature, equilibrium exists between Ag, O2, and Ag2O,i.e., the disintegration pressure of Ag2O has reached the oxygen pressure of 1 bar that wastaken as the basis. At higher temperatures the disintegration pressure of Ag2O becomeshigher and the metal oxide (Ag2O) will be disintegrated.From Figure 6.23 it can be concluded that the chosen metals, with the exception of silver,within the shown temperature range would form oxides. Because the oxidation takes place onthe surface, the oxide layer that is formed separates the two reaction partners, i.e., the metaland the oxygen. This oxide layer, which is material-specific, becomes thicker with time. Thereare several formulas expressing the dependence of the oxide layer thickness on time. Forhigher temperatures a parabolic law is usually used:y 2006 by Taylor & Francis Group, LLC.pA1 t(6:22)Free standard creating enthalpy G0, KJ/mol200OAg 204Ag+ O22uO 2C 2+ O24Cu2NiOO2Ni + 22004006002Zn800Ti +ZnO 4 Cr + O 223+ O2O2 2 Cr 2O 33TiO 22 Al 2O 334 Al + O 2MgO1000 32+ O22Mg1200050010001500Temperature T, 8C2000FIGURE 6.23 Temperature dependence of the free standard creating enthalpy of some oxidationreactions between 08C and the melting point of the relevant metal. (From G. Spur and T. Stoferle(Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)where y is the oxide layer thickness, A1 is a material-specific constant, and t is time. Thisparabolic law is valid for oxidation processes when the rate of oxidation depends on thediffusion of metal ions and oxygen ions through the oxide layer.When the oxide layer is porous, i.e., permeable for the gas, and therefore the metal andoxygen are not separated, build-up of the layer follows a linear law:y A2 t(6:23)where A2 is another material-specific constant.An oxide layer of a pure metal is constituted of uniform chemical compound if a singlevalency is involved, e.g., FeO. If more valencies are involved, the oxide layer consists ofsublayers with oxygen valencies increasing from the inside to the outside, e.g., FeO, Fe3O4,and Fe2O3, as shown in Figure 6.24. In Figure 6.24a, an oxide layer of pure iron, createdduring a 5-h annealing at 10008C (18328F), is shown. Relevant processes during developmentand build-up of the layer are schematically shown in Figure 6.24b. of SteelWhen metallic parts are heated above 5608C (10408F) (this is the temperature at which thecreation of wustite or FeO begins), after creation of the first part of the layer in the startingphase, the reaction follows by diffusion of Fe2 ions from the steel toward the outside and thediffusion of oxygen ions at the scalemetal interface toward the inside. As time passes, thelinear law of oxidation valid for the starting phase changes to a parabolic law.The growth of the oxide layer (scale) depends very much on the chemical composition ofthe steel. Different alloying elements, having different diffusion abilities, have differentinfluences on the oxidation process and the build-up of scale. The chemical composition ofthe original material on the surface is subject to changes. 2006 by Taylor & Francis Group, LLC.(a)(b)Fe2O3WstiteFeOIronFeMagnetiteFe3O4HematiteFe2O3Fe++OxygenO2O Fe++Fe+++eFeFeOaFe3O4ebecdFIGURE 6.24 Oxidation of pure iron. (a) Oxide layer and (b) processes during the build-up of the oxidelayer; a, transition of Fe2 2e from the metal into FeO; b, creation of FeO; c, creation of Fe3O4 andFe2O3; d, input of oxygen. (From. G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol.4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)According to their affinity for oxygen, the alloying elements in steel can be divided intothree groups with respect to their influence on the scaling process [2]:Group I contains those elements whose oxygen affinity is less than the affinity of therichest oxide compound wustite (FeO), e.g., Ni and Co. After saturation of the basicstite. Themetal with oxygen, the outer oxidation of iron begins with the creation of wualloying elements become richer at the scalemetal interface.Group II contains those elements whose oxygen affinity is greater than that of iron (Cr, Si,V, Al). After saturation of the basic metal with oxygen, inner oxidation begins. Becauseof the creation of internal oxides of alloying elements, a diffusion barrier builds upagainst the diffusion of metal and oxygen ions, hampering the development of scale.stite (Mo,Group III contains those elements whose oxygen affinity is similar to that of wuW). No inner oxidation takes place. The alloying elements become richer in the basicmetal at the scalemetal interface.Depth of the oxide layer, mm/yearFigure 6.25 shows the influence of Cr additions to a steel on the depth of scale (mm/year)at temperatures of 600, 700, and 8008C (1112, 1292, and 14728F). A particularly highoxidation resistance of steels may be achieved by Cr additions of 630% by mass.3224800 C16700 C8600 C00246Cr content, %8FIGURE 6.25 Influence of Cr on the oxidization of a steel at temperatures of 600, 700, and 8008C.(From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970.) 2006 by Taylor & Francis Group, LLC.6.1.5DECARBURIZATIONUnder conditions that cause the oxidation of iron, the oxidation of carbon is also to beexpected. Decarburization of a metal is based on the oxidation at the surface of carbon that isdissolved in the metal lattice. It should be noted that, depending on the carbon potential ofthe surrounding atmosphere, decarburization can take place independently of scaling. However, in heat treatment processes iron and carbon usually oxidize simultaneously. During theoxidation of carbon, gaseous products (CO and CO2) develop. In the case of a scale layer,substantial decarburization is possible only when the gaseous products can escape, i.e., whenthe equilibrium pressures of the carbon oxides are high enough to break the scale layer orwhen the scale is porous.The carbon consumed on the surface has to be replaced by diffusion from the inside.Hence the process of decarburization consists of three steps:1. Oxygen transport within the gas to the metal surface2. Carbon exchange at the gasmetal interface3. Diffusion of carbon within the metalGenerally the diffusion of carbon within the metal is the most important factor incontrolling the rate of decarburization, which after a short starting period follows a parabolictime law. When a mild steel is heated below 9108C (16708F), a surface layer of ferrite isformed that acts as a barrier to carbon transport owing to the very low solubility of carbon inferrite. Above 9108C (16708F) the steel remains austenitic throughout, and decarburizationbecomes severe. The model used to represent decarburization in the fully austenitic conditionis shown in Figure 6.26. The steel surface is continually oxidized to form a scale, while thecarbon is oxidized to form the gases CO and CO2. The scale is assumed to be permeable tothese gases, which escape to the atmosphere.The carbon content at a scalemetal interface is assumed to be in equilibrium with theoxygen potential of the scale, which at that position corresponds to the equilibrium betweeniron and wustite. The carbon concentration profile in the metal varies from the low surfaceconcentration to the original carbon content within the metal, as shown in Figure 6.26. Inusing this model, distances are measured from the original metal surface; the instantaneousscalemetal interface lies at the position x X at time t. This means that scaling has consumeda thickness X of metal during time t.SteelScaleCOriginal metalsurfaceC0CsOXxFIGURE 6.26 Model for decarburization in fully austenitic condition. (From ISI, Decarburization, ISIPublication 133, Gresham Press, Old Woking, Surrey, England, 1970.) 2006 by Taylor & Francis Group, LLC.To calculate the depth of decarburization, the distribution of carbon in the metal iscalculated by solving Ficks second law for the relevant boundary condition:dCd2 CD 2dtdxfor x > X(6:24)C C0 , x > 0; t 0(6:25)C Cs ,(6:26)x X; t > 0Equation 6.25 indicates that initially the carbon concentration was uniform throughout thespecimen; Equation 6.26 indicates that the carbon concentration at the metalscale interfaceis constant (in equilibrium with the scale). It is assumed that decarburization does not extendto the center of the specimen and that the diffusion coefficient of carbon in austenite isindependent of composition; enhanced diffusion down grain boundaries is neglected. Underthese conditions the solution at a constant temperature (for which the diffusion coefficient isvalid) readspC0 C erfc (x=2 2Dt)(6:27)C0 Cs erfc (kc =2D)1=2where C0 is the original carbon content of the metal, Cs is the carbon concentration at themetalscale interface, D is the carbon diffusion coefficient, t is time, erfc 1 erf (here erf isthe error function), and kc is the corrosion constant of the metal (kc X2/2t).Equation 6.27 provides the carbon content within the metal for x > X as a function of timeand position.The value of kc for the relevant steel in the relevant atmosphere is expressed askc 0:571 exp( 43,238=RT ) cm2 =s(6:28)Although the variation of the diffusion coefficient of carbon in austenite was ignored insolving Equation 6.24, it was found that the best agreement between calculated and measuredcarbon profiles was obtained when values relating to very low carbon content were used.Therefore, in this calculation, a diffusion coefficient for zero carbon content was used, whichreadsD(C 0) 0:246 exp(34,900=RT ) cm2 =s(6:29)A comparison of measured decarburization depths with values calculated using these datashowed that with 12 measurements of isothermal treatments between 1,050 and 1,2508C (1922and 22828F) for time between 900 and 10,800 s, the mean prediction was 97% of the measuredvalue [12]. It was found that the inner limit of the decarburized zone is placed, by metallographic examination, at the position where the carbon content is 92.5% of the original carboncontent.To ascertain the effect of scaling rate on decarburization, it may seem logical to try toreduce decarburization by reducing the oxidizing potential of the atmosphere. This is afallacy, as the carbon concentration at the metalscale interface is constant in equilibriumwith iron oxide as long as scale is present. However, the scaling rate can be affected bychanging the atmosphere, and this will affect the observed depth of decarburization. 2006 by Taylor & Francis Group, LLC.C, %0.900.925 C00.800.700.600.500.400.300. cm2/s0.4.1 1084.1 1070.05in0.070.09FIGURE 6.27 Effect of scaling rate on decarburization of an 0.85% C steel after 1.5 h at 10508C. Theposition where the carbon profile cuts the x-axis indicates the position of the scalemetal interface forthe three kc values. (From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking,Surrey, England, 1970.)To illustrate this, carbon profiles have been calculated and plotted in Figure 6.27 for a0.85% C steel heated for 1.5 h at 10508C (19228C). The carbon profiles are plotted relative tothe original metal surface, while the point at which a carbon profile cuts the x-axis indicatesthe position of the scalemetal interface for the related conditions. The curves refer to kcvalues of 0, 4.1 108, and 4.1 107 cm2/s. The depth of decarburization is determined bythe position at which C 0.925C0 (see the horizontal line drawn in Figure 6.27). From Figure6.27 it appears that with increased values of kc the scalemetal interface on the x-axis shiftsprogressively toward the inside of the metal, while the depth of decarburization (which is thehorizontal distance between the intersection of a carbon profile curve with the horizontal0.925C0 line and its intersection with the x-axis) is found to decrease as kc increases. Theseresults are shown in Table 6.3. This reveals an interesting situation where, by reducing theoxidation rate, the depth of decarburization is increased yet less metal is wasted.When scaling and decarburization take place simultaneously, decarburization is preventedduring the starting phase of scaling. It takes place substantially only after the equilibriumpressures of CO and CO2 increase at increased temperatures and the adhesion strength of thescale (because of its increased thickness) diminishes or the scale becomes porous. Effect of Alloying Elements on DecarburizationAlloying elements may affect decarburization due to their effect onTABLE 6.3Effect of Scaling Rate on DecarburizationTotal Depth of Metal Affected2Scaling Rate k (cm /s)04.1 1084.1 107Depth of Decarburized Layercmin.cmin.0.1190.1300.1500.0470.0510.0590.1190.1090.0840.0470.0430.033Source: From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970. 2006 by Taylor & Francis Group, LLC. transformation temperatureactivity of carbon in solutiondiffusion coefficient of carbon in solutionscaling characteristics of ironAlthough this subject is a complex one, even when considered only qualitatively, thefollowing statement is generally valid. Decarburization increases with (1) increased rate ofcarbon diffusion, (2) increased carbon activity, and (3) increased ferriteaustenite transformation temperature.A complication arises due to the fact that during scaling alloying elements tend toconcentrate either in the scale or in the metal at the scalemetal interface. In special cases,when strong carbide-forming elements are involved, decarburization may also be influencedby the rate of dissolution of carbides in the matrix. When the alloying elements are lessvaluable than iron, the possibility of oxidation arises; an external oxide layer may also beformed under circumstances that are normally protective to iron. If either type of oxideformation occurs, the concentration of the alloying element in solution is reduced at the metalsurface, and so the effect on carbon behavior will be altered correspondingly.When an external scale is formed, the effect of the alloying element on the scaling ratemust also be considered. If the scaling rate is increasing, then, in the absence of other factors,the observed depth of decarburization will be reduced.Although quantitative predictions are not possible, it is instructive to predict what theeffects of a few common alloying elements may be:Nickel will concentrate at the scalemetal interface and, although the scaling rate may notbe greatly affected, the solubility of carbon in the surface layers may be reduced, thusrestricting carbon diffusion outward and reducing the depth of decarburization.Manganese is taken into the scale in solid solution in the wustite and magnetite layers.Scaling rates are hardly affected, and any effect on decarburization will be restricted toits effects on carbon activity and the diffusion coefficient. Since manganese is denuded inthe surface layers of the metal, however, the effect may be only slight.Silicon also concentrates in the scale and forms fayalite, which reduces the scaling rate. Thisshould lead to deeper observed decarburization. Silicon also increases the activity ofcarbon and therefore increases the tendency of carbon to diffuse out to the scalemetalinterface. Thus the general effect expected of silicon is to increase decarburization.Chromium concentrates in the scale, forming spinel, depending on its concentration. Ingeneral, scaling rates are reduced. The formation of stable carbides introduces thepossibility of a slow carbide decomposition step into the mechanism. At the usualreheating temperatures and times, however, the chromium carbides may dissolve completely. In this case the effect of chromium would be to reduce the activity of carbon insolution, thus tending to reduce the rate of migration to the surface. There are thereforetwo conflicting factors. The lower scaling rate would tend to increase the observeddecarburization, whereas the reduction of carbon activity would tend to reduce it. Thelater factor may be expected to predominate and reduce decarburization. Definitions and Measurement of DecarburizationThe strength of a steel depends on the presence of carbides in its structure; therefore, loss ofcarbon from the surface softens and weakens the surface layers. In such a case the wearresistance is obviously decreased, and in many circumstances there can be a serious drop infatigue resistance. 2006 by Taylor & Francis Group, LLC.To avoid the real risk of failure or inferior performance of engineering components, it isessential to minimize decarburization at all stages in the processing of steel. This requiresinspection and the laying down of specifications for decarburization in various componentsand semiproducts. The decarburization limits in such specifications must be related to thefunction of the component and must enable checking by the use of agreed-upon preferablystandardized measuring techniques.There are several different definitions of decarburization. A rigorous definition is that thedepth of decarburization is the thickness of the layer in which the carbon content is less thanthat of the core, i.e., the distance from the surface to a boundary at which the carbon contentof the core is reached. This boundary corresponds to the asymptote of the graph of carboncontent vs. distance from the surface and is therefore somewhat diffuse and difficult to locatewith precision. Thus the depth of decarburization according to this definition is difficult tomeasure reproducibly. A functional definition is that the depth of decarburization is thethickness of the layer in which loss of carbon has a significant effect on properties that affectthe functioning of the final component. The limit of this layer can be expressed as a carboncontent or a hardness level. Finally, a practical definition of the depth of decarburization isthat it is the thickness of the layer in which the structure differs significantly from that of thecore. This definition is suitable when a metallographic examination is undertaken.There is a distinction between complete decarburization and partial decarburization.Complete decarburization leaves the surface layer entirely ferritic, which can be clearlydistinguished under the microscope. The depth of complete decarburization is the thicknessof the ferrite layer, i.e., the distance from the surface to the first particle of a second phase.In the zone of partial decarburization, the carbon content increases progressively from theferrite layer to the core and approaches the core composition asymptotically.The total thickness of the decarburization layer, i.e., the distance from the surface to theinner boundary of the core, is the total depth of decarburizationthe sum of the depths ofcomplete and partial decarburization. It should be noted also that the depth of decarburization may vary around the circumference of a component.A number of methods for measuring decarburization are available. The requirements thatsuch techniques have to meet are:1. Ability to measure a clearly defined depth of decarburization, e.g., compatibility withthe functional definition of the depth of decarburization2. Reproducibility of measurement3. Ease and convenience of measurementOptical metallography is the most useful and convenient method. A cross section of thecomponent or sample around the periphery is examined, and the depth of decarburization ismeasured from the surface to the practical boundaries of complete and partial decarburization. This method is suitable for ferritepearlite structures only.For the metallographic examination of high-speed steels, a method has been establishedthat depends on color staining by means of etching in alcoholic nitric acid (Nital). A polishedcross section of annealed high-speed steel is etched in 4% Nital. During the first 30 s in theetchant the specimen surface progresses through a gray color to a purplish-blue, which changessuddenly after about 60 s to a blue-green. Where the functional definition of decarburizationcalls for the development of the full hardness in the surface layers, the practical boundary isthe start of the general core structure, i.e., the edge of the blue-green zone.The arrest-quench method consists of austenitizing a very thin specimen in a neutralatmosphere or salt bath and quenching it in a hot bath held at an appropriate temperature.This is the Ms temperature corresponding to the carbon content at which it is desired toplace the boundary. The specimen is held at that temperature for about 5 s and is then 2006 by Taylor & Francis Group, LLC.water-quenched. During this short arrest the decarburized zone, which has an Ms temperatureabove the temperature of the bath, will partly transform to martensite; the core will remainaustenitic. As soon as martensite has formed in the decarburized zone, martensite needles willbegin to temper slightly. Thus, after water quenching the core will consist of fresh lightetching martensite, while the decarburized zone will contain dark-etching tempered martensite needles. A very sharp contrast is achieved at the boundary between the decarburized zoneand the core, and the boundary can be located with considerable accuracy at any desiredcarbon content below the original carbon content of the core.Figure 6.28 shows specimens taken from the same hot-rolled rod that have been arrestquenched at different temperatures to place the decarburization boundary at different carboncontents. The micrographs are placed on a graph of carbon content vs. depth of decarburization, and the carbon profile has been drawn through the microstructures. This technique iscompatible with the functional definition of the depth of decarburization based on a particular carbon content and gives very reproducible results.Microhardness measurement is a fairly convenient method for quantitatively accuratedeterminations of a functional decarburization limit by determining the variations of hardness with distance from the surface of the test piece. As this involves polishing a cross section,it is invariably preceded by a metallographic scan that facilitates the location of the best areafor the hardness survey.A graph of hardness vs. distance from the surface is plotted, and the deviation from thecore hardness can be detected. Chemical analysis of successive surface layers is the classicalreferee method for measuring decarburization. The sample has to be large enough to permitaccurate chemical analysis, and yet each surface layer must be fairly thin in order to give anadequate number of points on a graph of carbon content vs. distance from the surface. Thegraph of carbon content against distance can be used to indicate the first deviation from thecore composition or to locate any decarburization boundary. Complete decarburization is notvery easy to locate on this graph, because the carbon content of ferrite is too low for veryaccurate chemical analysis.Chemical analysis can be replaced by the carbon determination with a vacuum spectrograph. This has several advantages, particularly in speed and convenience and also because0.4Arrest temp.3308C3408CCarbon, %0.33508C3808C0.24408C0.1000. ofdecarburization, mmFIGURE 6.28 Depth of decarburization to various carbon contents, established by the arrest-quenchmethod. (From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey,England, 1970.) 2006 by Taylor & Francis Group, LLC.the sample size required is much smaller than for other methods. The only limitation is theneed to place the spark accurately on a flat area parallel to the original surface and at least15 mm in diameter. Successive layers have to be exposed by grinding, because the maximumdepth measured in one exposure is limited to about 500 mm.6.1.6 RESIDUAL STRESSES, DIMENSIONAL CHANGES, AND DISTORTIONResidual stresses are stresses in a body that is not externally loaded by forces and moments.They are in mechanical equilibrium within the body, and consequently the resultant force andthe resultant moment produced by residual stresses must be zero. Residual stresses areclassified, according to the area within which they are constant in magnitude and direction(i.e., in which they are homogeneous), into three categories:Residual stresses of the first kind are those homogeneous across large areas of thematerial, i.e., across several grains. Internal forces resulting from these stresses are inequilibrium with respect to any cross section, and mechanical moments resulting fromthese stresses are in equilibrium with respect to any axis. Any intervention in theequilibrium of forces and moments of a volume element containing such residual stresseswill change the elements macroscopic dimensions.Residual stresses of the second kind are those homogeneous across microscopically smallareas (one grain or subgrain region) and are in equilibrium across a sufficient number ofgrains. Macroscopic changes in the dimensions of a volume element possessing thesestresses may become apparent only if distinct disturbances of this equilibrium occur.Residual stresses of the third kind are those inhomogeneous across microscopically smallareas (within several atomic distances of single grains) and are in equilibrium acrosssubgrain regions. No macroscopic changes of the dimensions of the stressed material willresult when such equilibria are disturbed.Residual stresses of the first kind are called macroresidual stresses, and those of thesecond and third kinds are called microresidual stresses.Typical residual stresses of the third kind are stresses connected with dislocations andother lattice defects. An example of residual stresses of the second kind are stresses withingrains of a material consisting of two structural phases with different expansion coefficients.In practice, only residual stresses of the first kind are considered, and they are characterizedby the technological processes by which they originate. The main groups of residual stresses are:Casting residual stressesForming residual stressesWorking-out residual stressesHeat treatment residual stressesJoining residual stressesCoating residual stressesIn every stressed workpiece all three kinds of residual stresses are present.Figure 6.29 is a schematic presentation of all three kinds of residual stresses and theirsuperposition in a two-phase material after quenching. (RS IIII denote residual stresses ofthe first to third kinds, respectively.)Estimation of residual stresses in a workpiece is very important because they represent apreloading of the material. There is always a linear superposition of internal (residual) andexternal stresses, and the resulting stress affects the strength of the material and its deformation behavior. 2006 by Taylor & Francis Group, LLC.Phase AGrain olxxPhase BCut x xs RS Is RS IIs RS III+++s RSFIGURE 6.29 All three kinds of residual stresses in a two-phase material after quenching and their superposition (shown schematically). (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory andTechnology of Quenching, Springer-Verlag, New York, 1992.)In the case of a dynamic loading on a component, the residual stresses act as a constantpreloading. Tensile stresses decrease the fatigue strength, and compression stresses increase it.The fatigue strength of a component depends not only on the resulting stresses on the surfacebut also on the distribution of stresses across the section. Figure 6.30 shows schematically twocases with the same external stress (straight line c) and some fatigue strength (straight line a).The only difference is in the distribution of residual stresses (curve b). In case I, a high residualstress (compressive) on the surface rapidly decreases below the surface, while in case II theresidual stress, although smaller at the surface, decreases more slowly below the surface. Thecomponent can withstand the applied load only when the curve c b representing the sum ofthe external and residual stresses does not intersect the fatigue strength (straight line a). Incase I, in spite of higher compressive residual stresses at the surface, at a distance below thesurface the sum of external and residual stresses is higher than the fatigue strength, and acrack can be expected to form at this point. In case II, although the compressive residualstress is lower at the surface, its distribution below the surface is more favorable and the sumof external and residual stresses does not intersect the fatigue strength curve at any point.There is a further point to consider when dealing with resulting stresses. This is theirmultiaxis nature. In practice, the estimation of the sum of external and residual stresses iscomplicated by the difficulty of determining the direction of the stresses at the critical point ofthe workpiece. 2006 by Taylor & Francis Group, LLC.s+s+Case Icac+bcObCase IIODistance fromsurfacebac+bDistance fromsurfacessFIGURE 6.30 Schematic presentation of superposition of the external load and residual stresses at afatigue test. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEBDeutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) Stresses in the Case of Ideal Linear-Elastic Deformation BehaviorWhen a metallic body is heated or cooled, as soon as a temperature difference between thesurface and the core is established, residual stresses of the first kind occur. In heat treatment,quenching processes usually produce the biggest temperature gradients across the section andhence the greatest residual stresses. Let us therefore discuss thermal stresses due to local andtemporal differences in shrinking during quenching of ideal linear-elastic cylinders in whichno plastic deformation can arise.Transformation-free cooling of cylinders is accomplished by the development of a sequence of inhomogeneous temperature distributions, which, as a consequence of the thermalshrinking behavior, in turn cause locally and temporally different thermal strains andhence shrinking stresses. It is assumed that linear-elastic cylinders can elastically accommodate these stresses at all temperatures. At the beginning of quenching, the surface of such acylinder contracts more rapidly than its core. As a result, the surface zones of the cylinder aresubject to tensile stresses in the longitudinal and tangential directions, while radially compressive stresses are created, as shown in Figure 6.31. In order to establish equilibrium, thesestresses are counterbalanced by longitudinal, tangential, and radial compressive stresseswithin the core of the cylinder.ssrstllsrsrFIGURE 6.31 Thermal stresses in the surface zone and core of an ideal linear-elastic cylinder during rapid cooling (quenching). (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology ofQuenching, Springer-Verlag, New York, 1992.) 2006 by Taylor & Francis Group, LLC.TCoreSurfacet maxTt20 log tTmaxs shllog tSurface0log tCoreFIGURE 6.32 Top to bottom: timetemperature history, temperature difference between surface andcore, and development of longitudinal stresses during transformation-free quenching of an ideal linear elastic cylinder. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching,Springer-Verlag, New York, 1992.)Figure 6.32 shows the temperaturetime history at the very surface and at the core ofthe cylinder, the temperature difference between surface and core, and the developmentof longitudinal stresses during quenching of an ideal linear-elastic cylinder. The largesttemperature difference DTmax is attained at t tmax, where the slopes of temperaturetimecurves are identical for the core and the surface. Obviously, the surface reaches its maximumthermal stress before t tmax; the core, however, reaches its maximum later than t tmax.The magnitude of the developed longitudinal stresses depends on cylinder diameter, asshown in Figure 6.33 for cylinder diameters of 30, 50, and 100 mm, when the cylinders werequenched from 8008C (14728F) in water at 208C (688F). Because the maximum temperaturedifference between surface and core occurs later for the larger diameter cylinders, themaximum stresses also occur later for larger diameters. The longitudinal surface stressmaximum always occurs at t < tmax, whereas those of the core occurs later than tmax. Att < tmax, steep temperature gradients are present near the cylinder surface (see Figure 6.32),which cause high tensile stresses. In contrast, at t > tmax, relatively steep core temperaturegradients are established, which cause large compressive stresses in the core. Upon reachingthe temperature balance at 208C (688F) (t t20), the ideal linear-elastic cylinders are free ofresidual stresses. Transformational StressesLet us consider the development of pure transformational stresses in a material whosecoefficient of thermal expansion is zero. Furthermore, assume that if in the course ofquenching the martensite start temperature Ms is passed, complete martensitic transformation occurs, with corresponding volume increase. 2006 by Taylor & Francis Group, LLC.800D = 100 mmLongitudinal stress, N/mm260050400Surface30200020010050Core304006002-101 4 6 8 1246 8 10224 6 10Time t , sFIGURE 6.33 Dependence of longitudinal stresses at surface and core of ideal linear-elastic cylinders ontheir diameters, when quenched in water from 800 to 208C. Calculated for unalloyed steel with medium carbon content. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching,Springer-Verlag, New York, 1992.)The temperaturetime curves for the surface and core of a cylinder of such a material areshown in Figure 6.34. After passing the Ms temperature at time t t1, as a consequence oftransformation-induced volume increase, compressive transformational stresses develop atthe surface. These stresses within the surface zone must be compensated for by tensiletransformational stresses within the core of the cylinder. The magnitudes of both stressesincrease in the course of further surface cooling.TCoreSurfaceMst1s trlt2t 20 log tCore+0log tSurfaceFIGURE 6.34 Temperaturetime history and development of longitudinal transformation stresses, when quenching an ideal linear-elastic cylinder that transforms only to martensite. (From B. Liscic, H.M. Tensi,and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) 2006 by Taylor & Francis Group, LLC.s l sh + s l trCoreSurface0ShrinkingtiTransformationlog tCoreSurfaces l sh + sl trSurface0tit 20 log tCoreFIGURE 6.35 Combined thermal (shrinking) and transformation stresses during quenching of an ideal linear-elastic cylinder that transforms from austenite to martensite. (From B. Liscic, H.M. Tensi, andW. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.)When the core temperature reaches Ms at time t t2, a transformation-induced volumeincrease occurs in the core, which leads to a reduction of the tensile stresses present there. Thesurface compressive stresses are correspondingly reduced. After reaching temperature equalization at t t20, the same amounts of martensite are present across the whole cylinder, sothat finally a residual stress-free state is established. If, however, different amounts ofmartensite are formed within distinct areas, also under the idealized assumptions madehere, some transformational residual stresses will remain.In addition to the longitudinal stresses, tangential and radial residual stresses are causedby structural transformation. Within the surface zone, tangential compressive and radialtensile stresses are to be expected, while in the core all components should be tensile stresses.When thermal (shrinking) and transformational stresses act simultaneously duringquenching of an ideal linear-elastic cylinder that transforms from austenite to martensite,superposition of the two types of stresses occurs as shown in Figure 6.35. The upper graphshows the time dependence of the longitudinal components of thermal and transformationalstresses at surface and core. The lower graph shows the time dependence of the total stressafter the formal superposition of the two. The initiation of martensitic transformationimmediately reduces the absolute stress value within both core and surface.Further increasing martensitic transformation causes a stress inversion in both regions.Provided that the transformation occurs uniformly across the whole cylinder, at t t20 thetensile core stresses and the compressive surface stresses approach zero. Hence when temperature equalization is achieved in an ideal linear-elastic cylinder no residual stresses remain. Residual Stresses When Quenching Cylinders with Real ElasticPlasticDeformation BehaviorIn real practice there is no ideal linear-elastic deformation behavior as assumed above. Theyield strength (Ry) of metallic materials, which limits the elastic deformation range, is stronglytemperature-dependent and decreases with increasing temperature. 2006 by Taylor & Francis Group, LLC.600Rm16MnCr517CrNiMo6R y, R m, N/mm2400Ry2000200400600Temperature T, C8001000FIGURE 6.36 Yield strength (Ry) and tensile strength (Rm) of the steels DIN 16MnCr5 and DIN 17CrNiMo6 as a function of temperature. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theoryand Technology of Quenching, Springer-Verlag, New York, 1992.)At any temperature, plastic deformations will develop when stresses surpass thecorresponding yield strength. The ultimate tensile strength, which limits the uniaxialloading capacity of the material, is also temperature-dependent as shown in Figure 6.36for two low-alloy steels. During quenching of a cylinder, biaxial longitudinal andtangential stresses develop in its surface zone, whereas triaxial longitudinal, tangential, andradial stresses develop in the cylinder core. Plastic deformations can occur only if the localequivalent stresses equal or exceed the yield strength of the material at the correspondingtemperature.Equivalent stresses can be calculated according to various hypotheses. Assuming thevalidity of the van Mises criterion, the equivalent stress of a triaxial stress state, given bythe principal stresses s1, s2, s3, is1eq p [(1 2 )2 (2 3 )2 (3 1 )2 ]1=22(6:30)During quenching of a cylinder in its surface zone, s1 sl and s2 st, while in its cores1 sl, s2 st, and s3 sr.The condition for the onset of plastic deformation will be fulfilled when seq Ry. Thelocal shrinking and transformational stress components and consequently the equivalentstress (seq) depend on temperature, cooling conditions, geometry, and the mechanical andthermal properties of the material, and the yield strength (Ry) depends on temperature andthe structure of the material.The temperature dependence of the yield strength is obviously of particular importancefor the stresses that result upon quenching. Figure 6.37 shows the temperaturetime historyand development of yield strength for surface and core of a cylinder during quenching. Figure6.37a depicts the case of transformation-free cooling, and Figure 6.37b is valid for coolingwith martensitic transformation. To determine the occurrence of plastic deformations at anyinstant, the local yield strength must be compared with the local equivalent stress. Becauseplastic deformations never occur homogeneously over the whole cross section of the cylinder,residual stresses always remain after temperature equalizations. Plastic deformations can becaused by either thermal (shrinking) stresses or transformational stresses or by a combinationof the two. 2006 by Taylor & Francis Group, LLC.T, RyT, RyCoreCoreSurfaceSurfacesurfacecoreRyRysurfaceRyM1Iog tt max(a)MScoreRyt1(b)t2Iog tFIGURE 6.37 Temperaturetime history and development of yield strength for surface and core during quenching of a cylinder (a) without and (b) with martensitic transformation. (From B. Liscic, H.M. Tensi,and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) Thermal (Shrinking) Residual StressesFigure 6.38 shows the cooling curves for the surface and core of a cylinder during quenchingwithout martensitic transformation and the temperature- (and time-) dependent yieldstrengths, which at the same temperature are assumed to be identical for tensile and compressive loading. At the start of quenching, the surface temperature decreases faster than thecore temperature (Figure 6.38a). As a result, longitudinal tensile stresses develop at thesurface and compressive stresses develop at the core. If they were elastically accommodated,T, RyCoreSurfacesurfacecoreRyRy(a)log tt max sh, RlysurfaceSurface0Ryt maxCore(b)log tcoreRy shlSurfacet max0(c)log tCoreFIGURE 6.38 Longitudinal thermal (shrinking) residual stresses when quenching a cylinder. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, NewYork, 1992.) 2006 by Taylor & Francis Group, LLC.their development would be as shown in Figure 6.38b. However, because of the temperaturedependence of the yield strengths for surface and core, neither the surface nor the core canwithstand these stresses without plastic deformation, and so the surface zone is plasticallyextended and the core is plastically compressed. After the time t tmax, the temperature of thecore decreases faster than that of the surface, leading to a reduction of the magnitudes ofshrinking stresses in both regions. However, the stress values of core and surface reach zero atdifferent instants, as they can no longer coexist at the same time in a stress-free state becauseof plastic extension at the surface and plastic compression in the core. Upon further cooling,this extension and compression cause compressive and tensile stresses, respectively, which areopposed by those due to the temperature differences still existing between core and surface.These latter stresses ultimately vanish after reaching the temperature equalization at the endof quenching, and hence thermal (shrinking) residual stresses remain that are compressive atthe surface and tensile in the core, as depicted in Figure 6.38c. Transformational Residual StressesFigure 6.39 shows cooling curves for surface and core when quenching a cylinder that, uponcooling below the Ms temperature, transforms completely to martensite. For simplicity, it isassumed that no thermal (shrinking) stresses occur. Figure 6.39b shows the yield strengths forsurface and core, showing their strong increase with the onset of martensitic transformation.T, RyCoreSurfaceMscoreRysurfaceRyMf(a)s ltr, Ryt1t2log tcoreRyCore0t1t2log tSurface(b)surfaceRys ltrCore0(c)t1t2log tSurface FIGURE 6.39 Longitudinal transformation residual stresses when quenching a cylinder. (From B. Liscic,H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) 2006 by Taylor & Francis Group, LLC.The surface of the cylinder starts to transform to martensite at t t1. At that time thevolume expansion of the surface zone is impeded by the core not yet transformed. As a result,compressive transformational stresses are established at the surface that are compensated forby tensile stresses at the core. From Figure 6.39b it can be concluded that both areasplastically deform. In the course of further cooling, the tensile stressed core reaches Ms att t2. The immediate volume increase reduces both the tensile stresses of the core and thecompressive stresses of the surface. Due to the differently sized and opposing plastic deformations generated, the stresses at surface and core pass zero values at different time. Uponfurther cooling the still existing volume incompatibilities between surface and core createtransformational stresses of opposite sign to those that are produced by the plastic deformations. After reaching temperature equalization, compressive residual stresses remain in thecore and tensile residual stresses remain at the surface, as shown in Figure 6.39c.It should be noted also that transformation-induced plastic deformations that occur underlocal tensile or compressive stresses may enhance the local strains. Hardening Residual StressesWhen austenitized steel cylinders are quenched to room temperature, both thermal (shrinking) and transformational stresses develop, causing hardening residual stresses, which cannotbe described by simply superimposing the shrinking and transformational stresses. Of fundamental importance is the fact that any local martensitic transformation coupled with a volumeincrease always shifts the existing stress (irrespective of its sign) to more negative values. As areaction, for reasons of equilibrium, the unaffected material zones react with positive stresschanges. Structural transformations that occur in tensile-stressed material regions are therefore inevitable to reduce the stresses, while transformations that take place in compressivelystressed zones always enhance the (negative) values of the stresses. Consequently, because thethermal (shrinking) stresses of core and surface change sign in the course of cooling during thetime interval tc,0ts,0 as depicted in Figure 6.40a, the positions of the initiation time oftransformation at the surface (ts,i) and in the core (tc,i) relative to this time interval are ofkey importance for the hardening residual stresses that will remain at the end of quenching.The average time that passes before the quenching stresses invert ist0 (1=2)(ts,0 tc,0 )(6:31)Because for full-hardening steel cylinders the time for surface transformation (ts,i) alwaysoccurs earlier than the time for core transformation (tc,i), it is appropriate to distinguishbetween the following cases:t0 < ts,i < tc,i(Figure 6:40b)t0 % ts,i < tc,i(Figure 6:40c)ts,i < tc,i % t0(Figure 6:40d)ts,i < tc,i < t0(Figure 6:40e)Figure 6.40 shows schematically the development of longitudinal stresses as a function of timeand remaining longitudinal residual stress distributions across the section of cylinder 2006 by Taylor & Francis Group, LLC.Steel heat treatmentsls l RSSurfaceto0t c.0t solog tCore(a)slt 0 < t s, i < t c, i1Surface0s l RSlog tCore1(b)sls l RSSurfacet 0 t s, i < t c, i022log tCore(c)sls l RSSurface33t s, i < t c, i t 0 0log tCore(d)sls l RSSurface 44t s, i < t c, i < t 00log tCore(e)(Core)0(Surface)0.5AIA1.0FIGURE 6.40 Different possibilities of generation and development of hardening residual stresses (be)compared to pure thermal (shrinking) residual stresses (a), when quenching a cylinder with real elastic plastic deformation behavior. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technologyof Quenching, Springer-Verlag, New York, 1992.)specimens with real elasticplastic deformation behavior after complete temperature equalization at the end of the quenching process.Figure 6.40a shows a transformation-free quenching, and Figure 6.40bFigure 6.40e demonstrate the combined effects of thermal (shrinking) and transformational processes. Thenumbers 14 depict the initiation of transformation at the surface, while 14 represent that ofthe core. Figure 6.40b illustrates the case when both surface and core transform after t0. At the endof this cooling process, compressive stresses at the surface and tensile stresses in the core remain.Figure 6.40c illustrates the stress development in the case when the surface transformsslightly before t0 and the core transforms later. At the end of this cooling process, both core 2006 by Taylor & Francis Group, LLC.Temperatureand surface remain under compressive residual stresses, while the regions in between aresubjected to tensile residual stresses.Figure 6.40d illustrates the case when the surface transforms before t0 and the core atabout t0. At the end of this cooling process, tensile surface residual stresses and compressivecore residual stresses remain.Figure 6.40e illustrates the case when both surface and core transform before t0. In thiscase the start of transformation at the surface caused a rapid reduction of the tensile stresses.For reasons of equilibrium, the longitudinal stresses at the core must also change duringfurther cooling. Martensitic transformation in the core takes place when tensile stresses areacting there. This again causes stress inversions in the surface zone and in the core. At the endof this cooling process, tensile stresses at the surface and compressive stresses in the coreremain.When full hardening of equal-sized cylinders with different Ms temperatures is comparedwith respect to residual stress distributions, one finds that cylinders made of steels with lowMs temperatures show tensile surface residual stresses, whereas cylinders made of steel withhigh Ms temperatures give compressive surface residual stresses, as schematically illustrated inFigure 6.41.Because the high-temperature yield strength usually increases with decreasing Ms temperature, the largest tensile shrinking stresses develop at the surface of the steel with Ms,3 andthe smallest at the surface of the steel with Ms,1. The martensitic transformation, however,begins earliest for the steel with the highest and latest for the steel with the lowest Mstemperature.When high shrinking stresses and high Ms values act together, no secondary stressinversion occurs during further cooling, and ultimately compressive residual stresses remainwithin the surface zone.On the basis of the preceding discussion, the whole range of expected hardening residualstress distributions in quenched steel cylinders can be divided into three main groups, asMs,1Ms,2Ms,3Surface stresses0log tMs increasing320log t1FIGURE 6.41 Influence of different Ms temperatures on the development of surface residual stresses. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, SpringerVerlag, New York, 1992.) 2006 by Taylor & Francis Group, LLC.s l RSs l RSs l RSTransformationUnderUndercompressiontensionin thein thesurfacecore0TransformationUndercompressionin the surface00Shrinking typecTransition types00.5TransformationUnder Undertension compressionin the in thesurfacecorec1.00Transformation types0.51.0Ratio of cross sectionc0s0.51.0Cylinder diameters in mm for residual stresses ofSteelQuenching process850 CCk 45850 C20 C H2O60 C oilShrinking typeTransition type10030501530Transformation type510 FIGURE 6.42 Basic types of hardening residual stresses. (From B. Liscic, H.M. Tensi, and W. Luty(Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.)schematically illustrated in Figure 6.42. The arrows indicate how local transformations underexisting stress states will affect the residual stress distribution.It should be emphasized that the residual stress distributions that are created duringquenching of cylinders with different diameters but made of the same steel can be shiftedfrom the transformation type to the shrinking type with increasing cylinder diameter as wellas with the higher quenching intensity, i.e., higher cooling rates. Some cylinder diameter andquenchants are specified for the unalloyed steel DIN Ck45 where the basic residual stresstypes occur.The above statements and specifically the principle that local stresses are shifted to morenegative values due to transformation-induced volume increase also hold for all nonmartensitic transformations that are accompanied by volume changes. In the individual case, theeffect of volume changes on the final residual stress state depends on when the transformations start at the core and surface relative to time t0. Changes and Distortion during Hardening and TemperingAs a consequence of thermal (shrinking) stresses and transformational stresses, changes occurin both the shape and size of workpieces during hardening and tempering. Because there aremany factors that influence dimensional changes and distortion, the most difficult problem inpractice is to predict the amount of dimensional changes and distortion. It is likely thatcomputer modeling of the quenching process, which can account for the influence of allrelevant factors, will in the future enable more precise prediction. Let us therefore discussonly some basic mechanisms of dimensional changes and distortion during hardening andtempering. Influence of Thermal (Shrinking) StressesBecause of thermal (shrinking) stresses during quenching, generally all bodies whose shape isdifferent from a sphere tend by deformation to assume a spherical shape, which offers the 2006 by Taylor & Francis Group, LLC.least resistance during deformation. This means, at the practical level, that bodies that havethe shape of a cube will assume spherically distorted sides, bodies with the shape of a prismwill become thicker and shorter, and plates will shrink in area and become thicker. Thesedeformations are greater with greater temperature differences between the surface and thecore, i.e., with higher quenching intensity (which also corresponds to bigger differencesbetween the austenitizing temperature and the temperature of the quenchant); with greatercross-sectional size of the workpiece; and with smaller heat conductivity and smaller hightemperature strength of the material.The effect of thermal (shrinking) stresses can be studied in a low-carbon steel or anaustenitic steel, in which the martensitic transformation can be disregarded. Figure 6.43shows, according to Frehser and Lowitzer [15], the effect of different quenching intensitieson dimensional changes and distortion of plates made of low-carbon steel (0.10% C) afterquenching in water, oil, molten salt bath, and air. In Figure 6.43a the plate is solid, and inFigure 6.43b the plate has an inner square hole of 100 100 mm. The outer full lines denotethe original size of each plate. To illustrate the dimensional changes more clearly, they havebeen drawn to a larger scale (see the 0.4-mm scale). From this figure it is evident that the moredrastic the quench, the greater are the dimensional changes and distortion.Figure 6.44 shows that a greater difference between the austenitizing temperature and thetemperature of the quenchant causes greater dimensional changes and distortion. Figure 6.45shows the effect of the high-temperature strength of the material. The steel having the greatesthigh-temperature strength (18/8 steel) exhibits the highest dimensional stability. 0.4 mm change in size= 0.04 mm=20100(b)200(a)920 C/water920 C/oil920 C/molten bath at 220 C920 C/airFIGURE 6.43 Dimensional changes and distortion of plates made from low-carbon steel (0.10% C) aftercooling in water, oil, molten salt bath, and air. (From K.E. Thelning, Steel and Its Heat Treatment, 2nded., Butterworths, London, 1984.) 2006 by Taylor & Francis Group, LLC.6020060920 C/oil800 C/oil20200.04 mm=FIGURE 6.44 Effect of difference between the austenitizing temperature and the temperature of thequenchant on dimensional changes after quenching plates of low-carbon steel in oil. (From K.E.Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) Influence of Transformation StressesDuring heating and cooling, steels pass through various structural transformations accompanied by volume changes. These changes are usually studied by using a dilatometer and areregistered as changes in length of the specimen, as shown, for example, in Figure 6.46for eutectoid steel. During heating a continuous increase in length occurs up to Ac1, wherethe steel shrinks as it transforms to austenite. After the austenite formation is completed, thelength increases again. However, the expansion coefficient for austenite is not the same asthe expansion coefficient for ferrite.17% Cr steel920 C/water18/8 steel920 C/water2002000.1% C steel920 C/water4040Dimensional change scale: = 0.04 mmFIGURE 6.45 Dimensional changes and distortion after quenching steel plates of different compositionfrom 9208C in water. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984.) 2006 by Taylor & Francis Group, LLC.Change in lengthAc1T3T2T1+0200Ms0200400600Temperature, C8001000FIGURE 6.46 Dilatometer curves showing change in length during heating and rapid cooling of aeutectoid (0.8% C) steel. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984.)On cooling, thermal contraction takes place, and when martensite starts to form at the Mstemperature, the volume increases and the length of the specimen therefore increases. Aftercooling to room temperature, most martensitic steels contain some retained austenite, theamount of which increases with increased carbon content, with higher austenitizing temperature, and with the amount of some alloying elements dissolved during austenitization. Thelarger the quantity of retained austenite contained in the steel after hardening, the smaller theincrease in volume and in length of the specimen.Various structural constituents have different densities and hence different values ofspecific volume, as shown in Table 6.4. The amount of carbon dissolved in austenite, inmartensite, or in different carbides has a relatively strong effect on the specific volume as theformulas for calculating specific volume in this table indicate. When calculating the changes involume that take place during the transformations of different structural phases, the carboncontent must be taken into account, as shown in Table 6.5.TABLE 6.4Specific Volume of Phases Present in Carbon Tool SteelsPhase or PhaseMixtureAusteniteMartensiteFerriteCementitee-CarbideGraphiteFerrite cementiteLow carbon content martensite e-carbideFerrite e-carbideRange ofCarbon (%)020200.026.7 + 0.28.5 + 0.7100020.25202Calculated SpecificVolume at 208C (cm3/g)0.1212 0.0033 (%C)0.1271 0.0025 (%C)0.12710.130 + 0.0010.140 + 0.0020.4510.271 0.0005 (%C)0.1277 0.0015 (%C 0.25)0.1271 0.0015 (%C)Source: From K.E. Thelning, Steel and its Heat Treatment, 2nd ed., Butterworths, London, 1984. 2006 by Taylor & Francis Group, LLC.TABLE 6.5Changes in Volume during Transformation to Different PhasesTransformationSpheroidized pearlite ! austeniteAustenite ! martensiteSpheroidized pearlite ! martensiteAustenite ! lower bainiteSpheroidized pearlite ! lower bainiteAustenite ! upper bainiteSpheroidized pearlite ! upper bainiteChange in Volume (%)4.64 2.21 (%C)4.64 0.53 (%C)1.68 (%C)4.64 1.43 (%C)0.78 (%C)4.64 2.21 (%C)0Source: K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984.Taking as a basis the proportions of martensite and austenite, together with the amount ofcarbon dissolved therein, and using the data from Table 6.5, one can calculate the changes involume that occur during hardening. If the steel contains undissolved cementite, this volumehas to be deducted during the calculation. The following equation should be used:DV 100 Vc VaVa 1:68C ( 4:64 2:21C )100100V(6:32)where DV/V is the change in volume in %, Vc is the amount of undissolved cementite in vol%,Va is the amount of austenite in vol%, 100 Vc Va is the amount of martensite in vol%, andC is the carbon dissolved in austenite and martensite, respectively, in % by weight.The increase in volume during martensitic transformation depends not only on the carboncontent but also on the kind and amount of alloying elements in the steel. Consequently,different groups of steels undergo different changes in volume during hardening. The unalloyed water-hardening steels experience the greatest volume changes, followed by low-alloyoil-hardening steels, while the high-alloy ledeburitic Cr alloy steels show the least volumeincrease during hardening, as shown in Figure 6.47. The austenitizing temperature, as mentioned earlier, has an influence on the amount of retained austenite after hardening. Becausethe retained austenite producing volume contraction (compared to the original volume)counteracts the volume increase caused by martensitic transformation, the austenitizingtemperature may influence the volume changes during hardening.It should also be noted that engineering steels are not isotropic materials (because of therolling process they have undergone), which means that the linear change occurring duringhardening will not be the same in the direction of rolling as in the direction perpendicular to it. Dimensional Changes during TemperingDuring tempering, relaxation as well as structural transformations occur, which change thevolume of the hardened steel and its state of stress. Martensite decomposes to form ferrite andcementite, which implies that there is a continuous decrease in volume. The continuousdecomposition of martensite during tempering causes at the same time a continuous reduction in the state of stress. Figure 6.48 is a schematic presentation of the effect of changes ofstructural constituents on the volume changes during tempering of a hardened steel. Thedashed curves represent increases in volume during different tempering stages. The retainedaustenite, which in carbon steels and low-alloy steels is transformed to bainite in the secondstage of tempering at about 3008C (5728F), results in an increase in volume. 2006 by Taylor & Francis Group, LLC.Water-hardening steelse.g., DIN C100Oil-hardening steelse.g., DIN 90MnV8Air-hardeningsteelse.g., DIN X210Cr120.6 0.4 0.20 0.2 0.4 0.6Change in vol%0.81.0FIGURE 6.47 Volume changes of different steels during hardening when martensite is formed across thewhole section. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEBDeutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)When high-alloy tool steels are tempered at 5006008C (93211128F), very finely distributed carbides are precipitated. This gives rise to a stress condition that results in increasedhardness and greater volume. Simultaneously with the precipitation of carbides the alloycontent of the matrix is reduced, which implies that the Ms point of the retained austenite willbe raised to higher temperatures. During cooldown from the tempering, the retained austenitewill transform to martensite, which also results in an increase in volume.Figure 6.49 shows changes in length for different steels as a function of temperingtemperature. For low-alloy steels (105WCr6, see curve 1 of Figure 6.49), one can easilyrecognize the particular tempering stages. At low tempering temperatures (first temperingstage), a volume contraction takes place as a consequence of e-carbide precipitation. Athigher tempering temperatures (second tempering stage), transformation of the retainedIncrease in volumeDecompositionof martensiteto ferriteand cementiteRetained austeniteto martensiteRetained austeniteto bainite0100200Carbideprecipitation300400500600700800Tempering temperature, CFIGURE 6.48 Schematic presentation of the effect of changes of structural constituents on volumechanges during tempering of hardened steel. (From K.E. Thelning, Steel and Its Heat Treatment, 2nded., Butterworths, London, 1984.) 2006 by Taylor & Francis Group, LLC.Change in length, % IND IND IND IND IND IN105WC r640C rMOV21.14210C rW46X 100C rMoV 5.150N i C r13165C rMoV 46600Tempering temperature, CFIGURE 6.49 Change in length of different steels during tempering as a function of tempering temperature. (Designation of steels according to DIN.) (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)austenite again causes a certain volume increase, and in the third tempering stage theprogressive decomposition of martensite leads to the volume decrease.For high-alloy tool steels (e.g., 210CrW46, curve 3 of Figure 6.49), a stabilization ofaustenite is evident, so that the effect of the volume increase (due to austenitebainite oraustenitemartensite transformation) takes place only at higher temperatures. In most cases,as can be seen from Figure 6.49, a reduction in length, i.e., a volume decrease, can be foundafter tempering.It should be noted that the changes in length shown in Figure 6.49 represent only the orderof magnitude of the expected changes, because the actual value depends in each case on thespecific heat treatment conditions. The austenitizing temperature, which determines theamount of carbon dissolved and the amount of retained austenite, has a strong influenceon expected volume changes. PROCESSESSTRESS-RELIEF ANNEALINGStress-relief annealing is an annealing process below the transformation temperature Ac1,with subsequent slow cooling, the aim of which is to reduce the internal residual stresses in aworkpiece without intentionally changing its structure and mechanical properties.Residual stresses in a workpiece may be caused by1. Thermal factors (e.g., thermal stresses caused by temperature gradients within theworkpiece during heating or cooling)2. Mechanical factors (e.g., cold-working)3. Metallurgical factors (e.g., transformation of the microstructure)In processes that involve heat, residual stresses are usually caused by the simultaneousexistence of thermal and transformational stresses (e.g., during the solidification of liquidmetals, hot forming, hardening, or welding). Thermal stresses are always directly proportional to the existing temperature gradient, which further depends on the cross-sectional sizeand on the heating or cooling rate.In workpieces made of steel, for the above reasons, local residual stresses may amount tobetween about 10 N/mm2 and values close to the yield strength at room temperature. Theconsequences of residual stresses may include1.2.3.4.Dimensional changes and warpage of the workpieceFormation of macroscopic and microscopic cracksAsymmetric rotation of shaftsImpairment of the fatigue strength of engineering components 2006 by Taylor & Francis Group, LLC.Residual stresses in a workpiece can be reduced only by a plastic deformation in themicrostructure. This requires that the yield strength of the material be lowered below thevalue of the residual stresses. The more the yield strength is lowered, the greater the plasticdeformation and correspondingly the greater the possibility or reducing the residual stresses.The yield strength and the ultimate tensile strength of the steel both decrease with increasingtemperature, as shown in Figure 6.50 for a low-carbon unalloyed steel. Because of this, stressrelief annealing means a through-heating process at a correspondingly high temperature. Forplain carbon and low-alloy steels this temperature is usually between 450 and 6508C (842 and12008F), whereas for hot-working tool steels and high-speed steels it is between 600 and7508C (1112 and 13828F). This treatment will not cause any phase changes, but recrystallization may take place. Tools and machine components that are to be subjected to stress-reliefannealing should be left with a machining allowance sufficient to compensate for any warpingresulting from stress relief.When dealing with hardened and tempered steel, the temperature of stress-relief annealingshould be about 258C (778F) below that used for tempering. If the tempering temperature wasquite low, after stress-relief annealing quite a high level of residual stresses will remain. Insome other cases, for instance with a gray iron, the maximum temperature of the stress-reliefannealing should be limited because of possible strength loss. Therefore gray iron must not bestress-relief annealed above 5508C (10228F).In the heat treatment of metals, quenching or rapid cooling is the cause of the greatestresidual stresses. A high level of residual stress is generally to be expected with workpieces thathave a large cross section, are quenched at a high cooling rate, and are made of a steel of lowhardenability. In such a case high-temperature gradients will arise on the one side, and on theother side structural transformations will occur at different points of the cross section atdifferent temperatures and different times. In contrast to heat treatment processes with continuous cooling, processes with IT (e.g., austempering) result in a low level of residual stresses.To activate plastic deformations, the local residual stresses must be above the yieldstrength of the material. Because of this fact, steels that have a high yield strength at elevatedtemperatures can withstand higher levels of residual stress than those that have a low yieldstrength at elevated temperatures.5040800A3020600Rm10Elongation, %Yield strength and ultimate tensilestrength, MPa10000400sso200ssu0200 1000100 200 300 400 500 600Temperature, CFIGURE 6.50 Change in some mechanical properties of low-carbon unalloyed steel with increasingtemperature, according to Christen. A, Elongation; Rm, ultimate tensile strength; sso, upper yieldstrength; ssu, lower yield strength. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung vonStahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) 2006 by Taylor & Francis Group, LLC.Increase of the yield strength, N/mm2120Mo10080V60TiMn40Cu20Cr0300Ni350400450Temperature, C500550FIGURE 6.51 Increase in yield strength at elevated temperatures when 0.5% of each alloying elementindicated is added to an unalloyed steel. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)The level of yield strength at elevated temperatures depends on the alloying elements inthe steel. Figure 6.51 shows the increase in yield strength at temperatures of 3005508C (57210228F) when 0.5% of each element was added to an unalloyed steel. It can be seen from thisdiagram that additions of Mo and V are most effective in increasing the yield strength atelevated temperatures.To reduce residual stresses in a workpiece by stress-relief annealing, a temperature mustbe reached above the temperature corresponding to the yield strength that is adequate to themaximum of the residual stresses present. In other words, every level of residual stress in aworkpiece corresponds to a yield strength that in turn depends on temperature. In addition totemperature, soaking time also has an influence on the effect of stress-relief annealing, i.e., onthe reduction of residual stresses, as shown in Figure 6.52.The relation between temperature and soaking time during stress-relief annealing can bedescribed by Hollomons parameter:P T (C log t)(6:33)where P is Hollomons parameter (heat treatment processes with the same Hollomon parameter value have the same effect), C is the HollomonJaffe constant, T is temperature (K),and t is time (h).The HollomonJaffe constant can be calculated asC 21:3 (5:8 % carbon)(6:34)Figure 6.53 shows (according to LarsonMiller method) calculated values of the yield strengthat elevated temperatures (for 0.2% strain) for three grades of alloyed structural steels forhardening and tempering (designations according to DIN). Using this diagram, the abscissaof which represents the actual Hollomon parameter P, knowing the temperature and time of thestress-relief annealing, one can read off the level of residual stresses that will remain in theworkpiece after this annealing process, i.e., the level up to which the residual stresses will bereduced by this stress-relief annealing. If, for instance, for DIN 24CrMoV5.5 steel, a 2006 by Taylor & Francis Group, LLC.10Reduction of residual stresses, %2030401h10 h24 h48 h5060708090100200 300 400 500 600Temperature, C700FIGURE 6.52 Effect of soaking time (at different temperatures) of stress-relief annealing on thereduction of residual stresses for hardening and tempering steels. (From G. Spur and T. Stoferle(Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)400200Cr30V5NiMo10080.11605040rM7.45.5oVMoiCrab20C243028NYield strength or minimum residual stressafter stress-relief annealing, N/mm2P = T(20+log t)103h 1015Holding time, t(a)(b)0.1510201617 18 19 20 21Hollomon's parameter P2223550 600 650 700550 600550650700600 650550 600 650Temperature T, C700700FIGURE 6.53 Yield strength at elevated temperatures (for 0.2% strain) calculated according to theLarsonMiller method for three grades of alloyed structural steels for hardening and tempering(designations according to DIN). (a) Calculated values and (b) experimentally obtained values. (FromG. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser,Munich, 1987.) 2006 by Taylor & Francis Group, LLC.temperature of 6008C (11128F) and a soaking time of 10 h are chosen for stress-relief annealing,the residual stresses will, after this annealing, be reduced to a maximum of 70 N/mm2. Highertemperatures and longer times of annealing may reduce residual stresses to lower levels, as canbe seen from Figure 6.53.As in all heat treatment processes where Hollomons parameter is involved, selection of ahigher temperature may dramatically shorten the soaking time and contribute substantially tothe economy of the annealing process.Dealing with structural steels for hardening and tempering, the stress-relief process and thetempering process can be performed simultaneously as one operation, because Hollomonsparameter is also applicable to tempering. In such a case the stress-relief diagram may be usedin combination with the tempering diagram to optimize both the hardness and the level ofreduced residual stresses.The residual stress level after stress-relief annealing will be maintained only if the cooldown from the annealing temperature is controlled and slow enough that no new internalstresses arise. New stresses that may be induced during cooling depend on the cooling rate, onthe cross-sectional size of the workpiece, and on the composition of the steel. Figure 6.54shows the effect of cooling rate and cross-sectional diameter of forgings made of a CrMoNiVsteel on the level of tangential residual stresses after stress-relief annealing.A general conclusion about stress-relief annealing is the following: In the temperaturerange 4506508C (84212008F), the yield strength of unalloyed and low-alloyed steels islowered so much that a great deal of residual stress may be reduced by plastic deformation.The influence of the steel composition on the level of residual stresses after annealing can beconsiderable. While unalloyed and low-alloy steels with Ni, Mn, and Cr after stress-reliefannealing above 5008C (9328F) may get the residual stresses reduced to a low level, steelsalloyed with Mo or Mo V will retain a much higher level of the residual stresses after stressrelief annealing at the same temperature because of their much higher yield strength atelevated temperature.6.2.2NORMALIZINGmmm600mm.800m10080=1000mm120DiaTangential residual stresses, N/mm2Normalizing or normalizing annealing is a heat treatment process consisting of austenitizingat temperatures of 30808C (861768F) above the Ac3 transformation temperature (for60040mm0m20402000102030405060Average cooling rate to 400, C/h80FIGURE 6.54 Tangential residual stresses in a CrMoNiV alloy steel depending on the cooling rate andcross-section diameter. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2,Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.Temperature, C1000bAc3c800Ac1a600d4002000TimeFIGURE 6.55 Timetemperature regime of normalizing. a, Heating; b, holding at austenitizing temperature; c, air cooling; d, air or furnace cooling. (From G. Spur and T. Stoferle (Eds.), Handbuch derFertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)hypoeutectoid steels) followed by slow cooling (usually in air), the aim of which is to obtain afine-grained, uniformly distributed, ferritepearlite structure.Normalizing is applied mainly to unalloyed and low-alloy hypoeutectoid steels. Forhypereutectoid steels normalizing is performed only in special cases, and for these steels theaustenitizing temperature is 30808C (861768F) above the Ac1 transformation temperature.Figure 6.55 shows the thermal cycle of a normalizing process, and Figure 6.56 shows therange of austenitizing temperatures for normalizing unalloyed steels depending on theircarbon content. The parameters of a normalizing process are the heating rate, the austenitizing temperature, the holding time at austenitizing temperature, and the cooling rate.Normalizing treatment refines the grain of a steel that has become coarse-grained as aresult of heating to a high temperature, e.g., for forging or welding. Figure 6.57 shows theeffect of grain refining by normalizing a carbon steel of 0.5% C. Such grain refinement and12001147 CE1100Temperature, C1000G900 + Fe3C800 +723 C700 P00.4Pearlitea+Pearlite600500SKPearlite+Fe3C0.81.21.6Carbon content, %2.02.4FIGURE 6.56 Range of austenitizing temperatures for normalizing unalloyed steels depending on theircarbon content. (Temperature range above the line SE is used for dissolution of secondary carbides.)a, ferrite; g, austenite; Fe3C, cementite. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.FIGURE 6.57 Effect of grain refining by normalizing a carbon steel of 0.5% C. (a) As-rolled or forged,grain size ASTM 3 and (b) normalized, grain size ASTM 6. Magnification 500. (From K.E. Thelning,Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.)homogenization of the structure by normalizing is usually performed either to improve themechanical properties of the workpiece or (previous to hardening) to obtain better and moreuniform results after hardening. In some cases, normalizing is applied for better machinabilityof low-carbon steels.A special need for normalizing exists with steel castings because, due to slow cooling aftercasting, a coarse-grained structure develops that usually contains needlelike ferrite (Widmannstattens structure), as shown in Figure 6.58. A normalizing treatment at 7809508C(143617428F) (depending on chemical composition) removes this undesirable structure ofunalloyed and alloyed steel castings having 0.30.6% C.After hot rolling, the structure of steel is usually oriented in the rolling direction, as shownin Figure 6.59. In such a case, of course, mechanical properties differ between the rollingdirection and the direction perpendicular to it. To remove the oriented structure and obtainthe same mechanical properties in all directions, a normalizing annealing has to be performed.After forging at high temperatures, especially with workpieces that vary widely in crosssectional size, because of the different rates of cooling from the forging temperature, aheterogeneous structure is obtained that can be made uniform by normalizing.FIGURE 6.58 Structure of a steel casting (a) before normalizing and (b) after normalizing. (From H.J.Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag furGrundstoffindustrie, Leipzig, 1987.) 2006 by Taylor & Francis Group, LLC.FIGURE 6.59 Structure of DIN 20MnCr5 steel (a) after hot rolling and (b) after normalizing at 8808C.Magnification 100. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2,Warmebehandeln, Carl Hanser, Munich, 1987.)From the metallurgical aspect the grain refinement and the uniform distribution of thenewly formed ferritepearlite structure during normalizing treatment can be explained withthe following mechanism. At normalizing, the steel is subjected first to a a ! g (ferritepearlite to austenite) transformation, and after the holding time at austenitizing temperature,to a recurring g ! a (austenite to ferritepearlite) transformation. The effect of normalizingdepends on both austenitization and cooling from the austenitizing temperature.During austenitizing a far-reaching dissolution of carbides is aimed at, but this processcompetes with the growth of austenite grains after complete carbide dissolution, which is notdesirable. Besides the carbide dissolution, the degree of homogenization within the austenitematrix is important for obtaining a new arrangement of ferrite and pearlite constituents in thestructure after normalizing. Both dissolution and homogenizing are time- and temperaturedependent diffusion processes that are slower when the diffusion paths are longer (higherlocal differences in carbon concentration) and the diffusion rates are smaller (e.g., increasingamounts of alloying elements). Therefore, especially with alloyed steels, lower austenitizingtemperatures and longer holding times for normalizing give advantages taking into accountthe austenite grain growth. As shown in Figure 6.60, high austenitizing temperatures result ina coarse-grained austenite structure, which yields a coarse structure after normalizing.Holding time at austenitizing temperature may be calculated using the empirical formulat 60 D(6:35)where t is the holding time (min) and D is the maximum diameter of the workpiece (mm).When normalizing hypoeutectoid steels (i.e., steels with less than 0.8% C), during coolingfrom the austenitizing temperature, first a preeutectoid precipitation of ferrite takes place.With a lower cooling rate, the precipitation of ferrite increases along the austenite grainboundaries. For the desired uniform distribution of ferrite and pearlite after normalizing,however, a possibly simultaneous formation of ferrite and pearlite is necessary. Steels havingcarbon contents between 0.35 and 0.55% C especially tend to develop nonuniform ferritedistributions as shown in Figure 6.61. The structure in this figure indicates overly slowcooling in the temperature range of preeutectoid ferrite precipitation between Ar3 and Ar1.On the other hand, if the cooling through this temperature region takes place too fast, withsteels having carbon contents between 0.2 and 0.5%, formation of an undesirable needlelikeferrite (oriented at austenite grain boundaries), the so-called Widmannstattens structure,may result as shown in Figure 6.62. Formation of pearlite follows only after complete 2006 by Taylor & Francis Group, LLC.T2 > T1 > A1T2AusteniteT1A1PearliteHeatingCoolingFIGURE 6.60 Schematic presentation of the influence of austenitizing temperature on the grain size ofthe structure of a eutectoid steel after normalizing. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)precipitation of ferrite by transformation of the remaining austenite structure at temperatureAr1. It starts first at the boundaries of ferrite and austenite and spreads to the interior of theaustenite grains. The greater the number of the pearlitic regions formed, the more mutuallyhindered the pearlite grains are in their growth, and consequently the finer the grains of thenormalized structure. The influence of alloying elements on the austenite to ferrite andpearlite transformation may be read off from the relevant CCT diagram.Care should be taken to ensure that the cooling rate within the workpiece is in a rangecorresponding to the transformation behavior of the steel in question that results in a pureferritepearlite structure. If, for round bars of different diameters cooled in air, the coolingcurves in the core have been experimentally measured and recorded, then by using theappropriate CCT diagram for the steel grade in question, it is possible to predict the structureand hardness after normalizing. To superimpose the recorded cooling curves onto the CCTdiagram, the timetemperature scales must be equal to those of the CCT diagram.Figure 6.63 shows, for example, that the unalloyed steel DIN Ck45 will attain the desiredferritepearlite structure in the core of all investigated bars of different diameters cooled inFIGURE 6.61 Nonuniform distribution of ferrite and pearlite as a consequence of unfavorable temperature control during normalizing of unalloyed DIN C35 steel. Magnification 100. (From G. Spurand T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser,Munich, 1987.) 2006 by Taylor & Francis Group, LLC.FIGURE 6.62 Formation of needlelike ferrite at grain boundaries after normalizing of the unalloyedsteel DIN C35, because of too fast a cooling rate. Magnification 500. (From G. Spur and T. Stoferle(Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)air. On the other hand, as shown in Figure 6.64, the alloyed steel DIN 55NiCrMoV6 cooledin the same way in air will transform to martensite and bainite. In this case, to obtain adesired structure and hardness after normalizing, a much slower cooling of about 108C/h(508F/h), i.e., furnace cooling, has to be applied from the austenitizing temperature to thetemperature at which the formation of pearlite is finished (%6008C (%11008F)).6.2.3 ISOTHERMAL ANNEALINGHypoeutectoid low-carbon steels for carburizing as well as medium-carbon structural steelsfor hardening and tempering are often isothermally annealed, for best machinability, because1000Hardness HV900Ac370015Austenite10385mm203300000600300752 15150BainiteMs304008055=11 351050085Ac1454030 Ferrite6070 Pearlitem.dia6002010Temperature, C800Martensite2001007220Q1702 654 576 438 3481011Time, s2782441021222821310348 15min17410410560128 16 244hFIGURE 6.63 CCT diagram of the unalloyed steel DIN Ck45 (austenitizing temperature 8508C), withsuperimposed cooling curves measured in the core of round bars of different diameters cooled in air.(From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, CarlHanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.900Ac3800Ac1700100AP35 93m.mm400000600=1300753015050010Temperature, CD ia60053 5B75300Ms20200M100796Hardness HV08701011Time, s1021796786782103101753772743104102min454363 370105103285106s104FIGURE 6.64 CCT diagram of the alloyed steel DIN 55NiCrMoV6 (austenitizing temperature 9508C),with superimposed cooling curves measured in the core of round bars of different diameters cooled inair. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln,Carl Hanser, Munich, 1987.)a well-differentiated, nontextured ferritepearlite structure is the optimum structure formachinability of these steels. If low-carbon steels are soft annealed, they give long shavingswhen turned and a bad surface appearance (sometimes called smearing or tearing)because of the accumulation of the material on the tools cutting edge. On the other hand,nonannealed workpieces, having harder structural constituents like bainite, result in heavywear of the cutting edge when machined.An isothermally annealed structure should have the following characteristics: proportion of ferriteUniformly distributed pearlite grainsFine lamellar pearlite grainsShort pearlite lamellaeCoarse ferrite grainsFigure 6.65 shows the structure of a thin-wall die forging made of low-alloy steel forcarburizing (DIN 16MnCr5) after a normalizing anneal (Figure 6.65a) and after an isothermal annealing process (Figure 6.65b). The desired ferritepearlite structure originates duringan isothermal annealing, the principle of which is explained by Figure 6.66. This figure showsan IT diagram of a low-alloy steel for carburizing (DIN 15CrNi6) with superimposed coolingcurves for different cooling rates at continuous cooling. The slowest cooling rate of 3 K/minrelates to a furnace cooling, and the fastest cooling rate of 3000 K/min relates to a quenchingprocess. From the diagram in Figure 6.66 it can be clearly seen that bainite formation can beavoided only by very slow continuous cooling, but with such a slow cooling a textured(elongated ferrite) structure results (hatched area in Figure 6.66). There is only one way toavoid both the formation of bainite and the formation of a textured structure (see the openarrow in Figure 6.66), and this is the isothermal annealing process, which consists of 2006 by Taylor & Francis Group, LLC.FIGURE 6.65 Structure of a forging made of low-carbon steel for carburizing (DIN 16MnCr5) (a) afternormalizing and (b) after isothermal annealing. Magnification 200. (From G. Spur and T. Stoferle(Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)austenitizing followed by a fast cooling to the temperature range of pearlite formation(usually about 6508C (12008F)), holding at this temperature until the complete transformation of pearlite, and cooling to room temperature at an arbitrary cooling rate. The temperaturetime diagram of an isothermal annealing is given in Figure 6.67. The metallurgicalmechanism of a good isothermally annealed structure depends on the austenitizing conditionsas well as on the temperature and time of the isothermal transformation and on cooling fromthe austenitizing temperature to the isothermal transformation temperature.The austenitizing temperature and time should be high enough to completely dissolve allcarbides, to homogenize the austenite matrix, to stabilize the austenite structure, and achievea coarse-grained ferritepearlite structure after cooling. The undesired textured structureoriginates by preeutectoid ferrite precipitation along stretched phases acting as germs, forinstance manganese sulfides, carbon segregations, or aluminum nitride precipitations. Thesephases have been stretched as a consequence of a preliminary hot-forming process.To avoid the textured structure the steel has to contain as little sulfur, nitrogen, andaluminum as possible, and during austenitizing a complete dissolution of nitride precipitations and carbides should be achieved. Therefore the austenitizing temperature is adequatelyhigh, i.e., about 1008C (2128F) above Ac3, and the holding times are usually about 2 h.Field of textured structure10003 K/min30Temperature, C300P3000FAIsothermalannealingB500M3204000102101110Time, min250170 HV102103FIGURE 6.66 The principle of isothermal annealing. TTT diagram of the low-alloy steel for carburizingDIN 15CrNi6. (From J. Wunning, Harterei-Tech. Mitt. 32:4349, 1977, pp. 4349 [in German].) 2006 by Taylor & Francis Group, LLC.Temperature, C1000800Ac3Ac16004002000TimeFIGURE 6.67 Temperaturetime cycle of isothermal annealing. (From G. Spur and T. Stoferle (Eds.),Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)Another very important condition to avoid a textured structure is to realize a minimumcooling rate between the austenitizing temperature (%9508C (%17508F)) and the isothermaltransformation temperature (%6508C (12008F)). Thus, about 3008C (5728F) decrease shouldpass through at a minimum cooling rate of 2040 K/min. This means that the whole batch oftreated workpieces should be cooled from about 9508C (17508F) to about 6508C (12008F) inless than 10 min. During this cooling process an undercooling below the chosen isothermaltransformation temperature must be avoided to prevent the formation of bainite.The physical mechanism that accounts for the manner and magnitude of ferrite precipitation is the carbon diffusion during cooling from the austenitizing temperature. To achieve agood structure after isothermal annealing, all measures that reduce the carbon diffusion rateor restrict the diffusion time for carbon atoms during cooling are useful.Figure 6.68 shows three structures after isothermal annealing of the low-alloy steel DIN16MnCr5. It can be seen that cooling too slowly from the austenitizing temperature to thetransformation temperature results in an undesirable textured structure of ferrite and pearlite,and if during this cooling process an undercooling takes place (i.e., the transformationtemperature has been chosen too low) before the pearlite formation, then bainite will bepresent in the structure, which is not allowed.Big companies usually have internal standards to estimate the allowable degree of texturingof the isothermally annealed structures, with respect to machinability, as shown in Figure 6.69.The transformation temperature and the necessary transformation time for the steel in questionmay be determined by means of the appropriate IT diagram. Figure 6.70 shows such a diagramfor the steel DIN 17CrNiMo6. As can be seen, the lower the transformation temperature chosen,FIGURE 6.68 Different structures after isothermal annealing of the low-alloy steel DIN 16MnCr5 (left).Well-distributed ferritepearlite; correct annealing (center). Textured ferritepearlite structure; too slowcooling from the austenitizing to the transformation temperature (right). Ferrite pearlite bainite;undercooling before pearlite transformation. (From J. Wunning, Harterei-Tech. Mitt. 32:4349,1977, pp. 4349 [in German].) 2006 by Taylor & Francis Group, LLC.FIGURE 6.69 Internal standard of the German company Edelstahlwerke Buderus A.G.-Wetzlar forestimation of the allowable degree of texturing of the structure after isothermal annealing. Magnification 100. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)the sooner the transformation starts, up to a temperature (the so-called pearlite nose) at which theshortest time to start the transformation is achieved. Below this temperature, longer times areagain necessary to start the transformation. In the range of the pearlite nose temperature, finelamellar pearlite will be formed, and the time to complete pearlite transformation is the shortest.For unalloyed steels, the pearlite nose temperatures are between 550 and 5808C (1022 and10768F), while for alloyed steels they are between 640 and 6808C (1184 and 12568F). Theoptimum isothermal annealing temperature is 10208C (50688F) above the pearlite nosetemperature.The necessary transformation time depends on the alloying elements in the steel. In thepractice of isothermal annealing the holding time at the transformation temperature includesan adequate reserve because of compositional tolerances in different steel heats. Usually forlow-alloy steels for carburizing and structural steels for hardening and tempering the transformation times are below 2 h.From the technical standpoint, when a batch of workpieces has to be isothermallyannealed, the biggest problem is to realize sufficiently fast cooling from the austenitizing 2006 by Taylor & Francis Group, LLC.900Ac3Start of ferrite transformationTemperature, C880 Ac1700Start ofpearlitetransformationAustenite60084 91 8133500400Ms93Pearlite95End of transformationStart of transformation35300Bainite31Martensite200Hardness HRC10001Hardness HRB4610Time, s1021210348min1510410510660124824h1235days10FIGURE 6.70 Isothermal transformation (IT) diagram of the steel DIN 17CrNiMo6. Austenitizingtemperature 8708C. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2,Warmebehandeln, Carl Hanser, Munich, 1987.)temperature to the chosen transformation temperature without any undercooling. This cooling process depends on several factors, and the main factors include the workpiece crosssectional size, the loading arrangement, the temperature difference between the austenitizingtemperature and the temperature of the cooling medium, and the heat transfer coefficientbetween the workpieces surface and the ambient.6.2.4 SOFT ANNEALING (SPHEROIDIZING ANNEALING)Soft or spheroidizing annealing is an annealing process at temperatures close below or closeabove the Ac1 temperature, with subsequent slow cooling. The microstructure of steel beforesoft annealing is either ferritepearlite (hypoeutectoid steels), pearlite (eutectoid steels), orcementitepearlite (hypereutectoid steels). Sometimes a previously hardened structure existsbefore soft annealing. The aim of soft annealing is to produce a soft structure by changing allhard constituents like pearlite, bainite, and martensite (especially in steels with carboncontents above 0.5% and in tool steels) into a structure of spheroidized carbides in a ferriticmatrix.Figure 6.71 shows the structure with spheroidized carbides (a) after soft annealing of amedium-carbon low-alloy steel and (b) after soft annealing of a high-speed steel. Such a softstructure is required for good machinability of steels having more than 0.6% C and for all coldworking processes that include plastic deformation. Whereas for cold-working processes thestrength and hardness of the material should be as low as possible, for good machinabilitymedium strength or hardness values are required. Therefore, for instance, when ball bearingsteels are soft annealed, a hardness tolerance is usually specified. In the production sequence,soft annealing is usually performed with a semiproduct (after rolling or forging), and thesequence of operations is hot working, soft annealing, cold forming, hardening, and tempering.The required degree of spheroidization (i.e., 8090% of globular cementite or carbides) issometimes specified. To evaluate the structure after soft annealing, there are sometimesinternal standards, for a particular steel grade, showing the percentage of achieved globular 2006 by Taylor & Francis Group, LLC.(a)(b)FIGURE 6.71 Structures of (a) a medium-carbon low-alloy steel DIN 50CrMoV4 after soft annealing at7207408C and (b) a high-speed steel annealed at 8208C. Magnification 500. (From G. Spur andT. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich,1987.)cementite, as shown in Figure 6.72 for the ball bearing steel DIN 100Cr6. The degree ofspheroidization is expressed in this case as percentage of remaining lamellar pearlite.The physical mechanism of soft annealing is based on the coagulation of cementiteparticles within the ferrite matrix, for which the diffusion of carbon is decisive. Globularcementite within the ferritic matrix is the structure having the lowest energy content of allstructures in the ironcarbon system. The carbon diffusion depends on temperature, time, andthe kind and amount of alloying elements in the steel. The solubility of carbon in ferrite, whichis very low at room temperature (0.02% C), increases considerably up to the Ac1 temperature.At temperatures close to Ac1, the diffusion of carbon, iron, and alloying atoms is so great thatit is possible to change the structure in the direction of minimizing its energy content.The degree of coagulation as well as the size of carbides after soft annealing is dependentalso on the starting structure before annealing. If the starting structure is pearlite, the spheroidization of carbides takes place by the coagulation of the cementite lamellae. This process canbe formally divided into two stages. At first the cementite lamellae assume a knuckleboneshape, as shown in Figure 6.73. As annealing continues, the lamellae form globules at their endsand, by means of boundary surface energy, split up into spheroids, hence the name spheroidizing. During the second stage, some cementite (carbide) globules grow at the cost of fine carbideparticles, which disappear. In both stages, the rate of this process is controlled by diffusion. Thethicker the cementite lamellae, the more energy necessary for this process. A fine lamellarpearlite structure may more easily be transformed to a globular form.In establishing the process parameters for a soft (spheroidizing) annealing, a distinctionshould be drawn among hypoeutectoid carbon steels, hypereutectoid carbon steels, andalloyed steels. In any case the value of the relevant Ac1 temperature must be known. It canbe taken from the relevant IT or CCT diagram or calculated according to the formulaAc1 739 22(% C) 2(% Si) 7(% Mn) 14(% Cr) 13(% Mo) 13(% Ni) 20(% V), [ C](6:36)The temperature range for soft annealing of unalloyed carbon steels may be taken from theironcarbon diagram as shown in Figure 6.74. The holding time at the selected temperature isapproximately 1 min/mm of the workpiece cross section.For alloyed steels, the soft annealing temperature may be calculated according to theempirical formula 2006 by Taylor & Francis Group, LLC.FIGURE 6.72 Internal standard of the German company Edelstahlwerke Buderus A.G.-Wetzlar forevaluation of the degree of spheroidization after soft annealing of grade DIN 100Cr6 steel. Magnification 500. Amount of lamellar pearlite remaining 1, 0%; 2, 8%; 3, 20%; 4, 35%; 5, 60%; 6, 80%. (FromG. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser,Munich, 1987.)FIGURE 6.73 Schematic presentation of the process of transforming cementite lamella to spheroidsduring soft annealing. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984.) 2006 by Taylor & Francis Group, LLC.1050ETemperature, C1000950900850G800O750KP700S6500.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6Carbon content, wt%FIGURE 6.74 Temperature range for soft annealing of unalloyed steels having carbon contents of0.61.35% C. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)T 705 20(% Si % Mn % Cr % Mo % Ni % W) 100(% V) [ C](6:37)(a)(b)(c)Temperature, C Temperature, C Temperature, CThis formula is valid only up to the following values of the alloying elements: 0.9% C; 1.8% Si;1.1% Mn; 1.8% Cr; 0.5% Mo; 5% Ni; 0.5% W; and 0.25% V. If the steel has higher amountsof alloying elements, only these indicated maximum values are to be taken into account.Figure 6.75 shows possible temperaturetime regimes for soft annealing. The swingingregime (Figure 6.75c) is used to accelerate the transformation of cementite lamellae to globularform. Increasing the temperature above Ac1 facilitates the dissolution of cementite lamellae. Atsubsequent cooling below Ac1 this dissolution process is interrupted and the parts broken off(which has less resistance to boundary surface energy) coagulate more easily and quickly.On the basis of the investigations of Kostler, a degree of spheroidization e has beenestablished that gives the amount of globular cementite compared to the total amount of800700Ac1600500400800700Ac1600500400800700Ac1600500400FIGURE 6.75 Temperaturetime regimes at soft annealing. (a) Annealing at 208C below Ac1, forunalloyed steels and for alloyed steels with bainitic or martensitic starting structure; (b) annealing at108C above Ac1 (start) and decreasing temperature to 308C below Ac1 for alloyed steels; (c) swingingannealing +58C around Ac1 for hypereutectoid steels. (From G. Spur and T. Stoferle (Eds.), Handbuchder Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.Annealing temperature, K1000Degree of spheroidization e0.409600.500.600.800.959200.208808408000. 0.8 12468 10Annealing time, hFIGURE 6.76 Timetemperature diagram for soft annealing of the unalloyed steel DIN C35 (previouslydeformed 50%), to achieve the required degree of spheroidization. (After Kostler; see H.J. Eckstein(Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)cementite in a steel after soft annealing. e 1 means that 100% of the globular cementite (i.e.,no lamellar cementite) has remained. Because the degree of spheroidization depends on thetime and temperature of the soft annealing process, diagrams may be established thatcorrelate the degree of spheroidization with the time and temperature of soft annealing.Figure 6.76 shows such a diagram for the unalloyed steel DIN C35.The degree of spheroidization, especially above 80% (e 0.8), has considerable influence onultimate tensile strength, yield strength, and elongation, as shown in Figure 6.77 for the unalloyedeutectoid steel DIN C75. The hardness after soft annealing depends on the time and temperatureof spheroidization, as shown in Figure 6.78 for an unalloyed steel with 0.89% C.The machinability of steels with more than 0.6% C can be increased by soft annealing asshown in Figure 6.79, from which it can be seen that decreasing tensile strength and increasingthe degree of spheroidization allows a higher turning speed (v60) in m/min.The cooling after soft annealing should generally be slow. Depending on the steel grade,the cooling should be performed as follows:For carbon and low-alloy steels up to 6508C (12008F), with a cooling rate of 2025 K/h(furnace cooling)8502825750650Re221955016450A (L0 = 80 mm)Elongation, %Tensile strength (Rm)Yield strength (Re), MPaRmA35013100 20 40 60 80 100Degree of spheroidization, %250FIGURE 6.77 Change of ultimate tensile strength, yield strength, and elongation with increasingspheroidization of an unalloyed eutectoid steel, DIN C75. (From H.J. Eckstein (Ed.), Technologie derWarmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) 2006 by Taylor & Francis Group, LLC.Hardness, HRB13011090600 C625 C650 C700 C 675 C70050100150200Time, hFIGURE 6.78 Hardness of an unalloyed steel with 0.89% C after soft annealing, depending on thespheroidization time and temperature. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlungvon Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)For medium-alloy steels up to 6308C (11668F), with a cooling rate of 1520 K/h (furnacecooling)For high-alloy steels up to 6008C (11128F), with a cooling rate of 1015 K/h (furnace cooling)Further cooling below the temperatures indicated is usually performed in air6.2.5 RECRYSTALLIZATION ANNEALINGTurning speed (v60), m/minRecrystallization annealing is an annealing process at temperatures above the recrystallization temperature of the cold-worked material, without phase transformation, that aims atregeneration of properties and changes in the structure that exists after a cold-forming process20015010050abc500600700800900Tensile strength Rm, N/mm21000FIGURE 6.79 Influence of the ultimate tensile strength and degree of spheroidization on machinabilityof steels for carburizing and structural steels for hardening and tempering, expressed as 1 h turningspeed (v60) in m/min. (a) Spheroidization degree less than 30%; (b) spheroidization degree between 40and 60%; (c) spheroidization degree greater than 70%. (From G. Spur and T. Stoferle (Eds.), Handbuchder Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.FIGURE 6.80 Low-carbon steel with 0.05% C (a) after cold working with 20% reduction (hardness 135HV) and (b) after subsequent recrystallization annealing at 7508C (hardness 75 HV). Magnification200. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.)such as cold rolling, deep drawing, or wire drawing. Materials that are to be subjected to acold-forming process and subsequent recrystallization annealing must possess good coldforming ability. These materials include soft unalloyed steels, microalloyed steels for deepdrawing, microalloyed high-strength steels, unalloyed and alloyed carbon steels, stainlesssteels, and soft magnetic steels.The prerequisite to recrystallization on annealing is that the degree of deformation duringcold working has been large enough to produce the required number of defects in the crystalsto initiate nucleation, which is then followed by grain growth. Figure 6.80 shows the microstructure of a low-carbon steel (a) after cold working and (b) after subsequent recrystallization annealing. During cold working of metallic materials, by far the greatest amount of theenergy applied for deformation is transformed into heat, but a relatively small part (less than5%) of it remains accumulated in the material due to the formation of crystal lattice defects. Itis a known fact that every cold-working process (i.e., plastic deformation of the material)increases the dislocation density by some orders of magnitude. Because every dislocation is acrystal defect associated with internal stresses, the increase in the dislocation density causesthe accumulation of internal stresses (i.e., of internal energy) and thereby increases the freeenthalpy. Such a thermodynamically unstable material condition tends, at increased temperatures, to decrease the free enthalpy by rearranging and demolishing lattice defects. Thegreater the plastic deformation in a cold-forming process, the greater the strengthening of thematerial, which is characterized by an increase in tensile strength and yield strength and adecrease in elongation as shown in Figure 6.81. The material becomes harder and more800700600500400300010203040 50606060050Elongation, %700Yield strength, N/mm2Tensile strength, N/mm29005004003003020102001004001020 30400010203040 5060Degree of deformation at cold working, %FIGURE 6.81 Strengthening of a low-carbon steel by the cold-rolling process. (From G. Spur andferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)T. Sto 2006 by Taylor & Francis Group, LLC.brittle, so that in some cases a further step in the forming process cannot be applied without arecrystallization annealing. Also the anisotropy of the material, i.e., the dependence ofmechanical properties on the direction of the cold-forming process, can be annulated byrecrystallization annealing, by bringing the oriented grains that are deformed in one directionback to the original globular form.Thermic activation, i.e., increasing the temperature at recrystallization annealing, can beused to reestablish the original structure (before cold working) with the original density ofdislocations, which results in decreased hardness and strength and increased ductility andformability. The recrystallization annealing process includes the following phenomena: grainrecovery, polygonization, recrystallization, and grain growth. Grain RecoveryGrain recovery is a process of tempering a cold-worked metallic structure at low temperatures(1503508C (3006628F)) without causing any discernible changes in the microstructure. Itresults only in decreasing the internal stresses without substantially decreasing the strength ofthe material. However, during this process characteristic changes occur in the electrical resistanceand its temperature coefficient of the cold-worked material. The activation energy needed forgrain recovery depends on the degree of cold working. The higher the degree (i.e., the greater thedeformation), the less the activation energy required. The temperature of grain recovery correlates with the recrystallization temperature of the same material according to the formulaTGR TR 300 [ C](6:38) PolygonizationPolygonization of a cold-worked structure is the creation of a new polygonal arrangement ofedge dislocations in the metallic crystal lattice that takes place at temperatures close above thegrain recovery temperature. As shown in Figure 6.82, in such a case the applied thermalenergy is sufficient to rearrange the edge dislocations. In this case the originally bent slidingplanes take a polygonal shape, forming segments within a grain called subgrains. The anglesbetween subgrains are very small (about 18). As a consequence of a substantial energydischarge by discharge of internal stresses, material strength is decreased. Polygonizationtakes place primarily in heavily cold-worked structures, especially in ferritic matrices, belowthe recrystallization temperature.(a)a(b)bbFIGURE 6.82 Schematic presentation of polygonization. Arrangement of edge dislocations and slidingplanes (a) before polygonization and (b) after polygonization. a, Edge dislocations; b, sliding planes.(From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, CarlHanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.Hardness, HV10250ab200150100250350450550650750Temperature, CFIGURE 6.83 Decrease in hardness during recrystallization of a steel having 0.03% C, 0.54% Si, and 0.20%Mn that was cold rolled (805 deformation), as a function of annealing temperature. (Heating rate 208C/h.)a, Begin formation of new grains; b, end formation of new grains. (From G. Spur and T. Stoferle (Eds.),Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) and Grain GrowthThe process of recrystallization begins when the recrystallization temperature is overstepped.The recrystallization temperature of a material is the temperature at which the formation of newgrains begins within a cold-worked microstructure, as shown in Figure 6.83. From this figureone can conclude that for the steel in question the recrystallization temperature is 5208C(9688F). During recrystallization, as can be seen from Figure 6.83, hardness and strengthdecrease substantially while ductility increases. In practice, the recrystallization temperatureTR is often considered the temperature of a cold-worked material at which recrystallization iscompleted after 1 h of annealing. There is a correlation between the recrystallization temperature (TR) and the melting temperature (TM) of the material, which readsTR 0:4TM(6:39)Figure 6.84 shows that this correlation holds for practically all pure metals if bothtemperatures TR and TM are taken in deg. K. The recrystallization temperature can beRecrystallization temperature TR, C1500W1250ReMa1000750CrPd500BeNi Fe250MgAlPb0TaCbTiPtAuAg CuZnCd25005001000 1500 2000 2500 30003500Melting temperature TM, CFIGURE 6.84 Correlation between the recrystallization temperature and the melting temperature forpure metals. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.influenced by the degree of deformation during cold working, the heating rate, and thestarting microstructure.In contrast to the grain recovery process (which follows a parabolic law), the recrystallization process begins only after an incubation period (because of nucleation), startingslowly, reaching a maximum rate, and finishing slowly. The nuclei from which new grainsgrow are situated preferably at the grain boundaries of compressed cold-worked grains. Newgrains grow from these nuclei until they meet up with other grains. Recrystallization bringsabout the formation and movement of large-angle grain boundaries.Figure 6.85 is a schematic presentation of new grain formation and growth during therecrystallization process as a function of annealing time. As time passes, the new grains,starting from nuclei, grow unhindered within the cold-worked grains. Simultaneously, newnuclei are formed. At the movement of large-angle grain boundaries, new grains consume thepreviously deformed grains. The recrystallization process is locally finished when new neighboring grains collide with each other. The size, form, and orientation of the new structure, aswell as the condition of the lattice defects in it, differ substantially from those of the previousstructure. The recrystallization process itself can be hindered by precipitations, dispersions,and a second phase.The most important technological parameters of recrystallization annealing that influencethe rate of recrystallization and the material properties after recrystallization are:1. Material-dependent parametersthe chemical composition and the starting structure (including the degree of deformation)2. Process-dependent parametersannealing temperature, annealing time, and heating and cooling ratesThe course of a recrystallization process can be presented in an isothermal timetemperaturerecrystallization diagram as shown in Figure 6.86. As can be seen from this diagram,the higher the temperature of recrystallization annealing, the shorter the necessary annealingtime. The lower the degree of deformation at cold working, the higher the requiredt1t2 > t1t3 > t2t4 > t3t5 > t4t6 > t5FIGURE 6.85 Schematic presentation of new grain formation and growth during the recrystallizationprocess as a function of annealing time t. (From G. Spur and T. Stoferle (Eds.), Handbuch derFertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.TemperatureEnd of recrystallizationBeginning ofrecrystallizationHolding time (logarithmic)FIGURE 6.86 An isothermal timetemperaturerecrystallization diagram. (From H.J. Eckstein (Ed.),Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie,Leipzig, 1987.)Recrystallization temperature, Crecrystallization temperature, as shown in Figure 6.87. The higher the heating rate, thehigher the recrystallization temperature. It can be concluded from Figure 6.86 that withsubstantially longer annealing times, a full recrystallization can be achieved at relativelylow temperatures.The degree of deformation at cold working has a very important influence on the sizeof newly formed grains during recrystallization. If the cold working is carried out with avery low degree of deformation but without sufficient strengthening of the material toenable the process of recrystallization, a decrease in stresses can still be achieved by movementof the deformed grain boundaries. In this case grains with low dislocation densitygrow (because there are only a few nuclei) and a coarse-grained structure develops asshown in Figure 6.88. Consequently, there is a critical degree of deformation at coldworking that at subsequent recrystallization annealing leads to sudden grain growth, asshown in Figure 6.89 for a low-carbon steel. With an increase in the carbon content ofthe steel, this critical degree of deformation shifts from about 8 to 20% of deformation atcold working.1200110010009008007006005004000 10 20 30 40 50 60 70 80 90Degree of deformation at cold working, %FIGURE 6.87 Recrystallization temperature of a- and g-iron as a function of the degree of deformationat cold working. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2,Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.(a)(b)FIGURE 6.88 Development of coarse-grained structure during recrystallization of soft iron. (a) Microstructure before cold working and (b) microstructure after cold working with very low degree of deformation (10%) and subsequent recrystallization annealing at 7008C. Magnification 500. (From G. Spur andT. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich,1987.)6.3 HARDENING BY FORMATION OF MARTENSITE6.3.1 AUSTENITIZINGAustenitizing is the first operation in many of the most important heat treatment processes(hardening, carburizing, normalizing) on which the properties of heat-treated parts depend.Let us assume the bulk heat treatment of real batches of workpieces and consider themetallurgical and technological aspects of austenitizing. Metallurgical Aspects of AustenitizingGrain sizeThe way austenite is formed when a certain steel is heated depends very much on the steelsstarting microstructure. Let us take as an example an unalloyed eutectoid steel with 0.8% Cand follow the process of its austenitization using the schemes shown in Figure 6.90. At roomtemperature the cementite (Fe3C) plates of the pearlite are in direct contact with ferrite (a-Fe,see Figure 6.90a). The carbon atoms from cementite have a tendency to diffuse into the ferritelattice. The higher the temperature, the greater this tendency is. Upon heating, on reachingthe Ac1 temperature (7238C (13338F)), the transformation of ferrite into austenite (g-Fe)010 20 30 40 50 60 70 80 90 100Degree of deformation at cold working, %FIGURE 6.89 Grain growth in the range of the critical degree of deformation (at 10%) for a steel with0.06% C. Recrystallization temperature, 7008C. (From G. Spur and T. Stoferle (Eds.), Handbuch derFertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.-Fe-Fe-FeFe3C-FeAustenite(a)(b)(c)(d)(e)Austenite(f)FIGURE 6.90 Transformation of a pearlitic structure to austenite when heating an unalloyed eutectoidsteel of 0.8% C.starts immediately adjacent to the cementite plates (see Figure 6.90b). After that the cementiteplates start to dissolve within the newly formed austenite, becoming thinner and thinner(Figure 6.90c and Figure 6.90d). So two processes take place at the same time: the formationof austenite grains from ferrite and the dissolution of cementite plates in the austenite lattice.Experiments have shown that the process of ferrite-to-austenite transformation ends beforeall the cementite has been dissolved. This means that after all the ferrite has transformed intoaustenite, small particles of cementite will remain within the austenite grains (Figure 6.90e).Figure 6.91 shows the formation of austenite in a microstructure of eutectoid steel. Areas ofFIGURE 6.91 Formation of austenite (light patches) from pearlite as a function of time. (From G. Krauss,Steels: Heat Treatment and Processing Principles, ASM International, Materials Park, OH, 1990.) 2006 by Taylor & Francis Group, LLC.a2a1a122a2a112a233Austenitea3(a)Carbide1a3(b)(c)FIGURE 6.92 Nucleation sites for austenite formation in microstructures of (a) ferrite; (b) spheroidite;(c) pearlite. (From G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International,Materials Park, OH, 1990.)austenite formation are visible as white patches within the lamellar pearlitic structure. Some ofthe cementite persists in the form of spheroidized particles (the small dark spots in thewhite areas). They dissolve only with longer holding times at temperature. Once these cementiteparticles completely dissolve, the structure consists of only one phaseaustenite (see Figure6.90f). In this state, however, there are still differences in carbon concentration among particular austenite grains. In spots where cementite plates were previously to be found, the carbonconcentration is high, while in other spots far from cementite plates it is low.Equalizing of the carbon concentration proceeds gradually by diffusion, resulting in ahomogeneous austenite structure at the end of this process. The holding time at austenitizingtemperature necessary for this process is called the homogenization time. During pearliteaustenite transformation, several austenite grains are formed from one pearlite grain, i.e., thenewly formed austenite is fine-grained.Nucleation sites for austenite formation depend on the starting microstructure asshown in Figure 6.92. In ferrite the nucleation sites are situated primarily at grain boundaries. In spheroidized structures nucleation starts on carbide particles, whereas in pearliticstructures it starts primarily at the intersection of pearlite colonies but also at cementitelamellae. Kinetics of Transformation during AustenitizingFigure 6.93 shows the volume percent of austenite formed from pearlite in eutectoid steel as afunction of time at a constant austenitizing temperature. From the beginning of austenitizationVolume of austenite, %1007550250051015Time, s202530FIGURE 6.93 Volume percent austenite formed from pearlite in eutectoid steel as a function of time at aconstant austenitizing temperature. (From G. Krauss, Steels: Heat Treatment and Processing Principles,ASM International, Materials Park, OH, 1990.) 2006 by Taylor & Francis Group, LLC.a certain incubation time is necessary to form the first nuclei, and then the process proceeds at amore rapid rate as more nuclei develop and grow. At higher temperatures the diffusion rateincreases and austenite forms more rapidly, as shown in Figure 6.94.The duration of austenitizing process depends on the austenitizing temperature and thesteel composition. The influence of time at austenitization can best be explained by thediagrammatic illustrations shown in Figure 6.95. From Figure 6.95a and Figure 6.95b,which apply to eutectoid carbon steel of 0.8% C, one can see that if an austenitizingtemperature of 7308C (13468F) is maintained (after a rapid heating to this temperature), thetransformation will start in about 30 s. If instead an austenitizing temperature of 7508C(13828F) is chosen, the transformation will begin in 10 s, and if a temperature of 8108C(14908F) is selected, in about 1 s. The transformation of pearlite to austenite and cementite isin this case completed in about 6 s. If the steel is to be fully austenitic (all carbides dissolved,hatched area), it must be held at this temperature for about 2 h (7 103 s).Figure 6.95c and Figure 6.95d apply to a hypoeutectoid plain carbon steel of 0.45% C.They show that in this case at an austenitizing temperature of 8108C (14908F) the transformation from pearlite to austenite starts in about 1 s. In about 5 s the pearlite has beentransformed and the structure consists of ferrite, austenite, and cementite. About 1 minlater the carbon has diffused to the ferrite, which has thereby been transformed to austenite.Residual particles of cementite remain, however, and it takes about 5 h at this temperature todissolve them completely.Figure 6.95e and Figure 6.95f apply to a hypereutectoid steel containing 1.2% C. If thissteel is austenitized at 8108C (14908F), the pearlite starts to transform in about 2 s, and inabout 5 s the structure consists only of austenite and cementite. It is not possible for thecementite to be completely dissolved at this temperature. To achieve complete solution of thecementite, the temperature must be increased above Acm, in this case to at least 8608C(15808F).The holding time at austenitizing (hardening) temperature depends on the desired degreeof carbide dissolution and acceptable grain size, taking into account that the grain growthincreases with higher austenitizing temperatures and longer holding times. Since the amountof carbide is different for different types of steel, the holding time (from the metallurgicalpoint of view) depends on the grade of steel. However, carbide dissolution and the holdingtime are dependent not only on the austenitizing temperature but also the rate of heating to100Austenite, %80730 C751 C60402001101001000Time, sFIGURE 6.94 Effect of austenitizing temperature on the rate of austenite formation from pearlite in aeutectoid steel. (From G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International,Materials Park, OH, 1990.) 2006 by Taylor & Francis Group, LLC.TemperatureC900A3Acm800C900800AA+CA+PA1700700600600P5005000 0.2 0.4 0.6 0.8 1.0 1.2 % C 101 1(a)(b)TemperatureC900 A3AcmC90080080010 102 103 104 105 sA1700600500F+P+A500101 1(d)F+A +C700600AA+C0 0.2 0.4 0.5 0.8 1.0 1.2 % C(c)TemperatureC900 A3Acm10 102 103 104 105 sC900800800F+ PAA+CA+P+A1700600C700600P+C5000 0.2 0.4 0.5 0.8 1.0 1.2 % C(e)500101 1(f)10 102 103 104 105 sFIGURE 6.95 Structural transformations during austenitizing steels containing (a, b) 0.8% C; (c, d)0.45% C; (e, f) 1.2% C. A, austenite; C, cementite; F, ferrite; P, pearlite. (From K.E. Thelning, Steel andIts Heat Treatment, 2nd ed., Butterworths, London, 1984.)this temperature. Varying the rate of heating to this temperature will have an effect on the rateof transformation and dissolution of the constituents.The influence of the role of heating (and correspondingly of the holding time) on carbidedissolution, grain growth, and hardness after hardening for various grades of steel has beenstudied in detail and published in Refs. [18,19]. These timetemperatureaustenitizing diagrams (Zeit-Temperatur-Austenitisierung Schaubilder in German) have been produced either asisothermal diagrams (the steel specimens were heated rapidly at the rate of 1308C/s (2668F/s)to the temperature in question and held there for a certain predetermined time) or as continuousheating diagrams (the steel specimens were heated continuously at different heating rates). 2006 by Taylor & Francis Group, LLC.Consequently, isothermal diagrams may be read only along the isotherms, and the continuousheating diagrams may be read only along the heating rate lines.Figure 6.96 shows an isothermal type of timetemperatureaustenitizing diagram of gradeDIN 50CrV4 steel. From this type of diagram one can read off, for instance, that if the steel isheld at 8308C (15268F), after about 1 s, pearlite and ferrite will be transformed to austenite,but more than 1000 s is necessary to completely dissolve the carbides to achieve a homogeneous austenite.In practice, the continuous heating diagrams are much more important because everyaustenitizing process is carried out at a specified heating rate. Figure 6.97 shows a timetemperatureaustenitizing diagram of the continuous heating type for grade DIN Ck45 steel.The continuous heating was carried out at various constant rates ranging from 0.05 to24008C/s (32.09 to 43528F). If the heating rate was extremely slow (e.g., 0.228C/s (32.48F/s))to about 7758C (14278F), on crossing the Ac3 temperature after about 1 h all pearliteand ferrite would have been transformed to inhomogeneous austenite. At a heating rate13001200Homogenuous austeniteTemperature,C11001000ACC900Austenite + carbideAC3830800AC1Ferrite + pearlite + austeniteAC2Ferrite + pearliteHeating rate to hardening temp. 130 C/s7000.010.1110102103Time, sFIGURE 6.96 Isothermal timetemperatureaustenitizing diagram of the steel grade DIN 50CrV4(0.47% C, 0.27% Si, 0.90% Mn, 1.10% Cr). (From J. Orlich and H.J. Pietrzenivk (Eds.), Atlas zurWarmebehandlung der Stahle, Vol. 4, Zeit-Temperatur-Austenitisierung-Schaubilder, Part 2, VerlagStahleisen, Dusseldorf, 1976 [in German].) 2006 by Taylor & Francis Group, LLC.Heating rate, C/s2400 1000 3001003010310.220.05130012001100Temperature, CHomogeneous austenite1000900Ac3Inhomogeneousaustenite800Ferrite + pearliteAc1AusteniteAc2Ferrite + pearlite700101110102103104105Time, sFIGURE 6.97 Timetemperatureaustenitizing diagram for continuous heating of the steel grade DINCk45 (0.49% C, 0.26% Si, 0.74% Mn). (From J. Orlich, A. Rose, and P. Wiest (Eds.), Atlas zurWarmebehandlung der Stahle, Vol. 3, Zeit-Temperatur-Austenitisierung-Schaubilder, Verlag Stahleisen,Dusseldorf, 1973 [in German].)of 108C/s (508F/s) the pearlite and ferrite would have been transformed to inhomogeneousaustenite after crossing the Ac3 temperature at about 8008C (14728F) after only 80 s.A remarkable feature of such diagrams is that they show precisely the increase of Ac1 andAc3 transformation temperatures with increasing heating rates. This is especially importantwhen short-time heating processes like induction hardening or laser beam hardening, withheating rates ranging to about 10008C/s (18328F/s), are applied for surface hardening. In sucha case this diagram should be consulted to determine the required austenitizing temperature,which is much higher than in conventional hardening of the same grade of steel. For the steelin question, for example, the conventional hardening temperature would be in the range of8308508C (152615628F), but for induction or laser beam hardening processes the hardeningtemperatures required are between 950 and 10008C (1742 and 18328F). When heating at a rateof 10008C/s (18328F/s) to the austenitizing temperature of 10008C (18328F), only 1 s isnecessary, and the above-mentioned short heating time processes operate in approximatelythis time range. As Figure 6.97 shows, much higher temperatures are necessary to achieve the 2006 by Taylor & Francis Group, LLC.homogeneous austenite structure. In such a case one is, of course, concerned with the graingrowth.Figure 6.98 shows the grain growth (according to American Society for Testing andMaterials [ASTM]) when grade DIN Ck45 steel is continuously heated at different heatingrates to different austenitizing temperatures. Figure 6.99 shows the achievable Vickers hardness after hardening for grade DIN Ck45 steel austenitized at various heating rates to varioustemperatures. It shows, for example, that maximum hardness would be achieved uponaustenitizing the steel at 8508C (15628F) for about 900 s (or heating at a heating rate of18C/s (33.88F/s)), which corresponds to the field of homogeneous austenite (see Figure 6.97).The hardness after quenching, which depends on the amount of carbide dissolution, is alsodependent on the initial structure of the steel. This is illustrated in Figure 6.100. Figure 6.100ashows that a structure of spheroidized cementite (after soft annealing) of the hypoeutectoidDIN Cf53 carbon steel will attain the maximum hardness of 770 HV when heated at a rateof 18C/s (33.88F/s) to 8758C (16098F) (holding time 855 s or 14 min). The hardened andHeating rate C/s2400 10003001003010310.220.0513004 to 3120011008100091011 to 10Ac3900800Grain size (ASTM):Temperature, C46Ac1700101110102103104105Time, sFIGURE 6.98 Timetemperatureaustenitizing diagram for continuous heating showing the graingrowth of steel grade DIN Ck45. (From J. Orlich, A. Rose, and P. Wiest (Eds.), Atlas zur Warmebehandlung der Stahle, Vol. 3, Zeit-Temperatur-Austenitisierung-Schaubilder, Verlag Stahleisen, Dusseldorf, 1973 [in German].) 2006 by Taylor & Francis Group, LLC.Heating rate C/s2400 10003001003010310.220.05130012007808001000840Ac3900840840820800780800Ac1Hardness after quenching (HV):Temperature, C11007000.1110102103104105Time, sFIGURE 6.99 Timetemperatureaustenitizing diagram for continuous heating showing the achievablehardness after hardening steel grade DIN Ck45. (From J. Orlich, A. Rose, and P. Wiest (Eds.), Atlas zurWarmebehandlung der Stahle, Vol. 3, Zeit-Temperatur-Austenitisierung-Schaubilder, Verlag Stahleisen,Dusseldorf, 1973 [in German].)tempered structure (tempered martensite) of the same steel, as shown in Figure 6.100b,will attain the maximum hardness of 770 HV, however, if heated to 8758C (16098F) at therate of 10008C/s (18328F/s) (holding time less than 1 s). For this reason, when short-timeheating processes are used, the best results are achieved with hardened and tempered initialstructures.For eutectoid and hypereutectoid steel grades, which after quenching develop substantialamounts of retained austenite, the attainment of maximum hardness after quenching is morecomplicated. Figure 6.101 shows the hardness after quenching for the ball bearing hypereutectoid grade DIN 100Cr6 steel (1.0% C, 0.22% Si, 0.24% Mn, and 1.52% Cr). The maximumhardness of 900 HV after quenching is attained on heating to a very narrow temperaturerange, and furthermore this temperature range is displaced toward higher temperatures as theheating rate is increased. If this steel is quenched from temperatures that exceed the optimumrange, the resulting hardness is reduced owing the presence of an increasing amount ofretained austenite. 2006 by Taylor & Francis Group, LLC.Hardness after quenching, HV1900Rate of heating in C/s8007001101001000600700800900Hardness after quenching, HV1(a)10001100Temperature, C1200130012001300900Rate of heating in C/s8001 10 100 1000700600700(b)80090010001100Temperature, CFIGURE 6.100 Hardness after quenching as a function of the rate of heating and austenitizing temperature for grade DIN Cf53 steel (hypoeutectoid carbon steel) (a) for soft-annealed condition and (b)for hardened and tempered condition. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed.,Butterworths, London, 1984.)For plain carbon and low-alloy structural steels, which contain easily dissolved carbides, aholding time of 515 min after they have reached the hardening temperature is quite enoughto make certain that there has been sufficient carbide dissolution. For medium-alloy structural steels this holding time is about 1525 min. For low-alloy tool steels, it is between 10 and30 min; and for high-alloy Cr steels, between 10 min and 1 h. Aspects of AustenitizingIn heating metallic objects to their austenitizing (hardening) temperature, there are two kindsof heating rates to be distinguished: those that are technically possible and those that aretechnologically allowed.The technically possible heating rate is the heating rate the heating equipment couldrealize in actual use. It depends on1. The installed heating capacity of the equipment2. The heat transfer medium (gas, liquid, vacuum)3. The temperature difference between the heat source and the surface of the heatedobjects (workpieces put in a hot or cold furnace)4. The mass and shape of the workpiece (the ratio between its volume and superficialarea)5. The number of workpieces in a batch and their loading arrangement 2006 by Taylor & Francis Group, LLC.Hardness after quenching, HV11000Rate of heating in C/s9008007001 10100100060050070080090010001100Temperature, C12001300FIGURE 6.101 Hardness after quenching as a function of the rate of heating and austenitizing temperature for grade DIN 100Cr6 steel initially soft annealed. (From K.E. Thelning, Steel and Its HeatTreatment, 2nd ed., Butterworths, London, 1984.)The technologically allowed heating rate is the maximum heating rate that can be appliedin actual circumstances, taking into account the fact that thermal stresses that develop withinthe workpiece must not exceed the critical value because this could cause warping or cracking,since sections having different dimensions heat up at different speeds and large temperaturegradients can arise between the surface and the core of the workpiece. This heating ratedepends on1. The mass and shape of the workpiece (the ratio between its volume and superficialarea)2. The chemical composition of the material3. The initial microstructureWhen workpieces of heavy sections or of complicated shapes are heated, temperaturesbetween 250 and 6008C (482 and 11128F) are particularly dangerous, because in this temperature range the steel does not have enough plasticity to compensate for thermal stresses.If the heating of an object is asymmetrical, the object will warp. If thermal stresses aredeveloped that overstep the strength of the material (which is substantially lower at highertemperatures), cracks will result.If the heating rate is too high through the transformation temperature range (between Ac1and Ac3), warping may occur because of volume change of the structure lattice. The tendencyof a steel to crack during heating depends on its chemical composition. Carbon content hasthe decisive influence. The higher the carbon content, the greater the sensitivity to cracking.The complex influence of carbon and other alloying elements is expressed by the followingempirical formula termed the C equivalent (Cekv):Cekv C Mn Cr Mo Ni V Si 0:5 Ti W Al54310 555 10 10(6:40)where the element symbols represent wt% content. This formula is valid up to the followingmaximum values of alloying elements. 2006 by Taylor & Francis Group, LLC.C 0.9%Mn 1.1%Cr 1.8%Mo 0.5%Ni 5.0%VSiTiWAl0.25%1.8%0.5%2.0%2.0%The values of the alloying elements actually present are put into the formula in wt%. Ifthe amount of an alloying element exceeds the limit given above, then the indicated maximumvalue should be put into the formula.The higher the calculated Cekv value, the greater the sensitivity of the steel to cracking. Forinstance,Cekv 0.4: The steel is not sensitive to cracking (it may be heated quite rapidly).Cekv 0.40.7: The steel is medium sensitive to cracking.Cekv ! 0.7: The steel is very sensitive to cracking (when heating up a preheatingoperation should be included).The initial microstructure also has some influence on the technologically allowed heatingrate. A steel with a homogeneous microstructure of low hardness may be heated more rapidlythan a steel of high hardness with inhomogeneous microstructure.The thermal gradients and consequently the thermal stresses developed when heating toaustenitizing temperature can usually be diminished by preheating the workpiece to temperature lying close below the transformation temperature Ac1 and holding it there untiltemperature equalizes throughout the cross section.The theoretical timetemperature diagram of the austenitizing process is shown in Figure6.102. Practically, however, there is no such strict distinction between the heating and soakingTCCoSurfacereTa11preheating2heating up3heating through (thermal soaking)4structure homogenizing (metallurgicalsoaking)Ta austenitizing temperatureta austenitizing timeth32ta4FIGURE 6.102 Austenitizing process (theoretically). 2006 by Taylor & Francis Group, LLC.TemperatureFC1832 1000III1652 900III1472 8001292 7001112 600932 5001 4 in. I 25 100 mm2 7 in. II 50 175 mm4 8 in. III 100 200 mm752 400572 300392 200212 1003200246810 12 14 16 18 20 22 24 26 28 minHeating-up timeFIGURE 6.103 Timetemperature curves for steel bars of different diameters heated in a salt bath at10008C. Full line, measured temperature at surface; dashed line, measured temperature at center. (FromK.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.)times. Contrary to the generally widespread belief that the surface of the steel reaches thepreset temperature considerably earlier than the center, the closer the temperature of the steelapproaches the preset temperature, the smaller the temperature difference between surfaceand core, as shown in Figure 6.103. It can therefore be assumed that when the surface hasreached the preset temperature, part of the soaking time (depending on the cross-sectionalsize) has already been accomplished. Certainly, one has to be aware of the corner effectcorners, sharp edges, and thin sections reach the preset temperature much earlier than thecore of the workpiece.The most important parameters of every austenitizing process are:1. The austenitizing temperature2. The heat-up and soak time at austenitizing temperatureFor each grade of steel there is an optimum austenitizing (hardening) temperaturerange. This temperature range is chosen so as to give maximum hardness after quenchingand maintain a fine-grained microstructure. It can be determined experimentally as shownin Figure 6.104 and Figure 6.105. From Figure 6.104 it is clear that the lowest possiblehardening temperature for the steel in question is 8508C (15628F). A lower hardeningtemperature would result in the formation of bainite and even pearlite with inadequatehardness.When the hardening temperature is increased (see Figure 6.105), the grain size andthe amount of retained austenite increase. At 920 and 9708C (1688 and 17788F) the retainedaustenite may be discerned as light angular areas. On the basis of these experiments,the optimum hardening temperature range for the steel in question has been fixed at 8508808C (156216168F). The optimum hardening temperature range for unalloyed steels can bedetermined from the ironcarbon equilibrium diagram according to the carbon contentof the steel. This range is 30508C (861228F) above the Ac3 temperature for hypoeutectoidsteels and 30508C (861228F) above Ac1 for hypereutectoid steels, as shown in Figure 6.106.Because the curve SE in this diagram denotes also the maximum solubility of carbon 2006 by Taylor & Francis Group, LLC.FIGURE 6.104 Microstructures of a steel having 1% C, 1.5% Si, 0.8% Mn, and 1% Cr, hardened fromhardening temperatures between 800 and 8508C. Dimensions of test pieces: 30-mm diameter 100 mm.Magnification 400. (a) Hardening temperature 8008C, hardness 55 HRC; (b) hardening temperature8258C, hardness 61.5 HRC; (c) hardening temperature 8508C, hardness 66 HRC. (From K.E. Thelning,Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.)in austenite, it is clear that the higher the austenitizing (hardening) temperature, themore carbon can be dissolved in austenite. For alloyed steels the optimum austenitizing(hardening) temperature range depends on the chemical composition, because differentalloying elements shift the A1 temperature to either higher or lower temperatures. For 2006 by Taylor & Francis Group, LLC.FIGURE 6.105 Microstructures of steel having 1% C, 1.5% Si, 0.8% Mn, and 1% Cr, hardened fromhardening temperatures between 870 and 9708C. Dimensions of test pieces: 30-mm diameter 100 mm.Magnification 400. (a) Hardening temperature 8708C, hardness 62.5 HRC, retained austenite 12%; (b)hardening temperature 9208C, hardness 62 HRC, retained austenite 20%.; (c) hardening temperature9708C, hardness 61 HRC, retained austenite 28%. (From K.E. Thelning, Steel and Its Heat Treatment,2nd ed., Butterworths, London, 1984.)these steels, therefore, data from the literature on the optimum hardening temperature rangehave to be consulted.It should also be mentioned that increasing the austenitizing temperature causes thefollowing effects. 2006 by Taylor & Francis Group, LLC.Ironcarbon equilibrium diagramTCE1148G + Fe3C00+ AAustenitecm912A3727A1' + pearlitePearliteSPearlite + Fe3C0.772.11C, wt%The range of optimun hardening temperaturesFIGURE 6.106 Optimum hardening temperature range for unalloyed steels, depending on the carboncontent.1. It increases the hardenability of the steel because of the greater amount of carbidegoing into solution and the increased grain size.2. It lowers the martensite start temperature (Ms). Owing to the more complete carbidedissolution, the austenite becomes more stable and starts to transform upon quenching at lower temperature.3. It increases the incubation time, i.e., the time until the isothermal transformation topearlite or bainite starts. This is expressed as a shift in the start of transformationcurves in an IT diagram to later times.4. It increases the amount of retained austenite after quenching due to stabilization ofthe austenite, which at higher temperatures is more saturated with carbon fromdissolved carbides.Heat-up and soak time at austenitizing temperature is a very important parameter forbulk heat treatment because it not only determines the furnace productivity and economy(consumption of energy) but may also affect the properties of the treated workpieces. Untilrecently there was no reliable, objective method for accurately predicting heat-up and soaktimes for heat treatment cycles that took into account all workpiece characteristics, variationsin furnace design, and load arrangement. Current determinations of heat-up and soak timeare based on either a very conservative and general rule (e.g., 1 h/in. of cross section) or someempirical method, the results of which [20] differ substantially.By heat-up and soak time we mean the time it takes for the heated workpiece to go fromstarting (room) temperature to the preset temperature in its core. The main factors thatinfluence heat-up and soak time are diagrammed in Figure 6.107.On the basis of experiments with 26 specimens (cylinders, round plates, and rings ofvarious dimensions) made of unalloyed and low-alloy structural steels, Jost et al. [20] foundfrom core temperature measurements that the heat-up and soak time depends substantially onthe geometry of the heated workpiece and its mass. They found the heat-up and soak time tobe directly proportional to the mass/surface area (m/A, kg/m2) ratio, as shown in Figure 6.108.By regression analysis for their conditions (the specimens were heated in an electrically heatedchamber furnace of 8 kW capacity and 240 240 400 mm working space, to the hardening 2006 by Taylor & Francis Group, LLC.ActualconditionsWorkpieceFurnaceTypeWorking spaceHeating mode andinstalled capacityHeat transfermediumTemperaturedistributionTemperatureNumber of workpiecesLoading arrangementTraysShapesizeSurface areaMassHeat conductivityHeat-up and soak timeFIGURE 6.107 The main factors that influence the heat-up and soak time. (From S. Jost, H. Langer,D. Pietsch, and P. Uhlig, Fertigungstech. Betr. 26(5):298301, 1976 [in German].)temperature, 8708C (15988F)), they found that the heat-up and soak time (t) can be calculatedusing the equationt 0:42(m=A)3:7(6:41)The regression coefficients 0.42 and 3.7 are, of course, valid for their experimental conditionsonly. Comparison with their experimentally obtained results (see the points in Figure 6.108)showed a standard deviation of s2 1.4 min2, or s +1.2 min, indicating that this way ofpredicting heat-up and soak time in specific circumstances may be quite precise.The Jost et al. [20] approach may be used generally for prediction of heat-up and soaktimes according to the general expressiont a(m=A) b(6:42)provided that for a given situation the straight line of regression and relevant values of theregression coefficients a and b are fixed by means of some preliminary experiments. It shouldbe stressed, however, that the described results of this investigation are valid for singleworkpieces only.In another investigation [21], a method enabling heat treaters to accurately determine theheat-up and soak times for different loads treated in batch-type indirect fired furnaces was5040t, min302010030405060 7080m/A , Kg/m290100 110120FIGURE 6.108 Dependence of the heat-up and soak time on the mass/surface area ratio, (m/A). (FromS. Jost, H. Langer, D. Pietsch, and P. Uhlig, Fertigungstech. Betr. 26(5):298301, 1976 [in German].) 2006 by Taylor & Francis Group, LLC.developed. To develop the method, a statistical and experimental investigation of loadtemperature conditions was performed. A computer-aided mathematical model of heat andmass transfer throughout the furnace and load was developed. The computer model accurately predicts the suitable heat-up and soak times for various types of furnace loads, loadarrangements, workpiece shapes, and thermal properties. The treated loads were divided intoseveral groups in terms of workpiece allocation and aerodynamic patterns of the furnaceatmosphere, as shown in Figure 6.109.The experiments with six different loads were conducted in indirectly fired batch furnaces,the working space of which was of length 9151680 mm, width 6101420 mm, and height 6101270 mm. The furnaces were equipped with four burners firing into the trident burner tubeslocated on the side walls, with a circulating fan located on top of the furnace as shown in Figure6.110. The thermocouples were located in different parts of the load (measuring always thesurface temperature of the workpieces)on the top and bottom, in the core, at the corners, andon the surfaces facing radiant tubesto determine temperature variations across the load.As can be seen from Figure 6.110, the heat and mass transfer in the furnace and load are verycomplicated and are characterized by nonlinear three-dimensional radiation and convectionand by nonlinear heat conduction within the workpieces. In this case, the mathematical(a)Monolayer, horizontally oriented, ordered loadsPackedSpaced(b)Monolayer, horizontally oriented, random loads(c)Multilayer ordered and random loadsPacked(d)SpacedBulkVertically oriented loadsFIGURE 6.109 Load characterization. (a) Monolayer, horizontally oriented, ordered loads; (b) monolayer, horizontally oriented, random loads; (c) multilayer ordered and random loads. (d) Verticallyoriented loads. (From M.A. Aronov, J.F. Wallace, and M.A. Ordillas, System for prediction of heat-upand soak times for bulk heat treatment processes, Proceedings of the International Heat TreatmentConference on Equipment and Processes, April, 1820, 1994, Schaumburg, IL, pp. 5561.) 2006 by Taylor & Francis Group, LLC.Roof fanRadianttubesqqwc wrqtrqtcqtrqtcqwrqpcqwcqgcqprqgcqprLoadqpcRadiation from the radiant tubesConvection from the radiant tubesRadiation from the wallsConvection from the wallsConvection from the furnace gasesRadiation between partsConduction through the partsFIGURE 6.110 Heat transfer in the used furnace and load. (From M.A. Aronov, J.F. Wallace, andM.A. Ordillas, System for prediction of heat-up and soak times for bulk heat treatment processes,Proceedings of the International Heat Treatment Conference on Equipment and Processes, April, 1820,1994, Schaumburg, IL, pp. 5561.)model to describe the heat and mass exchange is a system of integral and differential nonlinearequations. The input parameters to the computer program were as follows:Geometrical data of the furnace and load: Furnace working space dimensions,radiant tube diameter and layout in the furnace, dimensions of the baskets, numberof trays in the basket, workpiece characteristic sizeType of load (according to load characterization, see Figure 6.109)Type of steel (carbon, alloyed, high-alloy)Load thermal propertiesLoad and furnace emissivitiesTemperature conditions (initial furnace and load temperature)Fan characteristic curve parametersComposition of protective atmosphereAs an example, maximum and minimum steel part temperatures for a test (heating ofshafts) together with the calculated data are shown in Figure 6.111. The experimental datashow that the temperature curve of the load thermocouple usually reaches the set furnacetemperature well within the soak time requirements. The experimentally determined soaktime is seen to be considerably shorter than the soak time defined by the heat treater. It wasfound that the discrepancy between soak times determined from the test data and calculationsdoes not exceed 8%, which is acceptable for workshop practice.The developed computer model was used for simulation of temperature conditions fordifferent load configurations, and a generalized formula and set of graphs were developed.The generalized equation for the soak time determination ists tsb k(6:43)where ts is the calculated soak time, min; tsb is soak time for baseline temperature conditions,min; and k is a correction factor for the type of steel.The basic soak time (tsb) is obtained from graphs derived from the computer simulation.Such a graph for packed loads is shown in Figure 6.112. Other load shapes and configurations 2006 by Taylor & Francis Group, LLC.Temperature, C900800700600500400300015304560 75Time, min90105 120135Furnace temp. + Load max. exper. Load min. exper.Load max. calcul. Load min. calcul.FIGURE 6.111 Computer simulation for heating of shafts. (From M.A. Aronov, J.F. Wallace, andM.A. Ordillas, System for prediction of heat-up and soak times for bulk heat treatment processes,Proceedings of the International Heat Treatment Conference on Equipment and Processes, April, 1820,1994, Schaumburg, IL, pp. 5561.)require different graphs. The correction factor k depends on the type of steel. The generalizedequation (Equation 6.43) for the heat-up and soak time determination was set into a userfriendly computer package that incorporates charts for the calculation. This resulted in astraightforward way of determining the soak time without the use of charts while allowing fora quick and accurate soak time calculation.6.3.2 QUENCHING INTENSITY MEASUREMENTANDEVALUATION BASED ON HEAT FLUX DENSITYIn designing the method for practical measurement, recording, and evaluation of the quenching and cooling intensity in workshop conditions, in contrast to the Grossmann H value concept, which expresses quenching intensity by a single number, the main idea of Liscic wasto express the quenching intensity by continuous change of relevant thermodynamic functionsduring the whole quenching process. Instead of recording only one cooling curve (as inlaboratory-designed tests) in the center of a small (usually 1/2 in.) cylindrical specimen, the300N=3Soak time (tsb), min250N=2N = Numberof traysN=4200N=1150N>41005000 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.57Load characteristic size, in.FIGURE 6.112 Thermal soak time for a packed load. (From M.A. Aronov, J.F. Wallace, and M.A.Ordillas, System for prediction of heat-up and soak times for bulk heat treatment processes, Proceedingsof the International Heat Treatment Conference on Equipment and Processes, April, 1820, 1994,Schaumburg, IL, pp. 5561.) 2006 by Taylor & Francis Group, LLC.heat flux density at the surface of a standard-size probe becomes the main feature inmeasuring, recording, and evaluating the quenching intensity.The first substantial difference between using the small laboratory specimen and using theprobe applied in the method described below is that when quenching, for example, in an oilquenching bath, because of its small mass and small heat capacity the former will cool downin 1530 s, whereas the latter will take 500600 s to cool to the bath temperature, allowing theentire quenching process of real components to be followed.This workshop-designed method is applicable to1. All kinds of quenchants (water and brine, oils, polymer solutions, salt baths, fluidbeds, circulated gases)2. A variety of quenching conditions (different bath temperatures, different agitationrates, different fluid pressures)3. All quenching techniques (direct immersion quenching, interrupted quenching,martempering, austempering, spray quenching)The method is sufficiently sensitive to reflect changes in each of the important quenchingparameters (specific character of the quenchant, its temperature, and mode and degree ofagitation).This method....Enables a real comparison of the quenching intensity among different quenchants,different quenching conditions, and different quenching techniquesProvides an unambiguous conclusion as to which of two quenchants will give a greaterdepth of hardening (in the case of the same workpiece and same steel grade) andenables the exact calculation of cooling curves for an arbitrary point on a round barcross section of a specified diameter, to predict the resulting microstructure andhardness (an exception is the case of delayed quenching, where the cooling rate isdiscontinuously changed; for an explanation see Ref. [23])Furnishes information about thermal stresses and possible superposition of structuraltransformation stresses that will occur during a quenching processProvides the basis for automatic control of the quenching intensity during thequenching processThe method itself, known in the literature as the temperature gradient method, is based onthe known physical rule that the heat flux at the surface of a body is directly proportional to thetemperature gradient at the surface multiplied by the thermal conductivity of the body material:qldTdx(6:44)where q is the heat flux density (W/m2) (i.e., quantity of heat transferred through a surfaceunit perpendicular to the surface per unit time); l is thermal conductivity of the body material(W/(m K)), and dT/dx is the temperature gradient at the probe surface perpendicular to thesurface (K/m).The essential feature of the method is a cylindrical probe [32] of 50-mm diameter 200 mm, instrumented with three thermocouples placed along the same radius at the halflength cross section as shown in Figure 6.113. As can be seen, the thermocouple at the surfacereproducibly measures the real surface temperature of the probe (Tn), which is importantto register all the phenomena that are taking place on the surface during quenching. Theintermediate thermocouple (Ti) measures the temperature at a point 1.5 mm below thesurface. The readings of Tn and Ti enable the heat treater to easily calculate the temperaturegradient near the surface of the probe at each moment of cooling. The central thermocouple 2006 by Taylor & Francis Group, LLC.Tn200TcTi100N51.5Diam. 50All dimensions in mm FIGURE 6.113 The Liscic-NANMAC probe (made by the NANMAC Corp., Framingham Center,MA) for measurement of the temperature gradient on the surface.(Tc), placed at the center of the cross section, indicates how long it takes to extract heat fromthe core and provides at every moment the temperature difference between the surface and thecore, which is essential for the calculation of thermal stresses.Specific features of probe are the following:1. The response time of the surface thermocouple is 105 s; the fastest transienttemperature changes can be recorded.2. The intermediate thermocouple can be positioned with an accuracy of +0.025 mm.3. The surface condition of the probe can be maintained by polishing the sensing tipbefore each measurement (self-renewable thermocouple).4. The body of the probe, made of an austenitic stainless steel, does not change instructure during the heating and quenching process, nor does it evolve or absorbheat because of structural transformation.5. The size of the probe and its mass ensure sufficient heat capacity and symmetricalradial heat flow in the cross-sectional plane where the thermocouples are located.6. The heat transfer coefficient during the boiling stage, which, according to Kobasko[22], depends on bar diameter, becomes independent of diameter for diametersgreater than 50 mm. 2006 by Taylor & Francis Group, LLC.When a test of the quenching intensity is performed, the probe is heated to 8508C (15628F)in a suitable furnace and transferred quickly to the quenching bath and immersed.The probe is connected to the temperature acquisition unit and a PC. Adequate softwareenables the storage of the temperaturetime data for all three thermocouples and thecalculation and graphical display of relevant functions. The software package consists ofthree modules:Module I. TEMP-GRAD (temperature gradient method)Module II. HEAT-TRANSF (calculation of heat transfer coefficient and coolingcurves)Module III. CCT-DIAGR (prediction of microstructure and hardness after quenching)As an example let us compare two different quenching cases:Case A. Quenching in a mineral oil bath at 208C (688F) without agitation (Figure6.114a through Figure 6.114f)Case B. Quenching in a 25% polyalkalene glycol (PAG) polymer solution at 408C(1048F) and 0.8 m/s agitation rate (Figure 6.115a through Figure 6.115f)By comparing Figure 6.114b and Figure 6.115b it is clear that case B involves delayedquenching with a discontinuous change in cooling rate, because in case A the time whenmaximum heat flux density occurs (tqmax) is 15 s whereas in case B it is 72 s.In case A (oil quenching), by 20 s after immersion (see Figure 6.114e) the extractedamount of heat was already 34 MJ/m2, and by 50 s, it was 50 MJ/m2, whereas in case B(high concentration polymer solution quenching; see Figure 6.115e) by 20 s, the extractedheat was only 5 MJ/m2 and by 50 s this value was only 20 MJ/m2. However, immediately afterthat in the period between 50 and 100 s, in case A the extracted amount of heat increased from50 to only 55 MJ/m2, whereas in case B it increased from 20 to 86 MJ/m2. Both of thecalculated integral ( q dt) curves, designated with the open square symbols in Figure 6.114eand Figure 6.115e, have been calculated as the area below the heat flux density vs. timecurves, designated similarly in Figure 6.114b and Figure 6.115b. That is, they represent theheat extracted only through the surface region between the point 1.5 mm below the surfaceand the surface itself.Comparing the time required to decrease the heat flux density from its maximum to a lowvalue of, e.g., 100 kW/m2 (see Figure 6.114b and Figure 6.115b), one can see that in case A 45 sis necessary, whereas in case B only 28 s is necessary. This analysis certifies that case B(quenching in PAG polymer solution of high concentration) is a quenching process withdelayed burst of the thick polymer film.Discontinuous change in cooling rate is inherent to this quenching regime. In this respectit is interesting to analyze the cooling rate vs. surface temperature diagrams of Figure 6.114fand Figure 6.115f. While in oil quenching (case A), the cooling rate at the surface of the probe(*) has a higher maximum than the cooling rates at 1.5 mm below the surface (&) and at thecenter (D), in case B the maximum cooling rate at 1.5 mm below surface (during a certainshort period between 350 and 3008C (662 and 5728F) surface temperature) is higher than themaximum cooling rate at the surface itself. This can also be seen in Figure 6.115a, whichshows that the slope of the cooling curve Ti between 500 and 3008C (932 and 8428F) is greaterthan the slope of the cooling curve for the very surface (Tn). This is another experimentalproof that in delayed quenching cooling rates below the workpiece surface can be higher thanat the surface itself.Another analysis, with respect to thermal stresses during quenching (on which the residualstresses and possible distortion depend), is possible by comparing Figure 6.114d andFigure 6.115d. This comparison shows that quenching in a PAG polymer solution of high 2006 by Taylor & Francis Group, LLC.concentration (case B), compared to oil quenching (case A), resulted in 27% lower maximumtemperature difference (read thermal stress) between the center and surface of the probe (*)or 36% lower maximum temperature difference between the center and the point 1.5 mmbelow the surface, (D), contributing to lower distortion than in oil quenching. Whereas withoil quenching the maximum temperature difference between the center and the point 1.5 mmbelow the surface (D) is higher than the maximum temperature difference between the point1000900TcTi700600MeasuredTemperature T, C800Tn500400300200100011n2i102050100200500 1000Time t, s3c1c-nCalculated(b)130002800260024002200200018001600140012001000800600400200001c-n22i-n3c-i51020tmax50100200500 1000Time t, sCalculatedHeat flux density q, kW/m2300028002600240022002000180016001400120010008006004002000Heat flux density q, kW/m23Tn = surface temperature, T i = temperature 1.5 mm below the surface,Tc = temperature in the center of the probe.(a)(c)2100 200 300 400 500 600 700 800 900 10001100 1200Temperature Tn, C2i-n3c-i FIGURE 6.114 Graphical display from Module I, TEMP-GRAD, when quenching the LiscicNANMAC probe in a 208C mineral oil bath without agitation. (a) Measured and recorded temperaturevs. time, T f(t); (b) calculated heat flux density vs. time, q f(t); (c) calculated heat flux density vs.surface temperature, q f(Tn). 2006 by Taylor & Francis Group, LLC.500400350300CalculatedTemperature differencesT, C450250200150100500(d)11c-n22i-n5102050100200500 1000Time t, s102050100200500 1000Time t, s3c-i1009080Calculated605040t0qtdt, (MJ/m2)703020100121c-n(e)2i-n53c-i504535302520CalculatedCooling rate dT, K/sdt40151050(f)1n0 100 200 300 400 500 600 700 800 900 10001100 1200Temperature Tn, C2i3cFIGURE 6.114 (Continued ) (d) Calculated temperature differences vs. time, DT f(t). (e) Calculatedintegral q dt heat extracted vs. time. (f) Calculated cooling rates vs. surface temperature dT/dt f(Tn).1.5 mm below the surface and the surface itself. (&), In the case of delayed quenching (caseB), the maximum temperature difference between the point 1.5 mm below the surface andthe surface itself (& in Figure 6.115d) is slightly higher than the maximum temperaturedefference between the center and the point 1.5 mm below the surface (D), which is reachedabout 20 s later. This also shows an abrupt heat extraction when the polymer film bursts.On the other hand, Figure 6.114d shows that in oil quenching the maximum temperaturedifference between the center and surface (*) occurs 20 s after immersion, when the surfacetemperature is 4508C (8428F) (see Figure 6.114a), i.e., above the temperature of the Ms point. In 2006 by Taylor & Francis Group, LLC.PAG polymer solution quenching (Figure 6.115d), the maximum temperature difference between the center and the surface (*) occurs much later, when the surface temperature hasalready fallen to 3608C (6808F) (see Figure 6.115a). In this respect, dealing with steels that havea high Ms temperature, water-based polymer solutions always run a higher risk of overlappingthermal stresses with those created due to austenite-to-martensite transformation.1000900TcTiTn700MeasuredTemperature T, C800600500400300200100011n1800160014001200100080060040020001(b)Heat flux density q, kW/m2Calculated3000280026002400220020001c-n3000280026002400220020001800160014001200100080060040020000(c)1c-n22i-n53c-i102030 100 200500 1000Time t, sTqmaxCalculatedHeat flux density q, kW/m2(a)25102050 100 200300 1000Time t, s2i3cTn = surface temperature, Ti = temperature 1.5 mm below the surface,TC = temperature in the center of the probe.100 200 300 400 500 600 700 800 900 1000 1100 1200Temperature Tn , C2i-n 3c-i Tqmax FIGURE 6.115 Graphical display from Module I, TEMP-GRAD, when quenching the Liscic-NANMAC probe in a PAG polymer solution of 25% concentration, 408C bath temperature, and 0.8 m/sagitation rate. (a) Measured and recorded temperature vs. time data, T f(t). (b) Calculated heat fluxdensity vs. time, q f(t). (c) Calculated heat flux density vs. surface temperature, q f(Tn). 2006 by Taylor & Francis Group, LLC.Temperature differences T, C500450400Calculated350300250200150100500(d)1251020501002003001000Time t, s1c-n 2i-n 3c-i200180Calculated140120100t0tx q dt, MJ/m2460806040200(e)1251c-n 2i-n 3c-i1020501002003001000Time t, s25151050(f)CalculateddTCooling rate dt , K/S2001n100 200 300 400 500 600 700 800 900 1000 1100 12002i3cTemperature Tn, CFIGURE 6.115 (Continued ) (d) Calculated temperature differences vs. time, DT f(t). (e) Calculatedintegral q dt heat extracted vs. time. (f) Calculated cooling rates vs. surface temperature,dT/dt f(Tn).The probability of crack formation can be seen at a glance by comparing the surfacetemperature of the probe at the moment the maximum heat flux density occurs (Tqmax). Asseen in Figure 6.114c, Tqmax is 5158C (9598F) for oil quenching (case A), while for waterbased polymer solution (case B), Tqmax is 3808C (7168F) (see Figure 6.115c). The lower thevalue of Tqmax, the higher is the risk of crack formation, especially with steel grades havinghigh Ms temperature. 2006 by Taylor & Francis Group, LLC.When direct immersion quenching is involved with continuous cooling (not delayedquenching with discontinuous cooling), the depth of hardening, when comparing two quenching processes, is determined as follows: The larger the values of qmax and q dt and the shorterthe time tqmax,, the greater will be the depth of hardening.Module II of the software package, HEAT-TRANSF, makes it possible (based on theinput of measured surface temperatures and calculated heat flux density on the very surface)to calculate (by a numerically solved method of control volumes) and graphically present1. The heat transfer coefficient between the probes surface and the surrounding fluidvs. time, a f(t) (Figure 6.116a)2. The heat transfer coefficient between the probes surface and the surrounding fluidvs. surface temperature, a f(Tn) (Figure 6.116b)Using the calculated values of a, the software program enables the calculation of coolingcurves at any arbitrary point of the round bar cross section of different diameters, as shown inFigure 6.117a and Figure 6.117b.The Module III of the software package, CCT-DIAGR, is used to predict the resultingmicrostructure and hardness after quenching of round bar cross sections of different diameters.This module contains an open data file of CCT diagrams in which users can store up to 100 CCTdiagrams of their own choice. The program enables the user to superimpose every calculatedcooling curve on the CCT diagram of the steel in question. From the superimposed cooling curvesshown on the computer screen, the user can read off the percentage of structural phases transformed and the hardness value at the selected point after hardening as shown by Figure 6.118.Heat transfer coefficient a , W/m2K18001600140012001000800600400200Heat transfer coefficient a , W/m2K(a)00.010.1110100Time, s100018001600140012001000800600400200(b) 0 0100200300400500600700800900Temperature, CFIGURE 6.116 Heat transfer coefficient (a) vs. time and (b) vs. surface temperature when quenching the Liscic-NANMAC probe (50-mm diameter 200 mm) in a 208C mineral oil bath without agitation. 2006 by Taylor & Francis Group, LLC.900800Temperature, 8C700600500400300200100(a)00.010.1110Time, s10010000.1110Time, s1001000900800Temperature, C700600500400300200100(b)00.01FIGURE 6.117 Comparison of measured (- - -) and calculated () cooling curves for the center of a50-mm diameter bar quenched in (a) mineral oil at 208C, without agitation and (b) 25% PAG polymersolution, 408C bath temperature, and 0.8 m/s agitation rate.CSiMn0.38Chemicalcomposition0.230.64PSCrCu0.019 0.013 0.99Mo0.170.16NiV0.08 <0.011000AISI 4140900Austenitizing temp. 850 C3/4 RCTemperature, C80030700SF2600127P5A40406070Ac3Ac16010200500B4002Ms37530085 75M20010058011053 521023428 27 230103220104105106Time, sFIGURE 6.118 CCT diagram of AISI 4140 steel with superimposed calculated cooling curves forsurface (S), three-quarter radius (3/4R) and center (C) of a round bar of 50-mm diameter. 2006 by Taylor & Francis Group, LLC.If for a round bar cross section of the chosen diameter the cooling curves are calculated atthree or five characteristic points (surface, (3/4)R, (1/2)R, (1/4)R, center), using the HEATTRANSF module, the CCT-DIAGR module enables the user to read off the hardness valuesafter quenching at those points and to obtain the hardness distribution curve displayedgraphically on the computer screen. In the case of delayed quenching with discontinuouschange of cooling rate, the prediction of structural transformations and hardness values afterquenching from an ordinary CCT diagram is not correct because the incubation timeconsumed (at any point of the cross section) before the cooling rate abruptly changes hasnot been taken into account.For a detailed explanation see Shimizu and Tamura [11].6.3.3 RETAINED AUSTENITEANDCRYOGENIC TREATMENTThe martensite start (Ms) and martensite finish (Mf) temperatures for unalloyed steels dependon their carbon content, as shown in Figure 6.119. As can be seen from this diagram, whensteels of more than 0.65% C are quenched the austenite-to-martensite transformation doesnot end at room temperature (208C (688F)) but at some lower temperature, even at temperatures much lower than 08C (328F). Consequently, after these steels are quenched to roomtemperature, a portion of austenite will remain untransformed; this is referred to as retainedaustenite. The greater the amount of carbon in the steel, the greater the amount of retainedaustenite after quenching, as shown in Figure 6.120c.Retained austenite, which is a softer constituent of the structure, decreases the steelshardness after quenching. If present in amounts of more than 10%, a substantial decrease inthe hardness of the quenched steel may result (see curve a of Figure 6.120a).When quenching hypereutectoid steels from the usual hardening temperature (Figure6.120b) i.e., from the g Fe3C region, the same hardness will result independently of carboncontent (curve b in Figure 6.120a), because the hardness of martensite depends only on thecarbon dissolved in austenite (g), which further depends (according to the solubility limit, lineSE) on the hardening temperature. The structure of hardened hypereutectoid steels thereforeconsists of martensite Fe3C retained austenite.When quenching hypereutectoid steels from the region of pure austenite (g), i.e., fromabove the Acm temperature (which is not usual), the structure after hardening consists only ofmartensite and retained austenite, and the hardness decreases with carbon content as shown600Temperature, 8C500400Ms300200Mf1002000.2 0.4 0.6 0.8 1.0 1.2 1.4C, wt%FIGURE 6.119 Martensite start (Ms) and martensite finish (Mf) temperatures vs. carbon content inunalloyed steels. 2006 by Taylor & Francis Group, LLC.c1000Hardness, HV10b800600(a) Quenched from regionto 0 C400a(b) Quenched from + Fe3Cregion to 0 C(c) After transformation to 100%martensite2000., C(a)1000Usual hardening temp.E900A cm + Fe3C+800700S600Retainedaustenite, vol%(b)4030201000.0(c) content, %FIGURE 6.120 (a) Hardness of carbon (unalloyed) steels depending on carbon content and austenitizingtemperature; (b) the range of usual hardening temperatures; (c) volume percent of retained austenite.(From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlagfur Grundstoffindustrie, Leipzig, 1987.)by curve a of Figure 6.120a. If the retained austenite is transformed (e.g., by subsequentcryogenic treatment) to 100% martensite, the hardness would follow curve c in Figure 6.120a.When after hardening the steel is kept at room temperature for some time or is heated tothe temperature range corresponding to the first tempering stage, the retained austenite isstabilized, which implies that it has become more difficult to transform when subjected tocryogenic treatment. The stabilization of retained austenite is assumed to be due to thedissolution, at the arrest temperature, of the martensite nuclei formed during cooling fromthe austenitizing temperature.When martempering is performed, i.e., the quenching process is interrupted somewherearound the Ms temperature, a similar stabilization of retained austenite occurs. When thecooling to room temperature is then continued, the same effect, in principle, results as thatobtained by the subzero cryogenic treatment with respect to the transformation of retainedaustenite to martensite.The initial amount of retained austenite (after quenching) is dependent to a very largeextent on the austenitizing (hardening) temperature. The higher the hardening temperature,the greater the amount of retained austenite, but greater amounts of retained austenite mayalso be transformed to martensite by subzero cryogenic treatment for the same stabilizingtemperature and same stabilizing time as shown in Figure 6.121. The stabilizing effect 2006 by Taylor & Francis Group, LLC.120xx80x5% transt.40xx40%3020100101Stablizing temperature, C(a)102103104Stablizing time, minx120xx80x20%x40x70% transt.010110230103x10430%x804050Stablizing time, min(b)1206040x50x40x70% transt.0(c)10110210360104Stablizing time, minFIGURE 6.121 Influence of stabilizing temperature and time on the amount of retained austenitethat transforms on being subzero treated at 1808C for the ball bearing steel AISI 52100. (a) Austenitizing temperature 7808C; 9.4% retained austenite after quenching; (b) austenitizing temperature 8408C;18% retained austenite after quenching; (c) austenitizing temperature 9008C; 27% retained austeniteafter quenching. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London,1984.)increases as the stabilizing temperature and time increase. After quenching from, say, 8408C(15448F) (Figure 6.121b), there is 18% retained austenite. If the subzero treatment is carriedout within 5 min after the temperature of the steel has reached 208C (688F), about 70% of theretained austenite will be transformed. If 40 min is allowed to pass before the subzerotreatment, 60% will be transformed, and after 50 h holding at 208C (688F) only 30% of theretained austenite will respond to the subzero cryogenic treatment.If the steel is held after quenching at a higher temperature, e.g., at 1208C (2488F), for only10 min before subzero treatment, only 30% of the retained austenite will be transformed. Inorder to transform the greatest possible amount of retained austenite, the subzero cryogenictreatment should be performed immediately after quenching before tempering.The question of whether the retained austenite in the structure is always detrimental orwhether in some cases it can be advantageous has still not been answered unambiguously.When dealing with carburized and case-hardened components, because of the high carboncontent in the case, the problem of retained austenite is always a real one, especially withsteels containing nickel. The higher alloy nickel steels, such as types AISI 4620, 4820, and 2006 by Taylor & Francis Group, LLC.2400Load, kg2000160050% retained 0% retainedausteniteaustenite16MnCr5 (1.13Mn,1.00Cr)14NiCr14 (3.67Ni,0.78Cr)20MoCr4(0.49Cr, 0.5Mo)1200800400104Ck15 (Carbon steel)105106107CyclesFIGURE 6.122 Improvement of contact fatigue of carburized and case-hardened steels containing 50%retained austenite, according to C. Razim. (From J. Parrish, Adv. Mater. Process. 3:2528, 1994.)9310, are particularly likely to have retained austenite in their microstructures after heattreatment because nickel acts as an austenite stabilizer.Tests performed by M. Shea (cited in Ref. [24]), showed marked improvement in tensilebending strain values when retained austenite was present in the 2030% range for AISI 8620and 4620 steels and up to 40% for AISI 3310 steel. This report indicated that the transformation of retained austenite in the range of more than 20% allows more plastic strain to beaccommodated before crack initiation because the austenite deforms and subsequently transforms to martensite.The graph in Figure 6.122, taken from work done by C. Razim, shows where largequantities of retained austenite (in the range of 50%) improve contact fatigue of carburizedand case-hardened steels. In another publication [24], several grades of carburized and casehardened steels were compared (both before and after subzero treatment), and a clearimprovement in bend ductility is reported for those having retained austenite.As a result of the above-mentioned investigations, when dealing with carburized and casehardened gears, an amount of 1020%, and in some instances up to 25%, of retained austeniteis not objectionable for most applications and may be beneficial. On the other hand, retainedaustenite can be detrimental, causing premature wear of sliding on the components surface orof sliding and rolling of gear teeth, because it is a softer constituent of the microstructure. Thepresence of retained austenite in cases of carburized and case-hardened components that areto be subsequently ground is definitely detrimental because under certain grinding conditionsit causes severe grinding burns and cracking. The susceptibility of carburized and casehardened components to cracking during grinding becomes greater, the greater the amountof retained austenite, this amount further depending on the steel grade and carburizingtemperature as shown in Figure Transforming the Retained AusteniteWhen steels are tempered, retained austenite decomposes to bainite during the secondtempering stage (2302808C (4465368F)). For high-alloy chromium steels, hot-work steels,and high-speed steels, the range of decomposition of retained austenite is shifted towardhigher temperatures. The product of decomposition, i.e., whether it is bainite or martensite,depends on the tempering temperature and time. Bainite formation occurs isothermally, i.e., at 2006 by Taylor & Francis Group, LLC.100Carburizing time 3 h60Carburizing temp.5018CrNi815CrNi6X20MnCr516MnCr520NiCrMo67020 MoCr4in C:1000X4095030XRetained austenite, vol%80X90X201090000. 1.8Cr content, wt%FIGURE 6.123 Influence of the Cr content of low-alloy steels for carburizing on the amount of retainedaustenite at carburizing. The amount of retained austenite was determined metallographically 0.05 mmbelow the specimens surface. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl,2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.)constant temperature during the tempering process, whereas martensite forms as the steel iscooling down from the tempering temperature.Subzero cryogenic treatment may be applied to transform the retained austenite tomartensite, substantially lowering its amount, sometimes to as little as about 1 vol%, whichcannot be detected metallographically but only by x-ray diffraction.Decreasing the amount of retained austenite achieves1. Increase in hardness and consequently in wear resistance2. More dimensional stability in the finished part (smaller change in dimensions due tostructural volume change in use)3. Less susceptibility to the development of cracks at grindingFigure 6.124 shows a heat treatment cycle that includes subzero treatment. The mostimportant parameters of the treatment are (1) the temperature below 08C (328F) that shouldTemperatureAustenitizationQuenchingTemperingTimeSubzero treatmentFIGURE 6.124 A heat treatment cycle including subzero treatment. 2006 by Taylor & Francis Group, attained and (2) the cooling capacity of the equipment. In some cases, temperatures of 80to 1008C (112 to 1488F) are sufficient, but for other steels, especially high-alloy ones,lower temperatures of 1408C (2808F) or even 1808C (2928F) are necessary. Holdingtime at low temperature is unimportant, because the transformation of retained austenite tomartensite does not depend on time, but only on the temperature to which the metal has beencooled. Only that portion of the retained austenite will be transformed to martensite thatcorresponds to the cooling temperature realized. Further transformation will take place onlyif the temperature is lowered further.There are four methods using different types of equipment for the subzero treatment:1. Evaporation of dry ice (CO2 in the solid state). This method is capable of reaching atmost 758C (1038F) or 788C (1088F) and is used for small quantities and smallmass of parts.2. Circulating air that has been cooled in a heat exchanger. This low-temperaturecascade system (Figure 6.125) cools the parts put in a basket with air circulated bya fan. The air flows from top to bottom, extracting the heat of the parts, exitingthrough a grate at the bottom of the basket and flowing further through the heatexchanger, which is cooled by two or four compressors. Such metal-treating freezershave been built with a capacity to cool a mass of 270680 kg of parts to 858C(1218F) in about 2 h.3. Evaporation of liquid nitrogen. For subzero treatment of relatively small quantitiesof parts down to 1808C (2928F), equipment such as that shown in Figure 6.126 isused. The parts to be cooled are put in the working space 1, and the liquid nitrogenis in the container 4. Because of heat coming through the walls, the pressure in thecontainer 4 increases. Using this pressure, when the valve 6 is opened, liquidnitrogen is injected into the working space, where it evaporates instantly. A fan 7circulates the evaporated nitrogen through the working space, taking the heat out ofparts and lowering their temperature. The amount of the injected liquid nitrogenand consequently the temperature of cooling can be controlled by adjusting thevalve 6. The overpressure that develops in the working space because of constantnitrogen evaporation is let out through the exhaust valve 8. A temperature of1808C (2928F) can be reached in less than 10 min.4. In a container connected to a cryogenerator. This method enables subzero treatmentof large quantities of parts to as low as 1908C (3108F). The cryogenerator poweredFIGURE 6.125 Low-temperature cascade system for subzero cooling by circulating air that has beencooled in a heat exchanger. 2006 by Taylor & Francis Group, LLC.FIGURE 6.126 Subzero treatment equipment with evaporation of liquid an electric motor works on the principle of the Stirling cycle. By continuouscirculation of air the working space with parts is gradually cooled to desired temperature. Figure 6.127 shows the cooling curve from room temperature to 1808C(2928F) and the natural reheating curve from 180 to 08C (292 to 328F) for theempty container of 100 dm3, connected to a cryogenerator. It can be seen that after 1 hof cooling a temperature of 1208C (1848F) has already been reached, but anadditional 1.5 h is necessary to reach 1808C (2928F). The natural reheating from180 to 08C (292 to 328F), as shown, takes much longer (about 20 h). AND TEMPERING OF STRUCTURAL STEELSMECHANICAL PROPERTIES REQUIREDA combined heat treatment process consisting of hardening plus tempering (to temperaturesbetween 450 and 6808C (842 and 12568F)) applied to structural steels (in German calledVergutung) is performed to achieve maximum toughness at a specified strength level. Toughness is a very important mechanical property, especially for components that must be able toC25K2980273Heati n gcurvecurveC oo l i n g50100223173150123180048 12 16 2093Heating hours32Cooling hours1FIGURE 6.127 Cooling curve and natural heating back curve of an empty container connected to acryogenerator. 2006 by Taylor & Francis Group, LLC.StrengthsN < sV < sHHardenedHDuctility eN eV > eH d,yToughness zN < Z V > ZHak, sB, dsHardenedandtemperedStrengthVNormalizedNssneghuTossnehugToBrittlenessNVDuctilityeFIGURE 6.128 Schematic presentation of ductility, toughness, and brittleness. (From E. Just, VDI-Ber.256:125140, 1976 [in German].)withstand dynamic loading or impact. The aim of hardening and tempering structural steelswill be better understood if one has a clear notion of the difference between toughness andductility.Ductility is the property denoting the deformability of a material and is measured in atensile test as elongation (A in %) and reduction of area (Z in %). It is a one-dimensionalproperty. Toughness of a material, however, is a two-dimensional property because it is anintegral (or product) of strength and ductility, as schematically shown in Figure 6.128. Steelsof the same ductility but different strength levels can differ in toughness. As Figure 6.128shows, a normalized steel (N) having the same ductility as a hardened and tempered steel (V)will have lower toughness because of its lower strength level. Toughness is measured inseparate tests as impact toughness (ak, J/cm2) or as fracture toughness (KIC, N/mm3/2). Thelower the ductility of a material, the more brittle it is. Total brittleness accordingly denoteszero ductility of the material.The aim of the hardening and tempering process can also be explained by means of thestressstrain diagram schematically shown in Figure 6.129. As hardened, a steel has high yieldstrength but low ductility, and a small area below the stressstrain curve (curve 2) indicateslow toughness. As-hardened and tempered (curve 3) steel has higher yield strength than in itsnormalized condition but also much higher ductility than in its hardened condition. Thegreatest area below the stressstrain curve indicates a substantial increase in toughnesscompared to either normalized or hardened conditions.For a certain steel grade, the relation between mechanical properties and the temperingtemperature can be read off from a diagram as shown in Figure 6.130 for the steelDIN 20CrNiMo2 (0.15% C, 0.20% Si, 0.88% Mn, 0.53% Cr, 0.50% Mo, 0.86% Ni). It canbe clearly seen from the lower part of this diagram how the impact toughness increases whenthe steel is tempered to a temperature above 5508C (10228F). Such diagrams enable preciseoptimization of the strength level and toughness by selection of the proper temperingtemperature. 2006 by Taylor & Francis Group, LLC.1. Normalized2. Hardened3. Hardened and tempered2Stress (s), N/mm231Strain (e)FIGURE 6.129 Stressstrain diagram of a steel after different heat treatments. 1, Normalized; 2,hardened; 3, hardened and tempered.The properties of a hardened and tempered steel correlate to a high degree with themicrostructure after hardening and tempering. Maximum toughness values are obtainedwhen tempering a structure that after quenching consists of fine-grained martensite (havinga grain size of ASTM ! 6) (see Figure 6.131). How different microstructures after differentheat treatment processes influence the impact toughness of 3.5% Ni steel at low temperaturesRm90090Rp0.28080070700ZContraction z, %1001000Impact toughness ak, J/cm2 Elongation, %, N/mm2Tensile strength RmYield strength Rp0.2, N/mm2110060600500302010200ISO-V specimensak(20 C)10050ak(50 C)0650700500550600Tempering temperature, CFIGURE 6.130 Hardening and tempering diagram of DIN 20CrNiMo2 steel. Hardening temperature9508C; quenched in water. Specimen from a plate of 25-mm thickness; testing direction longitudinal.(From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, CarlHanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.FIGURE 6.131 Microstructure of DIN 34CrNiMo6 steel after hardening and tempering. Temperedfine-grained martensite. Magnification 500. (From G. Spur and T. Stoferle (Eds.), Handbuch derFertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)is shown in Figure 6.132. The maximum toughness is achieved after tempering waterquenched specimens (tempered martensite). When testing the impact toughness at low temperatures, the so-called transition temperature (the temperature at which a substantial drop inimpact toughness begins) is of special interest. The lower the transition temperature, thehigher the toughness. Certainly, when hardening workpieces of big cross section, not onlymartensite is obtained, but also other constituents such as bainite, pearlite, and even preeutectoid ferrite, depending on the decrease in cooling rate at quenching, below the surfacetoward the core of the workpiece. So after tempering, besides tempered martensite, otherstructural constituents having lower toughness are present.Figure 6.133 shows the relationships among transition temperature, yield strength, andmicrostructure. For high strength values especially, the superiority of fine-grained martensitestructure with respect to toughness is evident.From a series of tests with hardened and tempered steels with about 0.4% C, Figure 6.134shows a general relation between the structural constituents and the properties characterizingductility (elongation and reduction of area) and impact toughness, respectively, for different levelsof yield strength. It is clear that tempered martensite always gives the best ductility and toughness.Impact energy (ISO-V)specimens120a90bc60d300150100500Temperature, 8C50FIGURE 6.132 Influence of different microstructure and respective heat treatments on the impacttoughness at low temperatures (ISO notch specimens) of a 3.5% Ni alloyed steel. a, Hardened byquenching in water and tempered; b, normalized and tempered; c, normalized only; d, hardenedby quenching in water only. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik,Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.B%%M/25%BM/50%B%7550lM/75B%0%25103050%F+P/5%0%10 F/70B%%F/9 B100%0%BBuPF+0%10FP50BGS200RTGSASTM 3GSTransition temperature, C10050MASTM 7100ISO-V longitudinal150200400600100080012001400Yield strength Rp, N /mm2FIGURE 6.133 Transition temperature as a function of yield strength and microstructure. F, Ferrite;P, pearlite; B, bainite; Bu, upper bainite; B1, lower bainite; M, martensite; GS, grain size (ASTM). (FromG. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser,Munich, 1987.)Elongation, %25M20B15F+P105(a)070MContraction Z, %60B5040F+P302010(b)Impact energy (20 C)0250(c)200M150100500300BF+P50070090011001300Yield strength Rp, N/mm2FIGURE 6.134 (a) Elongation; (b) reduction of area; and (c) impact toughness of hardened andtempered steels having about 0.4% C, as a function of structure constituents and yield strength. F,Ferrite; P, pearlite; B, bainite; M, martensite. Grain size: ASTM 67. Impact toughness: ISO notchspecimens. Testing direction: longitudinal. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 2006 by Taylor & Francis Group, LLC.175Impact energy (20 C)150Rm = 700 N/mm2125100758501000120014005025000. content, %FIGURE 6.135 Impact toughness as a function of tensile strength and carbon content for the structureof tempered martensite. Grain size: ASTM 67. (From G. Spur and T. Stoferle (Eds.), Handbuch derFertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.)When comparing the impact toughness of tempered martensite at different strength levels(different hardness levels), one can perceive the influence of carbon content. As shown inFigure 6.135, of steels for hardening and tempering, those with 0.20.3% C have the bestimpact toughness. When testing the impact toughness of a steel, one should be aware thattoughness is usually higher in the longitudinal direction (rolling direction) than in thetransverse direction. That is because some phases or nonmetallic inclusions that are presentin every steel (carbides, oxides, and sulfides) are stretched during rolling in the longitudinaldirection. In this way a textured structure originates that has lower impact toughness in thetransverse direction than in the longitudinal direction. As a measure of this effect, the factorof isotropy (the ratio of transverse impact toughness to longitudinal impact toughness) issometimes used.The great influence of the microstructure after hardening (before tempering) on theimpact toughness of a steel is evident from Figure 6.136. Appearance of preeutectoidPercentage of structurebefore temperingImpact toughness, J/cm220098 M15050 M50 B10080 M20 F + P50012050 M50 F + P804004080Temperature, CFIGURE 6.136 Influence of the microstructure after hardening (before tempering) on the impacttoughness of DIN 16MnCr5 steel. (From H.J. Spies, G. Munch, and A. Prewitz, Neue Hutte8(22):443445, 1977 [in German].) 2006 by Taylor & Francis Group, LLC.ferrite or ferrite and pearlite in the structure results in a substantial decrease in the impacttoughness.When selecting a structural steel for hardening and tempering, in order to better adapt themechanical properties to the requirements of the treated parts, the expected microstructuremust be considered. To be able to reproducibly influence mechanical properties, one shouldknow the relationships among the heat treatment regime, microstructure, and resultingmechanical properties.Unalloyed steels for hardening and tempering, because of their low hardenability, exhibita high degree of section sensitivity with respect to hardness distribution after hardening asshown in Figure 6.137. After quenching a bar specimen of 30-mm diameter of the steel inquestion in conventional quenching oil, a hardness of only 40 HRC was achieved at thesurface. When specimens of the same diameter were quenched in fast quenching oil, thehardness was 45 HRC; when quenched in 10% aqua-quench solution the hardness was 56HRC; and when quenched in water containing 5% Na2CO3, it was 58 HRC (see Figure 6.138).This example leads to two important conclusions:1. By using different quenchants and quenching conditions, different hardness distributions can be obtained with the same steel grade and same cross-sectional size.2. With an unalloyed steel (shallow-hardener), even when the most severe quenchant isused, for large cross-sectional sizes, the depth of hardening will be small and the corewill remain unhardened.Because of the second conclusion, when selecting a structural steel grade for hardeningand tempering, its hardenability must always be adapted to the workpieces cross-sectionalsize and the required strength level. Figure 6.139 shows the preferred fields for the applicationof some common steel grades for hardening and tempering according to the actual bardiameter and the strength level required. This recommendation is based on the assumptionthat a minimum impact toughness of about 50 J/cm2 at room temperature will be achieved.As can be seen from Figure 6.139, for bigger cross-sectional sizes (bigger diameters) and6010Hardnes, HRC502030403040502002520Surface151050Center510152025 mmSurfaceFIGURE 6.137 Hardness distribution (measured) on the cross section of bars of different diametersmade of unalloyed steel (0.52% C, 0.24% Si, 0.90% Mn, 0.06% Cr) quenched in conventional hardeningoil from 8608C. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London,1984.) 2006 by Taylor & Francis Group, LLC.601020Hardness, HRC503040405030200252015105Surface051015Center2025 mmSurfaceFIGURE 6.138 Hardness distribution (measured) on the cross section of bars of 1050-mm diametersmade of unalloyed steel (0.52% C, 0.24% Si, 0.90% Mn, 0.06% Cr) quenched from 8608C in watercontaining 5% Na2CO3. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths,London, 1984.)higher strength levels, steels of higher hardenability (i.e., with more alloying elements) arerequired.6.4.2 TECHNOLOGYOF THEHARDENINGANDTEMPERING PROCESSHardening, which is the first operation of the hardening and tempering process, will yield amartensitic structure (provided a correct austenitization and quenching with a cooling rategreater than the critical rate for the steel in question have been realized), the hardness ofwhich depends on the dissolved carbon content, according to the empirical formulap(6:45)H100 % mart % 60 C 20where C is the carbon content in wt%.Tensile strength, N/mm21500140030CrNiMo81300120042CrMo437Cr490080037Cr430CrNiMo8Ck4530CrNiMo842CrMo437Cr4Ck4570060030CrNiMo842CrMo41100100030CrNiMo842CrMo442CrMo437Cr4Ck45Ck22Ck22500to 16 mm>16 to 40 mm>40 to 100 mm >100 to 160 mm >160 to 250 mmBar diameterFIGURE 6.139 Applicability of steel grades for hardening and tempering according to required strengthlevel and bar diameter. Steel designations according to DIN. (From German standard DIN 17200.) 2006 by Taylor & Francis Group, LLC.The required critical cooling rate for unalloyed steels is about 2508C/s (4828F/s). Alloyedsteels have lower critical cooling rates or a higher hardenability, which means that the samequenching conditions yield a greater depth of hardening.As a measure of hardenability, the ideal critical diameter D1 (see Chapter 5) can beapplied. The actual depth of hardening, however, is influenced by, in addition to the alloyingelements, the austenitizing temperature (especially for steels containing carbides difficult todissolve) and quenching conditions. Consequently, two steels having the same D1 value maygive different depths of hardening.For a designer, therefore, information based only on the percentage of martensite does notseem practical, because even for the same D1 values the designer might get different depths ofhardening. Besides, microstructures having the same amount of martensite do not always givethe same hardness. The hardness of martensite depends on dissolved carbon content and maybe calculated for 50% martensite according to the empirical formulapH50% mart % 44 C 14for C < 0:7 %(6:46)More practical information for the designer, about expected depth of hardening, may beobtained for round bars from the correlation among the applied radius of the bar (R in mm),quenching intensity according to the Grossmann H factor (see Chapter 5), and the equivalentdistance (E in mm) on the relevant Jominy curve. According to Just [25], this correlation forthe surface of round bars readsEsurf R1=2 3 [mm] for R < 50 mm; E < 30 mm(3=4)H 3=4(6:47)and for the core of round bars:Ecore R[mm]2H 1 = 4(6:48)After calculating the equivalent Jominy distance E, one can read off the hardness from therelevant Jominy curve. Figure 6.140 shows this correlation as a diagram (for radii from 0 to50 mm and H values from 0.3 to 2) for convenient use.As already explained, the properties of hardened and tempered parts depend first of all onhow well the hardening operation has been performed. The higher the percentage of martensite at a specified point of the cross section after hardening, the better will be the propertiesafter subsequent tempering. To check the quality of hardening achieved, the degree ofhardening (S ) can be used. It is the ratio between the achieved (measured) hardness andthe maximum hardness attainable with the steel in question:SHH pHmax 60 C 20(6:49)where H is the actual hardness measured at a specified point of the cross section, in HRC, andHmax is the maximum attainable hardness in HRC. The degree of hardening is valid, ofcourse, for the point of the cross section where the hardness was measured.For highly stressed parts that are to be hardened and tempered to high strength levels,the required degree of hardening is S > 0.95, whereas for less stressed components values ofS > 0.7 (corresponding to about 50% martensite) are satisfactory. 2006 by Taylor & Francis Group, LLC.40REs =34430H33H0.30.4Jominy distance E, mm201012 Surface040Ec =2R4H0.30.412H302010Core0010203040Radius R, mm50FIGURE 6.140 Correlation among radius of round bars (quenched by immersion), quenching intensityH and equivalent Jominy distance for the surface and core of the bars. (From E. Just, VDI-Ber. 256:125140, 1976 [in German].)When hardening and tempering structural steels, the value of hardness after hardeningand tempering is usually specified. The required degree of hardening can also be expressed asa function of the hardness after hardening and tempering (Ht):S!11 8eHt =8(6:50)Figure 6.141 shows the minimum values of the required degree of hardening as a function ofhardness after hardening and tempering, limiting the allowed area.By specifying the required degree of hardening, one can avoid the risk of an incorrecthardening and tempering. It is known that too low a value of hardness after hardening (notenough martensite) can be covered up by tempering intentionally at a lower temperature.Although in such a case the final hardness after hardening and tempering may correspond tothe required value, the toughness and other mechanical properties important for dynamicallystressed parts will be insufficient because of an inadequate microstructure. Such a risk can beavoided by specifying the minimum degree of hardness for the critical point of the crosssection, which can easily be checked after hardening.In selecting sufficiently severe quenchants to obtain a high percentage of martensite andgreat depth of hardening, one has to be aware of the risk of cracking. Hardening cracks aredependent on:1. The shape of the workpiece (big differences in the size of the cross section, edges,and corners favor the formation of cracks)2. The heat treatment process (high austenitizing temperatures and severe quenchingconditions favor the formation of cracks)3. The steel grade itself (the lower the Ms temperature of the steel, the greater the riskof cracking) 2006 by Taylor & Francis Group, LLC.S=Hh20 + 60CAllowed region0.80.6S>11 + 8eHt /80.4Not allowed region040102030Hardness after tempering Ht1700150013000.010000.2800Degree of hardening S1.0N/mm250 HRCFIGURE 6.141 Minimum values of the degree of hardening required as a function of hardness afterhardening and tempering. (From E. Just, VDI-Ber. 256:125140, 1976 [in German].)The Ms temperature can be calculated using the formulaMs 548 440 C 14Si 26Mn 11Cr 9Mo 14Ni 2V [ C](6:51)where contents of alloying elements are in wt%.The carbon content, as is known, has the greatest influence on the Ms temperature and onthe risk of cracking. Tempering, which is the second important operation, decreases highhardnesses more than low hardnesses, as can be seen in Figure 6.142. This figure shows theJominy curve of the steel DIN 28Cr4 (0.240.31% C, 0.150.36% Si, 0.620.78% Mn, 0.751.07% Cr) in hardened condition and after tempering the Jominy specimen to 500 and 6008C(932 and 11128F). It can be seen that high hardness near the quenched end of the specimen has60Hh = 20 + 60 CHardness, HRC50Steel DIN 28Cr4Hardness after Jominy testTempered to 500 C 60 minTempered to 600 C 60 min4030201000102030405060Jominy distance, mmFIGURE 6.142 Influence of tempering temperature on level of hardness at various Jominy distances.(From E. Just, VDI-Ber. 256:125140, 1976 [in German].) 2006 by Taylor & Francis Group, LLC.been decreased much more by tempering than low hardness values at greater distances fromthe quenched end. With respect to the cross section of hardened real components, this effectmeans that tempering more or less equalizes the hardness differences between surface and core.It is known that the hardness after tempering is a linear function of tempering temperature (inthe range from about 320 to 7208C (608 to 13288F)) and a logarithmic function of temperingtime, according to the following formula [25], which is valid for a 100% martensite structure:Ht 102 5:7 103 [Tt (12 log t)] [HRC](6:52)where Tt is the tempering temperature (K) and t is tempering time (s).Tempering temperature and tempering time are consequently interchangeable with respectto resulting hardness; however, very short or very long tempering times do not yield optimumtoughness. To obtain the optimum toughness for chromium steels, the tempering timesshould be between 1 and 5 h.There is a firm relationship between the hardness after tempering and the hardness afterhardening. Spies et al. [26] have, by using multiple linear regression, quantified the influenceof hardness after hardening, chemical composition, and tempering temperature on hardnessafter tempering and developed the formulaHB 2:84Hh 75(% C) 0:78(% Si) 14:24(% Mn) 14:77(% Cr) 128:22(% Mo) 54:0(% V) 0:55T t 435:66(6:53)where HB is hardness after hardening and tempering (Brinell), Hh is hardness after hardening(HRC), and Tt is tempering temperature (8C).Equation 6.53 is valid for the following ranges:HhCSiMnCrTt2065 HRC0.200.54%0.171.40%0.501.90%0.031.20%5006508C (93212028F)According to the German standard DIN 17021, an average relation between the hardnessafter hardening (Hh) and the hardness after hardening and tempering (Ht) readsHh (Tt =167 1:2)Ht 17 [HRC](6:54)where Tt is tempering temperature (8C); and Ht is hardness after hardening and tempering(HRC). This formula is valid for 4908C (9148F) < Tt < 6108C (11308F) and for tempering timeof 1 h. Because, as already mentioned, high hardnesses decrease at tempering much morethan low hardnesses, the prediction is more precise if the degree of hardening (S) is accountedfor.Hh 8 (Ht 8) exp [S(Tt =917)6 ] [HRC]where S is the degree of hardening, S < and Tt is tempering temperature (K). 2006 by Taylor & Francis Group, LLC.(6:55)650 600Hardness after hardening Hh, HRCT 8C60500400S=150050650 600500 400S = 0.74030201000102030405060Hardness after tempering Ht, HRCFIGURE 6.143 Relationship among hardness after hardening, degree of hardening, tempering temperature, and hardness after tempering. (From E. Just, VDI-Ber. 256:125140, 1976 [in German].)Figure 6.143 shows a diagram from which it is possible to predict at a glance the hardnessrequired after hardening for a desired hardness after tempering, taking into account the actualtempering temperature and the necessary degree of hardening.It is also possible to calculate the necessary tempering temperature for a specified hardnessafter hardening and tempering when chemical composition and the degree of hardening areknown:pTt 647[S(60 c 20)=Ht 0:9]1=4 3:45SHt (537 561S )(%g C) 505S(%g V) 219S(% Mo) 75S(% Cr) 66S (% Si) 51(C)(6:56)where Ht is hardness after hardening and tempering [HRC]; S is the degree of hardening, S1.0; and alloying elements are given in wt%. This formula is valid for a tempering time of 2 h.There are also diagrams for practically every steel grade from which the temperingtemperature may be determined if the ultimate tensile strength or yield strength requiredafter hardening and tempering is known. Figure 6.144a and Figure 6.144b show suchdiagrams for the unalloyed steel DIN Ck45 after quenching in water (25-, 50-, and 100mm-bar diameter) and in oil (25- and 50-mm-bar diameter), respectively.6.4.3 COMPUTER-AIDED DETERMINATIONOFPROCESS PARAMETERSIncreasingly, modern heat treatment equipment incorporates microprocessors for automaticcontrol of temperaturetime cycles, protective or reactive atmosphere, material handling,and, to some extent, quenching operations. On the other hand, determination of the processparameters necessary to achieve the heat-treated properties required is normally based onempirical results.For routine often-repeated heat treatment processes (e.g., carburizing, hardening, andtempering), computer programs can be written to establish treatment parameters providedthat adequate data are available on workpiece geometry, material properties desired afterheat treatment, steel grade used, and the actual heat treatment equipment itself. The aims ofsuch an approach are to optimize the heat treatment operation by saving time and energy andto maintain close tolerances on the material properties imparted. The basic prerequisite is the 2006 by Taylor & Francis Group, LLC.Water quenched from 830850 CDiam. 25 mmN/mm2 kp/mm2Diam. 50 mmkp/mm2Diam. 100 mmkp/mm1370 1401401401180 120120120980 1001001007808059040200200(a)Z6039080Rm80Rm60ZRp0.2Z40A50500 550 600 650 700Rm60Rp0.2 4020A50500 550 600 650 700Rp0.220A50500 550 600 650 700Tempering temperature, CN/mm2Oil quenched from 840870 CDiam. 25 mmDiam. 50 mmkp/mm2kp/mm21370 1401401180 120120980 1001007808080Rm590390Z60Rp0.2406040A5200020200500 550 600 650 700(b)RmZRp0.2A50500 550 600 650 700Tempering temperature, CFIGURE 6.144 Tempering diagrams for the unalloyed steel DIN C45 when quenched (a) from 830 to8508C in water, for bar diameters 25, 50, and 100 mm and (b) from 840 to 8708C in oil, for bar diameters 25and 50 min. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.)availability of satisfactory mathematical models that enable the presentation and predictionof relevant metallurgical and physical phenomena. As described by Liscic and Filetin [27], the process of hardening and tempering has beendivided into operations of austenitization (A), quenching (Q), and tempering (T) as shown inFigure 6.145. Within the austenitization operation, the phases are (1) preheating, (2) heatingto austenitization temperature with workpiece equalization at that temperature, and (3)homogenization of the structure. Within tempering, the phases are (5) heating to the tempering temperature and temperature equalization, (6) soaking at tempering temperature, and (7)cooling down from the tempering temperature.Computer-aided determination of process parameters has a greater number of advantagescompared with earlier methods:1. Planning of the process and preparation of the technological documentation areincomparably faster and simpler. 2006 by Taylor & Francis Group, LLC.TemperatureSurfaceCoreHardening123ATempering45Q67TTimeFIGURE 6.145 Operations and phases in the process of hardening and tempering.2. Since the computer program takes into account the influences of all relevant factors,provided that all necessary input data are used, the resulting parameters and timetemperature profiles can be determined more precisely.3. The professional level of the technological documentation is always high andconsistent, irrespective of the ability and experience of an individual technologist.4. It is possible (by using appropriate subprograms and inserting the data for alternative equipment) to examine the potential energy savings or economy of using someother equipment for the same process.5. If modern heat treatment equipment with microprocessor control is available, theresulting treatment parameters can be distributed directly (online) to all units wherethe process will be performed and controlled automatically.The general scheme of computer-aided planning of the hardening and tempering process isshown in Figure 6.146. Use of a computer for this purpose requiresPreparatory stageDatabaseonsteelcharacteristicsProductionCalculations /subroutinesFeedbackinformationInput data:on steel grade (heat)Computer-aidedprocess planningParametersfor A, Q, TOn the workpieceOn requiredmechanical propertiesOn equipmentProcess operationAQTechnologicaldocumentationandtemperaturetimediagramPerformanceof hardeningand temperingCheckingofresultantpropertiesTFeedbackinformationEquipmentcharacteristicsData onquenchingconditionsFeedbackinformationon resultantpropertiesFIGURE 6.146 Scheme of computer-aided planning of the hardening and tempering process. (From B. Liscic and T. Filetin, Heat Treat. Met. 3:6266, 1987.) 2006 by Taylor & Francis Group, LLC.1. A database on characteristics of the steel grades treated2. A database on the equipment employed (especially data on quenching severitiesavailable)3. Subprograms stored in the computer memory for the necessary calculation ofparametersThe input data in a particular case are:1. Data on the steel grade in question2. Data on the workpiece (shape, critical cross section, surface condition, number ofpieces in a batch)3. Data on mechanical properties required, after hardening and tempering, at a specified point of the cross section (hardness or ultimate tensile strength, yield strength,ductility, impact toughness, minimum grade of hardening or minimum percentageof martensite after quenching)4. Data on the equipment used for all operations and phases of the process (preheating, austenitization, quenching, tempering)The database on steels contains the following information for every specified steel gradeor heat: chemical composition; carbon equivalent; austenitizing temperature; time for homogenization of the structure; Ms temperature; Jominy hardenability curve; holding time attempering temperature; and susceptibility to temper brittleness.To determine the parameters of the hardening and tempering process, the followingrelationships must be known and stored in the computer memory in the form of adequatemathematical equations:1. The effect of shape and cross-sectional size of the workpiece on the time necessary forheating and austenitization under the specific heat transfer conditions of the equipment employed. For the case of a 40-mm bolt made of grade BS 708A37 (En 19B) steel(see Figure 6.147), because of the high value of the carbon equivalent (0.82), apreheating stage at 4508C (8428F) was necessary. For calculation of preheating time,as well as the time for heating to austenitizing temperature and temperature equalization, formula 6.42 was used (see Section 6.3.1) whereby the regression coefficients aand b were experimentally determined for the equipment used.2. The influence of steel grade, cross-sectional size, and actual quenching conditions onthe depth of hardening. This is necessary for selection of optimum quenchingconditions (quenching medium, temperature, and agitation rate) to satisfy therequired degree of hardening. The method by which this selection was performedis described in Chapter 5 (see Section 5.6.4).3. The relationship between hardness after tempering and tempering temperature forthe steel in question.The necessary tempering temperature (Tt) was calculated by means of the formula!ln (Hh 8)=(Ht 8) 1=6273 [8C]Tt 917S(6:57)where Hh is hardness after hardening, HRC; Ht is required hardness after tempering, HRC;and S is degree of hardening. This equation is valid for tempering temperatures between 390and 6608C (734 and 12208F).The time necessary for heating up to the tempering temperature and for temperatureequalization through the cross section was calculated in the same manner as for austenitization, taking into account the data for the specific tempering furnace. The holding time attempering temperature was taken from the database for the steel grade in question. 2006 by Taylor & Francis Group, LLC.Technological documentation for the process:hardening and temperingWorkpiece: boltDimensions, mm: 40 dia. x 120Steel grade: BS 708A37Requirements:R = 900 N/mm2Hardening grade = .84Point on cross section: 1/4RTechnological parametersOperationEquipment and/or mediaTemperature( C)Time(min)PreheatingChamber furnaceair atmosphere45062AustenitizationChamber furnaceair atmosphere84047QuenchingOil-agitation 1 m/s2010TemperingPit furnace-circulating air627152In air20140Cooling after temperingTotal time = 411 min900T, C80070060050040030020010012345678hFIGURE 6.147 An example of the computer-generated parameters and timetemperature cycle for hardening and tempering a 40-mm diameter bolt made of BS 708A37 (En 19B) steel. (From B. Liscicand T. Filetin, Heat Treat. Met. 3:6266, 1987.)Cooling from the tempering temperature is carried out in air or inert gas in all cases wherethe steel is not prone to temper brittleness. If it is susceptible, faster cooling in oil or in airblast in necessary.By using the described subprograms, the input data are processed by means of storedequations into the following output data or process parameters:1. Temperature and time of preheating2. Temperature and time for austenitization3. Quenching conditions (quenching medium, its temperature, its agitation rate; timerequired for complete cooling of the workpiece) 2006 by Taylor & Francis Group, LLC.TABLE 6.6Comparison of Required and Measured Hardness at the Center of Cylinders Made of SteelGrade BS 708A37 (En 19B), after Hardening and Tempering under Computer-CalculatedConditionsDimensions (mm)Required ValuesLength120200320DiameterHardeningGrade (S)UltimateTensileStrengtha(N/mm2)0.950.840.65Measured ValuesHardnessafterTempering(HRC)12401000850305080TemperingTemperature (8C)UltimateTensileStrength(N/mm2)HardnessafterTempering (HRC)5195836211210105090037.332.627.7383124.5aCalculated from the hardness (DIN 50150). Source: From B. Liscic and T. Filetin, Heat Treat. Met. 3:6266 (1987).4. Temperature and time for tempering5. Mode of cooling from tempering temperature to room temperatureFigure 6.147 shows an example of computer-generated documentation for hardening andtempering a 40-mm diameter bolt made of BS 708A37 (En 19B) steel.Table 6.6 compares the required hardness (ultimate tensile strength) for the center of barsof 30-, 50-, and 80-mm diameters made of grade BS 708A37 steel with the measured hardnessafter hardening and tempering under computer-calculated conditions.6.5 AUSTEMPERINGThe austempering process (see Figure 6.148) consists of austenitization, quenching into a hotbath maintained between 260 and 4508C (500 and 8428F), holding at this temperature untilthe transformation of austenite to bainite is complete, and cooling to room temperature atTTTaTaTemperature, CTemperature, CAc3Ttr(a)Ac3FPT trBMs(b)Time, stTime, slog tFIGURE 6.148 Scheme of an austempering process (a) in timetemperature diagram and (b) in isothermal transformation (IT) diagram. 2006 by Taylor & Francis Group, LLC.will. Compared with the process of hardening and tempering, there are the following substantial differences:1. At austempering there is no austenite-to-martensite transformation, but the finalstructure (bainite) is obtained gradually during the isothermal transformation ofaustenite to bainite.2. After austempering there is no tempering.3. While hardening and tempering is a two-operation process, austempering is performed in one cycle only, which is an advantage for the automation of the processDealing with austempering one should use the IT diagram of the steel in question tooptimize the process parameters, among them first of all the transformation temperature (Ttr)and holding time at this temperature.Austempering of steel offers two primary potential advantages:1. Reduced distortion and less possibility of cracking2. Increased ductility and toughness, especially in the range of high strength (hardness)values between 50 and 55 HRCReduced distortion and less possibility of cracking are the result of lower thermal stresses,as well as lower transformational stresses compared to conventional hardening. Althoughat austempering there are also temperature differences between the surface and core ofthe workpiece, during quenching, these differences are substantially smaller, as shown inFigure 6.149, because the difference between the austenitizing temperature and the temperature of the quenching bath is much smaller (for 2004008C (3927528F)) than in conventionalhardening. Smaller temperature gradients across the section mean smaller thermal stresses.On the other hand, at austempering there is no momentary austenite-to-martensite transformation, connected with the volume increase, taking place at different moments at differentpoints of the cross section. Instead there is a gradual transformation from austenite to bainite,which takes place almost simultaneously in thin and thick cross sections. Both effectscontribute to much lower risk of cracking and distortion, thereby minimizing the productionof scraped parts and additional cost for straightening or grinding to repair the distortion.TemperatureThe greatest temperature differencebetween surface and core at IThe greatest temperature differencebetween surface and core at IICSS = surfaceC = coreCSTimeI Conventional hardening and temperingII AustemperingFIGURE 6.149 Temperature differences between surface and core of the workpiece in conventionalhardening and in austempering. (From K.H. Illgner, Fachber. Huttenpraxis Metallweiterverarb.17(4):281288, 1979 [in German].) 2006 by Taylor & Francis Group, LLC.Increased ductility and toughness as well as increased bendability and fatigue life are thestrongest reasons to apply austempering instead of hardening and tempering. Figure 6.150shows the relation of impact toughness and Brinell hardness (HB) of a CrMnSi steel afterconventional hardening and tempering and after austempering, as a function of temperingtemperature and austempering temperature, respectively. The most important difference isthat a good combination of hardness and toughness after conventional hardening andtempering is possible only at high tempering temperatures, which means low hardness,whereas at austempering a good combination of hardness and impact toughness may beachieved at high hardness values.Another comparison of impact toughness of a carbon steel after hardening and temperingand after austempering, as a function of hardness, is shown in Figure 6.151. It is evident thataustempering yields much better impact toughness, especially at high hardness, around 50HRC. It is necessary to emphasize that high toughness after austempering is possible onlyunder conditions of complete transformation of austenite to bainite. Table 6.7 shows acomparison of some mechanical properties of austempered and of hardened and temperedbars made of AISI 1090 steel. In spite of having a little higher tensile strength and hardness,austempered specimens have had remarkably higher elongation, reduction of area, andfatigue life.Figure 6.152 shows the fatigue diagram of DIN 30SiMnCr4 steel after conventionalhardening and tempering and after austempering. The increase in fatigue resistance valuesis especially remarkable for notched specimens.Regarding bendability, Figure 6.153, from an early work of Davenport [30], shows theresults of bending a carbon steel wire austempered and hardened and tempered to 50 HRC.When selecting a steel for austempering, IT diagrams should be consulted. The suitabilityof a steel for austempering is limited first of all with minimum incubation time (the distance ofthe transformation start curve from the ordinate). Another limitation may be the very longtransformation time. Figure 6.154 shows the transformation characteristics of four AISIgrades of steel in relation to their suitability for austempering. The AISI 1080 steel has onlylimited suitability for austempering (i.e., may be used only for very thin cross sections)550HardnessHardening and temperingHardnessHardness HB5001412104508400350300250Impacttoughness64ImpacttoughnessImpact toughness, kg/cm2Austempering60020200250300350400 300400500 550 600Austempering temperature, C Tempering temperature, CFIGURE 6.150 Impact toughness and hardness (HB) of five heats of a CrMnSi steel after conventional hardening and tempering and after austempering, as a function of tempering temperature andaustempering temperature, respectively. (From F.W. Eysell, Z. TZ Prakt. Metallbearb. 66:9499, 1972[in German].) 2006 by Taylor & Francis Group, LLC.Rod diam. 0.180 in.C0.74%Mn0.37%Si0.145%0.039%SP0.044%Energy absorbed in breaking impact strength, ft-lb50Each plotted pointrepresents theaverage ofseveral tests4030Quenchandtempermethod20Directmethodaustempered1004045505560Rockwell C hardness65FIGURE 6.151 Comparison of impact toughness of a carbon steel after conventional hardening andtempering and after austempering, as a function of hardness. (From G. Krauss, Steels: Heat Treatmentand Processing Principles, ASM International, Materials Park, OH, 1990.)because the pearlite reaction starts too soon near 5408C (10048F). The AISI 5140 steel is wellsuited to austempering. It is impossible to austemper the AISI 1034 steel because of theextremely fast pearlite reaction at 5405958C (100411038F). The AISI 9261 steel is not suitedto austempering because the reaction to form bainite is too slow (too long a transformationtime) at 2604008C (5007528F).The austempering process is limited to sections that can be cooled at a sufficient rate toprevent transformation to pearlite during quenching to the austempering bath temperature.Maximum section thickness is therefore important in determining whether or not a part canbe successfully austempered. For eutectoid or hypereutectoid carbon steels like AISI 1080, asection thickness of about 5 mm is the maximum that can be austempered to a fully bainiticstructure. Unalloyed steels of lower carbon content are restricted to a proportionately lesserthickness (except those containing boron). With increasing alloy content, heavier sections canbe austempered, in some alloy steels up to 25 mm cross section. When some pearlite ispermissible in the microstructure, even carbon steels can be austempered to sections significantly thicker than 5 mm. Table 6.8 lists section sizes and hardness values of austemperedparts made of various steels.Process parameters for the austempering process are:1. Austenitizing temperature and time2. Quenching intensity when cooling from the austenitizing temperature to the austempering bath temperature3. Temperature of transformation, i.e., the austempering bath temperature4. Holding time at austempering temperatureThe austenitizing temperature and time (as in any hardening process) are responsible forcarbide dissolution and homogenizing of the structure, which has a substantial influence onthe impact toughness of treated parts. 2006 by Taylor & Francis Group, LLC.TABLE 6.7Comparison of Some Mechanical Properties of Austempered and of Hardenedand Tempered Bars Made of AISI 1090 SteelPropertyaAustempered at 4008C (7508F)bQuenched and Temperedc1,415 (205)1,020 (148)11.530415105,000e1,380 (200)895 (130)6.010.238858,600fTensile strength, MPa (ksi)Yield strength, MPa (ksi)Elongation, %Reduction of area, %Hardness, HBFatigue cyclesdaAverage values.Six tests.cTwo tests.dFatigue specimens 21 mm (0.812 in.) in diameter.eSeven tests: range 69,050137,000.fEight tests: range 43,12095,220.Source: From ASM Handbook, 9th ed., Vol. 4, Heat Treating, ASM International, Materials Park, OH, 1991,p. 155.bQuenching must be severe enough to avoid any pearlite formation on cooling from theaustenitizing temperature to the temperature of the austempering bath. Molten nitritenitratesalts are used as quenching media for austempering. To increase the quenching severity,agitation and sometimes the addition of some percentage of water is used. When adding waterto a hot salt bath, care must be taken to prevent spattering. The higher the temperature of thesalt bath, the less water should be added. Because of evaporation, the amount of water addedmust be controlled.Bending fatigue strength s, N/mm2720680640Smoothspecimens600560Austempered sB = 1260 N/mm2Hardened + tempered sB = 1380 N/mm2Hardened + tempered sB = 1260 N/mm2520480440400Notchedspecimens3603202802345623 457 10Number of cycles7107FIGURE 6.152 Fatigue of DIN 30SiMnCr4 steel after conventional hardening and tempering and afteraustempering. (From F.W. Eysell, Z. TZ Prakt. Metallbearb. 66:9499, 1972 [in German].) 2006 by Taylor & Francis Group, LLC.Hardness: Rockwell C 50AustemperedQuenched and temperedFIGURE 6.153 Carbon steel wire (0.78% C, 0.58% Mn) of 4.6-mm diameter, (left) austemperedand (right) hardened and tempered to 50 HRC and bent under comparable conditions. (From E.S.Davenport, Steel, March 29, 1937.)800800500400F+CA+F+CA300Ms20010001 min1080110110(a)10 min210Time, s1 day10 h1h341010600A+F300MI2005140101Ms2001 min103410111010 min102Time, s6211031 day10 h1h104Ae3Ae170010102Time, s103104283438373137415055F+C500A+F+C400A300Ms200100520105(d)1 min92611105A600Temperature, CF+C3000(c)122525273543A+F+C100F+CA+F+CMs(b)Ae1500400A400010Hardness, HRCTemperature, C600A+F500800Ae313243126303744505153100665800700Ae11010210 min103Time, s1 day10 h1h104Hardness, HRCTemperature, C600Ae3 A700Temperature, C11323840404143505557Hardness, HRCAe1700Hardness, HRCTransformation temperature, i.e., austempering bath temperature, is one of the two mostimportant parameters as it directly influences the strength (hardness) level of the treatedparts. The higher the austempering temperature, the lower the strength (hardness) of theaustempered parts. The bainitic region can be divided according to austempering temperatureinto upper and lower bainite regions, the boundary between them at about 3508C (6628F).The structure of upper bainite in steels (consisting of parallel plates of carbides and ferrite) issofter and tougher, whereas the structure of lower bainite (needlelike, with small carbidesunder 608 within the needles) is harder and more brittle.65105FIGURE 6.154 Transformation characteristics of steel of AISI grades (a) 1080; (b) 5140; (c) 1034; and(d) 9261 in relation to their suitability for austempering (see text). (From ASM Handbook, 9th ed., Vol.4, Heat Treating, ASM International, Materials Park, OH, 1991.) 2006 by Taylor & Francis Group, LLC.TABLE 6.8Section Sizes and Hardness Values of Austempered Parts of Various Steel GradesSection SizeSteelnm1050106510661084108610901090e1095135040634150436551405160e875050100b35c7c6c13c5c20c4c16c16c13c25c3b26c3b8cSalt Temperaturein.Ms Temperaturea8Cb0.1250.187c0.281c0.218c0.516c0.187c0.820c0.148c0.625c0.625c0.500c1.0000.125b1.035c0.125b0.312c8F8C8FHardness (HRC)345ddddd315fddddd345315f315d655ddddd600fddddd655600f600d320275260200215210g235245285210330255285610525500395420410g45047554541063049054541475356536655585558576044.5 (avg.)57605356535652 max54 max434846.7 (avg.)47485760aCalculated.Sheet thickness.cDiameter of section.dSalt temperature adjusted to give maximum hardness and 100% bainite.eModified austempering; microstructure contained pearlite as well as bainite.fSalt with water additions.gExperimental value.bSource: From ASM Handbook, 9th ed., Vol. 4, Heat Treating, ASM International, Materials Park, OH, 1991, p. 155.Because not only the strength (hardness) level but also the impact toughness varies withaustempering temperature, the temperature of the austempering bath must be kept withinclose tolerance (+68C (+438F)).The holding time at austempering temperature should be sufficient to allow completetransformation. Therefore the IT diagram of the steel grade in question should be consulted.Allowing parts to remain in the bath for longer than the required time will increase the cost oftreatment but it is not harmful to the mechanical properties of austempered parts.REFERENCES1. G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, CarlHanser, Munich, 1987.2. H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlagfur Grundstoffindustrie, Leipzig, 1987. 3. B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation andcalculation of heat transfer data, Proceedings of the First International Conference on Quenching andControl of Distortion, Chicago, IL, September 1992, pp. 1726. 2006 by Taylor & Francis Group, LLC.4. A. Rose and W. Strassburg, Anwendung des Zeit-Temperatur-Umwandlungs-Schaubildes furkontinuierliche Abkuhlung auf Fragen der Warmebehandlung, Archiv. Eisenhuttenwes. 24(11/12):505514 (1953) (in German).5. H.P. Hougardy, Die Darstellung des Umwandlungsverhaltens von Stahlen in den ZTU-Schaubildern, Harterei-Tech. Mitt. 33(2):6370 (1978) (in German).6. E.Scheil, Arch. Eisenhuttenwes. 8:565567 (1934/1935) (in German).7. F. Wever and A. Rose (Eds.), Atals zur Warmebehandlung der Stahle, Vols. I and II, VerlagStahleisen, Dusseldorf, 1954/56/58.8. W. Peter and H. Hassdenteufel, Aussagefahigkeit der Stirnabschreckprufung und der Zeit-Temperatur-Umwandlungsschaubildes fur das Ergebnis der Hartung von Rundstaben, Stahl Eisen87(8):455457 (1967) (in German).9. M. Atkins, Atlas of Continuous Transformation Diagrams for Engineering Steels, British SteelCorporation, BSC Billet, Bar and Rod Product, Sheffield, U.K., 1977.10. E.A. Loria, Transformation behaviour on air cooling steel in A3A1 temperature range, Met.Technol., 490492 (1977).11. N. Shimizu and I. Tamura, Effect of discontinuous change in cooling rate during continuouscooling on pearlitic transformation behaviour of steel, Trans. ISIJ 17:469476 (1977).12. ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970. 13. B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag,New York, 1992.14. K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.15. J. Frehser and O. Lowitzer, The process of dimensional change during the heat treatment of toolsteels, Stahl Eisen. 77(18):12211233 (1957) (in German).nning, Verfahrenstechnik des Isothermgluhens, Harterei-Tech. Mitt. 32:4349 (1977) (in German).16. J. Wu17. G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International, Materials Park,OH, 1990.18. J. Orlich, A. Rose, and P. Wiest (Eds.), Atlas zur Warmebehandlung der Stahle, Vol. 3, ZeitTemperatur-Austenitisierung-Schaubilder Verlag Stahleisen, Dusseldorf, 1973 (in German).19. J. Orlich and H.J. Pietrzenivk (Eds.), Atlas zur Warmebehandlung der Stahle, Vol. 4, Zeit-Temperatur-Austenitisierung-Schaubilder Part 2, Verlag Stahleisen, Dusseldorf, 1976 (in German).20. S. Jost, H. Langer, D. Pietsch, and P. Uhlig, Rechnerische Ermittlung der Erwarmdauer bei derWarmebehandlung von Stahl, Fertigungstech. Betr. 26(5):298301 (1976) (in German).21. M.A. Aronov, J.F. Wallace, and M.A. Ordillas, System for prediction of heat-up and soak times forbulk heat treatment processes, Proceedings of the International Heat Treatment Conference onEquipment and Processes, April 1820, 1994, Schaumburg, IL, pp. 5561.22. N.I. Kobasko, Teplovie procesi pri zakalke stali, Mettaloved. Termiceskaja Obrab. Metalov. 3:26 (1968). 23. B. Liscic and G.E. Totten, Controllable delayed quenching, Proceedings of the International Conference on Equipment and Processes, April 1820, 1994, Schaumburg, IL, pp. 253262.24. J. Parrish, Retained austenite: new look at an old debate, Adv. Mater. Process. 3:2528 (1994).25. E. Just, Verguten-Werkstoffbeeinflussung durch Harten und Anlassen, VDI-Ber. 256:125140(1976) (in German).glichkeiten der Optimierung der Auswahl vergunch, and A. Prewitz, Motbarer26. H.J. Spies, G. Mutbarkeit, Neue Hutte 8(22):443445 (1977) (in German).Baustahle durch Berechnung der Hart-und Vergu 27. B. Liscic and T. Filetin, Computer-aided determination of the process parameters for hardening andtempering structural steels, Heat Treat. Met. 3:6266 (1987).28. K.H. Illgner, Qualitats-und Kostenvorteile mit Zwischenstufenvergutungsanlagen im Vergleich zunormalen Vergutungsanlagen, Fachber. Huttenpraxis Metallweiterverarb. 17(4):281288 (1979) (inGerman).29. F.W. Eysell, Die Zwischenstufenvergutung und ihre betriebliche Anwendung, Z. TZ Prakt. Metallbearb. 66:9499 (1972) (in German).30. E.S. Davenport, Heat treatment of steel by direct transformation from austenite, Steel, 29: (1937).31. ASM Handbook, 9th ed., Vol. 4, Heat Treating, ASM International, Materials Park, OH, 1991.32. U.S. Patent 2,829,185, regarding surface temperature measurement.33. Met. Prog., October 1963, p. 134. 2006 by Taylor & Francis Group, LLC.7Heat Treatment with GaseousAtmospheresJohann GroschCONTENTS7.17.27.3General Introduction ................................................................................................. 415Fundamentals in Common ........................................................................................ 417Carburizing ................................................................................................................ 4227.3.1 Introduction..................................................................................................... 4227.3.2 Carburizing and Decarburizing with Gases..................................................... 4227.3.2.1 Gas Equilibria ................................................................................... 4237.3.2.2 Kinetics of Carburizing ..................................................................... 4267.3.2.3 Control of Carburizing...................................................................... 4287.3.2.4 Carbonitriding................................................................................... 4317.3.3 Hardenability and Microstructures ................................................................. 4327.4 Reactions with Hydrogen and with Oxygen .............................................................. 4407.5 Nitriding and Nitrocarburizing.................................................................................. 4467.5.1 Introduction..................................................................................................... 4467.5.2 Structural Data and Microstructures .............................................................. 4487.5.2.1 Structural Data ................................................................................. 4487.5.2.2 Microstructures of Nitrided Iron ...................................................... 4507.5.2.3 Microstructures of Nitrided and Nitrocarburized Steels................... 4527.5.2.4 Microstructural Specialties................................................................ 4567.5.3 Nitriding and Nitrocarburizing Processes ....................................................... 4577.5.3.1 Nitriding............................................................................................ 4577.5.3.2 Nitrocarburizing................................................................................ 4607.5.3.3 Processing Effects on the Nitriding and Nitrocarburizing Results ... 4617.6 Properties of Carburized and Nitrided or Nitrocarburized Components .................. 463References .......................................................................................................................... 4697.1 GENERAL INTRODUCTIONHeat treatment of components is to date mostly accomplished in gaseous atmospheres, themore so if plasma and vacuum are regarded as special cases of gaseous atmospheres. Incomparison, heat treatment in solid or liquid media is negligible in numbers. Heat treatmentin gaseous atmospheres falls into two categories: processes with the aim of avoiding a masstransfer between the gaseous atmosphere and the material, and processes with the aim ofachieving just such a transfer. Mass transfer occurs when there is a difference in the potentialbetween the constituents of a gaseous atmosphere and those of the microstructure of acomponent. The direction of such a mass transfer is determined by the potential difference,which leaves two fundamental possibilities with regard to the component. One is the intake 2006 by Taylor & Francis Group, LLC.of elements of the gaseous atmosphere into the component and the other is the emission ofelements of the component into the gaseous atmosphere. This kind of heat treatment fallsunder heat treatment with gas, which is the subject of this chapter. The deposition ofconstituents of a gaseous atmosphere onto the surface of a component (coating), which isnot connected with the described mass transfer mechanism, is therefore excluded from thesubject of heat treatment with gas.Consequently, a protective gas is a gaseous atmosphere that is free from a potentialdifference with respect to those elements of both gas and steel that have the ability to transfermass. A central matter of concern with all the homogeneity treatments (annealing, austenitizing, tempering) is to prevent oxidation. Gas compositions suitable for reducing oxidation mayhave a potential difference against carbon; furthermore, a reduction of oxide layers makes italways possible for carbon to get into the gas. Protective gases with a reducing effect musthence be adjusted to the carbon content of the steel to prevent decarburization. Inert gasessuch as rare gas or pure nitrogen as well as high-quality vacua do not contain any reactantconstituents and thus prevent a mass transfer. In processes without a mass transfer, thegaseous atmosphere as a protective gas is an important and basic requirement of heattreatment but not used as a parameter to attain or alter certain properties of the component.Processes like these are therefore referred to as heat treatment in gas.Transport of matter and heat conduction can formally be calculated by applying the samerules. The heat conduction in steel, however, is of a much higher order than the transport ofmatter, which as a diffusion process causes a uniformly directed flow of atoms. Heattreatments with gases are therefore always isothermal processes. As a rule, the rather slowprocess of diffusion determines the time needed for the technical processing of a heattreatment with mass transfer. This in turn determines essential processing conditions.During a technically and economically justifiable treatment time, only the atoms that areinterstitially soluted in iron are absorbed in adequate quantity and sufficiently deep to meet thegiven requirements. Therefore elements used in heat treating are carbon, nitrogen, oxygen, andhydrogen. With these interstitally soluted atoms too, the exchange is limited to the case, withthe exception of thin sheets and hydrogen as the smallest element that diffuses most easily,where it is possible to influence the bulk material. Heat treatment with gases is therefore mostlya surface phenomenon. The corresponding thermochemical surface hardening processes withgases are carburizing and decarburizing, nitriding and denitriding, and soaking as well as thecombined processes carbonitriding and nitrocarburizing. Treatments with oxygen as the reactant cause almost always an oxide layer (controlled oxidizing, blueing) or lead to a reduction ofoxide layers. Boriding with gaseous boron sources is seldom done in practice because thepredominant medium, diborane B2F6, is highly toxic and the boron halides BBr3, BF3, orBCl3 are also seldom used due to their corrosive effect in humid condition. Plasma-assistedboriding with trimethyl borate B(OCH3)3 is still on the laboratory scale.A focus of the industrial heat treating of steel with gases is above all carburizing, surfaceheat treatment in the austenite phase field, nitrocarburizing and, to a lesser extent, nitriding,surface heat treatment in the ferrite phase field. Consequently, these heat treating processeswill be dealt with in detail as to the fundamental principles of introducing carbon, nitrogen, ora mixture of both into the case of a steel, as to the characteristics of the heat treatedmicrostuctures and the properties of carburized or nitrocarburized components. Oxidation,reduction, and the effect and composition of protective gases are often connected withdecarburizing processes. Hence this topic will be dealt with following the discussion oncarburizing. Some conditions of a desired oxidation will also be treated in connection withoxynitriding. A discussion of the fundamental principles of reactions in and with gaseousatmospheres and of diffusion in solid metals the above-mentioned processes have in commonwill precede the main chapters. 2006 by Taylor & Francis Group, LLC.7.2 FUNDAMENTALS IN COMMON [15]The absorption of material from a gaseous atmosphere occurs in several steps [6,7]:Processes in the gaseous atmosphere: Formation of transportable gas molecules andtransfer of these molecules through the gas phase onto the surface of the metal withsubsequent physical adsorption of the gas moleculesProcesses in the interface: Dissociation of the gas molecules and chemisorption of the gasatoms, penetration of the atoms through the surface of the metal with transition of theatoms from the state of chemisorption to the interstitially solute state in the solid solutionDiffusion of the atoms from the surface into the core of the materialThese steps are based on the premise that there is a potential difference between gas and steel.By analogy, the described steps are also valid for the emission of material, the atoms emittedfrom the solid solution recombine into molecules at the surface of the material and penetrateinto the gaseous atmosphere.Independent of the composition of the initial gases, the gaseous atmospheres used in heattreating at processing temperatures consist of the elementary molecules carbon monoxideCO, carbon dioxide CO2, hydrogen H2, water vapor H2O, oxygen O2, ammonia NH3, andsometimes also methane CH4, all of which are able to react with one another and with thecatalyzing surface of the component (and the furnace wall), thus releasing or absorbingcarbon, oxygen, nitrogen, and hydrogen. Reactions among the constituents of the gas aredescribed as homogeneous reactions; reactions between elements of the gas and elements ofthe component surface are described as heterogeneous reactions. The heterogeneous reactionsthat take place in the interface between gaseous atmosphere and component surface arechemophysical processes and responsible for the mass transfer. The reactions, i.e., thetransition from an initial state to a final state, are accompanied by a change in the energy u,which is determined by the first law of thermodynamicsdu q w(7:1)where q is the amount of heat added during the change of states and w the work done by thesystem. In many cases it suffices to just consider the work against the surrounding pressure(volume work):w P dv(7:2)du q P dv(7:3)Thus Equation 7.1 can be rewrittenCombined with the entropy derived from the second law of thermodynamicsds dqrevT(7:4)follows the basic equation for reversible thermodynamic processes, at constant pressure andconstant temperature:du T ds p dv 2006 by Taylor & Francis Group, LLC.(7:5)This relationship allows the derivation of thermodynamic potentials, one of which is the freeenthalpy, or Gibbs free energygupvT s(7:6)which is subsequently needed.In a closed system the gas reactions approach a dynamic equilibrium state, which isdetermined by pressure, temperature, gas composition, and material composition. In thisdynamic equilibrium state, the Gibbs free energy is at its minimum and reactions and reversereactions, on average, take the same amount of time, i.e., the total of locally absorbed andemitted particles equals zero, thus causing the net flow to cease. The rate of a chemicalreaction is in proportion to the active masses of the involved elements, which for gaseousmaterial are described by their partial pressures pA (volume of constituent A total pressurein the system). From the fact that reaction and reverse reaction take the same amount of time,it follows for a general reactionaA bB cC d D(7:7)(a through d are the stoichiometric factors of the reaction components A through D) that theequilibrium constant Kp of the process isKp p c pdDCp a pbBA(7:8)where, by agreement, the reaction products C and D are placed over the reactants A and B.Thus, the gas composition is replaced by the equilibrium constant Kp, which is independent ofpressure.The Gibbs free energy for standard state is thusdg0 R T ln dg0pc pdCD RT ln Kppa pbAB(7:9)Values of the Gibbs free energy for standard state reactions have been studied for manyreactions and can be found in special tables [8,9].The values of the partial pressures of the gaseous atmospheres discussed here varybetween 1017 and 1025 bar at treatment temperature and are thus for reasons of convenience often replaced by the activity ai of the gases by relating the partial pressure p to astandard state pressure. It is most usual to choose as standard state the partial pressure p0 1of the pure component in the same phase at the same temperature, which for carbon, forinstance, is the steam pressure pC of graphite (i.e., the activity of the graphite-saturated0austenite is by definition aCarbon 1 [10]).Equation 7.9 can thus be rewritten asdg0 R T lna C aDa A aB(7:10)These deductions are only valid for reactions where substance is neither added to noremitted from the system. On changing the amount of substance, the constituents must be takeninto consideration by means of their chemical potential, which as partial Gibbs free energy 2006 by Taylor & Francis Group, LLC.mi @gdniT , p, j@edni S,V , j(7:11)is defined with dni moles of the substance i. In this case it is necessary to completeEquation 7.5du T ds p dv Smi dni(7:12)The mass transfer within the interface is technically described by the mass transfer coefficient b,which determines how fast the particles move, with the mass transfer equation~m b(agas asurface )(7:13)and is therefore also called effective reaction rate constant. The direction of the mass transfer isdetermined by the activity gradient between gaseous atmosphere and the surface of the steel.The reaction rate constant indicates the total of mass transfer in the interface and comprises asa global value the effects of material, the microgeometrical state of the surface, flow conditions,pressure, and temperature on the mass transfer. The individual physicochemical reactions thatoccur in the interface cannot be described by the mass transfer coefficient.Single-phase systems are homogeneous when at thermodynamic equilibrium, differencesin the distribution of the involved atoms such as those caused during production are equalizedby matter flowing from regions of higher concentration toward regions of lower concentration. The cause of this flux, which is called diffusion, is the difference in the chemicalpotential mc of the diffusing substance. The partial molar Gibbs free energy, according toEquation 7.11, can be rewrittenmc dgdc(7:14)where c is the concentration of the diffusing substance.The potential difference is equalized by the flux~mD @ m ccgrad cRT @ c(7:15)where m is the number of atoms c that penetrate the area F in the time t, R is the general gasconstant (8.314 J=mol K), and D* the diffusion coefficient (or diffusivity).In multiphase systems with different chemical potentials it is likely that potential jumpsoccur at the phase boundaries; in this case it may happen that the flux is opposed to theconcentration gradient (uphill diffusion). On carbide formation, for instance, carbon diffusesfrom the saturated austenite into carbide with a higher concentration of carbon which,however, has a lower chemical potential in the carbide.In dilute solutions, i.e., when the amount of the diffusing material is small, it is possible touse, with adequate precision, the more easily accessible concentration gradient as a drivingforce. This approach is valid for the thermochemical surface treatment and leads to Ficksfirst law of diffusion [11]@c@c D@t@x 2006 by Taylor & Francis Group, LLC.(7:16)according to which the variation in time of the concentration depends on the concentrationgradient @ c=@ x parallel to the x-axis. The effective diffusion coefficient D has the unitarea=time and is usually expressed in cm2=s. Ficks first law is valid when there is change intime of the concentration gradient and thus none of the flux. Frequently, diffusion causes achange in the concentration gradient and thus becomes dependent on time and location. Thisis covered by Ficks second law [11]:@c@ @c D@t@x @xwith D f (c)(7:17)or, if D is independent of concentration and consequently of location,@c@2c D 2@t@xwith D 6 f (c)(7:18)In the case of a semiinfinite system, that is when the diffusion flow does not reach the end ofthe specimen as is the case in thermochemical surface treatment, Ficks second law as given inEquation 7.18, the van OrstrandDewey solution [1214] appliesxc(x,t) c0 (cs c0 ) 1 erf p2 Dt!(7:19)where c(x,t) is the concentration c at a distance x from the surface of a steel with the initialconcentration c0 following a diffusion time t, and cs is the surface concentration of thediffusing element (erf is the Gaussian error function). According to this relation, the depthof penetration increases in proportion to the root of the diffusion time, which leads to theempirical rule that to get double the depth of penetration it is necessary to quadruple thediffusion time.Ficks second law can also be resolved when substance is emitted, i.e., when cs is smallerthan c0, in the formx(7:20)c(x,t) cs (c0 cs ) erf p2 DtThe diffusion coefficient is given by the empirical relationshipQD D0 exp RT(7:21)with the element-dependent constant D0 and the activation energy Q of the diffusion. FromEquation 7.16 and Equation 7.19 it follows that the diffusion time which is needed for aspecific depth of penetration can only be reduced by higher temperatures T and an increase inthe concentration gradient @ c=@ x. The activation energy Q is dependent on the mechanismsof solid-state diffusion. The diffusion of the gases nitrogen, oxygen, hydrogen, and of carbon,which are located interstitially mainly on octahedral voids in the lattice, occurs primarily bythe interstitial mechanism that is the cause of the already mentioned rather fast diffusion ofthe above elements, which can still occur at room temperature and even lower temperatures.The diffusion coefficients DH, DC, DN, and DO have been thoroughly studied for a-iron.Figure 7.1 [15] offers a comparison of the magnitudes between hydrogen, carbon, oxygen, andnitrogen and of substituted atoms. Figure 7.2 [4] shows a detailed plot of the diffusion 2006 by Taylor & Francis Group, LLC.1500104400Temperature ( C)200 10025 050Coefficient of diffusion (m2/s)Hydrogen1091014Interstitial atoms N, C10191024Substitutional atoms1029012345103 1TKFIGURE 7.1 Diffusion coefficients of hydrogen and of interstitial and substitional elements in a-iron.(From E. Hornbogen, Werkstoffe, 2nd ed., Springer-Verlag, Berlin, 1979.)10610101081012101010141012101610140 in -Fe101810161020101810221020102410221026DO and DC(-Fe) (m2/s)104108D N (-Fe) (m2/s)1061024102801. 1TKFIGURE 7.2 Diffusion coefficients of C, N, and O in a-iron. (From Th. Heumann, Diffusion inMetallen, Springer-Verlag, Berlin, 1992.) 2006 by Taylor & Francis Group, LLC.coefficients of carbon, oxygen, and nitrogen. The diffusion coefficients in g-iron are approximately lowered by the second power of 10, and details on this will be dealt with in Section 7.3on carburizing. It ought to be noted that the diffusion proceeds faster alongside grainboundaries than in the matrix [3,4]. The directed exchange of matter requires a difference inpotential or activity that is established and maintained by the gas composition. With processes near equilibrium it is possible to relate them to their final state and to describe them byspecifically derived and easily obtainable values such as the carbon potential, and to controlthem accordingly. With processes far from equilibrium, where it is not possible to ascertainhow great the differences in potential or activity are, it has proved helpful to use processcharacteristics that allow to maintain a required gas composition, provided that the processing conditions are fixed.7.3 CARBURIZING7.3.1 INTRODUCTIONCarburizing produces a hard and compared with their dimensions often shallow surface onrelatively soft components when the surface microstructure of steels with a (core) carboncontent of