Steel Heat Treatment - Metallurgy And Technologies
832 Pages

Steel Heat Treatment - Metallurgy And Technologies

Course Number: METALLURGY 102, Spring 2012

College/University: Indian Institute of...

Word Count: 312219

Rating:

Document Preview

ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. Preface The first edition of the Steel Heat Treatment Handbook was initially released in 1997. The objective of that book was to provide the reader with well-referenced information on the subjects covered with sufficient depth and...

Unformatted Document Excerpt
Coursehero >> India >> Indian Institute of Technology, Kharagpur >> METALLURGY 102

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

2006 ß by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. Preface The first edition of the Steel Heat Treatment Handbook was initially released in 1997. The objective of that book was to provide the reader with well-referenced information on the subjects 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 for the designer or practicing engineer. However, since the initial release of the first edition of the Steel Heat Treatment Handbook, there have been various advancements in the field that needed to be addressed to assure up-to-date coverage of the topic. This text, Steel Heat Treatment: Metallurgy and Technologies, is part of a revision of the earlier text. Some of the chapters in this text are updated revisions of the earlier book and others are completely new chapters or revisions. These chapters include: Chapter Chapter Chapter Chapter Chapter Chapter Chapter Chapter 1. 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 revised versus the first edition of the Steel Heat Treatment Handbook. This new text is current with respect to heat treatment technology at this point at the beginning of the 21st century and is considerably 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, who helped to bring the first edition of Steel Heat Treatment Handbook into fruition was unable to assist in the preparation of this second edition. However, I have endeavored to keep the same consistency and rigor of coverage as well as be true to the original vision that we had for this text as a way of serving the heat treatment industry so that this book will be a value resource to the reader in the future. George E. Totten, Ph.D., FASM Portland State University Portland, Oregon ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. Editor George 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 of a number of books including Steel Heat Treatment Handbook, Handbook of Aluminum, Handbook of Hydraulic Fluid Technology, Mechanical Tribology, and Surface Modification and Mechanisms (all titles of CRC Press), as well as the author or coauthor of over 400 technical papers, patents, and books on lubrication, hydraulics, and thermal processing. He is a Fellow of ASM International, SAE International, and the International Federation for Heat Treatment and Surface Engineering (IFHTSE), and a member of other professional organizations including ACS, ASME, and ASTM. He formerly served as president of IFHTSE. 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. ß 2006 by Taylor & Francis Group, LLC. ß 2006 by Taylor & Francis Group, LLC. Contributors S.S. Babu Edison Welding Institute Columbus, Ohio Ronald Lesley Plaut University of Sao Paulo Sao Paulo, Brazil Elhachmi Essadiqi CANMET, Materials Technology Laboratory Ottawa, ON, Canada David Pye Pye Metallurgical Consulting, Inc. Meadville, Pennsylvania Johann Grosch Institut fuer Werkstofftechnik Technische Universitaet Berlin, Germany ˇˇ ´ ˇ Bozidar Liscic Faculty of Mechanical Engineering and Naval Architecture University of Zabreb Zabreb, Croatia Guoquan Liu Beijing University of Science and Technology Beijing, China Michiharu Narazaki Utsunomiya University Utsunomiya, Japan Arnold R. Ness Bradley University Peoria, Illinois Joseph W. Newkirk University of Missouri-Rolla, Rolla, Missouri Angelo Fernando Padilha University of Sao Paulo Sao Paulo, Brazil ß 2006 by Taylor & Francis Group, LLC. Paulo Rangel Rios Fluminense Federal University V. Redonda, Brazil Anil Kumar Sinha AKS Associates Fort Wayne, Indiana Anton Stich Technical University of Munich Munich, Germany Alexey V. Sverdlin Bradley University Peoria, Illinois Hans M. Tensi Technical University of Munich Munich, Germany Sanjay N. Thakur Hazen Powder Parts, LLC Hazen, Arkansas George E. Totten Portland State University Portland, Oregon Chengjian Wu Beijing University of Science and Technology Beijing, China ß 2006 by Taylor & Francis Group, LLC. Contents Chapter 1 Steel Nomenclature Anil Kumar Sinha, Chengjian Wu, and Guoquan Liu Chapter 2 Classification and Mechanisms of Steel Transformation S.S. Babu Chapter 3 Fundamental Concepts in Steel Heat Treatment Alexey V. Sverdlin and Arnold R. Ness Chapter 4 Effects of Alloying Elements on the Heat Treatment of Steel Alexey V. Sverdlin and Arnold R. Ness Chapter 5 Hardenability ˇˇ ´ ˇ Bozidar Liscic Chapter 6 Steel Heat Treatment ˇˇ ´ ˇ Bozidar Liscic Chapter 7 Heat Treatment with Gaseous Atmospheres Johann Grosch Chapter 8 Nitriding Techniques, Ferritic Nitrocarburizing, and Austenitic Nitrocarburizing Techniques and Methods David Pye Chapter 9 Quenching and Quenching Technology Hans M. Tensi, Anton Stich, and George E. Totten Chapter 10 Distortion of Heat-Treated Components Michiharu Narazaki and George E. Totten Chapter 11 Tool Steels Elhachmi Essadiqi Chapter 12 Stainless Steel Heat Treatment Angelo Fernando Padilha, Ronald Lesley Plaut, and Paulo Rangel Rios ß 2006 by Taylor & Francis Group, LLC. Chapter 13 Heat Treatment of Powder Metallurgy Steel Components Joseph W. Newkirk and Sanjay N. Thakur Appendices Appendix 1 Common Conversion Constants Appendix 2 Temperature Conversion Table Appendix 3 Volume Conversion Table Appendix 4 Hardness Conversion Tables: Hardened Steel and Hard Alloys Appendix 5 Recommended MIL 6875 Specification Steel Heat Treatment Conditions Appendix 6 Colors of Hardening and Tempering Heats Appendix 7 Weight Tables for Steel Bars ß 2006 by Taylor & Francis Group, LLC. 1 Steel Nomenclature Anil Kumar Sinha, Chengjian Wu, and Guoquan Liu CONTENTS 1.1 1.2 1.3 Introduction .................................................................................................................. 2 Effects of Alloying Elements......................................................................................... 2 1.2.1 Carbon ........................................................................................................... 3 1.2.2 Manganese ..................................................................................................... 3 1.2.3 Silicon ............................................................................................................ 4 1.2.4 Phosphorus .................................................................................................... 4 1.2.5 Sulfur ............................................................................................................. 4 1.2.6 Aluminum ...................................................................................................... 5 1.2.7 Nitrogen ......................................................................................................... 5 1.2.8 Chromium ...................................................................................................... 5 1.2.9 Nickel ............................................................................................................. 5 1.2.10 Molybdenum.................................................................................................. 5 1.2.11 Tungsten ........................................................................................................ 6 1.2.12 Vanadium....................................................................................................... 6 1.2.13 Niobium and Tantalum ................................................................................. 6 1.2.14 Titanium......................................................................................................... 6 1.2.15 Rare Earth Metals ......................................................................................... 7 1.2.16 Cobalt ............................................................................................................ 7 1.2.17 Copper ........................................................................................................... 7 1.2.18 Boron ............................................................................................................. 7 1.2.19 Zirconium....................................................................................................... 8 1.2.20 Lead ............................................................................................................... 8 1.2.21 Tin.................................................................................................................. 8 1.2.22 Antimony ....................................................................................................... 8 1.2.23 Calcium .......................................................................................................... 8 Classification of Steels .................................................................................................. 8 1.3.1 Types of Steels Based on Deoxidation Practice ............................................... 9 1.3.1.1 Killed Steels ....................................................................................... 9 1.3.1.2 Semikilled Steels............................................................................... 10 1.3.1.3 Rimmed Steels ................................................................................. 10 1.3.1.4 Capped Steels................................................................................... 11 1.3.2 Quality Descriptors and Classifications ......................................................... 11 1.3.3 Classification of Steel Based on Chemical Composition ............................... 13 1.3.3.1 Carbon and Carbon–Manganese Steels ........................................... 13 1.3.3.2 Low-Alloy Steels.............................................................................. 17 1.3.3.3 High-Strength Low-Alloy Steels ...................................................... 24 1.3.3.4 Tool Steels ....................................................................................... 27 1.3.3.5 Stainless Steels ................................................................................. 33 ß 2006 by Taylor & Francis Group, LLC. 1.3.3.6 Maraging Steels ............................................................................... Designations for Steels................................................................................................ 1.4.1 SAE-AISI Designations ................................................................................. 1.4.1.1 Carbon and Alloy Steels .................................................................. 1.4.1.2 HSLA Steels..................................................................................... 1.4.1.3 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 ........................................................................................... 1.6.1.1 The Designation for Steels with Yield Strength............................... 1.6.1.2 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 ........................................................................................................................... 1.4 1.1 44 45 46 46 47 47 47 50 50 51 51 54 66 66 66 66 66 84 85 86 86 86 86 87 INTRODUCTION According to the iron–carbon phase diagram [1–3], all binary Fe–C alloys containing less than about 2.11 wt% carbon* are classified as steels, and all those containing higher carbon content are termed cast iron. When alloying elements are added to obtain the desired properties, the carbon 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 Earth’s crust, (2) the high melting temperature of iron (15348C), (3) a range of mechanical properties, such as moderate (200–300 MPa) yield strength with excellent ductility to in excess of 1400 MPa yield stress with fracture toughness up to 100 MPa mÀ2 , 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 characteristics of steels, reviews the various systems used to classify steels, and provides extensive tabular data relating to the designation of steels. 1.2 EFFECTS OF ALLOYING ELEMENTS Steels contain alloying elements and impurities that must be associated with austenite, ferrite, and cementite. The combined effects of alloying elements and heat treatment produce an enormous 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 is calculated thermodynamically as 2.14 wt% [3]. ß 2006 by Taylor & Francis Group, LLC. would be difficult to include a detailed survey of the effects of alloying elements on the iron– carbon equilibrium diagram, allotropic transformations, and forming of new phases. This complicated subject, which lies in the domain of ferrous physical metallurgy, has been reviewed extensively in Chapter 2 of this handbook and elsewhere in the literature [4,5,8–12]. In this section, the effects of various elements on steelmaking (deoxidation) practices and steel characteristics will be briefly outlined. It should be noted that the effects of a single alloying element on either practice or characteristics is modified by the influence of other elements. 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 ferrite stabilizers) Alloying elements can be divided into two categories according to the interaction with carbon in steel: 1. Carbide-forming elements, such as Mn, Cr, Mo, W, V, Nb, Ti, and Zr. They go into solid solution in cementite at low concentrations. At higher concentrations, they form more 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 from carbide in steels, and normally found in the matrix [5,11,12]. To simplify the discussion, the effects of various alloying elements listed below are summarized separately. 1.2.1 CARBON The 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 the approximate range of 0.15–0.30% C may have poorer surface quality and require special processing to attain surface quality comparable to steels with higher or lower C contents. Carbon has a moderate tendency for macrosegregation during solidification, and it is often more significant than that of any other alloying elements. Carbon has a strong tendency to 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 main hardening element in all steels except the austenitic precipitation hardening (PH) stainless steels, managing steels, and interstitial-free (IF) steels. The strengthening effect of C in steels consists of solid solution strengthening and carbide dispersion strengthening. As the C content in steel increases, strength increases, but ductility and weldability decrease [4,5]. 1.2.2 MANGANESE Manganese (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 readily rimmed. Manganese is beneficial to surface quality in all carbon ranges (with the exception of extremely low-carbon rimmed steels) and reduction in the risk of red-shortness. Manganese favorably affects forgeability and weldability. ß 2006 by Taylor & Francis Group, LLC. Manganese is a weak carbide former, only dissolving in cementite, and forms alloying cementite 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 distortion during quenching [4,5,15]. The presence of alloying element Mn in steels enhances the impurities such as P, Sn, Sb, and As segregating to grain boundaries and induces temper embrittlement [5]. 1.2.3 SILICON Silicon (Si) is one of the principal deoxidizers used in steelmaking; therefore, silicon content also determines the type of steel produced. Killed carbon steels may contain Si up to a maximum 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 its strength without greatly decreasing ductility. Beyond 0.40% Si, a marked decrease in ductility is noticed in plain carbon steels [4]. If combined with Mn or Mo, silicon may produce greater hardenability of steels [5]. Due to the addition of Si, stress corrosion can be eliminated in Cr–Ni 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 a noncarbide former, and free from cementite or carbides; it dissolves in martensite and retards the decomposition of alloying martensite up to 3008C. 1.2.4 PHOSPHORUS Phosphorus (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 causes the temper embrittlement of alloying steels, especially in Mn, Cr, Mn–Si, Cr–Ni, and Cr–Mn steels. Phosphorus also increases the hardenability and retards the decomposition of martensite-like Si in steels [5]. High P content is often specified in low-carbon free-machining steels to improve machinability. In low-alloy structural steels containing ~0.1% C, P increases strength and atmospheric corrosion resistance. In austenitic Cr–Ni steels, the addition of P can cause precipitation effects and an increase in yield points [15]. In strong oxidizing agent, P causes grain boundary corrosion in austenitic stainless steels after solid solution treatment as a result of the segregation of P at grain boundaries [5]. 1.2.5 SULFUR Increased amounts of sulfur (S) can cause red- or hot-shortness due to the low-melting sulfide eutectics surrounding the grain in reticular fashion [15,16]. Sulfur has a detrimental effect on transverse ductility, notch impact toughness, weldability, and surface quality (particularly in the lower carbon and lower manganese steels), but has a slight effect on longitudinal mechanical properties. Sulfur has a very strong tendency to segregate at grain boundaries and causes reduction of hot ductility in alloy steels. However, sulfur in the range of 0.08–0.33% is intentionally added to free-machining steels for increased machinability [5,17] . Sulfur improves the fatigue life of bearing steels [18], because (1) the thermal coefficient on MnS inclusion is higher than that of matrix, but the thermal coefficient of oxide inclusions is lower than that of matrix, (2) MnS inclusions coat or cover oxides (such as ß 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 ALUMINUM Aluminum (Al) is widely used as a deoxidizer and a grain refiner [9]. As Al forms very hard nitrides with nitrogen, it is usually an alloying element in nitriding steels. It increases scaling resistance 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 the corrosion resistance in low-carbon corrosion-resisting steels. Of all the alloying elements, Al is 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 NITROGEN Nitrogen (N) is one of the important elements in expanded g-field group. It can expand and stabilize 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, fine nitrides 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 CHROMIUM Chromium (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 enhances abrasion resistance in high-carbon grades. Chromium carbides are hard and wear-resistant and increase the edge-holding quality. Complex chromium–iron carbides slowly go into solution in austenite; therefore, a longer time at temperature is necessary to allow solution to take place before quenching is accomplished [5,6,14]. Chromium is the most important alloying element in steels. The addition of Cr in steels enhances the impurities, such as P, Sn, Sb, and As, segregating to grain boundaries and induces temper embrittlement. 1.2.9 NICKEL Nickel (Ni) is a noncarbide-forming element in steels. As a result of the open g-phase field, Ni is 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 in steels [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 Cr–Ni austenitic stainless steels in nonoxidizing acid medium. 1.2.10 MOLYBDENUM Molybdenum (Mo) is a pronounced carbide former. It dissolves slightly in cementite, while molybdenum carbides will form when the Mo content in steel is high enough. Molybdenum can induce secondary hardening during the tempering of quenched steels and improves the creep strength of low-alloy steels at elevated temperatures. ß 2006 by Taylor & Francis Group, LLC. The addition of Mo produces fine-grained steels, increases hardenability, and improves fatigue strength. Alloy steels containing 0.20–0.40% Mo or V display a delayed temper embrittlement, but cannot eliminate it. Molybdenum increases corrosion resistance and is used to a great extent in high-alloy Cr ferritic stainless steels and with Cr–Ni austenitic stainless steels. High Mo contents reduce the stainless steel’s susceptibility to pitting [5,15]. Molybdenum has a very strong solid solution strengthening in austenitic alloys at elevated temperatures. Molybdenum is a very important alloying element for alloy steels. 1.2.11 TUNGSTEN Tungsten (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 forms very hard, abrasion-resistant carbides, and can induce secondary hardening during the tempering of quenched steels. It promotes hot strength and red-hardness and thus cutting ability. It prevents grain growth at high temperature. W and Mo are the main alloying elements in high-speed steels [5,13]. However, W and Mo impair scaling resistance. 1.2.12 VANADIUM Vanadium (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 vanadium carbides decrease hardenability [5]. Vanadium is a grain refiner, and imparts strength and toughness. Fine vanadium carbides and nitrides give a strong dispersion hardening effect in microalloyed steels after controlled rolling and controlled cooling. Vanadium provides a very strong secondary hardening effect on tempering, therefore it raises hot-hardness and thus cutting ability in high-speed steels. Vanadium increases fatigue strength and improves notch sensitivity. 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, and creep-resistant steels. It promotes the weldability of heat-treatable steels. The presence of V retards the rate of tempering embrittlement in Mo-bearing steels. 1.2.13 NIOBIUM AND TANTALUM Niobium (Nb) and tantalum (Ta) are very strong carbide and nitride formers. Small amounts of Nb can form fine nitrides or carbonitrides and refine the grains, therefore increasing the yield strength of steels. Niobium is widely used in microalloying steels to obtain high strength and good toughness through controlled rolling and controlled cooling practices. A 0.03% Nb in austenite can increase the yield strength of medium-carbon steel by 150 MPa. Niobiumcontaining nonquenched and tempered steels, including microalloyed medium-carbon steels and low-carbon bainite (martensite) steels, offer a greatly improved combination of strength and toughness. Niobium is a stabilizer in Cr–Ni austenitic steels to eliminate intergranular corrosion. 1.2.14 TITANIUM Titanium (Ti) is a very strong carbide and nitride former. The effects of Ti are similar to those of Nb and V, but titanium carbides and nitrides are more stable than those of Nb and V. It is widely used in austenitic stainless steels as a carbide former for stabilization to eliminate intergranular corrosion. By the addition of Ti, intermetallic compounds are formed in maraging steels, causing age hardening. Titanium increases creep rupture strength through formation of special nitrides and tends significantly to segregation and banding [15]. ß 2006 by Taylor & Francis Group, LLC. Ti, Nb, and V are effective grain inhibitors because their nitrides and carbides are quite stable and difficult to dissolve in austenite. If Ti, Nb, and V dissolve in austenite, the hardenability 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 METALS Rare earth metals (REMs) constitute the IIIB group of 17 elements in the periodic table. They are scandium (Sc) of the fourth period, yttrium (Y) of the fifth period, and the lanthanides of the sixth period, 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 and are difficult to separate in ore beneficiation and metal extraction so they are usually supplied as a mixture 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 REM on shape control and modification of inclusions would improve transversal plasticity and toughness, hot ductility, fatigue strength, and machinability. REMs tend strongly to segregate at the grain boundaries and increase the hardenability of steels [21,23]. 1.2.16 COBALT Cobalt (Co) is a noncarbide former in steels. It decreases hardenability of carbon steels, but by addition of Cr, it increases hardenability of Cr–Mo alloy steels. Cobalt raises the martensitic transformation temperature of Ms (8C) and decreases the amount of retained austenite in alloy steels. Cobalt promotes the precipitation hardening [5]. It inhibits grain growth at high temperature and significantly improves the retention of temper and high-temperature strength, resulting in an increase in tool life. The use of Co is generally restricted to high-speed steels, hot-forming tool steels, maraging steels, and creep-resistant and high-temperature materials [13,15]. 1.2.17 COPPER Copper (Cu) addition has a moderate tendency to segregate. Above 0.30% Cu can cause precipitation 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 exaggerates the surface defects inherent in resulfurized steels. However, Cu improves the atmospheric corrosion resistance (when in excess of 0.20%) and the tensile properties in alloy and low-alloy steels, and reportedly helps the adhesion of paint [6,14]. In austenitic stainless steels, a Cu content above 1% results in improved resistance to H2 SO4 and HCl and stress corrosion [5,15]. 1.2.18 BORON Boron (B), in very small amounts (0.0005–0.0035%), has a starting effect on the hardenability of steels due to the strong tendency to segregate at grain boundaries. The segregation of B in steels is a nonequilibrium segregation. It also improves the hardenability of other alloying elements. 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 no real increase in hardenability above 0.6% C [14]. The weldability of boron-alloyed steels is another reason for their use. However, large amounts of B result in brittle, unworkable steels. ß 2006 by Taylor & Francis Group, LLC. 1.2.19 ZIRCONIUM Zirconium (Zr) is added to killed HSLA steels to obtain improvement in inclusion characteristics, particularly sulfide inclusions, where modifications of inclusion shape improve ductility in transverse bending. It increases the life of heat-conducting materials. It is also a strong carbide former and produces a contracted austenite phase field [5,15]. 1.2.20 LEAD Lead (Pb) is sometimes added (in the range of 0.2–0.5%) to carbon and alloy steels through mechanical dispersion during teeming to improve machinability. 1.2.21 TIN Tin (Sn) in relatively small amounts is harmful to steels. It has a very strong tendency to segregate at grain boundaries and induces temper embrittlement in alloy steels. It has a detrimental effect on the surface quality of continuous cast billets containing small amounts of Cu [24]. Small amounts of Sn and Cu also decrease the hot ductility of steels in the austenite þ ferrite region [25]. 1.2.22 ANTIMONY Antimony (Sb) has a strong tendency to segregate during the freezing process, and has a detrimental effect on the surface quality of continuous cast billets. It also has a very strong tendency to segregate at grain boundaries and cause temper embrittlement in alloy steels. 1.2.23 CALCIUM Calcium (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 STEELS Steels can be classified by different systems depending on [4,6,8]: 1. Compositions, such as carbon (or nonalloy), low-alloy, and alloy steels 2. Manufacturing methods, such as converter, electric furnace, or electroslag remelting methods 3. Application or main characteristic, such as structural, tool, stainless steel, or heatresistant steels 4. Finishing methods, such as hot rolling, cold rolling, casting, or controlled rolling and controlled cooling 5. Product shape, such as bar, plate, strip, tubing, or structural shape 6. Oxidation practice employed, such as rimmed, killed, semikilled, and capped steels 7. 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) standards 9. Heat treatment, such as annealing, quenching and tempering, air cooling (normalization), and thermomechanical processing 10. Quality descriptors and classifications, such as forging quality and commercial quality Among the above classification systems, chemical composition is the most widely used basis for designation and is given due emphasis in this chapter. Classification systems based on oxidation practice, application, and quality descriptors are also briefly discussed. ß 2006 by Taylor & Francis Group, LLC. Ferrous alloys Classification by commercial name or application Classification by structure Steel Alloys without eutectic (<2% C on Fe−C diagram) Plain carbon steel Low-carbon steel (<0.2% C) Ferritic Medium-carbon steel (0.2−5% C) Ferritic−pearlitic High-carbon steel (>0.5% C) Pearlitic Low- and mediumalloy steel ≤10% alloying elements Martensitic Bainitic High-alloy steel >10% alloying elements Corrosion resistant Heat resistant Wear resistant Austenitic Precipitation hardened Austenitic− ferritic Duplex structure FIGURE 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 PRACTICE Steels, when cast into ingots, can be classified into four types according to the deoxidation practice or, alternatively, by the amount of gas evolved during solidification. These four types are called killed, semikilled, capped, and rimmed steels [6,8]. 1.3.1.1 Killed Steels Killed 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 in the upper central portion of the ingot, which is later cut off and discarded. All alloy steels, most ß 2006 by Taylor & Francis Group, LLC. low-alloy steels, and many carbon steels are usually killed. The continuous casting billets are also killed. The essential quality criterion is soundness [26–28]. Killed steel is characterized by a homogeneous 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. 1.3.1.2 Semikilled Steels Gas 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 killed steel, but less than in capped or rimmed steel. An ingot skin of considerable thickness is formed before the beginning of gas evolution. A correctly deoxidized semikilled steel ingot does not have a pipe but does have well-scattered large blow holes in the top-center half of the ingot; however, the blow holes weld shut during rolling of the ingot. Semikilled steels generally have a carbon content in the range of 0.15–0.30%. They find a wide range of uses in structural shapes, skelp, and pipe applications. The main features of semikilled steels are UJ variable degrees of uniformity in composition, which are intermediate between those of killed and rimmed steels and less segregation than rimmed steel, and (2) a pronounced tendency for positive chemical segregation at the top center of the ingot (Figure 1.2). 1.3.1.3 Rimmed Steels Rimmed steel is characterized by a great degree of gas evolution during solidification in the mold and a marked difference in chemical composition across the section and from the top to the bottom of the ingot (Figure 1.2). These result in the formation of an outer ingot skin or rim of relatively pure iron and an inner liquid (core) portion of the ingot with higher concentrations of alloying and residual elements, especially C, N, S, and P, having lower melting 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, and the steel does not retain any significant amount of highly oxidizable elements such as Al, Si, or Ti. Rimmed steels are cheaper than killed or semikilled steels for only a small addition of deoxidizer is required and is formed without top scrap. 1 2 Killed Semikilled 3 4 5 Capped 6 7 8 Rimmed FIGURE 1.2 Eight typical conditions of commercial steel ingots, cast in identical bottle-top molds, in relation to the degree of suppression of gas evolution. The dotted line denotes the height to which the steel originally was poured in each ingot mold. Based on the carbon, and more significantly, the oxygen content of the steel, the ingot structures range from that of a completely killed ingot (No. 1) to that of a violently 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.) ß 2006 by Taylor & Francis Group, LLC. 1.3.1.4 Capped Steels Capped steel is a type of steel with characteristics similar to those of a rimmed steel but to a degree intermediate between that of rimmed and semikilled steels. Less deoxidizer is used to produce a capped ingot than to produce a semikilled ingot [29]. This induces a controlled rimming action when the ingot is cast. The gas entrapped during solidification is excess of that required 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 time of gas evolution and prevents the formation of an excessive number of gas voids within the ingot. The capped ingot process is usually applied to steels with carbon contents greater than 0.15% 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 to seal the top of the ingot and to stop the rimming action [29]. Chemically capped steel is cast in open-top molds. The capping is accomplished by the addition of Al or ferrosilicon to the top of the ingot, causing the steel at the top surface to solidify rapidly. The top portion of the ingot is cropped and discarded. 1.3.2 QUALITY DESCRIPTORS AND CLASSIFICATIONS Quality descriptors are names applied to various steel products to indicate that a particular product possesses certain characteristics that make it especially well suited for specific applications or fabrication processes. The quality designations and descriptors for various carbon steel products and alloy steel plates are listed in Table 1.1. Forging quality and cold extrusion quality descriptors 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 requiring modest 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 bars are produced from special quality hot-rolled carbon steel bars. Alloy steel plate qualities are described by structural, drawing, cold working, pressure vessel, and aircraft qualities [27]. The various physical and mechanical characteristics indicated by a quality descriptor result from the combined effects of several factors such as (1) the degree of internal soundness, (2) the relative uniformity of chemical composition, (3) the number, size, and distribution of nonmetallic inclusions, (4) the relative freedom from harmful surface imperfections, (5) extensive testing 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 mill products with the desired characteristics. The degree of control over these and other related factors 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 more additional requirements as may be appropriate, namely macroetch test, special discard, restricted chemical composition, maximum incidental (residual) alloying elements, austenitic grain size, and special hardenability. These limitations could be applied forging quality alloy steel bars but not to merchant quality bars. Understanding the various quality descriptors is difficult because most of the prerequisites for qualifying steel for a specific descriptor are subjective. Only limitations on chemical composition ranges, residual alloying elements, nonmetallic inclusion count, austenitic grain size, and special hardenability are quantifiable. The subjective evaluation of the other attributes depends on the experience and the skill of the individuals who make the evaluation. Although the use of these subjective quality descriptors might appear impractical and imprecise, steel products made to meet the requirements of a specific quality descriptor can be relied upon to have those characteristics necessary for that product to be used in the suggested application or fabrication operation [6]. ß 2006 by Taylor & Francis Group, LLC. TABLE 1.1 Quality Descriptionsa of Carbon and Alloy Steels Carbon Steels Semifinished for forging Forging quality Special hardenability Special internal soundness Nonmetallic inclusion requirement Special surface Carbon steel structural sections Structural quality Carbon steel plates Regular quality Structural quality Cold-drawing quality Cold-pressing quality Cold-flanging quality Forging quality Pressure vessel quality Hot-rolled carbon steel bars Merchant quality Special quality Special hardenability Special internal soundness Nonmetallic inclusion requirement Special surface Scrapless nut quality Axle shaft quality Cold extrusion quality Cold-heading and coldforging quality Cold-finished carbon steel bars Standard quality Special hardenability Special internal soundness Nonmetallic inclusion requirement Special surface Cold-heading and coldforging quality Cold extrusion quality Hot-rolled sheets Commercial quality Drawing quality Drawing quality special killed Structural quality Cold-rolled sheets Commercial quality Drawing quality Drawing quality special killed Structural quality Porcelain enameling sheets Commercial quality Drawing quality Drawing quality special killed Long terne sheets Commercial quality Drawing quality Drawing quality special killed Structural quality Galvanized sheets Commercial quality Drawing quality Drawing quality special killed Lock-forming quality Electrolytic zinc coated sheets Commercial quality Drawing quality Drawing quality special killed Structural quality Hot-rolled strip Commercial quality Drawing quality Drawing quality special killed Structural quality Cold-rolled strip Specific quality descriptions are not ß 2006 by Taylor & Francis Group, LLC. Alloy Steels Mill products Specific quality descriptions are not applicable to tin mill products Alloy steel plates Drawing quality Pressure vessel quality Structural quality Aircraft physical quality Carbon steel wire Industrial quality wire Cold extrusion wires Heading, forging, and roll-threading wires Mechanical spring wires Upholstery spring construction wires Welding wire Hot-rolled alloy steel bars Regular quality Aircraft quality or steel subject to magnetic particle inspection Axle shaft quality Bearing quality Cold-heading quality Special cold-heading quality Rifle barrel quality, gun quality, shell or A.P. shot quality Carbon steel flut wire Stitching wire Stapling wire Carbon steel pipe Structural tubing Line pipe Oil country tubular goods Steel specialty tubular products Pressure tubing Mechanical tubing Aircraft tubing Hot-rolled carbon steel wire rods Industrial quality Rods for manufacture of wire intended for electric welded chain Rods for heading, forging, and rollthreading wire Rods for lock washer wire Rods for scrapless nut wire Rods for upholstery spring wire Rods for welding wire Alloy steel wire Aircraft quality Bearing quality Special surface quality Cold-finished alloy steel bars Regular quality Aircraft quality or steel subject to magnetic particle inspection Axle shaft quality Bearing shaft quality Cold-heading quality Special cold-heading quality Rifle barrel quality, gun quality, shell or A.P. shot quality Line pipe Oil country tubular goods Steel specialty tubular goods Pressure tubing Mechanical tubing Stainless and heatresisting pipe, pressure TABLE 1.1 (Continued ) Quality Descriptionsa of Carbon and Alloy Steels Carbon Steels provided in coldrolled strip because this product is largely produced for specific and use Alloy Steels tubing, and mechanical tubing Aircraft tubing pipe a In the case of certain qualities, P and S are usually finished to lower limits than the specified maximum. Source: From H. Okamoto, C–Fe, in Binary Alloy Phase Diagrams, 2nd ed., T.B. Massalski, Ed., ASM International, Materials Park, OH, 1990, pp. 842–848. 1.3.3 CLASSIFICATION OF STEEL BASED ON CHEMICAL COMPOSITION 1.3.3.1 Carbon and Carbon–Manganese Steels In addition to carbon, plain carbon steels contain the following other elements: Mn up to 1.65%, S up to 0.05%, P up to 0.04%, Si up to 0.60%, and Cu up to 0.60%. The effects of each of 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 and properties of the steel (see Section 1.2). However, variations in C content have the greatest effect on 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 contain up 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 various grades of standard carbon and low-alloy steels with the Society of Automotive Engineers and American Iron and Steel Institute (SAE-AISI) designations. Table 1.3 shows some representative standard carbon steel compositions with SAE-AISI and the corresponding Unified Numbering System (UNS) designations [6,8,30]. Low-carbon steels contain up to 0.25% C. The largest category of this class is flat-rolled products (sheet or strip), usually in the cold-rolled or subcritical annealed condition and usually with final temper-rolling treatment. The carbon content for high formability and high drawability steels is very low (<0.10% C) with up to 0.40% Mn. These lower carbon steels are used in automobile body panels, tin plates, appliances, and wire products. The low-carbon steels (0.10–0.25% C) have increased strength and hardness and reduced formability compared to the lowest carbon group. They are designated as carburizing or casehardening steels [9]. Selection of these grades for carburizing applications depends on the nature of the part, the properties required, and the processing practices preferred. An increase of 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-resistant surface but without any need for increased mechanical properties in the core, e.g., small shafts, plunges, or highly loaded gearing [8]. Rolled structural steels in the form of plates and ß 2006 by Taylor & Francis Group, LLC. 14 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.2 SAE-AISI Designation System for Carbon and Low-Alloy Steels Numerals and Digits Type of Steel and Nominal Alloy Content (%) Carbon steels 10xxa . . . . . . . . . Plain carbon (Mn 1.00 max) 11xx . . . . . . . . . Resulfurized 12xx . . . . . . . . . Resulfurized and rephosphorized 15xx . . . . . . . . . Plain carbon (max Mn range: 1.00–1.65) Manganese steels 13xx . . . . . . . . . Mn 1.75 Nickel steels 23xx . . . . . . . . . Ni 3.50 25xx . . . . . . . . . Ni 5.00 Molybdenum steels 40xx . . . . . . . . . Mo 0.20 and 0.25 44xx . . . . . . . . . Mo 0.40 and 0.52 Chromium–molybdenum steels 41xx . . . . . . . . . Cr 0.50, 0.80, and 0.95; Mo 0.12, 0.20, 0.25, and 0.30 a Type of Steel and Nominal Alloy Content (%) Nickel–chromium–molybdenum steels 43xx . . . . . . . . . Ni 1.82; Cr 0.50 and 0.80; Mo 0.25 43BVxx Ni 1.82; Cr 0.50; Mo 0.12 and 0.25; V 0.03 min 47xx . . . . . . . . . Ni 1.05; Cr 0.45; Mo 0.20 and 0.35 81xx . . . . . . . . . Ni 0.30; Cr 0.40; Mo 0.12 86xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.20 87xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.25 88xx . . . . . . . . . Ni 0.55; Cr 0.50; Mo 0.35 93xx . . . . . . . . . Ni 3.25; Cr 1.20; Mo 0.12 94xx . . . . . . . . . Ni 0.45; Cr 0.40; Mo 0.12 97xx . . . . . . . . . Ni 0.55; Cr 0.20; Mo 0.20 98xx . . . . . . . . . Ni 1.00; Cr 0.80; Mo 0.25 Nickel–molybdenum steels 46xx . . . . . . . . . Ni 0.85 and 1.82; Mo 0.20 and 0.25 48xx . . . . . . . . . Ni 3.50; Mo 0.25 Chromium steels 50xx . . . . . . . . . Cr 0.27, 0.40, 0.50, and 0.65 51xx . . . . . . . . . Cr 0.80, 0.87, 0.92, 0.95, 1.00, and 1.05 Numerals and Digits Type of Steel and Nominal Alloy Content (%) Chromium (bearing) steels ) 50xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 0.50 51xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 1.02 min C 1.00 52xxx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 1.45 Chromium–vanadium steels 61xx . . . . . . . . . . . . . . . . . . . . . . . . . . . Cr 0.60, 0.80, and 0.95; V 0.10 and 0.15 min Tungsten–chromium steel 72xx . . . . . . . . . . . . . . . . . . . . . . . . . . . W 1.75; Cr 0.75 Silicon–manganese steels 92xx . . . . . . . . . . . . . . . . . . . . . . . . . . . Si 1.40 and 2.00; Mn 0.65, 0.82, and 0.85; High-strength low-alloy steels Cr 0 and 0.65 9xx Boron steels . . . . . . . . . . . . . . . . . . . . . . . . . . . Various SAE grades xxBxx Leaded steels . . . . . . . . . . . . . . . . . . . . . . . . . . . B denotes boron steel xxLxx . . . . . . . . . . . . . . . . . . . . . . . . . . . L denotes leaded steel The 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 Technologies Nickel–chromium steels 31xx . . . . . . . . . Ni 1.25; Cr 0.65 and 0.80 32xx . . . . . . . . . Ni 1.75; Cr 1.07 33xx . . . . . . . . . Ni 3.50; Cr 1.50 and 1.57 34xx . . . . . . . . . Ni 3.00; Cr 0.77 Numerals and Digits TABLE 1.3 Standard Carbon Steel Compositions with SAE-AISI and Corresponding UNS Designations Plain Carbon Steel (Nonresulfurized, 1.0% Mn Max)a Cast or Heat Chemical Ranges and Limits (%)a UNS Number SAE-AISI Number C Mn P max S max G10060 G10080 G10090 G10100 G10120 G10150 G10160 G10170 G10180 G10190 G10200 G10210 G10220 G10230 G10250 G10260 G10300 G10330 G10350 G10370 G10380 G10390 G10400 G10420 G10430 G10450 G10490 G10500 G10550 G10600 G10640 G10650 G10700 G10740 G10750 G10780 G10800 G10840 G10850 G10860 G10900 G10950 1006 1008 1009 1010 1012 1015 1016 1017 1018 1019 1020 1021 1022 1023 1025 1026 1030 1033 1035 1037 1038 1039 1040 1042 1043 1045 1049 1050 1055 1060 1064 1065 1070 1074 1075 1078 1080 1084 1085 1086 1090 1095 0.08 max 0.10 max 0.15 max 0.08–0.13 0.10–0.15 0.12–0.18 0.12–0.18 0.14–0.20 0.14–0.20 0.14–0.20 0.17–0.23 0.17–0.23 0.17–0.23 0.19–0.25 0.22–0.28 0.22–0.28 0.27–0.34 0.29–0.36 0.31–0.38 0.31–0.38 0.34–0.42 0.36–0.44 0.36–0.44 0.39–0.47 0.39–0.47 0.42–0.50 0.45–0.53 0.47–0.55 0.52–0.60 0.55–0.66 0.59–0.70 0.59–0.70 0.65–0.76 0.69–0.80 0.69–0.80 0.72–0.86 0.74–0.88 0.80–0.94 0.80–0.94 0.80–0.94 0.84–0.98 0.90–1.04 0.45 max 0.50 max 0.60 max 0.30–0.60 0.30–0.60 0.30–0.60 0.60–0.90 0.30–0.60 0.60–0.90 0.70–1.00 0.30–0.60 0.60–0.90 0.70–1.00 0.30–0.60 0.30–0.60 0.60–0.90 0.60–0.90 0.70–1.00 0.60–0.90 0.70–1.00 0.60–0.90 0.70–1.00 0.60–0.90 0.60–0.90 0.70–1.00 0.60–0.90 0.60–0.90 0.60–0.90 0.60–0.90 0.60–0.90 0.50–0.80 0.60–0.90 0.60–0.90 0.50–0.80 0.40–0.70 0.30–0.60 0.60–0.90 0.60–0.90 0.70–1.00 0.30–0.50 0.60–0.90 0.30–0.50 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.3 (Continued) Standard Carbon Steel Compositions with SAE-AISI and Corresponding UNS Designations Free-Cutting (Resulfurized) Carbon Steel Compositionsa Cast or Heat Chemical Ranges and Limits (%) UNS Number SAE-AISI Number C Mn P max S G11080 G11100 G11170 G11180 G11370 G11390 G11400 G11410 G11440 G11460 G11S10 1108 1110 1117 1118 1137 1139 1140 1141 1144 1146 1151 0.08–0.13 0.08–0.13 0.14–0.20 0.14–0.20 0.32–0.39 0.35–0.43 0.37–0.44 0.37–0.45 0.40–0.48 0.42–0.49 0.48–0.55 0.50–0.80 0.30–0.60 1.00–1.30 1.30–1.60 1.35–1.65 1.35–1.65 0.70–1.00 1.35–1.65 1.35–1.65 0.70–1.00 0.70–1.00 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.08–0.13 0.08–0.13 0.08–0.13 0.08–0.13 0.08–0.13 0.13–0.20 0.08–0.13 0.08–0.13 0.24–0.33 0.08–0.13 0.08–0.13 Standard Resulfurized and Rephosphorized Carbon Steelsa UNS Number Gl2110 G12120 G12130 G12150 G12144 Cast or Heat Chemical Ranges and Limits, %(a) SAE-AISI Number C max P S Pb 0.13 0.13 0.13 0.09 0.15 1211 . . . . . . . . . 1212 . . . . . . . . . 1213 . . . . . . . . . 1215 . . . . . . . . . 12L14b . . . . . . . . . Mn 0.60–0.90 0.70–1.00 0.70–1.00 0.75–1.05 0.85–1.15 0.07–0.12 0.07–0.12 0.07–0.12 0.04–0.09 0.04–0.09 0.10–0.15 0.16–0.23 0.24–0.33 0.26–0.35 0.26–0.35 — — — — 0.15–0.35 Standard Nonresulfurized Carbon Steels (Over 1.0% Manganese) UNS Number G15130 G15220 G15240 G15260 G15270 G15360 G15410 G15480 G15510 G15520 G15610 G15660 1513 1522 1524 1526 1527 1536 1541 1548 1551 1552 1561 1566 Cast or Heat Chemical Ranges and Limits, % SAE-AISI Number ... ... ... ... ... ... ... ... ... ... ... ... C ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 0.10–0.16 0.18–0.24 0.19–0.25 0.22–0.29 0.22–0.29 0.30–0.37 0.36–0.44 0.44–0.52 0.45–0.56 0.47–0.55 0.55–0.65 0.60–0.71 Mn P max S max 1.10–1.40 1.10–1.40 1.35–1.65 1.10–1.40 1.20–1.50 1.20–1.50 1.35–1.65 1.10–1.40 0.85–1.15 1.20–1.50 0.75–1.05 0.85–1.15 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 Applicable 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 effect on machinability. b Contains 0.15–0.35% 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 Automotive Engineers, Warrendale, PA, pp. 1.01–1.189. a ß 2006 by Taylor & Francis Group, LLC. sections containing ~0.25% C, with up to 1.5% Mn and Al are used if improved toughness is required. When used for stampings, forgings, seamless tubes, and boilerplate, Al addition should be avoided. An important type of this category is the low-carbon free-cutting steels containing up to 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 0.30–0.55% C and 0.60–1.65% Mn are used where higher mechanical properties are desired. They are usually hardened and strengthened by heat treatment or by cold work. Low-carbon and manganese steels in this group find wide applications for certain types of cold-formed parts that need annealing, normalizing, or quenching and tempering treatment before use. The higher carbon grades are often cold drawn 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 section size and the mechanical properties needed after heat treatment [8]. These grades, generally produced as killed steels, are used for a wide range of applications that include automobile parts 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 grain refinement and improved toughness. In general, steels containing 0.40–0.60% C are used as rails, railway wheels, tires, and axles. High-carbon steels containing 0.55–1.00% C and 0.30–0.90% Mn have more restricted applications than the medium-carbon steels because of higher production cost and poor formability (or ductility) and weldability. High-carbon steels find applications in the spring industry (as light and thicker plat springs, laminated springs, and heavier coiled springs), farm implement industry (as plow beams, plowshares, scraper blades, discs, mowers, knives, and harrow teeth), and high-strength wires where improved wear characteristics and higher strength than those attainable with lower carbon grades are needed. UHC steels are experimental plain carbon steels with 1.0–2.1% C (15–32 vol% cementite) [32–34]. Optimum superplastic elongation has been found at about 1.6% C content [9]. These steels have the capability of emerging as important technological materials because they exhibit superplasticity. The superplastic behavior of these materials is attributed to the structure 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 -2 readily achieved by any of the four thermomechanical treatment routes described elsewhere [4]. 1.3.3.2 Low-Alloy Steels Alloy steels may be defined as those steels that owe their improved properties to the presence of one or more special elements or to the presence of large proportions of elements such as Mn and Si than are ordinarily present in carbon steels [26]. Alloy steels contain Mn, Si, or Cu in quantities greater than the maximum limits (e.g., 1.65% Mn, 0.60% Si, and 0.60% Cu) of carbon 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 is only greater than 1.00% Mn, 0.50% Si, or 0.10% Cu [7]. The alloying elements increase the mechanical and fabrication properties. Broadly, alloy steels can be divided into (1) low-alloy steels containing less than 5 wt% total noncarbon alloy addition, (2) medium-alloy steels containing 5–10 wt% total noncarbon alloy addition, and (3) high-alloy steels with more than 10 wt% total noncarbon alloy addition. Table 1.4 lists some low-alloy steel compositions with SAE-AISI and corresponding UNS designations. Low-alloy steels constitute a group of steels that exhibit superior mechanical properties compared 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 ß 2006 by Taylor & Francis Group, LLC. 18 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.4 Low-Alloy Steel Compositions Applicable to Billets, Blooms, Slabs, and Hot-Rolled and Cold-Finished Bars (Slightly Wider Ranges of Compositions Apply to Plates) Ladle Chemical Composition Limits (%)a UNS Number SAE Number Corresponding AISI Number C Mn P G13300 G13350 G13400 G13450 1330 1335 1340 1345 1330 1335 1340 1345 0.28–0.33 0.33–0.38 0.38–0.43 0.43–0.48 1.60–1.90 1.60–1.90 1.60–1.90 1.60–1.90 0.035 0.035 0.035 0.035 G40230 G40240 G40270 G40280 G40320 G40370 G40420 G40470 4023 4024 4027 4028 4032 4037 4042 4047 4023 4024 4027 4028 — 4037 — 4047 0.20–0.25 0.20–0.25 0.25–0.30 0.25–0.30 0.30–0.35 0.35–0.40 0.40–0.45 0.45–0.50 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 G41180 G41300 G41350 G41370 G41400 4118 4130 4135 4137 4140 4118 4130 — 4137 4140 0.18–0.23 0.28–0.33 0.33–0.38 0.35–0.40 0.38–0.43 G41420 G41450 G41470 G41500 G41610 4142 4145 4147 4150 4161 4142 4145 4147 4150 4161 G43200 G43400 G43406 G44220 G44270 4320 4340 E4340b 4422 4427 4320 4340 E4340 — — S Ni Cr Mo V 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — — — — — — — — — — — — 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.035–0.050 0.040 0.035–0.050 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — — — — — — — — — — — — 0.20–0.30 0.20–0.30 0.20–0.30 0.20–0.30 0.20–0.30 0.20–0.30 0.20–0.30 — — — — — — — 0.70–0.90 0.40–0.60 0.70–0.90 0.70–0.90 0.75–1.00 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — 0.40–0.60 0.80–1.10 0.80–1.10 0.80–1.10 0.80–1.10 0.08–0.15 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 — — — — — 0.40–0.45 0.41–0.48 0.45–0.50 0.48–0.53 0.56–0.64 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — 0.80–1.10 0.80–1.10 0.80–1.10 0.80–1.10 0.70–0.90 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.25–0.35 — — — — — 0.17–0.22 0.38–0.43 0.38–0.43 0.20–0.25 0.24–0.29 0.45–0.65 0.60–0.80 0.65–0.85 0.70–0.90 0.70–0.90 0.035 0.035 0.025 0.035 0.035 0.040 0.040 0.025 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 1.65–2.00 1.65–2.00 1.65–2.00 — — 0.40–0.60 0.70–0.90 0.70–0.90 — — 0.20–0.30 0.20–0.30 0.20–0.30 0.35–0.45 0.35–0.45 — — — — — Steel Heat Treatment: Metallurgy and Technologies Si 4615 4617 4620 4626 4615 — 4620 4626 0.13–0.18 0.15–0.20 0.17–0.22 0.24–0.29 0.45–0.65 0.45–0.65 0.45–0.65 0.45–0.65 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 max 0.15–0.25 0.15–0.35 0.15–0.35 0.15–0.35 1.65–2.00 1.65–2.00 1.65–2.00 0.70–1.00 — — — — 0.20–0.30 0.20–0.30 0.20–0.30 0.15–0.25 — — — — G47180 G47200 G48150 G48170 G48200 4718 4720 4815 4817 4820 4718 4720 4815 4817 4820 0.16–0.21 0.17–0.22 0.13–0.18 0.15–0.20 0.18–0.23 0.70–0.90 0.50–0.70 0.40–0.60 0.40–0.60 0.50–0.70 — 0.035 0.035 0.035 0.035 — 0.040 0.040 0.040 0.040 — 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.90–1.20 0.90–1.20 3.25–3.75 3.25–3.75 3.23–3.75 0.35–0.55 0.35–0.55 — — — 0.30–0.40 0.15–0.25 0.20–0.30 0.20–0.30 0.20–0.30 — — — — — G50401 G50441 G50460 G50461 G50501 G50600 G50601 50B40c 50B44c 5O46 50B46c 50B50c 5060 50B60c — 50B44 — 50B46 50B50 — 50B60 0.38–0.43 0.43–0.48 0.43–0.48 0.44–0.49 0.48–0.53 0.56–0.64 0.56–0.64 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — — — 0.40–0.60 0.40–0.60 0.20–0.35 0.20–0.35 0.40–0.60 0.40–0.60 0.40–0.60 — — — — — — — — — — — — — — G51150 G51170 G51200 G51300 G51320 G51350 G51400 G51470 G51500 G51550 G51600 G51601 5115 5117 5120 5130 5132 5135 5140 5147 5150 5155 5160 51B60c — 5117 5120 5130 5132 5135 5140 5147 5150 5155 5160 51B60 0.13–0.18 0.15–0.20 0.17–0.22 0.28–0.33 0.30–0.35 0.33–0.38 0.38–0.43 0.46–0.51 0.48–0.53 0.51–0.59 0.56–0.64 0.56–0.64 0.70–0.90 0.70–1.90 0.70–0.90 0.70–0.90 0.60–0.80 0.60–0.80 0.70–0.90 0.70–0.95 0.70–0.90 0.70–0.90 0.75–1.00 0.75–1.00 0.035 0.040 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–3.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — — — — — — — — — 0.70–0.90 0.70–0.90 0.70–0.90 0.80–1.10 0.75–1.00 0.80–1.05 0.70–0.90 0.85–1.15 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 — — — — — — — — — — — — — — — — — — — — — — — — G50986 G51986 G52986 50100b 51100b 52100b — E51100 E52100 0.98–1.10 0.98–1.10 0.98–1.10 0.25–0.45 0.25–0.45 0.25–0.45 0.025 0.025 0.025 0.025 0.025 0.025 0.15–0.35 0.15–0.35 0.15–0.35 — — — 0.40–0.60 0.90–1.15 1.30–1.60 — — — Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. G46150 G46170 G46200 G46260 — — — 19 Continued 20 ß 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 Ranges of Compositions Apply to Plates) Ladle Chemical Composition Limits (%)a SAE Number Corresponding AISI Number C Mn P S Si Ni Cr Mo V G61180 G61500 6118 6150 6118 6150 0.16–0.21 0.48–0.53 0.50–0.70 0.70–0.90 0.035 0.035 0.040 0.040 0.15–0.35 0.15–0.35 — — 0.50–0.70 0.80–1.10 — — 0.10–0.15 0.15 min G81150 G81451 G86150 G86170 G86200 G86220 G86250 G86270 G86300 G86370 G86400 8115 81B45c 8615 8617 8620 8622 8625 8627 8630 8637 8640 8115 81B45 8615 8617 8620 8622 8625 8627 8630 8637 8640 0.13–0.18 0.43–0.48 0.13–0.18 0.15–0.20 0.18–0.23 0.20–0.25 0.23–0.28 0.25–0.30 0.28–0.33 0.35–0.40 0.38–0.43 0.70–0.90 0.75–1.00 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.70–0.90 0.75–1.00 0.75–1.00 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.20–0.40 0.20–0.40 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.4O–0.70 0.30–0.50 0.35–0.55 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.08–0.15 0.08–0.15 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 — — — — — — — — — — — Steel Heat Treatment: Metallurgy and Technologies UNS Number 8642 8645 86B45c 8650 8655 8660 G87200 G87400 8720 8740 G88220 8642 8645 0.40–0.45 0.43–0.48 0.43–0.48 0.48–0.53 0.51–0.59 0.56–0.64 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.75–1.00 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.70 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 0.15–0.25 — — — — — — 8720 8740 0.18–0.23 0.38–0.43 0.70–0.90 0.75–1.00 0.035 0.035 0.040 0.040 0.15–0.35 0.15–0.35 0.40–0.70 0.40–0.70 0.40–0.60 0.40–0.60 0.20–0.30 0.20–0.30 — — 8822 8822 0.20–0.25 0.75–1.00 0.035 0.040 0.15–0.35 0.40–0.70 0.40–0.60 0.30–0.40 — G92540 G92600 G93106 9254 9260 9310b — 9260 — 0.51–0.59 0.56–0.64 0.08–0.13 0.60–0.80 0.75–1.00 0.45–0.65 0.035 0.035 0.025 0.040 0.040 0.025 1.20–1.60 1.80–2.20 0.15–0.35 — — 3.00–3.50 0.60–0.80 — 1.00–1.40 — — 0.08–0.15 — — — G94151 G94171 G94301 94B15c 94B17c 94B30c — 94B17 94B30 0.13–0.18 0.15–0.20 0.28–0.33 0.75–1.00 0.75–1.00 0.75–1.00 0.035 0.035 0.035 0.04 0.04 0.04 0.15–0.35 0.15–0.35 0.15–0.35 0.30–0.60 0.30–0.60 0.30–0.60 0.30–0.50 0.30–0.50 0.30–0.50 0.08–0.15 0.08–0.15 0.08–0.15 Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. G86420 G86450 G86451 G86500 G86550 G86600 — — — — — 8655 — a Small 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 the following maximum amount, copper to 0.35%, nickel to 0.25%, chromium to 0.20%, and molybdenum to 0.06%. b Electric furnace steel. c Boron content is 0.0005–0.003%. Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 1.01–1.189. 21 the hardenability in order to optimize the strength and toughness after heat treatment. In some instances, however, alloying elements are used to reduce environmental degradation under certain specified conditions. Low-alloy steels can be classified according to: (1) chemical composition such as nickel steels, nickel–chromium steels, molybdenum steels, chromium–molybdenum steels, and so forth, based on the principal alloying elements present and as described in Table 1.2, (2) heat treatment 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 some steels are employed in more than one heat-treated conditions some overlap exists among the low-alloy steel classifications. However, these grades can be divided into four major groups such as (1) low-carbon quenched and tempered (QT) steels, (2) medium-carbon ultrahighstrength steels, (3) bearing steels, and (4) heat-resistant Cr–Mo steels (see Table 1.5). Low-carbon QT steels (also called low-carbon martensitic steels) are characterized by relatively high yield strength with a minimum yield strength of 690 MPa (100 ksi) and good notch toughness, ductility, corrosion resistance, or weldability. The chemical compositions of low-carbon QT steels are listed in Table 1.5. These steels are not included in SAE-AISI classification. However, they are covered by ASTM designations, and a few steels, such as HY-80 and HY-100, are included in military (MIL) specifications. The steels listed are primarily 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 also used for cold-headed and cold-forged parts as fasteners or pins and heat-treated to the desired properties [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 typical compositions such as SAE-AISI 4130, high-strength 4140, deeper hardening higher-strength 4340, 300M (a modification of 4340 steel with increased Si content (1.6%) to raise the third transformation 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 V and 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 Cr–V 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 (0.10–0.20% C) case-hardened steels and high-carbon (~1% C) through-hardened or surface-induction hardened steels (see Table 1.5). Many of these steels are covered by SAE-AISI designations. Chromium–molybdenum heat-resistant steels contain 0.5–9% Cr, 0.5–1.0% Mo, and usually less than 0.20% C. They are ordinarily supplied in the normalized and tempered, quenched and tempered, or annealed condition. Cr–Mo steels are extensively used in oil refineries, oil and gas industries, chemical industries, electric power generating stations and fossil fuel and nuclear power plants for piping, heat exchangers, superheater tubes, and pressure vessels. Various product shapes and corresponding ASTM specifications for these steels are provided in Table 1.6. Nominal chemical compositions are given in Table 1.7. ß 2006 by Taylor & Francis Group, LLC. TABLE 1.5 Chemical Compositions for Typical Low-Alloy Steels Steel C Si Mn P S Ni Cr Mo Other Low-carbon quenched and tempered steels A 514/A 517 grade A 0.15–0.21 0.40–0.80 0.80–1.10 0.035 0.04 — 0.50–0.80 0.18–0.28 A 514/A 517 grade F 0.10–0.20 0.15–0.35 0.60–1.00 0.035 0.04 0.70–1.00 0.40–0.65 0.40–0.60 A 514/A 517 grade R A 533 type A A 533 type C HY-80 0.15–0.20 0.25 0.25 0.12–0.18 0.20–0.35 0.15–0.40 0.15–0.40 0.15–0.35 0.85–1.15 1.15–1.50 1.15–1.50 0.10–0.40 0.035 0.035 0.035 0.025 0.04 0.04 0.04 0.025 0.90–1.10 — 0.70–1.00 2.00–3.25 0.35–0.65 — — 1.00–1.80 0.15–0.25 0.45–0.60 0.45–0.60 0.20–0.60 HY-100 0.12–0.20 0.15–0.35 0.10–0.40 0.025 0.025 2.25–3.50 1.00–1.80 0.20–0.60 0.05–0.15 Znb 0.0025 B 0.03–0.08 V 0.15–0.50 Cu 0.0005–0.005 B 0.03–0.08 V — — 0.25 Cu 0.03 V 0.02 Ti 0.25 Cu 0.03 V 0.02 Ti Medium-carbon ultrahigh-strength steels 4130 4340 300M D-6a 0.28–0.33 0.38–0.43 0.40–0.46 0.42–0.48 0.20–0.35 0.20–0.35 1.45–1.80 0.15–0.30 0.40–0.60 0.60–0.80 0.65–0.90 0.60–0.90 — — — — — — — — — 1.65–2.00 1.65–2.00 0.40–0.70 0.80–1.10 0.70–0.90 0.70–0.95 0.90–1.20 0.15–0.25 0.20–0.30 0.30–0.45 0.90–1.10 Carburizing bearing steels 4118 5120 3310 0.18–0.23 0.17–0.22 0.08–0.13 0.15–0.30 0.15–0.30 0.20–0.35 0.70–0.90 0.70–0.90 0.45–0.60 0.035 0.035 0.025 0.040 0.040 0.025 — — 3.25–3.75 0.40–0.60 0.70–0.90 1.40–1.75 0.08–0.18 — — — — — Through-hardened bearing steels 52100 A 485 grade 1 A 485 grade 3 0.98–1.10 0.90–1.05 0.95–1.10 0.15–0.30 0.45–0.75 0.15–0.35 0.25–0.45 0.95–1.25 0.65–0.90 0.025 0.025 0.025 0.025 0.025 0.025 — 0.25 0.25 1.30–1.60 0.90–1.20 1.10–1.50 — 0.10 0.20–0.30 Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. Composition, wt%a — 0.35 Cu 0.35 Cu — — 0.05 V min 0.05–0.10 V a Single 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. 140–194. b 23 TABLE 1.6 ASTM Specifications for Chromium–Molybdenum Steel Product Forms Type Tubes Pipe Castings Plate A 182-F2 — — A 387-Gr2 1Cr–1/2Mo A 182-F12 A 336-F12 — A 335-P2 A 369-FP2 A 426-CP2 A 335-P12 A 369-FP12 — A 387-Gr12 11/4Cr–1/2Mo A 182-F11 A 336-F11/F11A A 541-C11C A 182-F22/F22a A 336-F22/F22A A 541-C22C/22D A 182-F21 A 336-F21/F21A A 217-WC6 A 356-Gr6 A 389-C23 A 217-WC9 A 356-Gr10 A 387-Gr11 — A 387-Gr21 1 Forgings 1 /2Cr– /2Mo 21/4Cr–1Mo 3Cr–1Mo 3Cr–1MoV 5Cr–1/2Mo 5Cr–1/2MoSi A 182-F21b A 182-F5/F5a A 336-F5/F5A A 473-501/502 — 5Cr–1/2MoTi 7Cr–1/2Mo — A 182-F7 A 473-501A 9Cr–1Mo A 182-F9 A 336-F9 A 473-501B A 426-CP12 A 199-T11 A 200-T11 A 213-T11 A 199-T22 A 200-T22 A 213-T22 A 199-T21 A 200-T21 A 213-T21 — A 199-T5 A 200-T5 A 213-T5 A 213-T5b A 213-T5c A 199-T7 A 200-T7 A 213-T7 A 199-T9 A 200-T9 A 213-T9 A 335-P11 A 369-FP11 A 426-CP11 A 335-P22 A 369-FP22 A 426-CP22 A 335-P21 A 369-FP21 A 426-CP21 — A 335-P5 A 369-FP5 A 426-CP5 A 335-P5b A 426-CP5b A 335-P5c A 335-P7 A 369-FP7 A 426-CP7 A 335-P9 A 369-FP9 A 426-CP9 — A 217-C5 A 387-Gr22 A 542 — A 387-Gr5 — — — — — A 387-Gr7 A 217-C12 A 387-Gr9 Source: From Anon., ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140–194. 1.3.3.3 High-Strength Low-Alloy Steels A general description of HSLA steel is as that containing: (1) low carbon (0.03–0.25%) content to obtain good toughness, formability, and weldability, (2) one or more of the strong carbide-forming microalloying elements (MAEs) (e.g., V, Nb, or Ti), (3) a group of solid solution strengthening elements (e.g., Mn up to 2.0% and Si), and (4) one or more of the additional MAEs (e.g., Ca, Zr) and the rare earth elements, particularly Ce and La, for sulfide inclusion shape control and increasing toughness [4,5,21,22,36,37]. In many other HSLA steels, small amounts of Ni, Cr, Cu, and particularly Mo are also present, which increase atmospheric corrosion resistance and hardenability. A very fine ferrite grain structure in the final product produced by a combination of controlled rolling and controlled cooling with an optimum utilization of microalloying additions, in HSLA steels, is an important factor in simultaneously increasing strength and toughness and decreasing the ductile–brittle transition temperature (to as low as À708C). Carbides (NbC, VC, TiC), nitrides (TiN, NbN, AlN), and carbonitrides (e.g., V(C,N), Nb(C,N), (Nb,V) CN, (Nb,Ti) CN) are the dispersed second-phase particles that act as grain size refiners or dispersive strengthening phases in HSLA steels. ß 2006 by Taylor & Francis Group, LLC. Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. TABLE 1.7 Nominal Chemical Compositions for Heat-Resistant Chromium–Molybdenum Steels Composition (%)a Type 1 /2Cr–1/2Mo 1Cr–1/2Mo 11/4Cr–1/2Mo 11/4Cr–1/2Mo 21/4Cr–1Mo 3Cr–1Mo 3Cr–1MoVb 5Cr–1/2Mo 7Cr–1/2Mo 9Cr–1Mo 9Cr–1MoVc UNS Designation C Mn S P Si Cr Mo K12122 K11562 K11597 K11592 K21590 K31545 K31830 K41545 K61595 K90941 — 0.10–0.20 0.15 0.15 0.10–0.20 0.15 0.15 0.18 0.15 0.15 0.15 0.08–0.12 0.30–0.80 0.30–0.60 0.30–0.60 0.30–0.80 0.30–0.60 0.30–0.60 0.30–0.60 0.30–0.60 0.30–0.60 0.30–0.60 0.30–0.60 0.040 0.045 0.030 0.040 0.040 0.030 0.020 0.030 0.030 0.030 0.010 0.040 0.045 0.030 0.040 0.040 0.030 0.020 0.030 0.030 0.030 0.020 0.10–0.60 0.50 0.50–1.00 0.50–1.00 0.50 0.50 0.10 0.50 0.50–1.00 0.50–1.00 0.20–0.50 0.50–0.80 0.80–1.25 1.00–1.50 1.00–1.50 2.00–2.50 2.65–3.35 2.75–3.25 4.00–6.00 6.00–8.00 8.00–10.00 8.00–9.00 0.45–0.65 0.45–0.65 0.45–0 65 0.45–0.65 0.87–1.13 0.80–1.06 0.90–1.10 0.45–0.65 0.45–0.65 0.90–1.10 0.85–1.05 a Single values are maximums. Also contains 0.02–0.030% V, 0.001–0.003% B, and 0.015–0.035% Ti. c Also contains 0.40% Ni, 0.18–0.25% V, 0.06–0.10% Nb, 0.03–0.07% N, and 0.04% Al. Source: From Anon., ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140–194. b 25 HSLA steels are successfully used as ship, plate, bar, structural sections, and forged bar products, and find applications in several diverse fields such as oil and gas pipelines; in the automotive, agricultural, and pressure vessel industries, in offshore structures and platforms and in the constructions of crane, bridges, buildings, shipbuildings, railroad, tank cars, and power transmission and TV towers [36]. 1.3.3.3.1 Classification of HSLA Steels Several special terms are used to describe various types of HSLA steels [37–39]: 1. Weathering steels: Steels containing ~0.1% C, 0.2–0.5% Cu, 0.5–1.0% Mn, 0.05–0.15% P, 0.15–0.90% Si, and sometimes containing Cr and Ni, exhibiting superior atmospheric corrosion resistance. Typical applications include railroad cars, bridges, and unpainted buildings. 2. Control-rolled steels: Steels designated to develop a highly deformed austenite structure by hot rolling (according to a predetermined rolling schedule) that will transform to a very 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 no pearlite in the microstructure. 4. Microalloyed steels: Conventional HSLA steels containing V, Ti, or Nb, as defined above. They exhibit discontinuous yielding behavior. 5. Acicular ferrite steels: Very low-carbon (typically 0.03–0.06%) steels with enough hardenability (by Mn, Mo, Nb, and B additions) to transform on cooling to a very fine, high-strength acicular ferrite structure rather than the usual polygonal ferrite structure. 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 and precipitations. They contain 0.02–0.2% C, 0.6–1.6% Mn, 0.3–0.6% Mo, and MAEs (such as V, Nb, Ti, and B), usually containing 0.4–0.7% Cr. The yield strength of these steels 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 0.05–0.25% C, 1.5–2.0% Mn, 0.20–0.50 Mo, and MAEs (such as Nb, Ti, V, and B). Some steels containing small amounts of Ni, Cr, and Cu, after rolling or forging, and directly quenching and tempering attain a low-carbon martensite structure with high yield strength (760–1100 MPa), high toughness (CVN 50–130 J), and superior fatigue strength [5,40,41]. 8. Dual-phase steels: Steels comprising essentially fine dispersion of hard strong martensite but sometimes also retained austenite or even bainite in a soft and fine-grained ferrite matrix. The volume function of martensite is about 20–30%. Steels are characterized by continuous yielding (i.e., no yield point elongation), low yield stress (the YS/UTS ratio being around 0.50), high UTS, superior formability, and rapid initial work-hardening rate. Additionally, they possess greater resistance to onset of necking (i.e., plastic instability) in the uniaxial sheet material forming process to provide large uniform strain [42–45]. Table 1.8 lists HSLA steels according to chemical composition and minimum machining property requirements. ß 2006 by Taylor & Francis Group, LLC. TABLE 1.8 Composition Ranges and Limits for SAE HSLA Steels Heat Composition Limits (%)a SAE Designationb 942X 945A 945C 945X 950A 950B 950C 950D 950X 955X 960X 965X 970X 980X C max Mn max P max 0.21 0.15 0.23 0.22 0.15 0.22 0.25 0.15 0.23 0.25 0.26 0.26 0.26 0.26 1.35 1.00 1.40 1.35 1.30 1.30 1.60 1.00 1.35 1.35 1.45 1.45 1.65 1.65 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.15 0.04 0.04 0.04 0.04 0.04 0.04 a Maximum 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 indicates that the steel contains niobium, vanadium, nitrogen, or other alloying elements. A second suffix K indicates that the steel is produced fully killed using fine-grain practice; otherwise, the steel is produced semikilled. Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 1.01–1.189. b 1.3.3.4 Tool Steels A 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 are designed to have high hardness and durability under severe service conditions. They comprise a wide range from plain carbon steels with up to 1.2% C without appreciable amounts of alloying elements to the highly alloyed steels in which alloying additions reach 50%. Although some carbon tool steels and low-alloy tool steels have a wide range of carbon content, most of the higher alloy tool steels have a comparatively narrow carbon range. A mixed classification system is used to classify tool steels based on the use, composition, special mechanical properties, or method of heat treatment. According to AISI specification, there are nine main groups of wrought tool steels. Table 1.9 lists the compositions of these tool steels with corresponding designated symbols [46], which are discussed herein. High-speed steels are used for applications requiring long life at relatively high operating temperatures such as for heavy cuts or high-speed machining. High-speed steels are the most important alloy tool steels because of their very high hardness and good wear assistance in the heat-treated condition and their ability to retain high hardness and the elevated temperatures often 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]. ß 2006 by Taylor & Francis Group, LLC. 28 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.9 Composition Limits of Principal Types of Tool Steels Compositiona (%) Designation AlSI UNS C Si Cr Ni Mo W V Molybdenum high-speed steels M1 T11301 0.78–0.88 M2 T11302 0.78–0.88; 0.95–1.05 M3 class 1 T11313 1.00–1.10 M3 class 2 T11323 1.15–125 M4 T11304 1.25–1.40 M7 T11307 0.97–1.05 M10 T11310 0.84–0.94; 0.95–1.05 M30 T11330 0.75–0.85 M33 T11333 0.85–0.92 M34 T11334 0.85–0.92 M35 T11335 0.82–0.88 M36 T11336 0.80–0.90 M41 T11341 1.05–1.15 M42 T11342 1.05–1.15 M43 T11343 1.15–1.25 M44 T11344 1.10–1.20 M46 T11346 1.22–1.30 M47 T11347 1.05–1.15 M48 T11348 1.42–1.52 M62 T11362 1.25–1.35 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.10–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.20–0.60 0.15–0.40 0.20–0.40 0.20–0.40 0.20–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.20–0.50 0.20–0.45 0.20–0.45 0.20–0.45 0.20–0.45 0.20–0.55 0.20–0.45 0.20–0.45 0.15–0.50 0.20–0.45 0.20–0.45 0.20–0.45 0.15–0.50 0.15–0.65 0.15–0.65 0.30–0.55 0.40–0.65 0.20–0.45 0.15–0.40 0.15–0.40 3.50–4.00 3.75–4.50 3.75–4.50 3.75–4.50 3.75–4.75 3.50–4.00 3.75–4.50 3.50–4.25 3.50–4.00 3.50–4.00 3.75–4.50 3.75–4.50 3.75–4.50 3.50–4.25 3.50–4.25 4.00–4.75 3.70–4.20 3.50–4.00 3.50–4.00 3.50–4.00 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 8.20–9.20 4.50–5.50 4.75–6.50 4.75–6.50 4.25–5.50 8.20–9.20 7.75–8.50 7.75–9.00 9.00–10.00 7.75–9.20 4.50–5.50 4.50–5.50 3.25–4.25 9.00–10.00 7.50–8.50 6.00–7.00 8.00–8.50 9.25–10.00 4.75–5.50 10.00–11.00 1.40–2.10 5.50–6.75 5.00–6.75 5.00–6.75 5.25–6.50 1.40–2.l0 — 1.30–2.30 1.30–2.10 1.40–2.10 5.50–6.75 5.50–6.50 6.25–7.00 1.15–1.85 2.25–3.00 5.00–5.75 1.90–2.20 1.30–1.80 9.50–10.50 5.75–6.50 1.00–1.35 1.75–2.20 2.25–2.75 2.75–3.75 3.75–4.50 1.75–2.25 1.80–2.20 1.00–1.40 1.00–1.35 1.90–2.30 1.75–2.20 1.75–2.25 1.75–2.25 0.95–1.35 1.50–1.75 1.85–2.20 3.00–3.30 1.15–1.35 2.75–3.25 1.80–2.10 Tungsten high-speed steels T1 T12001 T2 T12002 T4 T12004 T5 T12005 T6 T12006 T8 T12008 T15 T12015 0.10–0.40 0.20–0.40 0.10–0.40 0.20–0.40 0.20–0.40 0.26–0.40 0.15–0.40 0.20–0.40 0.20–0.40 0.20–0.40 0.20–0.40 0.20–0.40 0.20–0.40 0.15–0.40 3.75–4.50 3.75–4.50 3.75–4.50 3.75–5.00 4.00–4.75 3.75–4.50 3.75–5.00 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max — 1.00 max 0.40–1.00 0.50–1.25 0.40–1.00 0.40–1.00 1.00 max 17.25–18.75 17.50–19.00 17.50–19.00 17.50–19.00 18.50–21.00 13.25–14.75 11.75–13.00 0.90–1.30 1.80–2.40 0.80–1.20 1.80–2.40 1.50–2.10 1.80–2.40 4.50–5.25 0.65–0.80 0.80–0.90 0.70–0.80 0.75–0.85 0.75–0.85 0.75–0.85 1.50–1.60 Co — — — — — — — 4.50–5.50 7.75–8.75 7.75–8.75 4.50–5.50 7.75–8.75 4.75–5.75 7.75–8.75 7.75–8.75 11.00–12.25 7.80–8.80 4.75–5.25 8.00–10.00 — — — 4.25–5.75 7.00–9.50 11.00–13.00 4.25–5.75 4.75–5.25 Steel Heat Treatment: Metallurgy and Technologies Mn 0.20–0.60 0.20–0.60 3.75–4.50 3.50–4.30 0.30 max 0.30 max 3.90–4.75 4.00–4.90 — 0.75–1.50 0.80–1.25 1.65–2.25 Chromium hot-work steels H10 T20810 H11 T20811 H12 T20812 H13 T20813 H14 T20814 H19 T20819 0.35–0.45 0.33–0.43 0.30–0.40 0.32–0.45 0.35–0.45 0.32–0.45 0.25–0.70 0.20–0.50 0.20–0.50 0.20–0.50 0.20–0.50 0.20–0.50 0.80–1.20 0.80–1.20 0.80–1.20 0.80–1.20 0.80–1.20 0.20–0.50 3.00–3.75 4.75–5.50 4.75–5.50 4.75–5.50 4.75–5.50 4.00–4.75 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 2.00–3.00 1.10–1.60 1.25–1.75 1.10–1.75 — 0.30–0.55 — — 1.00–1.70 — 4.00–5.25 3.75–4.50 0.25–0.75 0.30–0.60 0.50 max 0.80–1.20 — 1.75–2.20 Tungsten hot-work steels H21 T20821 H22 T20822 H23 T20823 H24 T20824 H25 T20825 H26 T20826 0.26–0.36 0.30–0.40 0.25–0.35 0.42–0.53 0.22–0.32 0.45–0.55b 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.40 0.15–0.50 0.15–0.40 0.15–0.60 0.15–0.40 0.15–0.40 0.15–0.40 3.00–3.75 1.75–3.75 11.00–12.75 2.50–3.50 3.75–4.50 3.75–4.50 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max — — — — — — 8.50–10.00 10.00–11.75 11.00–12.75 14.00–16.00 14.00–16.00 17.25–19.00 0.30–0.60 0.25–0.50 0.75–1.25 0.40–0.60 0.40–0.60 0.75–1.25 — — — — — — Molybdenum hot-work steels H42 T20842 0.55–0.70b 0.15–0.40 — 3.75–4.50 0.30 max 4.50–5.50 5.50–6.75 1.75–2.20 — Air-hardening, medium-alloy, cold-work steels A2 T30102 0.95–1.05 A3 T30103 1.20–1.30 A4 T30104 0.95–1.05 A6 T30106 0.65–0.75 A7 T30107 2.00–2.85 A8 T30108 0.50–0.60 A9 T30109 0.45–0.55 Al0 T30110 1.25–1.50c 1.00 max 0.40–0.60 1.80–2.20 1.80–2.50 0.80 max 0.50 max 0.50 max 1.60–2.10 0.50 max 0.50 max 0.50 max 0.50 max 0.50 max 0.75–1.10 0.95–1.15 1.00–1.50 4.75–5.50 4.75–5.50 0.90–2.20 0.90–1.20 5.00–5.75 4.75–5.50 4.75–5.50 — 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 1.25–1.75 1.55–2.05 0.90–1.40 0.90–1.40 0.90–1.40 0.90–1.40 0.90–1.40 1.15–1.65 1.30–1.80 1.25–1.75 — — — — 0.50–1.50 1.00–1.50 — 0.15–0.50 0.80–1.40 — — 3.90–5.15 — 0.80–1.40 — — — — — — — — — High-carbon, high-chromium, cold-work steels D2 T30402 1.40–1.60 D3 T30403 2.00–2.35 D4 T30404 2.05–2.40 D5 T30405 1.40–1.60 D7 T304O7 2.15–2.50 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 0.60 max 11.00–13.00 11.00–13.50 11.00–13.00 11.00–13.00 11.50–13.50 0.30 max 0.30 max 0.30 max 0.30 max 0.30 max 0.70–1.20 — 0.70–1.20 0.70–1.20 0.70–1.20 — 1.00 max — — — 1.10 max 1.00 max 1.00 max 1.00 max 3.80–4.40 — — — — — — — 4.00–4.50 — — — 2.50–3.50 — Continued 29 0.15–0.45 0.15–0.45 Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. Intermediate high-speed steels M50 T11350 0.78–0.88 M52 T11352 0.85–0.95 30 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.9 (Continued ) Composition Limits of Principal Types of Tool Steels Compositiona (%) Designation AlSI UNS C Si Cr Ni Mo W V Co Oil-hardening cold-work steels O1 T31501 0.85–1.00 O2 T31502 0.85–0.95 O6 T31506 1.25–1.55c O7 T31507 1.10–1.30 1.00–1.40 1.40–1.80 0.30–1.10 1.00 max 0.50 max 0.50 max 0.55–1.50 0.60 max 0.40–0.60 0.50 max 0.30 max 0.35–0.85 0.30 max 0.30 max 0.30 max 0.30 max — 0.30 max 0.20–0.30 0.30 max 0.40–0.60 — — 1.00–2.00 0.30 max 0.30 max — 0.40 max — — — — Shock-resisting steels S1 T41901 S2 T41902 S5 T41905 S6 T41906 S7 T41907 0.10–0.40 0.30–0.50 0.60–1.00 1.20–1.50 0.20–0.90 0.15–1.20 0.90–1.20 1.75–2.25 2.00–2.50 0.20–1.00 1.00–1.80 — 0.50 max 1.20–1.50 3.00–3.50 0.30 max 0.30 max — — — 0.50 max 0.30–0.60 0.20–1.35 0.30–0.50 1.30–1.80 1.50–3.00 — — — — 0.15–0.30 0.50 max 0.35 max 0.20–0.40 0.20–0.30d — — — — — 0.10–0.90 0.25–0.80 0.50 max 0.50 max 0.70–1.20 0.60–1.20 — 1.25–2.00 0.25 max 0.50 max — — 0.10–0.30 0.20–0.30d — — 0.40–0.55 0.40–0.55 0.50–0.65 0.40–0.50 0.45–0.55 Low-alloy special-purpose tool steels L2 T61202 0.45–1.00b L6 T61206 0.65–0.75 Steel Heat Treatment: Metallurgy and Technologies Mn Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. Low-carbon mold steels P2 T51602 P3 T51603 P4 T51604 P5 T51605 P6 T51606 P20 T51620 P21 T51621 0.10 max 0.10 max 0.12 max 0.10 max 0.05–0.15 0.28–0.40 0.18–0.22 Water-hardening tool steels W1 T72301 0.70–1.50e W2 T72302 0.85–1.50e W5 T72305 1.05–1.15 0.10–0.40 0.20–0.60 0.20–0.60 0.20–0.60 0.35–0.70 0.60–1.00 0.20–0.40 0.10–0.40 0.40 max 0.10–0.40 0.40 max 0.10–0.40 0.20–0.80 0.20–0.40 0.75–1.25 0.40–0.75 4.00–5.25 2.00–2.50 1.25–1.75 1.40–2.00 0.50 max 0.10–4.50 1.00–1.50 — 0.35 max 3.25–3.75 — 3.90–4.25 0.15–0.40 — 0.40–1.00 — — 0.30–0.55 — — — — — — — — — — — — — — 0.15–0.25 — — — — — — 1.05–1.25Al 0.10–0.40 0.10–0.40 0.10–0.40 0.10–0.40 0.10–0.40 0.10–0.40 0.15 max 0.15 max 0.40–0.60 0.20 max 0.20 max 0.20 max 0.10 max 0.10 max 0.10 max 0.15 max 0.15 max 0.15 max 0.10 max 0.15–0.35 0.10 max — — — a All 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 be increased to 0.06 to 0.15% to improve machinability of group A, D, H, M, and T steels. b Available in several carbon ranges. c Contains free graphite in the microstructure. d Optional. e Specified 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. 757–779. 31 High-speed steels are grouped into molybdenum type M and tungsten type T. Type M tool steels contain Mo, W, Cr, V, Mo, and C as the major alloying elements, while type T tool steels 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 subgroup consisting of intermediate high-speed steels in the M group. The most popular grades among molybdenum types are M1, M2, M4, M7, M10, and M42, while those among tungsten types are T1 and T15. The main advantage of type M steels is their lower initial cost (approximately 40% cheaper than that of similar type T steels), but they are more susceptible to decarburizing, thereby necessitating better temperature control than type T steels. By using salt baths and sometimes surface coatings, decarburization can be controlled. The mechanical properties of type M and type T steels are similar except that type M steels have slightly greater toughness than type T steels at the same hardness level [4]. Hot-work tool steels (AISI series) fall into three major groups: (1) chromium-base, types H1–H19, (2) tungsten-base, types H20–H39, and (3) molybdenum-base, types H40–H59. The distinction is based on the principal alloying additions; however, all classes have medium carbon content and Cr content varying from 1.75 to 12.75%. Among these steels, H11, H12, H13 are produced in large quantities. These steels possess good redhardness and retain high hardness (~50 Rc) after prolonged exposures at 500–5508C. They are used extensively for hot-work applications, which include parts for aluminum and magnesium die casting and extrusion, plastic injection molding, and compression and transfer molds [47]. Cold-work tool steels comprise three categories: (1) air-hardening, medium-alloy tool steels (AISI A series), (2) high-chromium tool steels (AISI D series), and (3) oil-hardening tool steels (AISI O series). AISI A series tool steels have high hardenability and harden readily on air cooling. In the air-hardened and tempered condition, they are suitable for applications where improved toughness and reasonably good abrasion resistance are required such as for forming, blanking, and drawing dies. The most popular grade is A2. AISI D series tool steels possess excellent wear resistance and nondeforming properties, thereby making them very useful 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, and so forth. Among these steels, D2 is by far the most popular grade [47]. AISI O series tool steels 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 the most widely used. Shock-resisting tool steels (AISI S series) are used where repetitive impact stresses are encountered 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 and S7 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 hardenability and wear-resistance properties, type L6 and the low-carbon version of L2 are commonly used for a large number of machine parts. Mold steels (AISI P series) are mostly used in low-temperature die casting dies and in molds for the injection or compression molding of plastics [46]. Water-hardening tool steels (AISI W series): Among the three compositions listed, W1 is the most widely used as cutting tools, punches, dies, files, reamers, taps, drills, razors, woodworking tools, and surgical instruments in the quenched and tempered condition. ß 2006 by Taylor & Francis Group, LLC. 1.3.3.5 Stainless Steels Stainless steels may be defined as complex alloy steels containing a minimum of 10.5% Cr with or without other elements to produce austenitic, ferritic, duplex (ferritic–austenitic), 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 65–70% of the total U.S. stainless steel production and have occupied a dominant position because of their higher corrosion resistance such as strength and toughness at both elevated and ambient temperatures, excellent cryogenic properties, esthetic appeal, and varying specific combination and properties that can be obtained by different compositions within the group [50]. In general, austenitic stainless steels are Fe–Cr–Ni–C and Fe–Cr–Mn–Ni–N alloys containing 16–26% Cr, 0.75–19.0% Mn, 1–40% Ni, 0.03–0.35% C, and sufficient N to stabilize austenite at room and elevated temperatures. The 2xx series (Cr–Mn–Ni) steels contain N, 5.5–15.5% Mn, and up to 6% Ni, the 3xx (Cr–Ni) types contain higher amounts of Ni and up to 2% Mn. Mo, Cu, and Si may be added to increase corrosion resistance. Ti and Nb may be added to decrease the sensitivity of intergranular corrosion. The addition of Mo and N may increase halide-pitting resistance; Si and Cu may be added to increase resistance to stress corrosion 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 10.5–30% Cr with additions of Mn and Si and occasionally Mo, Ni, Al, Ti, or Nb to confer particular characteristics. As they remain ferritic at room and elevated temperatures, they cannot be hardened by heat treatment. The ductile– brittle transition temperature of ferrite stainless steels is higher than room temperature; if C þ 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 range 275–415 MPa (40–60 ksi). They are used because of their good ductility, good resistance to general liquid corrosion, and high-temperature oxidation, resistance to pitting and stress corrosion cracking, and generally lower cost than the austenitic grades [10]. As in the ferritic grade, S and Se may be added to improve machinability. The standard ferrite stainless steels are types 405, 409, 429, 430, 430F, 430F–Se, 434, 436, 439, 444, and 446 (Table 1.10). In addition, high-quality ferrite stainless steels are types E-Brite 26-1, MoNiT (25-4-4), AL29-4c, and AL29-4-2 (Table 1.11). Duplex stainless steels contain 18–29% Cr, 2.5–8.5% Ni, and 1–4% Mo, up to 2.5% Mn, up to 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. The yield strength increases with increasing ferrite content. The ultimate tensile strength rises to a maximum at 70–80% ferrite, then decreases as the ferrite goes to 100%. Compared to austenitic grades, they can offer improved yield strength (about two to three times greater) and greater resistance 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. In general, 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 popular duplex steels (Table 1.10 and Table 1.11). The new type SAF2507 contains ultralow carbon ß 2006 by Taylor & Francis Group, LLC. 34 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.10 Compositions of Standard Stainless Steels Type UNS Designation Austenitic types 201 202 205 301 302 302B 303 303Se 304 304H 304L 304LN 302Cu 304N 305 308 309 309S 310 310S 314 316 316F 316H 316L 316LN 316N Composition (%)a Mn Si Cr Ni P S Other S20100 S20200 S20500 S30100 S30200 S30215 S30300 S30323 S30400 S30409 S30403 S30453 S30430 S30451 S30500 S30800 S30900 S30908 S31000 S31008 S31400 S31600 S31620 S31609 S31603 S31653 0.15 0.15 0.12–0.25 0.15 0.15 0.15 0.15 0.15 0.08 0.04–0.10 0.03 0.03 0.08 0.08 0.12 0.08 0.20 0.08 0.25 0.08 0.25 0.08 0.08 0.04–0.10 0.03 0.03 5.5–7.5 7.5–10.0 14.0–15.5 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 1.00 1.00 1.00 1.00 1.00 2.0–3.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.50 1.50 1.5–3.0 1.00 1.00 1.00 1.00 1.00 16.0–18.0 17.0–19.0 16.5–18.0 16.0–18.0 17.0–19.0 17.0–19.0 17.0–19.0 17.0–19.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–20.0 17.0–19.0 18.0–20.0 17.0–19.0 19.0–21.0 22.0–24.0 22.0–24.0 24.0–26.0 24.0–26.0 23.0–26.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 3.5–5.5 4.0–6.0 1.0–1.75 6.0–8.0 8.0–10.0 8.0–10.0 8.0–10.0 8.0–10.0 8.0–10.5 8.0–10.5 8.0–12.0 8.0–12.0 8.0–10.0 8.0–10.5 10.5–13.0 10.0–12.0 12.0–15.0 12.0–15.0 19.0–22.0 19.0–22.0 19.0–22.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 0.06 0.06 0.06 0.045 0.045 0.045 0.2 0.2 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.20 0.045 0.045 0.045 0.03 0.03 0.03 0.03 0.03 0.03 0.15 min 0.06 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.10 min 0.03 0.03 0.03 S31651 0.08 2.00 1.00 16.0–18.0 10.0–14.0 0.045 0.03 0.25 N 0.25 N 0.32–0.40 N — — — 0.6 Mob 0.15 min Se — — — 0.10–0.16 N 3.0–4.0 Cu 0.10–0.16 N — — — — — — — 2.0–3.0 Mo 1.75–2.5 Mo 2.0–3.0 Mo 2.0–3.0 Mo 2.0–3.0 Mo; 0.10–0.16N 2.0–3.0 Mo; 0.10–0.16 N Steel Heat Treatment: Metallurgy and Technologies C S31700 S31703 S32100 S32109 N08330 S34700 S34709 S34800 0.08 0.03 0.08 0.04–0.10 0.08 0.08 0.04–0.10 0.08 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 1.00 1.00 1.00 1.00 0.75–1.5 1.00 1.00 1.00 18.0–20.0 18.0–20.0 17.0–19.0 17.0–19.0 17.0–20.0 17.0–19.0 17.0–19.0 17.0–19.0 11.0–15.0 11.0–15.0 9.0–12.0 9.0–12.0 34.0–37.0 9.0–13.0 9.0–13.0 9.0–13.0 0.045 0.045 0.045 0.045 0.04 0.045 0.045 0.045 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 348H S34809 0.04–0.10 2.00 1.00 17.0–19.0 9.0–13.0 0.045 0.03 384 S38400 0.08 2.00 1.00 15.0–17.0 17.0–19.0 0.045 0.03 Ferritic types 405 409 S40500 S40900 0.08 0.08 1.00 1.00 1.00 1.00 11.5–14.5 10.5–11.75 — 0.5 0.04 0.045 0.03 0.045 429 430 430F 430FSe 434 436 S42900 S43000 S43020 S43023 S43400 S43600 0.12 0.12 0.12 0.12 0.12 0.12 1.00 1.00 1.25 1.25 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 14.0–16.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 — — — — — — 0.04 0.04 0.06 0.06 0.04 0.04 0.03 0.03 0.15 min 0.06 0.03 0.03 439 S43035 0.07 1.00 1.00 17.0–19.0 0.50 0.04 0.03 442 444 S44200 S44400 0.2 0.025 1.00 1.00 1.00 1.00 18.0–23.0 17.5–19.5 — 1.00 0.04 0.04 0.03 0.03 446 S44600 0.20 1.50 1.00 23.0–27.0 — 0.04 0.03 3.0–4.0 Mo 3.0–4.0 Mo 5  %C min Ti 5  %C min Ti — 10  %C min Nb 8  %C min–1.0 max Nb 0.2 Co; 10  %C min Nb; 0.10 Ta 0.2 Co; 8  %C min– 1.0 max Nb; 0.10 Ta — Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. 317 317L 321 321H 330 347 347H 348 0.10–0.30 A1 6  %C min– 0.75 max Ti — — 0.6 Mob 0.15 min Se 0.75–1.25 Mo 0.75–1.25 Mo; 5  %C min– 0.70 max Nb 0.15 Al; 12  %C min–1.10 Ti 1.75–2.50 Mo; 0.025 N; 0.2 þ 4 (%C þ %N) min– 0.8 max (Ti þ Nb) 0.25 N Continued 35 Type UNS Designation 36 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.10 (Continued) Compositions of Standard Stainless Steels Composition (%)a C Mn Si Cr Ni P S Other 0.20 1.00 0.75 23.0–28.0 2.50–5.00 0.04 0.03 1.00–2.00 Mo Martensitic types 403 410 414 416 416Se 420 420F 422 S40300 S41000 S41400 S41600 S41623 S42000 S42020 S42200 0.15 0.15 0.15 0.15 0.15 0.15 min 0.15 min 0.20–0.25 1.00 1.00 1.00 1.25 1.25 1.00 1.25 1.00 0.50 1.00 1.00 1.00 1.00 1.00 1.00 0.75 11.5–13.0 11.5–13.5 11.5–13.5 12.0–14.0 12.0–14.0 12.0–14.0 12.0–14.0 11.5–13.5 — — 1.25–2.50 — — — — 0.5–1.0 0.04 0.04 0.04 0.06 0.06 0.04 0.06 0.04 0.03 0.03 0.03 0.15 min 0.06 0.03 0.15 min 0.03 431 440A 440B 440C S43100 S44002 S44003 S44004 0.20 0.60–0.75 0.75–0.95 0.95–1.20 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 15.0–17.0 16.0–18.0 16.0–18.0 16.0–18.0 1.25–2.50 — — — 0.04 0.04 0.04 0.04 0.03 0.03 0.03 0.03 — — — 0.6 Mob 0.15 min Se — 0.6 Mob 0.75–1.25 Mo; 0.75–1.25 W; 0.15–0.3 V — 0.75 Mo 0.75 Mo 0.75 Mo Precipitation-hardening types PH13-8Mo S13800 0.05 0.20 0.10 12.25–13.25 7.5–8.5 0.01 0.008 15-5PH S15500 0.07 1.00 1.00 14.0–15.5 3.5–5.5 0.04 0.03 17-4PH S17400 0.07 1.00 1.00 15.5–17.5 3.0–5.0 0.04 0.03 17-7PH S17700 0.09 1.00 1.00 16.0–18.0 6.5–7.75 0.04 0.04 a Single 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. 841–907. b 2.0–2.5 Mo; 0.90–1.35 Al; 0.01 N 2.5–4.5 Cu; 0.15–0.45 Nb 3.0–5.0 Cu; 0.15–0.45 Nb 0.75–1.5 Al Steel Heat Treatment: Metallurgy and Technologies Duplex (ferritic–austenitic) type 329 S32900 Composition (%)b UNS Designation C Ma Si Austenitic stainless steels Gall-Tough 203 EZ (XM-II) S20161 S20300 0.15 0.08 4.00–6.00 5.0–6.5 3.00–4.00 1.00 15.00–18.00 16.0–18.0 Nitronic 50 (XM-19) S20910 0.06 4.0–6.0 1.00 Tenelon (XM-31) Cryogenic Tenelon (XM-14) Esshete 1250 S21400 0.12 14.5–16.0 S21460 S21500 0.12 0.15 Type 216 (XM-17) S21600 Type 216 L (XM-18) Designationa Cr Ni P S 4.00–6.00 5.0–6.5 0.040 0.040 0.040 0.18–0.35 20.5–23.5 11.5–13.5 0.040 0.030 0.3–1.0 7.0–18.5 0.75 0.045 0.030 14.0–16.0 5.5–7.0 1.00 1.20 7.0–19.0 14.0–16.0 5.0–6.0 9.0–11.0 0.060 0.040 0.030 0.030 0.08 7.5–9.0 1.00 17.5–22.0 5.0–7.0 0.045 0.030 S21603 0.03 7.5–9.0 1.00 17.5–22.0 7.5–9.0 0.045 0.030 Nitronic 60 Nitronic 40 (XM-10) 21-6-9 LC Nitronic 33 (18-3-Mn) Nitronic 32 (18-2-Mn) 18-18 Plus S21800 S21900 S21904 S24000 S24100 S28200 0.10 0.08 0.04 0.08 0.15 0.15 7.0–9.0 8.0–10.0 8.00–10.00 11.50–14.50 11.00–14.00 17.0–19.0 3.5–4.5 1.00 1.00 1.00 1.00 1.00 16.0–18.0 19.0–21.5 19.00–21.50 17.00–19.00 16.50–19.50 17.5–19.5 8.0–9.0 5.5–7.5 5.50–7.50 2.50–3.75 0.50–2.50 — 0.040 0.060 0.060 0.060 0.060 0.045 0.030 0.030 0.030 0.030 0.030 0.030 303 Plus X (XM-5) MVMAc 304Bld S30310 S30415 S30424 0.15 0.05 0.08 2.5–4.5 0.60 2.00 1.00 1.30 0.75 17.0–19.0 18.5 18.00–20.00 7.0–10.0 9.50 12.00–15.00 0.020 — 0.045 0.25 min — 0.030 Other Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. TABLE 1.11 Compositions of Nonstandard Stainless Steels 0.08–0.20 N 0.5 Mo; 1.75–2.25 Cu 1.5–3.0 Mo; 0.2–0.4 N; 0.1–0.3 Nb; 0.1–0.3 V 0.35 N 0.35–0.50 N 0.003–0.009 B; 0.75–1.25 Nb; 0.15–0.40 V 2.0–3.0 Mo; 0.25–0.50 N 2.0–3.0 Mo; 0.25–0.50 N 0.08–0.18 N 0.15–0.40 N 0.15–0.40 N 0.20–0.40 N 0.20–0.45 N 0.5–1.5 Mo; 0.5–1.5 Cu; 0.4–0.6 N 0.6 Mo 0.15 N; 0.04 Ce 0.10 N; 1.00–1.25 B 37 Continued 38 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.11 (Continued) Compositions of Nonstandard Stainless Steels Composition (%)b UNS Designation C 304 HN (XM-21) Cronifer 1815 LCSi RA 85 Hc 253 MA S30452 S30600 S30615 S30815 0.04–0.10 0.018 0.20 0.05–0.l0 2.00 2.00 0.80 0.80 1.00 3.7–4.3 3.50 1.4–2.0 18.0–20.0 17.0–18.5 18.5 20.0–22.0 Type 309 S Cb S30940 0.08 2.00 1.00 Type 310 Cb S31040 0.08 2.00 254 SMO S31254 0.020 Type 316 Ti S31635 Type 316 Cb Designationa Ma Si Cr Ni S Other 8.0–10.5 14.0–15.5 14.50 10.0–12.0 0.045 0.020 — 0.040 0.030 0.020 — 0.030 22.0–24.0 12.0–15.0 0.045 0.030 1.50 24.0–26.0 19.0–22.0 0.045 0.030 1.00 0.80 19.50–20.50 17.50–18.50 0.030 0.010 0.08 2.00 1.00 16.0–18.0 10.0–14.0 0.045 0.030 S31640 0.08 2.00 1.00 16.0–18.0 10.0–14.0 0.045 0.030 Type 316 HQ — 0.030 2.00 1.00 16.00–18.25 10.00–14.00 0.030 0.015 Type 317 LM 17-14-4 LN S31725 S31726 0.03 0.03 2.00 2.00 1.00 0.75 18.0–20.0 17.0–20.0 13.5–17.5 13.5–17.5 0.045 0.045 0.030 0.030 Type 317 LN Type 370 S31753 S37000 0.03 0.03–0.05 2.00 1.65–2.35 1.00 0.5–1.0 18.0–20.0 12.5–14.5 11.0–15.0 14.5–16.5 0.030 0.040 0.030 0.010 0.16–0.30 N 0.2 Mo 1.0 Al 0.14–0.20 N; 0.03–0.08 Ce; 1.0 Al 10  %C min to 1.10 max Nb l0  %C min to 1.10 max Nb þ Ta 6.00–6.50 Mo; 0.50–1.00 Cu; 0.180–0.220 N 5  %(C þ N) min to 0.70 max Ti; 2.0–3.0 Mo; 0.10 N 10  %C min to 1.10 max Nb þ Ta; 2.0–3.0 Mo; 0.10 N 3.00–4.00 Cu; 2.00–3.00 Mo 4.0–5.0 Mo; 0.10 N 4.0–5.0 Mo; 0.10–0.20 N 0.10–0.22 N 1.5–2.5 Mo; Steel Heat Treatment: Metallurgy and Technologies P S38100 S63198 0.08 0.28–0.35 2.00 0.75–1.50 1.5–2.5 0.03–0.8 17.0–19.0 18.0–21.0 17.5–18.5 8.0–11.0 0.030 0.040 0.030 0.030 20Cb-3 N08020 0.07 2.00 1.00 19.0–21.0 32.0–38.0 0.045 0.035 20Mo-4 N08024 0.03 1.00 0.50 22.5–25.0 35.0–40.0 0.035 0.035 20Mo-6 N08026 0.03 1.00 0.50 22.00–26.00 33.00–37.20 0.03 0.03 Sanicro 28 N08028 0.02 2.00 1.00 26.0–28.0 29.5–32.5 0.020 0.015 AL-6X AL-6XN N08366 N08367 0.035 0.030 2.00 2.00 1.00 1.00 20.0–22.0 20.0–22.0 23.5–25.5 23.50–25.50 0.030 0.040 0.030 0.030 JS-700 N08700 0.04 2.00 1.00 19.0–23.0 24.0–26.0 0.040 0.030 Type 332 N08800 0.01 1.50 1.00 19.0–23.0 30.0–35.0 0.045 0.015 904L N08904 0.02 2.00 1.00 19.0–23.0 23.0–28.0 0.045 0.035 Cronifer 1925 hMo N08925 0.02 1.00 0.50 24.0–26.0 19.0–21.0 0.045 0.030 39 Continued Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. 18-18-2 (XM-15) 19-9 DL 0.1–0.4 Ti; 0.005 N; 0.05 Co — 1.0–1.75 Mo; 0.1–0.35 Ti; 1.0–1.75 W; 0.25–0.60 Nb 2.0–3.0 Mo; 3.0–4.0 Cu; 8  %C min to 1.00 max Nb 3.50–5.00 Mo; 0.50–1.50 Cu; 0.15–0.35 Nb 5.00–6.70 Mo; 2.00–4.00 Cu 3.0–4.0 Mo; 0.6–1.4 Cu 6.0–7.0 Mo 6.00–7.00 Mo; 0.18–0.25 N 4.3–5.0 Mo; 8  %C min to 0.5 max Nb; 0.5 Cu; 0.005 Pb; 0.035 S 0.15–0.60 Ti; 0.15–0.60 Al 4.0–5.0 Mo: 1.0–2.0 Cu 6.0–7.0 Mo; 0.8–1.5 Cu; 0.10–0.20 N 40 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.11 (Continued ) Compositions of Nonstandard Stainless Steels Composition (%)b Designationa UNS Designation Cronifer 2328 — C Ma Si Cr Ni P S Other 0.75 0.75 22.0–24.0 26.0–28.0 0.030 0.015 2.5–3.5 Cu; 0.4–0.7 Ti; 2.5–3.0 Mo 1.5–2.5 Mo 5  %C min to 0.75 max Ti 0.3 þ 9  (%C) min to 0.90 max Nb; 0.1–0.5 Ti; 0.03 N 0.75–1.5 Mo; 0.05–0.2 Nb; 0.015 N; 0.2 Cu 3.5–4.5 Mo; 0.2 þ 4 (%C þ %N) min to 0.8 max (Ti þ Nb); 0.035 N 2.5–3.5 Mo; 0.2 þ 4 (%C þ %N) min to 0.8 max (Ti þ Nb); 0.035 N 3.60–4.20 Mo; 0.20–1.00 Ti þ Nb and 6 (%C þ %N) min Ti þ Nb; 0.045 N 3.5–4.2 Mo; 0.15 Cu; 0.02 N; 0.025 max (%C þ %N) Ferritic stainless steels 18-2 FM (XM-34) Type 430 Ti S18200 S43036 0.08 0.10 1.25–2.50 1.00 1.00 1.00 17.5–19.5 16.0–19.5 — 0.75 0.040 0.040 0.15 min 0.030 Type 441 S44100 0.03 1.00 1.00 17.5–19.5 1.00 0.040 0.040 E-Brite 26-1 S44627 0.01 0.40 0.40 25.0–27.0 0.50 0.020 0.020 MoNiT (25-4-4) S44635 0.025 1.00 0.75 24.5–26.0 3.5–4.5 0.040 0.030 Sea-Cure (SC-1) S44660 0.025 1.00 1.00 25.0–27.0 1.5–3.5 0.040 0.030 AL29-4C S44735 0.030 1.00 1.00 28.0–30.0 1.00 0.040 0.030 AL29-4-2 S44800 0.01 0.30 0.20 28.0–30.0 2.0–2.5 0.025 0.020 Steel Heat Treatment: Metallurgy and Technologies 0.04 — — — — 0.04 0.02 0.06 0.03 ALFA IV — Sealmet 1 — 0.30 1.00 0.2–0.5 1.03 0.50 0.50 0.2–0.5 18.0 12.0 12.0–14.0 11.75–12.25 0.03 0.50 0.60 0.08 0.5–0.8 — — — 0.50 0.45 — — 0.040 0.030 — — 0.030 0.020 19.0–21.0 0.45 0.035 0.005 0.3–0.6 28.0–29.0 0.40 0.030 0.015 Duplex stainless steels 44LN S31200 0.030 2.00 1.00 24.0–26.0 5.50–6.50 0.045 0.030 DP-3 S31260 0.030 1.00 0.75 24.0–26.0 5.50–7.50 0.030 0.030 3RE60 2205 S31500 S31803 0.030 0.030 1.20–2.00 2.00 1.40–2.00 1.00 18.00–19.00 21.0–23.0 4.25–5.25 4.50–6.50 0.030 0.030 0.030 0.020 2304 S32304 0.030 2.50 1.0 21.5–24.5 3.0–5.5 0.040 0.040 Uranus 50 S32404 0.04 2.00 1.0 20.5–22.5 5.5–8.5 0.030 0.010 Ferralium 255 S32550 0.04 1.50 1.00 24.0–27.0 4.50–6.50 0.04 0.03 7-Mo-PlUS S32950 0.03 2.00 0.60 26.0–29.0 3.50–5.20 0.035 0.010 Martensitic stainless steels Type 410 S S41008 Type 410 Cb (XM-30) S41040 0.08 0.15 1.00 1.00 1.00 1.00 11.5–13.5 11.5–13.5 0.60 0.040 0.040 0.030 0.030 — 2.0 A1; 0.4 Ti 1.2 Al; 0.3 Ti 2.75–4.25 Al; 0.6 Ti 0.75–1.25 Al; 0.65–0.75 Nb; 0.3–0.5 Ti; 0.03 N 4.75–5.25 Al; 0.005–0.035 Ce; 0.03 N 0.04 N Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. 18 SR(c) 12 SR(c) 406 408 Cb 1.20–2.00 Mo; 0.14–0.20 N 2.50–3.50 Mo; 0.20–0.80 Cu; 0.10–0.30 N; 0.10–0.50 W 2.50–3.00 Mo 2.50–3.50 Mo; 0.08–0.20 N 0.05–0.60 Mo; 0.05–0.60 Cu; 0.05–0.20 N 2.0–3.0 Mo; 1.0–2.0 Cu; 0.20 N 2.00–4.00 Mo; 1.50–2.50 Cu; 0.10–0.25 N 1.00–2.50 Mo; 0.15–0.35 N — 0.05–0.20 Nb 41 Continued 42 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.11 (Continued) Compositions of Nonstandard Stainless Steels Composition (%)b Designationa UNS Designation C Ma Si Cr Ni P S S41050 S41500 S41610 0.04 0.05 0.15 1.00 0.5–1.0 1.5–2.5 1.00 0.60 1.00 10.5–12.5 11.5–14.0 12.0–14.0 0.60–1.1 3.5–5.5 — 0.045 0.030 0.060 0.030 0.030 0.15 min 0.030 S41800 S42010 S42023 0.15–0.20 0.15–0.30 0.3–0.4 0.50 1.00 1.25 0.50 1.00 1.00 12.0–14.0 13.5–15.0 12.0–14.0 1.8–2.2 0.25–1.00 — 0.040 0.040 0.060 0.030 0.060 Lapelloy S42300 0.27–0.32 0.95–1.35 0.50 11.0–12.0 0.50 0.025 0.025 Type 440 F Type 440 F–Se S44020 S44023 0.95–1.20 0.95–1.20 1.25 1.25 1.00 1.00 16.0–18.0 16.0–18.0 0.75 0.75 0.040 0.040 0.10–0.35 0.030 0.10 N 0.5–1.0 Mo 0.6 Mo 2.5–3.5 W 0.40–1.00 Mo 0.15 min Se; 0.6 Zr; 0.6 Cu 2.5–3.0 Mo; 0.2–0.3 V 0.08 N 0.15 min Se; 0.60 Mo Steel Heat Treatment: Metallurgy and Technologies E4 CA6NM 416 Plus X (XM-6) Type 418 (Greek Ascolloy) TrimRite Type 420 F–Se Other Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. Precipitation-hardening stainless steels PH14-4Mo S14800 0.05 1.00 1.00 13.75–15.0 7.75–8.75 0.015 0.010 PH15-7Mo (Type 632) S15700 0.09 1.00 1.00 14.0–16.0 6.5–7.75 0.040 0.030 AM-350 (Type 633) S35000 0.07–0.11 0.5–1.25 0.50 16.0–17.0 4.0–5.0 0.040 0.030 AM-355 (Type 634) S35500 0.10–0.15 0.5–1.25 0.50 15.0–16.0 4.0–5.0 0.040 0.030 Custom 450 (XM-25) S45000 0.05 1.00 1.00 14.0–16.0 5.0–7.0 0.030 0.030 Custom 455 (XM-16) S45500 0.05 0.50 0.50 11.0–12.5 7.5–9.5 0.040 0.030 2.0–3.0 Mo; 0.75–1.50 Al 2.0–3.0 Mo; 0.75–1.5 Al 2.5–3.25 Mo; 0.07–0.13 N 2.5–3.25 Mo; 0.07–0.13 N 1.25–1.75 Cu; 0.5–1.0 Mo; 8  %C min Nb 1.5–2.5 Cu; 0.8–1.4 Ti; 0.1–0.5 Nb; 0.5 Mo a XM designations in this column are ASTM designations for the listed alloy. Single values are maximum values unless otherwise indicated. c Nominal compositions. d UNS 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. 841–907. b 43 content (0.01–0.02% 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 dry CO2 and sour gas and oil products in the petrochemical industry, as welded tubing for heat exchanges, for handling chloride-containing coolants, and for handling hot brines and organic chemicals in the chemical, electric, and other industries [4]. Martensitic stainless steels contain 11.5–18% Cr, 0.08–1.20% C, and other alloying elements less than 2 to 3%. They can be hardened and tempered to yield strength in the range of 550–1900 MPa (80–275 ksi). The Cr content provides these steels with such high hardenability that they can be air hardened even in large sections. If they are to be heat treated for maximum 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, steam valves, 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, drive screws, 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 good corrosion resistance that are developed by a simple heat treatment comprising martensite formation and low-temperature aging (or tempering) treatment; the latter heat treatment step may be applied after fabrication. PH stainless steels can have a matrix structure of either 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 Table 1.11). A majority of these steels are classified by a three-digit number in the AISI 400 series or by a five-digit UNS designation. However, most of them are better known by their 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 used and 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 steels have a predominantly austenitic structure at the solution-annealing temperature, but they undergo 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 for increased formability before the hardening treatment. Important alloys are 17-7PH (UNS S17700) and PH15-7Mo (UNS S15700). These alloys are completely austenite in the as-quenched condition after solution annealing (which displays good toughness and ductility in the cold-forming operations), and eventually martensite can be obtained by conditioning treatment or thermomechanical treatment. Ultrahigh strength can be obtained in these steels by combinations of cold working and aging. Austenitic PH stainless steels possess austenitic structures in both the solution annealed and aged conditions. The most important steels in this class include A-286 (AISI 600 or UNS S66286), 17-10P, and 14–17Cu–Mo alloys. Of these grades, A-286 is the most extensively used in the aerospace applications. 1.3.3.6 Maraging Steels Maraging steels are a specific class of carbon-free (or small amounts) ultrahigh-strength steels that derive their strength not from carbon but from precipitation of intermetallic ß 2006 by Taylor & Francis Group, LLC. TABLE 1.12 Nominal Compositions of Commercial Maraging Steels Composition (%)a Grade Standard grades 18Ni(200) 18Ni(250) 18Ni(300) 18Ni(350) 18Ni(Cast) 12-5-3(180)c Ni 18 18 18 18 17 12 Cobalt-free and low-cobalt bearing grades Cobalt-free 18Ni(200) 18.5 Cobalt-free 18Ni(250) 18.5 Low-cobalt 18Ni(250) 18.5 Cobalt-free 18Ni(300) 18.5 Mo Co Ti Al Nb 3.3 5.0 5.0 4.2b 4.6 3 8.5 8.5 9.0 12.5 10.0 — 0.2 0.4 0.7 1.6 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.3 — — — — — — 3.0 3.0 2.6 4.0 — — 2.0 — 0.7 1.4 1.2 1.85 0.1 0.1 0.1 0.1 — — 0.1 — a All grades contain no more than 0.03% C. Some producers use a combination of 4.8% Mo and 1.4% Ti, nominal. c Contains 5% Cr. Source: From K. Rohrbach and M. Schmidt, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 793–800. b compounds and martensitic transformation [5,50–52]. The commonly available maraging steels contain 10–19% Ni, 0–18% Co, 3–14% Mo, 0.2–1.6% Ti, 0.1–0.2% Al, and some intermetallic compounds are Ni3 Ti, Ni3 Mo, Fe2 Mo, etc. Since these steels develop very high strength by martensitic transformation and subsequent age-hardening, they are termed maraging steels [53]. There are four types of maraging steels, namely 200, 250, 300, and 350; the number refers to the ultimate tensile strength in ksi (kpsi). The tensile strength is based on the Ti content, which varies between 0.2 and 1.85%. Table 1.12 lists the compositions of these grades [54]. In these grades, 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 ultrahigh strength and high toughness are essential and cost is not a major concern. Maraging steels have 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 gear 2. 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 the machine tools, carbide die holders, autofrettage equipment, etc. 1.4 DESIGNATIONS FOR STEELS A 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 ß 2006 by Taylor & Francis Group, LLC. specific document as well as in a particular country. In the steel industries, these terms have very specific uses: grade is used to describe chemical composition; type is used to denote deoxidation practice; and class is used to indicate some other attributes such as tensile strength 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 indicates tensile properties. In ASTM A 515, grade describes strength level; the maximum carbon content allowed by the specification is dependent on both the plate thickness and the strength level. In ASTM A 533, type indicates chemical analysis, while class denotes strength level. In ASTM A 302, grade identifies requirements for both chemical composition and tensile properties. ASTM A 514 and A 517 are specifications for high-strength quenched and tempered alloy steel plate for structural and pressure vessel applications, respectively; each has a number of grades for identifying the chemical composition that is capable of developing the required mechanical properties. However, all grades of both designations have the same composition limits. By far the most widely used basis for classification and designation of steels is the chemical composition. The most commonly used system of designating carbon and alloy steels in the United 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 Materials Specification (AMS) and American Petroleum Institute (API) designation. These designation systems are discussed below. 1.4.1 SAE-AISI DESIGNATIONS As stated above, the SAE-AISI system is the most widely used designation for carbon and alloy steels. The SAE-AISI system is applied to semifinished forgings, hot-rolled and cold-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 both carbon and low-alloy steels. 1.4.1.1 Carbon and Alloy Steels With few exceptions, the SAE-AISI system uses a four-digit number to designate carbon and alloy steels that is specific for chemical composition ranges. Certain types of alloy steels are designated by five digits (numerals). Table 1.2 shows an abbreviated listing of four-digit designations of the SAE-AISI carbon and alloy steels. The first digit, 1, of this designation indicates a carbon steel; i.e., carbon steels comprise 1xxx groups in the SAE-AISI system and are subdivided into four series due to the variance in certain fundamental properties among them. Thus, the plain carbon steels comprise 10xx series (containing 1.00% Mn maximum); resulfurized carbon steels comprise the 11xx series; resulfurized and rephosphorized carbon steels comprise the 12xx series; and nonresulfurized high-manganese (up to 1.65%) carbon steels are produced for applications requiring good machinability. Carbon and alloy steel designations showing the letter B inserted between the second and third digits indicate that the steel has 0.0005–0.003% boron. Likewise, the letter L inserted between the second and third digits indicates that the steel has 0.15–0.35% lead for enhanced machinability. Sometimes the prefix M is used for merchant quality steels and the suffix H is used to comply with specific hardenability requirements. In alloy steels, the prefix letter E is used 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 the designation (Table 1.2). Thus, a first digit of 2 denotes a nickel steel; 3, a nickel–chromium ß 2006 by Taylor & Francis Group, LLC. steel; 4, a molybdenum, chromium–molybdenum, nickel–molybdenum, or nickel–chromium– molybdenum steel; 5, a chromium steel; 6, a chromium–vanadium steel; 7, a tungsten– chromium steel; 8, a nickel–chromium–molybdenum steel; and 9, a silicon–manganese steel or a nickel–chromium–molybdenum steel. In the case of a simple alloy steel, the second digit represents 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-numeral designations indicate the approximate carbon content of the allowable carbon range in hundredths of a percent. For example, 1020 steel indicates a plain carbon steel with an approximate mean of 0.20% carbon, varying within acceptable carbon limits of 0.18 and 0.23%. Similarly, 4340 steels are Ni–Cr–Mo steels and contain an approximate mean of 0.40% carbon, varying within an allowable carbon range of 0.38–0.43%, and 51100 steel is a chromium steel with an approximate mean of 1.00% carbon, varying within an acceptable carbon range of 0.98–1.10% [4,30,55]. Potential standard steels are listed in SAE J1081 and Table 1.13. They are experimental steels to which no regular AISI-SAE designations have been assigned. The numbers consist of the prefix PS followed by a sequential number starting with 1. Some were developed to minimize the amount of nickel and others to enhance a particular attribute of a standard grade of alloy steel [30]. 1.4.1.2 HSLA Steels Several grades of HSLA steels have been described in the SAE Recommended Practice J410. Their chemical composition and minimum mechanical property requirements are provided in Table 1.8 [30]. 1.4.1.3 Formerly Listed SAE Steels A number of grades of carbon and alloy steels have been excluded from the list of standard SAE steels because of their inadequate applications. A detailed list of formerly used SAE carbon and alloy steels is given in SAE J1249, and producers of these steels should be contacted for their availability. 1.4.2 UNS DESIGNATIONS The UNS has been developed by the ASTM E 527, the SAE J1086, and several other technical societies, trade associations, and U.S. government agencies [29]. A UNS number, which is a designation of chemical composition and not a specification, is assigned to each chemical composition of the standard carbon and alloy steel grades for which controlling limits have been 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 that class. The prefix letter G signifies standard grades of carbon and alloy steels; the prefix letter H indicates standard grades that meet certain hardenability requirement limits (SAE-AISI H steels); 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 identifies miscellaneous steels and ferrous alloys; and the prefix W denotes welding filler metals (for example, W00001–W59999 series represent a wide variety of steel compositions) [56]. The first four 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 ß 2006 by Taylor & Francis Group, LLC. 48 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.13 SAE Potential Standard Steel Compositions Ladle Chemical Composition Limits (wt %) a SAE PS Number Mn P max 0.19–0.24 0.18–0.23 0.20–0.25 0.23–0.28 0.25–0.30 0.18–0.23 0.13–0.18 0.15–0.20 0.18–0.23 0.13–0.18 0.15–0.20 0.18–0.23 0.17–0.24 0.28–0.33 0.38–0.43 0.95–1.25 0.90–1.20 0.90–1.20 0.90–1.20 0.90–1.20 0.90–1.20 0.90–1.20 0.90–1.20 0.75–1.00 0.70–0.90 0.70–0.90 0.70–0.90 0.85–1.25 0.90–1.20 0.90–1.20 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 S max 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 Si Ni 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.20–0.40 — — — — — — — — 0.70–1.00 0.70–1.00 0.70–1.00 0.20 min — — Cr 0.25–0.40 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.40–0.60 0.45–0.65 0.45–0.65 0.45–0.65 0.45–0.65 0.20 min 0.40–0.60 0.45–0.65 Mo B 0.05–0.10 0.13–0.20 0.13–0.20 0.13–0.20 0.13–0.20 0.08–0.15 0.13–0.20 0.13–0.20 0.20–0.30 0.45–0.60 0.45–0.60 0.45–0.60 0.05 min 0.13–0.20 0.13–0.20 — — — — — 0.0005–0.003 — — — — — — — — — Steel Heat Treatment: Metallurgy and Technologies PS 10 PS 15 PS 16 PS 17 PS 18 PS 19 PS 20 PS 21 PS 24 PS 30 PS 31 PS 32 PS 33b PS 34 PS 36 C Steel Nomenclature ß 2006 by Taylor & Francis Group, LLC. PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS 38 39 40 54 55 56 57 58 59 61 63 64 65 66c 67 0.43–0.48 0.48–0.53 0.51–0.59 0.19–0.25 0.15–0.20 0.080–0.13 0.08 max 0.16–0.21 0.18–0.23 0.23–0.28 0.31–0.38 0.16–0.21 0.21–0.26 0.16–0.21 0.42–0.49 0.90–1.20 0.90–1.20 0.90–1.20 0.70–1.05 0.70–1.00 0.70–1.00 1.25 max 1.00–1.30 1.00–1.30 1.00–1.30 0.75–1.10 1.00–1.30 1.00–1.30 0.40–0.70 0.80–1.20 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.035 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 1.00 max 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 0.15–0.35 — — — — 1.65–2.00 1.65–2.00 — — — — — — — 1.65–2.00 — 0.45–0.65 0.45–0.65 0.45–0.65 0.40–0.70 0.45–0.65 0.45–0.65 17.00–19.00 0.45–0.65 0.70–0.90 0.70–0.90 0.45–0.65 0.70–0.90 0.70–0.90 0.45–0.75 0.85–1.20 0.13–0.20 0.13–0.20 0.13–0.20 0.05 min 0.65–0.80 0.65–0.80 1.75–2.25 — — — — — — 0.08–0.15 0.25–0.35 — — — — — — — — — — 0.0005–0.003 — — — — a Some PS steels may be supplied to a hardenability requirement. Supplied to a hardenability requirement of 15 HRC points within the range of 23–43 HRC at J4 (4/16 in. distance from quenched end), subject to agreement between producer and user. c PS 66 has a vanadium content of 0.10–0.15%. Source: From Numbering System, Chemical Composition, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 1.01–1.189. b 49 composition requirements, such as boron, lead, or nonstandard chemical ranges. Table 1.3 and Table 1.4 list the UNS numbers corresponding to SAE-AISI numbers for various standard carbon and alloy steels, respectively, with composition ranges. 1.5 SPECIFICATIONS FOR STEELS A specification is typically an acronym or abbreviation for a standards organization plus a specific written statement of both technical and commercial requirements that a product must satisfy. It is a document that restrains or controls procurement and is issued by that standards organization. All material specifications contain general and specific information [57]. Any reasonably 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 supplier or the user. An informative title and a statement of the required form may be employed instead 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 additional prerequisites might be necessary. It may also include the type of steel and the steelmaking processes allowed. Quantitative requirements recognize permissible composition ranges and all physical and mechanical properties necessary to characterize the material. Testing methods employed to check 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 finish on flat-rolled products as well as special packaging, identification, and loading instructions. Engineering societies, trade associations, and institutes whose members make, specify, or purchase steel products publish standard specifications; many of them are well recognized and 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 particular specification-writing group is limited to its own specialized field. Organization Association of American Railroads American Bureau of Shipbuilding Aerospace Materials Specification (of SAE) American National Standards Institute American Petroleum Institute American Railway Engineering Association American Society of Mechanical Engineers American Society for Testing and Materials American Welding Society Society of Automotive Engineers 1.5.1 Acronym AAR ABS AMS ANSI API AREA ASME ASTM AWS SAE ASTM (ASME) SPECIFICATIONS The most widely used standard specifications for steel in the United States are those published by ASTM, many of which are complete specifications, usually adequate for procurement ß 2006 by Taylor & Francis Group, LLC. purposes. These specifications frequently apply to specific products, which are usually oriented toward the performance of the fabricated end product. They begin with the prefix ASTM, followed by letter A, identifying a ferrous material, then a number indicating the actual specification, which may be followed by letters or numbers subdividing the material by analysis. The AISI code is sometimes used for this purpose. Finally the year of 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 by analysis. ASTM specifications represent a consensus drawn from producers, specifiers, fabricators, and users of steel mill products. In many cases, the dimensions, tolerances, limits, and restrictions in the ASTM specifications are the same as the corresponding items of the standard practices in the AISI steel product manuals. Many of the ASTM specifications have been adopted by the American Society of Mechanical Engineers (ASME) with slight or no modifications. ASME uses the prefix S with the ASTM specifications; for example, ASME SA 213 and ASTM A 213 are the same. Steel products can be distinguished by the ASTM specification number, which denotes their method of production. Sometimes, citing the ASTM specification is not sufficient to completely 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 this specification, the grade/AISI-SAE designation and class (the required strength level) must also be quoted. The ASTM specification A 434 also covers, by reference, two standards for test methods (A 370 for mechanical testing and E 112 for grain size determination) and A 29 specifying general requirements for bar products. SAE-AISI designations for the chemical compositions of carbon and alloy steels are sometimes 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 specify the 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 SPECIFICATIONS AMS, published by SAE, are procurement documents, not design specifications. The majority of the AMS pertain to materials intended for aerospace applications. These specifications generally include mechanical property requirements and limits that are significantly more severe than those for materials or steel grades with identical compositions but meant for nonaerospace applications. Their compliance will ensure procurement of a specific form and condition or a specific material (or steel grade) or process. Table 1.15 and Table 1.16 show the AMS 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 similar MIL or federal (FED) specifications [58]. 1.5.3 MILITARY AND FEDERAL SPECIFICATIONS MIL specifications and standards are produced and adopted by the U.S. Department of Defense. MIL specifications are used to define materials, products, and services. MIL standards provide procedures for design, manufacturing, and testing instead of giving only a particular material description. MIL specifications begin with the prefix MIL, followed by a ß 2006 by Taylor & Francis Group, LLC. 52 ß 2006 by Taylor & Francis Group, LLC. TABLE 1.14 ASTM Specifications That Cover SAE-AISI Designations A 29 A 108 A 295 A 304 A 322 A 331 A 434 A 506 A 507 A 510 A 534 A 535 A 544 A 545 A 546 A 547 A 548 A 549 A 575 A 576 Carbon steel wire rods and coarse round wire Carburizing steels for antifriction bearings Special quality ball and roller bearing steel Scrapless nut quality carbon steel wire Cold-heading quality carbon steel wire for machine screws Cold-heading quality medium high carbon steel wire for hexagon-head bolts Cold-heading quality alloy steel wire for hexagon-head bolts Cold-heading quality carbon steel wire for tapping or sheet metal screws Cold-heading quality carbon steel wire for wood screws Merchant quality hot-rolled carbon steel bars Special quality hot-rolled carbon steel bars A 646 A 659 A 682 A 684 A 689 A 711 A 713 A 752 A 827 A 829 A 830 Premium quality alloy steel blooms and billets for aircraft and aerospace forgings Commercial quality hot-rolled carbon steel sheet and strip Cold-rolled spring quality carbon steel strip, generic Untempered cold-rolled high-carbon steel strip Carbon and alloy steel bars for springs Carbon and alloy steel blooms, billets, and slabs for forging High-carbon spring steel wire for heat-treated components Alloy steel wire rods and coarse round wire Carbon steel plates for forging and similar applications Structural quality alloy steel plates Structural quality carbon steel plates Source: From Anon., Carbon and alloy steels, SAE J411, 1993 SAE Handbook, Vol. 1, Materials Society of Automotive Engineers, Warrendale, PA, pp. 2.01–2.04. Steel Heat Treatment: Metallurgy and Technologies A 505 Carbon and alloy steel bars, hot rolled and cold finished Standard quality cold-finished carbon steel bars High carbon–chromium ball and roller bearing steel Alloy steel bars having hardenability requirements Hot-rolled alloy steel bars Cold-finished alloy steel bars Hot-rolled or cold-finished quenched and tempered alloy steel bars Hot-rolled and cold-rolled alloy steel sheet and strip Regular quality hot-rolled and cold-rolled alloy steel sheet and strip Drawing quality hot-rolled and cold-rolled alloy steel sheet and strip TABLE 1.15 AMS Number, Title of Specification, and Equivalent UNS Number, Proprietary/AISI-SAE Alloy, and Similar Specification for Wrought Carbon Steels AMS No. 5100H 5020C Title of Specification 5030F 5031C Bars, screw stock, free machining, cold drawn Bars, forgings, and tubing, 1.5Mn 0.25Pb (0.32–0.39C), free cutting Bars, forgings, and tubing, 0.14–0.20C, free cutting Bars, forgings, and tubing, 1.5Mn (0.32–0.39C), free cutting Wire, welding, 1.05Cr 0.55Ni 1.0Mo 0.07V (0.26–0.32C), vacuum melted, environmentcontrolled packaging Wire, welding, 1.05Cr 0.55Ni 1.0Mo 0.07V (0.34–0.40C), vacuum melted, environmentcontrolled packaging Wire, welding, 0.78Cr l.8Ni 0.35Mo 0.20V (0.33–0.38C), vacuum melted, environmentcontrolled packaging Wire, welding, 0.06 carbon maximum Welding electrodes, covered, steel, 0.07–0.15C 5032E 5036G 5040J 5022L 5024F 5027C 5028B 5029B 5042J 5044G 5045F 5046A 5047D 5050J 5053G 5060F 5061D 5062E 5069E 5070G 5075E 5077E 5080H UNS No. Alloy G12120 G11374 1212 111.37 G11170 G11370 1117 1137 K24728 D6AC K23725 Similar Specification D6AC K23577 K00606 W06013 S6013 Wire, 0.18–0.23C, annealed Sheet and strip, aluminum coated, low carbon G10200 1020 Sheet and strip, 0.15 carbon maximum, deep drawing grade Sheet and strip, 0.15 carbon maximum, forming grade Sheet and strip, 0.15 carbon maximum, half hard temper Sheet and strip, 0.25 carbon maximum, hard temper Sheet, strip, and plate, annealed G10100 1010 G10100 1010 G10100 1010 G10200 G10200 G10250 G10100 1020 1020 1025 1010 G10100 G10100 G10150 K00802 K02508 1010 1010 1015 G10180 G10220 G10250 1018 1022 1025 MIL-T-5066 G10250 1025 MIL-T-5066 G10350 1035 Sheet and strip, 0.08–0.13C, Al killed, deep-forming grade Tubing, seamless, 0.15 carbon maximum, annealed Tubing, welded, 0.13 carbon maximum, annealed Bars, forgings, and tubing, 0.13–0.18C Bars and wire, low carbon Bars, forgings, tubing, sheet, strip, and plate, low carbon Bars, forgings, and tubing, 0.15–0.20C Bars and forgings, 0.18–0.23C Tubing, seamless, 0.22–0.28C, cold drawn and stress relieved Tubing, welded, 0.22–0.28C, normalized or stress relieved Bars, forgings, and tubing, 0.31–0.38C FED-QQ-E-450, Type 6013 FED-QQ-W-461 MIL-S-4174, Type 1, Grade B MIL-S-7952 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.15 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Proprietary/AISI-SAE Alloy, and Similar Specification for Wrought Carbon Steels AMS No. UNS No. Alloy G10350 G10500 G10800 G10900 G10700 1070 G10740 G10950 G10950 1074 1095 1095 5132G Bars, 0.90–1.30C G10950 Similar Specification 1035 1050 1080 1090 5120J 5121G 5122G Tubing, seamless, 0.31–0.38C, stress relieved Sheet, strip, and plate, 0.47–0.55C, annealed Wire, carbon, spring temper, cold drawn, 0.75–0.88C Wire, spring quality music wire, 0.70–1.00C, cold drawn Wire, valve spring quality, 0.60–0.75C, hardened and tempered Strip, 0.68–0.80C Sheet and strip, 0.90–1.40C Strip, 0.90–l.04C, hard temper 1095 5082E 5085D 5110F 5112J 5115G Title of Specification MIL-S-7947 MIL-S-7947, hard temper Source: 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 then the 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 in Table 1.15 and Table 1.16. Federal (QQ) specifications are identical to the MIL, except that they are provided by the General Services Administration (GSA) and are used by federal agencies as well as by MIL establishments when there are no separate MIL specifications available. Federal specifications begin with the prefix FED-QQ, followed by the letter and code numbers. Examples of federal 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 SPECIFICATIONS The API fosters the development standards, codes, and safe practices within the petroleum industry. 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 producing operations), 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 for dimensions, weight, lengths, strengths, threaded ends, plain ends, belled ends, and thread protectors, 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 and higher. It provides the standards for pipe suitable for use in conveying gas, water, and oil on both the oil and natural gas industries [60]. API Spec 5LC covers seamless, centrifugal cast, and welded corrosion-resistant alloy line pipe (austenitic stainless steels, martensitic stainless steels, ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6250H 6255A 6256A 6257 6260L 6263H 6264G 6265H 6266G 6267D 6270L 6272H 6274L 6275F 6276F Title of Specification UNS No. Alloy Bars, forgings, and tubing, 1.5Cr 3.5Ni (0.07–0.13C) Bars, forgings, and tubing, 1.1Si 1.45Cr 1.0Mo 0.08A1 (0.16– 0.22C), premium air quality, double vacuum melted Bars, forgings, and tubing, 1.0Cr 3.0Ni 4.5Mo 0.08A1 0.38V (0.10–0.16C), premium air quality, double vacuum melted Bars, forgings, and tubing, 1.6Si 0.82Cr 1.8Ni 0.40Mo 0.08V (0.40–0.44C), consumable Electrode vacuum remelted, normalized and tempered Bars, forgings, and tubing, carburizing grade, 1.2Cr 3.2Ni 0.12Mo (0.07–0.13C) Bars, forgings, and tubing, carburizing grade, 1.2Cr 3.2Ni 0.12Mo (0.11–0.17C) Bars, forgings, and tubing, carburizing grade, 3.2Ni 1.2Cr 0.12Mo (0.14–0.20C) Bars, forgings, and tubing, 1.2Cr 3.25Ni (0.07–0.13C), vacuum consumable electrode remelted Bars, forgings, and tubing, 0.50Cr 1.82Ni 0.25Mo 0.003B 0.06V (0.08–0.13C) Bars, forgings, and tubing, 1.2Cr 3.25Ni 0.12Mo (0.07–0.13C), electroslag remelted or vacuum remelted, consumable electrode Bars, forgings, and tubing, 0.5Cr 0.55Ni 0.20Mo (0.11–0.17C) Bars, forgings, and tubing, 0.50Cr 0.55Ni 0.20Mo (0.15–0.20C) Bars, forgings, and tubing, 0.50Cr 0.55Ni 0.20Mo (0.18–0.23C) Bars, forgings, and tubing, 0.40Cr 0.45Ni 0.12Mo 0.002B (0.15– 0.20C) Bars, forgings, and tubing, 0.50Cr 0.55Ni 0.20Mo (0.18–0.23C), consumable electrode vacuum melted K44910 3310 K21940 CBS 600 K71350 CBS 1000M G93106 9310 G93150 9315 K44414 9317 G93106 9310 K21028 43BV12 G93106 9310 G86150 8615 G86170 8617 G86200 8620 G94171 94B17 G86200 Similar Specification 8620 MIL-S-7393, Composition I Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6277D 6278A 6280H 6281G 6282G 6290F 6292F 6294F 6299C 6300C 6302E 6303E 6304G MAM 6304 6305B Alloy Similar Specification Title of Specification UNS No. Bars, forgings, and tubing, 0.50Cr 0.55Ni 0.20Mo (0.18–0.23C), vacuum arc or electroslag remelted Bars, forgings, and tubing, 4.1Cr 3.4Ni 4.2Mo 1.2V (0.11–0.15C), premium aircraft quality for bearing applications, double vacuum melted Bars, forgings, and rings, 0.50Cr 0.55Ni 0.20Mo (0.28–0.33C) Tubing, mechanical, 0.50Cr 0.55Ni 0.20Mo (0.28–0.33C) Tubing, mechanical, 0.50Cr 0.55Ni 0.25Mo (0.33–0.38C) Bars and forgings, carburizing grade, 1.8Ni 0.25Mo (0.11–0.17C) Bars and forgings, carburizing grade, 1.8Ni 0.25Mo (0.14–0.20C) Bars and forgings, carburizing grade, 1.8Ni 0.25Mo (0.17–0.22C) Bars, forgings, and tubing, 0.50Cr 1.8Ni 0.25Mo (0.17–0.23C) Bars and forgings, 0.25Mo (0.35–0.40C) Bars, forgings, and tubing, low alloy, heat resistant, 0.65Si 1.25Cr 0.50Mo 0.25V (0.28–0.33C) Bars and forgings, low alloy, heat resistant, 0.65Si 1.25Cr 0.50Mo 0.85V (0.25–0.30C) Bars, forgings, and tubing, low alloy, heat resistant, 0.95Cr 0.55Mo 0.30V (0.40–0.50C) G86200 8620 G86300 8630 G86300 8630 G87350 8735 G46150 4615 G46170 4617 G46200 4620 H43200 4320H G40370 4037 K23015 17-22A(S) K22770 17-22A(V) K14675 17-22A MIL-S-24502 Bars, forgings, and tubing, low alloy, heat resistant, 0.95Cr 0.55Mo 0.30V (0.40–0.50C) Bars, forgings, and tubing, low alloy, heat resistant, 0.95Cr 0.55Mo 0.30V (0.40–0.50C), vacuum arc remelted K14675 17-22A MIL-S-24502 K14675 17–22A MIL-S-6050 MIL-S-7493, Composition 4615 MIL-S-7493 Composition 4617 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6308A 6312E 6317F 6320J 6321D 6322K 6323H 6324E 6325F 6327G 6328H 6330E 6331 6342H 6348A 6349B 6350H 6351E 6352F Alloy Similar Specification Title of Specification UNS No. Bars and forgings, 0.90Si 1.0Cr 2.0Ni 3.2Mo 2.0Cu 0.10V (0.07–0.13C), vacuum arc or electroslag remelted Bars, forgings, and tubing, 1.8Ni 0.25Mo (0.38–0.43C) Bars and forgings, 1.8Ni 0.25Mo (0.38–0.43C), heat treated, 125 ksi (862 MPa) tensile strength Bars, forgings, and rings, 0.50Cr 0.55Ni 0.25Mo (0.33–0.38C) Bars, forgings, and tubing, 0.42Cr 0.30Ni 0.12Mo 0.003B (0.38–0.43C) Bars, forgings, and rings, 0.50Cr 0.55Ni 0.25Mo (0.38–0.43C) Tubing, mechanical, 0.50Cr 0.55Ni 0.25Mo (0.38–0.43C) Bars, forgings, and tubing, 0.65Cr 0.70Ni 0.25Mo (0.38–0.43C) Bars and forgings, 0.50Cr 0.55Ni 0.25Mo (0.38–0.43C), heat treated, 105 ksi (724 MPa) tensile strength Bars and forgings, 0.50Cr 0.55Ni 0.25Mo (0.38–0.43C), heat treated, 125 ksi (862 MPa) tensile strength Bars, forgings, and tubing, 0.50Cr 0.55Ni 0.25Mo (0.48–0.53C) Bars, forgings, and tubing, 0.65Cr 1.25Ni (0.33–0.38C) Wire, welding, 0.50Cr 0.55Ni 0.20Mo (0.33–0.38C), vacuum melted, environment controlled packaging Bars, forgings, and tubing, 0.80Cr 1.0Ni 0.25Mo (0.38–0.43C) Bars, 0.95Cr 0.20Mo (0.28–0.33C), normalized Bars, 0.95Cr 0.20Mo (0.38–0.43C), normalized Sheet, strip, and plate, 0.95Cr 0.20Mo (0.28–0.33C) Sheet, strip, and plate, 0.95Cr 0.20Mo (0.28–0.33C), spheroidized Sheet, strip, and plate, 0.95Cr 0.20Mo (0.33–0.38C) K71040 Pyrowear, alloy 53 K22440 4640 K22400 4640 G87350 8735 K03810 81B40 G87400 8740 G87400 8740 K11640 8740 Mod G8740 8740 MIL-S-6049 G8740 8740 MIL-S-6049 K13550 8750 MIL-S-6049 K22033 G87350 8735 G98400 9840 G41300 4130 MIL-S-6758 G41400 4140 MIL-S-5626 G41300 4130 MIL-S-18729 G41300 4130 G41350 4135 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6354D 6356D 6357G 6358F 6359F 6360J 6361C 6362D 6365H 6370K 6371H 6372H 6373C 6374A 6375 6378E 6379A 6381E Alloy Similar Specification Title of Specification UNS No. Sheet, strip, and plate, 0.75Si 0.62Cr 0.20Mo 0.10Zr (0.10–0.17C) Sheet, strip, and plate, 0.95Cr 0.20Mo (0.30–0.35C) Sheet, strip, and plate, 0.50Cr 0.55Ni 0.25Mo (0.33–0.38C) Sheet, strip, and plate, 0.50Cr 0.55Ni 0.25Mo (0.38–0.43C) Sheet, strip, and plate, 0.80Cr 1.8Ni 0.25Mo (0.38–0.43C) Tubing, seamless, 0.95Cr 0.20Mo (0.28–0.33C), normalized or stress relieved Tubing, seamless round, 0.95Cr 0.20Mo (0.28–0.33C), 125 ksi (860 MPa) tensile strength Tubing, seamless, 0.95Cr 0.20Mo (0.28–0.33C), 150 ksi (1034 MPa) tensile strength Tubing, seamless, 0.95Cr 0.20Mo (0.33–0.38C), normalized or stress relieved Bars, forgings, and rings, 0.95Cr 0.20Mo (0.28–0.33C) Tubing, mechanical, 0.95Cr 0.20Mo (0.28–0.33C) Tubing, mechanical, 0.95Cr 0.20Mo (0.33–0.38C) Tubing, welded, 0.95Cr 0.20Mo (0.28–0.33C) Tubing, seam-free, round, 0.95Cr 0.20Mo (0.28–0.33C), 95 ksi (655 MPa) tensile strength Wire, welding, 0.50Cr 0.55Ni 0.20Mo (0.18–0.23C), vacuum melted, environment controlled packaging Bars, 1.0Cr 0.20Mo 0.045Se (0.39–0.48C), die drawn, 130 ksi (896 MPa) yield strength, free machining Bars, die drawn, 0.95Cr 0.20Mo 0.05Te (0.40–0.53C), tempered, 165 ksi (1140 MPa) yield strength Tubing, mechanical, 0.95Cr 0.20Mo (0.38–0.43C) K11914 NAX 9115-AC G41320 4132 G87350 8735 G87400 8740 G43400 4340 G41300 4130 MIL-T-6736 Condition N G41300 4130 MIL-T-6736 G41300 4130 G41350 4135 MIL-T-6736 Condition HT MIL-T-6735 G41300 4130 MIL-S-6758 G41300 4130 MIL-T-6736 G41350 4135 G41300 4130 G41300 4130 G86200 8620 K11542 4142H Mod K11546 4140 Mod G41400 4140 MIL-T-6736 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6382K 6385E 6386B 6390C 6395D 6396B 6406C 6407E 6408 6409 6411D 6412J 6413H 6414F 6415M MAM 6415 6417D Title of Specification UNS No. Alloy Bars, forgings, and rings, 0.95Cr 0.20Mo (0.38–0.43C) Sheet, strip, and plate, low alloy, heat resistant, 1.25Cr 0.50Mo 0.65Si 0.25V (0.27–0.33C) Sheet and plate, heat treated, 90 ksi and 100 ksi yield strength Tubing, mechanical, 0.95Cr 0.20Mo (0.38–0.43C) Sheet, strip, and plate, 0.95Cr 0.20Mo (0.38–0.43C) Sheet, strip, and plate, 0.80Cr 1.8Ni 0.25Mo (0.49–0.55C), annealed Sheet, strip, and plate, 2.1Cr 0.58Mo 1.6Si 0.05V (0.41– 0.46C), annealed Bars, forgings, and tubing, 1.2Cr 2.0Ni 0.45Mo (0.27–0.33C) Bars and forgings, tool, hotwork, 5.2Cr 1.5Mo 1.0V (0.35–0.45C), electroslag remelted (ESR) or consumable electrode vacuum arc remelted (VAR), annealed Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.25Mo (0.38–0.43C), special aircraft quality cleanliness, normalized and tempered Bars, forgings, and tubing, 0.88Cr 1.8Ni 0.42Mo 0.08V (0.28– 0.33C), consumable electrode remelted Bars and forgings, 0.80Cr 1.8Ni 0.25Mo (0.35–0.40C) Tubing, mechanical, 0.80Cr 1.8Ni 0.25Mo (0.35–0.40C) Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.25Mo (0.38–0.43C), vacuum consumable electrode remelted Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.25Mo (0.38–0.43C) Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.25Mo (0.38–0.43C) Bars, forgings, and tubing, 0.82Cr 1.8Ni 0.40Mo 1.6Si 0.08V (0.38–0.43C), consumable electrode remelted G41400 4140 K23015 17–22A/S K11856 Similar Specification — G41400 4140 G41400 MIL-S-5626 4140 K22950 K34378 X200 K33020 IIS-220 T20813 H-13 G43400 4340 K23080 4340 Mod G43370 4337 G43370 4337 G43400 4340 G43400 4340 MIL-S-5000 G43400 4340 MIL-S-5000 K44220 300M MIL-S-5000 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6418G 6419C 6421C 6422F 6423D 6424B 6425 6426D 6427H 6428D 6429D 6430D 6431J 6432A 6433D UNS No. Bars, forgings, tubing, and rings, 0.30Cr 1.8Ni 0.40Mo 1.3Mn 1.5Si (0.23–0.28C) Bars, forgings, and tubing, 0.82Cr 1.8Ni 0.40Mo 0.08V 1.6Si (0.40– 0.45C), consumable electrode vacuum remelted Bars, forgings, and tubing, 0.80Cr 0.85Ni 0.20Mo 0.003B (0.35–0.40C) Bars, forgings, and tubing, 0.80Cr 0.85Ni 0.20Mo 0.003B 0.04V (0.38–0.43C) Bars, forgings, and tubing, 0.92Cr 0.75Ni 0.52Mo 0.003B 0.04V (0.40–0.46C) Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.25Mo (0.49–0.55C) Bars, forgings, and tubing, 0.30Cr 1.8Ni 0.40Mo 1.4Mn 1.5Si (0.23–0.28C), consumable vacuum electrode remelted Bars, forging, and tubing, 1.0Cr 0.58Mo 0.75Si (0.80–0.90C), consumable electrode melted Bars, forgings, and tubing, 0.88Cr 1.8Ni 0.42Mo 0.08V (0.28–0.33C) Bars, forgings, and tubing, 0.80Cr 1.8Ni 0.35Mo 0.20V (0.32–0.38C) Bars, forgings, tubing, and rings, 0.78Cr 1.8Ni 0.35Mo 0.20V (0.33–0.38C), consumable electrode vacuum melted Bars, forgings, tubing, and rings, 0.78Cr 1.8Ni 0.35Mo 0.20V 0.75Mn (0.32–0.38C) Bars, forgings, and tubing, 1.05Cr 0.55Ni 1.0Mo 0.11V (0.45–0.50C), consumable electrode vacuum melted Bars, forgings, and tubing, 1.05Cr 0.55Ni 1.0Mo 0.12V (0.43–0.49C) Sheet, strip, and plate, 0.80Cr 1.8Ni 0.35Mo 0.20V 0.75Mn (0.33–0.38C) K32550 Hy-Tuf MIL-S-7108 K44220 300M MIL-S-8844 — Alloy Similar Specification Title of Specification 98B37 Mod K11940 98BV40 Mod K24336 98BV40 Mod K22950 — K32550 Hy-Tuf K18597 52CB K23080 4330 Mod K23477 4335 Mod K33517 4335 Mod K33517 4335 Mod K24728 D6 K24728 D6A K33517 4335 Mod MIL-S-8949 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6434D 6435C 6436B 6437D 6438D 6439B 6440J 6442E 6443E 6444H 6445E 6446C 6447D 6448F 6449C Alloy Similar Specification Title of Specification UNS No. Sheet, strip, and plate, 0.78Cr 1.8Ni 0.35Mo 0.20V (0.33–0.38C) Sheet, strip, and plate, 0.78Cr 1.8Ni 0.35Mo 0.20V (0.33–0.38C), vacuum consumable electrode melted, annealed Sheet, strip, and plate, low alloy, heat resistant, 0.65Si 1.25Cr 0.50Mo 0.85V (0.25–0.30C), annealed Sheet, strip, and plate, 5.0Cr 1.3Mo 0.50V (0.38–0.43C) Sheet, strip, and plate, 1.05Cr 0.55Ni 1.0Mo 0.12V (0.45–0.50C), consumable electrode vacuum melted Sheet, strip, and Plate, 1.05Cr 0.55Ni 1.0Mo 0.12V (0.42–0.48C), consumable electrode vacuum melted, annealed Bars, forgings, and tubing, 1.45Cr (0.98–1.10C), for bearing applications Bars and forgings, 0.50Cr (0.98–1.10C), for bearing applications Bars, forgings, and tubing, 1.0Cr (0.98–1.10C), consumable electrode vacuum melted Bars, forgings, and tubing, 1.45Cr (0.98–1.10C), premium aircraft quality, consumable electrode vacuum melted Bars, forgings, and tubing, 1.05Cr 1.1Mn (0.92–1.02C), consumable electrode vacuum melted Bars, forgings, and tubing, 1.0Cr (0.98–1.10C), electroslag remelted Bars, forgings, and tubing, 1.4Cr (0.98–1.10C), electroslag remelted Bars, forgings, and tubing, 0.95Cr 0.22V (0.48–0.53C) Bars, forgings, and tubing, 1.0Cr (0.98–1.10C), for bearing applications K33517 4335 Mod K33517 4335 Mod K22770 17–22A(V) T20811 H-H K24728 D6 K24728 D6AC G52986 52100 G50986 50100 G51986 51100 G52986 52100 K22097 51100 Mod G51986 51100 G52986 52100 G61500 6150 MIL-S-8503 G51986 51100 MIL-S-7420 MIL-S-8949 MIL-S-7420 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6450F 6451A 6452A 6453 6454B 6455G 6456A 6457A 6458F 6459B 6460D 6461G 6462F 6463B 6464E Alloy Similar Specification Title of Specification UNS No. Wire, spring, 0.95Cr 0.22V (0.48–0.53C), annealed and cold drawn Wire, spring, 1.4Si 0.65Cr (0.51–0.59C), oil tempered Wire, welding, 0.95Cr 0.20Mo (0.38–0.43C), vacuum melted, environment controlled packaging Wire, welding, 0.30 Cr 1.8Ni 0.40Mo (0.23–0.28C), vacuum melted, environment controlled packaging Sheet, strip, and plate, 1.8Ni 0.8Cr 0.25Mo (0.38–0.43C), consumable electrode melted Sheet, strip, and plate, 0.95Cr 0.22V (0.48–0.53C) Wire, welding, 0.8Cr 1.8Ni 0.25Mo (0.35–0.40C), vacuum melted, environment controlled packaging Wire, welding, 0.95Cr 0.20Mo (0.28–0.33C), vacuum melted, environment controlled packaging Wire, welding, 1.25Cr 0.50Mo 0.30V 0.65Si (0.28–0.33C), vacuum melted, environment controlled packaging Wire welding, 1.0Cr 1.0Mo 0.12V (0.18–0.23C), vacuum induction melted Wire, welding, 0.62Cr 0.20Mo 0.75Si 0.10Zr (0.10–0.17C) Wire, welding, 0.95Cr 0.20V (0.28–0.33C), vacuum melted, environment controlled packaging Wire, welding, 0.95Cr 0.20V (0.28–0.33C) Wire, welding, 18.5Ni 8.5Co 5.2Mo 0.72Ti 0.10A1, vacuum environment controlled packaging Electrodes, welding, covered, 1.05Mo 0.20V (0.06–0.12C) G61500 6150 G92540 9254 G43406 E4340 K 32550 Hy-Tuf G43400 4340 G61500 6150 MIL-S-18731 4340 Mod MIL-R-5632, Type III K13147 4130 MIL-R-5632, Type I K23015 17–22A(S) MIL-R-5632, Type II K22720 K11365 NAX-915-AC K13148 6130 K13149 6130 K93130 Mar 300 W10013 10013 (AWS) MIL-E-6843, Class E-10013 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6465B 6466D 6467C 6468B 6469A 6470J 6471D 6472C 6473 6475F 6476 6477 6485G 6487G 6488E Alloy Similar Specification Title of Specification UNS No. Wire, welding, 2.0Cr 10Ni 8.0Co 1.0Mo 0.02A1 0.06V (0.10–0.14C), vacuum melted, environment controlled pakaging Wire, welding, corrosion resistant, 5.2Cr 0.55Mo Electrode, welding, covered, 5Cr 0.55Mo Wire, welding, 1.0Cr 3.8Co 0.45Mo 0.08V (0.14–0.17C), vacuum melted, environment controlled packaging Wire, welding, 1.75Mn 0.80Cr 2.8Ni 0.85Mo (0.09–0.12C), vacuum melted, environment controlled packaging Bars, forgings, and tubing, nitriding grade, 1.6Cr 0.35Mo 1.1A1 (0.38– 0.43C) Bars, forgings, and tubing, nitriding grade, 1.6Cr 0.35Mo 1.2A1 (0.38– 0.43C), consumable electrode vacuum melted Bars and forgings, nitriding grade, 1.6Cr 0.35Mo 1.1 A1 (0.38– 0.43C), hardened and tempered, 112 ksi (772 MPa) tensile strength Wire, welding, 0.88Cr 1.8Ni 1.6Co 0.42Mo 0.08V (0.28–0.33C), vacuum melted, environment controlled packaging Bars, forgings, and tubing, nitriding grade, 1.1Cr 3.5Ni 0.25Mo 1.25A1 (0.21–0.26C) Bars, forgings, and tubing, 0.50Cr 0.12Mo (0.89–1.01C), for bearing applications Bars, forgings, and tubing, 0.80Cr (0.90–1.03C), for bearing applications Bars and forgings, 5.0Cr 1.3Mo 0.50V (0.38–0.43C) Bars and forgings, 5.0Cr 1.3Mo 0.50V (0.38–0.43C), consumable electrode vacuum melted Bars and forgings, 5.0Cr 1.3Mo 0.5V (0.38–0.43C) K91971 HY-180 S50280 Type 502 W50210 Type 502 K91461 HP 9-4-20 — — K24065 135 Mod K24065 135 Mod K24065 135 Mod MIL-S-6709 T20811 H-11 FED-QQ-T-570 Class H-11 T20811 H-11 T20811 H-11 MIL-S-6709 K52355 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6490D 6491A 6501A 6512C 6514C 6518A 6519A 6520B 6521A 6522A 6523C UNS No. Bars, forgings, and tubing, 4.0Cr 4.2Mo 1.0V (0.77–0.85C), premium aircraft quality for bearing applications, consumable electrode vacuum melted Bars, forgings, and tubing, 4.1Cr 4.2Mo 1.0V (0.80–0.85C), premium aircraft quality for bearing applications, Double vacuum melted Wire, welding, maraging steel, 18Ni 8.0Co 4.9Mo 0.40Ti 0.10A1, vacuum induction melted, environment controlled packaging Bars, forgings, tubing, and rings, 18Ni 7.8Co 4.9Mo 0.40Ti 0.10A1, consumable electrode melted, annealed Bars, forgings, tubing, and rings, Maraging, 18.5Ni 9Co 4.9Mo 0.65Ti 0.10A1, consumable electrode melted, annealed T11350 M-50 T11350 M-50 K92890 Maraging 250 K92890 Maraging 250 MIL-S-46850 Type III, Grade K93120 Maraging 300 MIL-S-46850 Type 300 MIL-S-13881, Type II, Class I K92890 Maraging 250 K93120 Maraging 300 K92571 AF-1410 K91472 HP 9–4–20 Sheet, strip, and plate, maraging, 19Ni 3.0Mo 1.4Ti 0.10A1, double vacuum melted, solution heat treated Bars, forgings, tubing, and rings, maraging, 19Ni 3.0Mo 1.4Ti 0.10A1, double vacuum melted, annealed Sheet, strip, and plate, maraging 250, 18Ni 7.8Co 4.9Mo 0.40Ti 0.10A1, consumable electrode melted, Solution heat treated Sheet, strip, and plate, 18.5NI 9.0Co 4.9Mo 0.65Ti 0.10A1, consumable electrode melted, solution heat treated plate, 2.0Cr 10Ni 14Co 1.0Mo (0.13–0.17C), vacuum melted, normalized and overaged Sheet, strip, and plate, 0.75Cr 9.0Ni 4.5Co 1.0Mo 0.09V (0.17–0.23C), vacuum consumable electrode melted, annealed Alloy Similar Specification Title of Specification MIL-S-46850 Grade 300 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.16 (Continued ) AMS Number, Title of Specification, and Equivalent UNS Number, Nearest Proprietary or AISI-SAE Grade, and Similar Specification for Wrought Alloy Steels AMS No. 6524C 6525A 6526C 6527B 6528 6529 6530H 6532 6533 6535G 6543B 6544B 6546D 6550H Title of Specification UNS No. Alloy Wire, welding, 1.0Cr 7.5Ni 4.5Co 1.0Mo 0.09V (0.29–0.34C), consumable electrode vacuum melted Bars, forgings, tubing, and rings, 0.75Cr 9.0Ni 4.5Co 1.0Mo 0.09V (0.17–0.23C), consumable electrode vacuum melted Bars, forgings, tubing, and rings, 1.0Cr 7.5Ni 4.5Co 1.0Mo 0.09V (0.29–0.34C), consumable electrode vacuum melted, annealed Bars and forgings, 2.0Cr 10Ni 14Co 1.0Mo (0.15–0.19C), vacuum melted, normalized and overaged Bars, 0.95Cr 0.20Mo (0.28–0.33C), special aircraft quality cleanliness, normalized Bars, 0.95Cr 0.20Mo (0.38–0.43C), special aircraft quality cleanliness, normalized Tubing, seamless, 0.50Ni 0.55Cr 0.20Mo (0.28–0.33C) Bars and forgings, 3.1Cr 11.5Ni 13.5Co 1.2Mo (0.21–0.25C), vacuum melted, annealed Wire, welding, 2.0Cr 10Ni 14Co 1.9Mo (0.13–0.17C), vacuum melted, environment controlled packaging Tubing, seamless, 0.50Cr 0.55Ni 0.20Mo (0.28–0.33C) Bars and forgings, 2.0Cr 10Ni 8.0Co 1.0Mo (0.10–0.14C), double vacuum melted, solution heat treated Plate, maraging, 2.0Cr 10Ni 8.0Co 1.0Mo (0.10–0.14C), double vacuum melted, heat treated Sheet, strip, and plate, 0.48Cr 0.80Ni 4.0Co 0.48Mo 0.09V (0.24–0.30C), consumable electrode melted, annealed Tubing, Welded, 0.55Cr 0.50Ni 0.20Mo (0.28–0.33C) K91313 HP 9–4–30 K91472 HP 9–4–20 K91283 HP 9–4–30 K92571 AF 1410 G41300 4130 G41400 4140 G86300 8630 K92580 Aermet 100 K92571 AF 1410 G86300 8630 K92571 Similar Specification AF 1410 K91970 K91122 HP 9–4–25 G86300 8630 Source: From 1994 SAE AMS Index, Society of Automotive Engineers, Warrendale, PA. ß 2006 by Taylor & Francis Group, LLC. MIL-T-6734 duplex stainless steels, and nickel-base alloys), dimensions, weights, process of manufacture, chemical and mechanical property requirements, and testing and inspection methods [61]. API Spec 5LD covers seamless, centrifugal cast, and welded clad steel line pipe and lined steel pipe with increased corrosion-resistant properties. The clad and lined steel line pipes are composed of a base metal outside and a corrosion resistent alloy (CRA) layer inside the pipe; the base material conforms to API Spec 5L, except as modified in the API Spec 5LC document. This specification provides standards for pipe with improved corrosion resistance suitable for use in conveying gas, water, and oil in both the oil and natural gas industries [62]. 1.5.5 ANSI SPECIFICATIONS ANSI standard begins with the prefix ANSI, followed by an alphanumeric code with an uppercase letter, subsequently followed by one to three digits and additional digits that are separated by decimal points. ANSI standards can also have a standard developer’s acronym in 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, and ANSI/API Spec 5LC-1991 [57,60–62]. 1.5.6 AWS SPECIFICATIONS AWS standards are used to support welding design, testing, quality assurance, and other related joining functions. These standards begin with the prefix AWS followed by the letter and 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). 1.6 INTERNATIONAL SPECIFICATIONS AND DESIGNATIONS Since steelmaking technology is available worldwide, familiarity with international specifications and designations for steels is necessary. Table 1.17 cross-references SAE steels with those of a selected group of international specifications and designations, which are described in the following paragraphs. More elaborate information on cross-referencing is available in Refs. [7,15,57,63]. 1.6.1 ISO DESIGNATIONS The International Organization for Standardization (ISO) system has standard designation for steel. 1.6.1.1 The Designation for Steels with Yield Strength The designations are preceded by the letter and followed by yield strength value (MPa). The prefix of nonalloy structural steel is letter S, for example, S235. The prefix of nonalloy engineering 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 of yield strength is 355–690 MPa, for example, E355 , . . . , . . . , E690, where E355 and E690 are different steel grades. ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Carbon steels 1005 1.0288, D5-2 1.0303, QSt32-2 1.0312, D5-1 1.0314, D6-2 1.0393, ED3 1.0394, ED4 1.1012, RFe120 1006 1.0311, D7-1 1.0313, D8-2 1.0317, RSD4 1.0321, St23 1.0334, StW23 1.0335, StW24 1.0354, St14Cu3 1.0391, EK2 1.0392, EK4 1.1009, Ck7 1008 1.0010, D9 1.0318, St28 1.0320, St22 1.0322, USD8 1.0326, RSt28 1.0330, St2, St12 1.0333, St3, St13 1.0331, RoSt2 1.0332, StW22 1.0336, USt4, USt14 1.0337, RoSt4 1.0344, St12Cu3 1.0347, RRSt13 1.0357, USt28 1.0359, RRSt23 1.0375, Feinstblech T57, T61, T65, T70 1.0385, Weissblech T57, T61, T65, T70 1.0744, 6P10 1.0746, 6P20 1.1116, USD6 1010 1.0204, UQSt36 1.0301, C10 1.0328, USD10 1.0349, RSD9 Japan (JIS) United Kingdom (BS) France (AFNOR NF) China (GB) ISO 05F — — — — 970 015A03 — — 970 030A04 970 040A04 970 050A04 A35-564 XC6FF G3445 STKM11A (11A) 1449 3CR 1449 3CS 1449 3HR 1449 3HS 1717 ERW101 3606 261 A35-551 XC10 XC6 XC6FF 08F 08 08A1 CC8X G4051 S10C G4051 S9Ck 1449 40F30, 43F35, 46F40, 50F45, A33-101 AF34 CC10 C10 10F 10 C10 CE10 C11x Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Japan (JIS) 1.1121, Ck10 1.1122, Cq10 1012 1.0439, RSD13 1013 1.0036, USt37-2 1.0037, St37-2 1.0038, RSt37-2 1.0055, USt34-1 1.0057, RSt34-1 1.0116, St37-3 1.0218, RSt41-2 1.0219, St41-3 1.0307, StE210.7 1.0309, St35.4 1.0315, St37.8 1.0319, RRStE210.7 1.0356, TTSt35 1.0417 1.0457, StE240.7 1.0401, C15 1.1132, CQ15 1.1135, Ck16A1 1015 G4051 S12C — G4051 F15Ck G4051 S15C United Kingdom (BS) 60F55, 68F62, 75F70 (available in HR, HS, CS conditions) 1449 4HR, 4HS, 4CR, 4CS 970 040A10 (En2A, En2A/1, En2B) 970 045A10, 045M10 (En32A) 970 050A10 970 060A10 980 CEW1 1449 12HS, 12CS 1501 141-360 970 040A12 (En2A, En2A/1, En2B) 970 050A12 970 060A12 3059 360 3061 360 3603 360 970 040A15 970 050A15 970 060A15 France (AFNOR NF) China (GB) A33-101 AF37 A35-551 XC12 C12 — — A35-551 XC12 CC12 — — XC15 15 ISO C15 C15E4 C15M2 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 1016 1017 Fed. R. of Germany (DIN) 1.1140, Cm15 1.1141, Ck15 1.1144 1.1148, Ck15A1 1.0419, RS144.2 1.0467, 15Mn3 1.0468, 15Mn3A1 1.1142, GS-Ck16 — 1018 1.0453, C16.8 1019 1020 — 1.0402, C22 1.0414, D20-2 1.0427, C22.3 1.0460, C22.8 1.1149, Cm22 1.1151, Ck22 1021 1022 — 1.0432, C21 1.0469, 21Mn4 1.0482, 19Mn5 1.1133, 20Mn5, GS-20Mn5 1.1134, Ck19 Japan (JIS) United Kingdom (BS) France (AFNOR NF) China (GB) ISO 970 080A15, 080M15 970 173H16 — G4051 S17C — — G4051 S20C G4051 S20CK 3059 440 3606 440 970 080A15, 080M15 970 170H15 970 173H16 1449 17HS, 17CS 970 040A17 970 050A17 970 060A17 970 080A17 — 970 040A20 970 050A20 (En2C, En2D) 970 060A20 — 970 070M20 970 080A20 — 3111 Type 9 970 120M19 970 170H20 15Mn A35-551 XC18 A35-552 XC18 A35-566 XC18 A35-553 XC18S A35-554 XC18S A33-101 AF42 C20 — A35-551 XC18 A35-552 XC18 A35-566 XC18 A35-553 C20 A35-553 XC18S A35-554 XC18S CC20 A35-551 21B3 A35-552 21B3 A35-553 21B3 A35-557 21B3 A35-566 21B3 A35-551 20MB5 A35-552 20 MB5 A35-553 20MB5 A35-556 20MB5 A35-557 20MB5 A35-566 20MB5 A35-566 20M5 — — — 20 C20 CC21K — — 20Mn — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Japan (JIS) 1023 1.1150, Ck22.8 1.1152, Cq22 G4051 S22C 1025 1.0406, C25 1.0415, D25-2, D26-2 1.1158, Ck25 1.1155, GS-Ck25 1.1156, GS-Ck24 G4051 S25C 1029 1.0562, 28Mn4 1030 1.0528, C30 1.0530, D30-2 1.1178, Ck30 1.1179, Cm30 1.1811, G-31Mn4 G3445 STKM15A (15A), STKM15C (15C) G4051 S28C G4051 S30C 1035 1.0501, C35 1.0516, D35-2 1.1172, Cq35 1.1173, Ck34 1.1180, Cm35 1.1181, Ck35 G4051 S35C 1037 1.0520, 31Mn4 1.0561, 36Mn4 G4051 S35C 1038 No international equivalents 1.1190, Ck42A1 1026 1039 — — United Kingdom (BS) 1449 2HS, 22CS 970 040A22 (En2C, En2D) 970 050A22 970 060A22 970 080A22 — 970 070M26 970 080A25 970 080A27 970 060A27 970 080A27 (En5A) France (AFNOR NF) China (GB) ISO — — — A35-552 XC25 A35-566 XC25 — 25 C25 C25E4 25Mn A33-101 AF50 CC28 C30 — C25M2 — — 1449 30HS, 30CS 970 060A30 970 080A30 (En5B) 970 080M30 (En5) 1717 CDS105/ 106 970 060A35 970 080A32 (En5C) 970 080A35 (En8A) 980 CFS6 3111 type 10 970 080M36 970 170H36 A35-552 XC32 A35-553 XC32 30 C30 C30 A33-101 AF55 A35-553 C35 A35-553 XC38 A35-554 XC38 XC35 XC38TS C35 35 C35 C35E4 C35M2 970 060A40 970 080A40 (En8C) 970 080M40 (En8) 40M5 A35-552 XC38H2 A35-553 38MB5 — 35Mn — 40Mn — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Japan (JIS) United Kingdom (BS) 970 170H41 1040 1.0511, C40 1.0541, D40-2 1.1186, Ck40 1.1189, Cm40 G4051 S40C 1042 1.0517, D45-2 G4051 S43C 1043 1.0558, GS-60.3 G4051 S43C 1044 1045 1.0517, D45-2 1.0503, C45 1.1184, Ck46 1.1191, Ck45, GS-Ck45 1.1192, Cq45 1.1194, Cq45 1.1201, Cm45 1.1193, Cf45 1.0503, C45 1.0519, 45MnAl 1.1159, GS-46Mn4 — G4051 S45C G5111 SCC5 1046 1049 — — G3445 STKM17A (17A) G3445 STKM17C (17C) 1287 1449 40HS, 40CS 3146 Class 1 Grade C 3146 Class 8 970 060A40 970 080A40 (En8C) 970 080M40 (En8) 970 060A42 970 080A42 (En8D) 970 060A42 970 080A42 (En8D) 970 080M46 — 970 060A47 970 080A47 970 080M46 3100 AW2 970 080M46 970 060A47 970 080A47 France (AFNOR NF) A35-556 38MB5 A35-557 XC38H2 XC42, XC42TS A33-101 AF60 C40 China (GB) 40 A35-552 XC42H1 A35-553 C40 CC45 XC42, XC42TS A35-552 XC42H2 — A33-101 AF65 A35-552 XC48H1 A35-553 XC45 A35-554 XC48 XC48TS C45 45M4TS A35-552 XC48H1 A35-552 XC48H2 XC48TS A35-552 XC48H1 A35-554 XC48 XC48TS ISO C40 C40E4 C40M2 — — — — — — C45 C45E4 C45M2 45 45Mn — — — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 1050 1053 1055 1059 1060 1064 1065 Fed. R. of Germany (DIN) 1.0540, C50 1.1202, D53-3 1.1206, Ck50 1.1210, Ck53 1.1213, Cf53 1.1219, Cf54 1.1241, Cm50 1.1210, Ck53 1.1213, Cf53 1.1219, Cf54 1.0518, D55-2 1.0535, C35 1.1202, D53-3 1.1203, Ck55 1.1209, Cm55 1.1210, Ck53 1.1213, Cf53 1.1219, Cf54 1.1220, D55-3 1.1820, C55W 1.0609, D58-2 1.0610, D60-2 1.0611, D63-2 1.1212, D58-3 1.1222, D63-3 1.1228, D60-3 1.0601, C60 1.0642, 60Mn3 1.1221, Ck60 1.1223, Cm60 1.1740, C60W 1.0611, D63-2 1.0612, D65-2 1.0613, D68-2 1.1222, D63-3 1.1236, D65-3 1.0627, C68 1.0640, 64Mn3 1.1230, Federstahldraht FD 1.1233 1.1240, 65Mn4 1.1250, Federstahldraht VD 1.1260, 66Mn4 Japan (JIS) G4051 S50C G4051 S53C United Kingdom (BS) China (GB) A35-553 XC50 50 52M4TS A35-553 XC54 50Mn 3100 AW3 970 060A57 970 070M55 970 080A52 (En43C) 970 080A57 A33-101 AF70 A35-552 XC55H1 A35-552 XC55H2 A35-553 XC54 XC55 C55 55 970 060A62 A35-553 XC60 1449 60HS 1449 60CS 970 060A57 970 080A57 A35-553 XC60 — 970 060A62 970 080A62 (En43D) — — 970 060A67 G4051 S53C G4051 S53C G4051 S55C — G4051 S58C 1549 50HS 1549 50CS 970 060A52 970 080A52 (En43C) 970 080M50 (En43A) 970 080A52 (En43C) France (AFNOR NF) XC65 C50 C50E4 C50M2 — C55 C55E4 — 60 — C60 C60E4 C60M2 — 65 ISO SL, SM Type DC 970 080A67 (En43E) Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 1069 1070 1074 1075 1078 1080 Fed. R. of Germany (DIN) Japan (JIS) United Kingdom (BS) France (AFNOR NF) 1.0615, D70-2 1.0617, D73-2 1.0627, C68 1.1232, D68-3 1.1237 1.1249, Cf70 1.1251, D70-3 1.1520, C70W1 1.1620, C70W2 1.0603, C67 1.0643, 70Mn3 1.1231, Ck67 — — A35-553 XC68 XC70 70 — XC70 70 Type DC 1.0605, C75 1.0645, 76Mn3 1.0655, C74 1.1242, D73-3 1.0614, D75-2 1.0617, D73-2 1.0620, D78-2 1.1242, D73-3 1.1252, D78-3 1.1253, D75-3 1.0620, D78-2 1.0622, D80-2 1.0626, D83-2 1.1252, D78-3 1.1253, D75-3 1.1255, D80-3 1.1262, D83-3 1.1525, C80W1 1.1259, 80Mn4 1.1265 D85-2 — A35-553 XC75 XC70 75 — A35-553-XC75 XC70 75 — — — G4801 SUP3 — 1084 1.1830, C85W — 1085 1.0647, 85Mn3 1.1273, 90Mn4 1.1819, 90Mn4 — 1449 70HS, 70CS 970 060A72 970 070A72 (En42) 970 080A72 970 070A72 (En42) 970 080A72 — 970 060A78 XC80 1449 80HS, 80CS 970 060A78 970 060A83 970 070A78 970 080A78 970 080A83 970 060A86 970 080A86 970 080A83 XC80 China (GB) — ISO — 80 — 85 — XC85 — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 1086 Fed. R. of Germany (DIN) Japan (JIS) 1141 1.0616, C85, D85-2 1.0626, D83-2 1.0628, D88-2 1.1262, D83-3 1.1265, D85-3 1.1269, Ck85 1.1272, D88-3 1.1273, 90Mn4 1.1819, 90Mn4 1.1282, D95S3 1.0618, D95-2 1.1274, Ck 101 1.1275, Ck100 1.1282, D95S3 1.1291, MK97 1.1545, C105W1 1.1645, C105W2 No international equivalents — G4804 SUM42 1144 1.0727, 45S20 G4804 SUM43 1146 1151 1.0727, 45S20 1.0728, 60S20 1.0729, 70S20 1090 1095 1140 12L14 1.0736, 9SMn36 — France (AFNOR NF) China (GB) 970 050A86 A35-553 XC90 — 1449 95HS 1449 95CS 970 060A96 1449 95HS 1449 95CS 970 060A99 — T9A — A35-553 XC100 T10 — A35-562 45MF4 Y40Mn G4801 SUP4 — — G4804 SUM23 — 970 212A42 (En8DM) 970 216A42 970 212A42 (En8DM) 970 212M44 970 216M44 970 225M44 970 226M44 970 212M44 — — 970 220M07 (En1A) 970 230M07 970 240M07 (En1A) 970 240M07 (En1B) — 85 ISO — Resulfurized/rephosphorized carbon steels 1211 No international equivalents 1212 1.0711, 9S20 G4804 SUM21 1.0721, 10S20 1.1011, RFe160K G4804 SUM22 1213 1.0715, 9SMn28 1.0736, 9SMn36 1.0740, 9SMn40 1215 United Kingdom (BS) Type DC 44SMn28 A35-562 45MF6 — 45MF4 — — — 10F2 12MF4 S200 A35-561 S250 S250 — — — A35-561 S300 — — — Y15Pb 44SMn28 — — 9S20 11SMnPb28 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Alloy steels 1330 — 1335 1.5069, 36Mn7 1340 1.5223, 42MnV7 1345 1.0625, StSch90C 1.0912, 46Mn7 1.0913, 50Mn7 1.0915, 50MnV7 1.5085, 51Mn7 1.5225, 51MnV7 4023 1.5416, 20Mo3 4024 1.5416, 20Mo3 4027 1.5419, 22Mo4 4028 — 4032 1.5411 4037 4042 4047 4118 4130 4135 1.2382, 43MnSiMo4 1.5412, GS-40MnMo4 3 1.5432, 42MnMo7 1.2382, 43MnSiMo4 1.5432, 42MnMo7 No international equivalents 1.7211, 23CrMoB4 1.7264, 20CrMo5 — 1.2330, 35CrMo4 1.7220, 34CrMo4 1.7220, GS34CrMo4 1.7226, 34CrMoS4 1.7231, 33CrMo4 Japan (JIS) United Kingdom (BS) France (AFNOR NF) — — — — — — — — — — — — 30Mn2 35Mn2 40Mn2 45Mn2 28Mn6 36Mn6 42Mn6 — — — 970 605M30 970 605A32 970 605H32 970 605M30 970 605M36 (En16) 3111 Type 2/1 3111 Type 2/2 970 605A37 970 605H37 — — — — — — — — — — — — — — — — — — — — 970 708H20 970 708M20 — — — — — G5111 SCMnM3 — — G4052 SCM15H G4105 SCM21H G4052 SCM418H G4105 SCM418H G4105 SCM1 G4105 SCM432 G4105 SCM2 G4105 SCM430 G4106 SCM2 G4054 SCM3H G4054 SCM435H G4105 SCM1 G4105 SCM432 G4105 SCM3 G4105 SCM435 1717 CDS110 970 708A30 970 708A37 970 708H37 A35-552 30CD4 A35-556 30CD4 A35-557 30CD4 35CD4 A35-552 35CD4 A35-553 35CD4 A35-556 35CD4 A35-557 34CD4 China (GB) G20 CrMo ISO 36Mo3E — 18CrMo4 18CrMo4E 30CrMo 30CrMoA — 35CrMo 35CrMoV 34CrMo4 34CrMo4E 34CrMoS4 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 4137 Fed. R. of Germany (DIN) 1.7225, GS42CrMo4 Carbon–manganese steels 1513 1.0424, Schiffbaustahl CS:DS 1.0479, 13Mn6 1.0496, 12Mn6 1.0513, Schiffbaustahl A32 1.0514, Schiffbaustahl B32 1.0515, Schiffbaustahl E32 1.0549 1.0579 1.0583, Schiffbaustahl A36 1.0584, Schiffbaustahl D36 1.0589, Schiffbaustahl E 36 1.0599 1.8941, QStE260N 1.8945, QStE340N 1.8950, QStE380N 1522 1.0471, 21MnS15 1.0529, StE350-Z2 1.1120, GS-20Mn5 1.1138, GS-21 Mn5 1.1169, 20Mn6 1.8970, StE385.7 1.8972, StE415.7 1.8978 1.8979 Japan (JIS) G4052 SCM4H G4052 SCM440H G4105 SCM4 — G4106 SMn21 United Kingdom (BS) France (AFNOR NF) China (GB) ISO — — 3100 type 5 970 708A37 970 708H37 970 709A37 40CD4 42CD4 A35-552 38CD4 A35-557 38CD4 1449 40/30 HR 1449 40/30 HS 1449 40/30 CS 1453 A2 2772 150M12 970 125A15 970 130M15 970 130M15 (En201) 12M5 A33-101 AF50-S A35-501 E35-4 A35-501 E36-2 A35-501 E36-3 15Mn — 1503 221–460 1503 223–409 1503 224–490 3146 CLA2 980 CFS7 A35-551 20MB5 A35-552 20M5 A35-556 20M5 A35-552 20MB5 A35-553 20MB5 A35-556 20MB5 A35-557 20MB5 A35-566 20MB5 20Mn — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) 1524 1.0499, 21Mn6A1 1.1133, 20Mn5, GS20Mn5 1.1160, 22Mn6 1526 — Japan (JIS) G4106 SMn21 G5111 SCMn1 — 1527 1.0412, 27MnSi5 1.1161, 26Mn5 1.1165, 30Mn5 1.1165, GS-30Mn5 1.1170, 28Mn6 G5111 SCMn2 1536 1.0561, 36Mn4 1.1165, 30Mn5 1.1165, GS-30Mn5 1.1166, 34Mn5 1.1167, 36Mn5, GS-36Mn5 1.1813, G-35Mn5 G4052 SMn1H G4052 SMn433H G4106 SMn1 G4106 SMn433 G5111 SCMn2 G5111 SCMn3 1541 1.0563, E 1.0564, N-80 1.1127, 36Mn6 1.1168, GS-40Mn5 1548 1.1128, 46Mn5 1.1159, GS-46Mn4 1.0542, StSch80 1.0624, StSch90B 1.1226, 52Mn5 1.0908, 60SiMn5 1.1233 1.1240, 65Mn4 1.1260, 66Mn7 G4106 SMn2, SMn438 G4052 SMn2H, SMn438H G4106 SMn3, SMn443 G4052 SMn3H, SMn443H G5111 SCMn5 — 1551 1552 1561 1566 United Kingdom (BS) 1456 Grade A 970 150M19 (En14A, En14B) 970 175H23 980 CDS9, CDS10 970 120M28 1453 A3 1456 Grade B1, Grade B2 3100 A5 3100 A6 970 150M28 (En14A, En14B) 1045 3100 A5, A6 970 120M36 (En15B) 970 150M36 (En15) 970 135M44 970 150M40 France (AFNOR NF) — China (GB) ISO 20Mn2 — A35-566 25MS5 — 25Mn — 30Mn — A35-552 32M5 A35-552 38MB5 A35-553 38MB5 A35-556 38MB5 A35-557 38MB5 40M5 45M5 A35-552 40M6 35Mn — — — — — — — 24M4TS 55M5 — — — — — — 40Mn2 SL, SM 45Mn SL, SM 50Mn2 SL, SM SL, SM 60Mn 65Mn SL, SM — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Resulfurized carbon steels 1108 1.0700, U7S10 1.0702, U10S10 1110 1.0703, R10S10 1117 — 1118 — G4804 SUM12 G4804 SUM11 G4804 SUM31 — 1137 Japan (JIS) 1139 1.3563, 43CrMo4 1.7223, 41CrMo4 1.7225, 42CrMo4 1.7225, GS-42CrMo4 1.7227, 42CrMoS4 4142 G4804 SUM41 1.0726, 35S20 4140 — — 1.3563, 43CrMo4 1.7223, 41CrMo4 G4052 SCM4H G4052 SCM440H G4103 SNCM4 G4105 SCM4 G4105 SCM440 — United Kingdom (BS) — — 970 210A15 970 210M17 (En32M) 970 214A15 970 214M15 (En202) 970 214M15 (En201) 970 212M36 (En8M) 970 216M36 (En15AM) 970 225M36 970 212A37 (En8BM) 970 212M36 (En8M) 970 216M36 (En15AM) 970 225M36 3100 Type 5 4670 711M40 970 708A40 970 708A42 (En19C) 970 708H42 970 708M40 970 709A40 970 709M40 970 708A42 (En19C) 970 708H42 970 709A42 France (AFNOR NF) China (GB) A35-562 10F1 — — — 10S20 — — — — — Y20 — 35MF4 A35-562 35MF6 ISO Y35 — 35MF4 A35-562 35MF6 — — 40CD4 A35-552 42CD4, 42CDTS A35-553 42CD4, 42CDTS A35-556 42CD4, 42CDTS A35-557 42CD4, 42CDTS 40CD4 A35-552 42CD4, 42CDTS A35-553 42CD4, 42CDTS A35-556 42CD4, 42CDTS A35-557 42CD4, 42CDTS 42CrMo 42CrMo4 42CrMo4E 42CrMoS4 42CrMo 40CrMnMo 42CrMo4 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) 4145 1.2332, 47CrMo4 4147 4320 1.2332, 47CrMo4 1.3565, 48CrMo4 1.7228, 50CrMo4 1.7228, GS50CrMo4 1.7230, 50CrMoPb4 1.7328, 49CrMo4 1.3565, 48CrMo4 1.7238, 49CrMo4 1.7228, 50CrMo4 1.7228, GS50CrMo4 1.7230, 50CrMoPb4 1.7238, 49CrMo4 1.7229, 61CrMo4 1.7266, GS-58CrMnMo4 GS-58CrMnMo4 43 — 4340 1.6565, 40NiCrMo6 4150 4161 E4340 4422 4427 40CrNiMoA 45CrNiMo VA 1.6562, 40NiCrMo 73 45CrNiMo VA 40CrNiMoA 1.5419, 22Mo No international equivalent Japan (JIS) G4052 SCM5H G4052 SCM445H G4105 SCM5, SCM445 G4052 SCM5H G4052 SCM445H G4105 SCM5, SCM445 United Kingdom (BS) France (AFNOR NF) China (GB) ISO 970 708H45 A35-553 45SCD6 A35-552 45SCD6 — — 970 708A47 A35-552 45SCD6 A35-553 45SCD6 A35-571 50SCD6 — — — — A35-571 50SCD6 G4801 SUP13 3100 BW4 3146 CLA12 Grade C G4103 SNCM23 G4103 SNCM420 G4103 SNCM420H G4103 SNCM8 G4103 SNCM439 G4108 SNB23-15 G4108 SNB24-15 36CrNiMo6 — — 4670 818M40 970 2S.119 — — 50CrMo4 — — 20NCD7 A35-565 18NCD4 A35-565 20NCD7 — 970 2S.119 — — — — 23D5 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Japan (JIS) 4615 4617 — — — — 4620 — — 4626 — — 4718 4720 4815 4817 4820 50B40 50B44 5046 50B46 50B50 5060 5115 No international equivalent — No international equivalent No international equivalent No international equivalent 1.7003, 38Cr2 1.7023, 28CrS2 — 1.3561, 44Cr2 No international equivalent 1.7138, 52MnCrB3 1.2101, 62SiMnCr4 1.7131, 16MnCr5, GS-16MnCr5 1.7139, 16MnCrS5 1.7142, 16MnCrPb5 1.7160, 16MnCrB5 United Kingdom (BS) — 970 665A17 970 665H17 970 665M17 (En34) 970 665A19 970 665H20 970 665M20 970 665A24 (En35B) France (AFNOR NF) China (GB) ISO 15ND8 — — — — — 2ND8 — — — — — — — 18NCD4 — — G4052 SMnC3H G4052 SMnC443H G4106 SMnC3 G4106 SMnC443 G5111 SCMnCr4 — — — — — — — A35-552 38C2 A35-556 38C2 A35-557 38C2 A35-552 42C2 A35-556 42C2 A35-557 42C2 45C2 — — — — — — — 55C2 — — 61SC7 A35-552 60SC7 16MC5 A35-551 16MC5 — — 15Cr 15CrMn — — G4052 SCr21H G4052 SCr415H G4104 SCr21 G4104 SCr415 970 526M60 (EnH) 970 527A17 970 527H17 970 527M17 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 5117 5120 5130 Fed. R. of Germany (DIN) 1.3521, 17MnCr5 1.7016, 17Cr3 1.7131, 16MnCr5, GS-16MnCr5 1.7139, 16MnCrS5 1.7142, 16MnCrPb5 1.7168, 18MnCrB5 1.2162, 21MnCr5 1.3523, 19MnCr5 1.7027, 20Cr4 1.7028, 20Cr5 4 1.7121, 20CrMnS3 3 1.7146, 20MnCrPb5 1.7147, GS20MnCr5 1.7149, 20MnCrS5 1.8401, 30MnCrTi4 Japan (JIS) United Kingdom (BS) — — 18Cr4 A35-551 16MC5 G4052 SCr22H G4052 SCr420H G4052 SMn21H G4052 SMn421H G4104 SCr22 G4104 SCr420 — A35-551 20MC5 A35-552 20MC5 20Cr 20CrMn 970 530A30 (En18A) 970 530H30 28C4 30Cr 970 530A32 (En18B) 970 530A36 (En18C) 970 530H32 3111 Type 3 970 530A36 (En 18C) 970 530H36 A35-552 32C4 A35-553 32C4 A35-556 32C4 A35-557 32C4 G4052 SCr2H G4052 SCr430H G4104 SCr2 G4104 SCr430 G4104 SCr3 G4104 SCr435 5132 1.7033, 34Cr4 1.7037, 34CrS4 5135 1.7034, 37Cr4 1.7038, 37CrS4 1.7043, 38Cr4 G4052 SCr3H G4052 SCr435H 5140 1.7035, 41Cr4 1.7039, 41CrS4 1.7045, 42Cr4 G4052 SCr4H G4052 SCr440H G4104 SCr4 G4104 SCr440 3111 Type 3 970 2S.117 970 530A40 (En 18D) 970 530H40 970 530M40 France (AFNOR NF) 38C4 A35-552 38Cr4 A35-553 38Cr4 A35-556 38Cr4 A35-557 38Cr4 A35-552 42C4 A35-557 42C4 A35-556 42C4 China (GB) ISO — — — 20Cr4 20MnCr5 20Cr4E 20CrS4 — — 35Cr 34Cr4 34CrS4 40Cr 38Cr 41Cr4 41CrS4 41Cr4E Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) Japan (JIS) 5147 1.7145, GS50CrMn4 4 — 5150 1.7145, GS50CrMn4 4 1.8404, 60MnCrTi4 — 5155 1.7176, 55Cr3 5160 1.2125, 65MnCr4 51B60 E50100 — 1.2018, 95Cr1 1.3501, 100Cr2 1.2057, 105Cr4 1.2109, 125CrSi5 1.2127, 105MnCr4 1.3503, 105Cr4 1.2059, 120Cr5 1.2060, 105Cr5 1.2067, 100Cr6 1.3505, 100Cr6 1.3503, 105Cr4 1.3514, 101Cr6 1.3520, 100CrMn6 No international equivalents 1.8159, GS-50Cr V4 E51100 E52100 6118 6150 8115 81B45 8615 8617 G4801 SUP11 G4801 SUP9 G4801 SUP9A — — — — G4801 SUP10 No international equivalents No international equivalents — — — — United Kingdom (BS) 3100 BW2 3100 BW3 3146 CLA12 Grade A 3146 CLA12 Grade B 3100 BW2 3100 BW3 3146 CLA12 Grade A 3146 CLA12 Grade B — France (AFNOR NF) ISO 45Cr — 50Cr — 55CrMnA — 60CrMnA — 60CrMnBA GCr6 — — — GCr9 — 970 534A99 (En31) 970 535A99 (En31) — GCr15 970 735A 50 (En47) 970 S.204 A35-552 50CV4 970 527A60 (En48) 970 527H60 — — — — 970 805A17 970 805H17 970 805M17 (En 361) 50C 4 China (GB) — A35-571 55C3 — — A35-565 100C2 50CrVA 1 50CrV4 A35-553 50CV4 A35-571 50CVA 15NCD2 15NCD4 18NCD4 18NCD6 — — — — Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) Fed. R. of Germany (DIN) 8620 1.6522, 20NiCrMo2 1.6523, 21NiCrMo2 1.6526, 21NiCrMoS2 1.6543, 21NiCrMo2 2 8622 1.6541, 23MnNiCrMo5 2 Japan (JIS) United Kingdom (BS) G4052 SNCM21H G4052 SNCM220H G4103 SNCM21 G4103 SNCM220 2772 806M20 970 805A20 970 805H20 970 805M20 (En362) — 2772 806M22 970 805A22 970 805H22 970 805M22 8625 — — 970 805H25 970 805M25 8627 8630 8637 No international equivalents 1.6545, 30NiCrMo2 2 — — — 8640 1.6546, 40NiCrMo2 2 — — 970 945M38 (En100) 3111 Type 7, 2S.147 970 945A40 (En 100C) 8642 France (AFNOR NF) China (GB) ISO 20CrNiMo G20CrNiMo 20NiCrMo2 20NiCrMo2E 20NiCrMoS2 — — 30NCD2 40NCD3 — — — — 40NCD2 40NCD2TS 40NCD3TS 40NCD3 — 41CrNiMo2E 18NCD4 20NCD2 A35-551 19NCDB2 A35-552 19NCDB2 A35-551 20NCD2 A35-553 20NCD2 A35-565 20NCD2 A35-566 20NCD2 23NCDB4 A35-556 23MNCD5 A35-556 23NCDB2 A35-566 22NCD2 25NCD4 A35-556 25MNCD6 A35-566 25MNDC6 No international equivalents No international equivalents No international equivalents 8645 86B45 Continued ß 2006 by Taylor & Francis Group, LLC. TABLE 1.17 (Continued) Cross-Reference to Steels United States (SAE) 8650 8655 8660 8720 8740 8822 9254 9260 Fed. R. of Germany (DIN) No international equivalents No international equivalents — No international equivalents 1.6546, 40NiCrMo2 2 No international equivalents — — E9310 1.6657, 14NiCrMo1 3 4 G10CrNi3Mo 94B15 Japan (JIS) United Kingdom (BS) — 3111 Type 7, 2S.147 China (GB) ISO — — — — — 970 805A60 970 805H60 — France (AFNOR NF) — G4801 SUP7 — 60S7 61S7 94B30 16NCD13 — 60Si2Mn 60Si2MnA 56SiCr7 59Si7 61SiCr7 No international equivalents No international equivalents No international equivalents 94B17 — — 970 250A58 (En45A) 970 250A61 (En45A) 970 832H13 970 832M13 (En36C) S.157 40NCD2 40NCD2TS 40NCD3TS — Source: From Anon., Classification and designation of carbon and low-alloy steels, ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 140–194; H. Lin, G. Lin, and Y. Ma, Eds., Designation and Trade Name Handbook of Steels Worldwide, mechanical Industry Press, Beijing, 1997. 1.6.1.2 The Designation for Steels with Chemical Composition The prefix of nonalloy heat-treatable steel is C, and followed by numerical value. The numerical value is 100 times of the average of carbon content, for example, C25 is the steel containing an approximate mean of 0.25% C. The additional suffix EX or MX indicated the quality steel or the high quality steel. The designation for structural alloy steels and spring steels are similar to that of DIN systems. 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). ß 2006 by Taylor & Francis Group, LLC. The designation for bearing steels is indicated by a numeric code. Types 1–5 represent high-carbon chromium-bearing steels (fully hardened hearing steels), Types 10–16 are the surface-hardened bearing steels, Types 20–21 denote stainless-bearing steels, and Types 30–32 are the high-temperature bearing steels. Stainless steels are also indicated by a numeric code. For example, the ferritic stainless steels include Type 1Ti, 1, 2, 8, . . . , martensitic stainless steels include Type 3, 4, 5, . . . , austenitic stainless steels include Types 10–24 and A2–A4, etc. Heat-resistant steels are indicated with prefix letter H and followed by numeric code. For example, types H1–H7 are ferritic steels and types H10–H18 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 would not 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 1.17. 1.6.2 GB DESIGNATIONS (STATE STANDARDS OF CHINA) The state standard of China for steels is called Guojia Biaozhun, abbreviated as GB. The GB designations for nonalloy common steels and HSLA steels use the prefix letter Q, followed by the yield strength value (MPa). For example, Q235, Q345, Q390 denote nonalloy 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 the steel containing 0.45% C. Alloy elements in steel use the descriptive code with chemical symbols, and followed by its average content. As the average content is less than 1.5%, it is indicated only with the chemical symbol, for example, 34CrNi3Mo containing 0.30–0.40% C, 0.70–1.10% Cr, 2.75–3.25% Ni, and 0.25–0.40% 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 contains an 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 average content is less than 1.0%, it uses numeric code ten times carbon content. For example, CrMn steel contains 1.30–1.50% C, 1.30–1.60% Cr, 0.45–0.75% Mn, and 9Mn2V steel contains 0.85– 0.95% C, 1.70–2.00% Mn, 0.10 to 0.25% V. The descriptive method for the alloy element is the same in alloy structural steels. Carbon content is not indicated in high-speed steels, and only uses the descriptive code with chemical symbols and followed by alloy element content. For example, tungsten highspeed steel 18-4-1 (T1) is represented by W18Cr4V, and W–Mo 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 content and followed by chemical symbols of alloy elements with their content in stainless steels and heat-resistant steels, but microalloy elements only show the chemical symbols. For example, steel 9Cr18MoV contains 0.85–0.95% C, 17–19% Cr, 0.0–1.3% Mo, 0.07–0.12% V. If the carbon 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. ß 2006 by Taylor & Francis Group, LLC. 1.6.3 DIN STANDARDS DIN standards are developed by Deutsches Institut fuer Normung in Germany. All German steel standards and specifications are represented by the letters DIN and followed by an alphanumeric or a numeric code. An uppercase letter sometimes precedes this code. German designations are reported in one of the following two methods. One method uses the descriptive code with chemical symbols and numbers in the designation. The second, called the Werkstoff number, uses numbers only, with a decimal after the first digit. There are four figures after the decimal 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 the decimal point. Examples of both methods are provided in Table 1.17, which cross-references DIN designations and indicates chemical composition for DIN steels. However, standards for heat-resistant steels are prefixed with the letter SEW (Stahl-Eisen-Werkstoff Blaetter, steel-iron material 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 Ni–Cr–Mo steel that contains 0.35–0.45% C, 0.9–1.4% Cr, 0.5–0.7% Mn, 0.2–0.3% Mo, 1.4–1.7% Ni, 0.035% S, and 0.15–0.35% Si; DIN 17200 1.1149 or DIN 17200 Cm22 (G10200) is a nonresulfurized carbon steel containing 0.17–0.24% C, 0.3–0.6% Mn, 0.035 max P, 0.02–0.035% S, and 0.4% max Si. 1.6.4 JIS STANDARDS JIS standards are developed by the Japanese Industrial Standards Committee (JISC) in Tokyo, Japan. The specifications begin with the prefix JIS, followed by a letter G for carbon and low-alloy steels. This is followed by a space and series of numbers and letters indicating the 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 a low-carbon tube steel containing 0.12% C, 0.35% Si, 0.60% Mn, 0.04% P, and 0.04% S; JIS G3445 STK 17A (G10490) is a medium-carbon nonresulfurized steel containing 0.45–0.55% C, 0.40–1.0% Mn, 0.04% P, 0.04% S, and 0.04% Si; JIS G4403 SKH2 (AISI T1 grade) is a tungsten high-speed tool steel containing 0.73–0.83% C, 3.8–4.5% Cr, 0.4% Mn, 0.4% Si, 0.8–1.2% V, and 17–19% W; and JIS G4403 SKH59 (AISI M42 grade) is a molybdenum ultrahard high-speed tool steel containing 1–1.15% C, 7.5–8.5% Co, 3.5–4.5% Cr, 0.4% max Mn, 9–10% Mo, 0.5% max Si, 0.9–1.4% V, 1.2–1.9% W, 0.25% max Ni, 0.03% max P, 0.03% max S, and 0.25% Cu. 1.6.5 BS STANDARDS BS standards are developed by the British Standards Institute (BSI) in London, England. The letters 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 British standards. Some example of BS designations with equivalent AISI designations in parentheses are given: BS 970 708A30 (4130) is a Cr–Mo low-alloy steel containing 0.28–0.33% C, 0.9–1.2% Cr, 0.4–0.6% Mn, 0.15–0.25% Mo, 0.035% P, 0.04% S, and 0.1–0.35% Si; and BS 970 304S15 (304) is a wrought austenitic stainless steel (sheet, strip, plate) containing 0.06% C, 17.5–19% Cr, 0.5–2.0% Mn, 8–11% Ni, 0.05% P, 0.03% S, and 0.2–1.0% Si. 1.6.6 AFNOR STANDARDS AFNOR 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 ß 2006 by Taylor & Francis Group, LLC. NF, followed by an alphanumeric code constituting an uppercase letter followed by a series of digits, which are subsequently followed by an alphanumeric sequence. For example, resulfurized (free-cutting) steel is listed in AFNOR NF A35–562 standard or specification, and 35MF6 designation (equivalent to SAE 1137) represents the steel bar containing 0.33– 0.39% C, 1.30–1.70% Mn, 0.10–0.40% Si, 0.040% P, and 0.09–0.13% S; whereas 45MF4 designation (equivalent to SAE 1146) contains 0.42–0.49% C, 0.8–1.1% Mn, 0.1–0.4% Si, 0.04% max P, 0.09–0.13% S. Similarly, AFNOR NF A35-573 Z6CN 18.09 is a wrought (SAE 304) stainless steel (sheet, strip, plate) and contains 0.07% C, 17–19% Cr, 2% Mn, 8–10% Ni, 0.04% P, 0.03% S, and 1% Si. REFERENCES 1. 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. 8840–8843. 3. H. Okamoto, C–Fe, in Binary Alloy Phase Diagrams, 2nd ed., T.B. Massalski, Ed., ASM International, Materials Park, OH, 1990, pp. 842–848. 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. 140–194. 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 of Automotive Engineers, Warrendale, PA, pp. 2.01–2.04. 9. G. Krauss, Steels—Heat 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, Steels—Microstructure and Properties, Adward Arnold, London, 1982. 13. H.E. Boyer, in Fundamentals of Ferrous Metallurgy, Course 11, Lesson 12, Materials Engineering Institute, 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 Society for 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): 26–31 (1988). 22. S. Zhang and C. Wu, in Proceedings of 3rd International Congress on Heat Treatment of Materials, Shanghai, 1983, pp. 3.25–3.36. 23. T.Y. Hsu (Zuyao Xu), ISIJ Int. 38: 1153–1164 (1998). 24. M.J.U.T. Van Wijngaarden and G.P. Visagie, in Proceedings of the 79th Steelmaking Conference, Vol. 79, Pittsburgh Meeting, March 24–27, pp. 627–631. 25. H. Matsuka, K. Osawa, M. Ono, and M. Ohmura, ISIJ Int. 37: 255–262 (1997). ß 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 Alloy Steel, 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. 1075–1138. 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 of Automotive Engineers, Warrendale, PA, pp. 1.01–1.189. 31. 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: 2688–2692 (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. 32–42. 35. T.A. Phillip, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 430–448. 36. A.K. Sinha, in Production of Iron/Steel and High Quality Product Mix (Conf. Proc. 1991), ASM International, Materials Park, OH, 1992, pp. 195–206. 37. L.F. Porter and P.E. Repas, J. Metals 34(4): 14–21 (1982). 38. L.F. Porter, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford, 1986, pp. 2157–2162. 39. H. Zhang, in Chinese Encyclopedia of Metallurgy, Metallic Materials, Metallurgical Industry Press, Beijing, 2001, pp. 152–154. 40. P. Charlier and L. Bacher, Proc. Seventh Int. Conf. Strength of Metals and Alloys, Vol. 2, Montreal, Canada, August 12–16, 1986, Pergamon Press, Oxford, p. 1019. 41. P. Charlier and A. Guimier, Proc. Int. Sem. Automotive Steels, Moscow, April 19–21, 1988, p. 167. 42. A.H. Nakagawa and G. Thomas, Metall. Trans. 16A: 831–840 (1985). 43. D.Z. Yang, E.L. Brown, D.K. Matlock, and G. Krauss, Metall. Trans. 16A: 1523–1526 (1985). 44. R.C. Davies, Metall. Trans. 10A: 113–118 (1979). 45. A.R. Marder, Metall. Trans. 13A: 85–92 (1982). 46. A.M. Bayer and L.R. Walton, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 757–779. 47. M.G.H. Wells, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford, 1986, pp. 5115–5120. 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, Materials Park, OH, 1990, pp. 841–907. 50. R.F. Decker, J.T. Each, and A.J. Goldman, Trans. ASM 55: 58 (1962). 51. S. Floreen, Met. Rev. 13: 115–128 (1968). 52. S. Floreen, in Encyclopaedia of Materials Science and Engineering, Pergamon Press, Oxford, 1986, pp. 5171–5177. 53. T. Morrison, Metall. Mater. Technol. 8: 80–85 (1976). 54. K. Rohrbach and M. Schmidt, in ASM Handbook, 10th ed., Vol. 1, ASM International, Materials Park, OH, 1990, pp. 793–800. 55. Heat Treater’s Guide—Standard 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 Petroleum Institute, Washington, D.C. 60. Specification for Line Pipe, API Specification 5L, 40th ed., November 1, 1992, American Petroleum Institute, Washington, D.C. ß 2006 by Taylor & Francis Group, LLC. 61. Specification for CRA Line Pipe, API Specification 5LC, 2nd ed., August 1, 1991, American Petroleum 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. ß 2006 by Taylor & Francis Group, LLC. 2 Classification and Mechanisms of Steel Transformation* S.S. Babu CONTENTS 2.1 2.2 2.3 Introduction ................................................................................................................ 91 Phase Transformation Mechanisms ............................................................................ 92 Microstructure Evolution during Austenite Decomposition....................................... 96 2.3.1 Allotriomorphic Ferrite .................................................................................... 96 ¨ 2.3.2 Widmanstatten Ferrite...................................................................................... 97 2.3.3 Bainite............................................................................................................... 99 2.3.4 Pearlite............................................................................................................. 101 2.3.5 Martensite........................................................................................................ 103 2.4 Microstructure Evolution during Reheating .............................................................. 106 2.4.1 Tempered Martensite....................................................................................... 107 2.4.1.1 Carbon Segregation and Aging of Martensite .................................. 107 2.4.1.2 First Stage of Tempering................................................................... 107 2.4.1.3 Second Stage of Tempering............................................................... 107 2.4.1.4 Third Stage of Tempering ................................................................. 108 2.4.1.5 Fourth Stage of Tempering ............................................................... 108 2.4.2 Austenite Formation ....................................................................................... 109 2.5 Summary of Steel Microstructure Evolution ............................................................. 109 2.6 Prediction of Microstructure Evolution during Heat Treatment ............................... 110 2.6.1 Calculation of Multicomponent Multiphase Diagrams................................... 112 2.6.2 Calculation of Diffusion-Controlled Growth .................................................. 113 2.7 Summary.................................................................................................................... 116 2.8 Acknowledgments ...................................................................................................... 118 References .......................................................................................................................... 118 2.1 INTRODUCTION The microstructure of steels consists of a spatial arrangement of crystalline aggregates of different phases. The size, shape, distribution, composition, and crystal structure of these phases 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 the industrial 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 reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. ß 2006 by Taylor & Francis Group, LLC. by 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 by changing the composition and cooling rate. In low-alloy steels, the most important phase change is from austenite to a-ferrite. For example, for a given composition of low-alloy steel, a ferrite–pearlite microstructure can be obtained by slow cooling. With an increase in cooling ¨ rate, Widmantstatten or upper bainite microstructure can be obtained from austenite. With a further increase in the cooling rate, a hard martensite microstructure can be obtained. Pearlite contains a lamellar aggregate of ferrite and cementite phases, upper bainite contains ferrite platelets separated by austenite or carbides, and martensite contains carbon-supersaturated ferrite 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 ferrite–pearlite microstructure, which contains ferrite þ 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 the farthest from equilibrium. Therefore, tempering martensite at high temperature can be used to attain intermediate-phase mixtures that are closer to final equilibrium, thus providing another methodology to control the microstructure and properties of the steel. The methods of controlling steel microstructure through alloy modification and heat treatment have been followed by metallurgists and this will continue to be the goal of material scientists. The focus of this chapter 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 been illustrated in a simple Fe–C metastable diagram (Figure 2.1) that describes the stability regions for ferrite, austenite, and cementite structures. On cooling from the liquid region, the first phase to solidify is d-ferrite. With continued cooling, the d-ferrite transforms to g-austenite. With further cooling, the austenite transforms to a-ferrite and cementite. Basic and applied research in the past has led to a fundamental understanding of these structural changes and their relationship to microstructure evolution in alloys ranging from simple Fe–C systems to complex Fe–C–X steels (where X stands for many different substitutional alloying elements, including manganese, nickel, chromium, silicon, and molybdenum). The relationships between crystal structure changes, the interface structure, the phase morphologies (i.e., ferrite or bainite), 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 is possible to predict transformation kinetics quantitatively as a function of steel composition. In this chapter, a framework to classify transformation mechanisms is introduced. Then, the steel microstructures are classified and discussed based on this framework. Finally, a brief introduction is given to the application of computational thermodynamics and kinetics to describe phase transformations in low-alloy steels. 2.2 PHASE TRANSFORMATION MECHANISMS The various phase transformations that occur in steels can be classified according to a thermodynamic basis, a microstructural basis, and a mechanistic basis [1]: 1. The thermodynamic basis classifies the phase transformation based on the derivatives of the Gibbs free energy (G) of the system with temperature change at a constant pressure. 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 ß 2006 by Taylor & Francis Group, LLC. bcc Liquid δ-ferrite 1600 γ +L 1400 δ+L δ+γ fcc 1200 Temperature (8C) γ-austenite bcc 1000 α-ferrite 800 γ +θ α+γ 600 α-ferrite + θ-cementite 400 200 0.00 0.2 0.4 0.6 0.8 Weight percentage carbon 1.0 FIGURE 2.1 Illustration of body-centered cubic (bcc) ferrite and face-centered cubic (fcc) austenite with a calculated metastable Fe–C 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 is classified 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 ferrite phase in an Fe–Al system. 2. The microstructural basis relates to the origin of a product phase from a parent phase. If the 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 small entity with a sharp interface and if it grows into the product phase, it is classified as a heterogeneous 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 during transformation. The heterogeneous transformation in a material can occur by three mechanisms [2]: (a) athermal growth through glissile interfaces (e.g., austenite to martensite formation), (b) thermally activated growth (e.g., pearlite formation from austenite), and (c) growth controlled by heat transport (e.g., solidification). Details of each subclassification of growth mechanisms have been discussed by Christian [2] and are 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 ß 2006 by Taylor & Francis Group, LLC. 94 ß 2006 by Taylor & Francis Group, LLC. Heterogeneous transformations Athermal growth (glissile interfaces) Coherent interface Semi-coherent interface Martensite (steels) Mechanical twinning Low-angle boundary No long-range transport (interface controlled) Polymorphic transitions growth from vapour Recrystallization grain growth Order–disorder changes Long-range transport Discontinous reactions diffusion + interface controlled? Continuous reactions Interface controlled Crystal growth from vapour Solidification melting Diffusion controlled Precipitation dissolution Eutectoidal reaction Discontinuous precipitation Precipitation dissolution FIGURE 2.2 Classification of heterogeneous transformation based on mechanisms of growth. Most relevant phase transformation mechanisms for steel heat treatment 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 Technologies Coherent martensite Growth controlled by heat transport Thermally activated growth controlled by glissile interface motion, thermally activated interface movement, or heat transport. The relevant mechanism for steel microstructure evolution during heat treatment is a heterogeneous phase transformation involving glissile interface (displacive) or long-range transport (reconstructive). The reconstructive transformation involves mixing of atoms on either side of the parent and product phase interface through diffusion. The displacive transformation involves coordinated atom motion in the parent phase to change the parent crystal structure to the product crystal structure. These reconstructive and displacive transformation mechanisms in steels are discussed with a schematic illustration of austenite decomposition to ferrite or martensite as shown in Figure 2.3 [3]. Let us assume that the austenite fcc lattice is represented by the square motif shown in the upper left part of Figure 2.3 bounded by a rectangle a–b–c–d. Various substitutional atoms such as silicon, manganese, chromium, and molybdenum are schematically described by different symbols. Initial arrangements of some atoms are denoted by numbers 1 to 6. During the decomposition of austenite to ferrite or martensite, a crystal structure change takes place at the interface between austenite and the product phase (represented by line e–f). 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 change from fcc to bcc occurs at the e–f interface and is accompanied by somewhat random hopping of atoms across the interface. Because of this hopping of the atoms across the interface, the original atomic arrangements (implied by the change in the position of 1–6 numbered atoms) are lost. As a result, the transformation involves the reconstruction of an austenite lattice into a ferrite lattice, which leads to no macroscopic shape change of the crystal and mixing or separation of atoms on either side of the austenite and ferrite that would lead to a change in the composition of the phase. However, there is a distinct change in volume. Based on the above facts, the rate transformation is expected to be controlled by long-range diffusion of substitutional atoms in both the ferrite and austenite lattices. In practice, the transformations in steels are complicated by the presence of carbon, an interstitial element, which is not shown in the Figure 2.3. The growth rate of ferrite (which has low solubility of carbon) is controlled by diffusion of both carbon in austenite and substitutional atoms in ferrite and austenite. Under reconstructive transformation, diffusion of all atoms occurs during nucleation and growth and is usually sluggish below 6008C [4]. Austenite Austenite a d f 6 1 2 3 4 5 c a d e 1 b Interface 6 2 3 4 Displacive transformation 5 c b Atomic correspondence IPS shape change with a significant shear component Diffusionless Martensite Austenite 1 d 2 a 5 e f 4 Interface Reconstructive transformation 3 No atomic correspondence No shape change with shear component Possible composition change 6 b c Ferrite FIGURE 2.3 Schematic illustration of crystal structure change due to reconstructive and displacive transformations. (Adapted from H.K.D.H. Bhadeshia, Worked Examples in the Geometry of Crystals, The Institute of Materials, London, 1987.) ß 2006 by Taylor & Francis Group, LLC. The schematic illustration of martensite formation is shown in the upper right side of Figure 2.3, and the bct lattice is represented by a diamond motif. The main difference in this case, compared to the fcc to bcc transformation, is that the crystal structure change from fcc to bct lattice is brought about by coordinated atom movement at the interface e–f, or in other words, a shear deformation of the original fcc lattice. As a result, the original atomic correspondences within the parent phase are maintained in the product phase. For example, the row of atoms 1–6 is just displaced laterally parallel to interface e–f, and the atomic sequence remains the same. This mode of transformation involving a shear deformation is called invariant plane strain (IPS). IPS leads to a macroscopic shape change, as indicated in the schematic illustration. This macroscopic change (from a–b–c–d to a–e–b–c–f–d) manifests itself as a surface relief on the polished surface of the sample after the decomposition of austenite to martensite. For more crystallographic details, the reader is referred to the literature [5–7]. 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 the magnitude of chemical free energy increases above the strain energy requirement. The large strain energy also induces the lenticular or plate-shaped region of martensite to minimize the strain 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 composition between the austenite and martensite phases (i.e., it is a diffusionless mode of transform¨ ation). As a result, the growth rate of martensite is usually athermal. The Widmanstatten and bainitic ferrite transformations occur at higher temperatures than the martensite transformation. They exhibit displacive transformation for a crystal structure change and varying degrees of diffusion of interstitial carbon during nucleation and growth. In the following sections, the steel microstructure evolution during cooling (austenite decomposition) and heating (tempering and austenite formation) will be described based on reconstructive and displacive transformation mechanisms. 2.3 MICROSTRUCTURE EVOLUTION DURING AUSTENITE DECOMPOSITION There have been numerous reviews of austenite decomposition over the past five decades [8–10]. The ferrite morphologies, which form during austenite decomposition, were originally ´ classified by Dube [11,12]. A change in morphology from one to another was found to occur as the austenite decomposition temperature was lowered. The common ferrite morphologies ¨ are grain boundary allotriomorphic ferrite, idiomorphic ferrite, Widmanstatten ferrite, and intragranular ferrite [13,14]. Other microstructures that form at lower transformation temperatures are pearlite, bainite, and martensite. The evolution of each one of the above microstructure is discussed below. 2.3.1 ALLOTRIOMORPHIC FERRITE Decomposition of austenite to allotriomorphic ferrite occurs over a wide range of temperatures below the austenite to austenite þ ferrite phase boundary (see Figure 2.1). This section is not intended to be a comprehensive review of this subject, but rather a short summary of extensive literature. The reader can consult the classic reviews by Aaronson and other researchers for details [10,13]. Allotriomorphic ferrite usually nucleates along the austenite–austenite grain boundary. It first grows laterally along the boundary and then can proceed perpendicularly into the austenite grain. The nucleation and growth involve a reconstructive mode of crystal structure change that leads to an absence of any macroscopic shape change, and only a volume change ß 2006 by Taylor & Francis Group, LLC. is observed. During nucleation, the allotriomorphic ferrite crystals exhibit a preferred orientation relationship with one of the austenite grains (g1). The orientation relationship is usually of the Kurdjumov–Sachs (KS) type: {1 1 1}g =={1 1 0}a and h1 " 0ig ==h1 " 1ia : 1 1 The KS relationship indicates that the close-packed planes of austenite and ferrite are parallel to each other and that the close-packed directions of austenite and ferrite are parallel to each other. On some occasions, the ferrite may deviate slightly from having close-packed parallel directions, as shown by the Nishiyama–Wasserman (NW) orientation relationship: {1 1 1}g =={1 1 0}a and " h1 1 2ig ==h1 " 0ia : 1 The KS or NW orientations allow for the existence of a semicoherent boundary between austenite and ferrite and thereby minimize the surface energy required to nucleate the ferrite at the austenite boundary. A small free energy is sufficient to satisfy the KS and NW orientations with one of the austenite grains for the nucleation of a ferrite, and then the ferrite 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 disordered boundary will exist and rapidly grow into the austenite. An example of such ferrite growth in an Fe–C–Si–Mn steel is shown in Figure 2.4. The optical micrograph (Figure 2.4a) shows the allotriomorphic ferrite grains along the prior austenite grain boundary (marked by arrow). The remaining microstructure is bainitic ferrite. A transmission electron microscopy image of this sample (see Figure 2.4b) shows an allotriomorphic ferrite grain situated along the boundary (marked as B) of two austenite grains (g1 and g2 ). Electron diffraction analysis of the above structure indicated that the allotriomorphic ferrite had a KS orientation relationship 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 protuberances on the g1 side of the allotriomorphic ferrite. However, the interface on the g2 side showed disordered boundary with no specific structure [15]. The rate of growth on the g2 side is more rapid than that on the g1 side. This apparent difference in growth rate was shown by classic in situ transmission electron microscopy experiments by Purdy [16]. The observed semicoherent boundary on only one side of the austenite grain also plays a critical role in the development ¨ of secondary Widmanstatten ferrite, which is described in the following section. ¨ 2.3.2 WIDMANSTATTEN FERRITE ¨ Widmanstatten ferrite describes a ferrite morphology in the form of side plates or laths and grows into austenite grains with a KS orientation relationship. The lath ferrite, which forms ¨ from the austenite grain boundary, is referred to as primary Widmanstatten ferrite. The ferrite that nucleates on the preexisting allotriomorphic ferrite is referred to as secondary ¨ ¨ Widmanstatten ferrite [14]. Nucleation of secondary Widmanstatten ferrite occurs mostly on the semicoherent interphase boundary between the ferrite and the austenite. A typical ¨ microstructure 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 the steel cools from high temperature, the allotriomorphic ferrite forms with a KS–NW 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 and grows into the austenite grain. This nucleation and growth process leads to apparent sawtooth morphology. ß 2006 by Taylor & Francis Group, LLC. a 20 μm b B γ2 γ1 1 μm FIGURE 2.4 (a) Optical micrograph showing the presence of allotriomorphic ferrite (marked by arrow) and bainite microstructure in an Fe–C–Si–Mn steel. (b) Transmission electron micrograph showing the presence 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 KS–NW orientation relationship and was in a random orientation relationship with the g2 grain. ¨ Two growth mechanisms of Widmanstatten ferrite have been proposed. In the first mechanism, the growth is attributed to lateral movement of semicoherent interfaces by ¨ small steps (ledges) in the interface [17,18]. In the second mechanism, the Widmanstatten ferrite may grow through a displacive transformation mechanism [19]. Many experimental ¨ results are in agreement with the second mechanism. Thin, wedge-shaped Widmanstatten ferrite is produced due to cooperative, back-to-back growth of two ferrite crystallographic ¨ variants. The Widmanstatten ferrite plates grow into untransformed austenite by extension ¨ along their length. Since Widmanstatten ferrite forms at temperatures well below that for allotriomorphic ferrite, growth occurs by a paraequilibrium (PE) mode (i.e., the ratio of the ß 2006 by Taylor & Francis Group, LLC. a Allotriomorphic ferrite Widmanstätten ferrite 20 μm Original allotriomorphic ferrite b Widmanstätten ferrite Prior austenite grain boundary γ1 γ2 ¨ FIGURE 2.5 (a) Optical micrograph showing secondary Widmanstatten ferrite microstructure in an Fe– C–Mn steel weld obtained through continuous cooling. The growth of these plates occurs on only one side of the allotriomorphic ferrite. (b) Schematic illustration showing possible mechanism for the ¨ formation of Widmanstatten ferrite microstructure on only one side of the original allotriomorphic ferrite. iron concentration to the substitutional atom concentration is frozen in both the parent and the product phases). The rate of growth is controlled only by carbon diffusion in the austenite ¨ ahead of the plate. It is possible to estimate the growth rate of Widmanstatten ferrite with diffusion-controlled growth models of Bhadeshia et al. [20] and Trivedi [21,22] by considering only the carbon diffusion. 2.3.3 BAINITE The growth of bainitic ferrite has been discussed in the literature based on either reconstructive or displacive transformation mechanisms. In the reconstructive definition, the growth of bainite is the product of diffusional, noncooperative, competitive ledgewise growth of ferrite and cementite into austenite during eutectoid decomposition with cementite appearing in a nonlamellar form [23]. In this definition, the transformation kinetics is related to the rate of ß 2006 by Taylor & Francis Group, LLC. ledge movement at the interface and is controlled by carbon diffusion. In the displacive definition, a subunit of ferrite forms from the austenite with complete supersaturation of carbon through a displacive transformation involving IPS deformation. Carbon diffusion to austenite occurs through a post-transformation event. The overall transformation kinetics is then related to the nucleation of this ferrite subunit. The bainitic subunits are expected to show surface relief, no substitutional partitioning, an incomplete reaction phenomenon, and a KS–NW orientation relationship [24]. They are also expected to show compliance with an externally applied elastic stress on the austenite similar to martensite [25]. The reader can consult the recent discussions by Hillert [26] for current research in bainite transformation mechanisms. In this section, the bainitic transformation mechanisms are discussed from the displacive mode of transformation point of view [4]. The bainite microstructure consists of aggregates of ferrite plates separated by thin films of austenite, martensite, or cementite. These aggregates of plates are called sheaves [4]. A typical bainitic sheaf in an Fe–Cr–C steel is shown in Figure 2.6a. This microstructure was attained by austenitization followed by isothermal transformation below the bainitic start temperature Bs [25]. The small ferrite plates that make up the sheaves are often referred to as subunits, 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 a lath structure. The presence of subunits within the sheaf can also be seen by transmission electron microscopy. Transmission electron microstructures of an Fe–C–Si–Mn steel (see Figure 2.6b) and an Fe–Cr–C steel (see Figure 2.6c) show the subunit ferrite plates within a a b 1 μm 50 μm Bainitic sheaf d c 1 μm Austenite/ carbide Subunit Austenite grain boundary FIGURE 2.6 (a) Optical micrograph from an Fe–Cr–C steel showing a bainitic sheaf structure. (b) Transmission electron micrograph from an Fe–C–Si–Mn steel showing many sheaves of bainitic ferrite. (c) Another transmission electron micrograph of a bainitic sheaf structure from an Fe–Cr–C steel. (d) Schematic illustration of the growth mechanism by subunit nucleation and growth. (Adapted from H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., The Institute of Materials, London, 2001.) ß 2006 by Taylor & Francis Group, LLC. bainitic sheaf. The first ferrite subunit that nucleates on the austenite–austenite grain boundary grows to a particular size, and successive plates form to produce the microstructure shown in Figure 2.6d. The plates are separated by untransformed austenite, which can transform to martensite on cooling to low temperature. In certain conditions, this austenite can decompose into cementite, giving rise to a classical ferrite þ cementite mixture of bainitic microstructure. 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 bainite forms at temperatures higher than the martensite start (Ms ) temperature. Because of the IPS that accompanies the bainitic transformation, a large number of dislocations are observed within the austenite and the ferrite. These dislocations limit the growth of each subunit to a certain size. Due to the displacive mode of transformation, the substitutional and interstitial atoms are configurationally frozen and there is no redistribution of these elements between austenite and bainitic ferrite during transformation. However, after the bainite reaches a certain size, the carbon mobility is sufficient to redistribute to austenite while substitutional atoms remain frozen in place. This carbon partitioning from the bainite to austenite enriches the austenite with the progress of bainitic transformation. The successive nucleation of subunits from this carbon-enriched austenite continues to occur. The successive spatial alignment of nucleated subunits occurs due to an autocatalytic mechanism that leads to a typical sheaf structure. When the carbon enrichment of austenite increases above a critical level, the displacive mode of bainite formation cannot be sustained (the free energy of ferrite is equal to that of austenite) and the reaction stops (i.e., an incomplete reaction). This residual austenite may remain untransformed in 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 formers such 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 bainite subunits 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, the lower bainite microstructure is similar to that of tempered martensite, the main difference is the occurrence of only one crystallographic variant* of the carbide in lower bainite. When these carbides are cementite, the major axis of the cementite is aligned at ~608 to the long axis of the ferrite 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 and lower bainites form from austenite through a displacive transformation. The only difference is that, in the case of upper bainite, the carbon from the bainitic subunit partitions into austenite before 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. This competition between carbon escape from supersaturated ferrite plate and precipitation of carbide within the ferrite plates is schematically shown in Figure 2.7 [4]. 2.3.4 PEARLITE A typical pearlite microstructure has many colonies of a lamellar mixture of ferrite and cementite (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 ferrite and 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. ß 2006 by Taylor & Francis Group, LLC. Carbon supersaturated plate Carbon diffusion into austenite and carbide precipitation in ferrite Carbon diffusion into austenite Carbon precipitation from austenite Upper bainite (high temperature) Lower bainite (low temperature) FIGURE 2.7 Schematic illustration of upper and lower bainite formation mechanism. The dark regions between the plates represent carbides that form in the residual austenite. The dark lines represent carbides that form within the ferrite plates. (Adapted from H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., The Institute of Materials, London, 2001.) a 5003 10 μm Austenite grain boundary b Ferrite Pearlite colony Cementite FIGURE 2.8 (a) A typical optical microstructure of a pearlite in an Fe–C–Mn steel. Several distinct colonies are marked with arrows. (b) Schematic illustration of pearlite colony growth. ß 2006 by Taylor & Francis Group, LLC. crystallographic relationships. The most common orientation relationships are Pitsch–Petch relationship [14]: À Á  à 51 3 (0 0 1)cementite == " 2 " ferrite ; (0 1 0 )cementite 2 À 3 from 1 1 " ferrite ; (1 0 0 )cementite 2 À 3 from [1 3 1 ]ferrite , and the Bagaryatski relationship, ½1 0 0 Šcementite ==[0 1 "]ferrite ; [0 1 0 ]cementite ==[" 1 1]ferrite ; (0 0 1)cementite ==(2 1 1)ferrite : 1 1 The pearlite grows into austenite by cooperative growth of ferrite and cementite. Neither the ferrite nor the pearlite shows any preferred crystallographic orientation with the austenite into which they are growing. The colony interface with austenite is an incoherent high-energy interface. As a result, the pearlite that nucleates on preexisting allotriomorphic ferrite always chooses the ferrite side (high-energy interface) that has no crystallographic relationship with an ¨ austenite grain. In contrast, Widmanstatten ferrite and bainite always nucleate on low-energy interfaces (see Section 2.3.2 and Section 2.3.3). There is a well-known relationship between pearlite colonies and transformation conditions. The spacing of lamellae decreases with a decrease in the transformation temperature. It is possible to determine the lamellar spacing by equating the increase in interfacial energy to a decrease in energy due to transformation. The pearlite transformation occurs very close to thermodynamic equilibrium (i.e., partitioning of alloying elements occurs under local equilibrium), and often occurs at higher temperatures and 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 MARTENSITE Martensite forms from austenite through a displacive transformation. This section briefly outlines the morphology, nucleation, and growth of martensite and the crystallographic aspects of the martensite transformation. This section is not a comprehensive review of this subject; 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 below the Ms . The transformation from austenite to martensite proceeds as the temperature is reduced below 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 some austenite may be retained if only cooled to room temperature. The martensite transformation is also sensitive to external and internal stresses. The martensite plates are related to austenite through a KS orientation relationship and show very preferred habit planes. The habit plane of martensite in low-carbon steel is parallel to {1 1 1}austenite and in high-carbon steel it is parallel to {2 2 5}austenite . Because the martensite transformation occurs through a shear mechanism and 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 structure cannot 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 an austenite crystal structure is represented by a shape bounded by ABCD in (I), on applying a shear deformation (IPS deformation), the observed martensite shape is attained as shown in II. However, this leads to the wrong crystal structure. Nevertheless, another homogeneous shear (III) can be applied that leads to correct the crystal structure (bct). However, because the ß 2006 by Taylor & Francis Group, LLC. I II D III A Invariant plane strain Austenite A Observed shape (wrong structure) C B Homogeneous shear D D Martensite (wrong shape) C B C B A Lattice invariant deformation A A D D Twin boundary IV C B B Twinned martensite C Slipped martensite FIGURE 2.9 Schematic illustration of the phenomenological theory of martensite formation (bcc or bct structure) from austenite (fcc structure), showing the intermediate stages before the fcc structure changes to 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 lattice invariant deformations. Therefore, in the next step, an inhomogeneous lattice invariant deformation produces a slipped or twinned martensite that matches the observed shape. In summary, the change in crystal structure from austenite to martensite is achieved through two IPSs and an inhomogeneous lattice invariant strain. Because this is a phenomenological theory, it does not predict the sequences of these deformations. Rather, it defines a method by which the austenite crystal can be transformed to a martensite crystal. Another important feature of martensite transformation in carbon containing steels is that the tetragonal distortion of bct structure (a ¼ b 6¼ c) increases with carbon content, given by the following relation [14]: c ¼ 1 þ 0:045  (wt% C): a An increase in tetragonality also leads to the hardening of martensite. Martensite that forms in low-carbon steel (<0.5 wt% C) shows mainly dislocations, whereas martensite that forms from high-carbon steels shows twins. A typical example of dislocated martensite formed from 0.05 wt% C steel is shown in Figure 2.10. The martensite was attained even with the presence of a low concentration of carbon by rapid quench conditions attained through resistance spot welding. Transmission electron microscopy shows lath martensite with a high dislocation density. ß 2006 by Taylor & Francis Group, LLC. a 50 μm b 2 μm FIGURE 2.10 (a) Optical microstructure of lath martensite observed in a low-carbon steel obtained by rapid cooling in a weld. (b) Transmission electron micrograph of the same microstructure exhibiting parallel and long lath martensite plates with high dislocation density. ß 2006 by Taylor & Francis Group, LLC. The kinetics of martensite is often discussed in terms of nucleation and growth that are related to the thermodynamic free energy of austenite and martensite, defect density, and glissile interfaces between austenite and martensite. The Ms temperature is closely linked to thermodynamic parameter T0 (i.e., the temperature at which the Gibbs free energy of austenite equals that of the martensite with the same composition). However, a certain amount of undercooling is needed before martensite forms from austenite. This undercooling À 0Á is defined by the driving force of austenite to martensite DGg!a and by the amount of strain 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 independent of the carbon concentration. It is now possible to use thermodynamic models to calculate the Ms temperature* by using similar principles for any multicomponent steel [27]. The next step is to describe the nucleation rate of martensite in austenite. Application of homogeneous nucleation theory principles clearly shows that martensite nucleation cannot be described based on pure atomic vibrations due to thermal energy. As a result, heterogeneous nucleation theories 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 the creation of these dislocation loops is the formation of stacking faults in the closely packed planes of austenite [7]. Once the nuclei of martensite is established, the next step is the growth of these crystals. Because the growth of the martensite is modeled as the glide of parallel dislocations, the rate of growth is rapid. Research has shown three types of martensite growth exist: athermal, burst, and isothermal transformations [7,14]. In athermal growth, martensite formation is only a function of undercooling below the Ms temperature and not the time spent at each temperature. This transformation starts exactly at Ms temperature. This mode of transformation also exhibits a strong dependence on thermal cooling conditions. If the quenching of austenite to a low temperature was interrupted and aged at a temperature in between the Ms and Mf temperatures, then further martensite formation would not occur until further undercooling is instituted. This austenite stabilization is related to the carbon segregation to nucleation sites for martensite. Burst transformation involves the formation of a certain fraction of martensite (ranging from a few percent to 50%) from austenite below Ms temperature, instead of a gradual increase as in the case of athermal growth. The time interval for a burst of martensite transformation is about 1 ms. Burst transformation is attributed to an extreme form of autocatalysis (i.e., the formation of one martensite plate triggers the formation of another). Isothermal transformation involves an increase in the martensite fraction with holding time at a given temperature. This mode of transformation is also related to autocatalytic events. The isothermal cooling transformation diagrams (or time temperature transformation [TTT] diagrams) for isothermal martensite formation show a typical C-curve behavior, indicating that both thermal and athermal characters are present. Athermal or burst martensite formation is observed in steels; isothermal martensite formation is observed only in iron alloys that do not contain carbon. 2.4 MICROSTRUCTURE EVOLUTION DURING REHEATING The previous section focused on microstructure evolution due to austenite decomposition that occurs during cooling from the austenitization temperature. In most heat treatments, microstructure evolution during reheating is also important. For example, the tempering of martensite to impart toughness is achieved by reheating the martensite to a high temperature. * Software to calculate Ms temperature can be downloaded freely from www.msm.cam.ac.uk=map. ß 2006 by Taylor & Francis Group, LLC. There is a need to understand the various phase transformations that occur during tempering to achieve an optimum combination of strength and toughness. Austenite formation itself is also important. For example, in the case of induction heat treatment, the depth of hardening is related to the depth of austenite formation. In this section, various phase transformations that occur during tempering and austenite formation are highlighted. 2.4.1 TEMPERED MARTENSITE In principle, various physical processes that lead to tempering of martensite start in as soon as martensite is heated to about room temperature [7]. These physical processes are associated with different temperature ranges. The transformation mechanisms in these stages are predominantly reconstructive modes with short- or long-range transport of atoms. However, in certain cases, decomposition may be possible by just the movement of carbon atoms with only a small distortion of the martensite lattice through displacive transformation [28,29]. 2.4.1.1 Carbon Segregation and Aging of Martensite During 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 C and Fe16 C2 . The carbon atoms may segregate to dislocations or may diffuse to interlathretained austenite. Ohmori and Tamura have postulated that carbon segregation to defects such as dislocations may be the overwhelming factor for the observation of carbon-rich and carbon-depleted regions in the martensite [31]. The evidence for carbon clustering or spinodal decomposition has been obtained indirectly through electron diffraction or atom probe field ion microscopy [28,32]. 2.4.1.2 First Stage of Tempering In 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 be replaced by other elements. The e-carbide has a close-packed hexagonal structure and occurs as narrow laths or rods on cube planes of the martensite with Jack’s orientation relationship [7]: À Á 1 1 2 (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 formed during 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 and would undergo further decomposition on heating to higher temperatures. 2.4.1.3 Second Stage of Tempering In the temperature range of 200 to 3508C, the retained austenite in the steel decomposes into ferrite and cementite. This decomposition was detected successfully by x-ray diffraction and dilatometric and specific volume measurements [7]. The kinetics of this decomposition are related to carbon diffusion in austenite. The untransformed austenite may undergo transformation on application of strain and thus affect the toughness of the steel. However, the fraction of retailed austenite is usually low in steels containing less than 0.2 wt% C. The ß 2006 by Taylor & Francis Group, LLC. formation of stable carbides that is typical to third stage of tempering may overlap in the second stage of tempering. 2.4.1.4 Third Stage of Tempering In the temperature range of 250 to 7508C, cementite precipitates within the martensite. The composition of the cementite is Fe3 C. In alloyed steels, it is referred as M3 C, where M corresponds to substitutional alloying additions (e.g., Cr, Mn) in addition to Fe. Cementite ¨ has an orthorhombic crystal structure and usually occurs as Widmanstatten plates. An example of tempered martensite in 300-M steel is shown in Figure 2.11a. The orientation relationship between ferrite and cementite is of the Bagaryatski type: 1 ½1 0 0Šcementite ==[0 1 "]ferrite ; [0 1 0]cementite ==[" 1 1]ferrite ; (0 0 1)cementite ==(2 1 1)ferrite : 1 The habit planes of cementite can be parallel to either {0 1 1}a or {1 1 2}a of ferrite. The nucleation of cementite may occur at e-carbide and may grow by dissolution of the e-carbide. In high-carbon steels, the cementite precipitates along the twin boundaries of martensite. Other sites for nucleation of cementite are the prior austenite grain boundaries or interlath boundaries. With the formation of cementite, most of the carbon in martensite is removed from solid solution. As a result, the tetragonality of bct structure is lost. Early stages of cementite growth occur only by carbon diffusion with no significant partitioning of substitutional alloying elements [32,33]. However, with extended tempering, redistribution of alloying elements also occurs between ferrite and cementite [32]. In addition, the plate-like cementite particles may coarsen and spheroidize with extended tempering. At this stage, the recovery and recrystallization of martensite laths may also be initiated. In high-carbon martensite, higher order carbides such as M5 C2 (x-carbide) can also form. It has been shown that higher order carbides are actually polytypes of the basic trigonal prism basis of cementite structure [34]. 2.4.1.5 Fourth Stage of Tempering Tempering at higher temperatures (>7008C) leads to the precipitation of more equilibrium alloy carbides such as M7 C3 and M23 C6 . In steels containing Cr, Mo, V, and Ti, these carbides are associated with hardening of the steel that is called secondary hardening. The precipitation of these carbides also leads to the dissolution of cementite. An example of alloy carbide formation by tempering at 6008C in 300-M steel is shown in Figure 2.11b. At this a 0.5 μm b 0.5 μm FIGURE 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) after tempering at 6008C for 1 min, showing alloy carbides. ß 2006 by Taylor & Francis Group, LLC. stage, the recrystallization of martensite lath is more complete, and there is a tendency for the formation of equiaxed grains and extensive grain growth. 2.4.2 AUSTENITE FORMATION The early work on austenite formation by Robert and Mehl [35] focused on the nucleation of austenite from a ferrite–pearlite mixture. Various researchers have since shown the complexity of austenite formation from a two-phase mixture of ferrite and cementite and have made attempts at modeling austenite formation as a function of composition and microstructure [36–41]. On heating steel, with a spheroidized ferrite þ cementite mixture, the austenite phase nucleates at the ferrite–cementite boundary. With further heating, the austenite phase consumes the cementite and then grows into ferrite through diffusion-controlled growth. In a pearlitic microstructure, the austenite may nucleate in the cementite and grow into the colony by dissolving both ferrite and cementite. A typical micrograph of austenite growth into a pearlite colony is shown in Figure 2.12a. A schematic illustration of the growth is shown in Figure 2.12b. Recent work has shown that it is possible to model the above phenomenon with computational tools [42]. It is possible to construct TTT diagrams for the austenite growth for any steel to evaluate the effect of the initial microstructure. These diagrams do not show a C-curve behavior because both the driving force for austenite formation and the diffusivity increase with temperature. This results in monotonic acceleration of austenite formation as the temperature increases. The rate of austenite formation also increases with the presence of residual austenite in the microstructure, as demonstrated by Yang and Bhadeshia, who studied the austenite growth kinetics in a bainitic microstructure [43,44]. The microstructure contained residual austenite films, and there was no requirement for nucleation of austenite and the austenite films grew with an increase in the temperature. After the completion of austenite formation, continued heating leads to grain growth of austenite. The grain growth is also affected by the presence of fine carbonitride precipitates. With the presence of these precipitates the grain boundaries are pinned and therefore, grain growth characteristics are 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 formation from ferrite occurs by a diffusion-controlled reconstructive transformation mechanism. However, at rapid rates, the transformation of ferrite to austenite may occur by interface controlled or by displacive transformation [47]. 2.5 SUMMARY OF STEEL MICROSTRUCTURE EVOLUTION An 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 a parent lattice into a product lattice occurs through a noncoordinated motion of atoms across the interface. The growth mostly occurs by nucleation and growth of product phases. Reconstructive transformations are slow below 6008C. The formation of allotriomorphic ferrite, idiomorphic ferrite, massive ferrite, pearlite, carbide, and austenite all belong to the category of reconstructive transformation. Displacive transformation involves coordinated atom, causing a change from a parent crystal to a product crystal. This change is achieved by IPS deformation with a large shear component, leading to a plate or lath shape. During this transformation, the substitutional atoms do not diffuse. However, displacive transformations occurring at high temperatures (above Ms ) may involve varying amounts of interstitial carbon ¨ diffusion during nucleation and growth. In the case of Widmanstatten ferrite formation, both nucleation and growth involve carbon diffusion. In contrast, in bainitic ferrite formation, ß 2006 by Taylor & Francis Group, LLC. a Allotriomorphic ferrite Pearlite colony Austenite/pearlite boundary Allotriomorphic ferrite 10003 5 μm Pearlite colony b Austenite FIGURE 2.12 (a) Optical micrograph of a steel sample with ferrite þ pearlite microstructure heated to an 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 and eventually 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 use theoretical formulations to describe this varying carbon supersaturation during displacive transformation within a coupled diffusional-displacive transformation framework [48–50]. 2.6 PREDICTION OF MICROSTRUCTURE EVOLUTION DURING HEAT TREATMENT The final properties of heat-treated parts depend on the microstructure that evolves during the heat-treating process. The microstructure evolution is in turn controlled by the complex ß 2006 by Taylor & Francis Group, LLC. Reconstructive Displacive Diffusion of all atoms during nucleation and growth Sluggish below about 850 K Invariant-plane strain shape deformation with large shear component No iron or substitutional solute diffusion and thin plate shape Allotriomorphic ferrite Idiomorphic ferrite Massive ferrite No change in bulk composition Pearlite Cooperative growth of ferrite and cementite Widmanstätten ferrite Carbon diffusion during paraequilibrium nucleation and growth Bainite and acicular ferrite Carbon diffusion during paraequilibrium nucleation No diffusion during growth Martensite Diffusionless nucleation and growth Austenite formation Full austenitization and intercritical annealing Tempering reaction Cementite and alloy precipitation, laves phase precipitation and coarsening FIGURE 2.13 Classification of steel microstructure evolution during heating and cooling based on the mechanism of transformation. (Adapted from H.K.D.H. Bhadeshia, Bainite in Steels, 2nd ed., the Institute 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, in conjunction with an extensive experimental database, often achieves the required properties. However, this approach has the following limitations: (1) it is rigid and cannot be changed easily to meet the ever-increasing demand for optimization of cost and quality and to meet the requirements of strict codes and standards and (2) it cannot be extended to newly developed steels. In this regard, tools that can predict the evolution of microstructure as a function of composition and heat-treatment history will be useful. These predictive tools should be capable of addressing the effect of alloy composition on the stability of various phases. In addition, the effect of time–temperature histories on the growth and dissolution of these phases 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 transformation diagrams as a function of steel composition (see Figure 2.14). The figure on the left-hand side shows an iron-rich corner of the Fe–C phase diagram showing the phase stability at different temperature. The schematic figure in the middle shows the initiation of austenite formation from ferrite as a function of heating curves. The schematic figure in the right shows the initiation of ferrite formation from austenite as a function of different cooling rate. With the ß 2006 by Taylor & Francis Group, LLC. Temperature ( C) 1000 Austenite CHT (α + Fe3C to γ) Cooling Ferrite + austenite 800 CCT (γ to α) 600 Heating 400 Ferrite + cementite 200 0.0 0.1 0.2 0.3 24 0.4 −1 10 Weight percentage carbon 24 24 100 101 Time (s) 102 10−1 100 101 Time (s) 102 FIGURE 2.14 Schematic illustration, showing the importance of continuous heating transformation (CHT) and continuous cooling transformation (CCT) diagrams with reference to an Fe–C metastable binary diagram. framework of microstructural evolution presented in the earlier sections and state-of-the-art computational and kinetic models presented in the following sections, it is possible to predict microstructural evolution during heat treatment of steels as a function of composition and thermal history. 2.6.1 CALCULATION OF MULTICOMPONENT MULTIPHASE DIAGRAMS The stability of a phase is governed by its free energy, which is a function of temperature and its constitution. A generic description of free energy of a solid-solution phase, f, Gf , is given by the following equation [51]: Gf ¼ Gf þ Gf Àmix þ Gf Àmix , o ideal excess (2:1) where Gf is the free energy contribution from pure components in that phase, Gf Àmix is the o ideal contribution from ideal mixing, and Gf Àmix is the contribution due to nonideal interexcess actions between the components. Many thermodynamic models describe various phases (see Ref. [15]). With the description of this Gf for all phases that can form in a given alloy, it is possible to estimate equilibrium fractions of each phase and their constitution at a given temperature. This estimation is performed by minimization of the free-energy curves of the various phases. This procedure also allows for determination of the tie line, which is schematically shown in Figure 2.15. The tie line is represented by a common tangent and is mathematically represented by the following equation: ma ¼ mb ; ma ¼ mb , A B B A (2:2) where mf is the chemical potential of element i in the f-phase. The chemical potentials i are obtained from the free-energy expressions given in Equation 2.1 with the following equation: ß 2006 by Taylor & Francis Group, LLC. β-phase Free energy (J/mol) α-phase xβ B mβ A mβ B mα B xα B Tie line mα A A B Concentration of B FIGURE 2.15 Schematic illustration of the methodology used to calculate phase stability in a binary alloy: free energy of an a-phase and a b-phase and the common-tangent construction for the description of the tie line and chemical potential of elements A and B. mf ¼ Gf À xf B A   ›G f ; mf ¼ Gf þ 1 À xf : B B ›x f ›xf B B ›Gf (2:3) Equation 2.2 and Equation 2.3 can be extended to multicomponent systems by invoking the equality of the chemical potentials of all the components. The following equation determines the governing equilibrium between a- and g- phases in the Fe–Cr–Ni–C alloy system: mg ¼ ma ; mg ¼ ma ; mg ¼ ma ; mg ¼ ma : C Fe Cr Ni Fe C Cr Ni (2:4) Using the relations in Equation 2.4 and mathematical methods, a phase equilibria between the g- and a-phases as a function of Cr, Ni, and C concentrations can be calculated. This methodology has proven to be useful for describing the phase stability in multicomponent alloys such as aluminum alloys, stainless steels, low-alloy steels, and Nibase alloys [52–57]. For example, a simple Fe–C binary diagram is compared with an (Fe,Cr,Mo)–C quasibinary phase diagram in Figure 2.16. The diagram shows that a simple ferrite þ austenite þ cementite phase equilibrium is modified to ferrite þ austenite þ cementite þ M23 C6 þ M6 C þ M3 C2 þ MC due to the addition of Cr and Mo to the Fe–C systems. ThermoCalc software was used to construct these diagrams with solid solution database [58]. 2.6.2 CALCULATION OF DIFFUSION-CONTROLLED GROWTH Although the phase stability calculation allows the estimation of the equilibrium microstructure at a particular temperature, microstructure control in most heat-treatment processes relies on the kinetics of product-phase formation from the parent phase. For example, in dual phase low-alloy steels, it is important to understand the kinetics of austenite formation to control 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 is formed during processing. Both phenomena can be described by diffusion-controlled growth models by coupling thermodynamic models with diffusion-controlled growth calculations. The methodology is presented below. ß 2006 by Taylor & Francis Group, LLC. Ferrite + austenite M23C6 1000 Temperature ( C) Austenite Austenite Ferrite + austenite 800 Ferrite + M23C6 600 Ferrite + M23C6 + M6C 400 Ferrite + M23C6 + M6C + Laves Ferrite + cementite Ferrite + M23C6 + M3C2 + MC 200 0.0 0.1 0.2 0.3 Weight percentage carbon 0.4 0.1 0.2 0.3 0.4 Weight percentage carbon FIGURE 2.16 Comparison of a calculated Fe–C binary-phase diagram with an Fe–Cr–Mo–C 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 phase involves diffusion of partitioning elements. If local equilibrium exists between the parent and product phases at the interface, the interfacial concentrations can be given by the tie lines drawn in the phase diagram. Given this condition, it is possible to describe the movement of this interface as a function of temperature and time by solving Fick’s second law and by maintaining a mass balance at the interface. The governing equation for a-phase formation in g-phase in one-dimensional geometry is given in the following example [59,60]: À m,a CI m,g Á À CI   Àm Á m (dl =dt) ¼ Dm dCg =dx À Dm dCa =dx : g a (2:5) m ,a In the above equation, (dl =dt) is the rate of interface movement or velocity. The terms CI m,g and CI are the interface concentrations of element m in the a- and g-phases. The terms Dm a m and Dm are the diffusivities of element m in a- and g-phases. The terms ðdCg =dxÞ and À mg Á 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 multicomponent alloy, Equation 2.5 must be satisfied for all elements, and a unique velocity of the interface is obtained. This restriction leads to the selection of the interface compositions as determined by the 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 the diffusion equations and require close coupling with thermodynamic models. This is the methodology implemented in the DicTra software [61] and other published works. An example calculation is presented below. The final microstructure of a stainless steel weld often contains ferrite and austenite at room temperature. This microstructure is far from equilibrium. However, given a high-temperature heat treatment, equilibration of this microstructure will occur. During this heat treatment, the ferrite may grow or dissolve, depending upon the alloy composition, the initial state, and the heat treatment temperature. Examples are shown in Figure 2.18. In case A, the initial austenite composition was Fe–15%Cr–20%Ni * Ideally, the chemical potential gradient needs to be used instead of concentration gradient. ß 2006 by Taylor & Francis Group, LLC. Concentration α CIm,α γ CIm,γ dl dt Distance FIGURE 2.17 Schematic illustration of interface concentrations and diffusion in both a- and b- phases to describe diffusion-controlled transformations. (wt%), and the initial ferrite composition was Fe–20%Cr–10%Ni (wt%). The initial volume percentage of ferrite was 9%. The calculations predicted rapid dissolution of ferrite if this weld was aged at 13008C. In case B, the initial austenite composition was Fe–21%Cr–11%Ni (wt%), and the initial ferrite composition was Fe–30%Cr–4.5%Ni (wt%). The initial volume percentage of ferrite was again set to 9%. The calculations predicted that the volume percentage of ferrite would increase from 9 to 25% if the weld was aged at 13008C. In addition, the calculations showed different stages where the kinetics of this transformation were controlled by either solute diffusion in ferrite or solute diffusion in austenite. This condition is due to the large differences in the diffusivity of Cr and Ni in the ferrite and austenite phases. The above methodology has been applied to continuous cooling and heating conditions. Such calculations allow for the design of postweld heat treatment of stainless steel welds. However, the local equilibrium assumption used in the above model leads to some paradoxes for simulating growth below ~6008C. At temperatures below 6008C, the diffusion profile of substitutional atoms (see Figure 2.17) may be extremely steep, with a width that is smaller than interatomic distances. Hultgren and other authors have addressed this issue [62– 65]. It is hypothesized that under these conditions, the concentration ratio of substitutional atoms to iron atoms will remain configurationally frozen and that interface growth will be controlled purely by carbon (interstitial) diffusion. The carbon concentration at the interface will now be determined by a phase-boundary calculation similar to the Fe–C 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 to correlate well with the accelerated ferrite growth observed under large undercooling in most heat treatment and welding conditions. In steels, PE growth rates are always higher than 25 9% Ferrite 91% Austenite Ferrite % 20 15 10 5 Case A Case B 0 0.01 0.1 1 10 Time (s) 100 1000 FIGURE 2.18 Variation of ferrite percentage while aging at 1573 K for two initial conditions: (A) corresponds to Fe–15Cr–20Ni (wt%) austenite and Fe–20Cr–10Ni (wt%) ferrite; (B) corresponds to Fe– 21Cr–11Ni (wt%) austenite. (The ferrite composition was Fe–30Cr–4.5Ni [wt%] ferrite.) ß 2006 by Taylor & Francis Group, LLC. 10−5 Thickening rate (m s−0.5) 4 2 10−6 4 2 10−7 4 Paraequilibrium Local equilibrium 2 10−8 600 650 700 750 Temperature (8C) 800 FIGURE 2.19 Calculated parabolic thickening rate of ferrite growth into austenite in an Fe–0.1C– 2.25Cr–1Mo steel under (.) local equilibrium and (*) paraequilibrium. growth rates calculated by local equilibrium models. This discrepancy is demonstrated by a comparison of the parabolic thickening rates of ferrite into austenite in an Fe–Cr–Mo–C steel at different temperatures (see Figure 2.19). It is now possible to use models to predict TTT and continuous cooling transformation (CCT) diagrams with commercial software that are available on the Internet and from the commercial organizations [66–68]. Typical TTT and CCT diagrams calculated with the JMatPro software are shown in Figure 2.20a and Figure 2.20b, respectively. A TTT and CCT diagram for 0.1% transformation to ferrite calculated by using an Internet tool is shown in Figure 2.20c. The methods of calculation for each tool are different and are constantly refined. As a result, these results must be used as guidance rather than as accurate predictions. However, with the advent of new models [69] and detailed thermodynamic and diffusion data, an accurate description of transformation in steels should be possible. The final goal of these predictive approaches is to couple these microstructure models with thermal and structural models to describe the overall response of steel structures to heat treatment as envisioned by Kirkaldy [70] (see Figure 2.21). In this integrated model, it would be possible to describe transient thermal fluctuations in a steel part during heat treatment in a furnace and during quenching or cooling. In addition, during such thermal responses, it should be possible to describe the thermal stress evolution based on the constitutive relations of different phases. By coupling all the above parameters with a microstructure 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 different heat treatments as a function of steel composition. 2.7 SUMMARY In 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 important microstructures observed during steel heat treatment, including allotriomorphic ferrite, Wid¨ manstatten ferrite, bainite, pearlite, martensite, and carbide formation during tempering and austenite 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. ß 2006 by Taylor & Francis Group, LLC. (a) Composition (wt%): Fe: 97.93 Mn: 1.45 Si: 0.25 C: 0.37 Transitions: Pearlite: 710.68 C Bainite: 565.04 C Ferrite: 769.61 C Martensite: 337.77 C TTT HSLAS 800 Temperature (8C) 700 600 500 Ferrite (0.1%) Pearlite (0.1%) Bainite (0.1%) Pearlite (99.9%) Bainite (99.9%) 400 300 0.1 1000 10 Time (s) Grain size: 6.0 ASTM Austenitization: 819.61 C Composition (wt%): Fe: 97.93 Mn: 1.45 Si: 0.25 C: 0.37 Transitions: Pearlite: 710.69 C Bainite: 565.04 C Ferrite: 769.62 C Martensite: 337.78 C CCT HSLAS (b) 900 Temperature (8C) 800 700 600 Ferrite (0.1%) Pearlite (0.1%) Bainite (0.1%) Pearlite (99.9%) Bainite (99.9%) 100.0 C/s 10.0 C/s 1.0 C/s 0.1 C/s 500 400 300 0.1 10 1000 Time (s) Grain size: 6.0 ASTM Austenitization: 819.61 C (c) TTT 1100 CCT Temperature (K) 1000 900 800 Ba 700 Ma ORNL 600 10−1 100 101 102 103 104 Time (s) FIGURE 2.20 Calculated (a) time temperature transformation (TTT) diagram and (b) continuous cooling transformation (CCT) diagram for a steel using JMatPro1 software for an Fe–0.37C–0.25Si– 1.45Mn steel. (c) TTT and CCT transformation diagram is calculated using the tool found on the Internet at http:= engm01.ms.ornl.gov. = ß 2006 by Taylor & Francis Group, LLC. Chemical composition Austenitizing condition Austenite grain size Thermal boundary conditions Heat transfer coefficients Thermophysical properties Transformation behavior Transient temperature field Creep deformation ion at rm t ) ) fo , re ns f (T ctu tra = ru of ion , st at (T at he rm = f sfo ies an ert p ro .p Tr nt (Heat generation due to deformation) Thermal stress te Component geometry FE - mesh La at M [transformation Mechanical response displ. s, e = f (x,y,z,t ) Transformation kinetics Phase diagram TTT diagram T = f (x,y,z,t ) Microstructural = f (s )] s − ε curve development = f (T, structure) Transformation strain phases = f (x,y,z,T,t ) Transformation plasticity Mechanical boundary cond. loading, material properties Hardness distribution FIGURE 2.21 Modeling phase transformations in steels, in response to thermomechanical conditions. (Adapted from J.S. Kirklady, Scand. J. Metall., 20, 1991.) 2.8 ACKNOWLEDGMENTS Part of the results presented in this paper is based on the research sponsored by the Division of Materials Sciences and Engineering and Assistant Secretary for Energy Efficiency and Renewable 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 this chapter. Finally, the author would like to thank Drs. M.K. Miller and Dr. J.M. Vitek of ORNL for helpful suggestions and review of the manuscript. REFERENCES 1. J. Agren, Scr. Mater., 46, 2002, 893–8. 2. J.W. Christian, The Theory of Transformations in Metals and Alloys, 2nd ed., Part 1, Pergamon Press, 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. Martensite—A 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, 550–95. 9. A. Hultgren, Trans. ASM, 39, 1947, 915–1005. 10. H.I. Aaronson, Decomposition of Austenite by Diffusional Processes, Interscience, New York, 1962. ß 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 and J.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, 477–86. 17. H.I. Aaronson, Metall. Trans. A, 24A, 1993, 241–76. 18. G. Spanos, W.T. Reynolds, Jr., and R.A. Vandermeer, Metall. Trans. A, 22A, 1991, 1367–80. 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, 1271–83. 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, 134–80. 24. H.K.D.H. Bhadeshia and J.W. Christain, Metall. Trans. A, 21A, 1990, 767–97. 25. S.S. Babu and H.K.D.H. Bhadeshia, Mater. Sci. Eng., A156, 1992, 1–9. 26. M. Hillert, Scr. Mater., 47, 2002, 137–8. 27. H.K.D.H. Bhadeshia, Met. Sci., 16, 1982, 159–65. 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, 2749–65. 29. B.P.J. Sandvik, Metall. Trans. A, 13A, 1982, 789–800. 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, 2717–37. 31. Y. Ohmori and I. Tamura, Metall. Trans. A, 23A, 1992, 2147–58. 32. S.S. Babu, K. Hono, and T. Sakurai, Metall. Trans. A, 25A, 1994, 499–508. 33. R.C. Thomson and M.K. Miller, Acta Mater., 46, 1998, 2203–13. 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, 613–50. 36. R.R. Judd and H.W. Paxton, Trans. AIME, 242, 1968, 206–15. 37. G. Molinder, Acta Metall., 4, 1956, 565–71. 38. M. Hillert, K. Nilsson, and L.E. Trondahl, JISI, 209, 1971, 49–66. 39. T. Akbay, R.C. Reed, and C. Atkinson, Acta Metall. Mater., 42, 1994, 1469–80. 40. C. Garcia De Andres, F.G. Caballero, C. Capdevilla, and H.K.D.H. Bhadeshia, Scr. Mater., 39, 1998, 791–6. 41. F.G. Caballero, C. Capdevilla, and C. Garcia De Andres, Mater. Sci. Tech., 17, 2000, 1114–8. 42. A. Jacot and M. Rappaz, Acta Mater., 47, 1999, 1645–51. 43. J.R. Yang, C.Y. Huang, Mater. Chem. Phys., 35, 1993, 168–75. 44. J.R. Yang and H.K.D.H. Bhadeshia, Mater. Sci. Eng. A, 131, 1991, 99–113. 45. M. Chapa, S.F. Medina, V. Lopez, and B. Fernandez, ISIJ Int., 42, 2002, 1288–96. 46. T. Gladman, Heat Treat. Met., 21, 1994, 11–4. 47. W.J. Kaluba, R. Taillard, and J. Foct, Acta Mater., 46, 1998, 5917–27. 48. G.B. Olson, H.K.D.H. Bhadeshia, and M. Cohen, Acta Metall., 37, 1989, 381–9. 49. G.B. Olson, H.K.D.H. Bhadeshia, and M. Cohen, Metall. Trans. A, 21A, 1990, 805–9. 50. S.A. Mujahid and H.K.D.H. Bhadeshia, J. Phys. D, 34, 2001, 2573–80. 51. N. Saunders and A.P. Miodownik, CALPHAD-Calculation of Phase Diagrams—A Comprehensive Guide, 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, 763–74. 53. S.S. Babu, S.A. David, and M.K. Miller, Appl. Surf. Sci., 94=95, 1996, 280–87. 54. S.A. David, J.M. Vitek, S.S. Babu, L. Boatner, and R.W. Reed, Sci. Tech. Weld. Join., 2, 1996, 79–88. 55. P. Hofer, M.K. Miller, S.S. Babu, S.A. David, and H. Cerjak, Metall. Mater. Trans. A, 31A, 2000, 975–984. ß 2006 by Taylor & Francis Group, LLC. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. M.K. Miller, S.S. Babu, and M.G. Burke, Mater. Sci. Eng., A270, 1999, 14–8. S.S. Babu, S.A. David, J.M. Vitek, and R.W. Reed, Sci. Tech. Weld. Join., 6, 2001, 31–40. B. Sundman, B. Jansson, and J.O. Andersson, CALPHAD, 9, 1985, 153–90. J.M. Vitek, S.A. Vitek, and S.A. David, Metall. Mater. Trans. A, 26A, 1995, 2007–25. H.K.D.H. Bhadeshia, Prog. Mater. Sci., 29, 1985, 321–86. ˚ J. Agren, ISIJ Int., 32, 1992, 292–6. A. Hultgren, Jerkontorets Ann., 135, 1951, 403. J.B. Gilmour, G.R. Purdy, and J.S. Kirkaldy, Metall. Trans., 3, 1972, 3213–22. G. Ghosh and G.B. Olson, Acta Mater., 50, 2002, 2009–19. E. Kozeschnik, J. Phase Equilibria, 21, 2000, 336–41. http:==www.msm.cam.ac.uk=map http:==engm01.ms.ornl.gov, Modeling Microstructure Development in Welds, Internet online computational tool, Oak Ridge National Laboratory, Oak Ridge, Tenn. 68. JMatPro, http:==www.thermotech.co.uk 69. J. Odqvist and M. Hillert, Acta Mater., 50, 2002, 3211–25. 70. J.S. Kirkaldy, Scand. J. Metall., 20, 1991, 50–61. ß 2006 by Taylor & Francis Group, LLC. 3 Fundamental Concepts in Steel Heat Treatment Alexey V. Sverdlin and Arnold R. Ness CONTENTS 3.1 3.2 3.3 3.4 3.5 3.6 Introduction ............................................................................................................... 122 Crystal Structure and Phases ..................................................................................... 122 3.2.1 Crystal Structure of Pure Iron........................................................................ 122 3.2.2 Iron–Carbon Equilibrium Diagram ............................................................... 123 3.2.2.1 Metastable Fe–Fe3C Equilibrium Diagram...................................... 123 3.2.2.2 Stable Fe–C Equilibrium Diagram................................................... 124 3.2.3 Effect of Carbon............................................................................................. 125 3.2.4 Critical (Transformation) Temperatures ........................................................ 127 Structural Transformations in Steel ........................................................................... 128 3.3.1 Austenite–Pearlite Transformation................................................................. 128 3.3.2 Structure of Pearlite........................................................................................ 129 3.3.3 Transformation of Austenite in Hypo- and Hypereutectoid Steels ................ 130 3.3.4 Martensite Transformation ............................................................................ 131 3.3.5 Morphology of Ferrous Martensites .............................................................. 133 3.3.6 Bainite Transformation .................................................................................. 134 3.3.7 Morphology of the Bainite Transformation................................................... 135 3.3.8 Tempering....................................................................................................... 135 Kinetics of Austenite Transformation ....................................................................... 137 3.4.1 Isothermal Transformation Diagrams............................................................ 137 3.4.2 Continuous-Cooling Transformation Diagrams ............................................ 138 3.4.2.1 Transformations That Take Place under Continuous Cooling of Eutectoid Steels ............................................................................ 139 3.4.2.2 Transformations of Austenite on Cooling in the Martensite Range ............................................................................. 141 3.4.3 Derivation of the Continuous-Cooling Transformation Diagram from the Isothermal Transformation Diagram .............................................. 142 3.4.4 Continuous-Cooling Transformation Diagram as a Function of the Bar Diameter........................................................................................ 143 3.4.5 Definition of Hardenability ............................................................................ 145 Grain Size .................................................................................................................. 146 3.5.1 Structure of Grain Boundaries ....................................................................... 146 3.5.1.1 Structural Models............................................................................. 147 3.5.2 Determination of Grain Size .......................................................................... 149 3.5.3 Austenite Grain Size Effect and Grain Size Control ...................................... 150 3.5.4 Grain Size Refinement.................................................................................... 152 Strengthening Mechanism in Steel ............................................................................. 153 ß 2006 by Taylor & Francis Group, LLC. 3.6.1 3.6.2 3.6.3 3.6.4 3.6.5 Solid Solution Strengthening .......................................................................... 153 Grain Size Refinement.................................................................................... 154 Dispersion Strengthening ............................................................................... 154 Work Hardening (Dislocation Strengthening)................................................ 158 Thermal Treatment of Steels .......................................................................... 159 3.6.5.1 Annealing ......................................................................................... 159 3.6.5.2 Quenching (Strengthening Treatment) ............................................. 162 3.6.5.3 Tempering......................................................................................... 163 References .......................................................................................................................... 163 Further Reading ................................................................................................................. 163 3.1 INTRODUCTION The purpose of heat treatment is to cause desired changes in the metallurgical structure and thus in the 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, and therefore structural changes in iron–carbon alloys are considered in this chapter. In general, the most stable steel structures are produced when a steel is heated to the high-temperature austenitic state (to be defined later) and slowly cooled under near-equilibrium conditions. This type of treatment, often referred to as annealing or normalizing, produces a structure that has a low level of residual stresses locked within the part, and the structures can be predicted from an equilibrium diagram. However, the properties that interest heat treaters the most are those exhibiting high strength 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 their properties. It is essential that heat treaters have a clear understanding of the structures that can be produced in steel under different treatment conditions that they can apply in their equipment. 3.2 3.2.1 CRYSTAL STRUCTURE AND PHASES CRYSTAL STRUCTURE OF PURE IRON Iron in the solid state is known in two allotropic states. (Allotropy is the phenomenon of an element having different crystal lattices depending on the particular temperature and pressure.) 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 into g-iron crystals possessing a face-centered cubic (fcc) lattice. The g crystals retain stability up to temperature of 14008C (25008F). Above this temperature they again acquire a bcc lattice and are usually called d crystals. The d crystals differ from a crystals only in the temperature region 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 heated to 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 7708C is 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 conventionally symbolized 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 are observed in carbon alloys of iron. The subscript 2 denotes a magnetic transformation of the a-phase, while the subscripts 3 and 4 stand for transformation of a to g and g to d. ß 2006 by Taylor & Francis Group, LLC. Temperature, C 1600 1539 1400 δ-Fe Ac4 1400 1200 1000 900 800 700 600 1528 δ-Fe Ar4 γ-Fe γ-Fe 910 906 Ar3 Ac3 770 Ar2 α-Fe Ac2 α-Fe Time FIGURE 3.1 Heating and cooling curves for pure iron. In going from one form to another, iron is capable of undercooling. This causes a difference in the position of transformation points on heating and cooling. The difference depends on the cooling rate (Figure 3.1) and is termed hysteresis. The letters ‘‘c’’ and ‘‘r’’ indicate 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 the volume of the material. Sometimes this gives rise to stresses that exceed the elastic limit and lead to failure. The density of g-Fe is about 4% higher than that of a-Fe. 3.2.2 IRON–CARBON EQUILIBRIUM DIAGRAM The structure of iron–carbon alloys can contain either pure carbon (graphite) or a chemical compound (cementite) as the carbon-enriched component. Cementite is present even in relatively slowly cooled alloys: a long holding at elevated temperatures is required to decompose cementite to iron and graphite. For this reason the iron–carbon diagram is usually treated as the iron–iron carbide diagram. The former is stable, whereas the latter is metastable. The iron–carbon diagram is shown in Figure 3.2. Dashed lines stand for the stable Fe–C diagram, and solid lines denote the metastable Fe–Fe3C diagram. 3.2.2.1 Metastable Fe–Fe3C Equilibrium Diagram As shown in Figure 3.2, the lattices of allotropic forms of iron (d, g, and a) serve as sites of formation of d, g, and a solid solutions of carbon in iron (the same symbols are adopted for the designation of solid solutions). When carbon-depleted alloys crystallize, crystals of the d solid solution precipitate at the liquidus AB and solidus AH. The d solid solution has a bcc lattice. At the maximum carbon temperature of 14908C (27148F), the d solution contains 0.1% C (point H). At 14908C a peritectic 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 the liquid (at the liquidus BC and solidus IE). At 11308C (20668F) the limiting solubility of carbon 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 carbide from the g solution. As the carbon content is raised, the temperature at which the g lattice transforms to the a lattice lowers, and the transformation takes place over the temperature interval corresponding to the curves GS and GP. ß 2006 by Taylor & Francis Group, LLC. A L 1400 δ + L B H δ γ+δ 1300 N D D γ+L 1200 Fe3C + L (1.98) C (1135 ) F Temperature, 8C E E 2.0 1100 1130 F C 4.3 1000 900 800 700 γ + Fe3C G 9108 α+γ P 768 P (0.69) S 738 K S 0.80 723 K 0.025 500 210 200 A0 α + Fe3C 100 0 1 2 3 4 5 Carbon, % FIGURE 3.2 Iron–carbon diagram. The a-phase precipitation curve GS intersects the iron carbide precipitation curve ES. The point S is a eutectoid point with the coordinates 7238C (13338F) and 0.80% C. At this point a saturated a solution and Fe3C precipitate simultaneously form the eutectoid concentration g solution. The lattice of the a solid solution is identical to the lattice of the d solid solution. At the eutectoid 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 it equals a small fraction of a percent (point D). When the carbon content is 2.0–4.3%, crystallization starts with precipitation of the g solution at the line BC. An increase in the carbon content to over 4.3% causes precipitation of iron carbide at the line CD. Precipitation of the surplus primary phase in all alloys containing over 2.0% C is followed by a eutectic crystallization of the g solution and iron carbide at point C, whose coordinates are 11308C (20668F) and 4.3% C. The line Ao is associated with a magnetic transformation, that is, a transition from the ferromagnetic to the paramagnetic state. Table 3.1 describes structural components of the iron–carbon system. 3.2.2.2 Stable Fe–C Equilibrium Diagram Given very low rates of cooling, carbon (graphite) can crystallize directly from the liquid. In this case, a eutectic mixture of austenite and graphite is formed instead of the eutectic of austenite and cementite. As is seen in Figure 3.2, the dashed lines symbolizing the iron– graphite system are at higher temperatures than the lines of the iron–cementite system. This testifies to the greater stability and closeness to a full equilibrium of the iron–graphite system. ß 2006 by Taylor & Francis Group, LLC. TABLE 3.1 Components of the Iron–Carbon System Phase or Mixture of Phases Name Solid solution of carbon in a (d) iron Solid solution of carbon in g-iron Iron carbide (Fe3C) Eutectoid mixture of carbon solid solution in g-iron with iron carbide Eutectoid mixture of carbon solid solution in a-iron with iron carbide Ferrite Austenite Cementite Ledeburite Pearlite The conclusion is also supported by the fact that heating of high-carbon alloys with a large amount of cementite leads to its decomposition: Fe3C ! 3Fe þ C. At intermediate rates of cooling, part of the alloy can crystallize according to the graphite system and the other part according to the cementite system. Phase equilibrium lines in the diagrams of both the systems can be displaced depending on particular cooling rates. A most pronounced displacement can be observed for the lines of precipitation of the carbon solid solution in g-Fe (austenite). For this reason the diagram holds completely true only with respect to the alloys that are subjected to a relatively slow cooling rate. 3.2.3 EFFECT OF CARBON A maximum solubility of carbon in a-Fe is observed at 7218C (13308F) and is equal to 0.018% C. Subject to quenching, carbon can remain in the a solid solution, but soon precipitation of phases commences, by an aging mechanism. In the a solid solution, carbon can form either (1) a homogeneous solution, a statically uniform interstitial distribution (a rare case), or (2) an inhomogeneous solution, with the formation 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 clusters thus formed represent an obstacle to movement of dislocations during plastic deformation and are responsible for an inhomogeneous development of the deformation at the onset of plastic flow. To analyze the influence of the carbon content on iron–carbon alloys, every structural component should be characterized. Slowly cooled alloys comprise ferrite and cementite or ferrite and graphite. Ferrite is plastic. In the annealed state, ferrite has large elongation (about 40%), is soft (Brinell hardness is 65–130 depending on the crystal dimension), and is strongly ferromagnetic up to 7708C (14188F). At 7238C (1338F), 0.22% C dissolves in ferrite, but at room 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 weakly magnetic up to 2108C (4108F), is a poor conductor of electricity and heat, and has a complicated rhombic lattice. Usually a distinction is made between primary cementite, which crystallizes from the liquid at the line CD; secondary cementite, which precipitates from the g solution at the 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 does not melt even at temperatures of 3000–35008C (5430–63308F). It possesses a hexagonal lattice with the axis relation c > 2: a ß 2006 by Taylor & Francis Group, LLC. FIGURE 3.3 Steel microstructure of ferrite and tertiary cementite at grain boundaries, 500Â. Austenite is soft (but is harder than ferrite) and ductile. Elongation of austenite is 40–50%. It has lower conductivity of heat and electricity than ferrite, and is paramagnetic. Austenite possesses an fcc lattice. The structure of the steel containing 0–0.02% C comprises ferrite and tertiary cementite (Figure 3.3). A further increase in the carbon content leads to the appearance of a new structural component—a eutectoid of ferrite and cementite (pearlite). Pearlite appears first as separate 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 along with pearlite. The secondary cementite is shaped as needles (Figure 3.5). The amount of cementite increases as the carbon content is increased. At 2% C it occupies 18% of the field of vision of the microscope. A eutectic mixture appears when the carbon content exceeds 2%. In rapidly cooled steels, not all the surplus phase (ferrite or cementite) has time to precipitate before a eutectoid is formed. Alloys with 3.6% C contain ledeburite (a eutectic mixture of carbon solid solution in g-Fe and iron carbide). An electron microscopic image of the carbides is shown in Figure 3.6. The alloys would be more properly classified with hypoeutectic white cast irons. FIGURE 3.4 Steel microstructure of pearlite, 500Â. ß 2006 by Taylor & Francis Group, LLC. FIGURE 3.5 Steel microstructure of secondary cementite (needles) and pearlite, 500Â. 3.2.4 CRITICAL (TRANSFORMATION) TEMPERATURES Carbon has a pronounced effect on transformations of iron in the solid state. The positions of the lines GS and NI in the iron–carbon equilibrium diagram show that an increase in the carbon content leads to lowering of the point A3 and raising of the point A4 with respect to their counterparts depicted in Figure 3.2 for pure carbon. So carbon extends the temperature range of the d-phase. When a eutectoid (pearlite) is formed, heating and cooling curves exhibit a stop, which is designated as the point A1 (Ac1 on heating and Ar1 on cooling). This phenomenon takes place at 0.9% C (point S in the Fe–C diagram). Precipitation of ferrite in hypoeutectoid steels (on crossing the line GOS) shows up in heating and cooling curves as an inflection symbolized as the point A3. The point corresponds to the g ! a transformation in pure iron. Precipitation of cementite (crossing of the line ES), which precedes the eutectoid precipitation, is seen in the cooling curve as a weak inflection designated as the point Acm (Ac,cm on heating and Ar,cm on cooling). Addition of carbon has little influence on the magnetic transformation temperature (point A2). Therefore, the line MO corresponds to the magnetic transformation in alloys with 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. If the carbon content is higher than the one corresponding to point S, then the magnetic transformation coincides with the temperature A1. FIGURE 3.6 Steel microstructure of electron microscopic image of iron carbides, 3000Â. ß 2006 by Taylor & Francis Group, LLC. Cementite undergoes a magnetic transformation. Whatever the carbon content, the transformation takes place at a temperature of 210–2208C (410–4308F). It occurs without a marked hysteresis, as does the magnetic transformation of pure iron at point A2. 3.3 STRUCTURAL TRANSFORMATIONS IN STEEL When a steel part is hardened, it is heated to a high temperature in order to convert the entire structure to the austenite phase. As discussed earlier, austenite is a single-phase structure of iron and carbon stable at high temperatures. If the steel were cooled slowly, the austenite would transform to pearlite, which is the equilibrium phase at room temperature. A pearlitic structure is an annealed structure and is relatively soft with low physical properties. If the steel is cooled very rapidly, a very hard and strong structure called martensite forms that is a metastable phase of carbon dissolved in iron. It may be tempered to produce lower hardness structures that are less brittle. Intermediate cooling rates will produce other structures referred to as bainites, although this type of structure is only produced in quantity in an alloy steel. Eutectoid carbon steels produce predominantly martensite or pearlite, depending on the cooling rate. 3.3.1 AUSTENITE–PEARLITE TRANSFORMATION Transformation of the fcc lattice of austenite to the bcc lattice of ferrite is hampered due to the presence of dissolved carbon in austenite. The austenite lattice has enough space to accommodate carbon atoms at the centers of unit cells. The bcc lattice of ferrite has no such space. For this reason the solubility of carbon is lowered considerably on transition from austenite to ferrite. During the b ! a transformation, almost all carbon precipitates from the austenite lattice. In accordance with the metastable Fe–Fe3C diagram, it precipitates as iron carbide (cementite). This transformation can be described by three interconnected processes: 1. Transformation of the g-Fe lattice to the a-Fe lattice 2. Precipitation of carbon as the carbide Fe3C (cementite) 3. Coagulation of the carbides At the temperature of point A1 processes 1 and 2 proceed almost simultaneously, with the formation of a lamellar mixture of ferrite and cementite. Atoms of dissolved carbon are distributed randomly in the lattice. For this reason cementite nucleates in carbon-rich regions and ferrite in carbon-depleted regions that have little if any carbon. Such a redistribution of carbon is realized through diffusion and depends on temperature and time. When hypoeutectoid steels containing less than 0.8% C are subjected to slow cooling, the transformation starts with the formation of ferrite at grain boundaries. The boundaries act as ferrite crystallization centers. Carbon is forced inside the crystallite. As ferrite precipitates, a concentration 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 grain boundaries 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 as ÀQ=RT D ¼ D0 : Presenting an appropriate cooling rate, undercooling can be enhanced to such an extent that formation of pearlite becomes impossible. ß 2006 by Taylor & Francis Group, LLC. In the range of low temperatures, the transformation mechanism and the character of the formed 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 continuous transition from one transformation mechanism to another can take place over these temperature ranges. The processes strongly depend on the content of alloying elements, especially of carbon, in steel. 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 of a mixture of ferrite and carbide. Free ferrite or carbides can precipitate at the austenite grain boundaries. Here the formation and growth of both phases are controlled by diffusion processes (diffusion crystallization). Diffusion of iron and other alloying elements also plays 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 STRUCTURE OF PEARLITE A mechanical mixture of ferrite and carbide plates is formed on transformation in the pearlite range. 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 also depends on the diffusion rate, which decreases with temperature. The growth of pearlite islets depends in the main on the diffusion rate of carbon and iron atoms. The other decisive factors are the degree of supersaturation and the free energy advantage during the ferrite formation. Pearlite islets grow not only through the formation of new plates but also by way of 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 of nucleation of ferrite and cementite plates and branching of the plates of both phases lead to the 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 steel improve with a decrease in that spacing. The formation rate of cementite and ferrite crystallization centers in the pearlite range accelerates as the temperature is lowered. The plate-to-plate spacing decreases, and the fineness of the structure increases. In the eutectoid steel, the pearlite transformation takes place on cooling to 600–7008C (1100–13008F). In this case, the plate-to-plate spacing equals 0.5–1 mm. Precipitation of austenite over the temperature interval of 650–6008C (1200–11008F) provides the plate-to-plate distance of 0.4–0.2 mm. In this case, the eutectoid is finer pearlite. When austenite precipitates over the temperature interval of 600–5008C (1100–9308F), an extremely fine 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 the pearlite colony. A decrease in the colony dimension is accompanied by a growth of the impact strength and decrease of brittleness. The critical brittleness temperature depends on the pearlite morphology as  1 Tcr ¼ f pffiffiffi d where d is the pearlite colony dimension. Thus a relatively high strength pearlite is formed in the case of the breaking of ferrite and cementite plates, forming a high density of dislocations inside the ferrite. ß 2006 by Taylor & Francis Group, LLC. A better fracture strength of pearlite is achieved through spheroidization of cementite particles. The spheroidization can be facilitated by deformation of pearlite, subsequent heating, and holding at a temperature near Ac1. Another method providing relatively high strength and ductility of pearlite consists in deformation during pearlite transformation. This leads to the formation of a polygonal structure and spheroidization of cementite. The yield stress of the ferrite–pearlite mixture depends on the properties of ferrite and pearlite in an additive manner: s2 ¼ fa sa þ (1 À fa )sp , 0 where fa is the volume fraction of ferrite, sa is the yield stress of pearlite, and sp is the yield stress of pearlite. 3.3.3 TRANSFORMATION OF AUSTENITE IN HYPO- AND HYPEREUTECTOID STEELS The transformation of austenite in eutectoid composition steels was considered above. In hypo- and hypereutectoid steels, the pearlite transformation should be preceded by precipitation of excess phases—ferrite and secondary cementite (see the Fe–C equilibrium diagram in Figure 3.2). The relative amount of the structurally free excess phase depends on the degree of austenite undercooling. The amount of excess ferrite or cementite decreases with an increase in the cooling rate. Given a sufficient degree of undercooling, the formation of an excess phase as an independent structural component can be avoided. When a hypoeutectoid steel containing a small amount of eutectoid austenite is subjected to slow cooling, eutectoid ferrite grows on the grains of excess ferrite and eutectoid cementite is left as structurally free interlayers at grain boundaries. In a hypereutectoid steel, the eutectoid can also be subject to structural degeneration. Cementite, which is formed as a result 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 are found alongside. This eutectoid transformation, which is accompanied by separation of the phases, is referred to as abnormal. In normal eutectoid transformation, ferrite and cementite grow cooperatively in the form of colonies with a regular alternation of the two phases. In the case of abnormal transformation, a coarse mixture of ferrite and cementite does not have a characteristic eutectoid structure. During a eutectoid transformation the mechanism can change from abnormal to normal. Therefore, with a rapid cooling and a correspondingly great undercooling of austenite, the abnormal transformation can be suppressed altogether. Consider the forms and structure of excess ferrite in hypoeutectoid steels. The ferrite is found in two forms: compact equiaxial grains and oriented Widmannstatten plates (Figure 3.7). Compact precipitates of hypoeutectoid ferrite appear predominantly at austenite grain boundaries, whereas Widmannstatten plates are formed inside grains. The Widmannstatten ferrite is observed only in steels with less than 0.4% C and rather coarse grains of austenite. As the dimensions of austenite grains decrease, the share of ferrite in the form equiaxial grains grows. The Widmannstatten ferrite is formed over the temperature interval from A3 (508C or 908F) to 600 to 5508C (1112–10228F). With an increase in the carbon content of steel, the share of the Widmannstatten ferrite in the structure lowers. It is assumed that the Widmannstatten ferrite is formed owing to a shear g ! a rearrangement of the lattice, which is accompanied by an ordered interrelated movement of atoms. Equiaxial grains of ferrite grow by a normal diffusive rearrangement of the lattice with a disordered transition of atoms across the g=a boundary. ß 2006 by Taylor & Francis Group, LLC. FIGURE 3.7 Structure of excess ferrite in hypereutectoid steel, 500Â. One of the methods used to strengthen steels consists in providing a structure with hypoeutectoid ferrite containing dispersed carbide precipitates. To produce such a structure, the steel should be heated until special carbides dissolve in austenite and then cooled rapidly so as to preclude the usual precipitation of carbide directly from austenite before hypoeutectoid ferrite starts forming. 3.3.4 MARTENSITE TRANSFORMATION The 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 martensite transformation in carbon steels are as follows: Temperature, C 1. The martensite transformation is realized on rapid cooling of steel from a temperature above A1 in, for example, water. In this case, diffusive precipitation of austenite to a mixture of two phases (ferrite and carbide) is suppressed. The concentration of carbon in martensite corresponds to that in austenite. The main difference between the martensite transformation 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 is formed over a certain temperature interval. The particular temperature is determined by the carbon content of the steel (Figure 3.8). 3. Termination of cooling over the temperature interval Ms–Mf suspends formation of martensite. This feature distinguishes the martensite transformation from the pearlite 800 600 400 Ms 200 Mf 0 200 Fe 0.4 0.8 1.2 1.6 C Quantity, % FIGURE 3.8 Martensite start Ms and finish Mf temperatures versus carbon content. ß 2006 by Taylor & Francis Group, LLC. transformation. In the latter case, transformation continues to the end at a constant temperature below the point A1, and the final result is a complete disappearance of austenite 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 no incubation period. A certain amount of martensite is formed instantaneously below the temperature Ms. 5. On cooling below Ms, the amount of martensite increases rapidly owing to the quick formation of new plates. The initially formed plates do not grow with time. This feature also distinguishes the martensite transformation from its pearlite counterpart; in the latter case new colonies nucleate and old colonies continue growing. 6. The martensite lattice is regularly oriented relative to the austenite lattice. A certain orientation relationship exists between the lattices. With the pearlite transformation, lattices of the phases comprising the eutectoid mixture exhibit a random orientation with respect to the starting austenite grain. The temperature Ms characterizes an alloy of a certain composition that has been subjected to a particular pretreatment. In a given steel, the martensite transformation starts at the same temperature whatever the cooling rate. That temperature depends on the alloy composition 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 in austenite if the quenching temperature is elevated. Consequently, the carbon concentration of austenite increases and the Ms point lowers. The martensite formation is characterized by a shear mechanism of the austenite lattice rearrangement. The martensitic (shear) mechanism of phase transformation is distinguished by an ordered interrelated movement of atoms to distances shorter than the interatomic spacing, and the atoms do not exchange places. An atom in the initial phase preserves its neighbors in the martensite phase. This is the main feature specific to a shear rearrangement of the lattice. This character of the lattice rearrangement provides coherence of the boundary between the 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 the matrix even at low temperatures. The atoms move cooperatively to distances shorter than the 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 boundary between the martensite crystal and the starting matrix. Slipping movement of the boundary is rendered 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, after coherence is broken, little if any growth of the martensite crystal is observed. The polymorphous transformation of solid solutions by the martensitic mechanism is characterized by the absence of diffusive redistribution of the components. We consider the conditions necessary for the martensitic mechanism by which the high-temperature phase transforms 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 case of a disordered rearrangement of the lattice, elastic deformation is determined by changes in the volume only, whereas with the martensite transformation, it additionally depends on coherence of the lattices of the initial and martensite crystals. As the degree of undercooling is increased, the disordered rearrangement rate of the lattice increases, achieves a maximum, ß 2006 by Taylor & Francis Group, LLC. and then drops. When g-Fe is undercooled to 911 to 7508C (1670–13808F), the normal g ! a transformation is realized, while below 7508C (13808F) the martensite g ! a transformation takes place. To realize the martensitic mechanism of polymorphous transformation in iron, samples should be strongly overheated in the g range and then cooled very quickly to suppress development of the normal transformation. 3.3.5 MORPHOLOGY OF FERROUS MARTENSITES Consider the crystallogeometry of the rearrangement of the fcc lattice of austenite to the bcc tetragonal 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-axis and a simultaneous increase in dimensions along the a-axis. The degree of the tetragonal distortion 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 iron alloys: those due to Kurdyumov and Zacks, Nishiyama, and Treninger and Trojano. The Kurdyumov–Zacks 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 Treninger–Trojano relationship is intermediate between the first two relationships. Several hypotheses are available as to the character of martensite nucleation. Most of them suggest a heterogeneous nucleation at special defect sites in the starting matrix. It was shown experimentally that the sites do not include grain and subgrain boundaries, as these are not places of preferable nucleation of martensite. They might be stacking fault arising in the g-phase during splitting of dislocations. According to other hypotheses, the sites include special configuration dislocation pile-ups or separate dislocations, which are the sources of fields of internal stresses. This decreases the work on critical nucleus formation. By morphology, martensite can be divided into two basic types: plate and massive martensite. They are different in shape, mutual arrangement of crystals, substructure, and habit plane. Plate (needle) martensite is found most frequently in high-carbon steels and carbon-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, 1000Â. ß 2006 by Taylor & Francis Group, LLC. Plates that appear first pass throughout the unit, dividing it into separate parts. But they cannot cross the matrix grain boundary. Therefore, the plate dimension is limited by the dimension 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 the austenite 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 structureless martensite, and it is most desirable. Massive (lath) martensite can be observed in low- and medium-carbon steels. Crystals of this type of martensite are shaped as interconnected plates having approximately the same orientation. The habit plane of laths is close to the {1 1 1}A plane. Plates of massive martensite are separated with low-angle boundaries. An electron microscopic image of massive martensite is given in Figure 3.10. As is seen, a package of plates is the main structural component. Several martensite packages can be formed in an austenite grain. 3.3.6 BAINITE TRANSFORMATION The bainite transformation is intermediate between pearlite and martensite transformations. The kinetics of this transformation and the structures formed exhibit features of both diffusive pearlite transformation and diffusionless martensite transformation. A mixture of the a-phase (ferrite) and carbide is formed as a result of the bainite transformation. The mixture is called bainite. The bainite transformation mechanism involves g ! 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 upper bainite to low-carbon martensite. Closeness of the bainite transformation to its pearlite and martensite counterparts can be explained as follows. The diffusive movement of atoms of the basic component, iron, is almost completely suppressed over the bainite transformation range. Then the g ! a formation of ferrite is difficult because pearlite precipitation is suppressed. However, carbon diffusion is rather active and causes precipitation of carbides. Over the intermediate range the g-phase crystals are formed through coherent growth similarly 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 can FIGURE 3.10 Electron microscopic image of lath martensite, 20,000Â. ß 2006 by Taylor & Francis Group, LLC. precipitate only from the carbon-depleted g-phase. Thus the growth rate of the a-phase crystals depends on the carbon diffusive removal rate. In this case, the martensite start point 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 MORPHOLOGY OF THE BAINITE TRANSFORMATION A distinction is drawn between upper and lower bainite, which are formed in the upper and lower parts of the intermediate temperature range. The conventional boundary between the bainites is close to 3508C (6608F). Upper bainite has a feathery structure, whereas lower bainite exhibits an acicular morphology, which is close to that of martensite. The difference in the structures of upper and lower bainites is attributed to a different mobility of carbon in the upper and lower parts of the bainite temperature range. An electron microscopic analysis showed that the a-phase substructure of upper bainite resembles the substructure of massive martensite in low-carbon steels, while the a-phase structure of lower bainite approximates the structure of martensite in high-carbon steels. In upper bainite, carbide particles can precipitate both at lath boundaries and inside laths. This fact suggests that here carbides precipitate directly from austenite. In lower bainite, carbide is found inside the a-phase. This means that carbide is formed during precipitation of a supersaturated solid solution of carbon in the a-phase. Both upper and lower bainites exhibit a high density of dislocations inside the a-phase. Cementite is the carbide phase in upper bainite, and e-carbide in lower bainite. As the holding time is increased, e-carbide turns into cementite. The austenite grain dimensions have no effect on the martensite transformation kinetics. 3.3.8 TEMPERING The main processes that take place during tempering are precipitation and recrystallization of martensite. Quenched steel has a metastable structure. If subjected to heating, the structure becomes closer to equilibrium. The character of the processes that occur during tempering is determined by three major features of quenched steel: strong supersaturation of the martensite solid solution, high density of crystal lattice defects (dislocations, low- and large-angle boundaries, twin interlayers), and the presence of retained austenite. The main process taking place during tempering of steels is precipitation of martensite accompanied by formation of carbides. Depending on the temperature and duration of tempering, 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 the substructure 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 increases with an increase in the carbon content of steel. This leads to lowering of the temperature Ms and transition from massive martensite to plate martensite. The amount of retained austenite also 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 ß 2006 by Taylor & Francis Group, LLC. homogeneously in the matrix. The heterogeneous nucleation of the segregated carbon occurs either during quenching or immediately after it. Flat homogeneous clusters of carbon atoms that are not connected with lattice defects are formed at tempering temperatures below 1008C (2128F). Their formation is due to considerable displacements of iron atoms and the appearance of elastic distortions. As the tempering temperature 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 above 1008C (2128F). It possesses a hexagonal lattice and appears directly from carbon clusters when the carbon concentration is increased. Metastable e-carbide can also precipitate directly from the a solution. At low temperatures e-carbide precipitates as very fine (10–100 nm) plates or rods (Figure 3.11). With an increase in tempering temperature or time, e-carbide particles 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 of martensite accompanied by deposition of excess carbide is accomplished during quench cooling in 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 of cementite nucleation have been known. First, it precipitates directly from a supersaturated a solid solution. Cementite particles grow at the expense of the dissolution of less stable carbides. Second, cementite appears as a result of transformation of the intermediate carbide lattice 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 (660–7508F). 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 retained austenite accompanied by deposition of carbides. Precipitation occurs over the temperature interval 200–3008C (400–5708F). During tempering, retained austenite transforms into lower bainite. A decrease in the carbon concentration of the a-phase during carbide formation causes changes in the phase structure. Martensite precipitation can conventionally be divided into two 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 areas of unprecipitated martensite with a high-carbon concentration. This results in a nonuniform FIGURE 3.11 Electron microscopic image of the e-carbide, 50,000Â. ß 2006 by Taylor & Francis Group, LLC. content of carbon in different areas of the martensite and consequently inhomogeneity of martensite with respect to its tetragonality. In areas with precipitated carbide, tetragonality is lower than in unprecipitated areas. Two solid solutions with different carbon concentrations coexist. 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 areas containing martensite with the initial carbon concentration. Carbide particles do not grow at this stage. At the second stage of martensite precipitation (150–3008C; 300–5708F) the a solution is depleted of carbon owing to diffusive growth of carbide particles. But the process proceeds very slowly. Therefore, the precipitation kinetics are characterized by a rapid depletion of the a solution in carbon (the timespan decreases as the annealing temperature is increased). Subsequently, depletion of the solid solution in carbon stops. At 3008C (5708F) about 0.1% C is left in the a solution. Above this temperature no difference between the lattice of the 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 of excess carbon and transformation of martensite to ferrite is finished. As mentioned earlier, plates (needles) of quenched martensite have a high density of dislocations, which is comparable to the density of the deformed material. However, recrystallization centers and their development to recrystallized grains have not been observed. This is because carbide particles pin dislocations and large-angle boundaries. It is only above 6008C (11128F), when the density of the particles decreases owing to their coagulation, that recrystallization growth of grains takes place at the expense of migration of large-angle boundaries. Therewith the morphological features of lath martensite vanish. These processes are hampered in high-carbon steels in comparison with low-carbon alloys, because the density of carbides is greater in high-carbon steels. The acicular structure is retained up to the tempering 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 tempering temperature is raised. 3.4 KINETICS OF AUSTENITE TRANSFORMATION 3.4.1 ISOTHERMAL TRANSFORMATION DIAGRAMS To understand the kinetics of transformations to austenite, it is important to follow the process at a constant temperature. To this end, a diagram was constructed that characterizes the isothermal process of austenite precipitation. In this diagram, the transformation time is the abscissa on the logarithmic scale and the temperature is plotted on the ordinate (see Figure 3.12). From this diagram, the incubation period (left-hand curve) can be determined and also the time required for completion of the process (right-hand curve). The instant an alloy 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 transformation curve behavior at the initial and final sections of the curve. Therefore, the lines of the isothermal transformation (IT) diagram usually correspond to a certain final volume that underwent 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 intermediate curves 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 ß 2006 by Taylor & Francis Group, LLC. Temperature of quenching 885 C A1 800 A1 700 Temperature, C 600 500 400 300 200 Martensite 100 0 0.5 1 10 102 60 1 2 103 4 min 8 15 30 1 104 105 60 2 46 8 16 24 Time FIGURE 3.12 Isothermal transformation diagram. finish curves to the right, i.e., toward greater duration. This phenomenon can be observed if the quenching heating temperature rises as a result of a decrease in the number of foreign inclusions, 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 a given steel the temperature that corresponds to a maximum transformation rate (the so-called nose of the sigmoid curve) does not, as a rule, change significantly. 3.4.2 CONTINUOUS-COOLING TRANSFORMATION DIAGRAMS Continuous-cooling transformation (CCT) diagrams consider the transformation kinetics of a eutectoid steel. The major transformation that takes place during annealing cooling of steel is a eutectoid precipitation of austenite into a mixture of ferrite and carbide. The eutectoid transformation kinetics are given by IT diagrams of austenite (Figure 3.13) at a temperature of 7278C (13408F). The structure obtained after tempering below 3008C (5728F) is called tempered martensite. An acicular structure is observed after tempering at 300–4508C (572– 8428F). Tempering over the temperature interval of 450–6008C (842–11128F) exhibits a pronounced dot structure. The structure obtained after tempering below 3008C (5728F) is called tempered martensite. Austenite is in a thermodynamically stable equilibrium with the ferrite–cementite mixture. Stability of undercooled austenite is defined by a period of time during which the appearance of precipitation products in the diagram cannot be registered by conventional 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 steel microstructure. 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 structures formed in a eutectoid steel depending on particular undercooling conditions. ß 2006 by Taylor & Francis Group, LLC. g in ol Co Lines of starting and finishing of transformation Ai A in Pearlite molt O3 + N AU + F + C of KN b en P lt bath Temperature, 8C e ac rn fu Hold in furnace g in g in sa lin Coo Coolin Au 450 C Upper bainite aNO 3 230 C 250 C Lower bainite Martensite M 10 100 1000 Time, log FIGURE 3.13 Isothermal transformation diagram for a eutectoid composition steel. A, stable austenite; Au, undercooled austenite; F, ferrite; C, carbide. 3.4.2.1 Transformations That Take Place under Continuous Cooling of Eutectoid Steels As 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 rise in the degree of undercooling, precipitation of hypoeutectoid ferrite is suppressed. The amount of pearlite increases and the carbon content becomes less than that in pearlite of the 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 steels containing greater amounts of carbon, the precipitation of ferrite cannot be suppressed even if the carbon content decreases. Ferrite precipitation precedes the formation of pearlite A r3 Upper critical rate Temperature, 8C A r2 A r1 Lower critical rate Ar Martensite point (A r ) Rate of cooling FIGURE 3.14 Schematic diagram showing changes in location of the critical points depending on the particular cooling rate. ß 2006 by Taylor & Francis Group, LLC. even at a maximum transformation rate, but the amount of ferrite will be less than that was formed at smaller undercooling. These propositions are valid for the precipitation of cementite in hypereutectoid steels, but it can be suppressed even at relatively small undercooling. In this case, the carbon content of pearlite becomes higher than that in the eutectoid steel. As a result of suppression of the hypoeutectoid ferrite precipitation under continuous cooling from the region of the g solid solution, the point Ar3 lowers much faster than the point Ar1 as the cooling rate is increased. Given a certain cooling rate, both points merge into 0 one point A 2 (see Figure 3.14), which corresponds to the formation of a fine plate structure of the pearlite type free of ferrite. Under continuous cooling the transformation process can also be pictured as diagrams in temperature–time coordinates (Figure 3.15). Hence the behavior of cooling curves should be analyzed to obtain characteristics of the transformation processes. In this diagram, the ferrite and pearlite start lines are shifted toward longer periods of time compared to the IT diagram of Figure 3.13. This is due to an increase in the temperature interval necessary for preparing the transformation processes in the austenite lattice. As a result, only part of the incubation period, which is required for the IT to start, is effective. In this case, the incubation period is the mean of the effective lengths of time corresponding to different periods of time in the given range. This proposition can be used to calculate the behavior of the transformation start line 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 pearlite transformation kinetics consists not only in the presence of an incubation period but also in the character of the volume increase during isothermal soaking: the fraction of the transformed volume 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 completely during the bainite transformation. Every point in the bainite finish curve corresponds to a certain amount of retained austenite. Similar to the pearlite transformation, the bainite transformation can 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 martensite 1000 900 Ac Temperature, C 800 700 600 A 75 P 12 500 400 3 Ac Ms 300 F 35 35 35 65 1 35 65 65 20 65 B 7 10 30 30 M 200 100 58 0 10 46 27 23 102 20 103 220 104 105 106 Time, s FIGURE 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. ß 2006 by Taylor & Francis Group, LLC. Temperature, C 800 700 600 500 400 300 200 2 1 3 1 101 102 103 Log time (t ) 4 104 FIGURE 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 austenite is inhomogeneous with respect to the carbon content, martensite is formed predominantly in carbon-enriched regions. In the case of high-alloy steels, isothermal curves can be separated by a temperature interval in which undercooled austenite is highly stable. In this interval, pearlite precipitation does not take place for many hours, while undercooling is insufficiently great for the bainite transformation (Figure 3.16). In carbon steels, the bainite transformation proceeds concurrently with the pearlite transformation. Products of the pearlite transformation dominate at higher temperatures, and those of the bainite transformation at lower temperatures. 3.4.2.2 Transformations of Austenite on Cooling in the Martensite Range The martensite component in the steel structure appears when the cooling rate achieves a certain value. The minimum cooling rate at which the martensite component is formed is called the lower critical rate of cooling. The rate at which transformations by the pearlite and bainite mechanisms are suppressed completely is referred to as the upper critical rate of cooling (quenching). If the conditions of austenite formation (austenitization temperature and the holding time at this temperature) and the cooling conditions (cooling rate should exceed the upper critical rate) are constant, the location of the martensite point Ms depends only 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 on subsequent cooling to lower temperatures. As the temperature of the quenching medium is lowered, the amount of formed martensite rises first rapidly and then slowly. With an increase in the quenching heating temperature (austenitization temperature), the transformation also shifts toward lower temperatures (Figure 3.17) as more of the alloying elements are taken into solution. A certain amount of martensite may be formed during isothermal holding, but it is not high in carbon steels. Retained austenite is stabilized during isothermal holding. As a result, more martensite is formed during subsequent cooling. Formation of martensite stops at the point Mf. Figure 3.18 shows a relationship between some factors that influence the stabilization of martensite. As is seen, if continuous cooling is stopped at the temperature Th1 and a holding time is allowed at this temperature, the formation of martensite starts after passing through a certain temperature interval rather than immediately when cooling is resumed. Subject to cooling below the point M 0 s, further formation of martensite takes place. If holding is realized at a lower temperature, Th2, the effect of stabilization is enhanced, because further formation of martensite commences at the temperature Ms2 after passing through a greater temperature ß 2006 by Taylor & Francis Group, LLC. Volume fraction of martensite 100 845 80 60 925 40 1040 20 0 20 50 100 150 Temperature, C 200 250 FIGURE 3.17 Curves showing variation of the relative amount of martensite as a function of the transformation 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 of martensite 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 curve that intersects the curve that corresponds to the relative amount of martensite formed under continuous cooling. The point of intersection of the curves, ss, means that stabilization of austenite is impossible at a higher temperature. 3.4.3 DERIVATION OF THE CONTINUOUS-COOLING TRANSFORMATION DIAGRAM THE ISOTHERMAL TRANSFORMATION DIAGRAM FROM Quantity of martensite When solving practical problems involved in thermal treatment of steel, it is often necessary to know how the continuous cooling rate affects the structure formed as a result of austenite transformation. To this end, attempts were made to establish the relationship between the transformation kinetics of austenite under isothermal conditions (IT diagram) and under continuous cooling (CCT diagram). The attempts started from the concept of additivity of the transformation 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 pearlite transformation is continuous. 1 2 3 4 M s2 Th2 M s1 Th1 ds Ms Temperature, 8C FIGURE 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 isothermal holding at Th; 4, M0 s curve, ts is the limiting stabilization temperature. ß 2006 by Taylor & Francis Group, LLC. If the pearlite transformation is preceded by precipitation of eutectoid pearlite or the pearlite and bainite transformations occur concurrently, calculated data are at a discrepancy with 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 period causes acceleration of the subsequent intermediate transformation at the expense of preparatory processes. 2. Precipitation of hypoeutectoid ferrite alters the austenite composition. This delays the subsequent intermediate transformation. 3. Partial transformation of austenite over the intermediate range decreases the rate of the said transformation at lower temperatures and facilitates an increase in retained austenite. 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 austenite in different ways. For this reason, special methods of constructing thermokinetic transformation diagrams of austenite subject to continuous cooling were elaborated for noneutectoid steels (see Figure 3.15). From these diagrams it is possible to determine the critical rate of quenching cooling or continuous cooling that is necessary to complete a particular stage of austenite precipitation. 3.4.4 CONTINUOUS-COOLING TRANSFORMATION DIAGRAM OF THE BAR DIAMETER AS A FUNCTION When steel is subjected to martensitic hardening, it should be cooled from the quenching temperature so that on undercooling to a temperature below the Ms point austenite has no time to precipitate and form a ferrite–carbide mixture. To achieve this, the cooling rate should be less than the critical value. The critical cooling rate is the minimum rate at which austenite does not precipitate to a ferrite–carbide mixture. Of course, the cooling rate of steel products is nonuniform over their cross section. It can be higher than the critical rate on the surface and lower than the critical rate at the center. The critical cooling rate at different points of a product can be directly determined from an IT diagram (Figure 3.19). In the first approximation, it is given by the slope of the tangent to the C curve that denotes the austenite precipitation onset. This method gives a value that is (a) (b) Vcooling t 1 2 e Unquenching field Quenching field Center Surfac Temperature, 8C Vcritical Surface 1 r 2 r Center Time,t FIGURE 3.19 Diagram showing distribution of the cooling rate over the cross section of a sample and the corresponding IT curve. ß 2006 by Taylor & Francis Group, LLC. 800 Temperature, C 700 25 600 % Ferrite 40 35 35 2 3 500 65 20 C B A % Pearlite 60 65 5 D E 400 10 40 40 1 30 300 5 % Bainite 50% 95% Martensite 200 100 23 52 48 46 57 0 1 30 27 25 2 10 23 21 20 17 103 10 104 105 106 60 s 1 (a) 2 4 8 15 30 min 60 1 2 4 6 8 16 24 h Time 1 2 4 7 10 Days 60 A Hardness, HRC 50 40 B C 30 E D 20 10 0 0 5 10 (b) 15 20 25 30 35 40 45 50 55 60 65 70 75 Distance from cooling surface, MM FIGURE 3.20 (a) Time–transformation temperature diagram for continuous cooling of steel containing 0.38% C compared with (b) the process of the sample’s cooling during face quenching. Numerals at the bottom 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 if one uses thermokinetic diagrams (Figure 3.20). Intercepts of the cooling curves with the lines of the thermokinetic diagrams show the start and finish temperatures of the corresponding transformation. From the transformation diagram it is possible to determine, for example, the rate that will provide 50% martensite in the structure or the rates at which the entire transformation occurs in the pearlite range, i.e., hardening is excluded altogether. Because the data on the critical hardening rate depend on cooling time and should be associated with a particular temperature (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 is possible to straightforwardly determine the critical cooling time K: Km for fully martensitic hardening; 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 ß 2006 by Taylor & Francis Group, LLC. sample during end-face hardening are known, the outcome of hardening can be determined from the transformation diagram. It should be remembered that a transformation diagram is valid only for particular conditions of melting and homogenization. Deviations in the composition or grain dimensions cause changes in the trend of thermodynamic curves. This is 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 decreases with an increase in the extent of grain boundaries. 3.4.5 DEFINITION OF HARDENABILITY Hardness, HRC The depth of the hardened zone is termed hardenability. This is one of the most important characteristics of steel. Since the cooling rate is nonuniform along the cross section of a sample (see Figure 3.19), austenite can pass into martensite in surface layers only, while at the center 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 the core of a large-diameter product is lower than the critical value and therefore the core is not martensitically hardened. Martensite is present in the surface layer only. After hardening treatment, a bulky part with a large cross section may exhibit the entire range of structures: a smooth transition from martensite near the surface through troostite– martensite 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 in the cooling rate are incurred by the diameter of samples. Considering what has been said above, to achieve a through hardening of bulky products or 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 to determine this rate. The diagrams were plotted for different grades of steel, taking into account the progress of cooling in different sections and in different hardening media. Note that hardenability depends on the steel composition, specifically on the carbon content. Hardenability of each grade of steel is presented as a hardenability band (Figure 3.21). These diagrams have been plotted for almost all existing grades of steel. They show how to achieve hardening of a product made of a particular steel. 65 60 55 50 45 40 35 30 25 20 15 10 0 5 10 15 20 25 30 35 40 45 Distance from cooling surface, mm FIGURE 3.21 Steel hardenability band. ß 2006 by Taylor & Francis Group, LLC. 50 Hardenability of steel is also characterized by transformation time–temperature curves (IT curves). The more the curve is shifted to the right along the abscissa axis, the greater is the hardenability of the steel. This is explained by the fact that the rightward shift of the IT curve is due to better stability of austenite. An improvement in the stability of undercooled austenite and hence an increase in the critical hardening rate lead to a greater depth of hardening. Then hardenability depends on all the factors that improve the stability of undercooled austenite. For example, the stability of austenite can be raised by alloying steel with chromium and tungsten. These elements lower the austenite precipitation 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 iron–carbon alloy. Elevation of the hardening temperature favors an increase in the hardening depth thanks to the homogenization of austenite and enlargement of austenite grains. Refinement of grains impairs hardenability as grain boundaries affect the stability of austenite. The hardening depth also depends on the hardening medium used. The greater the intensity of cooling, the greater the depth of hardening. Besides, the hardening depth depends on the cross-sectional diameter of the products. The critical diameter is that of the greatest cross section that lends itself to through hardening in a given hardening medium. The critical diameter 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 through hardening, the properties do not differ along the cross section of a product. Otherwise they decrease from the surface to the center. Let us analyze the influence of hardenability on the properties of steels that were tempered after hardening. A high temperature favors equalization of hardness along the cross section. However, the structure of weakly hardenable steels remains inhomogeneous; a grain structure will appear on the surface, where martensite is formed during quenching, while a lamellar structure will remain at the center. A grain structure will be present along the entire cross section of a through-hardening steel. This determines the character of changes 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 a through-hardening steel these quantities remain unchanged along the cross section. The properties of tempered steels (fracture stress, yield stress, impact strength, reduction of area) are impaired if ferrite precipitates during quenching. The mechanical properties of a product depend on its cross-sectional area. To obtain the best mechanical properties in the tempered state, a grain structure should be provided along the entire cross section; i.e., through hardenability should be ensured in the quenched state. 3.5 3.5.1 GRAIN SIZE STRUCTURE OF GRAIN BOUNDARIES When analyzing any processes or properties associated with grain boundaries, it is necessary to know the structure of the material. The overwhelming majority of structural materials are polycrystalline. They comprise a set of grains separated by boundaries. The grain boundary is one of the basic structural elements in polycrystalline materials. The grain boundary represents an interface between two differently oriented crystals. This is the region of crystal imperfection. It is capable of moving and adsorbing impurities. The boundary has a high diffusive permeability. In polycrystalline materials, the boundaries determine the kinetics of many processes. For example, movement of grain boundaries controls the process of recrystallization. A high ß 2006 by Taylor & Francis Group, LLC. FIGURE 3.22 Electron microscopic image of (a) low-angle boundary and (b) large-angle boundary, 50,000Â. diffusive permeability of grain boundaries determines the kinetics of diffusion-dependent processes at moderate temperatures. Grain boundaries adsorb impurities. Embrittlement of metal material is connected with enrichment of grain boundaries in impurities. Grain boundaries may conventionally be divided into two large groups: low-angle and large-angle boundaries. Low-angle boundaries (or subgrain boundaries with an angle of less than 108) represent networks or walls of dislocations. The structure of large-angle boundaries is much more complicated. Figure 3.22 shows both types of grain boundaries. The progress in understanding the structure of grain boundaries is connected with elaboration of the models describing the observed microscopic properties of the boundaries. 3.5.1.1 Structural Models The pioneering structural model is the model of an amorphous boundary. It allows an explanation of the value of the surface tension G3 and the grain boundary slip. In terms of this model, it was assumed that the usual boundary with a large angle has random regions of incontingency similar to a liquid. The width of the regions does not exceed three atomic diameters. In later models, amorphous portions of the boundaries were added with crystalline portions. According to Mott and coworkers [1,2], G3 represents portions of good and poor contingency. In the opinion of Smoluchowski [3,4], even when the boundary angle exceeds 158, dislocations combine themselves into groups and form incontingency regions separated by undistorted areas. If the misorientation angle is greater than 358, then G3 is a solid region of incontingency. Geisler and Hill [5] and Hargreaves [6] described the grain boundary in ß 2006 by Taylor & Francis Group, LLC. terms of the model of a transition lattice. According to this model, a certain system in the arrangement of atoms exists in G3. The arrangement corresponds to a minimum energy possible 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 are common to the crystal lattices of both grains. The boundaries lying in close-packed planes of such superlattices will be most favorable with respect to energy. If the misorientation angle is small, the coincidence is upset. Coinciding atoms are present in the boundary plane for some discrete values of the grain misorientation angles. Boundaries that meet the conditions required for the coinciding atoms to appear are called partial contingency (or special) boundaries. Direct experimental studies of G3 are scarce. Microscopy studies and transmission electron microscopy 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 which grain boundaries are formed. A grain structure, and correspondingly G3, can be formed as a result of crystallization from the liquid state, phase transformations in the solid state, or recrystallization 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 oriented rather than chaotic character. Correspondingly, the spectrum of boundaries in a cast material should differ from the random distribution. In the case of recrystallization in the solid state, i.e., polymorphous transformations of metals 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, the formed boundaries should have strictly defined and crystallographically determined misorientations rather than random orientations. Many boundaries that appear during a polymorphous transformation are close to special boundaries of coincident sites. Experimental studies into misorientations of the crystals formed during phase transformations in chromium–nickel steel and titanium alloy showed that misorientations at real boundaries agree with theoretical ones (what is meant here is the calculation of crystallographically determined misorientations for the fcc–hcp and hcp–fcc, hcp–bcc and bcc–hcp, and fcc–bcc transformations). Then the crystallogeometry of the boundaries resulting from polymorphous transformations is controlled by orientation relationships of the phases formed. In the case of recrystallization processes, the grain structure depends on the stage of recrystallization at which annealing was stopped. During primary recrystallization the formation of the structure starts with the appearance of nuclei, that is, dislocation-free portions of the matrix. They are surrounded by large-angle G3. The proposed models of nucleation assume that nuclei of new grains are formed near the initial G3 owing to a rearrangement of intergrain lattice dislocations. However, it has been established recently that new grains with large-angle boundaries can be formed without participation of intergrain dislocations but rather during splitting of initial boundaries. This process can be accounted for in the following way. After plastic deformation the grain boundaries are in a nonequilibrium state owing to the trapping of lattice dislocations. During annealing the grain boundary structure regains the equilibrium state at the expense of splitting of the boundaries. Splitting of the boundaries during recrystallization is caused by lowering of the total energy of the grain system because high-energy boundaries are replaced by low-energy ones. Here mutual misorientations depend on misorientation of the nuclei in the deformed matrix. At subsequent states of recrystallization the grains become coarser owing to migration of the boundaries. One would expect that the average statistical trend of the process should be ß 2006 by Taylor & Francis Group, LLC. toward formation of low-energy special boundaries. However, the available experimental data are contradictory. This fact suggests that in addition to the tendency to a thermodynamic equilibrium, kinetic factors (different mobility of the boundaries, their pinning by impurities and precipitates) play an important role in the process of structure formation during 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 recrystallization temperature and time. In contrast, in commercially pure aluminum the fraction of special boundaries decreases. The state of grain boundaries in a material depends not only on their misorientation but also on the content of lattice defects. For this reason the boundaries of recrystallization nuclei are not in equilibrium; they are formed in the regions of the deformed matrix with an excess density of dislocations of like sign. Rearrangement of the dislocations within the nucleus boundaries is not complete. A nonequilibrium state of the boundaries is also preserved during their migration through the deformed matrix as the matrix absorbs lattice dislocations. Then the boundaries are nonequilibrium in ultrafine grain materials formed at the early stage of recrystallization. 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: grain boundaries formed at a low-tempered (<0.3–0.4Tmelt) deformation and those formed under deformation 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 grains of 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 whole volume. On the average, two out of three boundaries are large-angle ones. In this case, the fraction of special boundaries is small. When subject to deformation at high temperatures, the formation of grain boundaries is due to development of recrystallization processes directly during deformation. This phenomenon is called a dynamic recrystallization. The grains formed during a dynamic recrystallization are large-angle ones. Data on crystallogeometrical parameters of these boundaries are very scarce. From the above discussion it appears that, depending on their origin, grain boundaries have 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 DETERMINATION OF GRAIN SIZE The size of the grain that is formed under a given treatment is determined from microsections after their etching. For carbon and alloyed steels the following reagent is used: 1–5 ml HNO3 þ 100 ml ethyl or methyl alcohol. Austenitic steel is etched in a copper sulfate chloride solution containing 10 g copper sulfate, 50 ml hydrochloric acid, and 50 ml water. When carbon low-alloy steels are etched, the reagents turn pearlite dark and make visible the ferrite grain boundaries, the martensite structure, and tempering products. The etching rate rises with the amount of nitric acid. The etching time is from several seconds to a minute. Etching of austenitic steel reveals the austenite structure and the austenite grain boundaries. Carburization is also used to establish the austenite grain boundaries. In this case, samples are heated to 9308C (17008F) in a carburizing medium (e.g., a mixture of 40% BaCO3 and 60% charcoal), cooled, and etched. ß 2006 by Taylor & Francis Group, LLC. In addition, an oxidation method is used according to which microsections are heated in vacuum to a temperature 20–308C (35–558F) higher than the quenching temperature and are soaked for 3 h. Subsequently air is fed to the furnace for 30–60 s, and the samples are cooled in water. Before quenching it is recommended to heat samples in a borax melt at 930–9508C (1700–17508F) for 30–40 s and then cool them in water. After these treatments microsections are polished and etched in a 15% solution of hydrochloric acid in ethyl alcohol. Grain boundaries are seen as the oxide network. Apart from this, use is made of the method of etching austenite grain boundaries, the method 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 in composition 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 the grain number N corresponds to the formula n ¼ 8  2n, with n the number of grains per 1 mm2 of 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 or from a photomicrograph within the area bounded by a circle 79.8 mm in diameter. At 100 magnification this value corresponds to a microsection area of 0.5 mm2. The total number of grains is calculated from the formula m100 ¼ m þ 0.5m1, where m is the number of grains inside the circle and m1 is the number of grains intersected by the circle. The number of grains per mm2 of the microsection is M À 2m100. If a magnification other than 100 power is used, M ¼ 2(g=100)2mg (g the magnification power used and mg the number of grains counted at this magnification power). The mean number of grains (Mmean) is calculated using three characteristic areas. The mean area (Smean) and diameter (dmean) of the grains are calculated using the formula Smean ¼ 1=Mmean and dmean ¼ 1(Mmean )1=2 : The values of equiaxial grains are characterized by the mean nominal diameter, which is determined on the focusing screen of the microscope or from a photomicrograph. For this purpose several arbitrary straight lines are drawn so that every line intersects at least ten grains. The number of intersections on the length of all the lines is counted. Finally, the mean diameter 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 used to determine dmean if one assumes that the grain is spherical in shape (x ¼ pD2=4): Smean ¼ k2 «(pD2=4) m=«m. Then the number of grains (N ) per mm2 is found from the formula N ¼ 1=Smean. 3.5.3 AUSTENITE GRAIN SIZE EFFECT AND GRAIN SIZE CONTROL The austenite grain boundary structure that is produced on heating above the critical points is important because the austenite transformation products formed during cooling (martensite, pearlite, etc.) appear inside austenite crystals. A coarse austenite grain determines a coarse plate 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. ß 2006 by Taylor & Francis Group, LLC. As is known, a coarse-grain structure of steel (ferrite–pearlite, martensite, etc.) is characterized by lower mechanical properties. For this reason a fine-grain structure of steel is preferable in practice. Then the primary task is to produce fine-grain austenite. Since austenite appears during heating of a ferrite–carbide mixture, growth centers of the austenite phase are very numerous, and initially austenite grains are extremely small, on the order of 10–20 mm. But with an increase in the heating temperature or holding time in the austenite range, the grains begin to grow intensively. Two types of steels exist: hereditarily coarse-grained steels and hereditarily fine-grained steels. This difference is due to the grain growth kinetics with an increase in temperature. In hereditarily coarse-grained steels a grain gradually and rather uniformly becomes larger as the temperature is raised above Ac3. In hereditarily fine-grained steels, fine grains are preserved up to about 9508C (17508F). On transition through the coarsening temperature, separate grains start growing intensively and variations in grain size arise. Near 1100–12008C (2000– 22008F), grains of hereditarily fine-grained steels may be even larger than those of hereditarily coarse-grained steels. Such differences in the growth of grains in steels are explained by the differences in number and state of disperse nonmetal inclusions such as, above else, aluminum nitrides, certain carbides, and oxides. These articles retard movement of grain boundaries until temperatures are reached at which the particles dissolve in austenite. The barrier effect of the 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 growth of austenite grains is not observed for 8 h after carburization at 9258C (17008F), the steel is assumed to be a hereditarily fine-grained one. Extrapure steels, those produced with a minimum amount of foreign impurities, nitrogen and oxygen, are distinguished by a rapid growth of grains above the critical point Ac3. In the case of the usual commercial steels, a grain 20–25 mm in size corresponds to standard heating for quenching, normalization, or annealing. As the temperature is elevated to 1200–12508C (2200–22508F), the grain size reaches 0.1 mm, and in large forgings and welds, grains of several millimeters in size occur. In ingots and castings, grains can be as large as several centimeters. If a steel is heavily alloyed with elements that stabilize austenite, the austenite structure is fixed during cooling to or below room temperature and the steel grain is equal to the initial austenite grain. If austenite passes to pearlite, then, for example, for a hypereutectoid steel one should take into account the size of the pearlite colony, which is characterized by the same crystallographic orientation of ferrite and cementite plates. A pearlite colony usually differs in size from an austenite grain. Several pearlite colonies are formed in every grain. So an austenite grain is broken into several grains. This is also true of the ferrite–pearlite structure of a hypoeutectoid steel. But in the latter case a network of excess ferrite is formed at grain boundaries. This suggests a connection between a grain of a thermally treated steel and the initial austenite grain. When steel undergoes quenching, a large number of martensite crystals appear in every austenite grain. They are connected with the initial austenite grain by certain orientation relationships. For this reason a correlation is easily seen between the initial austenite grain and grains of the quenched steel. Refinement of the initial grain under heating above Ac3 results 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 complicated by structural inheritance. When crystallographically ordered structures of bainite or martensite are heated, austenite can also be formed, under certain conditions, in a crystallographically ordered way. Therefore, under heating above Ac3 the austenite grain is equal in size to a coarse ß 2006 by Taylor & Francis Group, LLC. grain of steel. In this case, refinement of the crystal structure as a result of phase recrystallization during the a ! g transformation does not take place. After the a ! g transformation the structure 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 refined rather 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 grain becomes still coarser. Note once again the unusual character of this two-stage process, which includes first a reestablishment of the initial austenite grain and then a refinement of the grain with temperature. Plastic deformation inhibits the structural inheritance. This is due to the appearance of globular 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 as same-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 initial austenite grains. With an increase in the heating rate, sections of austenite with a different orientation appear. If isothermally heated these sections grow larger and absorb the restored austenite and excess ferrite. The greater the number of such sections formed at a faster rate of heating, the finer the austenite grain. Formation of these sections cannot be explained in terms of the austenite recrystallization because their growth stops as soon as the excess ferrite dissolves completely. They appear in a somewhat overheated and therefore nonequilibrium ferrite–austenite structure. Note an anomalous dependence of the point Ac3 on the heating rate. Increasing the heating rate of the steel allows completion of austenitization at lower temperature. 3.5.4 GRAIN SIZE REFINEMENT Tchernov [7] was the first to show that it was possible to refine a coarse-grained structure in 1868. Since that time the procedure has been widely used for treatment of steel products. The grain refinement, which takes place on heating steels above the temperature Ac3, is related to a transition to the austenite state through nucleation of numerous centers of the austenite phase. Development of these centers leads to formation of a relatively fine-grained structure. Above Ac3, the cross-sectional size of the grain is 10–30 mm. Initially the grain size is independent of the grain of the starting structure; it can be very fine irrespective of whether the starting structure of the steel was fine or coarse. A fine-grained structure of the restored austenite provides a fine-grained structure of cooled steel whatever structural components are formed—pearlite, bainite, or martensite. This is due to the fact that all the transformation products 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 the formation of fine-needle martensite. This underlies the grain refinement effect that is associated with hearing above Ac3. Heating the steel above Ac3 during full annealing, normalization, or quenching is followed by recrystallization. Given an initially coarse-grained structure, ß 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 expected correction of the structure during the g ! a transformation does not take place. Refinement of crystallites is especially pronounced when transformation to the austenite state starts in many centers inside the initial structure. The formed centers should have a random orientation, which is not connected with the orientation of the a-phase in the initial structure. Normally such centers are sufficiently great in number that the grain size does not exceed 15–30 mm. Breaking of an austenite grain into pearlite colonies, each of which can be considered an independent grain, also represents refinement of steel during pearlite precipitation of austenite. 3.6 STRENGTHENING MECHANISM IN STEEL 3.6.1 SOLID SOLUTION STRENGTHENING Solid solution strengthening is a phenomenon that occurs when the number of impurity atoms in the lattice of the basic element is so small that they are incapable of forming both stable and metastable precipitation phases under any thermal treatment conditions. Nevertheless the impurity atoms favor improvement of mechanical properties. This can be accounted for by the following. The presence of impurity atoms in the matrix lattice leads to distortion of the lattice because of the difference in size between the atomic radii of the impurity and the basic component. This in turn leads to the appearance of elastic deformation fields, which retard movement of dislocations in slip planes under the action of applied stresses. In addition, the impurity atoms can inhibit movement of dislocations by forming impurity atmospheres around them. 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 lattice of the a-iron, on the structure and properties of a-iron. Solubility of carbon in a-iron is much lower 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, the cubic lattice of the a-iron suffers, upon introduction of carbon atoms, a tetragonal distortion similar to the one of the martensite lattice, except that in the former case the distortion is much smaller. In addition, implantation of carbon atoms causes the entire lattice of the a-iron to expand somewhat. For example, at a carbon content of 0.015% the lattice constant increases 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 solution was detected. The yield stress rises most dramatically with an increase in the carbon concentration from 10À7 to 10À4À10À3%. The influence that carbon exerts on plastic deformation resistance of the a-phase is due to both its strong interaction with dislocations and pinning of the dislocations and elastic deformations arising as a result of the tetragonal distortion of the a-phase lattice after implantation of carbon atoms. What is more, the presence of carbon in lattices of different structural components formed during thermal treatment of steel also leads to changes in their mechanical characteristics. For example, the location of implanted carbon atoms predominantly in one of the sublattices of interstitial sites during the martensite formation brings about additional tetragonal distortions of the martensite crystal lattice. This enhances plastic deformation resistance owing to the 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 carbon with an increase in the carbon content. This is accompanied by impairment of the plastic characteristics of the steel and lowering of the fracture stress. For example, if the carbon content is raised from 0.25% to 0.4% in a steel with 5% Cr, after quenching and low tempering ß 2006 by Taylor & Francis Group, LLC. the tensile strength increases from 1600 to 2000 MPa, the fracture stress of samples with a purpose-produced crack decreases from 1300 to1000 MPa, and the impact strength drops from 0.3 to 0.04 J m2. The influence of carbon dissolved in the a-phase on the mechanical properties of steel is also observed in the case of the ferrite–pearlite transformation. The factors responsible for this phenomenon were analyzed above. Both in the homogeneous a-phase and the ferrite– pearlite mixture, the yield stress rises most sharply when the carbon concentration of ferrite is raised from 10À7 to 10À4–10À3%. A direct examination of the crystal structure of the a-phase formed over the temperature interval of 250–3008C (482–5728F) during the intermediate (bainite) transformation also revealed a tetragonal structure with the c=a axis ratio equal to 1.006 and 1.008 at carbon contents of 1 and 1.2%, respectively. This attests to dissolution of part of the carbon in the a-phase and suggests that the solid solution strengthening of the phase is one of the factors providing the high strength properties of intermediate transformation products. 3.6.2 GRAIN SIZE REFINEMENT In Section 3.5.4 the possibility of refining steel grains by phase recrystallization under heating to a temperature above Ac3 was considered. Although austenite passes to other phases during cooling, its grain size represents an important characteristic of steel. This is due to the fact that all structural components are formed within each separate crystal. The smaller the austenite grains, the finer the network of excess ferrite at their boundaries and the smaller the 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 already precipitated. Impact strength is especially sensitive to the austenite grain size, and it decreases with grain enlargement. A decrease in the dimensions of pearlite colonies inside the initial austenite grain favors a rise in impact strength also. Although the grain size has a considerable effect on impact strength, its influence is small if any 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, the inherited size has no effect. However, the technological process of heat treatment is determined by the inherited grain. For example, a hereditarily fine-grained steel may be deformed at a higher temperature with the assurance that the coarse-grained structure will not occur. 3.6.3 DISPERSION STRENGTHENING In the majority of metal alloys, precipitation of supersaturated solid solutions formed during quenching is followed by precipitation of disperse particles enriched in atoms of the alloying components. It was found that the strength (hardness) of the alloys increases with the precipitation of these particles. The increment in the value of these characteristics increases as the dispersion and volume fraction of the particles increase. This phenomenon has been referred to as dispersion strengthening. Precipitation of supersaturated solid solutions occurs during heating (aging) of quenched alloys. The study of precipitation processes is ultimately aimed at elaboration of the most efficient methods for strengthening aging materials. In a general case, strengthening results from 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 atomic clusters [GP zones] or different-structure precipitate particles) that retard their movement. The character of interaction between moving dislocations and precipitates of the second phase can be different depending on the phase morphology and structure. ß 2006 by Taylor & Francis Group, LLC. The total effect of aging on the strength properties of alloys is determined by (1) the strength of the precipitates formed, (2) the volume fraction of precipitates, (3) the degree of precipitate 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 should precipitate 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 diffusion processes, 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 carbon can correspond to its maximum solubility of 0.018%. During subsequent storage at room temperature (natural aging) carbon tends to precipitate from the solid solution. Carbonenriched regions appear predominantly in defective sections of the matrix. Precipitation of carbon from a supersaturated solid solution during natural aging results in improvement of its strength characteristics and hardness. However, plastic characteristics—reduction of area, specific elongation, and impact strength—are impaired (see Figure 3.23). A clearly pronounced yield 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 an extent that in several hours of aging it reaches the value obtained after several days of natural aging (Figure 3.24). This is due to intensification of diffusion processes with an increase in temperature. As the temperature is elevated further, precipitation of the supersaturated a solid solution proceeds still faster. The total process of carbon precipitation from the supersaturated solid solution in a-iron comprises several consecutive processes. Mechanical characteristics and hardness are not sensitive to structural changes that take place during the aging of alloys. Sharp changes in properties indicate alterations in the structural state of the steel. 210 190 70 170 60 HB s HB 150 50 d,y % 60 s −2 80 b s−kG mm 130 40 30 y 40 20 20 10 0 0 d 4 8 12 16 20 Time, days FIGURE 3.23 Curves showing strengthening of fcc crystals. ß 2006 by Taylor & Francis Group, LLC. 24 28 HB 200 190 50 C 180 170 160 100 C 150 140 20 C 130 200 C 120 150 C 110 300 C 0 4 8 12 (a) 20 C 16 20 24 1 3 (b) Time, h 6 9 12 15 18 Time, days FIGURE 3.24 The aging temperature dependence of hardness of carbon steel. A maximum change in mechanical properties during precipitation is achieved only if excess crystals in a highly disperse state precipitate. Subsequent coagulation of the crystals leads to degradation of the properties (Figure 3.24 and Figure 3.25). As the temperature is raised above 1008C (2128F), carbides begin to homogeneously precipitate directly from the solid solution. The precipitating phase has the carbide lattice below 2008C and the cementite lattice above 3008C (5728F). A transition from one phase to the other is realized over the temperature interval of 200–3008C (392–5728F). The onset of transition from atomic clusters near dislocations to precipitation of the e-carbide remains to be ascertained. The temperature and time during which the e-carbide crystals precipitate from the inhomogeneous solid solution depend on the degree of the solution supersaturation and concentration of vacancies. Coagulation of the e-carbide crystals lowers the increment in hardness, fracture stress, and yield stress as the effect of breaking the slip dislocations diminishes. Above 2008C (3928F), 100 40 C 22 20 75 18 25 Degree of precipitation 50 16 0 100 65 C 20 75 50 18 25 16 0 100 ss 14 130 C 20 75 50 18 25 16 0 14 Time, h FIGURE 3.25 Variation of the yield stress as a result of carbon precipitation from the a-iron lattice at different temperatures. ß 2006 by Taylor & Francis Group, LLC. where precipitation of the particles is detected even by metallographic methods, hardness stops increasing. If a naturally aged sample of steel is heated at a temperature of 100–2008C (212–3928F), a decrease in hardness can be observed. This is due to the phenomenon of recovery where the phase nuclei that were formed at room temperature dissolve on heating to higher temperatures. 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 very low content of carbon. In steels containing over 0.4% C, the effects considered above are obscured by the influence of cementite particles formed during the pearlite transformation. Besides, nucleation of the precipitating phase can be inhibited owing to migration of carbon to the cementite–ferrite interfaces. As a result, the amount of carbon concentrated at lattice defects 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 of heterogeneous 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 observed in other aging alloys. Mechanical properties change during aging after cold deformation in the same way as after quenching. That is, the yield stress, the fracture stress, and hardness are altered. With an increase in aging time, specific elongation and reduction of area decrease and the tendency to brittle fracture is enhanced. The rate of change is greater than in a quenched alloy. What is more, the character of the changes is different. Whereas in the case of aging after quenching, hardness reaches a maximum and then drops, after cold deformation hardness does not decrease with the aging time (Figure 3.26). As the aging temperature is raised, the maximum hardness of a quenched alloy lowers, while after cold deformation hardness is independent of the aging temperature. This is explained by the fact that a considerable amount of carbon is concentrated near dislocations. Few if any clusters nucleate in the matrix homogeneously. Consequently, clusters cannot grow at the expense of other clusters, i.e., they cannot coagulate. As the solubility of carbon in g-iron is also susceptible to changes, one can also expect the effect of dispersion strengthening. However, the g-phase is not fixed during quenching but undergoes martensite transformation. For this reason an additional amount of carbon transferred to the solid solution at the line ES will just enhance the precipitation of martensite and retained austenite during tempering of steel. Still, an increase in hardness as a result of carbide precipitation is observed in purely austenitic steels. Hardness HB 90 85 100 C 80 40 C 75 70 100 C 65 0.1 1 10 Time of aging, h 100 500 FIGURE 3.26 Aging after quenching from 7208C (13288F) (- - - -) and after 10% cold deformation (––) of cast steel. ß 2006 by Taylor & Francis Group, LLC. 3.6.4 WORK HARDENING (DISLOCATION STRENGTHENING) An important method used to strengthen steels is deformation strengthening. stresses (stresses of the first kind). Usually these stresses are very high. Changes in properties that occur under cold deformation can be rectified during subsequent heating. The greater the degree of deformation, the lower the heating temperature. Depending on the temperature and time of annealing, various structural changes take place in a cold-deformed material. The changes are divided into recovery and recrystallization processes. Recovery is a totality of any spontaneous processes of variation in the density and distribution of defects before the onset of recrystallization. If recovery proceeds without the formation and migration of subgrain boundaries inside the recrystallized grains, it is called restoring. If subgrain boundaries are formed and migrate inside the crystallites, recovery is referred to as polygonization. Restoring does not include an incubation period. Properties start changing right at the beginning of annealing. Restoring is accompanied by a redistribution of point defects whose ß 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 a temperature 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 stress fields, and therefore the wall formation process is energetically favorable. A wall composed of like-sign dislocations represents a low-angle boundary separating neighboring subgrains with a 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, the subgrains 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 persist even after long holding at this temperature. Starting from a certain annealing temperature, the structure changes drastically. New rather equilibrium grains are observed along with extended deformed grains. They differ from the grains of the deformed matrix by having a more perfect internal structure. While the density of dislocations in a strongly deformed matrix is 1011À1012 cmÀ2, after recrystallization it lowers to 106À108 cmÀ2. As distinct from the polygonized structure, recrystallized grains are separated from the matrix with large-angle boundaries. The formation and growth of grains with a more perfect structure that are surrounded by large-angle boundaries at the expense of initially deformed grains of the same phase is called primary recrystallization. Recrystallization begins with an incubation period. The recrystallization rate first increases from zero to a maximum and then decreases owing to an ever rising number of new grains in contact with one another. Inclusions of insoluble impurities (carbides, nitrides) lower the tendency to growth of recrystallized grains. This is especially important in the case of ferritic steels, which are prone to grain growth. Another phase may precipitate during recrystallization in alloys that were subjected to a strong cold deformation. Sometimes the intensive growth of individual crystals can be observed after a strong deformation 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. With an 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 the degree of deformation. At a given degree of deformation, higher the recrystallization temperature, the coarser the initial grain. After recrystallization an initially coarse-grained structure gives a larger grain than a fine-grained structure does. In iron–carbon alloys, coarse grains are formed until the appearance of new grains associated with a polymorphous transformation. Under critical conditions of recrystallization the grain size decreases with an increase in the 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 and Ac3 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 an improved long-time strength at lower temperature, while a coarse-grained material exhibits this property at higher temperatures. A required size can be obtained by a proper choice of the deformation and recrystallization conditions. In the case of steels, where no transformations take place (pure ferritic or austenitic steels) this combination of technological operations is the only opportunity to influence the grain size. ß 2006 by Taylor & Francis Group, LLC. 3.6.5.2 Quenching (Strengthening Treatment) Quenching refers to cooling from the temperature range of the solid solution at such a rate that transformations in the primary and bainite ranges are suppressed and martensite is formed. In this state, steels are characterized by the greatest hardness. A distinction is made between (a) normal quenching, which is used mainly for treatment of medium- and high-carbon steels and (b) quenching after a thermochemical treatment (carburization, hightemperature cyaniding), which is used for low-carbon steels. 3.6.5.2.1 Normal Quenching To provide a required cooling rate during quenching, various cooling media and methods are employed. Water, oil, or air can serve as the cooling medium. Many alloyed steels, which are characterized by a high stability of austenite, are subjected to step quenching. With this method of quenching, the temperature drop is less than in the case of direct cooling to room temperature and consequently quenching stresses are less. A certain amount of austenite is retained during quenching even in steels with a relatively small content of carbon. For this reason it is impossible to impart the maximum hardness to a product. Since austenite is stable at room temperature and passes to martensite at lower temperatures, 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 or quenching 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, while the core is cold and remains unquenched on subsequent rapid cooling. The selection of steel for surface quenching must be governed by the sensitivity of the metal to quick heating and cooling. For this reason the carbon concentration is limited to 0.7%. Otherwise cracks are formed. Among the main quenching defects are excessive holding and overheating. They show up as enlargement of martensite needles and coarse-grain fracture. This leads to a high brittleness of quenched products and the formation of cracks. Cracks often form at the boundaries of initial austenite grains. A low quenching temperature or too short a holding time at the given temperature causes incomplete quenching. In this case, a quenched metal is insufficiently hard. 3.6.5.2.2 Thermochemical Treatment Carburization is associated with surface saturation of steel with carbon and nitrogen. These elements quickly dissolve in iron by the interstitial method and are capable of rapid diffusion to a considerable depth. Products made of low-carbon (up to 0.25%) steels are subject to carburization. Carburization is carried out at 900–9508C (1650–17508F) and sometimes 1000– 10508C (1800–19008F). Gas carburization is used mostly, under which steel is heated in the atmosphere generated from natural gas (containing predominantly CH4) or from liquid hydrocarbons (kerosene, gasoline, etc.). Carburization is aimed at enrichment of the surface layer 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 layer is greater than the specific volume of the core, and therefore considerable compression stresses 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 a cyanide-containing salt bath. The carbon–nitrogen ratio in the diffusion layer is controlled by changing the medium’s composition and the processing temperature. Advantages of cyaniding over carburization consist in a shorter process time and improved wear and corrosion resistance (owing to the presence of nitrogen in the surface layer). ß 2006 by Taylor & Francis Group, LLC. 3.6.5.3 Tempering The 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 in the bainite range that is more disperse than the one formed after cooling in air. But the most disperse 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 stress–yield stress ratio may be taken as a measure of the tempering efficiency. The tempering efficiency depends on the cross-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 exceeded over the entire cross section, full tempering does not require this procedure. For example, in a quenched steel that has martensite in the surface zone and pearlite in the core, the hardness of the core sometimes may be higher than that of the surface zone after tempering. This is especially the case during a short tempering when precipitation of carbides from martensite proceeds faster than the coagulation of pearlite plates. Tempering of hypoeutectoid steels, which do not contain free ferrite, provides a uniform improved structure. In the presence of ferrite precipitates, the fracture stress–yield stress ratio decreases and the impact strength is smaller than in the surface zone. Therefore, in selecting the content of carbon and alloying elements and particular conditions of austenitization and cooling, the size of the product to be tempered must be considered. For tempering to yield adequate properties, it often suffices to suppress the formation of ferrite during continuous cooling. Only when a very high fracture stress is required an abrupt cooling is used for tempering. In this case, susceptibility to full tempering can be improved by raising the quenching temperature and thus enlarging the austenitic grain size. REFERENCES 1. 2. 3. 4. 5. 6. 7. N.F. 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 READING M.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 Steels— Handbook, 3rd ed., Vols. 1–3, 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 (in Russian). 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 (in Russian). ß 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 (in Russian). R. Zimmerman and K. Gunter, Metallurgy and Materials Science—Handbook, Metallurgiya, Moscow, 1982 (in Russian). ß 2006 by Taylor & Francis Group, LLC. 4 Effects of Alloying Elements on the Heat Treatment of Steel Alexey V. Sverdlin and Arnold R. Ness CONTENTS 4.1 4.2 4.3 4.4 4.5 Effects of Alloying Elements on Heat Treatment Processing of Iron–Carbon Alloys............................................................................................... 166 4.1.1 g- and a-Phase Regions.................................................................................. 166 4.1.2 Eutectoid Composition and Temperature ...................................................... 169 4.1.3 Distribution of Alloying Elements ................................................................. 171 4.1.4 Alloy Carbides................................................................................................ 172 Effect of Alloying Elements on Austenite Transformations ...................................... 174 4.2.1 Influence of Alloying on Ferrite and Pearlite Interaction .............................. 175 4.2.2 Effect on Martensite Transformation............................................................. 177 4.2.3 Retained Austenite ......................................................................................... 179 4.2.4 Effect on Bainite Transformation................................................................... 181 4.2.5 Transformation Diagrams for Alloy Steels .................................................... 183 Hardening Capacity and Hardenability of Alloy Steel .............................................. 185 4.3.1 Hardness and Carbon Content....................................................................... 185 4.3.2 Microstructure Criterion for Hardening Capacity ......................................... 187 4.3.3 Effect of Grain Size and Chemical Composition ........................................... 189 4.3.4 Boron Hardening Mechanism ........................................................................ 193 4.3.5 Austenitizing Conditions Affecting Hardenability ......................................... 195 Tempering of Alloy Steels.......................................................................................... 196 4.4.1 Structural Changes on Tempering.................................................................. 196 4.4.2 Effect of Alloying Elements............................................................................ 197 4.4.3 Transformations of Retained Austenite (Secondary Tempering) ................... 198 4.4.4 Time–Temperature Relationships in Tempering ............................................ 199 4.4.5 Estimation of Hardness after Tempering ....................................................... 199 4.4.6 Effect of Tempering on Mechanical Properties .............................................. 200 4.4.7 Embrittlement during Tempering ................................................................... 201 Heat Treatment of Special Category Steels................................................................ 201 4.5.1 High-Strength Steels ....................................................................................... 201 4.5.2 Boron Steels.................................................................................................... 202 4.5.3 Ultrahigh-Strength Steels ............................................................................... 202 4.5.4 Martensitic Stainless Steels............................................................................. 204 4.5.5 Precipitation-Hardening Steels ....................................................................... 205 4.5.5.1 Structural Steels................................................................................ 205 4.5.5.2 Spring Steels ..................................................................................... 206 4.5.5.3 Tool Steels ........................................................................................ 207 4.5.5.4 Heat-Resistant Alloys....................................................................... 208 4.5.6 Transformation-Induced Plasticity Steels ....................................................... 208 ß 2006 by Taylor & Francis Group, LLC. 4.5.7 Tool Steels ...................................................................................................... 209 4.5.7.1 Carbon Tool Steels ........................................................................... 209 4.5.7.2 Alloy Tool Steels .............................................................................. 209 4.5.7.3 Die Steels .......................................................................................... 210 4.5.7.4 High-Speed Steels ............................................................................. 210 Further Reading ................................................................................................................. 211 4.1 EFFECTS OF ALLOYING ELEMENTS ON HEAT TREATMENT PROCESSING OF IRON–CARBON ALLOYS A steel that contains, in addition to iron and up to 2% carbon, specially introduced chemical elements not found in a usual carbon steel is called an alloy steel. Chemical elements purposely added into steel are termed alloying elements. Steels may contain various numbers of alloying elements, and accordingly they are classified as ternary steels, which have, along with Fe and C, one specially introduced alloying element; quaternary steels, which contain two additional alloying elements, and so on. Alloying elements impart a wide variety of microstructures 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, are usually 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 in changes in the mechanical, chemical, and physical properties of the steel. Improvement of the properties of steel in accordance with its designated purpose is the main goal of alloying. The level to which the properties of steel are changed by alloying depends on the amount of alloying elements 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 steel should begin with consideration of the relationship between particular alloying elements, and Fe and C. What should be considered is the effect of alloying elements on the critical points of iron and steel, and also the distribution of the alloying elements in the steel. 4.1.1 g- AND a-PHASE REGIONS The position of the critical points A3 and A4 and the location of the eutectoid temperature A1 are of great significance because they determine the lowest temperature to which a steel should be heated for quenching, annealing, or normalization as well as the temperatures of the maxima in the cure showing the precipitation rate of undercooled austenite. The processes 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 A3 and A4. The alloying elements are accordingly divided into two large groups, each in turn broken down into two subgroups. Addition of the elements from the first group is followed by lowering of the critical point A3 and a simultaneous rise of the point A4. This effect is shown schematically in Figure 4.1 and is most vividly pictured in Fe–Mn and Fe–Ni equilibriums. It is seen that with an increase in the content of the alloying element, the g-phase region broadens considerably, and starting from a certain concentration the alloys are found in the state of the g-solid solution until they melt. 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 limited solubility in iron. Given a certain concentration of these elements in iron alloys, chemical compounds are formed and eutectic or eutectoid transformations are observed. In other ß 2006 by Taylor & Francis Group, LLC. Fe–C system Scheme α 1,528 1,600 Temperature, C 1,400 1,400 A4 1,200 1,000 906 A3 γ γ 800 600 400 α 200 0 10 20 30 40 50 60 70 80 90 100 Alloying element, % (a) α (b) Ni, % FIGURE 4.1 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with extended g-phase range and unlimited solubility. words, heterogeneous regions appear in diagrams of the iron-alloying elements system. The heterogeneous regions limit the g-phase occurrence range. This type of phase diagram of alloys is exemplified in Figure 4.2. As is seen, with an increase in the concentration of the alloying element in the alloy, the critical point A3 lowers and A4 rises. As a result, the range of g-solid solutions widens. But then, owing to the formation of heterogeneous regions, the g-phase narrows and, finally, vanishes. Equilibrium diagrams of this type (exhibiting first a wide range of the g-phase and then a narrowing of the phase caused by the appearance of heterogeneous regions) are found for N, C, Cu, Zn, Au, Re, etc. As distinct from the elements of the first group, elements entering the second group elevate the point A3 and lower the point A4 as their content in the alloy is raised. This leads initially to narrowing and then to a complete closing of the region of the g-solid solution as shown schematically in Figure 4.3. This shift of the critical points of alloys is induced by such elements as Cr, Mo, W, Si, T, Al, and Be (Cr group). These elements can be placed in the first subgroup of the second group of alloying elements. The second subgroup includes elements Fe–C system Scheme Temperature, C 1529 1400 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 α A4 γ 906 A3 α A1 α γ α 0 1 2 3 Carbon, % 0 10 20 30 40 Alloying element, % (a) (b) Fe3C, % FIGURE 4.2 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with extended g-phase range and limited solubility. ß 2006 by Taylor & Francis Group, LLC. Temperature, 8C Scheme 1529 1400 A4 Fe–Cr system 1800 1600 1400 α 1200 γ 1000 800 α 600 400 200 α γ 906 A3 α Alloying element, % (a) (b) 0 10 20 3040 50 60 70 80 90 100 Cr, % FIGURE 4.3 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with closed g-phase range. whose introduction causes the appearance of other phases in the equilibrium diagrams before the g-phase range is closed. It follows from Figure 4.4 that in this case the narrow range of the g-phase is limited by adjacent heterogeneous regions. Equilibrium diagrams of this type (with a narrow range of the g-phase and its limitation by an adjacent heterogeneous region) are due to Zr, Ta, Nb, Ce, etc. The above-described division of alloying elements into two large groups can be applied to ternary and more complex systems. The first basic ternary diagram is obtained when iron is alloyed with two elements, each leading to broadening of the g-phase range in binary iron alloys. Such alloys can be exemplified by Fe–Co–Ni, Fe–Co–Mn, and Fe–Ni–Mn. The second basic diagram covers iron alloys with two elements, which close the g-phase range. An example of these alloys is the Fe–Cr–Mo system, but it includes, along with regions of 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 the elements widens the g-phase range and the other element closes it. An example is the Fe–Cr–Ni system, which is important in technical terms. Thus even ternary systems may include purely ferritic (a-phase) and purely austenitic (g-phase) alloys as well as alloys possessing a multiphase structure. Scheme Fe–Ta system 1800 1600 Temperature C 1529 1400 α γ 1400 1200 γ 1000 906 800 α (a) α Alloying element, % (b) 600 α 0 10 20 30 40 Ta, % FIGURE 4.4 Scheme (a) and equilibrium diagram (b) for Fe and alloying elements with narrow g-phase range limited by adjacent heterogeneous region. ß 2006 by Taylor & Francis Group, LLC. 4.1.2 EUTECTOID COMPOSITION TEMPERATURE AND The aforementioned division of alloying elements into groups according to their influence on allotropic transformations in alloys of the iron-alloying element system allows one to predict to 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 form to another, it can be expected that the elements extending the g-phase range (Ni group) will lower the a ! g iron transition point Ac3, while the elements narrowing the g-phase range (Cr group) 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 of the critical point Ac1. As is seen, the elements narrowing the g-phase range do raise the critical point 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 relationship between the limiting concentration necessary to close the g-phase range in iron-alloying element alloys and the degree of elevation of the point Ac1. The lower the concentration of the element at which 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 critical points in the same direction, the critical points usually lower or elevate to a greater extent than would be the case if only one of the elements exerted its influence. But here the result cannot be presented by a simple dependence. If a carbon steel contains alloying elements that have an opposite effect on the position of the critical points during heating, the influence of the elements shows up differently depending on their quantitative ratio. Table 4.1 gives values of the critical points during heating and cooling for some multialloy steels. As is seen, rather high Ac1 and Ac3 points are characteristic of chromium–silicon and chromium– silicon–molybdenum steels of the heat-resistant type, high-chromium steels of the stainless type, chromium–molybdenum–vanadium steels, and others. The data of Table 4.1 are also interesting in that they show a simultaneous effect of the most significant alloying elements on the critical points Ar3 and Ar1 under constant-rate cooling. In particular, a very sharp lowering of these points in multialloy steels is caused by molybdenum. Molybdenum is responsible for a drastic drop of the critical points under cooling in steels that contain chromium and nickel at the same time. This last fact is especially important for structural steel. Eutectoid temperature, 8C 1300 Ti 1200 Mo Si 1100 1000 W Cr 900 800 700 600 Ni 500 0 2 Mn 4 6 8 10 12 14 16 18 Alloying elements, % FIGURE 4.5 Effect of alloying elements on the eutectoid transformation temperature Ac1. ß 2006 by Taylor & Francis Group, LLC. 170 ß 2006 by Taylor & Francis Group, LLC. TABLE 4.1 Position of Critical Points during Heating and Cooling of Some Multialloy Steels Critical Pointsb (8C) Chemical Compositiona (%) Heating Cooling C Mn Si Cr Ni Mo V W Ac1 Ac3 Ar3 Ar1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17c 0.20 1.10 0.06 0.36 0.45 0.30 0.34 0.25 0.22 0.35 0.36 0.29 0.60 0.28 0.15 0.12 0.13 — — — — — — — 1.26 1.15 — — — — — — — — 2.25 1.55 — — — — — — — — — — — — — — — 1.50 4.83 5.33 0.96 1.00 4.82 4.72 — — — — 0.64 0.75 0.86 11.27 12.19 13.15 — — — — — — — — — 2.73 2.87 3.44 1.40 1.97 0.13 — — — 0.51 0.57 0.20 0.35 2.20 1.65 0.41 1.18 0.39 0.39 — 0.22 0.35 — 0.58 — — — — — 0.26 0.42 — — — — — — — — — — — — — — — — — 0.93 — — 0.19 — — — — — — — 820 845 830 770 775 885 780 745 750 710 715 715 730 740 810 830 890 860 880 880 810 790 920 820 840 840 765 760 770 765 785 860 870 900 800 865 820 745 700 745 310 750 540 495 515 600 540 490 770 700 810 715 760 750 680 650 710 225 690 430 400 400 525 630 390 715 — 785 a The 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). c No. 17 also has 0.25% Al. b Steel Heat Treatment: Metallurgy and Technologies No. Eutectoid carbon content, % 0.80 0.60 0.40 0.20 0 Cr Mo Si W Mn Ti 2 Ni 4 6 8 10 12 14 16 18 Alloying 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 respect not only to temperature but also concentration. Figure 4.6 illustrates how the content of alloying elements in steel affects the carbon concentration at the eutectoid point. As can be seen from the figure, all the alloying elements shift the eutectoid point to the left, i.e., toward lowering of the carbon concentration, and consequently decrease the carbon content of alloy pearlite. In analogy to the shift of the eutectoid point to the left, the addition of most alloying elements in steel is followed by a leftward displacement of the point E in the Fe–C equilibrium diagram, which determines the solubility limit of carbon in austenite. The point E is shifted most by Cr, Si, W, Mo, V, and Ti, which are arranged here in ascending order of their influence. All these elements narrow the g-phase range in alloys of the iron-alloying element system. If a carbon steel contains a certain amount of an alloying elements, point E is displaced to the left to such an extent that even at a carbon concentration of several tenths of a percent the steel structure may have ledeburite, which is present in pure iron–carbon steels only when the carbon content is over 1.7%. It is of interest to note that the more strongly an element shifts the points E and S, the lower the element concentration at which it closes the g-phase range in the iron-alloying element diagram. Therefore a leftward shift of the points E and S can be considered as the tendency 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 a shift 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 DISTRIBUTION OF ALLOYING ELEMENTS In commercial alloy steels, which are multicomponent systems, alloying elements can be found (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 in cementite or in the form of independent compounds with carbon (special carbides); or (5) as a solution in iron. As to the character of their distribution in steel, alloying elements may be divided into two groups: 1. Elements that do not form carbides in steel, such as Ni, Si, Co, Al, Cu, and N 2. Elements that form stable carbides in steel, such as Cr, Mn, Mo, W, V, Ti, Zr, and Nb The law determining the manner in which elements of the first group are distributed in steel is 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 ß 2006 by Taylor & Francis Group, LLC. solutions with iron. The only exceptions are Cu and N. Copper dissolves in a-iron at normal temperatures in amounts of up to 1.0%. If the Cu content exceeds 7%, iron will contain copper in the free state as metal inclusions. Similar behavior is typical of the alloying elements that do not dissolve in solid iron at all (e.g., Pb or Ag). Nitrogen also has a limited solubility in ferrite. When the N content is higher than 0.015%, nitrogen is found in steel in the form of chemical compounds with iron or some alloying elements (V, Al, Ti, Cr). These chemical compounds 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 are not used in the usual commercial steels. Therefore it can be assumed that the common quantity-produced steels do not have intermetallic compounds of alloying elements with iron or with each other. Intermetallic compounds are formed in high-alloy steels, a fact that is of great significance for these steels. Alloying elements, whose affinity for oxygen is greater than that of iron, are capable of forming oxides and other nonmetal compounds. When added at the very end of the steel melting 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 that alloying elements that are deoxidizers are introduced at the final stages of the steel melting process, the majority of oxides have no time to coagulate or to pass to slag, and as a result they are retained in the solid steel as fine nonmetal inclusions. In addition to a great affinity for oxygen, some alloying elements have a greater affinity for sulfur than iron does, and upon introduction into steel, they form sulfides. Compared to noncarbide-forming elements, alloying elements that form stable carbides in steel exhibit a much more complicated distribution. They can be found in the form of chemical 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 of other carbide-forming elements. If a steel contains a relatively small amount of carbon and a great quantity of an alloying element, then obviously carbon will be bound to carbides before the carbide-forming elements are completely used. For this reason excess carbide-forming elements will be found in the solid solution. If a steel has a large amount of carbon and little of the alloying elements, the latter will be present in steel mainly as carbides. Carbide formation is treated in detail in the next section. Note in conclusion that most alloying elements, except C, N, O, B, and metalloids standing far from iron in the periodic table, dissolve in great amount in iron. The elements standing to the left of iron in the periodic table are distributed between iron (base) and carbides; those to the right of iron (Co, Ni, Cu, etc.) form solutions with iron only and do not enter into carbides. Thus one can state that alloying elements dissolve predominantly into basic phases (ferrite, austenite, cementite) of iron–carbon alloys or form special carbides. 4.1.4 ALLOY CARBIDES Carbides are formed in steels only by iron and metals that stand to the left of iron in the periodic table: Mn, Cr, W, V, Zr, Nb, Ti. Here the elements are arranged in accordance with their affinity for carbon. The elements at the left end of the row form relatively unstable carbides that dissociate readily on heating. In contrast, the elements at the right end of the row form extremely stable carbides that dissociate at temperatures much higher than the critical points of steel. Similar to iron, the above-mentioned carbide-forming elements refer to the elements of transition groups but possess a less perfect d-electron band. The further left a carbide-forming element stands in the periodic table, the less perfect is its d-band. ß 2006 by Taylor & Francis Group, LLC. There is reason to believe that during carbide formation carbon donates its valence electrons to fill the d-electron band of the metal atom, while valence electrons of the metal form a metal bond, 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 the general principles of carbide formation in steels: only metals whose d-electron band is filled less than that of iron are carbide-forming elements. Their activity as carbide-forming elements is greater and the stability of the carbide phases formed is the higher, the less perfect is the d-band of the metal atom. This principle allows specifying conditions of carbide formation in steels in the presence of several carbide-forming elements, the sequence of dissolution of various carbides in austenite, 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 will increase in going from Mn and Cr to Mo, W, V, Ti, and other elements whose d-bands are less perfect than those of Mn and Cr. This means that if a steel contains, e.g., both chromium and vanadium, 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 all carbide-forming elements except Fe, Mn, and Cr, the ratio of atomic radii of carbon and metal is less than 0.59. It is known that if the ratio of atomic radii of a transition group metal and a metalloid with a small atomic radius (C, N, H) is less than 0.59, special types of compounds called interstitial phases can be formed. The carbon=metal ratio of most carbide-forming alloying elements is lower than 0.59, and therefore the elements and carbon are capable of forming interstitial phases. It was found that the following carbide compounds may be formed in steels: Carbides of Group I Fe3C Mn3C Cr23C6, Cr7C3 Fe3Mo3C Fe3W3C Carbides of Group II Mo2C W2C, WC VC TiC NbC TaC, Ta2C ZrC However, the above carbides are not found in steels in pure form. Carbides of all alloying elements contain iron in solution, and if other alloying elements are present, they include these elements too. Thus, in a chromium–manganese steel the carbide (Cr, Mn, Fe)23C6, which contains iron and manganese 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 the presence of titanium and niobium, for example, rather than two kinds of carbides forming, a single carbide will be formed that includes titanium and niobium on equal terms. For this reason possible variants of carbide formation are fewer, and actually we have only six kinds of carbides in steels: Carbides of Group I M3C M23C6 M7C3 M6C ß 2006 by Taylor & Francis Group, LLC. Carbides of Group II MC M2C where M denotes a sum of carbide-forming (metal) elements. The carbides placed in group I possess a complicated crystal structure; an example is cementite. A specific structural feature of the carbides of group II as interstitial phases is a simple crystal lattice. They usually crystallize with a great carbon deficiency. It is worth noting that interstitial phases dissolve poorly in austenite and may not pass into solid solution even at high temperatures. This distinguishes interstitial phases from the carbides of group I, which readily dissolve in austenite on heating. All carbide phases have a high melting temperature and high hardness. Interstitial phases surpass the carbides of group I in this respect. 4.2 EFFECT OF ALLOYING ELEMENTS ON AUSTENITE TRANSFORMATIONS The overwhelming majority of alloy steels are used after heating to the austenite stage, quenching, and subsequent annealing. During quenching and annealing, austenite transforms with three types of transformations possible: pearlite transformation (often called diffusive transformation, precipitation to the ferrite–carbide 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 constant temperature 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 undercooled austenite. 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 continuous temperature variation rather than under isothermal conditions. Under continuous cooling, transformations occur at a lower temperature and take a longer time than in the isothermal case. Alloying elements have considerable influence on the kinetics and mechanism of all three types of transformations of undercooled austenite: pearlite, bainite, and martensite transformations. However, these elements influence austenite precipitation in different ways. Alloying elements that dissolve only in ferrite and cementite without the formation of special 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 the kinetics of isothermal transformations. Thus, the alloying elements forming group I carbides influence the austenite precipitation rate differently at different temperatures: At 1300–9308F (700–5008C) (pearlite formation), they slow the transformation. At 930–7508F (500–4008C), they dramatically slow the transformation. At 750–5708F (400–3008C) (bainite formation), they speed up the transformation. Thus, steels alloyed with carbide-forming elements (Cr, Mo, Mn, W, V, etc.) have two maxima of the austenite isothermal precipitation rate separated by a region of relative ß 2006 by Taylor & Francis Group, LLC. A1 Temperature A1 Cr, W, V, Mosteel Ni, Mn, Sisteel Carbon steel Carbon steel M M Time (a) (b) Time FIGURE 4.7 Diagrams of isothermal austenite precipitation. (a) Carbon steel and steel alloyed with noncarbide-forming elements; (b) carbon steel and steel alloyed with carbide-forming elements. stability of undercooled austenite (Figure 4.7). The isothermal precipitation of austenite has two clearly defined intervals of transformation: (1) to a lamellar structure (pearlite transformation) and (2) to a needle-like structure (bainite transformation). At temperatures lower than those indicated above and given greater degrees of undercooling as in the g ! a transformation start temperature, the martensite transformation develops in alloy steels. As a result, a supersaturated a-iron-based solid solution is formed. The remainder of this section considers in more detail the influence of alloying elements on the mechanism and kinetics of austenite precipitation for all three types of transformations. 4.2.1 INFLUENCE OF ALLOYING ON FERRITE AND PEARLITE INTERACTION The most important practical feature of alloying elements is their capacity to decrease the austenite precipitation rate in the region of the pearlite transformation, which shows up as a rightward shift of the line in the isothermal austenite precipitation diagram. This favors a deeper hardening and undercooling of austenite up to the range of martensite transformation under 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, special carbides and a ferrite–cement mixture (pearlite) are formed. Particular alloying elements and their amounts in the initial g-solid solution determine the rates of the individual steps of 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, as distinct from the martensite transformation (under greater undercooling), which proceeds through ordered shear. As mentioned above, all alloying elements dissolved in austenite, except Co, slow the pearlite transformation and shift the top section of the isothermal austenite precipitation curve to the right. The nature of the increase in stability of undercooled austenite under the influence of alloying elements is rather complicated. Whereas in carbon steels the pearlite transformation is associated with the g ! a rearrangement of the lattice and diffusion redistribution of carbon, ß 2006 by Taylor & Francis Group, LLC. in alloy steels these processes can be supplemented with the formation of special carbides and diffusion redistribution of alloying elements dissolved differently in ferrite and carbide. Not only do austenite-dissolving elements have small diffusion coefficients of their own which are sometimes several orders of magnitude smaller than that of carbon, but also some of them (e.g., Mo, W) slow the diffusion of carbon in the g lattice. Besides, some of the elements (e.g., Cr, Ni) retard the g ! a rearrangement, which is part of the pearlite transformation. Depending on the steel composition and degree of undercooling, the decisive role 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 and carbides. 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 eutectoid ferrite (pearlite). Every stage yields special carbides whose type depends on the austenite composition. If a steel contains carbide-forming elements (V, Nb, Ti, Zr) that pass into solid solution during austenitization, then a carbide of one type, MeC (VC, NbC, TiC, ZrC), is formed at all the stages. The scheme of austenite precipitation and the carbide formation process during pearlite transformation in steels with V, Nb, Ti, and Zr are as follows: g ! MeCA þ g0 ase þ ge MeCf þ aee Fe3 C þ ase ! MeCe þ aee Here g is the initial undercooled austenite; MeCA is the carbide precipitated in austenite; g0 is austenite 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 in excess ferrite; ge is austenite of eutectoid composition; Fe3C is the eutectoid cementite (pearlite); ase is the supersaturated eutectoid ferrite (pearlite); aee is the equilibrium eutectoid ferrite; 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 in austenite decreases with decreasing temperature. As is seen from the scheme, after the polymorphous 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 process lasts for a few seconds. In steels containing other carbide-forming alloying elements (Cr, Mo, W), the carbide formation process is much more complicated. Depending on their content in austenite, these elements 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 carbide formation. 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., the process of their redistribution during carbide formation. As mentioned above, the process of ß 2006 by Taylor & Francis Group, LLC. carbide formation is limited by the mobility of the carbide-forming elements. With a decrease in temperature their diffusion mobility diminishes and special carbides are not formed below 400– 5008C (750–9308F). At lower temperatures the intermediate (bainite) transformation takes place, and at higher undercooling rates martensite (diffusionless) transformation occurs. 4.2.2 EFFECT ON MARTENSITE TRANSFORMATION As in carbon steels, the martensite transformation in alloy steels takes place under rapid cooling from temperatures higher than the equilibrium temperature of the g ! a transformation (A1). At the martensite transformation temperature both the diffusion movement of metal atoms of iron and alloying elements and that of metalloid atoms of carbon and nitrogen are 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. In carbon-containing steels, the solid solution is supersaturated mainly with carbon, and in noncarbon-containing alloy steels, with alloying elements. The content of carbon and alloying elements 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 range of cooling rates. The martensite transformation kinetics of most carbon and structural and tool alloy steels is athermal in character. The athermal martensite transformation is characterized by a smooth increase in the amount of martensite as the temperature is lowered continuously in the martensite interval Ms–Mf, where Mf is the martensite finish temperature. As a rule, this transformation takes 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 substructure of the martensitic-quenched steel. At temperatures Ms below room temperature, lamellar (plate) martensite is formed in quenched iron–carbon and alloy steels. Crystals of this martensite are shaped as fine lenticular plates. In steels with the martensite point Ms higher than room temperature, lath martensite is formed during quenching. Crystals of this martensite have the form of approximately equally oriented thin plates, which are combined into more 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 transformation kinetics, the morphological type of martensite, the substructure of martensitic-quenched steels, and other phenomena are connected to a great extent with the martensite-start temperature 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 practical importance is also the martensite finish temperature Mf. Experiments concerned with the influence of alloying elements on the position of the martensite point show that Co and Al elevate the martensite start temperature, Si has little if any effect, and all the other elements decrease Ms (Figure 4.8). ß 2006 by Taylor & Francis Group, LLC. 700 Temperature, C 600 500 400 300 Ms 200 100 0 Mf −100 −200 0 0.4 (a) 0.8 1.2 1.6 2.0 Carbon content, % AI 300 0.76 % C Co Temperature M s, 8C 250 0.9% C 200 Si 150 Mn 1% C 100 Cu Ni V Cr 50 Mn 0 0 (b) 1 2 3 4 5 6 Amount of alloying element, % FIGURE 4.8 The influence of the content of (a) carbon and (b) alloying elements at 1% C on the martensite point position. The quantitative influence of alloying elements is approximately as follows (per 1 wt% of the alloying element): Element Mn Cr Ni V Si Mo Cu Co Al Shift of point Ms (8C) À45 À81 À35 À63 À26 À47 À30 À54 0 0 À25 À45 À7 À13 þ12 þ22 þ18 (8C) þ32 (8C) These data are given for carbon steels containing 0.9–1.0% C. For a wider range of C content, the quantitative influence of the elements can be different. In particular, it was established that the smaller the C content, the weaker is the influence of the alloying elements on the position of point Ms. The martensite start temperature of medium-carbon alloy steels can be estimated using empirical formula MH (8C) ¼ 520 À 320(% C) À 50(% Mn) À 30(% Cr) À 20[%(Ni þ Mo)] À 5[%(Cu þ Si)], where % C, % Mn, etc. are the contents of the corresponding elements in weight percent. ß 2006 by Taylor & Francis Group, LLC. The results of calculations by this formula for steels containing 0.2–0.8% C are in good agreement with the experimental data. However, for multialloy steels this formula does not always yield reliable data because if a steel contains several alloying elements it is impossible to determine their combined effect on the martensite point by simple summation. Thus, for example, Mn lowers the point Ms to a greater extent than Ni, but in a steel with a high Cr content its effect is weaker than that of Ni. The martensite point Ms is affected mostly by C dissolved in austenite (Figure 4.8). The transformation finish temperature Mf intensively decreases too, as the C content is increased up 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 is mainly a change in the relative thermodynamic stability of g- and a-phases of iron, because the martensite transformation itself is a g ! a transformation. 4.2.3 RETAINED AUSTENITE A characteristic feature of the martensite transformation in steels, whatever its character (athermal, explosive, or fully isothermal), is that transformation of austenite to martensite is never complete. Figure 4.9 shows the amount of martensite formed when the temperature is decreased continuously in the martensite range Ms–Mf (martensite curve) for the athermal type of martensite transformation. The transformation starts at the point Ms, and the amount of 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 lower the amount of retained austenite. Investigations show that martensite curves of different steels, both carbon and steels and steels alloyed with different elements and in different amounts, exhibit approximately the same 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 martensite temperature range, i.e., on the Ms–Mf temperature difference; it increases as the range narrows. But the martensite range itself depends on the position of the martensite point Ms: the range narrows as the point Ms lowers. 100 A, % Martensite content, % 90 80 50 40 30 20 0 ß 2006 by Taylor & Francis Group, LLC. Mf 60 10 FIGURE 4.9 Martensite curve. A, % 70 Ms 20 Temperature, C −100 100 Mn Cr Heating 11508C 90 Retained austenite, % 80 50 70 g tin C 12 Cr a He 60 50 Ni 40 Cu 30 W 20 V Co 10 AI 0 2 1 3 4 5 6 7 8 Alloying 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 the quenching of steels should qualitatively and, to a great extent, quantitatively correspond to their influence on the position of the martensite start point Ms. Available experimental data show that alloying elements that lower and raise the martensite start point increase and decrease the amount of retained austenite, respectively (Figure 4.10). Besides, a certain sequence in the arrangement of the elements is observed from the point of view of their quantitative influence. In particular, the largest amount of retained austenite in accordance with their influence on the position of the martensite point is due to Mn, Cr, Ni, etc. As illustrated in Figure 4.11, the influence of these elements on the martensite range follows the same sequence. Martensite range, 8C 400 0.6% C 300 Ni Cr Ni 200 Mn 1.0% C Cr 100 Mn 0 2 4 6 Alloying element content, % 8 FIGURE 4.11 The influence of alloying elements on the martensite range. ß 2006 by Taylor & Francis Group, LLC. In alloy steels, the martensite point Ms lowers most and the martensite range narrows under the influence of carbon. Therefore the influence of carbon on the amount of retained austenite is much stronger than that of alloying elements. An increase in the C content of chromium–nickel 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 the amount of retained austenite to ~6% only. The fact that carbon promotes the greatest retention 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 of retained austenite than the law of summation suggests. However, in multiple alloy steels too, the relationship between lowering of the martensite point under the influence of a given element and an increase in the amount of retained austenite caused by the same element persists in the main. In addition to the content of carbon and alloying elements, other factors can influence the amount of retained austenite formed during quenching of steel. The most important of these is 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 it affects 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 austenite to isothermally formate martensite at temperatures a little lower than the point Ms is realized here. At temperatures close to the martensite finish temperature Mf but within the interval Ms–Mf, when a rather significant amount of martensite has been formed already, acceleration of 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 temperature makes retained austenite less prone to subsequent transformation. With slow cooling the austenite stabilizing processes have time to near completion and the transformation proceeds more slowly. Austenite stabilization is associated with stress relaxation. The longer the holding time, the greater the stress relaxation and the greater the degree of the metal cooling needed to accumulate stresses required for the martensite transformation to continue. The quenching temperature can influence largely, either directly or indirectly, the amount of retained austenite. Its direct effect can be connected with thermal stresses facilitating the transformation of austenite. An indirect effect of the quenching temperature is associated with enrichment of intercrystallite boundaries of austenite in carbon and alloying elements and, primarily, with the transfer of carbides, ferrite, and other phases to the solution. If a steel is heated to a temperature falling within the interval between the critical point Ac1 and the temperature of full dissolution of ferrite or carbides, the heating temperature will determine the content of carbon and alloying elements in austenite. If carbides dissolve above Ac1, then the amount of retained austenite will increase with quenching temperature. If the quenching temperature is elevated above Ac1 and excess ferrite dissolves (with resulting decrease in the austenite concentration), then the martensite point will occupy the lowest position when a steel is quenched from temperatures slightly higher than the point Ac1. Correspondingly, the amount of retained austenite must be the largest at these temperatures and must subsequently decrease until the temperature of full ferrite dissolution is reached. 4.2.4 EFFECT ON BAINITE TRANSFORMATION The bainite transformation (stage II transformation) takes place in carbon steels under the precipitation curve of undercooled austenite (C curve) in the interval of approximately 500– 2508C (930–4808F). This is called the intermediate transformation. It occurs in between the pearlite and martensite transformations. The kinetics of this transformation and the structures produced are similar to those observed during the diffusion pearlite or diffusionless ß 2006 by Taylor & Francis Group, LLC. 800 Temperature, C 700 2 600 1 500 400 3 300 4 200 1 101 102 103 104 Time, s FIGURE 4.12 The austenite precipitation diagram of alloy steels with separate C curves of pearlite and bainite transformations. 1, Start of pearlite formation; 2, finish of pearlite formation; 3, start of bainite formation; 4, finish of bainite formation. martensite transformation. A mixture of the a-phase (ferrite) and carbide is formed as a result of 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 maximum transformation rate (minimum incubation period) depends mainly on the chemical position of the steel. For alloy steels, C curves of pearlite and bainite transformations can be separated by a temperature interval of a highly stable undercooled austenite where pearlite does not precipitate for many hours, while undercooling is insufficient for the bainite transformation (Figure 4.12). Alloying elements affect the kinetics of the intermediate transformation, although to a lesser degree than in the case of the pearlite transformation. In some alloy steels, the isothermal 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 that range. In steels alloyed with 2% Si or Cr, the transformation of austenite stops even at the lowest 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 in the intermediate range. Figure 4.13 shows the influence of some alloying elements on these parameters for high-carbon steels with 1.0% C. As is seen, Mn and Cr strongly influence the kinetics of the intermediate transformation, increasing the duration of the incubation period and 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 the pearlite transformation, does not have a pronounced effect on the kinetics of the intermediate transformation. The intermediate transformation in alloy steels consists of a diffusion redistribution of carbon in austenite, diffusionless g ! a transformation, and formation of carbides, namely ß 2006 by Taylor & Francis Group, LLC. 104 1% Cr–Mn 1% Cr–0.3% Mn τ, s 103 1% Cr–Mo 102 1% Cr–W 101 1 (a) 1 2 3 4 5 6 440 tms, 8C 400 1% Cr–W 380 1% Cr–0.3% Mn 320 1% Cr–Mo 1% Cr–Mn 280 1 (b) 2 3 4 5 6 10 1% Cr–0.3% Mn nmax, %/s 1.0 1% Cr–W 1% Cr–Mo 0.1 0.01 1% Cr–Mn 0.001 (c) 1 2 3 4 5 67 Alloying elements, % (by mass) FIGURE 4.13 The influence of Cr, Mo, W, and Mn (a) on the bainite period t at a minimum stability of austenite; (b) on the temperature of minimum stability tms, and (c) on the maximum transformation rate vmax in the intermediate range. e-carbide (a type of Fe carbide) and cementite. Owing to the low diffusion mobility of metallic alloying elements, which are substitutional impurities, special carbides are not formed during the intermediate transformation. The content of alloying elements in the e-carbide and cementite of bainite is the same as in the initial austenite. Alloying elements do not undergo redistribution during the bainite transformation. 4.2.5 TRANSFORMATION DIAGRAMS FOR ALLOY STEELS The kinetics of austenite transformation, i.e., the form of the precipitation diagram, depends on a variety of factors, primarily on the chemical composition of austenite. Depending on the alloying of a steel, it is possible to distinguish six basic versions of the diagram of isothermal precipitation 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 and intermediate stages are not separated. When these steels are subjected to continuous cooling, three types of structures—martensite, martensite and a ferrite–carbide mixture, and only a ferrite–carbide mixture—can be formed depending on the cooling rate. ß 2006 by Taylor & Francis Group, LLC. Temperature, C A1 700 A1 A1 1 3 4 2 500 4 Mf 300 Ms Ms Ms 25% Mf 100 (a) Temperature, C 2 3 500 50% 75% A1 3 2 Ms 25% (c) A1 4 300 Mf 75% (b) A1 700 50% 2 5 Mf 25% 75% Ms 50% 100 1 10 102 103 104 105 (d) 1 10 102 103 104 105 Time, 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 0.4–0.35% C) containing carbide-forming elements; (c) steels alloyed with Cr, Ni, Mo, and W and having a low content of carbon (up to 0.2–0.25% C); (d) alloy steels containing carbide-forming elements (over 0.4–0.5% 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 ferrite–carbon mixture; 4, start of formation of the intermediate 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 pearlite and intermediate transformations. Each of the ranges is characterized by its own C-shaped curves. When the carbon content of structural steels is up to 0.4–0.5%, the stage I transformation is shifted to the right relative to the stage II transformation (Figure 4.14b); if the carbon content is higher, stage I is found to the left of stage II (Figure 4.14d). Chromium–nickel–molybdenum and chromium–nickel–tungsten steels containing 0.15– 0.25% carbon (Figure 4.14c) are characterized by a rather high stability of undercooled austenite in the pearlite range and a low stability of undercooled austenite in the bainite range. As a consequence, stage I is absent from the austenite precipitation diagram. In high-alloy chromium steels, the intermediate transformation may be strongly inhibited and 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 below room temperature and stages I and II precipitation practically do not take place owing to a high content of Cr, Ni, Mn, and C (Figure 4.14f). Thanks to the high content of carbon in the austenite of these steels, excess special carbides may be formed on undercooling. It is worth noting that the aforementioned distinction of the diagrams is conventional to a certain measure as they do not cover a great variety of isothermal and thermokinetic precipitation diagrams of supersaturate austenite. ß 2006 by Taylor & Francis Group, LLC. 4.3 HARDENING CAPACITY AND HARDENABILITY OF ALLOY STEEL As noted in Section 4.2, at great rates of cooling when the cooling curves do not touch the region 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 fully martensitically hardened. Martensitic transformation of austenite results in a supersaturated solid solution of carbon in a-Fe; the higher the carbon content of the austenite, the more supersaturated the solution. Compared with other austenitic transformation products (pearlite and upper and lower bainite), martensite possesses the greatest hardness and gives very hard steels. The ability of a steel to increase in hardness during quenching is called its hardenability or hardening capacity. The hardening capacity is characterized by the maximum hardness that can be obtained on the surface of a given steel product by quenching. To achieve maximum hardness it is necessary to observe basic conditions: the rate of cooling should be equal to or higher than the critical rate at which quenching gives martensite alone (inevitability with some retained austenite, of course, but without bainite); all carbon at the quenching temperature should be in the solid solution in austenite (the quenching temperature should be above the critical points Ac1 and Ac3 by 30–508C (80–1208F) for hypereutectoid and hypoeutectoid steels, respectively). Alongside the notion of hardening capacity, broad use is made in practice of the notion of hardenability, though these two characteristics depend on different factors and are achieved in different ways. The hardening capacity of a steel is determined by the factors affecting the hardness of martensite, while its hardenability is determined by those affecting the quantity of the martensite obtained and the hardness penetration depth. Upon quenching, steel can feature high hardening capacity and low hardenability at the same time. Such a steel would correspond to the schematic curve 1 in Figure 4.15. If for a workpiece of the same diameter D cooled under the same conditions, the distribution of hardness over the cross section is characterized by curve 2; such a steel possesses medium or poor hardening capacity but good hardenability. Finally, steel that corresponds to curve 3 would possess high hardening capacity and high hardenability. 4.3.1 HARDNESS AND CARBON CONTENT The hardening capacity of a steel whose general characteristic could be maximum hardness depends mainly on the carbon content and, to a lesser extent, on the amount of alloying elements and austenite grain size. Increasing the carbon content of martensite increases its 2 Hardness 3 1 D FIGURE 4.15 Distribution of hardness over the cross section of workpiece for three steels differing in hardenability and hardening capacity. ß 2006 by Taylor & Francis Group, LLC. hardness (Figure 4.7). Note that the hardness of a quenched steel and the hardness of martensite 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 carbon concentration of 0.8–0.9% C and then decreases due to an increase in the volume fraction of soft retained austenite (Figure 4.16). For the above carbon content of steel, the martensite point Ms drops significantly, which leads to an increase in the proportion of retained austenite in quenched steel. Steel with 1.9% C quenched from a temperature higher than Ast has the same hardness as quenched steel with 0.1% C. If hypereutectoid steels are quenched from a temperature of Ac1 þ (20–308C) (70–908F), as is common practice, all hypereutectoid steels would have practically the same austenite composition at the same quenching temperature and level of hardness (Figure 4.16, curve b). Another important feature of the dependence of steel hardness on carbon content is that an increase in the carbon content to ~0.6% results in a most dramatic rise in the maximum hardness; then the curve becomes less steep. This is probably associated with the very nature of high martensite hardness in steel. The martensite transformation of austenite results in a supersaturated solid solution of carbon in a-Fe. An increase in the carbon content of martensite weakens, rather than strengthens, the interatomic bonds. This is due to an increase in the distance between iron atoms brought about by implanted carbon atoms. Carbon nevertheless increases the hardness of martensite, which is explained primarily by the fact that carbon atoms implanted into the a-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 martensite crystals surround dislocations (atmospheres around dislocations), thus pinning them. This leads to a general increase in plastic deformation resistance despite the fact that carbon weakens interatomic bonding in the martensite lattice. In steels with a high martensite start point Ms such as carbon steels containing less than 0.5% C [Ms > 3008C (5708F)], quench cooling over the martensite range is characterized by the most favorable conditions for partial precipitation of martensite with the release of disperse carbide particles. Moreover, in all steels hardened at normal rates, carbon has time to segregate as the steel cools above the point Ms. The carbon segregates of austenite are inherited by martensite, and since the latter is already supersaturated with carbon, these segregates become nucleation sites of carbide particles. This is in agreement with the fact that at very high cooling rates the hardness of 70 c b Hardness, HRC 60 50 a 40 30 20 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 C, % FIGURE 4.16 Hardness of carbon and alloy steels depending on the carbon content and quenching temperature. (a) Quenching above Ac3; (b) quenching above Ac1 (7708C); (c) microhardness of martensite. ß 2006 by Taylor & Francis Group, LLC. FIGURE 4.17 Microstructure of plate martensite. Light shading ¼ retained austenite 50,000Â. martensite crystals is two thirds of that obtained at normal cooling rates. High hardness of martensite may also be due to the fact that carbon makes a noticeable contribution to covalent bonding whose main property is high plastic deformation resistance. Owing to the above strengthening mechanisms, carbon has such a strong strengthening effect 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 is determined by the concentration of carbon. To conclude, extreme strengthening of steels during martensitic hardening is due to the formation of a carbon-supersaturated a-solution, an increase in the density of dislocations during the martensite transformation, the formation of carbon atom atmospheres around dislocations, and precipitation of disperse carbide particles from the a-solution. 4.3.2 MICROSTRUCTURE CRITERION FOR HARDENING CAPACITY Studies of the structure of hardened carbon steels and carbon-free iron-based alloys revealed two main morphological types of martensite: plate and lath. These two types of martensite differ 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 iron alloys with a high concentration of the second element—for instance, Fe–Ni alloys with a Ni content higher than 28%. Martensite crystals are shaped as thin lenticular plates. Such a shape corresponds to the minimum energy of elastic distortions when martensite is formed in the austenite 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 form frame-like ensembles. Plates that are formed first (near the point Ms) extend through the entire 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, new martensite plates are formed in the austenite sections, the size of the plates limited by the size of the matrix sections. In the course of transformation the austenite grain splits into still smaller sections, in which smaller and smaller martensite plates are formed. In the case of a small austenite grain caused by, for instance, a small overheating of steel above Ac3, the martensite 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 is most desirable. After quenching, martensite retains some austenite between its plates at room temperature (Figure 4.18). ß 2006 by Taylor & Francis Group, LLC. FIGURE 4.18 Microstructure of lath martensite. 50,000Â. Lath martensite (which is also called massive, high-temperature, or nontwinned) is a widespread 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 alloys—for instance, Fe–Ni alloys with a Ni concentration of less than 28%. Crystals of lath martensite are shaped like thin plates of about the same orientation, adjacent to each other and forming more or less equiaxial laths. The plate width within the lath is about the same everywhere, ranging from several micrometers to fractions of a micrometer (commonly 0.1–0.2 mm); i.e., it can reach or even exceed the resolution limit of a light microscope. Inside the martensite laths there are interlayers of retained austenite 20–50 nm thick. One austenite grain can contain several martensite laths. The formation of lath martensite has all the main features specific to the martensite transformation, including the formation of a relief on a polished surface. Transmission electron microscopy reveals a rather complicated fine structure of martensite crystals with many 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 the midrib is an area with a dense arrangement of parallel fine twin interlayers. The twinning plane in martensite of iron-based alloys is commonly {1 1 2}M. Depending on the composition of the alloy and martensite formation conditions, the thickness of the twinned interlayers may form several tenths of a nanometer to several tens of nanometers. On both sides of the central twinned zone there are peripheral areas of martensite plates that contain dislocations of relatively low density (109–1010 cmÀ2). The substructure of lath martensite is qualitatively different from the substructure of plate martensite in that there is no zone of fine twin interlayers. It is a complex dislocation structure characterized by high-density dislocation pileups with densities on the order of 1011–1012 cmÀ2, i.e., the same as in a metal subjected to strong cold deformation. The laths of lath martensite often consist of elongated slightly misoriented subgrains. Twin interlayers can occur in lath martensite, but their density is much lower than in the midrib in plate martensite, 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 higher density of imperfections occurring under local plastic deformation due to martensite crystals. Flat dislocation pileups, dislocation tangles, and stacking faults may be observed in austenite around martensite crystals. ß 2006 by Taylor & Francis Group, LLC. At present it is believed that the decisive role in the formation of plate martensite belongs to accommodating (complementary) twinning deformation, whereas for the lath type this role is played by slip. As temperature is decreased, resistance to slip increases at a higher rate than that of resistance to twinning; therefore, the martensite transformation at low and high temperatures results in twinned and lath martensite, respectively. In alloys a decrease in temperature 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 lath martensite; the numerals show the second component content in percent. System Fe–C Fe–Ni Fe–C2 0.6 0.6–2.0 Lath type, M Plate, M Fe–N 0.7 0.7–2.5 29 29–34 10 — Chromium–nickel, manganese, chromium–manganese, and other steel alloys with lowenergy stacking faults contain hexagonal e-martensite with plates in parallel to planes {1 1 1}g k {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 undergone martensite 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 EFFECT OF GRAIN SIZE AND CHEMICAL COMPOSITION As noted in Section 4.3.1, the hardening capacity of a steel, i.e., its ability to undergo martensitic hardening, depends mainly on its carbon content and to a lesser extent on its content of alloying elements and the size of austenite grains. At the same time these two factors—grain 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 zone can penetrate. It is reasonable, therefore, to consider the effects of grain size and chemical composition of austenite on hardening capacity and hardenability separately. FIGURE 4.19 Microstructure of e-martensite. 50,000Â. ß 2006 by Taylor & Francis Group, LLC. In the case of austenite–martensite transformation, martensite plates develop inside the austenite grain, extending from one side to the other. If the steel is considerably overheated above the critical point Ac3, then coarse grains of austenite are formed; this results in larger martensite plates than normal. In quenched steel the volume fraction of retained austenite is higher if its structure is dominated by large martensite plates. Although the hardness of martensite is practically independent of the plate size, an increase in the total soft austenite content of quenched steel leads to a decrease in its maximum hardness, i.e., impairs its hardening capacity. Moreover, the plastic properties of steel, particularly its toughness, also deteriorate 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 be achieved with a fine-grain austenite structure, which is produced at small overheating of steel at 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 elements tend to shift the isothermal austenite precipitation curves to the right and hence decrease the critical rate of cooling, it is easier to obtain maximum hardness for an alloy steel than for an ordinary 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 be applied to an alloy steel containing elements that form stable carbides by heating below the carbide dissolution point, then the carbon content of the major martensite mass will be lower than the total carbon content of the steel. As a result, the maximum achievable hardness for this carbon content will not be attained because the hardening capacity of the steel will deteriorate. It should be noted that high-carbon steel, such as tool grade steel, features high hardness even 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 hardness also decreases. This decrease is small and is due mainly to the fact that hardness versus carbon curves for carbon steels and alloy steels are rather smooth at carbon concentrations higher than 0.6% (Figure 4.16); it is compensated for by a considerably lower amount of retained austenite and the high hardness of the carbides themselves. The situation is different for structural steels containing less than 0.4% C. The maximum hardness curve is so steep (Figure 4.16) that even a small decrease in the concentration of carbon in martensite in an alloy steel due to incomplete dissolution of special carbides would lead to a considerable reduction in the hardness of the martensite. The steepness of the curve over the range of 0.1–0.4% C shows that from the viewpoint of hardening capacity it is essential to heed even the smallest fluctuations in the carbon content, specifically fluctuations within the quality limits, up to individual ingots. Therefore the carbon content limits for specific structural steels should be as narrow as possible, although this may encounter some technological difficulties. The indirect effect of alloying components on hardenability is not great. It is therefore possible to construct a general curve for the dependence of maximum hardness on the carbon content for carbon and alloy steels (Figure 4.16). This is understandable because martensite of an alloy steel is a combined solid solution in which atoms of the alloying elements replace iron atoms in the lattice, while carbon atoms are implanted into this lattice. Carbon atoms introduced into the a-Fe lattice impede the slip of dislocations in martensite and thereby increase its hardness. A certain increase in the hardness of martensite due to alloying elements can be expected only because of the strengthening of a-Fe during quenching. The possibility of quench ß 2006 by Taylor & Francis Group, LLC. 240 Cr 220 Hardness, Hv 200 Ni 180 Mn Nb 160 Ti Si 140 V W AI 120 Co 100 80 0 0.8 1.6 2.4 3.2 4.0 5.0 Alloying elements in α-Fe, % FIGURE 4.20 Effect of dissolved alloying elements on hardness of ferrite after quenching from 12008C in water. strengthening an alloyed ferrite has been studied in detail. From Figure 4.20, one can judge roughly how different elements increase the hardness of ferrite with ~0.01% C upon quenching from 12008C in water. Alloying elements causing substantial strengthening of ferrite should also increase the hardness of martensite in a quenched steel, though this increase in hardness 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 will increase with a decrease in the critical rate of quenching. This rate, in turn, depends on the resistance 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 lower the critical rate of quenching and the higher the hardenability of the steel products. Thus, the factors 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) the chemical composition of the steel (composition of austenite, to be more exact); (2) austenite grain size; and (3) the homogeneity of austenite. Under otherwise equal conditions, coarse austenite grains improve the hardening capacity of steel. This circumstance is connected with the extent of grain boundaries; the extent is less, the coarser the grain. Since nucleation centers are formed primarily along the austenite grain boundaries during austenite precipitation above the point Ms, it is always easier to undercool austenite with coarse grains, thereby increasing hardenability. To estimate hardenability, in practice, use is made of the quantity called the critical diameter. The critical diameter (Dcr) is the maximum diameter of a bar permitting through hardening for a given cooling medium. To avoid putting hardenability in dependence on the ß 2006 by Taylor & Francis Group, LLC. Critical harding rate, C/s 1,400 I 1,200 1,000 800 600 400 II 200 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 C, % FIGURE 4.21 Curves showing the effect of carbon on the critical rates of quenching. I, low-heating temperatures; II, high-heating temperatures. method of cooling and type of coolant, use is made of still another notion, the ideal critical diameter (D1), which is the diameter of a maximum section allowing through hardening in an ideal cooling liquid that is absorbing heat at an infinitely great rate. It was established that the grain 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. This is due primarily to the fact that carbon and alloying elements affect the critical rate of quenching. Figure 4.21 shows the effect of carbon on the critical rate of quenching for carbon steel. It can be seen that the minimum rate of quenching is observed in steels that are close to eutectoid with respect to carbon content (curve I). A decrease in the carbon content of steel below 0.4% leads to a sharp increase in the critical rate to the extent that at a certain minimum carbon content martensitic hardening becomes virtually impossible. An increase in the critical rate of hypereutectoid steel with an increase in the carbon content is explained by the presence of cementite nuclei facilitating the austenite precipitation. Hence, the trend of curve I is related 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 a continuous decrease in the critical rate of quenching (curve II), with a resulting rise in hardenability. The effect of alloying elements on hardenability can be estimated by the degree of increase or decrease in stability of undercooled austenite in the pearlite and intermediate ranges. With the 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 is made of such additives such as Mn, Ni, Cr, and Mo. Particularly effect is complex alloying whereby 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 iron–nickel steel. It can be seen that Mn, Cr, and Mo additives improve hardenability to a considerable extent. The improving effect of alloying on hardenability is used in two ways. First, alloying ensures through hardening across sections inaccessible for carbon steels. Second, in the case of small-section products, replacing carbon steel with an alloy steel permits less radical cooling regimes. Small-diameter carbon steel products can be hardened by quenching in water. 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 water can be replaced with softer hardening in emulsion, oil, or even air. ß 2006 by Taylor & Francis Group, LLC. Ideal critical diameter D, mm With % Mn 175 With % Cr 1 1.5 With % Mo 0.25 0.5 150 125 0.5 1 100 0.1 75 0.7 50 0 0 0.3 25 0 0 1 2 3 0 1 2 3 0 1 2 3 Ni, % FIGURE 4.22 The effect of manganese, chromium, and molybdenum on the hardenability of steels with different nickel contents. 4.3.4 BORON HARDENING MECHANISM It 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 steels increases considerably upon introduction of boron in amounts of thousandths of a percent. A further 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 preliminarily well 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, or zirconium. Figure 4.23 shows hardenability curves for a low-carbon steel without boron, with boron added, with boron and vanadium, and with boron, vanadium, and titanium. It can be seen that the extent of the martensite range for all the steels with boron added is greater than that in steels without boron; however, it is practically the same for steels with several additives. At the same time, the extent of the half-martensite structure range is much greater if the steel contains other additives along with boron. Hardness, HRC 70 4 60 3 50 40 1 2 30 20 0 2 4 6 8 10 12 Length from end, mm 14 16 FIGURE 4.23 Curves of hardenability for steels with 0.44–0.43% C and various small amounts of additives. 1, Without boron; 2, with an addition of boron; 3, with boron and vanadium added; 4, with boron, vanadium, and titanium added. ß 2006 by Taylor & Francis Group, LLC. The effect of boron on hardenability decreases with an increase in the carbon content. If the 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 for low-carbon steels of various applications. It does not have any effect on the hardenability of tool steels or on high-carbon carburized layers. The above relationship between the carbon content and the effect of boron on the hardenability of steel is due to the fact that the two elements have the same effect on austenite precipitation. Boron increases the length of the austenite precipitation incubation period, thereby decreasing the critical temperature of quenching. 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 the quenching temperature is sufficiently high [850–9008C (1560–16508F)]. Various suggestions have 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 produce its effect at very low concentrations above, above which no further effect on hardenability is observed. 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 to increase the grain size, it produces the same hardenability-improving effect in steels with small grains. Others explain the fact that boron increases the stability of austenite and consequently improves hardenability by the fact that it increases the coefficient of surface tension at the austenite–new-phase nucleus interface. Therefore more energy is required for the formation of 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 frequently explained by the fact that this element is surface-active in austenite. There is experimental evidence that boron is segregated at the boundaries of austenite grains and dissolves in insignificant amounts in this layer. Since boron forms an interstitial solid solution with iron, 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. This impedes the formation of new-phase nuclei of critical size, which are formed primarily at the austenite grain boundaries. The resistance of austenite to precipitation improves, thereby increasing the hardenability of the steel. The cessation of the boron effect on hardenability with increasing boron content is due to its low limiting solubility in g-iron at a given temperature [about 0.003% at 10008C (18328F)]. As soon as the boron solubility limit at the grain interface is achieved, any further increase in its total content leads to the formation of iron–boron 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 compound particles 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 grain caused 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 have any effect on hardenability and even may impair it. Based on this viewpoint, one can explain some other specific effects of boron on the hardenability of steel that were noted above. Thus, boron increases the duration of the austenite precipitation incubation period only, the duration determined by the formation of critical nuclei at grain boundaries but not affecting the length of austenite precipitation. From this viewpoint it is also clear that boron, like carbon, facilitates enlargement of austenite grains under heating. An increase in the quenching temperature first improves hardenability owing to an increase in the concentration of boron at the austenite grain boundaries to its ß 2006 by Taylor & Francis Group, LLC. solubility limit at these boundaries. A further increase in the quenching temperature and enrichment of the grain boundaries in boron can lead to the formation of boron–iron compounds. 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 boron and carbon have virtually the same effect on hardenability. In high-carbon steels, the effect of boron becomes practically negligible owing to its poor solubility in iron. 4.3.5 AUSTENITIZING CONDITIONS AFFECTING HARDENABILITY The austenite condition prior to quenching (chemical composition, grain size, homogeneity of austenite) has the decisive effect on the hardening capacity and, especially, hardenability of steel. Other factors are secondary or derive from the basic three. These factors, in turn, are determined by the carbon content, type, and amount of alloying elements at the time of quenching, the quenching temperature (austenitizing temperature), and the holding time at a given 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, and recrystallization of the austenite grains. To improve the hardening capacity and hardenability of steel, the austenitizing conditions should be such as to ensure that a maximum amount of carbon passes from the ferrite– carbide mixture to the solution and, at the same time, no marked growth of grains occurs as a result of overheating, as this would lead to a high brittleness and the formation of quenching cracks. The quenching temperature should be maintained as constant as possible, and the holding 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-grain carbides need more time than thin-plate and fine-grain ones. Steels alloyed with elements forming special carbides should be heated to a temperature considerably exceeding Ac3. Small carbides available in the structure impede enlargement of grains and the nuclei of the new phase facilitate transformation of austenite in the pearlite range and increase the critical rate of quenching, thus decreasing the hardenability of the steel. As the quenching temperature and time are increased, the critical rate of quenching decreases and, accordingly, hardenability rises, because carbides and other inclusions playing the role of new-phase nuclei dissolve most. The degree of austenite homogeneity and dispersion of local carbon pileups (which can act as nuclei during transformation in the pearlite range) can have a strong effect on hardenability and hardening capacity. When the quenching temperature is increased, carbides dissolve together with other minute, sometimes hardly measurable, quantities of inclusions such as nitrides and sulfides, which can also serve as nuclei during transformation. Finally, when the quenching temperature and holding time are increased, enlargement of austenite 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 in the critical rate of quenching and hardenability improves. Nearly all of the alloying elements impede the growth of austenite grains. The exception is manganese, 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 this retarding effect is believed to be the formation of low-soluble carbides, nitrides, and other phases, 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 ß 2006 by Taylor & Francis Group, LLC. Nb−C 0.1% C 0.4% C Zr−C % Zr 0.1% C Ti−C % Ti 0.4% C % Nb Content of metal in austenite, % %V V−C 0.1% C 0.1% C 0.8% C 0.4% C 1.2% C 0.4% C 1.2% C 1.2% C 0.8% C 0.8% C 0.8% C 1.2% C 800 900 1000 1000 1200 1000 1200 1000 1200 Temperature, C FIGURE 4.24 Solubility of carbides in austenite at various temperatures depending on the carbon content (shown as % C on the curves) for (a) vanadium; (b) niobium; (c) titanium; (d) zirconium. show 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 great effect on the solubility of carbides. Figure 4.24 shows relevant data on V, Nb, Ti, and Zr carbides. An increase in the temperature of carbide solubility in austenite with a rise in the carbon content is due to the greater activity of carbon at its higher concentrations in a solid solution and higher thermodynamic activity. It should be noted that C, N, and Al are not bound to carbides or nitrides, but found in the solid solution of austenite facilitate the growth of 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 tendency toward the growth of austenite grains because they contain different amounts of low-soluble disperse particles of carbides, nitrides, and other phases, which are barriers to the growth of austenite grains. The distribution and size of these particles depend both on steelmaking conditions and preliminary heat treatment. Thus, the tendency of steel to grain size growth under heating depends on, in addition to its composition, the metallurgical quality and process, 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 quenching temperature for a sufficiently long time. This refers to cat steel where liquation is the highest and also to forged and rolled steels. Longer quenching times increase hardenability owing to the elimination of residual liquation and fluctuation in homogeneity of austenite. Note in conclusion that hardening capacity and hardenability are not important by themselves in practical applications. They are important if they can improve overall properties of steels in accordance with practical needs. 4.4 4.4.1 TEMPERING OF ALLOY STEELS STRUCTURAL CHANGES ON TEMPERING Structural changes on tempering were considered in detail in Chapter 3. Therefore, this section briefly considers only the most characteristics effects. Tempering is a thermal martensitic treatment of quenched steels. The basic process that takes place during tempering is martensite precipitation. The first structural change during ß 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 above 1008C (2128F). During the third stage, cementite precipitates above ~2508C (~4808F). At the final stage of tempering above 3508C (6608F), cementite particles coagulate and spheroidize. Consider the changes that take place in martensite and austenite structures at different stages 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 to unity. Tetragonal martensite transforming to a cubic form is called tempered martensite. The decrease of tetragonality is connected with precipitation of carbon from the solution. Heating of steels over 2008C (3908F) and up to 3008C (5708F) activates transformation of retained austenite to a heterogeneous mixture composed of a supersaturated a-solution and the Fe3C carbide. This means that retained austenite transforms to tempered martensite, Feg (C) ! Fea (C) þ Fe3 C By the end of transformation (~3008C; 5708F), the retained austenite contains about 0.15–0.20% C. Heating above 3008C (5708F) leads to a further precipitation of carbon and the relaxation of internal stresses arising from previous transformations. Complete precipitation of carbon was found to occur at 4008C (7508F). A further increase in temperature leads only to coagulation of ferrite and cementite particles. During tempering, cementite acquires a globular form when a ferrite–cementite mixture is formed from martensite. The different form of cementite in the ferrite–cementite mixture determines the difference in properties. 4.4.2 EFFECT OF ALLOYING ELEMENTS The influence of alloying elements on transformations during tempering depends on whether they dissolve in ferrite and cementite or form special carbides. The diffusion mobility of atoms of alloying elements dissolved in a-Fe by the substitutional method is many orders of magnitude lower than the diffusion mobility of carbon atoms dissolved by the interstitial method. So at a temperature below 4008C (7508F) no diffusion redistribution of alloying elements in the matrix takes place. First the e-carbide and then cementite precipitate from the a-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) at a temperature below 1508C (3008F) is affected little by alloying elements. At this stage of tempering, nucleation of carbide particles depends basically on supersaturation of the a-solution with carbon. The second stage of martensite precipitation is strongly influenced by a number of alloying 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 is retained up to temperatures of 450–5008C (840–9308F). Additions of Cr, W, Mo, V, Co, and Si bring about this effect. A delay in martensite precipitation can be explained by two factors. First, one of the alloying elements lowers the rate of carbon diffusion in the a-solution. Second, the other elements can increase the strength of interatomic bonds in the a-solution lattice. This will prevent the atoms from crossing the a-solution carbide interface. Both factors impede precipitation of martensite. ß 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) is formed first, then Me23C6 appears, and finally Me6C emerges. The sequence of their formation can be written as Fe3 C ! Me2 C þ Me23 C6 ! Me6 C Alloying elements affect the coagulation rate of carbide particles. Nickel accelerates the coagulation rate while chromium, molybdenum, vanadium, and other elements slow it down. Owing to the low diffusion rate of alloying elements, the coagulation of special carbides proceeds slowly. Even alloyed cementite (Fe, Cr)C3 coagulates much more slower than Fe3C in a carbon steel. Additions of alloying elements slow down recrystallization and polygonization. Atoms of these elements form impurity atmospheres near dislocations and prevent their movement during polygonization. Disperse particles of special carbides retard movement of large-angle boundaries during polygonization. 4.4.3 TRANSFORMATIONS OF RETAINED 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 the point Ms, thus decreasing the amount of retained austenite. Others (e.g., silicon) have no influence on Ms. However, the majority of elements decrease the martensite point and increase the amount of retained austenite in quenched steel. Up to 60% of retained austenite is left in high-carbon steels during quenching and 10–15% in a large number of structural alloy steels. During tempering of carbon and low-alloy steels, retained austenite transforms over the temperature interval of 230–2808C (440–5408F) or at lower temperatures if the holding time is extended. Alloying elements, especially Cr and Si, inhibit that transformation, shifting it to higher temperatures and longer tempering time. The transformation kinetics of retained austenite during tempering is similar to those of undercooled austenite. Steels with two clearly distinguished transformation ranges (pearlite and bainite) also exhibit two regions of fast transformation of retained austenite during tempering that are separated by a zone of high stability of retained austenite. When alloy steels are tempered at 500–6008C (930–11108F), in many cases the transformation of retained austenite is not complete. The retained austenite that did not precipitate at these tempering temperatures transforms during cooling from those temperatures (secondary quenching). The phenomenon is most pronounced in high-speed and high-chromium steels. The secondary martensite transformation during cooling after tempering is caused by the depletion of austenite in carbon and alloy elements in the course of tempering. As a result, the temperature of retained austenite Ms during cooling is increased. Secondary quenching (double tempering) is also observed in structural steels. It takes place if the primary quenching is accompanied by a partial intermediate transformation leading to an increase in the carbon content of austenite. During tempering at 500–5508C (930–10208F), retained austenite with a high content of carbon yields carbides intensively, and the martensite transformation temperature Ms increases. As a result, the secondary martensite transformation takes place during cooling after tempering. In high-alloy steels, for example high-speed steels, even a very long tempering at high temperatures does not completely eliminate the retained austenite. To obtain a full ß 2006 by Taylor & Francis Group, LLC. transformation, it is necessary to perform double tempering. Double tempering favors additional precipitation of special carbides and decreases the degree of austenite alloying. This causes another increase in the transformation temperature Ms. Sometimes multiple tempering is required to realize the most complete transformation of retained austenite. 4.4.4 TIME–TEMPERATURE RELATIONSHIPS IN TEMPERING The kinetics of structural transformations during tempering is described by temperature–time curves 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 the curve and diagram, the nontransformed austenite (30%) was taken to be 100%. It is found that at 6008C (11108F) 5% of primary austenite transforms in 7 min and 5% of retained austenite transforms in 30 s. However, after 10–15% 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 precipitate immediately at these tempering temperatures transforms during cooling after tempering. The transformation rate of retained austenite in the intermediate range is much higher than that of the initial austenite; 25% of initial austenite transforms at 3008C (5708F) in 75 min and the same amount of retained austenite in 15 s; 75% of the initial austenite transforms in 220 min and 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 at 4008C (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 decreases with increasing temperatures. 4.4.5 ESTIMATION OF HARDNESS AFTER TEMPERING Hardness decreases noticeably when alloy steels and addition-free steels are subjected to tempering at 500–6008C (930–11108F). This decrease is due to the precipitation of martensite and coagulation of cementite. However, when the tempering temperature is higher, the 1 2 600 25% 500 5 75 25 5 25 75 t, 8C 400 300 25 5 75 200 100 1 23 5 10 20 30 50 1 235 s m Time FIGURE 4.25 Time–temperature transformation curves. ß 2006 by Taylor & Francis Group, LLC. 10 20 30 1 23 h 5 10 hardness 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 elements and carbon (a maximum hardness often corresponds to the clusters) and the replacement of relatively coarse particles of cementite by much more disperse precipitates of special carbides (TiC, VC, Mo2C, W2C). When these particles coagulate, hardness decreases. Particles of Me6C are rather coarse and do not add to strengthening. The chromium additive causes a small secondary hardening. This is connected with a rapid coagulation of the Cr7C3 carbide at 5508C (10208F) as opposed to Mo2C and especially W2C. During secondary hardening an increase in the yield stress is accompanied by an increase in toughness owing to dissolution of coarse cementite particles. 4.4.6 EFFECT OF TEMPERING ON MECHANICAL PROPERTIES The manner in which structural changes that take place during tempering affect the properties of steels depends on the particular tempering conditions. The general tendency of changes in mechanical properties of carbon steels during tempering is that as the tempering temperature is 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) are improved. However, these properties change nonmonotonically, and the variation depends on the tempering temperature intervals. Low-temperature tempering (120–2508C; 250–4808F) is used for treatment of highstrength structural and tool steels. Medium-temperature tempering (350–4508C; 660–8408F) is applied mainly to spring steels to achieve high elasticity. High-temperature tempering (450– 6508C; 840–12008F) is widely used for products made of structural steels combining a relatively 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 is possible to obtain a very favorable combination of strength (sB, s0.2), ductility (d, c), and impact strength under low-temperature tempering (160–2008C; 320–3908F). In low-carbon martensitic steels containing chromium, manganese, nickel, and molybdenum, the tensile strength remains unchanged up to 400–5008C (750–9308F). 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 impact strength increase under high-temperature tempering. These data suggest that mechanical properties of every type of steel exhibit certain specific features that vary with the tempering temperature. These features are determined by the influence of alloying elements on the kinetics of phase transformations: change of the martensite point Ms, stabilization of retained austenite, and carbide formation. The influence of structural evolution on properties during tempering can be most fully understood through the example of a maraging (martensite-aging) alloy containing 0.02– 0.03% C and also Co, Mo, Ti, and Al. Alloying with cobalt increases the temperature Ms and provides 100% martensite after cooling. Tensile strength reaches 1000–1100 MPa. The subsequent tempering (aging) at temperatures of 450–5008C (840–9308F) results in considerable strengthening. Thus, sB can reach 1900–2100 MPa, s0.2 ¼ 1800–2000 MPa, and d ¼ 8–10%. 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 the matrix. The size of particles is approximately 100 nm. Coagulation of the precipitates with an increase in temperature leads to lowering of the strength characteristics and increasing of the ductility. ß 2006 by Taylor & Francis Group, LLC. 4.4.7 EMBRITTLEMENT DURING TEMPERING When carbon and alloy steels are tempered over the temperature interval of 250–4008C (480– 7508F), a dramatic drop in impact strength is observed. If the steel is subjected to a higher temperature tempering and then tempering is repeated at 250–4008C, the brittle state is not recovered. Therefore this phenomenon has been called irreversible tempering brittleness. Such tempering 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 it does not occur at all. The embrittlement may be caused by nonuniform precipitation of martensite at the second stage of tempering. This structure has a lower resistance to dynamic loads. This effect is enhanced when the initial grain boundaries of austenite get saturated with impurities under quenching heating. The alloying elements, which retard the second stage of martensite precipitation, shift the interval of irreversible brittleness toward higher temperatures. Another drop in impact strength is found at tempering temperatures of 450–6008C (840– 11108F). A very significant feature of embrittlement is its reversibility under high-temperature tempering. If a steel that has undergone tempering embrittlement is heated to a temperature above 6008C (11108F) and then cooled rapidly, its impact strength is restored. Therefore such brittleness is termed reversible. In the state of reversible tempering embrittlement, steel possesses a structure that consists of ferrite and carbide. When subjected to impact tests, fracture occurs mainly along the boundaries of the initial austenite grains. Embrittlement over a certain temperature interval is typical not only of martensitically hardened steels. It also shows up, although to a lesser degree, in steels with the bainite structure 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 0.2–0.3%) weaken tempering embrittlement. The presence of Sb, P, Sn, and As in industrial steels makes these steels most susceptible to tempering embrittlement. 4.5 HEAT TREATMENT OF SPECIAL CATEGORY STEELS 4.5.1 HIGH-STRENGTH STEELS Low-alloy steels are most often used as construction materials. The combination of high strength 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 normalization and after quenching with tempering. Alloying makes it possible to perfect the properties of steels without using quenching with tempering because 1. 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 or normalization. 3. Steel grains and microstructure components become finer, and changes occur in the morphology 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 into three groups: 1. Low-carbon steels (0.11–0.22% C) used in the hot-rolled or normalized states. Thermal treatment of such steels (quenching and tempering) only slightly improves their strength characteristics. ß 2006 by Taylor & Francis Group, LLC. 2. Low-carbon steels (0.05–0.18% C) strengthened by disperse precipitation of carbides and carbonitrides are used in the normalized or hot-rolled states. 3. Medium-carbon steels (0.25–0.50% C). The required level of properties is achieved in such steels by quenching and high tempering. The main alloying element in these steels is manganese. Additional alloying of manganese steels with Mo, Nb, and V results in formation of needle ferrite. Owing to their needle structure, these steels combine high strength, high viscosity, and cold resistance. Therefore, the steels with 0.05% C, 2% Mn, 0.4% Mo, 0.01–0.02% Nb, or 0.07% V exhibit the following strength characteristics: sB ¼ 650–750 MPa, s0.2 ¼ 630 MPa, d ¼ 33%. These steels are more often used in the normalized state and seldom in the hot-rolled state. Low-alloy steels with improved strength characteristics include steels containing mainly manganese and silicon. Tensile strength sB in these steels is more than 600 MPa and can be as high as 1800 MPa. Their ductility and viscosity depend on their carbon content and on the types of treatment. High-strength low-alloy steels with sB ¼ 1800 MPa are used in the hotrolled or cold-worked state. However, in this case, they are characterized by low impact strength. In the normalized state their strength decreases to 800 MPa, and after quenching and tempering to 600 MPa, the impact strength increasing simultaneously. 4.5.2 BORON STEELS Alloying 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) and tempering at 8008C (14708F). The strength and yield limits the increase simultaneously. At the same time the viscosity of such steels decreases more than in the usual austenite steels. The precipitation of borides, because of the high temperature of disperse hardening (8008C; 14708F), has a beneficial effect on the properties of refractory alloys (chromium, chromium–nickel, chromium–nickel–cobalt alloys). At test temperatures up to 7008C (12908F) but still below the temperature of boride precipitation, the refractory characteristics of steels appreciably improve even at a very lowboron content. In low-carbon steels with 16–30% Cr and 6.5–30% Ni, the effect of disperse hardening associated with the presence of boron was not observed. But boron binds the elements stabilizing austenite, thus favoring the formation of martensite. Carbon steels with 0.2–0.3% C, on the other hand, are hardened considerably owing to the formation of boride–carbide precipitates. Addition of boron to cemented steels improves their hardenability and increases the strength of the core. Boron somewhat accelerates carburizing, but its influence on the case lessens with increasing carbon content. The greatest effect of boron was observed in steels with 0.7–0.8% C. The influence of boron is enhanced as the quenching temperature is raised. However, the sensitivity of steel to overheating also increases. Therefore, boron steels usually contain small quantities of titanium and vanadium, which have a favorable effect on the structure of steels when they are heated to high temperatures. 4.5.3 ULTRAHIGH-STRENGTH STEELS Low-carbon steels with a martensite structure have been developed recently that, upon cooling in air, undergo subsequent dispersion hardening at 400–5008C (750–9308F). The tensile strength of such steels is in the range 2200–2500 MPa. As a rule, they are alloyed with 12–18% Ni, up to 10% Cr, 3–5% Mo, and 0.6–1.0% Ti. These martensite-aging steels are distinguished ß 2006 by Taylor & Francis Group, LLC. by 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 of substitutional solid solution in the course of alloying, strengthening brought about by the martensite g ! a transformation, and strengthening connected with different stages of the solid solution precipitation accompanied by formation of segregates and disperse particles of metastable 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 fine substructure with a high density of dislocations is formed. The particles of intermetallics ˚ 50–100 A in size are found at the stage of maximum strengthening. These particles are coherently connected with the matrix. The high resistance of martensite-aging alloys to brittle fracture is determined by the high viscosity of the matrix—the low-carbon martensite alloyed with Ni and Co, which enhance the mobility of dislocations. In addition, a high density of dislocations in martensite is responsible for high dispersion and homogeneous distribution of phases precipitated during aging. Alloying with Mo suppresses the precipitation of particles of strengthening phases at the grain boundaries and prevents intergrain brittle fracture. Alloying with Co increases the martensite start temperature Ms and ensures a 100% martensite structure. At the same time, alloying with cobalt reduces the solubility of molybdenum and fosters dispersion hardening. At low-carbon content and moderate cooling rates, the martensite structure in martensiteaging alloys is obtained through relatively high degrees of alloying. At 10–18% Ni, the point Ms lowers so significantly that the g transformation can be realized only according to the martensite mechanism. Compared with the high-strength manganese steels considered in Section 4.5.1, the martensite-aging steels are distinguished by a greater degree of alloying of the g-solid solution. This promotes almost complete transformation of austenite to martensite. A wide range of alloying elements ensures a stronger solid-solution strengthening and increases the volume fraction of disperse particles precipitated during phase aging. The above-mentioned three 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 affect their structure. The most advantageous of these is low-temperature thermomechanical treatment (LTMT), which consists of deformation of the undercooled austenite in the region of high stability and subsequent quenching. Undercooling of austenite is used to achieve deformation below the temperature of its recrystallization. Such treatment allows the attainment of advanced mechanical properties. The results gained at LTMT can be achieved by such factors as the composition of the steel, the temperature of austenitization, the rate of cooling to the deformation temperature, the temperature of deformation and holding time at this 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 of the bainite transformation. In conformity with this, the steels undergoing LTMT should contain austenite-stabilizing elements. LTMT strengthening is usually employed for highalloy steels with 1–7% Cr, 1–5% Ni, 0.5% V, 2.5% Mo, and 2% Si and sometimes with other additions as well. The strengthening of steels in LTMT depends on their carbon content. The strengthening effect of carbon is more pronounced in LTMT than in conventional quenching. ß 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 is increased by 7 + 2 MPa. A decrease in deformation temperature results in a more intensive strengthening of steel in LTMT. The strength characteristics of steel after a small (up to 30%) deformation are lower and less sensitive to changes in the deformation temperature than at high degrees of compression. At deformations of up to 20–30%, LTMT leads to a sharp drop in ductility; with a rise in the degree of compression above this value, the ductility begins to increase. There is a critical degree of deformation in LTMT above which the ductility of steel is sufficient. As the temperature of deformation increases, the ductility also increases. In LTMT, the steel should be tempered as after the usual quenching. It is the opinion of the majority of researchers that the strengthening effect of LTMT is retained up to 350–4008C (660–7508F). If steel is alloyed with Mo, V, or W, the strengthening effect of LTMT persists up to 5008C (9308F). The tensile strength of steel alloyed with tungsten is 2600 MPa after quenching at 3508C (6608F) and 2450 MPa after quenching at 5008C (9308F). The rate of cooling 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 treatment has 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 martensite crystals and changes in their morphology. This can be attributed to the emergence of a cellular structure during deformation of undercooled austenite. The sites of dislocation pileups in austenite remain the sites where dislocations accumulate in martensite after the transformation. Upon LTMT deformation, the fragmentation of austenite crystals results in the fragmentation of the martensite structure. Individual fragments measuring fractions of a micrometer mutually disoriented through 10–158 are joined with each other by dense disloca˚ tion pileups. These fragments, in turn, consist of 100–200 A fragments disoriented relative to each other through angles greater than 18. Thus, one of the possible mechanisms of strengthening in LTMT is connected with the creation of a high density of structural imperfections in austenite as a result of deformation and the inheritance by martensite of the dislocation structure of the work-hardened austenite. This mechanism provides the most comprehensive explanation for the high strength of martensite obtained with LTMT. 4.5.4 MARTENSITIC STAINLESS STEELS Pure iron and low-alloy steels are not resistant to corrosion in the atmosphere, water, or many other media. The resistance of steel to corrosion can be enhanced by alloying it with various elements. High strength of such steels is achieved primarily by quenching to obtain the martensite structure and through its subsequent aging. In martensitic stainless steels, the amount of martensite necessary for strengthening is formed after high-temperature heating and subsequent cooling to room temperature at a relatively small content of alloying components. The majority of alloying additions improve the resistance of martensite by lowering the point Ms. The possibilities for anticorrosion alloying of martensitic steels are limited. In austenitic–martensitic steels (transition class), quenching does not lead to the complete transformation 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 ß 2006 by Taylor & Francis Group, LLC. temperatures below Mg; (2) plastic deformation below Ms; and (3) heating in the region of the most intensive precipitation of alloyed carbides from austenite (700–7508C; 1290–13808F); when the matrix is depleted in alloying elements, the resistance of austenite decreases. The austenitic–martensitic steels admit a high degree of alloying and therefore afford more possibilities for achieving total corrosion resistance and high strength. Such alloying elements 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 another element depends on their combination. Cold plastic deformation initiates the martensite transformation. The less stable the austenite and the lower the deformation temperature, the quicker the transformation. The strength of austenitic–martensitic and martensitic stainless steels increases with decreasing deformation temperature. If after treatment these steels contain 70–90% martensite, their yield stress can amount to 700–1000 MPa and their tensile strength to 1100–1400 MPa. Further improvement of strength is achieved by aging the martensite. This enhancement of strength is attributed to segregation of the Guinier–Preston 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 quenching proceeds below the temperature Mc1 under rather high compression. On the one hand, this accelerates the martensite transformation, and on the other hand, aging takes place in the martensite strengthened by deformation (sometimes also in the presence of the deformationstrengthened austenite). After complete thermal treatment, steels have the following characteristics: s0.2 ¼ 830–1200 MPa, sB ¼ 1200–1300 MPa. The conditions of quenching are set with allowance for complete dissolution of carbides subject to the absence of excessive grain growth. Deep freezing treatment after quenching ensures a more complete transformation of austenite to martensite. The amount of martensite can be as high as 70–90%. The conditions of aging should provide the required set of mechanical properties and corrosion resistance. The maximum strength values are attained as a result of aging in the temperature range 450–5008C (840–9308F). At the same time, the best corrosion resistance is attained at the lower aging temperature range of 350–2808C (660–5408F) (high total corrosion resistance is obtained at the stage preceding precipitation of strengthening phases). 4.5.5 PRECIPITATION-HARDENING STEELS As is known, steels are classified into structural, spring, tool, and heat-resistant alloy steels in conformity with their application. This section considers the behavior of precipitationhardening alloys in each of these groups. 4.5.5.1 Structural Steels Low-carbon manganese steels (0.1–0.2% C) containing 1.3–1.7% Mn, 0.10–0.20% V, about 0.1% 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 the formation of disperse precipitates of vanadium and titanium carbonitrides or aluminum nitrides. 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 ß 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 strengthening phases 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 (~0.15–0.3%). Aluminum (up to 0.05%) can also be present in these steels. The properties of the steels under consideration are formed in the process of rolling during precipitation of disperse particles of the strengthening phase and grain refinement. The conditions of rolling should ensure maximum dissolution of components that subsequently cause the formation of disperse particles. These particles strengthen ferrite, which leads to grain refinement. The rolling temperature of low-pearlite steels depends on their composition and the strength and viscosity requirements. A high-heating temperature (~12008C; 21908F) ensures more complete dissolution of vanadium and niobium. This contributes 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 case of heating at 1050–11008C (1920–20108F). The temperature at the end of rolling of low-pearlite steels is usually reduced to 700–8008C (1300–14758F). This is due to (1) a decrease in the austenite grain size; (2) an increase in the degree of dispersion of the strengthening phase and, hence, enhancement of the hardening effect; and (3) displacement of the g ! a transformation to the region of lower temperatures, which results in a finer ferrite grain. 4.5.5.2 Spring Steels Alloys based on Fe–Ni, Fe–Ni–Cr, Co–Ni–Cr, Ni–Cr, and other systems, predominantly with titanium and aluminum or niobium additions, are used for spring steels strengthened by precipitation hardening. The particles of strengthening phases in these alloys precipitate during aging (tempering). Additional improvement of the strength properties of these alloys can be achieved through plastic deformation between quenching and aging. In this case the precipitation of the supersaturated solid solution may proceed according to a discontinuous mechanism. If the discontinuous precipitation cells completely occupy each grain (which is possible for a very fine grain structure), a very strong strengthening of alloys takes place. In the process of aging, additional refinement of the initial grain occurs during discontinuous precipitation. Strengthening is observed in alloyed martensite-aging steels under developing of disperse particles of precipitating phases. A large number of steels differing in composition and properties are used in the industry. In addition to 0.4–0.8% C, they contain at least two of such alloying elements as Si, Cr, V, Mo, Mn, more rarely Ni and W. Isothermal quenching with subsequent tempering is advantageous for these steels, especially for those containing silicon. The maximum elastic limit in alloyed steels is attained with tempering at 300–3508C (570– 6608F). These tempering conditions correspond to the conditions of a sufficiently complete precipitation of austenite accompanied by preservation of a high density of dislocations, since disperse particles of carbides hammer the redistribution and annihilation of dislocations. In addition, 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 be lowered owing to the higher carbon content. This brings about a significant improvement of the strength characteristics of such steels, the highest properties achieved when strong carbide-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 (1600–21008F) depending on their titanium ß 2006 by Taylor & Francis Group, LLC. content; the greater the titanium content, the higher the quenching temperature. The finer the grain, the higher the properties of these steels. Fine grain can be attained either by multiple g ! a transformations or through deformation and recrystallization processes. To reduce the quantity of austenite retained upon quenching, a deep freezing treatment (À708C; À948F) is employed. Then comes aging at 4508C (8408F) for 6 h, during which NiTi or Ni3Ti phases precipitate. Austenitic steels are also strengthened as a result of precipitation hardening. They may contain chromium, nickel, titanium, or molybdenum. Upon quenching, these alloys have the structure of the g-solid solution with chromium, titanium, and titanium carbonitride inclusions. The properties of aged alloys depend on the quenching temperature, which determines the degree of supersaturation of the solid solution, and on the cooling rate, which should be as high as possible. Aging may proceed by the discontinuous or continuous mechanism. 4.5.5.3 Tool Steels Tool steels strengthened by precipitation-hardening tempering on the basis of the initial martensite structure are used to manufacture dies for cold deformation of steels. As a result of tempering, the hardness and strength characteristics of steels are enhanced when strengthening phases (carbides) precipitate from martensite. The retained austenite, a phase with low hardness, transforms to martensite. These processes increase the yield stress under compression but reduce viscosity. Precipitation-hardening strengthening is also characteristic of heat-resistant steels. The structure of these steels represents a martensite matrix with particles of the strengthening phases—carbides or intermetallics—precipitating during tempering. The main principles of heat treatment of precipitation-hardening tool steels are now considered 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 760–7808C (1400–14358F) for carbon steels, 780–8108C (1435–14908F) for alloy steels, and 830–8708C (1525–16008F) for high-alloy chromium steels, with 2–3 h as the holding time . Quenching of tool steels is aimed at obtaining martensite with a high concentration of carbon and alloying components with retained fine-grain structure. That is why quenching is carried out at temperatures corresponding to complete dissolution of the basic carbides in austenite. These temperatures, however, should not be conducive to austenite grain growth. Usually, the quenching temperature corresponds to the temperature of heating; it is a little higher than A1 for steels in which the main carbide phase is cementite; up to 1000– 10608C (1832–19408F) for steels with a chromium-based carbide phase of types Me7C3 and Me23C6, and 1080–11008C (1975–20108F) for steels with greater carbide content of the type Me23C6. At tempering, the assigned level of properties is achieved by changing the structure of the quenched steel. Heating a quenched steel during tempering to 150–2008C (300–3908F) causes the precipitation of small e-carbide plates from martensite and reduces the carbon concentration. Such tempering only slightly impairs the steel’s hardness but significantly improves its strength and viscosity. Heating to 250–2808C (480–5358F) during tempering noticeably decreases the carbon concentration in martensite and enhances the strength and viscosity characteristics of the steel. This tempering permits almost complete removal of the retained austenite. An appreciable increase in the hardness of steels results from the precipitation of a large number of small carbide particles (intermetallics of alloying elements of the types Me2C, Me23C6, MeC, and Me7Me6) from martensite. ß 2006 by Taylor & Francis Group, LLC. 4.5.5.4 Heat-Resistant Alloys A vast group of heat-resistant alloys consists of austenitic steels strengthened with carbides and intermetallics. To increase heat resistance, elements that strengthen the solid solution and induce precipitation hardening are introduced into iron–nickel-based alloys. These elements include Cr, Mo, W, Nb, V, Ti, and Al. To acquire high heat resistance, the alloys undergo double quenching. The purpose of the first quenching is to obtain grains of a certain size and to 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 obtain ˚ disperse precipitates of the g0 -phase 200–500 A in diameter. As a result of these quenching procedures, the strengthened alloy contains a certain number of larger precipitates along with fine inclusions. This structure ensures high strength and the necessary margin of ductility. 4.5.6 TRANSFORMATION-INDUCED PLASTICITY STEELS There are numerous examples for improving the plasticity of load-bearing samples under the influence of phase transformations of the diffusion and shear types. The term transformationinduced plasticity was proposed to denote an improvement of plasticity under martensite transformation. 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 stresses that exceed the yield stress of austenite. In the segment where localization of flow sets in, martensite deformation occurs. This segment is stronger than austenite, and because of this the flow extends to the neighboring segments of the sample. Thus, the quasiequilibrium flow in TRIP steels is due to high 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, the volume that undergoes such a transformation reaches a certain level and does not increase further. Therefore, the greatest overall effects of plasticity improvement are observed at cyclic temperature changes that lead to multiple occurrences of the phase transformation. TRIP is found when the temperature of the sample is changed (within limits exceeding the temperature range of the phase transformation), with a constant load. The load applied is usually 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 deformation value in one cycle is directly proportional to the applied stress. An increase in the applied stress 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 at comparatively high stresses. As the volumetric effect of phase transformation increases, the deformation 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 204–6488C (400–12008F), a sample of Fe–15.4% Ni alloy with an initial grain size of 150 mm became 160% longer, with no neck formed, owing to the reversible martensite transformation under load. The deformation mechanisms typical of TRIP have not been clearly established because of difficulties in using direct structural methods during phase transformation when the structure of 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 during volumetric changes ß 2006 by Taylor & Francis Group, LLC. 2. Weakening of bonding forces between atoms at the interface at the moment of transformation 3. Changes in the form associated with realization of particular orientations of the formed martensite 4. Summation over the phase and applied stress, which determines the plastic deformation of the weaker phase (ferrite in the case of a!g transformation) 5. Formation of ultrafine grain in the course of phase transformation 4.5.7 TOOL STEELS Tool steels can be classified into four groups according to their application: (1) steels for cutting 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 tool steels contain a minimum of 0.6% C. The main requirement imposed on steels used in severe conditions (high-speed steels) is stable hardness under long heating. All tool steels fall into four categories: (1) carbon tool steels, (2) alloy tool steels, (3) die steels, and (4) high-speed steels. 4.5.7.1 Carbon Tool Steels Carbon tool steels contain 0.60–0.74% C, 0.25–0.35% Mn, and 0.30% Si. The quenching temperature of these steels is chosen in conformity with the Fe–C equilibrium diagram. The tetrahedral structure of martensite and internal stresses in quenched steels bring about considerable 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. Steels with the structure of lamellar pearlite are difficult to machine. Therefore, with the help of annealing at a temperature slightly above Ac1, easily worked steels with the structure of globular 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 of through-hardened steels. 4.5.7.2 Alloy Tool Steels Compared 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 by subsequent tempering. If it is necessary to preserve hardness at the level of the quenched state, the tempering temperature should not exceed 150–1708C (300–3408F). In all cases where quenching should be accompanied by minimum deformation during the pearlite ! martensite transformation (pearlite is the initial structure in this process), lowdeformation tool steels are used. Such steels can be obtained by alloying with elements that increase the amount of retained austenite in the quenched state, namely, chromium and manganese. These steels contain about 12% Cr and ~1.5% C. The formation of a large amount of carbides (Cr,Fe)7C3 significantly improves their wear resistance. These high-chromium steels belong to the ledeburitic class. In the cast state, the initial carbides form the eutectic ledeburite. In forging, the eutectic breaks down and the structure of the steels consists of sorbite-forming pearlite with inclusions of excess carbides. When heated for quenching, the carbides dissolve in austenite. The highest hardness of the steel is achieved upon quenching at ~10508C (~19208F). ß 2006 by Taylor & Francis Group, LLC. To obtain high hardness, the steel is quenched in oil. The retained austenite precipitates in the process of cold treatment and tempering. Owing to the greater stability of martensite compared to other steels, the tempering temperature is increased to 200–2208C (390–4308F). 4.5.7.3 Die Steels Dies operating in the cold state need high-hardness steels. The steel to be used in hot pressing should have low sensitivity to local heating. Different grades of steels—from carbon to complex alloy steels—are used in the production of dies. Carbon steel is used for dies operating under mild conditions and alloy steel for dies operating under severe conditions. Carbon steel contains 0.6–1% C. Alloy steel includes 0.3–0.7% C, Cr, Si, and sometimes Ni. Dies made of alloy steels are usually quenched from ~8508C (15608F) in oil with subsequent tempering at 500–5508C (930–10208F). Hardness of the steel amounts to 350–400HB. Dies for cold pressing are quenched in the temperature range of 860–10508C (1580–19208F) (depending on the steel grade) in oil with subsequent tempering at 200–3008C (390–5708F). Depending on the tempering temperature and the steel grade, the steel hardness is within Rc ¼ 56–62. 4.5.7.4 High-Speed Steels High-speed steels must not only possess high hardness in the hot state, but also be able to retain it during long heating (red hardness). To preserve hardness during heating, it is necessary to hamper the process of carbide coagulation. For this purpose, special carbides should be formed. Such carbides can be produced if the steel is alloyed with 3% Cr. The special 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 at temperatures over 5008C (9308F). All high-speed steels are rated in the ledeburite class and in the cast state have the structure 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 annealed state, three types of carbides are observed: coarse primary carbides, smaller secondary carbides, 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 (800–8508C; 1470–15608F) is not accompanied by structural changes. Above this point, the eutectoid transforms to austenite, the secondary carbides 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 carbides in austenite. Carbon and alloying elements contained in austenite lower the martensite point and increase the content of retained austenite. At quenching temperatures above 10008C (18328F) the martensite point decreases to 08C (328F) or lower. This peculiarity is taken advantage 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 to 100–2008C (212–3908F) causes a small compression, since the tetragonal martensite transforms to the cubic modification. At 300–4008C (570–7508F), hardness deteriorates owing to a decrease in the work hardening of retained austenite. At 500–6008C (930–11108F), finely disperse carbides precipitate from austenite. Cooling the steel from these temperatures brings about the secondary formation of martensite: depleted austenite transforms to martensite in larger quantities. ß 2006 by Taylor & Francis Group, LLC. The higher the tempering temperature or the longer the tempering time, the greater the amount of retained austenite transformed to martensite. Complete transformation of austenite can be attained by multiple tempering. The microstructure of the quenched and tempered steel should consist of finely dispersed martensite and carbides. The quenching temperature of high-speed steel should be as high as possible, but at the same time it should not allow intensive grain growth (1260–12808C; 2300–23408F). During quenching the steel may be cooled comparatively slowly owing to a low-critical rate of quenching (in air or oil). Tempering is an obligatory operation and is usually realized at 560–5808C (1040–10758F) 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 READING Belous, 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. ß 2006 by Taylor & Francis Group, LLC. 5 Hardenability ˇc ´ ˇidar Lis ˇic Boz CONTENTS 5.1 5.2 5.3 Definition of Hardenability ....................................................................................... 213 Factors Influencing Depth of Hardening................................................................... 215 Determination of Hardenability................................................................................. 217 5.3.1 Grossmann’s Hardenability Concept.............................................................. 217 5.3.1.1 Hardenability in High-Carbon Steels.............................................. 220 5.3.2 Jominy End-Quench Hardenability Test ........................................................ 228 5.3.2.1 Hardenability Test Methods for Shallow-Hardening Steels ........... 230 5.3.2.2 Hardenability Test Methods for Air-Hardening Steels................... 233 5.3.3 Hardenability Bands ....................................................................................... 237 5.4 Calculation of Jominy Curves from Chemical Composition ..................................... 240 5.4.1 Hyperbolic Secant Method for Predicting Jominy Hardenability .................. 243 5.4.2 Computer Calculation of Jominy Hardenability ............................................ 247 5.5 Application of Hardenability Concept for Prediction of Hardness after Quenching..... 249 5.5.1 Lamont Method ............................................................................................. 253 5.5.2 Steel Selection Based on Hardenability .......................................................... 256 5.5.3 Computer-Aided Steel Selection Based on Hardenability .............................. 257 5.6 Hardenability in Heat Treatment Practice ................................................................. 264 5.6.1 Hardenability of Carburized Steels................................................................. 264 5.6.2 Hardenability of Surface Layers When Short-Time Heating Methods Are Used......................................................................................................... 266 5.6.3 Effect of Delayed Quenching on the Hardness Distribution .......................... 267 5.6.4 A Computer-Aided Method to Predict the Hardness Distribution after Quenching Based on Jominy Hardenability Curves ....................................... 268 5.6.4.1 Selection of Optimum Quenching Conditions ................................ 273 References .......................................................................................................................... 275 5.1 DEFINITION OF HARDENABILITY Hardenability, in general, is defined as the ability of a ferrous material to acquire hardness after austenitization and quenching. This general definition comprises two subdefinitions: the ¨ ability to reach a certain hardness level (German: Aufhartbarkeit) and the hardness distribu¨ tion within a cross section (German: Einhartbarkeit). The ability to reach a certain hardness level is associated with the highest attainable hardness. It depends first of all on the carbon content of the material and more specifically on the amount of carbon dissolved in the austenite after the austenitizing treatment, because only this amount of carbon takes part in the austenite-to-martensite transformation and has relevant influence on the hardness of martensite. Figure 5.1 shows the approximate relationship between the hardness of the structure and its carbon content for different percentages of martensite [1]. ß 2006 by Taylor & Francis Group, LLC. 80 Martensite Hardness, HRC 60 99.9% 90% H max 50% 40 20 0 HRC99.9 = 35 + 50 . %C HRC90 = 30 + 50 . %C HRC50 = 23 + 50 . %C 0 0.2 0.4 0.6 0.8 10% carbon content FIGURE 5.1 Approximate relationship between hardness in HRC and carbon content for different ¨ percentages 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 hardness from the surface of a specified cross section toward the core after quenching under specified conditions. It depends on carbon content and the amount of alloying elements dissolved in the austenite during the austenitizing treatment. It may also be influenced by the austenite grain size. Figure 5.2 shows the hardness distributions within the cross sections of bars of 100 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 other two grades in oil, the W1 steel has the lowest hardenability because it does not contain alloying elements. The highest hardenability in this case is that of the D2 steel, which has the greatest amount of alloying elements. 70 AISI D2 Hardness, HRC 60 50 AISI 01 40 AISI W1 30 20 0 0 10 1/2 20 30 11/2 Depth below surface 1 40 50 mm 2 in. FIGURE 5.2 Hardness distributions within cross sections of bars of 100 mm diameter for three different kinds of steel, after quenching. Steel W1 was water-quenched; the rest were oil-quenched. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984, p. 145.) ß 2006 by Taylor & Francis Group, LLC. When a steel has high hardenability it achieves a high hardness throughout the entire heavy 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 in Figure 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 hardenability while at the same time achieving much lower hardness values across a cross section. This can be best compared by using Jominy hardenability curves (see Section 5.3.2). Hardenability is an 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 HARDENING Depth of hardening is usually defined as the distance below the surface at which a certain hardness level (e.g., 50 HRC) has been attained after quenching. Sometimes it is defined as the distance 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 hardening depends on the following factors: 1. Shape and size of the cross section 2. Hardenability of the material 3. Quenching conditions Quenching conditions include not only the specific quenchant with its inherent chemical and physical properties, but also important process parameters such as bath temperature and agitation rate. The cross-sectional shape has a remarkable influence on heat extraction during quenching and consequently on the resulting hardening depth. Bars of rectangular cross sections always achieve less depth of hardening than round bars of the same cross-sectional size. Figure 5.3 is a diagram that can be used to convert square and rectangular cross sections to equivalent circular cross sections. For example, a 38-mm square and a 25  100-mm rectangular cross section are each equivalent to a 40-mm diameter circular cross section; a 60  100-mm rectangular 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 the same quenching conditions is shown in Figure 5.4A. Steeper hardness decreases from surface to core and substantially lower core hardness values result from quenching a larger cross section. Figure 5.4B shows the influence of hardenability and quenching conditions by comparing an unalloyed (shallow-hardening) steel to an alloyed steel of high hardenability when each is quenched in (a) water or (b) oil. The critical cooling rate (ncrit) of the unalloyed steel is higher than the critical cooling rate of the alloyed steel. Only those points on the cross section that have been cooled at a higher cooling rate than ncrit could transform to martensite and attain high hardness. With unalloyed steel this can be achieved up to some depth only by quenching in water (curve a); oil quenching (curve b) provides essentially no hardness increase. With alloyed steel, quenching in water (because of the high cooling rate of water) produces a cooling 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 of hardening. Only the core region remains unchanged. ß 2006 by Taylor & Francis Group, LLC. in. mm 240 250 8 200 mm in. 210 200 180 250 10 240 190 180 170 7 160 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 16 160 150 6 Thickness mm f 240 230 220 220 140 140 5 130 120 110 100 120 4 100 90 80 80 3 70 60 60 50 2 40 40 30 1 20 16 20 0 0 20 1 40 60 2 80 3 100 4 120 5 140 160 6 180 200 7 8 220 240 260 280 9 10 11 9 8 7 6 Diameter 9 5 4 3 2 1 300 mm 12 in. Breadth FIGURE 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.) 40 20 (B) Unalloyed steel Alloyed steel ncrit aa bb ncrit 0 Surf. Core Hardness, HRC 10 mm 10 mm Martensite content 93% 90% 50% 60 Cooling rate Hardness, HRC (A) 60 40 Martensite content 99% 90% 50% a a b b 20 0 Surf. Core FIGURE 5.4 Influence of (A) cross-sectional size and (B) hardenability and quenching conditions on the depth of hardening. (a) Water quenching; (b) oil quenching, ncrit, critical cooling rate. (From ¨ G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, Warmebehandeln, Carl Hanser, Munich, 1987, p. 1012.) ß 2006 by Taylor & Francis Group, LLC. 5.3 DETERMINATION OF HARDENABILITY 5.3.1 GROSSMANN’S HARDENABILITY CONCEPT Grossmann’s method of testing hardenability [3] uses a number of cylindrical steel bars of different diameters hardened in a given quenching medium. After sectioning each bar at midlength and examining it metallographically, the bar that has 50% martensite at its center is selected, and the diameter of this bar is designated as the critical diameter (Dcrit). The hardness value corresponding to 50% martensite will be determined exactly at the center of the bar of Dcrit. Other bars with diameters smaller than Dcrit have more than 50% martensite in the center of the cross section and correspondingly higher hardness, while bars having diameters larger than Dcrit attain 50% martensite only up to a certain depth as shown in Figure 5.5. The critical diameter Dcrit is valid for the quenching medium in which the bars have been quenched. If one varies the quenching medium, a different critical diameter will be obtained for the same steel. To identify a quenching medium and its condition, Grossmann introduced the quenching intensity (severity) factor H. The H values for oil, water, and brine under various rates of agitation are given in Table 5.1[4]. From this table, the large influence of the agitation rate on the 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 a given steel that would produce 50% martensite at the center when quenched in a bath of quenching intensity H ¼ 1. Here, H ¼ 1 indicates a hypothetical quenching intensity that reduces 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 the conversion of any value of critical diameter Dcrit for a given H value to the corresponding value for the ideal critical diameter (DI) of the steel in question [2]. For example, after quenching in still water (H ¼ 1.0), a round bar constructed of steel A has a critical diameter (Dcrit) of 28 mm according to Figure 5.6. This corresponds to an ideal critical diameter (DI) of 48 mm. Another round bar, constructed of steel B, after quenching in oil (H ¼ 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 hardenability than steel A. This indicates that DI is a measure of steel hardenability that is independent of the quenching medium. Hardness, HRC 60 40 HRCcrit ≅50% M D crit 20 0 f80 f60 f50 f40 FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. TABLE 5.1 Grossmann Quenching Intensity Factor H H Value (in.21) Method of Quenching Oil Brine 0.25–0.30 0.30–0.35 0.35–0.40 0.40–0.50 0.50–0.80 0.80–1.10 No agitation Mild agitation Moderate agitation Good agitation Strong agitation Violent agitation Water 1.0 1.0–1.1 1.2–1.3 1.4–1.5 1.6–2.0 4.0 2.0 2.0–2.2 5.0 200 20 120 80 0.01 40 0 360 1. 0. 0 80 48 320 40 0.40 32 Steel A 0.20 24 Steel B 16 0.10 8 0 Quenching intensity H 80 120 160 200 240 280 Ideal critical diameter D I, mm 2. 40 ∞ 0 Quenching intensity H 160 0 Critical diameter Dcrit, mm 0. 0.10 10 5. .0 0 Critical diameter Dcrit, mm 240 5. 0 2. 1.0 0 0. 60 0. 40 ∞ Source: Metals Handbook, 8th ed., Vol. 2, American Society for Metals, Cleveland, OH, 1964, p. 18. 0.01 0 8 16 24 32 40 48 56 Ideal critical diameter D I, mm 64 72 FIGURE 5.6 The chart for converting the values of the critical diameter Dcrit into the ideal critical diameter 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.) ß 2006 by Taylor & Francis Group, LLC. 0.40 0.38 10.0 Grain size ASTM 0.36 0.34 9.5 9.0 4 0.32 8.5 8.0 6 0.26 7 6.5 0.24 8 6.0 7.5 7.0 0.22 5.5 0.20 5.0 0.18 DI , mm 5 0.28 DI , in. 0.30 4.5 0.16 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Carbon content, % FIGURE 5.7 The ideal critical diameter (DI) as a function of the carbon content and austenite grain size for 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 that steel for various quenching media. For low- and medium-alloy steels, hardenability as determined by DI may be calculated from the chemical composition after accounting for austenite grain size. First, the basic hardenability of the steel as a function of carbon content and austenite grain size is calculated from Figure 5.7 according to the weight percent of each element present. For example: if a steel has an austenite grain size of American Society for Testing 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) is DI ¼ 0.17. The total hardenability of this steel is DI ¼ 0:17  1:2  3:3  3:4  1:6 ¼ 3:7 in: (5:1) For these calculations, it is presumed that the total amount of each element is in solution at the austenitizing temperature. Therefore the diagram in Figure 5.8 is applicable for carbon contents above 0.8% C only if all of the carbides are in solution during austenitizing. This is not the case, because conventional hardening temperatures for these steels are below the temperatures necessary for complete dissolution of the carbides. Therefore, decreases in the basic hardenability are to be expected for steels containing more than 0.8% C, compared to values in the diagram. Later investigations by other authors produced similar diagrams that account for this decrease in the basic hardenability that is to be expected for steels with more than 0.8% C, compared to the values shown in Figure 5.8 [6]. Although values of DI calculated as above are only approximate, they are useful for comparing the hardenability of two different grades of steel. The most serious objection to Grossmann’s hardenability concept is the belief that the actual quenching intensity during the entire quenching process can be described by a single H value. It is well known that the heat transfer coefficient at the interface between the metal ß 2006 by Taylor & Francis Group, LLC. Multiplying factor Multiplying factor 3.8 8.4 Mn 3.4 7.6 Cr 3.0 Ni 6.8 2.6 6.0 Mo 2.2 5.2 Si 1.8 4.4 Mn (continued) 1.4 0.8 1.2 1.6 2.0 1.0 0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 3.6 4.0 Alloy content, % FIGURE 5.8 Multiplying factors for different alloying elements when calculating hardenability as DI 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 the quenching process for a vaporizable fluid. Another difficulty is the determination of the H value for a cross-sectional size other than 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 18Cr8Ni round bar quenched in water from 8458C [7]. It is evident that the H value determined in this way passed through a maximum with respect to terminal temperatures. It is also evident that H values at the centers of round bars decreased with increasing diameter. Values of the quenching intensity factor H do not account for specific quenchant and quenching characteristics such as composition, oil viscosity, or the temperature of the quenching bath. Table of H values do not specify the agitation rate of the quenchant either uniformly or precisely; that is, the uniformity throughout the quench tank with respect to mass flow or fluid turbulence is unknown. Therefore, it may be assumed that the tabulated H values available in the literature are determined under the same quenching conditions. This assumption, unfortunately, is rarely justified. In view of these objections, Siebert et al. [8] state: ‘‘It is evident that there cannot be a single H-value for a given quenching bath, and the size of the part should be taken into account when assigning an H-value to any given quenching bath.’’ 5.3.1.1 Hardenability in High-Carbon Steels The hardenability effect of carbon and alloying elements in high-carbon steels and the case regions of carburized steels differ from those in low- and medium-carbon steels and are influenced significantly by the austenitizing temperature and prior microstructure (normalized or spheroidize-annealed). Using Grossmann’s method for characterizing hardenability in terms of the ideal critical diameter DI, multiplying factors for the hardenability effects of Mn, Si, Cr, Ni, Mo, and Al were successfully derived [9] for carbon levels ranging from 0.75 to 1.10% C in single-alloy and multiple-alloy steels quenched at different austenitizing temperatures from 800 to 9308C. These austenitizing temperatures encompass the hardening temperatures of hypereutectoid tool steels, 1.10% C bearing steels, and the case regions of ß 2006 by Taylor & Francis Group, LLC. Temperature, C 300 500 700 3.2 H value, in.−1 2.8 A — 1/2-in. (13-mm) round B — 1-in. (25-mm) round C — 1-1/2-in. (38-mm) round D — 2-1/4-in. (57-mm) round E — 3-in. (76-mm) round Water temperature 60 F (16 C) Center couples A 2.4 0.13 0.12 0.11 0.10 0.09 0.08 2.0 0.07 B 1.6 0.06 C H value, mm−1 100 0.05 1.2 0.04 D 0.8 0.03 E 0.02 0.4 0.01 0.0 200 600 400 1000 800 1400 1200 1600 Temperature, F FIGURE 5.9 Change of the H value with temperature and size of the round bar. Calculated from cooling curves measured at the center of bars made of 18Cr8Ni steel quenched in water from 8458C, according to Carney and Janulionis. (From D.J. Carney and A.D. Janulionis, Trans. ASM 43:480–496, 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 solution could vary with the prior microstructure and the austenitizing conditions. The hardenability of 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 because hardenability 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 for machining 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 of which 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 much more difficult to dissolve when the steel is heated for hardening. Therefore the amount of alloy 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 spheroidized prior microstructure actually yields higher hardenability than a prior normalized microstructure, at least for austenitizing temperatures up to approximately 8558C. This effect occurs because larger carbides are not as efficient nuclei for early pearlite formation upon cooling as fine and lamellar carbides and the nuclei are present in lower numbers. With either prior microstructure, if strict control is maintained over austenitizing temperature and time, the solution of carbon and alloy can be reproduced with sufficient consistency to permit the ß 2006 by Taylor & Francis Group, LLC. Indicated hardenability D I 7 6 50% Martensite 5 4 95% Martensite 3 2 1 0 99.9% Martensite 1 2 3 4 5 6 Hardenability D I, 50% martensite 7 FIGURE 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 whether pearlite or bainite would limit hardenability because the effects of some elements on these reactions 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 with reference 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 be bainite. The 50% martensite hardenability criterion (usually used when calculating DI) was selected by Grossmann because this structure in medium-carbon steels corresponds to an inflection in the hardness distribution curve. The 50% martensite structure also results in marked contrast in etching between the hardened and unhardened areas and in the fracture appearance of these areas in a simple fracture test. For many applications, however, it may be necessary to through-harden to a higher level of martensite to obtain optimum properties of tempered martensite in the core. In these instances, D1 values based on 90, 95, or 99.9% martensite must be used in determining the hardenability requirements. These D1 values can be either experimentally determined or estimated from the calculated 50% martensite values using the relationships shown in Figure 5.10, which were developed for medium-carbon low-alloy steels [10]. A curve for 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 D1 values were obtained from the measured Jominy curves using the conversion curve modified by Carney shown in Figure 5.12. The measured DI values were plotted against the percent content of various elements in the steel. These curves were then used to adjust the DI value of the steels whose residual content did not conform to the base composition. Once the DI value of each analysis was adjusted for residuals, the final step was to derive the multiplying factors for each element from the quotient of the steels Dà and that of the base as follows: I fMn ¼ where DI is the initial reference value. ß 2006 by Taylor & Francis Group, LLC. Dà at x % Mn I DI (5:2) D I, Normalized prior structure, in. 4 3 2 1 2 3 4 5 D I, Annealed prior structure, in. 6 FIGURE 5.11 Correlation between hardenability based on normalized and spheroidize-annealed prior structures in alloyed 1.0% C steels. (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) Excellent agreement was obtained between the case hardenability results of carburized steels assessed at 1.0% carbon level and the basic hardenability of the 1.0% C steels when quenched from the normalized prior structure. It was thus confirmed that all multiplying factors 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. The multiplying factors shown in Figure 5.15 through Figure 5.18 were principally determined in compositions where only single-alloy additions were made and that were generally pearlitic in initial transformation behavior. Consequently, these multiplying factors may be applied to 70 60 DI 50 40 80 30 70 20 60 32 2.0 10 2 4 6 40 48 56 Sixteenths 3.0 in. 64 4.0 8 10 12 14 16 18 20 22 24 26 28 30 32 Distance from end-quenched end—sixteenths .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 Distance from end-quenched end—in. 2.0 FIGURE 5.12 Relationship between Jominy distance and DI. (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) ß 2006 by Taylor & Francis Group, LLC. Normalized prior structure Base : DI — 1.42 6 1.00 Mo Multiplying factor 5 0.90 4 0.80 Carbon factor 3 0.70 Si-Multi-alloy steels 2 N 0.75 1.00 1.25 1.50 Mn Cr-Carburizing steels Cr Si-Single-alloy steels Ni %C 1 Mn, Cr, Si 0 0 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 Percent element FIGURE 5.13 Multiplying factors for calculation hardenability of high-carbon steels of prior normalized structure. (From 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.) the calculation of hardenability of all single-alloy high-carbon compositions and to those multialloyed compositions that remain pearlitic when quenched from these austenitizing conditions. This involves all analyses containing less than 0.15% Mo and less than 2% total of Ni plus Mn and also less than 2% Mn, Cr, or Ni when they are present individually. Of course, all of the factors given in Figure 5.15 through Figure 5.18 also apply to the calculation of case hardenability of similar carburizing steels that are rehardened from these temperatures following air cooling or integral quenching. Annealed prior structure 6 Base : DI — 1.42 1.00 Multiplying factor 5 0.90 4 0.80 Carbon factor Mo 3 0.70 1.00 0.75 Si-Multi-alloy steels 2 Mn 1.25 %C 1.50 Si-Single-alloy steels Cr Ni Ni 1 Mn, Cr, Si 0 0 0.25 0.50 0.75 1.00 1.25 1.50 Percent element 1.75 2.00 2.25 FIGURE 5.14 Multiplying factors for calculation of hardenability of high-carbon steels of prior spheroidize-annealed structure. (From 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.) ß 2006 by Taylor & Francis Group, LLC. 1.00 0.90 4 5 Multiplying factor 0.80 1700 6 7 0.70 1525 1575 8 0.60 1475 0.50 0.40 0.20 0.30 0.40 0.50 0.60 0.70 0.80 Percent carbon 0.90 1.00 1.10 FIGURE 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 to ASTM 8. (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) 3.5 Kramer Manganese 3.0 1700 1575 2.5 Multiplying factor 1525 2.0 1475 1.5 1.0 Chromium Kramer 1700 2.5 1700 2.0 1525−1575 1.5 1475 1.0 0 0.25 0.50 0.75 1.00 1.25 Percent element 1.50 1.75 2.00 FIGURE 5.16 Effect of austenitizing temperature on multiplying factors for Mn and Cr at high-carbon levels (Kramer data for medium-carbon steels). (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) ß 2006 by Taylor & Francis Group, LLC. 1700 Molybdenum Multiplying factor 5.0 4.0 Kramer 1575 3.0 1475 1525 2.0 1.0 0 0.25 0.50 0.75 1.00 Percent molybdenum 1.25 1.50 FIGURE 5.17 Effect of austenitizing temperature on multiplying factors for Mo at high carbon levels. (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) Aluminum 2.0 1475–1700 1.5 Kramer 1.0 1700 Multi−alloy Multiplying factor Silicon 2.5 1700 2.0 Kramer 1475 1525 Multi−alloy 1575 1.5 1475−1575 1.0 Nickel 2.0 Kramer 1700 1.5 1475−1575 1.0 0 0.25 0.50 0.75 1.00 1.25 Percent element 1.50 1.75 2.00 FIGURE 5.18 Effect of austenitizing temperature on multiplying factors for Si, Ni, and Al at high-carbon levels. (Arrow on Al curve denotes maximum percentage studies by Kramer.) (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) ß 2006 by Taylor & Francis Group, LLC. For steels containing more Mo, Ni, Mn, or Cr than the above percentages, the measured hardenability will always be higher than calculated with the single-alloy multiplying factors because these steels are bainitic rather than pearlitic and also because synergistic hardenability effects have been found to occur between certain elements when present together. The latter effect was specifically noted between Ni and Mn, especially in steels made bainitic by the 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 for Mn and Ni, as the independence of alloying element effects is implicit in the Grossmann multiplying 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 steels that are spheroidize-annealed prior to hardening. However, the calculated DI value must be converted 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 as high as 660 mm (26.0 in.). 50 .80 80 40 60 30 Mn .70 40 30 20 20 % Nickel 30 .60 .50 .0 .30 Mn 40 Combinded Ni x Mn multiplying factor 20 10 .80 14758F (8008C) 80 40 30 Mn 60 .70 40 30 20 20 .60 .50 .40 .30 Mn % Nickel 30 40 20 10 15258F (8308C) .80 80 50 30 Mn 60 40 40 20 20 % Nickel 30 .70 40 .60 .50 .40 .30 Mn 30 20 15758F (8558C) 10 1.0 1.2 1.4 1.6 1.8 Percent nickel FIGURE 5.19 Combined multiplying factor for Ni and Mn in bainitic high-carbon steels quenched from 800 to 8558C, to be used in place of individual factors when composition contains more than 1.0% Ni and 0.15% Mo. (From C.F. Jatczak, Metall. Trans. 4:2267–2277, 1973.) ß 2006 by Taylor & Francis Group, LLC. 5.3.2 JOMINY END-QUENCH HARDENABILITY TEST The end-quench hardenability test developed by Jominy and Boegehold [12] is commonly referred to as the Jominy test. It is used worldwide, described in many national standards, and available 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 wide range 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 supporting fixture (Jominy apparatus) and quenched from the lower end by spraying with a jet of water under specified conditions as illustrated in Figure 5.20. The cooling rate is the highest at the end 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 the quenched end. After quenching, two parallel flats, approximately 0.45 mm below surface, are ground on opposite sides of the specimen and hardness values (usually HRC) are measured at 1=16 in. intervals from the quenched end and plotted as the Jominy hardenability curve (see Figure 5.21). When the distance is measured in millimeters, the hardness values are taken at every 2 mm from the quenched end for at least a total distance of 20 or 40 mm, depending on the steepness of the hardenability curve, and then every 10 mm. On the upper margin of the Jominy hardenability diagram, approximate cooling rates at 7008C may be plotted at several distances from the quenched end. 1/2 in. (12.7 mm) 1/8 in. (3.2 mm) 1−1/8 in. (29 mm) 458 1−1/32 in. (26.2 mm) Unimpeded water jet 4 in. (102 mm) 1-in. (25.4-mm) round specimen 2-1/2 in. (64 mm) Water at 75 ± 5°F (24 ± 2.8°C) 1/2-in. (12.7-mm) i.d. orifice 1/2 in. (13 mm) From quick-opening valve FIGURE 5.20 Jominy specimen and its quenching conditions for end-quench hardenability test. ß 2006 by Taylor & Francis Group, LLC. 270 70 18 5.6 K/s 489" 124" 32.3" 10" °F/s 1/16 4/16 8/16 16/16 Cooling rates Distance from quenched end, in. 60 Hardness, HRC 50 40 30 20 10 1.0 0 0 2.0 3.0 in. 75 mm 25 50 Distance from quenched end FIGURE 5.21 Measuring hardness on the Jominy specimen and plotting the Jominy hardenability curve. (From G. Krauss, Steels Heat Treatment and Processing Principles, ASM International, Metals Park, OH, 1990.) Figure 5.22 shows Jominy hardenability curves for different unalloyed and low-alloyed grades of steel. This figure illustrates the influence of carbon content on the ability to reach a certain hardness level and the influence of alloying elements on the hardness distribution expressed as hardness values along the length of the Jominy specimen. For example, DIN Ck45, an unalloyed steel, has a carbon content of 0.45% C and exhibits a higher maximum hardness (see the value at 0 distance from the quenched end) than DIN 30CrMoV9 steel, 60 Hardness, HRC 50CrV4 30CrMoV9 40 50CrMo4 42MnV7 37MnSi5 20 Ck45 0 0 40 60 20 Distance from quenched end, mm 80 FIGURE 5.22 Jominy hardenability curves (average values) for selected grades of steel (designations according to German DIN standard). (From G. Spur (Ed.), Handbuch der Fertigungstechnik, Band 4=2, ¨ Warmebehandeln, Carl Hanser, Munich, 1987, p. 1012.) ß 2006 by Taylor & Francis Group, LLC. Distance from quenched end, mm 10 20 30 40 70 50 Hardness, HRC 60 50 40 30 20 10 0 4 8 12 16 20 24 Distance from quenched end, 1/16 in. 28 32 FIGURE 5.23 Reproductibility of the end-quench hardenability test. Hardenability range (hatched area between curves) based on tests by nine laboratories on a single heat of SAE 4068 steel. (From C.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 a higher 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 (core hardenability) and for structural steels, which are typically through-hardened in oils and tempered. The Jominy end-quench test is suitable for all steels except those of very low or very high hardenability, i.e., D1 < 1.0 in. or D1 > 6.0 in. [8]. The standard Jominy end-quench test cannot be used for highly alloyed air-hardened steels. These steels harden not only by heat extraction through the quenched end but also by heat extraction by the surrounding air. This effect 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 of an end-quench hardenability test performed by nine laboratories on a single heat of SAE 4068 steel [8]. Generally, quite good reproducibility was achieved, although the maximum difference may be 8–12 HRC up to a distance of 10 mm from the quenched end depending on the slope of the curve. Several authors who have investigated the effect of deviations from the standard test procedure have concluded that the most important factors to be closely controlled 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 transfer time from the furnace to the quenching fixture are not as critical. 5.3.2.1 Hardenability Test Methods for Shallow-Hardening Steels If 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 hardness decreases rapidly for every incremental increase in Jominy distance. Therefore the standard Jominy specimen with hardness readings taken at every 1=16 in. (1.59 mm) cannot describe precisely the hardness trend (or hardenability). To overcome this difficulty it may be helpful ß 2006 by Taylor & Francis Group, LLC. to (1) modify the hardness survey when using standard Jominy specimens or (2) use special L specimens. 5.3.2.1.1 Hardness Survey Modification for Shallow-Hardening Steels The essential elements of this procedure, described in ASTM A255, are as follows: 1. The procedure in preparing the specimen before making hardness measurements is the same as for standard Jominy specimens. 2. An anvil that provides a means of very accurately measuring the distance from the quenched end is essential. 3. Hardness values are obtained from 1=16 to 1=2 in. (1.59–12.7 mm) from the quenched end at intervals of 1=32 in. (0.79 mm). Beyond 1=2 in., hardness values are obtained at 5=8, 3=4, 7=8, and 1 in. (15.88, 19.05, 22.23, and 25.4 mm) from the quenched end. For readings within the first 1=2 in. from the quenched end, two hardness traverses are made, both with readings 1=16 in. apart: one starting at 1=16 in. and completed at 1=2 in. from the quenched end, and the other starting at 3=32 in. (2.38 mm) and completed at 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 bed that will accommodate the indentations of the flat surveyed first. The second hardness traverse is made after turning the bar over. If the fixture does not have such a grooved bed, two pairs of flats should be ground, the flats of each pair being 1808 apart. The two hardness surveys are made on adjacent flats. 5. For plotting test results, the standard form for plotting hardenability curves should be used. 5.3.2.1.2 The Use of Special L Specimens To 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 except that the stream of water rises to a free height of 100+5 mm (instead of the 63.55 mm with a standard specimen) above the orifice, without the specimen in position. (a) (b) f32 f25 f32 100 ± 0.5 97 ± 0.5 25 25 50 10 97 ± 0.5 100 ± 0.5 f5 f125 f125 f20 f20 f25 f25 FIGURE 5.24 L specimens for Jominy hardenability testing of shallow-hardening steels. All dimensions in millimeters. ß 2006 by Taylor & Francis Group, LLC. h1 h 2 S S h3 h4 Hardness, HRC h5 A B C D E h6 h 7 F G C K 1 16 A A Diameter, 1 in. S = average surface hardness h1, h2, h3, etc. = average hardness at depths indicated C = Average center hardness Then Area of A = s + h1 1/ 16 2 Area of B = h1 + h2 1/ 16 2 Total area = 2(A + B + C + D + E + F + G + K ) S C = 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. 473–474.) [15] 5.3.2.1.3 The SAC Hardenability Test The SAC hardenability test is another hardenability test for shallow-hardening steels, other than carbon tool steels, that will not through-harden in sizes larger than 25.4 mm (1 in.) in diameter. 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 at the specified temperature of 1 h and cooling in air, it is austenitized by being held at temperature for 30 min and quenched in water at 24+58C, where it is allowed to remain until 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 tempering that 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. The average of these readings then becomes the surface reading. Next, a series of HRC readings are taken on the cross section in steps of 1=16 in. (1.59 mm) from the surface to the center of the specimen. From these readings, a quantitative value can be computed and designated by a code 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 60-54-43 indicates a surface hardness of 60 HRC, a total Rockwell-inch area of 54, and a center hardness of 43 HRC. The computation of the total Rockwell-inch area is shown in Figure 5.25. 5.3.2.1.4 Hot Brine Hardenability Test For steels of very low hardenability, another test has been developed [15] that involves quenching several specimens 2.5 mm (0.1 in.) thick and 25 mm (1.0 in.) square in hot brine at controlled temperatures (and controlled quench severity), and determining the hardness and percent martensite of each specimen. The brine temperature for 90% martensite structure expressed as an equivalent diameter of a water-quenched cylinder is used as the hardenability ß 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 steels using this method, a linear regression equation has been derived for estimating hardenability from chemical composition and grain size that expresses the relative contribution of carbon and alloying elements by additive terms instead of multiplicative factors. 5.3.2.2 Hardenability Test Methods for Air-Hardening Steels When a standard Jominy specimen is used, the cooling rate at a distance of 80 mm from the quenched end (essentially the opposite end of the specimen) is approximately 0.7 K=s. The hardenability of all steel grades with a critical cooling rate greater than 0.7 K=s can be determined by the standard Jominy end-quench hardenability test as a sufficient decrease in hardness 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=s there will be no substantial change in the hardness curve because martensite will be obtained at every distance along the Jominy specimen. This is the case with air-hardening steels. To cope with this situation and enable the use of the Jominy test for air-hardening steels, the mass of the upper part of the Jominy specimen should be increased [16] by using a stainless steel cap as shown in Figure 5.26. In this way, cooling rates of the upper part of the specimen are 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 be fixed with the cap. When austenitizing, a leg is installed on the lower end of the specimen as shown in Figure 5.26 to equalize heating so that the same austenitizing conditions exist along the entire test specimen. 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 (together with the cap), the leg should be removed. Figure 5.27 illustrates cooling rates when quenching a standard Jominy specimen and a modified specimen with added cap. This diagram illustrates the relationship between the cooling times from the austenitizing temperature to 5008C and the distance 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 mm from the quenched end, the cooling time curves for the standard specimen and the modified 70 f f 66 All dimensions in mm 27 4 8 58 ff 42 65 47.5 87 Cap 8 Leg 26 ff 30 32 f f 43 45 f FIGURE 5.26 Modification of the standard Jominy test by the addition of a cap to the specimen for testing the hardenability of air-hardening steels. (From A. Rose and L. Rademacher, Stahl Eisen 76(23):1570–1573, 1956 [in German].) ß 2006 by Taylor & Francis Group, LLC. Jominy distance from the quenched end, mm 110 100 Modified Jominy specimen (added cap) Standard Jominy specimen 90 Austenitizing temperature: 8008C 80 8008C 9008C 10008C 11008C 70 60 50 40 30 20 10 0 0 400 500 600 100 200 300 Cooling time from austenitizing temp. to 5008C, s FIGURE 5.27 Cooling times between austenitizing temperature and 5008C for the standard Jominy specimen and for a specimen modified by adding a cap. (From A. Rose and L. Rademacher, Stahl Eisen 76(23):1570–1573, 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 slower cooling rates relative to the standard specimen. By adding the cap, the cooling time is nearly doubled, 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 standard specimen and the other with the modified specimen, for the hot-working tool steel DIN 45CrMoV67 (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 the modified 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 be seen only if a quenched Jominy specimen is tempered at a temperature that will result in a secondary hardening effect. Figure 5.29 illustrates this for the tool steel DIN 45CrVMoW58 70 Austenitizing temp. 970 C 60 Hardness, HRC 50 40 Standard Jominy specimen Modified Jominy specimen (added cap) 30 20 Depth of the ground flat 1 mm 10 0 0 50 60 70 80 10 20 30 40 Jominy distance from the quenched end, mm FIGURE 5.28 Jominy hardenability curves of grade DIN 45CrMoV67 steel for a standard specimen and for a specimen modified by adding a cap. (From A. Rose and L. Rademacher, Stahl Eisen 76(23):1570– 1573, 1956 [in German].) ß 2006 by Taylor & Francis Group, LLC. 70 Austenitizing temperature 1100 C 60 Hardness, HRC 50 40 Not tempered Tempered at: 300 C 550 C 30 20 Modified Jominy specimen (added cap) Depth of the ground flat 1 mm 10 0 0 10 20 30 40 50 60 70 Jominy distance from the quenched end, mm 80 FIGURE 5.29 Jominy hardenability curves of grade DIN 45CrVMoW58 steel after quenching (solid curve) 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):1570–1573, 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 the quenched end decreases. Within this region martensitic structure is predominant. At about 25 mm from the quenched end the hardness curve after tempering becomes equal to the hardness curve after quenching. After tempering to 5508C, however, the hardness is even more decreased up to a distance of 17 mm from the quenched end, and for greater distances a hardness increase up to about 4 HRC units can be seen as a result of the secondary hardening effect. This increase in hardness can be detected only when the modified Jominy test is conducted. Another approach for measuring and recording the hardenability of air-hardening steels is the Timken Bearing Company Air hardenability Test [17]. This is a modification of the air-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 a cylindrical bar 6 in. in diameter by 15 in. long, leaving 4 in. of each test bar exposed. The total setup is heated to the desired hardening temperature for 4 h. The actual time at temperature is 45 min for the embedded bar sections and 3 h for the sections extending outside the large cylinder. The test bar is then cooled in still air. The large cylindrical bar restricts the cooling of the exposed section of each test bar, producing numerous cooling conditions along the bar length. 4 in. 1 in.–8 Thread 61/2 in. 11/8 in. 1.0 in. Diameter 6 in. Diameter 875 in. Diameter 800 in. Diameter 10 in. 1 in. 15 in. FIGURE 5.30 Timken Roller Bearing Company air hardenability test setup. Two test specimens with short threaded sections as illustrated are fixed in a large cylindrical bar. (From C.F. Jatczak, Trans. ASM 58:195–209, 1965.) ß 2006 by Taylor & Francis Group, LLC. The various positions along the air-hardenability bar, from the exposed end to the opposite 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 are shown in Figure 5.31. The actual cooling rates corresponding to each bar position are shown. 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. To prevent 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 center of a 13-in. diameter bar cooled in still oil and even larger cylindrical bars cooled in water. Co W Si Cr Ni Mo V Norm. temp. 8F Quench temp. 8F 0.61 - - 0.67 1.30 0.18 0.47 0.26 Ann 1750 0.54 1.06 - 0.53 1.26 - 0.52 0.27 " 1750 0.38 0.40 - - 0.85 4.87 0.11 1.34 0.60 " 1850 A115 0.39 0.52 - - 0.85 5.12 - 5.10 0.68 " 1850 Lapelloy 18287 0.31 1.07 - - 0.27 11.35 0.43 2.85 0.24 " 1900 HTS-1100 A117 0.44 0.42 - 1.70 0.51 1.48 1.01 " 1900 6 in.φ312 in. 6 in.φ312 in. 0.83 12 6 in.φ36 in. 1.0 22 51/2 in.φ312 in. 2 in.φ312 in. 3 in.φ312 in. 2 in.φ312 in. 3 in.φ312 in. 4 in.φ312 in. 32 2.0 1.5 1.2 3 in.φ33 in. 4 in.φ34 in. 5 in.φ35 in. 21/2 in.φ3221/2 in. 48 63 3.0 72.5 Cooling rate in °/F min 2.4 2 in.φ32 in. Size round with same still air cooling rate Equiv. A/V ratio 1.39 13 Halmo 5 in.φ312 in. 12887 51/2 in.φ312 in. H-11 7 in.φ37 in. 0.31 Interface 0.29 A120 0.85 10420 5 in.φ312 in. "+Co 13 Mn 4 in.φ312 in. C 16 1722 AS Code 11/2 in.φ312 in. Heat no. Size round with same as quenched hardness 11/2 in.φ312 in. 11/4 in.φ312 in. Type Rockwell C hardness scale 65 60 55 50 45 40 35 30 25 0 2 4 6 8 10 12 14 16 18 20 Distance in 1/2 in. units from large end of air hardenability bar FIGURE 5.31 Chemistry and air-hardenability test results for various Cr–Mo–V steels. (From C.F. Jatczak, Trans. ASM 58:195–209, 1965.) ß 2006 by Taylor & Francis Group, LLC. 65 1340H 60 Limits for steel made to chemical specifications Hardness, HRC 55 50 45 40 Standard H- band 35 30 25 20 0 2 4 6 8 10 12 14 16 18 20 22 24 Distance from quenched end, 1/16 in. 26 28 30 32 FIGURE 5.32 Hardenability band for SAE 1340H steel. 5.3.3 HARDENABILITY BANDS Because of differences in chemical composition between different heats of the same grade of steel, 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 by the letter H following the composition code. Figure 5.32 shows such a hardenability band for 1340H 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 represents the 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 the majority of steels are purchased according to hardenability bands. Suppliers guarantee that 93 or 95% of all mill heats made to chemical specification will also be within the hardenability band. The H bands were derived from end-quench data from a large number of heats of a specified composition range by excluding the upper and lower 3.5% of the data points. Steels may be purchased either to specified composition ranges or to hardenability limits defined by H bands. In the latter case, the suffix H is added to the conventional grade designation, for example 4140H, and a wider composition range is allowed. The difference in hardenability between an H steel and the same steel made to chemical specifications is illustrated in Figure 5.32. These differences are not the same for all grades. High-volume production of hardened critical parts should have close tolerance of the depth of hardening. The customer may require, at additional cost, only those heats of a steel grade that 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 ¼ 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 Jominy distance should be closer to the quenched end. For thick sections, greater Jominy distances are important. . The minimum and maximum distance from the quenched end where a desired hardness value occurs. For example, ß 2006 by Taylor & Francis Group, LLC. 70 Hardness, HRC 60 A C D 50 B B 40 C 30 A D 20 0 4 8 12 16 20 24 28 Distance from quenched surface, 1/16 in. 32 FIGURE 5.33 Different ways of specifying hardenability limits according to SAE. J45 ¼ 7=16 À 14=16 in: (B--B, Figure 5.33) . Two maximum hardness values at two desired Jominy distances. For example, J52 ¼ 12=16 in: ( max ); . (5:4) J38 ¼ 16=16 in: (max) (5:5) Two minimum hardness values at two desired Jominy distances. For example, J52 ¼ 6=16 in: ( min ); J28 ¼ 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 structure–volume fraction diagram (see Figure 5.34) for the same steel is available, the effective 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 structure— volume fraction diagram can also be used for the preparation of the transformation diagram when limits of the hardenability of a steel are determined. If the structure—volume fraction diagram is not available, the limit values of hardness or the effective depth of hardening can be estimated form the hardenability band using the diagram shown in Figure 5.35. Hardness depends 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 may vary 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 having maximum hardenability and a heat with minimum hardenability can be determined from the following examples. For Cmin ¼ 0.31% and 50% martensite, a hardness of 38 HRC can be determined 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 ¼ 0.39% and 50% martensite, a hardness of 42 HRC can be determined from Figure 5.35. This hardness corresponds to the upper curve of the hardenability band and is found at 20 mm from the quenched end. In this example, the Jominy hardenability (measured up to 50% martensite) for this steel varies between 4 and 20 mm. Using conversion charts, differences in the depth of hardening for any given diameter of round bars quenched under the same conditions can be determined. ß 2006 by Taylor & Francis Group, LLC. Hardness, HRC 60 50 40 30 20 100 P 75 F B Structure proportion, % 50 Ms 25 0 100 P 75 B 50 25 Ms F 0 10 20 30 0 40 50 Distance from quenched end of the Jominy specimen, mm FIGURE 5.34 Hardenability band and structure–volume fraction diagram of SAE 5140 steel. ˇˇ ´ F ¼ ferrite, P ¼ pearlite, B ¼ bainite, Ms ¼ martensite. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.) 70 Martensite 99.9% 95 90 80 50% Hardness, HRC 60 50 40 C Ni 30 MnSi CrSi CrNiMo Mo CrMo Cr Maximum hardness after Burns, Moore and Archer Hardness at different percentages of martensite after Hodge and Orehoski 20 10 CrNi 0 0.1 0.2 0.3 0.4 0.5 0.6 Carbon content, wt % 0.7 0.8 0.9 FIGURE 5.35 Achievable hardness depending on the carbon content and percentage of martensite in the ˇˇ ´ structure. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, SpringerVerlag, Berlin, 1992.) ß 2006 by Taylor & Francis Group, LLC. 60 Max. hardness difference 32 HRC at J = 10 mm Hardness, HRC Gradient of hardness 25 HRCmin at J = 7.5 mm 22 HRC/5 mm 50 47 HRCmin at J = 2.5 mm 50% Martensite 40 30 20 37 Mn Si 5 10 0 10 20 30 40 50 60 Distance from quenched end, J, mm 38 HRCmin at 4 mm 42 HRCmax at 20 mm (Cmin = 0.31%; (Cmax = 0.39%; at 38 HRCmin) at 42 HRCmax) 50% martensite 50% martensite Hardenability: J(50 M) = 4–20 mm C 31–39; J 4–20 FIGURE 5.36 Hardenability band of DIN 37MnSi5 steel and the way technologically important ˇˇ ´ information can be obtained. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.) Effective depth of hardening is not the only information that can be derived from the hardenability band. Characteristic features of every hardenability band provide information on 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 quenched end, 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 the hardness gradient. In Figure 5.36, this is illustrated by the minimum hardenability curve for the steel in question where there is a high gradient of hardness (22 HRC for only 5 mm difference in the Jominy distance). High hardness gradients indicate high sensitivity to cooling rate variation. 5.4 CALCULATION OF JOMINY CURVES FROM CHEMICAL COMPOSITION The first calculations of Jominy curves based on the chemical composition of steels were performed in the United States in 1943 [21,22]. Later, Just [23], using regression analysis of fictitious Jominy curves from SAE hardenability bands and Jominy curves of actual heats from the USS Atlas (USA) and MPI-Atlas (Germany), derived expressions for calculating the hardness at different distances (E) from the quenched end of the Jominy specimen. It was found that the influence of carbon depends on other alloying elements and also on the cooling rate, 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 other alloying elements have the factor 0 at this distance. This implies that the hardness at a Jominy distance of 0 is governed solely by the carbon content. The influence of other alloying elements generally increases from 0 to values of their respective factors up to a Jominy distance of about 10 mm. Beyond this distance, their influence is essentially constant. Near the quenched end the ß 2006 by Taylor & Francis Group, LLC. influence of carbon prevails, while the influence of other alloying elements remains essentially constant beyond a Jominy distance of about 10 mm. This led Just to propose a single expression for the whole test specimen, except for distances shorter than 6 mm: pffiffiffiffi pffiffiffiffi J6À80 ¼ 95 C À 0:00276E 2 C þ 20Cr þ 38Mo þ 14Mn þ 5:5Ni þ 6:1Si þ 39V pffiffiffiffi þ 96P À 0:81K À 12:28 E þ 0:898E À 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 within the 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 the equation for the maximum attainable hardness with 100% martensite, is pffiffiffiffi Hmax ¼ 60 C þ 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 end of the Jominy specimen, other authors argue that beyond a Jominy distance of 65 mm the continuous decrease in cooling rate at the Jominy test cannot be ensured even for low-alloy steels because of the cooling effect of surrounding air. Therefore, newer calculation methods rarely go beyond a Jominy distance of 40 mm. Just [23] found that a better fit for existing mutual correlations can be achieved by formulas 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 than for structural steels for hardening and tempering. Therefore, separate formulas for casehardening steels were derived: pffiffiffiffi pffiffiffiffi J6À40 (case-hardening steels) ¼ 74 C þ 14Cr þ 5:4Ni þ 29Mo þ 16Mn À 16:8 E þ 1:386E þ 7HRC (5:9) and for steels for hardening and tempering, pffiffiffiffi J6À40 (steels for hardening and tempering) ¼ 102 C þ 22Cr þ 21Mn þ 7Ni þ 33Mo pffiffiffiffi À 15:47 E þ 1:102E À 16HRC (5:10) In Europe, five German steel producers in a VDEh working group jointly developed formulas that adequately define the hardenability from different production heats [24]. The goal was to replace various existing formulas that were used individually. Data for some case-hardening steels and some low-alloy structural steels for hardening and tempering have been compiled, and guidelines for the calculation and evaluation of formulas for additional families of steel have been established. This work accounts for influential factors from the steel melting process and for possible deviations in the Jominy test itself. Multiple linear regression methods using measured hardness values for Jominy tests and actual chemical compositions were also included in the analyses. The number of Jominy curves of a family of steel grades necessary to establish usable formulas should be at least equal to the square of the total number of chemical elements used for the calculation. Approximately 200 curves were suggested. To obtain usable equations, all Jominy curves for steel grades that had similar transformation characteristics (i.e., similar continuous cooling transformation [CCT] diagram) ß 2006 by Taylor & Francis Group, LLC. TABLE 5.2 Regression Coefficients for the Calculation of Jominy Hardness Values for Structural Steels for Hardening and Tempering Alloyed with about 1% Cr Jominy Distance (mm) 1.5 3 5 7 9 11 13 15 20 25 30 Regression Coefficients Constant C Si Mn 29.96 26.75 15.24 À7.82 À27.29 À39.34 À42.61 À42.49 À41.72 À41.94 À44.63 57.91 58.66 64.04 81.10 94.70 100.78 95.85 88.69 78.34 72.29 72.74 2.29 3.76 10.86 19.27 22.01 21.25 20.54 20.82 17.57 18.62 19.12 3.77 2.16 4.87 10.24 14.70 16.06 17.75 20.18 20.73 21.42 S Cr À41.85 À73.79 À37.76 À65.81 À81.41 2.86 12.29 21.02 24.82 25.39 26.46 25.33 23.85 24.08 24.39 Mo Ni 6.66 38.31 52.63 54.91 47.16 7.51 7.69 10.75 Cu N À2.65 À2.59 30.41 38.97 26.95 35.99 27.57 Al 83.33 59.87 À115.50 À176.82 À144.07 4.56 8.58 7.97 9.0 8.89 9.96 9.64 9.71 ¨bben, H. Rohloff, P. Schu V. Schu ¨ler, ¨ler, and H.J. Wieland, Source: R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. Lu ¨ Harterei Tech. Mitt. 47(3):183–188, 1992. when hardened were used. Therefore, precise equations for the calculation of Jominy hardness values were derived only for steel grades of similar composition [24]. The regression coefficients for a set of equations to calculate the hardness values at different Jominy distances from 1.5 to 30 mm from the quenched end are provided in Table 5.2. The chemistry of the steels used for this study is summarized in Table 5.3. The regression coefficients in Table 5.2 do not have the same meaning as the hardenability factors in Equation 5.7, Equation 5.9, and Equation 5.10; therefore, there is no restriction on the calculation of Jominy hardness values at less than 6 mm from the quenched end. Because the regression coefficients used in this method of calculation are not hardenability factors, care should 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. The TABLE 5.3 Limiting Values of Chemical Composition of Structural Steels for Hardening and Tempering Alloyed with about 1% Cra Content (%) C Min. Max. Mean s Si Mn P S Cr Mo Ni Al Cu N 0.22 0.47 0.35 0.06 0.02 0.36 0.22 0.07 0.59 0.97 0.76 0.07 0.005 0.037 0.013 0.005 0.003 0.038 0.023 0.008 0.80 1.24 1.04 0.10 0.01 0.09 0.04 0.02 0.01 0.28 0.13 0.05 0.012 0.062 0.031 0.007 0.02 0.32 0.16 0.05 0.006 0.015 0.009 0.002 a Used in calculations with regression coefficients of Table 5.2. ¨bben, H. Rohloff, P. Schu ¨ler, V. Schu and H.J. Wieland, ¨ler, Source: R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. Lu ¨ Harterei Tech. Mitt. 47(3):183–188, 1992. ß 2006 by Taylor & Francis Group, LLC. Steel 41Cr4 (DIN) 60 0.9 2.1 s= Σ Δ2 n−2 50 2.5 s = 2.94 HRC 1.1 Hardness, HRC 40 Δ = calculated hardness − measured hardness 3.6 30 60 0.7 0.1 s = 7.45 HRC 50 −7.4 −10.5 40 Δ = calculated hardness − measured hardness 30 0 10 20 Distance from the quenched end, mm −1.4 30 FIGURE 5.37 Comparison between measured (O) and calculated (.) hardness values for a melt with adequate 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):183–188, 1992.) upper curve for a heat of DIN 41Cr4 steel, having a residual scatter of s ¼ 2.94 HRC, shows an adequate consistency, while the lower curve for another heat of the same steel, with a residual scatter of s ¼ 7.45 HRC, shows inadequate consistency. Such checks were repeated for every Jominy distance and for every heat of the respective steel family. During this process it was found that the residual scatter depends on the distance from the quenched end and that calculated Jominy curves do not show the same precision (compared to measured curves) at all 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 the residual scatter of the calculated results, it was concluded ‘‘that properly calibrated predictors offer a strong advantage over testing in routine applications’’ [25]. When judging the precision of a calculation of Jominy hardness values, hardenability predictors are expected to accurately predict (+1 HRC) the observed hardness values from the chemical composition. However, experimental reproducibility of a hardness value at a fixed Jominy distance near the inflection point of the curve can be 8–12 HRC (see Figure 5.23 for J10mm). Therefore it was concluded ‘‘that a properly calibrated hardenability formula will always anticipate the results of a purchaser’s check test at every hardness point better than an actual Jominy test’’ [25]. 5.4.1 HYPERBOLIC SECANT METHOD FOR PREDICTING JOMINY HARDENABILITY Another method for predicting Jominy end-quench hardenability from composition is based on the four-parameter hyperbolic secant curve-fitting technique [26]. In this method, it is ß 2006 by Taylor & Francis Group, LLC. Hardness dissipation, HRC 4 3 2 Cr family of steels CrMo family of steels MnCr family of steels C family of steels 1 0 0 30 10 20 Distance from the quenched end, mm 40 FIGURE 5.38 Residual scatter between measured and calculated hardness values versus distance to the quenched 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):183–188, 1992.) assumed that the Jominy curve shape can be characterized by a four-parameter hyperbolic secant (sech) function (SECH). The SECH curve-fitting technique utilizes the equation DHx ¼ A þ B{sech[C (x À 1)D ] À 1} (5:11) DHx ¼ (A À B) þ B{sech[C (x À 1)D ]} (5:12) or alternatively IH = A DH(P ) B = IH − DH∞ DH(P ) DH(∞) x=P Jominy Position, x FIGURE 5.39 Schematic showing the relationships between the hyperbolic secant coefficients A and B and Jominy curve characteristics. (From W.E. Jominy and A.L. Boegehold, Trans. ASM 26:574, 1938.) ß 2006 by Taylor & Francis Group, LLC. where the hyperbolic secant function for any y value is sechy ¼ 2 ey þ eÀ y (5:13) where x is the Jominy distance from the quenched end, in 1/16 in., DHx the hardness at the Jominy distance x, and A, B, C, D are the four parameters, which can be set such that DHx conforms closely to an experimental end-quench hardenability curve. The relationship between 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 quenched end. The parameter B corresponds to the difference between the upper and lower asymptotic hardness values, respectively (DH1). This means that for a constant value of A, increasing the value of B will decrease the lower asymptotic hardness, as shown in Figure 5.40a. 60 A = 50 C = 0.05 D=2 50 40 30 B = 20 20 (a) B = 10 B = 30 B variation 10 60 Hardness, HRC 50 (b) A = 50 B = 20 C = 0.05 40 D = 2.0 30 D = 0.5 D = 3.5 20 D variation at High C value 10 60 A = 50 B = 20 C = 0.05 50 D = 0.5 40 D = 2.0 D = 3.5 30 20 D variation at Low C value (c) 10 0 0 4 8 12 16 20 24 28 32 36 Distance 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 and A.L. Boegehold, Trans. ASM 26:574, 1938.) ß 2006 by Taylor & Francis Group, LLC. 60 56 C 0.2 52 Mn 0.68 Si 0.32 Ni 1.59 Cr 0.51 Mo 0.45 Grain size 8 48 Hardness, HRC 44 40 36 32 28 ID 5273 24 A = 46.83 B = 21.11 C = 0.1859 D = 0.9713 20 16 12 0 4 8 12 16 20 24 28 32 Distance from quenched end of specimen in 1/16 in. 36 FIGURE 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 in the calculated Jominy curve. If the A, B, and C parameters are constant, lowering the value of parameter D will cause the inflection point to occur at greater Jominy distances, as shown in Figure 5.40b and Figure 5.40c. A similar result will be obtained if parameters A, B, and D are kept constant, and parameter C is shown by comparing Figure 5.40b and Figure 5.40c. In fact, it may be appropriate to set C and D to a constant value characteristic of a grade of steels and 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 general expression such as Equation 5.11 or Equation 5.12. For example, if a significant amount of carbide precipitation were to occur in the bainite or pearlite cooling regime, a ‘‘hump’’ in the Jominy curve might be observed that could not be calculated. To calculate the values of the four SECH parameters for each experimentally obtained Jominy curve, the minimum requirement is a data set from which the predictive equations will be developed. This data set should contain compositions of each steel grade (or heat), with associated values of Jominy hardness at different end-quench distances, as determined by the experiment. Other metallurgical or processing variables such as grain size or austenitizing temperature can also be included. The data set must be carefully selected; the best predictions will 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 linear–nonlinear regression analysis program using least squares was used to calculate separate values of the four parameters for each experimental Jominy curve in the regression data 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 hyperbolic secant 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 ß 2006 by Taylor & Francis Group, LLC. TABLE 5.4 Multiple Regression Coefficients for Backward-Elimination Regression Analysis Dependent Variable (SECH Parameter) B A Ind. Var. C*C*C (Constant) C D Coeff. Ind. Var. Coeff. Ind. Var. Coeff. Ind. Var. Coeff. 481.27031 41.44362 Cr*Mo Mn*Si GS Ni*Ni*Ni (Constant) À28.17764 À61.55499 À1.71674 À1.35352 60.23736 Cr*Mo Mn*Si Ni*Ni*Ni C*C*C* 33.57479 À0.79950 À1.04208 À0.04871 À14.85249 0.92535 Cr*Mo Mn*Si Ni*Ni*Ni C*C*C (Constant) 1.19695 1.97624 0.09267 33.57479 À0.26580 parameters A, B, C, and D have been obtained for each heat as described above, four separate equations with these parameters as dependent variables are constructed using multiple regression 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, 31 variables were arbitrarily defined for possible selection as independent variables in the multiple regression analysis. The list of these variables consisted of all seven single-element and grain size terms, the seven squares and seven cubes of the single element and grain size terms, and all 10 possible two-way element interaction terms that did not include carbon or grain size. Based on the multiple regression coefficients from Table 5.4, the following four equations for SECH parameters were developed for the regression data set of 40 carburizing steels: A ¼ 481C3 þ 41:4 (5:14) B ¼ À28:7CrMo À 61:6MnSi À 1:72GS À 1:35Ni3 þ 60:2 (5:15) C ¼ À0:8CrMo À 1:04MnSi À 0:05Ni3 À 14:9C3 þ 0:93 (5:16) D ¼ 1:2CrMo þ 1:98MnSi þ 0:09Ni3 þ 33:6C3 À 0:27 (5:17) where an element name denotes percentage of that element in the steel and GS denotes grain size. Equation 5.14 through Equation 5.17 are valid for steel compositions in the range of 0.15–0.25% C, 0.45–1.1% Mn, 0.22–0.35% Si, 0–1.86% Ni, 0–1.03% Cr, and 0–0.76% Mo, with ASTM grain sizes (GS) between 5 and 9. After the four parameters are calculated, they are substituted into Equation 5.11 or Equation 5.12 to calculate distance hardness (DH) at each Jominy distance x of interest. To validate this method, 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] and Just [28] prediction methods). The SECH predictions were not as accurate as distance hardness predictions based on the two methods developed earlier because of the limited size and sparsely populated sections (not homogeneously distributed) of the initial data set. 5.4.2 COMPUTER CALCULATION OF JOMINY HARDENABILITY The application of computer technology has greatly enhanced the precision of these calculations. Commercial software is available for the calculation of Jominy hardness. For example, ß 2006 by Taylor & Francis Group, LLC. (a) 60 (b) 60 Processed J1 47.8 Processed J32 18.9 Inflection point 4.6 HRC at inflection point 36.2 Slope at inflection point −4.7 45 Hardness, Rc Hardness, Rc Measured Processed 30 15 45 30 15 4 8 12 16 20 24 28 32 4 Jominy depth, 1/16 in. 8 12 16 20 24 28 32 Jominy depth, 1/16 in. FIGURE 5.42 Outputs from Minitech Predictor data processing program for best fit to measured Jominy data. (a) Initial trial; (b) final trial. (From J.S. Kirkaldy and S.E. Feldman, J. Heat. Treat. 7:57–64, 1989.) the Minitech Predictor [25] is based on the initial generation of an inflection point on the Jominy curve. Figure 5.42 shows a typical output of the Minitech Predictor operating in the data processing mode. Input values are Jominy hardness values, chemical composition, and estimated grain size. The Minitech program generates a predicted Jominy curve (Jn) and a predicted inflection point distance from quenched end x’ and displays a comparison of the predicted and experimentally obtained curves as shown in Figure 5.42a. A weighting pattern Jn is accessed that specifies a weight of 1.5 for all distances from n ¼ 1 to n ¼ 2x’ and a weight of 0.75 for n > 2x’ to n ¼ 32 mm (or any limit of the data). Using an effective carbon content and grain size as adjustable parameters, the theoretical curve is then iterated about J’ and x’ to n minimize 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 data as shown in Figure 5.42b. Jominy distance (mm) Hardness (HV) 1.5 5.0 9.0 460 370 270 HV is the Vickers pyramid hardness. Calculated Jominy hardness curves are used to replace Jominy testing by equivalent predictions for those steel grades (e.g., very shallow-hardening steels) that it is difficult or impossible to test. Although the accurate prediction of hardenability is important, it is more important 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 three points within H band for SAE 862OH as shown in Figure 5.43 [29]. Using these data, the steel mill will first compare the customer’s specification against two main criteria: ß 2006 by Taylor & Francis Group, LLC. (a) (b) 500 (SAE 8620H) (SAE 8620H) 500 Customer demand Customer demand 400 300 300 200 Hardness, HV 400 200 0 10 20 30 0 10 20 30 Distance from the quenched end, mm FIGURE 5.43 (a) Customer’s specification of hardenability within an H band for SAE 8620H. (b) Jominy curve for finished heat. (From T. Lund, Carburizing Steels: Hardenability Prediction and Hardenability Control in Steel-Making, SKF Steel, Technical Report 3, 1984.) 1. That the hardenability desired is within limits for the steel grade in question 2. That the specified points fall on a Jominy curve permissible within the analysis range for SAE 862OH, i.e., the specified points must provide a physically possible Jominy curve When 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 chemical adjustments. The computer program is linked directly to the ferroalloy selection and dispensing system. By successive adjustments, the heat is refined to a chemical composition that meets the required hardenability specification within the chemical composition limits for the steel grade in question. The use of calculated Jominy curves for steel manufacturing process control is illustrated in the following example. Quality control analysis found that the steel heat should have a manganese value of 0.85%. During subsequent alloying, the analysis found 0.88% Mn. This overrun in Mn was automatically compensated for by the computer program, which adjusted hardenability by decreasing the final chromium content slightly. The resulting heat had the measured Jominy curve shown in Figure 5.43b. In this case, the produced steel does not deviate from the required specification by more than +10 HV at any Jominy distance below 19 mm. 5.5 APPLICATION OF HARDENABILITY CONCEPT FOR PREDICTION OF HARDNESS AFTER QUENCHING Jominy curves are the preferred methods for the characterization of steel. They are used to compare the hardenability of different heats of the same steel grade as a quality control method in steel production and to compare the hardenability of different steel grades when selecting steel for a certain application. In the latter case, Jominy curves are used to predict the depth of hardening, i.e., to predict the expected hardness distribution obtained after hardening parts of different cross-sectional dimensions after quenching under various conditions. Such predictions are generally based on the assumption that rates of cooling prevailing ß 2006 by Taylor & Francis Group, LLC. Cooling rate at 700 C (1300 F) 270 170 110 70 43 490 305 195 125 77 23 42 18 33 14 12 10 9 7.8 6.9 6.1 5.6 26 21.4 18 16.3 14 12.4 11 10.0 4.6 8.3 3.9 7.0 4 100 Diameter of bar, mm C/s F/s 5 125 Three-quarter radius Surface Half-radius Center 75 3 50 2 25 Diameter of bar, in. (a) 31 56 1 Quenched in water at 60 m/min (200 ft/min) 0 0 0 2 4 6 8 10 12 14 16 18 20 4.6 8.3 3.9 7.0 Equivalent distance from quenched end, 1/16 in. Cooling rate at 700 C (1300 F) 270 170 110 70 43 490 305 195 125 77 (b) 31 56 23 42 18 33 14 12 10 9 7.8 6.9 6.1 5.6 26 21.4 18 16.3 14 12.4 11 10.0 C/s F/s 5 125 Three-quarter radius Half-radius 4 75 3 Surface Center 50 2 25 Diameter of bar, in. Diameter of bar, mm 100 1 Quenched in oil at 60 m/min (200 ft /min) 0 0 0 2 4 6 8 10 12 14 16 18 20 Equivalent distance from quenched end, 1/16 in. FIGURE 5.44 Correlation of equivalent cooling rates at different distances from the quenched end of the 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 with the cooling rates prevailing at different locations on the cross sections of bars of different diameters. If the cooling rates are equal, it is assumed that equivalent microstructure and hardness can be expected after quenching. The diagrams shown in Figure 5.44 have been developed for this purpose. These diagrams provide a correlation of equivalent cooling rates at different distances from the quenched end of 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 the specified quenching conditions only. Figure 5.44a is valid only for quenching in water at an agitation rate of 1 m=s, and the diagram in Figure 5.44b is valid only for quenching in oil at an agitation rate of 1 m=s. ß 2006 by Taylor & Francis Group, LLC. in. mm 2 50 f 100 mm (4 i n.) 45 Distance below surface of bar 40 f 75 mm (3 i n.) 1½ 35 30 1 f 50 mm (2 i n.) 25 f 38 mm (11/2 i n.) 20 15 f 25 mm (1 i n.) 1/2 10 f 125 mm (1/2 i n.) 5 0 0 0 5 10 15 20 25 30 35 40 1 ½ 1½ Distance from the quenched end 45 50 mm 2 in. FIGURE 5.45 Relationship between cooling rates at different Jominy distances and cooling rates at different points below the surface of round bars of different diameters quenched in moderately agitated 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 distances and 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 same cooling rate found at a Jominy distance of 14 mm prevails at a point 2 mm below the surface of a 75-mm diameter bar, at 10 mm below the surface of 50-mm diameter bar, and at the center 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 the section after hardening. This type of diagram is also valid for only the specified quenching conditions. To correlate the hardness at different Jominy distances and the hardness at the center of round bars of different diameters that are quenched in various quenchants under different quenching conditions, the critical diameter (Dcrit), the ideal critical diameter (DI), and Grossmann’s quenching intensity factor H must be used. The theoretical background of this approach is provided by Grossmann et al. [5], who calculated the half-temperature time (the time necessary to cool to the temperature halfway between the austenitizing temperature and the temperature of the quenchant). To correlate Dcrit and H, Asimow et al. [31], in collaboration with Jominy, defined the half-temperature time characteristics for 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. If the microstructure of this steel is determined as a function of Jominy distance, the ideal critical diameter can be determined directly from the curve at that distance where 50% martensite is observed as shown in Figure 5.46. The same principle holds for Dcrit when different quenching conditions characterized by the quenching intensity factor H are involved. Figure 5.47 shows the relationship between the diameter of round bars (Dcrit and DI) and the distance from the quenched end of the Jominy specimen for the same hardness (of 50% martensite) at the center of the cross section after quenching under various conditions [31]. ß 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 2.0 7 160 Ideal critical diameter D I, in. Ideal critical diameter D I, mm 0 6 140 120 5 100 4 80 60 3 2 40 1 20 0 30 40 10 20 Distance from quenched end, mm 0 50 FIGURE 5.46 Relationship between the distance from the quenched end of the Jominy specimen and the ideal critical diameter. (From M. Asimov, W.F. Craig, and M.A. Grossmann, SAE Trans. 49(1):283–292, 1941.) The application of the Figure 5.47 diagram can be explained for the two steel grades shown 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 48 HRC. The lower part of the diagram depicts two H curves taken from the diagram in Figure 5.47. One is for vigorously agitated brine (H ¼ 5.0), and the other for moderately agitated oil (H ¼ 0.4). From both diagrams in Figure 5.48, it is seen that quenching the grade 50CrMo4 steel in vigorously agitated brine provides a hardness of 48 HRC in the center of the cross section of a round bar of 110-mm diameter. Quenching the same steel in moderately agitated oil provides this hardness at the core of round bars of only 70-mm diameter. The unalloyed grade 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 in moderately agitated oil provides this hardness in the center of a round bar of only 10 mm diameter. Diameter D crit or D I, mm 200 a 150 H ∝ 5 2 1 0.8 0.4 0.2 100 50 b c 0.02 0 0 10 20 30 40 50 60 Distance from the quenched end, mm FIGURE 5.47 Relationship between the round bar diameter and the distance from the quenched end of the Jominy specimen, giving the hardness in the center of the cross section after quenching under different quenching conditions, a, water; b, oil; c, air. (From M. Asimov, W.F. Craig, and M.A. Grossmann, SAE Trans. 49(1):283–292, 1941.) ß 2006 by Taylor & Francis Group, LLC. Hardness, HRC Diameter, mm 60 50CrM04 40 20 Ck45 48 or 45 HRC Hardness at 50 % martensite 0 150 H = 5.0 100 H = 0.4 50 0 0 10 20 30 40 50 60 Distance from quenched end, mm FIGURE 5.48 Determining the critical diameter of round bars (i.e., the hardness of 50% martensite at the center) from the Jominy hardenability curves of two steel grades quenched in vigorously agitated brine (H ¼ 5.0) and in moderately agitated oil (H ¼ 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 METHOD The diagram shown in Figure 5.47 permits the prediction of hardness only at the center of round bars. Lamont [32] developed diagrams relating the cooling rate at a given Jominy distance to that at a given fractional depth in a bar of given radius that has been subjected to a given Grossmann quenching intensity (H) factor. Analytical expressions have been developed for the Lamont transformation of the data to the appropriate Jominy distance J: J ¼ 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 ¼ 0 at the center; r=R ¼ 1 at the surface), and H the Grossmann quenching intensity factor. These expressions [33] are valid for any value of H from 0.2 to 10 and for bar diameters up to 200 mm (8 in.). Lamont developed diagrams for the following points and fractional depths on the cross section of round bars: r=R ¼ 0 (center), r=R ¼ 0.1, r=R ¼ 0.2, . . . , r=R ¼ 0.5 (half-radius), r=R ¼ 0.6, . . . , r=R ¼ 1.0 (surface). Each of these diagrams is always used in connection with Jominy hardenability curve for the relevant steel. Figure 5.49 through Figure 5.51 show the Lamont diagram for r=R ¼ 0 (center of the cross section), r=R ¼ 0.5, and r=R ¼ 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 hardness at 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 a hardness of 55 HRC at a Jominy distance of 10 mm, then the maximum diameter of the bar that will achieve this hardness at half-radius when quenched in oil with H ¼ 0.35 will be 28 mm. This result is obtained by using the diagram in Figure 5.50 for r=R ¼ 0.5 and taking the vertical line at a Jominy distance of 10 mm to the intersection with the curve for H ¼ 0.35, giving the value of 28 mm on the ordinate. ß 2006 by Taylor & Francis Group, LLC. 0 10 15 20 30 37.5 40 50 mm 1.0 in. Bar diameter, mm 140 120 ∞ 5.0 2.0 1.5 1.0 0.70 r =0 R 6.0 r 5.0 0.50 R 100 80 60 50 40 20 0.35 4.0 0.20 3.0 2.0 Quenching intensity H 160 1.0 0 ¼ 0 ½ ¾ 1 1¼ 1½ 1¾ Distance from the quenched end, in. 2 FIGURE 5.49 Relation between distance from the quenched end of Jominy specimen and bar diameter for the ratio r=R ¼ 0, i.e., the center of the cross section, for different quenching intensities. (From J.L. Lamont, Iron Age 152:64–70, 1943.) 2. To determine the hardness at a specified location when the diameter of the bar, the quenching intensity H, and the steel grade are known. For example, if a 120-mm diameter bar is quenched in still water (H ¼ 1.0), the hardness at the center (r=R ¼ 0) will be determined at a distance of 37.5 mm from the quenched end on the Jominy curve of the relevant steel grade (see Figure 5.49). 220 200 Bar diameter 180 160 in. 0 10.0 10 20 30 40 50 mm r = 0.5 R 9.0 8.0 7.0 ∞ 5.0 2.0 1.5 1.0 0.70 r R 6.0 140 0.5 120 5.0 100 4.0 80 Quenching intensity H mm 240 3.0 0.35 0.2 60 40 28 20 0 2.0 1.0 0 ½ 1 1½ Distance from the quenched end, in. 2 FIGURE 5.50 Relation between distance from the quenched end of Jominy specimen and bar diameter for the ratio r=R ¼ 0.5, i.e., 50% from the center, for different quenching intensities. (From J.L. Lamont, Iron Age 152:64–70, 1943.) ß 2006 by Taylor & Francis Group, LLC. mm 240 220 in. 10.0 9.0 8.0 200 7.0 160 30 50 mm 0.6 ∞ 5 2 1.5 6.0 1.0 Bar diameter 40 140 0.7 0.5 5.0 0.35 120 100 80 76 60 Quenching intensity H 180 10 15 20 4.0 0.2 3.0 2.0 40 20 0 1.0 0 r = 0.8 R r R 1/2 1 11/2 Distance from the quenched end, in. 2 FIGURE 5.51 Relation between distance from the quenched end of Jominy specimen and bar diameter for the ratio r=R ¼ 0.8, i.e., 80% from the center, for different quenching intensities. (From J.L. Lamont, Iron Age 152:64–70, 1943.) 3. To select adequate quenching conditions when the steel grade, the bar diameter, and the location on the cross section where a particular hardness should be attained are known. For example, a hardness of 50 HRC, which corresponds to the distance of 15 mm from the quenched end on the Jominy curve of the relevant steel grade, should be attained at the center of a 50-mm diameter bar. The appropriate H factor can be found by using Figure 5.49. In this case, the horizontal line for a 50-mm diameter and the vertical line for a 15-mm Jominy distance intersect at the point that corresponds to H ¼ 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 appropriate transformation diagram. For example, if 50 HRC hardness, which corresponds to a distance of 15 mm from the quenched end on the Jominy curve of the relevant steel grade, should be attained at 7.6 mm below the surface of a 76-mm diameter bar, then r 38 À 7:6 ¼ ¼ 0: 8 R 38 (5:19) This calculation indicates that the diagram for r=R ¼ 0.8 (Figure 5.51) should be used. In this case, the horizontal line for 76 mm diameter intersects the vertical line for 15-mm Jominy distance on the interpolated curve H ¼ 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 when the bar diameter and steel grade and its Jominy curve and quenching intensity H are ß 2006 by Taylor & Francis Group, LLC. known. For this calculation, diagrams for every ratio r=R from the center to the surface should be used. The following procedure should be repeated with every diagram. At the point where the horizontal line (indicating the bar diameter in question) intersects the relevant H curve, the vertical line gives the corresponding distance from the quenched end on the Jominy curve from which the corresponding hardness can be read and plotted at the corresponding fractional depth. Because some simplifying assumptions 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. 5.5.2 STEEL SELECTION BASED ON HARDENABILITY The selection of a steel grade (and heat) for a part to be heat-treated depends on the hardenability that will yield the required hardness at the specified point of the cross section after 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 to those steel grades with known hardenability bands or Jominy curves. This is true first of all for structural steels for hardening and tempering and also for steels for case hardening (to determine core hardenability). If the diameter of a shaft and the bending fatigue stresses it must be able to undergo are known, engineering analysis will yield the minimum hardness at a particular point on the cross section that must be achieved by hardening and tempering. Engineering analysis may show that distortion minimization requires a less severe quenchant, e.g., oil. Adequate toughness after tempering (because the part may also be subject to impact loading) may require 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 the required hardness after tempering. This is done by using a diagram such as the one shown in Figure 5.52. For example, if a hardness of 35 HRC is required after hardening and then tempering at 5008C at the critical cross-sectional diameter, the minimum hardness after quenching must be 45 HRC. As-quenched hardness, HRC 60 55 Tempered 600 C/ 60 min 50 45 40 35 Tempered 500 C/ 60 min 30 25 15 20 25 30 35 40 Tempered hardness, HRC 45 FIGURE 5.52 Correlation between the hardness after tempering and the hardness after quenching for structural steels (according to DIN 17200). ß 2006 by Taylor & Francis Group, LLC. Hardness, HRC 70 %C 0.6 0.5 0.4 0.3 60 50 0.2 40 30 30 40 50 60 70 80 Portion of martensite % 90 100 FIGURE 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. 473–474, 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 correlates hardness 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 critical position 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 a particular 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 is 45 mm in diameter and that the critical point on the cross section (which was determined from engineering analysis of resultant stresses) is three fourths of the radius. To determine if a particular steel grade, e.g., AISI 4140H, will satisfy the requirement of 45 HRC at (3=4)R after oil quenching, the diagram shown in Figure 5.54a should be used. This diagram correlates cooling rates along the Jominy end-quench specimen and at four characteristic locations (critical points) on the cross section of a round bar when quenched in oil at 1 m=s agitation rate (see the introduction to Section 5.5 and Figure 5.44). Figure 5.54a shows that at (3=4)R the shaft having a diameter of 45 mm will exhibit the hardness that corresponds to the hardness at a distance of 6.5=16 in. (13=32 in.) from the quenched end of the Jominy specimen. Step 3. Determine whether the steel grade represented by its hardenability band (or a certain heat represented by its Jominy hardenability curve) at the specified distance from the quenched end exhibits the required hardness. As indicated in Figure 5.54b, the minimum hardenability curve for AISI 4140H will give a hardness of 49 HRC. This means that AISI 4140H has, in every case, enough hardenability for use in the shaft example above. This graphical method for steel selection based on hardenability, published in 1952 by Weinmann and coworkers, can be used as an approximation. Its limitation is that the diagram shown in Figure 5.54a provides no information on the quality of the quenching oil and its temperature. Such diagrams should actually be prepared experimentally for the exact conditions that will be encountered in the quenching bath in the workshop; the approximation will be valid only for that bath. 5.5.3 COMPUTER-AIDED STEEL SELECTION BASED ON HARDENABILITY As in other fields, computer technology has made it possible to improve the steel selection process, making it quicker, more intuitive, and even more precise. One example, using a ß 2006 by Taylor & Francis Group, LLC. Diameter of bar, mm (a) 125 Three-quarter radius Half-radius 100 Surface 75 50 45 Center 25 Quenched in oil at 1 m/s 0 0 2 4 6 8 10 12 14 16 Distance from quenched end, 1/16 in. 18 20 49 HRC minimum hardenability of 4140H meets the requirement Hardness, HRC (b) 70 4140H 60 50 40 30 0 2 4 6 8 10 12 14 16 Distance from quenched end, 1/16 in. 18 20 FIGURE 5.54 Selecting a steel of adequate hardenability. (a) equivalent cooling rates (and hardness after quenching) for characteristic points on a round bar’s 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. 473–474, p. 493.) software package developed at the University of Zagreb [35], is based on a computer file of experimentally determined hardenability bands of steels used in the heat-treating shop. The method is valid for round bars of 20–90-mm diameters. The formulas used for calculation of equidistant locations on the Jominy curve, described in Ref. [23], were established through regression analysis for this range of diameters. The essential feature of this method is the calculation of points on the optimum Jominy hardenability 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 cross section [(3=4)R, (1=2)R, (1=4)R, or center]. The input data for the computer-aided selection process are the following: . . . . . Diameter of the bar (D mm) Surface hardness HRC Hardness at a critical point HRC Quenching intensity factor I (I equals the Grossmann quenching intensity factor H as given in Table 5.1) Minimum percentage of martensite required at the critical point The first step is to calculate the equidistant locations from the quenched end on the Jominy curve (or Jominy hardenability band). These equidistant locations are the points on the Jominy curve that yield the required as-quenched hardness. The calculations are performed as follows [23]: ß 2006 by Taylor & Francis Group, LLC. On the surface: Es ¼ D0:718 5:11I 1:28 (5:20) At (3=4)R: E3=4R ¼ D1:05 8:62I 0:668 (5:21) E1=2R ¼ D1:16 9:45I 0:51 (5:22) E1=4R ¼ D1:14 7:7I 0:44 (5:23) At (1=2)R: At (1=4)R: At the center: Ec ¼ D1:18 8:29I 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 of the critical points (Ecrit), using the hardenability band of the relevant steel, the hardness values achievable with the Jominy curve of the lowest hardenability (Hlow) and the hardness values achievable with the Jominy curve of the highest hardenability (Hhigh) for both Es and 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 with the steel in question: S¼ H Hmax (5:25) H high s H high HRC crit H low s H low s 0 Es E crit Jominy distance, mm FIGURE 5.55 Determination of minimum and maximum hardness for equidistant location Es and Ecrit from a relevant hardenability band. (After T. Filetin, Strojarstvo 24(2):75–81, 1982 [in Croatian].) ß 2006 by Taylor & Francis Group, LLC. TABLE 5.5 Correlation between Degree of Hardening S and Percentage of Martensite in As-Quenched Structure Percent Martensite Degree of Hardening S 50–60 60–70 70–80 80–85 85–90 90–95 95–97 97–100 0.70–0.74 0.74–0.76 0.76–0.78 0.78–0.81 0.81–0.86 0.86–0.91 0.91–0.95 0.95–1.00 Source: T. Filetin and J. Galinec, Software programme for steel selection 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 of the hardenability band, taking the value for Hmax from the relevant Jominy curve at distance 0 from the quenched end (J ¼ 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 the as-quenched structure as shown in Table 5.5. It is also possible to determine whether the required percentage of martensite can be achieved 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 of hardening) may be given. For statically stressed parts, S < 0.7; for less dynamically stressed parts, 0.7 < S < 0.86; and for highly dynamically stressed parts, 0.86 < S < 1.0. In this way, a direct comparison of the required S value with values calculated for both Jominy curves at the Ecrit 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 of the hardenability band (Slower). In this case all heats of this steel will satisfy the requirement. 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 of the hardenability band (Supper). In this case, none of the heats of this steel can satisfy the requirement. This steel must not be selected because its hardenability is too low for the case in question. 3. The value of required degree of hardening (S) is somewhere between the values for the degree of hardening achievable with the upper and lower curves of the hardenability band (Supper and Slower, respectively). In the third case, the position of the S required, designated as X, is calculated according to the formula: X¼ ß 2006 by Taylor & Francis Group, LLC. S À Slower Supper À Slower where X is the distance from the lower curve of the hardenability band on the ordinate Ecrit to the actual position of S required, which should be on the optimum Jominy curve. This calculation divides the hardenability band into three zones: The lower third, X 0.33 The middle third, 0.33 < X The upper third, 0.66 < X 0.66 All heats of a steel grade where the Jominy curves pass through the zone in which the required S point is situated can be selected as heats of adequate hardenability. This zone is indicated in a graphical presentation of the method. Once the distance X is known, the optimum Jominy hardenability curve can be drawn. The only requirement is that for every distance from the quenched end the same calculated ratio (X) that indicates the same position of the Jominy curve relative to the lower and upper hardenability curves of the hardenability band is maintained. The following example illustrates the use of this method in selecting a steel grade for hardening and tempering. A 40-mm diameter shaft after hardening and tempering should exhibit a surface hardness of Hs ¼ 28 HRC and a core hardness of Hc ¼ 26 HRC. The part is exposed to high dynamic stresses. 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 (in this case the core) as shown in Figure 5.56. Next, the required percentage of martensite at the critical point after quenching (in this case 95%, because of high dynamic stresses) and the quenching 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 for which the hardenability band is stored in the file and presents the results on the screen as shown in Figure 5.57. This is a list of all stored steel grades regarding suitability for the application being calculated. Acceptable steel grades, suitable from the upper, middle, or lower third of the hardenability band, and unacceptable steel grades with excessively high hardenability are determined. Selection of steel in hardened and tempered condition Diameter, mm (0–90):40 Critical point on the cross-section: Core Required value: <1> 3/4R <2> 1/2R <3> 1/4R <4> Core <1> Hardness, HRC (20–50) <2> Tensile strength, N/mm2 (750–1650) Hardness, HRC – On the surface: – At the critical point: 28 H s 26 H c FIGURE 5.56 Input data for computer program. (From T. Filetin and J. Galinec, Software programme for steel selection based on hardenability, Faculty of Mechanical Engineering, University of Zagreb, 1994.) ß 2006 by Taylor & Francis Group, LLC. Results of steel selection JUS C4181 C4730 C4731 C4781 C4732 C4782 C4733 C4738 C4734 AISI 4130 E4132 4140 4150 Not suitable Not suitable Suitable heats from upper third of band Suitable heats from upper third of band Suitable heats from middle third of band Suitable heats from middle third of band Suitable heats from middle third of band Too high hardenability Too high hardenability FIGURE 5.57 List of computer results. (From T. Filetin and J. Galinec, Software programme for steel selection 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 be obtained. This gives the optimum Jominy hardenability curve for the case required and indicates the desired zone of the hardenability band. In addition, the necessary tempering temperature can be calculated according to the formula: Ttemp ¼ 917 vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u À8 uln ¼ Hcrit 6 À8 t Htemp (5:26) S where Ttemp is the absolute tempering temperature (K) (valid for 4008C < Ttemp < 6608C), Hcrit the hardness after quenching at the critical point HRC (taken from the optimum Jominy curve at the distance for the critical point), Htemp the required hardness after tempering at the critical point HRC, and S the degree of hardening (ratio between hardness on the optimum Jominy curve at the distance Ecrit and at the distance E ¼ 0). 70 Hardness, HRC 60 50 40 30 20 10 0 5 E s 10 E crit15 20 25 30 35 40 Jominy distance, mm FIGURE 5.58 Graphical presentation of the optimum Jominy hardenability curve. (From T. Filetin and J. Galinec, Software programme for steel selection based on hardenability, Faculty of Mechanical Engineering, University of Zagreb, 1994.) ß 2006 by Taylor & Francis Group, LLC. Tensile strength (Rm, N=mm2 ) is also calculated at the relevant points using the formula: Rm ¼ 0:426H 2 þ 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 ¼ (0:8 þ 0:1S )Rm þ 170S À 200 [N=mm2 ] (5:28) A5 ¼ 0:46 À (0:0004 À 0:00012S )Rm [%] (5:29) Z ¼ 0:96 À (0:00062 À 0:00029S )Rm [%] (5:30) Rd ¼ (0:25 þ 0:45Z )Rm [N=mm2 ] (5:31) KU ¼ [460 À (0:59 À 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 found suitable, the mechanical properties for the surface and for the critical cross section point can be calculated. The computer output is shown in Figure 5.59. Compared to the previous steel selection processes, these computer-aided calculations have the following advantages: (AISI 4140) Heats from the middle third of the band Calculated tempering temperature: 643 C Mechanical properties Yield strength: Rp0.2, N/mm2 Tensile strength: Rm, N/mm2 Bending fatigue strength: Rd, N/mm2 Elongation: A5, % Contraction: Z, % Impact engergy: KU, J Surface 793 920 499 20 65 125 Critical point 735 874 474 21 65 123 FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. 1. Whereas the previously described graphical method is valid for only one specified quenching 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 more precise. 3. Calculations of the exact tempering temperature and all mechanical properties after tempering at the critical point that give much more information and facilitate the steel selection are possible. 5.6 5.6.1 HARDENABILITY IN HEAT TREATMENT PRACTICE HARDENABILITY OF CARBURIZED STEELS Carburized 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 case hardness improves the fatigue durability. Historically, it was thought that core hardenability was required for the selection of carburizing steels and heat treatment of carburized parts and that 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; therefore the historical view of core hardenability may not be correct. In fact, hardenability of both case and core is essential for proper selection of the optimum steel grade and the heat treatment of carburized parts. It is now also known that the method of quenching after carburizing, i.e., direct quenching or reheat and quench, influences case hardenability. The case hardenability of carburized steel is determined by using the Jominy end-quench test. Standard Jominy specimens are carburized in a carburizing medium with a high C potential for sufficient time to obtain a carburized layer of the desired depth. In addition to the Jominy specimens, two bars of the same steel and heat, the same surface finish, and the same dimensions (25 mm diameter) are also carburized under identical conditions. These bars are used to plot the carbon gradient curve shown in Figure 5.60a, which is produced by chemical analysis of chips obtained from machining of the carburized layer at different layer thicknesses. In this way, as shown in Figure 5.60a, the following carbon contents were found as a function of case depth: 1.2 Measured carbon content curve carburized at 925 C for 4.5 h 1.0 d1 = 0.2 0.7% C d4 = 0.57 %C 0.8 1.0% C 0.8% C 0.6 d3 = 0.45 0.4 0.2 (a) d1 0 d2 d3 d2 = 0.32 0.9% C d4 0.2 0.4 0.6 0.8 1.0 Depth from surface, mm 1.2 (b) FIGURE 5.60 (a) Measured carbon gradient curve after gas carburizing at 9258C for 4.5 h. (b) Grinding ˇˇ ˇ of the carburized Jominy specimen. (From T. Filetin and B. Liscic, Strojarstvo 18(4):197–200, 1976 [in Croatian].) ß 2006 by Taylor & Francis Group, LLC. 1.0% C at 0.2 mm depth (distance from the surface of the bar)—d1 0.9% C at 0.32 mm depth—d2 0.8% C at 0.45 mm depth—d3 0.7% C at 0.57 mm depth—d4 One of the carburized Jominy specimens should be end-quenched in the standard way using the Jominy apparatus directly from the carburizing temperature (direct quenching), and the other should be first cooled to room temperature and then reheated and quenched from a temperature 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 the depths, d1, d2, d3, and d4, as shown in Figure 5.60b. Hardness is measured in the standard way on each of the ground surfaces, and the corresponding Jominy curves are plotted. Figure 5.61a Direct quenching from 925 C 900 600 55 500 50 HV 800 HRC 700 0.9 67 1.0 65 0.7 0.8 % C 60 45 400 300 0.17 200 40 35 30 25 20 100 0 0 2 4 6 8 10 121416 20 (a) 24 28 32 36 40 44 48 Jominy distance, mm Indirect quenching from 820 C 900 67 0.7 65 700 0.9 %C 60 HV 600 55 500 HRC 0.8 800 50 1.0 400 45 40 35 30 25 20 300 200 100 0 0 2 4 6 8 10 121416 20 (b) 24 28 32 36 40 44 48 Jominy distance, mm FIGURE 5.61 Jominy case hardenability curves of carburized DIN 16MnCr5 steel (a) after direct quenching from 9258C and (b) after reheating followed by quenching from 8208C. (From T. Filetin ˇˇ ˇ and B. Liscic, Strojarstvo 18(4):197–200, 1976 [in Croatian].) ß 2006 by Taylor & Francis Group, LLC. provides an example of Jominy hardenability curves for the carburizing steel grade DIN 16MnCr5 (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 the carburizing temperature, 9258C. Figure 5.61b provides Jominy curves for the same carburized case after indirect quenching (reheated to 8208C). From both diagrams of Figure 5.61 the following conclusions can be drawn: 1. The hardenability of the core is substantially different from the hardenability within the carburized case. 2. The best hardenability of the carburized case is found for this steel grade at 0.9% C with direct quenching and at 0.8% C with indirect quenching (reheat and quench). Consequently, the carburizing process should be controlled so that after carburizing a surface carbon content of 0.9% is obtained for direct quenching and one of 0.8% for indirect quenching. 5.6.2 HARDENABILITY OF SURFACE LAYERS WHEN SHORT-TIME HEATING METHODS ARE USED When short-time (zero time) heating processes for surface hardening are used, e.g., flame hardening, induction hardening, or laser hardening, the same metallurgical reactions occur as in conventional hardening except that the heating processes cycle must be much shorter than that of conventional hardening. Heating time for these proceses vary by one to three orders of magnitude; approximately 100 s for flame hardening, 10 s or less for induction hardening, and 1 s or less for laser hardening. This means that the heating rates are very high. Problems associated 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 does not occur between normal temperatures Ac1 and Ac3 as in conventional hardening because the high heating rate produces nonequilibrium systems. The Ac1 and Ac3 temperatures are displaced to higher temperatures as shown in Figure 5.62. Although an austenitizing temperature may be sufficiently high to form austenite under slow heating conditions (conventional hardening), the same temperature level may not be sufficient to even initiate austenization under high heating rates [37]. Therefore, substantially higher austenitizing temperatures are used with flame, induction, and laser hardening (especially the latter) than for conventional hardening of the same steel. 2. For quench hardening, the austenitization must dissolve and uniformly distribute the carbon of the carbides in the steel. This is a time-dependent diffusion process (sometimes called homogenization), even at the high temperatures used in short-time heating methods. At very high heating rates, there is insufficient time for diffusion of carbon atoms from positions of higher concentrations near carbides to the positions of lower concentrations (areas that originated from practically carbon-free ferrite). This diffusion depends 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 are undesirable in this regard. Tempered martensite, having small and finely dispersed carbides, provides the shortest paths for carbon diffusion and is therefore most desirable. Figure 5.62a illustrates a time temperature transformation diagram for continuous heating at different heating rates when austenitizing an unalloyed steel with 0.7% C with a starting structure of ferrite and lamellar pearlite. Figure 5.62b shows a similar diagram for ß 2006 by Taylor & Francis Group, LLC. 900 Austenite 880 Temperature, C 860 840 820 Austenite + carbide 800 780 Au 760 ste 740 720 (a) + fer r it e + c a r Ac3 bid e Ac1 Ferrite + pearite 700 680 0.1 n it e 1 10 102 Time, s 900 103 104 Austenite 880 Temperature, C 860 Acm 840 820 Austenite + carbide 800 780 Au 760 740 720 (b) n it e Ac1e + f e r rit e + c a r b i d e Ac1b Tempered martensite 700 680 10−1 ste 100 101 102 Time, s 103 104 105 FIGURE 5.62 Time temperature transformation diagram for continuing heating with different heating rates, 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 heating methods are involved, Der Peddinghaus Erfahrungsaustausch, Gevelsberg, 1957, pp. 13–19 [in German].) a starting structure of tempered martensite. A comparison of the two diagrams illustrates the influence of starting structure on the austenitizing process. Whereas for the ferrite–pearlite starting structure at maximum heating rate the upper transformation temperature Ac3 is 8658C, 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 better hardenability than the austenite of a pearlite–ferrite starting structure. The practical consequence of this is that prior to surface hardening by any short-time heating process, if the steel is in the hardened and tempered condition, maximum hardened case depths are possible. If the annealed material has a coarse lamellar structure, or even worse, globular carbides, minimum hardening depths are to be expected. 5.6.3 EFFECT OF DELAYED QUENCHING ON THE HARDNESS DISTRIBUTION Delayed quenching processes have been known for a long time. Delayed quenching means that austenitized parts are first cooled slowly and then after a specified time they are quenched at a much faster cooling rate. Delayed quenching is actually a quenching process in which a discontinuous change in cooling rate occurs. In some circumstances, depending on steel ß 2006 by Taylor & Francis Group, LLC. AISI 4140 Batch No.73456 13 11 16 17 15 50 55 12 14 50 HRC HRC 55 1 45 45 R 3/4R 1/2R 1/4R 0 1/4R 1/2R 3/4R R 50 mm Diameter FIGURE 5.63 Measured hardness distribution on the cross section of 50 mm diameter  200 mm bars ˇˇ ´ 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 heat transfer data. ASM Quenching and Distortion Control, Proceedings of First International Confererence On Quenching and Control of Distortion, Chicago, IL, 22–25 Sept. 1992, pp. 17–26.) hardenability and section size, the hardness distribution in the cross section after delayed quenching does not have a normal trend (normally the hardness decreases continuously from the surface toward the core) but instead exhibits an inverse trend (the hardness increases from the surface toward the core). This inverse hardness distribution is a consequence of the discontinuous change in the cooling rate and related to the incubation period (at different points in the cross section) before changing the cooling rate. This process has been explained theoretically by Shimizu and Tamura [40,41] in Figure 5.63. In every experiment, the delay in quenching was measured as the time from immersion to the moment when maximum heat flux density on the surface (tqmax) occurred. As shown in Figure 5.63 and Table 5.6 for AISI 4140 steel with a section 50 mm in diameter, when the delay 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 inverse or inverse to normal hardness distribution was obtained. In experiments where tqmax was less than 15 s, a normal hardness distribution resulted. Besides the inherent hardenability of a steel, delayed quenching may substantially increase the depth of hardening and may compensate for lower hardenability of the steel [39]. Interestingly, none of the available software programs for predicting as-quenched hardness simulates the inverse hardness distribution because they do not account for the length of the incubation period before the discontinuous change in cooling rate at different points in the cross section. 5.6.4 A COMPUTER-AIDED METHOD TO PREDICT THE HARDNESS DISTRIBUTION AFTER QUENCHING BASED ON JOMINY HARDENABILITY CURVES The objective here is to describe one method of computer-aided calculation of hardness distribution. This method, developed at the University of Zagreb [44], is based on the Jominy ß 2006 by Taylor & Francis Group, LLC. TABLE 5.6 Time from Immersion (tqmax) until Maximum Heat Flux Density under Various Quenching Conditions for AISI 4140 Bars (50 mm Diameter 3 200 mm)a Figure 63 Curve No. 1 11 12 13 14 15 16 17 Quenching Conditions tqmax (s) Mineral oil at 208C, without agitation Polymer solution (PAG) 5%; 408C; 0.8 m=s Polymer solution (PAG) 15%; 408C; 0.8 m=s Polymer solution (PAG) 25%; 408C; 0.8 m=s Polymer solution (PAG) 20%; 358C; 1 m=s Polymer solution (PAG) 10%; 358C; 1 m=s Polymer solution (PAG) 5%; 358C; 1 m=s Polymer solution (PAG) 20%; 358C; 1 m=s 14 16 33 70 30 12 13 47 a See Figure 63. ˇˇ ´ Source: B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation and calculation of heat transfer data. ASM Quenching and Distortion Control, Proceedings of First International Confererence On Quenching and Control of Distortion, Chicago, IL, 22–25 Sept. 1992, pp. 17–26. hardenability curves. Jominy hardenability data for steel grades of interest are stored in a databank. In this method, calculations are valid for cylindrical bars 20–90 mm in diameter. Figure 5.64 shows the flow diagram of the program, and Figure 5.65 is a schematic of the step-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 by quenching it under specific conditions. Step 3. Measure the hardness (HRC) on the specimen’s cross section in the middle of the length. Step 4. Store in the file the hardness values for five characteristic points on the specimen’s cross section (surface, (3=4)R, (1=2)R, (1=4)R, and center). If the databank already contains the 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 on the Jominy curve (Es, E3=4R, E1=2R, E1=4R, Ec) that have the same hardness values as those measured at the characteristic points on the specimen’s 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 specimen’s diameter Dspec and on known E values: " D0:718 spec Is ¼ 5:11Es " I3=4R ¼ ß 2006 by Taylor & Francis Group, LLC. #0:78 D1:05 spec 8:62E3=4R (5:33) #1:495 (5:34) Start Input: Steel grade, quenching conditions Database—stored data into files: – Jominy hardenability Search in database—files Additional experiments yes – Quenching intensity recorded as functions: T = f (t ), q = f (t ), q = f (Ts) – Measuring of quenching intensity no All data available? – Hardness distribution on the test specimens cross section – Hardening of test specimen Input parameters: Steel grade; quenching medium and conditions; Jominy hardenability data; hardness on the test specimens cross section Reading of the corresponding Jominy distances (Ei ) Calculation of the "hypothetical quenching intensity" within the test specimens cross section Ii = f (Dsp, Ei) Input of the actual diameter, D Calculation of Jominy distances corresponding to the diameter, D (E'i = f (D, li )) Reading of the hardness data from the Jominy hardenability curve for Jominy distances corresponding to: E'S, E'3R/4, E'R/2, E'R/4, E'C Results obtained: Hardness data in five points on the cross section of the bar. Hardness curve, graphically yes Another diameter? no yes Another steel grade and/or quenching conditions no Stop FIGURE 5.64 Flowchart of computer-aided prediction of hardness distribution on cross section of ˇˇ ´ quenched round bars. (From B. Liscic, H.M. Tensi, and W. Luty, Theory and Technology of Quenching, Springer-Verlag, Berlin, 1992.) " I1=2R D1:16 spec ¼ 9:45E1=2R " I1=4R ¼ ß 2006 by Taylor & Francis Group, LLC. D1:14 spec 7:7E1=4R #1:495 (5:35) #2:27 (5:36) Hardenability ß 2006 by Taylor & Francis Group, LLC. HRC HRC HRC Actual diameter D Hs Measured H 3R/4 hardness H Jominy curve for the relevant steel grade R /2 HR /4 HC 120 H's C R/4 R/2 3R/4 Predicted hardness distribution S H'c Test specimen 50-mm diameter Es Ec E3R/4 ER/2 ER/4 E's Distance from quenched end E'c Distance from E' quenched end E'3R /4 E'R /2 R /4 lx Step 1 to step 4 Step 5 Step 6 Step 8 C R /2 S R /4 3 R /4 Step 9 ˇˇ ´ 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 and Technology of Quenching, Springer-Verlag, Berlin, 1992.) 271 " Ic ¼ D1:18 spec #2:27 8:29Ec (5:37) Equatoane, and D.H. Breen, The Hardenability of Steels, ASM International, Cleveland, OH, 1997. 9. C.F. Jatczak, Hardenability in high carbon steel, Metall. Trans. 4:2267–2277 (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). ß 2006 by Taylor & Francis Group, LLC. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. These 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. 473–474. R.A. Grange, Estimating the hardenability of carbon steels, Metall. Trans. 4:2231 (1973). ¨ A. Rose and L. Rademacher, Weiterentwicklung des Stirnabschreckversuches zur Prufung der ¨ ¨ Harbarkeit von tiefer einhartenden Stahlen, Stahl Eisen 76(23):1570–1573 (1956) (in German). C.F. Jatczak, Effect of microstructure and cooling rate on secondary hardening of Cr–Mo–V steels, Trans. ASM 58:195–209 (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. ˇˇ ´ 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):85–100 (1968). ¨ ¨ ¨ R. Caspari, H. Gulden, K. Krieger, D. Lepper, A. Lubben, H. Rohloff, P. Schuler, V. Schuler, and ¨ ¨ H.J. Wieland, Errechnung der Hartbarkeit im Stirnabschreckversuch bei Einsatz und Vergutungs¨ ¨ stahlen, Harterei Tech. Mitt. 47(3):183–188 (1992). J.S. Kirkaldy and S.E. Feldman, Optimization of steel hardenability control, J. Heat. Treat. 7:57–64 (1989). J.M. Tartaglia, G.T. Eldis, and J.J. Geissler, Hyperbolic secant method for predicting Jominy hardenability; an example using 0.2 C–Ni–Cr–Mo steels, J. Heat. Treat. 4(4):352–364 (1986). J.M. Tartgalia and G.T. Eldis, Metall. Trans. 15A(6):1173–1183 (1984). E. Just, Met. Prog. 96(5):87–88 (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 quenched round bars, SAE Trans. 49(1):283–292 (1941). J.L. Lamont, How to estimate hardening depth in bars, Iron Age 152:64–70 (1943). D.V. Doane and J.S. Kirkaldy (Eds.), Hardenability Concepts with Applications to Steel, Proceedings of a Symposium, Chicago, Oct. 24–26, 1977, The Metallurgical Society of AIME, 1978. T. Filetin, A method of selecting hardenable steels based on hardenability, Strojarstvo 24(2): 75–81 (1982) (in Croatian). T. Filetin and J. Galinec, Software programme for steel selection based on hardenability, Faculty of Mechanical Engineering, University of Zagreb, 1994. ˇˇ ´ T. Filetin and B. Liscic, Determining hardenability of carburizing steels, Strojarstvo 18(4):197–200 (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 Peddinghaus Erfahrungsaustausch, Gevelsberg, 1957, pp. 13–19 (in German). ˇˇ ´ B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation and calculation of heat transfer data. ASM Quenching and Distortion Control, Proceedings of First International Confererence On Quenching and Control of Distortion, Chicago, IL, 22–25 Sept. 1992, pp. 17–26. N. Shimizu and I. Tamura, Effect of discontinuous change in cooling rate during continuous cooling on pearlitic transformation behavior of steel, Trans. ISIJ 17:469–476 (1977). N. Shimizu and I. Tamura, An examination of the relation between quench-hardening behavior of steel and cooling curve in oil, Trans. ISIJ 18:445–450 (1978). E. Just, Hardening and tempering—influencing steel by hardening, VDI Ber. 256:125–140 (1976) (in German). W. Gerber and U. Wyss, Hardenability and ability for hardening and tempering of steels, Von Roll Mitt. 7(2=3):13–49 (1948) (in German). ˇˇ ´ B. Liscic and T. Filetin, Computer-aided evaluation of quenching intensity and prediction of hardness distribution, J. Heat. Treat. 5(2):115–124 (1988). ß 2006 by Taylor & Francis Group, LLC. 6 Steel Heat Treatment ˇc ´ ˇidar Lis ˇic Boz CONTENTS 6.1 6.2 6.3 Fundamentals of Heat Treatment.............................................................................. 278 6.1.1 Heat Transfer .................................................................................................. 278 6.1.2 Lattice Defects................................................................................................. 285 6.1.3 Application of TTT (IT) and CCT Diagrams ................................................. 287 6.1.3.1 Isothermal Transformation Diagram ................................................ 287 6.1.3.2 Continuous Cooling Transformation Diagram................................. 288 6.1.3.3 Heat Treatment Processes for Which an IT or CCT Diagram May Be Used ..................................................................... 292 6.1.3.4 Using the CCT Diagram to Predict Structural Constituents and Hardness upon Quenching Real Workpieces ............................. 293 6.1.3.5 Special Cases and Limitations in the Use of CCT Diagrams............ 298 6.1.4 Oxidation......................................................................................................... 301 6.1.4.1 Scaling of Steel .................................................................................. 304 6.1.5 Decarburization ............................................................................................... 306 6.1.5.1 The Effect of Alloying Elements on Decarburization ....................... 308 6.1.5.2 Definitions and Measurement of Decarburization............................ 309 6.1.6 Residual Stresses, Dimensional Changes, and Distortion ............................... 312 6.1.6.1 Thermal Stresses in the Case of Ideal Linear-Elastic Deformation Behavior ...................................................................... 314 6.1.6.2 Transformational Stresses ................................................................. 315 6.1.6.3 Residual Stresses when Quenching Cylinders with Real Elastic–Plastic Deformation Behavior............................................... 317 6.1.6.4 Dimensional Changes and Distortion during Hardening and Tempering .................................................................................. 324 Annealing Processes ................................................................................................... 330 6.2.1 Stress-Relief Annealing.................................................................................... 330 6.2.2 Normalizing..................................................................................................... 334 6.2.3 Isothermal Annealing ...................................................................................... 339 6.2.4 Soft Annealing (Spheroidizing Annealing) ...................................................... 344 6.2.5 Recrystallization Annealing............................................................................. 349 6.2.5.1 Grain Recovery ................................................................................. 351 6.2.5.2 Polygonization................................................................................... 351 6.2.5.3 Recrystallization and Grain Growth................................................. 352 Hardening by Formation of Martensite .................................................................... 355 6.3.1 Austenitizing.................................................................................................... 355 6.3.1.1 Metallurgical Aspects of Austenitizing.............................................. 355 6.3.1.2 Technological Aspects of Austenitizing ............................................ 364 ß 2006 by Taylor & Francis Group, LLC. 6.3.2 Quenching Intensity Measurement and Evaluation Based on Heat Flux Density ........................................................................................... 374 6.3.3 Retained Austenite and Cryogenic Treatment................................................. 384 6.3.3.1 Transforming the Retained Austenite ............................................... 387 6.4 Hardening and Tempering of Structural Steels.......................................................... 390 6.4.1 Mechanical Properties Required...................................................................... 390 6.4.2 Technology of the Hardening and Tempering Process.................................... 397 6.4.3 Computer-Aided Determination of Process Parameters.................................. 402 6.5 Austempering ............................................................................................................. 407 References .......................................................................................................................... 413 6.1 FUNDAMENTALS OF HEAT TREATMENT 6.1.1 HEAT TRANSFER Heat treatment operations require a direct or an indirect supply of energy into the treated workpieces and its subsequent removal to effect the heating and the cooling, respectively, of these pieces. Because this chapter deals only with heat treatment operations involving the whole 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 a plate 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): Q¼ dQ dT ¼ ÀlF , x ¼ 0, . . . , s=2 dt dx (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). F F t = t2 > t1 Tu t = t0 t = t1 > t0 t = t2 > t1 t = t1 > t0 t = t0 Q Q Q Q Tu X X S (a) S (b) FIGURE 6.1 Temperature distribution on the cross section of a plate (a) during heating and (b) during cooling. t0, Beginning of temperature change; Tu, surrounding temperature; Q, heat flux; s, thickness of ¨ ¨ the plate. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. From the other side, based on the first law of thermodynamics, dQ ¼ r dx FCp zdT (6:2) where r is the density of the workpiece (kg/m3) and Cp the specific heat capacity of the workpiece at constant pressure (J/(kg K)). From Equation 6.1 it follows that   d dQ d2 T d dQ ¼ ÀlF 2 ¼ dx dt dx dt dx (6:3) From Equation 6.2 it follows that  d dQ dT ¼ ÀrFCp dt dx dt (6:4) and for the one-dimensional heat flux, the time-dependent temperature distribution inside the workpiece is dT l d2 T d2 T ¼ ¼a 2 2 dt rCp dx dx (6:5) where a ¼ l/rCp (m2/s) is the temperature diffusivity. In the case of three-dimensional heat flux, Equation 6.5 reads ! dT d2 T d2 T d2 T þ þ 2 ¼a T ¼a dt dx2 dy2 dz (6:6) Á where ¼ d2 d2 d2 þ 2þ 2 dx2 dy dz Á is the Laplace operator. Equation 6.5 and Equation 6.6 are the temperature conduction equations in which the temperature diffusivity represents the amount of the time-dependent temperature change of a workpiece because of nonstationary heat conduction. A heat flux dQ flowing through a surface of area F is, according to Fourier’s heat conduction law, proportional to the temperature gradient at the relevant position: dQ ¼ Àl dT F dt dx (6:7) or, expressed as heat flux density per unit time (s) and unit surface (m2), q ¼ Àl dT ¼ Àl 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 material’s 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 in reality it depends on temperature. Figure 6.2 shows the temperature dependence of heat ß 2006 by Taylor & Francis Group, LLC. TABLE 6.1 Approximate Values for Heat Conductivity l, W/(m K) Metals Copper Aluminum Brass Gray iron Steel Liquids Water Oil Stiff inorganic materials Chamotte Brick Concrete Mineral wool 0.6–0.7 0.1–0.2 0.5–1.2 0.8 0.8–1.4 0.05–0.1 Gases Air (20–20008C) H2 (20–20008C) 350 170–230 80–115 58 40 0.026–0.11 0.18–0.75 ¨ ¨ Source: From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987. conductivity for groups of steel. As can be seen, the biggest differences in heat conductivity among different steel grades are at room temperature. Whereas for unalloyed steels the heat conductivity decreases with increasing temperature, for high-alloy steels it slightly increases with increasing temperature. At about 9008C (16528F), the value of l is almost the same for all steel grades. The specific heat capacity Cp depends also on temperature. The transport of thermal energy through a solid body, described by the heat conduction equation (Equation 6.8), extends naturally beyond the body surface; i.e., heat transfer takes place between the body and its environment. This heat transfer is expressed as the amount of heat exchanged between the surface of the body and its environment per unit surface area and per unit time. According to Newton’s law of cooling, the amount of heat exchanged between a body and its environment depends on the difference between the body surface temperature and the temperature of its environment. The relevant heat flux density is q¼ dQ ¼ a (TK À TU ) dF dt for T K > T U (6:9) where TK is the body surface temperature, TU is the temperature of the environment, and a is the heat transfer coefficient, W/(m2 K). 80 Heat conductivity l, W/m K a 60 b 40 c d 20 0 0 200 400 600 800 Temperature T, 8C 1000 FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. The actual conditions of the heat transfer in each case are represented by the relevant heat transfer coefficient a, which depends on 1. 2. 3. 4. 5. The shape and cross-sectional size of the body The position of the body (standing or lying) The surface condition of the body The physical properties of the body’s material The physical properties of the surrounding fluid (density, specific heat capacity, dynamic viscosity) 6. The agitation (flow) rate of the surrounding fluid During every heating or cooling process, the temperature difference between the body surface and its environment becomes smaller with time, i.e., the exchanged heat quantity becomes smaller. The heat transfer coefficient a is therefore not a constant but varies with the body surface temperature. Gases, liquids, and vacuums are the environments in which heat transfer occurs during heat 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 or gaseous 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 to another. Therefore heat convection is possible only in liquids and gases. The amount of heat transferred 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 temperature difference, the movement is called free or natural convection. When the movement of particles of 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 convection contributing to the heat transfer depends on the temperature gradient within the fluid, and the contribution 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 aK can be calculated, according to Eckstein [2], by using the formula aK ¼ (4:64 þ 3:49  10À3 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 cylinder surface; and v is the air velocity (m/s). The third heat transfer mechanism is heat radiation. Solid bodies, liquids, and gases can all transfer heat in the form of radiation. This kind of heat transfer does not depend on any heat transfer carrier; therefore, it can take place in vacuum also. Heat radiation is in the form of electromagnetic waves whose length is in the range of 0.3–500 mm. When radiation strikes the surface of a body, part of it will be absorbed, part of it will be reflected, and the rest may pass through the body. Every body emits radiation corresponding to its temperature. The body that, at a certain temperature, emits or absorbs the largest amount of radiation is called a blackbody. All other bodies emit or absorb less radiation than the blackbody. The ratio of the 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 according to the Stefan–Boltzmann law q ¼ «sT 4 ß 2006 by Taylor & Francis Group, LLC. (6:11) where « is the emission-relation coefficient, « < 1.0; s is the Stefan–Boltzmann constant, s ¼ 5.67  10À8 W/(m2 K4); and T is the absolute temperature (K). If two bodies mutually exchange radiant heat, then not only is the warmer body emitting heat to the colder one, but the colder body is also emitting heat to the warmer body, so that the transferred heat consists of the difference of the amounts of heat absorbed by the two bodies. The total heat transferred by radiation from one body having a surface area F1 to another solid body of any surface area can be calculated according to 4 4 Q ¼ «1;2 sF1 (T1 À 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 absolute temperature of the emitting body; and T2 is absolute temperature of the absorbing body. In industrial furnaces, heat is transferred substantially by simultaneous heat convection and heat radiation. The heat transferred by heat conduction (in fluids) is negligible compared to 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 with a formula similar to Newton’s law q ¼ a« (T1 À T2 ) (6:13) The heat transfer coefficient for radiation a« can be calculated by combining Equation 6.12 and Equation 6.13: a « ¼ «1 , 2 s 4 4 T1 À T2 T1 À T2 (6:14) The total heat transfer coefficient is then a ¼ a k þ 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 transfer coefficient depends in each case on many specific factors, discussed earlier, but also depends strongly on the surface temperature of the workpiece. It is a temperature-dependent and location-dependent value that changes during heat transfer as the body surface temperature equalizes to the environment’s temperature. According to Eckstein [2], the complex heat transfer 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. With increasing temperature, heat transfer by radiation becomes more important, and at about 8008C (14728F) it reaches 80% of the total heat transfer. Especially in operations that employ immersion quenching in liquids, where two-phase heat transfer takes place with high heat flux densities under nonstationary conditions, the heat transfer coefficient value changes very much. Therefore nowadays when heat transfer calculations are carried out by computer a temperature-dependent function of the heat transfer coefficient instead of an average value should be used. One practical way to obtain this function in each actual case is to measure the surface temperature of an adequately instrumented probe (cylinder or plate of adequate dimensions) placed appropriately in the quenching tank and, from the measured surface temperature vs. time history, calculate the corresponding heat flux density and heat transfer coefficient vs. temperature functions. ß 2006 by Taylor & Francis Group, LLC. TABLE 6.2 Average Values of the Heat Transfer Coefficient a When Cooling or Quenching in Liquid or Gaseous Media a [W/(m2 K)] Medium Furnace Still air Moving air Compressed air Air–water mixture Hardening oil Liquid lead Water 15 30 40 70 520 580 1200 3500 ¨ Source: From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB ¨ Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987. Figure 6.3 shows, as an example, the heat transfer coefficient vs. surface temperature for quenching 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, the temperature can be measured near the surface and, by using the inverse heat conduction method and an adequate mathematical procedure finite element method (FEM), the surface temperature of the probe can be calculated. To explain the dependence between the heat transfer conditions and the temperature fields in solid bodies, let us consider the heating of a plate of thickness s (see Figure 6.4). At the beginning of heating (t ¼ 0), the plate has a temperature TK ¼ 0 and is suddenly transferred in standing position into a furnace, where the environmental temperature is TU. Equal amounts of heat are transferred from both sides of the plate. Because boundary conditions of the third 2000 Heat transfer coefficient a, W/m2K 1800 1600 1400 1200 1000 800 600 400 200 0 0 100 200 300 400 500 600 700 800 900 o Temperature, C FIGURE 6.3 Heat transfer coefficient vs. surface temperature when quenching a stainless steel cylinder of 50-mm diameter  200 mm in still oil of 208C, calculated from the measured surface temperature–time ˇ´ ˇˇ ´ history. (From B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation and calculation of heat transfer data, Proceedings of the First International Conference on Quenching and Control of Distortion, Chicago, IL, September 1992, pp. 17–26.) ß 2006 by Taylor & Francis Group, LLC. l a 2 l a T TU 1 t3 t2 t1 x2 0 − s −x1 2 −x 0 < t1 < t2 < t3 TK (x, t = 0) = 0 +x1 + s 2 +x FIGURE 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 a point at temperature TU outside the plate. The straight line connecting this point and the relevant surface temperature is the tangent on the temperature distribution curve at the body surface. As time progresses, both the surface temperature and the temperature in the middle of the plate increase. The temperature gradient inside the body is different for different l/a ratios and changes over time. If the heat conductivity l of the body’s material is small or the heat transfer between the environment and the body surface is large, the ratio l/a is small and heat accumulates in the surface region because the amount of heat transferred is greater than the amount transported by conduction into the body’s interior. The smaller the ratio l/a, the faster the surface temperature equalizes to the temperature of the environment. The relevant changes of temperature at points x1 and x2 and the value and the change in the temperature gradient over time are also greater. This can be seen in Figure 6.5 when comparing the curves for (l/a)1 small to (l/a)2 big. If the heat conductivity l is big or the heat transfer coefficient a is small, i.e., TU Temperature Temperature gradient 1 2 3 Curve 4: T (x2) = f[(l/a)2, t ] 4 6 5 Curve 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: dT dx x1 dT Curve 6: dx x1 = f[(l/a)1, t ] = f[(l/a)2, t ] Time FIGURE 6.5 Change of the plate temperatures at points x1 and x2 and temperature gradients at point x1 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, 2nd ¨ ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) ß 2006 by Taylor & Francis Group, LLC. the heat is transported from the surface to the core of the body faster than it is transferred from the environment to the body’s surface, the l/a ratio becomes big, and the temperature in the interior of the body increases relatively faster than the surface temperature. Other factors that should be taken into account when analyzing such heat transfer problems are the shape and cross-sectional size of the body. As to the influence of different shapes of workpieces it should be borne in mind that at constant heat transfer conditions and constant thermal properties of the material and equal temperature of the environment, the temperature change with time depends on the surface-to-volume ratio of the body. The greater the ratio is, the greater is the temperature change over time. 6.1.2 LATTICE DEFECTS Generally the lattice of a metal crystal contains imperfections, i.e., aberrations from a perfect atomic arrangement. These imperfections may be divided, from the geometrical standpoint, into the following categories: Zero-dimensional or point imperfections One-dimensional or linear Two-dimensional or superficial Three-dimensional or spatial The 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 is built, with four different types of lattice point defects. At position b one atom is missing; this defect is called a vacancy. Atom c occupies a place between the regular places in the lattice and it is called an interlattice atom; in this case c is the same kind of atom as the regular atoms. In position d, a strange atom (of an alloying element) with a larger diameter has taken the place of a regular atom; therefore it is called a substitutional atom. A practical example of this is a manganese atom dissolved in iron. In position e, a strange atom with a much smaller diameter than the regular atoms of the lattice is inserted between regular atoms in a position that is not occupied by regular atoms. It is called an interstitial atom. A practical example of this is a carbon atom dissolved in iron. Both substitutional and interstitial atoms cause local deformations and microstresses of the crystal lattice. Figure 6.6b shows at f a linear lattice defect. A row of atoms in the outlined atomic plane terminates at this point. If we imagine the outlined atomic plane as a section through a crystal that stretches perpendicular to the plane shown, then the row of atoms terminating at f becomes a half-plane of atoms that has been inserted between regular planes of atoms and ends inside the crystal. The boundary line of the inserted half-plane of atoms that stretches through the greater lattice region, perpendicular to the plane shown, is a linear lattice defect called 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 g–g represents schematically a low-angle grain boundary that consists of edge dislocations arranged regularly one under the other. The inserted half-plane of each edge dislocation terminates at the associated atom shown in black. The dashed area is a section through a low-angle grain boundary between neighboring parts of the crystal lattice that are inclined at a low angle to each other. The line h–h represents a twinning boundary. It is characterized by the fact that the atoms on both sides of the boundary are symmetrically distributed, and therefore neighboring parts of the crystal 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 of atoms is missing. This zone of missing atoms could have developed by way of an accumulation of vacancies. It can be stretched to other planes of atoms perpendicular to the outlined one. The imperfection k is a more or less irregular distribution of atoms between two neighboring parts of ß 2006 by Taylor & Francis Group, LLC. a b g c (b) (a) h f d gh e (c) k l m i (d) n o p q FIGURE 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, interlattice atom; 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, phase ¨ boundary; 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 the lattice. It is called a high-angle grain boundary, or simply a grain boundary. The superficial imperfection at l is a section through a zone of strange atoms that stretches in two dimensions perpendicular to the plane shown. The boundary plane m between two different lattices is called a phase boundary, which is also a two-dimensional lattice defect. Figure 6.6d shows schematically the characteristic three-dimensional lattice defects. In many metal alloys, within the lattice of grains under specific thermodynamic conditions, new lattice regions with changed structure are formed. Such a lattice defect shown at n is called a precipitate. The spatial imperfection at o is called an inclusion. Such inclusions, which develop unfailingly during the production of alloys, are nonmetallic or intermetallic compounds. Like precipitates, inclusions have their own structure and phase and are separated by a phase boundary from the surrounding lattice. Microcrack is denoted by p, a spatial imperfection that is created by three edge dislocations that came to a phase boundary and formed 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 crystal’s lattice is ß 2006 by Taylor & Francis Group, LLC. shown; this is called a micropore. Such defects can originate from the accumulation of either vacancies or gases. Of all the lattice defects discussed above, vacancies and edge dislocations are especially important 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 a consequence of the intermittent movement of the atoms), compared to the case of a perfect iron crystal (whisker), dislocations facilitate the plastic deformation of the material. 6.1.3 APPLICATION OF TTT (IT) AND CCT DIAGRAMS Time–temperature–transformation 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 process that will yield the desired microstructure and hardness. The use of the either type of diagram requires that the user be acquainted with its specific features, possibilities, and limitations. 6.1.3.1 Isothermal Transformation Diagram Figure 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 in the time–temperature diagram (the abscissa of which is always in logarithmic scale) are valid only under conditions of fast quenching from the austenitizing temperature to the chosen transformation temperature and subsequent holding at that temperature. This is the way %C %Si %Mn %P %S %Cr %Cu %Ni Method of melting Mc Quaid–Ehn b.S-M. %V 4 Temperature, C 0.43 0.41 0.82 0.041 0.015 1.22 0.14 0.04 0.11 Austenitizing temperature = 880 C AC3 (0.25 /min) 800 AC1 (0.25 /min) Ferrite-start Pearlite-start 700 22 Pearlite-finish 27 F P 35 39 1% 600 99% 500 42 Bainite-start B 400 39 40 40 MS 99% 46 300 51 Bainite-finish 200 Hardness HRC 100 0 1 101 s 60 102 1 2 103 8 15 30 60 min 1 Time 104 105 4 2 4 6 8 16 24 h FIGURE 6.7 Isothermal transformation (IT) diagram of DIN 50CrV4 steel. (From A. Rose and ¨ W. Strassburg, Archiv. Eisenhuttenwes. 24(11/12):505–514, 1953 [in German].) ß 2006 by Taylor & Francis Group, LLC. 1.0 0.8 M 0.6 0.4 0.2 0 0 500 1000 Time, s 1500 2000 FIGURE 6.8 Relation between the amount of transformed structure (M) and time in IT diagrams, accord¨ ing to Equation 6.16. (From H.P. Hougardy, Harterei-Tech. Mitt. 33(2):63–70, 1978 [in German].) the diagram itself was developed. Therefore the IT diagram may be read only along the isotherms. The beginning and end of transformation of ferrite, pearlite, and bainite in isothermal processes take place according to the function M ¼ 1 À exp ( Àbt n ) (6:16) where M is the fraction of phase transformed; t is time (s); and b ¼ 2  10À9 and n ¼ 3. Because, as shown in Figure 6.8, this function starts and ends very flat, the actual beginning and end of transformation are difficult to determine exactly. Therefore, an agreement is reached, according to which the curve marking the beginning of transformation denotes 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 only on the temperature below the Ms point. Hougardy [5] gave the following formula (valid for structural steels for hardening and tempering) for this transformation: Ma ¼ 1 À 0:929 exp[ À 0:976  10À2 (Ms À T )1:07 ] (6:17) where Ma is the amount of martensite, Ms is the martensite start temperature, and T is a temperature below Ms. Some IT diagrams, when read along the isotherms, enable the user to determine the percentages 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 holding time) is fast quenched to 6508C (12008F) and held at this temperature, after 12 s ferrite starts to form. After 30 s the formation of pearlite begins. After 160 s the transformation is completed with 5 vol% of ferrite and 95 vol% of pearlite formed. The hardness achieved is about 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 be taken as +108C (508F), and with respect to the time ordinates, as +10%. 6.1.3.2 Continuous Cooling Transformation Diagram Figure 6.10 shows the CCT diagram of the same heat (as Figure 6.7) of the low-alloy DIN 50CrV4 steel. ß 2006 by Taylor & Francis Group, LLC. C 0.44 Chemical composition Mn 0.80 Si 0.22 P 0.030 1000 S 0.023 Cr 1.04 Cu 0.17 Mo 0.04 Ni 0.26 V <0.01 Austenitizing temp. 840 C Holding time 5 min 900 Ac3 800 Ac1 Temperature, C 700 F A 20 95 P 600 31 70 32 500 B 38 400 M s 300 50% 90% M 200 100 0 1 10 102 103 104 105 106 Time, s FIGURE 6.9 Isothermal transformation (IT) diagram of DIN 41Cr4 steel. (From H.P. Hougardy, ¨ Harterei-Tech. Mitt. 33(2):63–70, 1978 [in German].) Bez. %C %Si %Mn %P %S %Cr %Ni %Cu %V 1 0.43 0.41 0.82 0.041 0.015 1.22 0.04 0.14 0.11 Method of melting b.S.-M. Austenitizing temp. = 8808C 25 22 F Temperature, 8C 600 4 Grain size (ASTM) = 10 –11 800 700 MC Quoid–Ehn 15 20 10 80 5 85 1 1 30 P 25 75 76 78 % Ferrite Ac3 (0.258 min) % Pearlite Ac1 500 400 .. = Hardness HRC B MS 300 5 10 20 30 40 30 30 50 200 % Bainite % Martensite 90 100 58 0 101 1 s 57 57 53 52 46 4147 33 31 25 60 102 1 2 103 4 8 15 30 60 min 12 14 20 104 4 68 h 105 106 1624 1 2 3 4 6 10 Days FIGURE 6.10 Continuous cooling transformation (CCT) diagram of DIN 50CrV4 steel. (From A. Rose ¨ and W. Strassburg, Archiv. Eisenhuttenwes. 24(11/12):505–514, 1953 [in German].) ß 2006 by Taylor & Francis Group, LLC. When comparing the curves for the start of transformation in CCT and IT diagrams for the same heat and steel grade (Figure 6.7 and Figure 6.10), we found that in the CCT diagram the curves are slightly shifted to longer times and lower temperatures. For example, in the IT diagram 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 CCT diagram of Figure 6.10, however, the shortest transformation start time for ferrite is 32 s at 6208C (11508F) and the corresponding time for bainite is 20 s at 3808C (7168F). This indicates that in CCT processes the transformation starts later than in IT processes. The basis of this phenomenon is related to the incubation time and can be found in a hypothesis of Scheil [6]. It should also be noted that with higher austenitizing temperature the curves denoting the start of transformation of a particular phase can be shifted to longer times. Figure 6.11 shows the CCT diagram of DIN 16MnCr5 steel after austenitizing at 8708C (16008F) (a), and after austenitizing at 10508C (19228F) (b). In the latter case, the regions of ferrite and pearlite are shifted to longer times. It is necessary, therefore, when using a CCT diagram, to ascertain that the 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-mm diameter  2 mm for high cooling rates, and 4.5-mm diameter  15 mm for medium and low cooling rates) are austenitized and cooled within a dilatometer with different cooling rates. Start and finish of transformation of relevant phases with each cooling curve are recorded and these points are connected to obtain the regions of transformation for the relevant phases (see Figure 6.10). Therefore, a CCT diagram can be read only in the way in which 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 slow cooling 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 numbers along the cooling curve (usually marked in CCT diagrams of German origin). The numbers at the 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 for grade DIN 50CrV4 steel, if cooling proceeds at the rate marked with Â, 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 the area (region) of a phase that the cooling curve intersects is by no means a measure of the amount of transformed phase. Sometimes a CCT diagram can be supplemented with a diagram showing portions of each structural phase and hardness after quenching; see the lower part of Figure 6.12. The abscissa of this diagram denoting time enables the cooling time to 5008C (9328F) to be determined for every cooling curve. To determine the portions of structural phases and hardness after quenching, one should follow the relevant cooling curve until its intersection with the 5008C (9328F) isotherm and from this point down along the vertical line read the phase portions and hardness after quenching. For example, for cooling curve C, which intersects the 5008C (9328F) isotherm at 135 s, the readings are 4% ferrite, 7% pearlite, 78% bainite, and 11% 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 that was used for its construction. The influence of different heats (having slightly different compositions) of the same grade of steel on the position of transformation curves in the relevant CCT diagram, as an example, is shown in Figure 6.13. ß 2006 by Taylor & Francis Group, LLC. Chemical composition vol% C Si Mn P S Al, ges. Cr Mo Ni V 0.16 0.22 1.12 0.030 0.008 0.015 0.99 0.02 0.12 0.01 1000 Austenitizing temp. 8708C (Holding time 10 min) heated in 3 min 900 Ac3 800 Ac1 Temperature, 8C 700 F A 2 600 20 5 1 B Ms 5 60 65 60 35 300 25 72 M 200 7 100 188 423 (a) 28 23 27 500 400 65 72 35 65 50 60 P 62 5 70 412 181 268 260 229 207 182 165 305 0 1000 Austenitizing temp. 10508C (Holding time 10 min) heated in 3 min 900 Ac3 Temperature, 8C 800 Ac1 F 700 A 1 600 5 2 12 50 50 30 P 65 35 2 10 500 400 B Ms 60 10 60 70 68 65 55 300 35 35 M 200 100 283 425 (b) 0 0.7 1 s Time 422 340 10 10 1 263 276 290 2 206 250 195 423 103 105 104 10 100 1000 min FIGURE 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 Warme¨ ¨ behandlung der Stahle, Vols. I and II, Verlag Stahleisen, Dusseldorf, 1954/56/58.) ß 2006 by Taylor & Francis Group, LLC. C 0.44 Chemical composition Si 0.22 Mn P S Cr 0.80 0.030 0.023 1.04 1000 Cu 0.17 Mo 0.04 Austenitizing temp. 8408C Holding time 8 min 900 Ac3 800 D Temperature, 8C 700 Ni V 0.26 <0.01 A C E 4 600 a 8 18 7 2 500 b 82 c 60 62 66 Ac1 40 38 34 50 d 400 e 3 16 17 300 58 78 40 200 58 60 0 100 80 36 28 52 44 39 34 20 18 16 M P 60 40 60 40 F B 20 20 HRC 0 0 1 10 102 103 104 Hardness, HRC Portion of structure, % 100 105 Time, s FIGURE 6.12 CCT diagram of 41Cr4 steel (top) with the diagram at bottom showing portions of each structural phase and hardness after quenching depending on the cooling time to 5008C. (From H.P. ¨ Hougardy, Harterei-Tech. Mitt. 33(2):63–70, 1978 [in German].) As for IT diagrams, the accuracy of a CCT diagram, according to Hougardy [5], with respect to the position of isotherms is +108C (508F) and with respect to time ordinates +10% of the relevant time. 6.1.3.3 Heat Treatment Processes for Which an IT or CCT Diagram May Be Used Taking 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 be used. The first is isothermal annealing to obtain a coarse ferritic–pearlitic structure, for better machinability (Figure 6.14a). In this case, the IT diagram gives the crucial information, the optimum temperature at which annealing should take place to achieve the full transformation in 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 in question is applicable for this process, i.e., has enough hardenability (which means that its start 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 the hardness desired and read the minimum time needed at this temperature to achieve the full transformation to bainite. ß 2006 by Taylor & Francis Group, LLC. 900 800 Heat 1 0.44% C and 1.20% Cr Heat 2 0.41% C and 1.06% Cr Temperature, 8C 700 600 500 400 300 200 100 0 1 10 102 Time, s 103 104 FIGURE 6.13 Influence of the difference in composition between two heats of DIN 41Cr4 steel on the ¨ position of transformation curves in the relevant CCT diagram. (From H.P. Hougardy, Harterei-Tech. Mitt. 33(2):63–70, 1978 [in German].) The third process is the martempering process (Figure 6.14c), an interrupted quenching in a hot bath to obtain the martensite structure with minimum stress and distortion. The applicability of a steel for the martempering process may be checked in the same way as above. In this case the diagram gives information about the necessary temperature of the hot bath and the maximum time the parts can be immersed in it (in order to obtain only martensite) 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 which only a CCT diagram may be used. The first case (Figure 6.15a) is direct quenching to obtain full martensitic structure. In this case the diagram enables the user to determine the critical cooling rate for the steel in question. The second case (Figure 6.15b) is a continuous slow cooling process, e.g., cooling in air after normalizing annealing. In this case the diagram enables the user to select the cooling rate required to yield the desired hardness of the ferritic– pearlitic structure at room temperature. The percentage of ferrite and pearlite can be read as described above if the diagram allows it. The third case (Figure 6.15c) represents any continuous cooling regime that results in more than two structural phases. In any of these cases the diagram enables the user to determine the portion of each phase and the hardness after quenching. 6.1.3.4 Using the CCT Diagram to Predict Structural Constituents and Hardness upon Quenching Real Workpieces Each CCT diagram describes only those transformations of the structure that occur along the cooling curves of specimens used for its construction. This means that a CCT diagram is valid only for the cooling conditions under which it was constructed. The cooling law for the specimens of small diameter and small volume that were used in constructing the CCT diagram can, according to Rose and Strassburg [4], be described by the exponential function ß 2006 by Taylor & Francis Group, LLC. Ac3 Ac1 B M log t Ac3 Ac1 A r F P B Ms log t Ac3 Ac1 A F P ter Surface Temperature, C Ms Cen (c) 800 700 600 500 400 300 200 100 0 P A ce Surfa (b) 900 800 700 600 500 400 300 200 100 0 F te Cen Temperature, C Temperature, C 800 700 600 500 400 300 200 100 (a) 0 B Ms log t FIGURE 6.14 Isothermal heat treatment processes for which only IT diagrams may be used. (a) Isothermal annealing; (b) austempering; (c) martempering. T ¼ T0 eÀ t (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 values upon quenching depends on the extent to which the cooling law at different points in the cross section of real workpieces corresponds to the cooling curves of specimens drawn in the CCT diagram. 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 in different points of a round bar’s cross section correspond in form to the cooling curves in CCT diagrams to the extent that the structural transformation, i.e., the resulting structural constituents and hardness values, can be compared. Figure 6.16 shows how a hardness distribution can be predicted by using this correspondence. If the temperature–timescales of the measured cooling curves and the CCT diagram are the same (as in Figure 6.16a and Figure 6.16b), then by using a transparent sheet of paper the ß 2006 by Taylor & Francis Group, LLC. Temperature, C Ac3 Ac1 A V crit 800 700 600 500 400 300 200 100 0 (a) F P Ms B M log t Ac3 Temperature, C 800 700 600 500 400 300 200 100 0 (b) Ac1 A F P B Ms M 220 HV 180 HV log t Temperature, C Ac3 (c) 800 700 600 500 400 300 200 100 0 Ac1 A 10 F 3 Ms B P 70 M 30 HRC log t FIGURE 6.15 Heat treatment processes with continuous cooling for which only CCT diagrams may be used. (a) Direct quenching to obtain full martensitic structure; (b) slow cooling to obtain a ferrite– pearlite structure of required hardness; (c) continuous cooling regime where a mixed structure is obtained. measured cooling curves on CCT diagrams of different steel grades can be superimposed, and in this way steel grade can be selected that will develop the required structure and hardness at the desired point of the cross section. The accuracy of such prediction from a CCT diagram decreases as the radius of the workpiece’s cross section increases. According to Peter and Hassdenteufel [8], sufficiently exact predictions are possible using CCT diagrams for round bars up to 100 mm in diameter when quenching in oil and up to 150 mm in diameter when quenching in water. It appears that the main problem in the practical use of CCT diagrams for prediction of structural constituents and hardness upon quenching is to establish exactly the cooling curve for the specified point on the workpiece’s cross section. This can be done either by calculation (if symmetrical parts and one-dimensional heat flow are involved and the boundary conditions are known) or experimentally (for asymmetric parts) by measuring the ß 2006 by Taylor & Francis Group, LLC. 60 0 I II III IV 600 400 0 I II III IV Hardness, HRC Temperature, C 800 1 mm 5 mm 10 mm 15 mm 25 mm Below surface 200 40 20 0 (c) 0 1 10 (a) 102 Time, s 103 Core 10 20 30 0 Distance from surface, mm 104 Temperature, C 800 % Ferrite 40 20 F P % Pearlite 15 60 20 65 23 l ll lll lV 20 B l 10 40 30 5% Bainite 50 95% Martensite HRC A 600 0 400 200 Ms M 57 0 1 10 (b) 52 46 36 27 102 Time, s 103 104 FIGURE 6.16 Prediction of hardness distribution on a round bar cross section. (a) Cooling curves measured at different points below surface, as indicated. (b) CCT diagram of the relevant steel with superimposed cooling curves from (a). (c) Hardness distribution on the bars’ cross section upon quenching, obtained by reading the hardness values from (b). temperature–time history with a thermoelement. The correspondence between cooling curves of real workpieces and cooling curves drawn on CCT diagrams also enables the reverse, to draw conclusions about the cooling history (curve) at a specified point of the cross section of a workpiece of any shape and size based on metallographic analysis of the structure and measured hardness upon quenching. When CCT diagrams (of American origin) are used, the manner of predicting structural constituents 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 of the 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 the quenched end (the heavier line in the diagram), one can read that after 25 s of cooling, the Jominy specimen made of AISI 3140 steel at this distance from the quenched end starts to develop ferrite, after 30 s, pearlite; and after 45 s bainite. After 90 s of cooling 50% of the austenite is already transformed. After 140 s, when the temperature at this point has fallen to 3158C (5998F), formation of martensite begins. The corresponding Jominy curve at the bottom of Figure 6.17a shows that this cooling curve (at this Jominy distance) with the steel in question will yield a hardness of 48 HRC. To correlate this hardness to different points of the round bars’ cross sections of different diameters, an auxiliary diagram (valid in this case only for quenching in moderately agitated oil) such as that shown in Figure 6.17b should be used. From this diagram one can see that the same hardness of 48 HRC can be met, after quenching in moderately agitated oil, 9 mm below the surface of a round bar of 75-mm diameter. ß 2006 by Taylor & Francis Group, LLC. Temperature 871 1600 0.41 C, 0.86 Mn, 0.26 Si, 1.28 Ni, 0.71 Cr 760 1400 6.4 1/8 3/16 1/16 12.7 19.3 3/8 1/2 25.4 3/4 1/4 mm 11/2 in. F1 A 649 1200 1% 3 P1 1 2 F+ A 10% F+P+A 4 538 1000 2 50% F+P+B+A 427 800 5 6 A 3 75% Ms 316 600 93 200 1 2 Transformed M+A — austenite — ferrite — pearlite — bainite — martensite 204 400 1% 1/16 5 10 10% 20 50 Cooling time, s 50% 75% 100 1/4 3/8 1/2 3/4 200 1/8 3/16 500 1000 75 S Hardness rockwell, C 65 55 45 35 25 15 5 0 4 0 8 12 16 20 24 28 32 36 40 in. 16 6.4 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 below surface of bar in. mm 2 50 ∅ 100 mm(4 in.) 45 11/2 40 ∅75 mm(3 in.) 35 30 1 ∅ 50 mm(2 in.) 25 ∅ 38 mm(11/2 in.) 20 15 1/2 10 5 ∅25 mm(1 in.) ∅125 mm(1/2 in.) 0 0 0 (b) 5 10 15 1/2 20 25 30 35 40 11/2 1 Jominy distance 45 50 mm 2 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 Jominy distances and rate of cooling in moderately agitated oil of round bars of 12.5–100-mm diameter. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. There also exist CCT diagrams of another type, developed by Atkins [9]; an example is given in Figure 6.18. These diagrams were developed by cooling and quenching round bars of different diameters in air, oil, and water, recording their cooling curves in the center of the bar, and later simulating these cooling curves in a dilatometric test in order to identify the transformation temperatures, microstructures, and hardness. These diagrams therefore refer only to the center of a bar. Instead of a timescale on the abscissa, these diagrams have three parallel scales, denoting bar diameters cooled in air, quenched in oil, and quenched in water. 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), denoting different cooling rates. For example, to determine the microstructure developed and resulting hardness in the center of a 10-mm bar of the steel in question when cooling it in air, one takes the vertical line at 10-mm diameter on the scale for air cool (see Figure 6.18), starts in the austenite region and proceeds downward. Transformation in this case (unalloyed steel grade with 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 of bainite is indicated before transformation is complete. If oil quenching of a 10-mm bar is now considered, the 10-mm position should be located on the oil-quenched bar diameter scale in Figure 6.18. Again starting in the top region and following the vertical line down, it is seen that in this case bainite is the first phase to form from austenite at 5608C (10408F). At 3308C (6268F), after about 40% transformation, the remaining 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 martensite starting at 3608C (6808F) and finishing at 1508C (3008F). Relevant hardness values after quenching (and in some cases after tempering to different specified temperatures) can be read following the same vertical line further down into the hardness after transformation diagram. An examination of the left-hand side of the upper diagram in Figure 6.18 for the steel in question shows that martensite will form on air cooling with bars up to 0.18 mm in diameter, on oil 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 be assessed at a glance. Figure 6.19a is a CCT diagram for a very low hardenability steel previously rolled and austenitized at 9508C (12428F). It shows early transformation to ferrite and pearlite (even with oil and water quenching of smallest diameters). Figure 6.19b shows a similar diagram for a high-hardenability steel previously rolled, softened at 6008C (11128F), and austenitized at 8308C (15268F). In this case the austenite changed predominately to martensite and bainite over a wide range of bar diameters and quenching rates. Diagrams of this type representing 172 steel grades have been published in the British Steel Corporation (BSC) atlas [9]. 6.1.3.5 Special Cases and Limitations in the Use of CCT Diagrams When dealing with carburized steels, one should be aware that, because of the big difference in carbon content between the core (%0.2%) and the case (%0.8%), the CCT diagram for the case will be totally different from the one for the core of the same steel, as shown in Figure 6.20 and Figure 6.21. The increased carbon content in the case increased the hardenability and caused the pearlite and bainite regions to be shifted to much longer times. The ferrite region disappeared, and the Ms point was lowered. Cooling at the same rate results in different portions of structural constituents and substantially different hardness values. ß 2006 by Taylor & Francis Group, LLC. Austenitized at 860 C previous treatment rolled Analysis, wt% C 900 Si Mn 0.38 0.20 0.70 S Cr Mo Water quench Ac3 Ni Al Nb V Air cool Oil quench 800 700 P 0.020 0.020 Start A 10% 50% 90% Ac1 F Finish P 600 B C 500 400 300 M 200 100 1000 500 200 100 50 20 10 5 2 1 C per min Cooling rate at 750 C 0 0.2 mm 0.1 Bar diameter 5 0.5 10 10 1 2 5 20 20 10 20 100 50 50 100 50 150 200 150 200 100 300 300 200 500 1000 2000 mm Air 500 500 mm Oil mm Water 800 700 Hardness after transformation Hardenability band BS 970 080 136 600 500 HV 400 300 As cooled 60 T 500 C 1 h T 600 C 1 h T 700 C 1 h 50 HRC 40 30 20 200 10 100 FIGURE 6.18 CCT diagram for rolled steel austenitized at 8608C. (From M. Atkins, Atlas of Continuous Transformation Diagrams for Engineering Steels, British Steel Corporation, BSC Billet, Bar and Rod Product, Sheffield, U.K., 1977.) ß 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 oil quenching. The left-hand part of Figure 6.22a shows the start of transformation as in the conventional CCT diagram for the steel in question. The right-hand part (Figure 6.22b) of this 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 bainite regions to longer times. Such a cooling mode enhances hardenability and results in higher 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 incubation time before the change in cooling rate occurs and is theoretically explained by Shimizu and Tamura [11]. Austenitized at 9508C previous treatment rolled sofened 600∗C Analysis, wt % C Si 0.06 900 Mn P S Cr Mo Ni Al Nb V 0.30 Ac3 Start A 10% 50% 800 90% 700 F Ac1 Finish F P B 600 C 500 M 400 300 200 100 500 1000 200 100 50 Cooling rate at 8508C 20 10 5 2 1 8C per min 0 mm 0.1 Bar diameter (a) 0.2 0.5 5 1 10 2 5 20 10 20 50 50 100 100 150 200 300 200 500 1000 2000 mm Air 500 mm Oil 10 20 50 100 150 200 300 500 mm Water FIGURE 6.19 Examples of (a) a low-hardenability and (b) a high-hardenability steel as depicted in CCT diagrams. (According to M. Atkins, Atlas of Continuous Transformation Diagrams for Engineering Steels, British Steel Corporation, BSC Billet, Bar and Rod Product, Sheffield, U.K., 1977.) ß 2006 by Taylor & Francis Group, LLC. Austenitized at 830 C previous treatment rolled, softened 6008C 1 h Analysis, wt % C Si Mn 0.40 0.25 0.60 Cr Mo Ni 0.020 0.020 0.65 P S 0.55 2.55 Al Nb V 900 800 Ac3 A 700 A c1 Start F C 600 500 10% 400 50% 90% 300 Ms B Finish 200 M 100 1000 500 200 100 50 Cooling rate at 700 C 20 10 5 2 1 C per min 0 0.2 mm 0.1 Bar diameter (b) 5 0.5 10 1 2 20 5 50 10 100 20 150 200 50 300 100 200 500 1000 2000 mm Air 500 mm Oil 10 20 50 100 150 200 300 500 mm Water FIGURE 6.19 (Continued) 6.1.4 OXIDATION Oxidation takes place as an undesirable accompanying phenomenon during every heat treatment of metals in a noninert atmosphere. Chemical reactions that occur during the oxidation of a metal are generally expressed by the formula 2 2x ! Me þ O2 ÀÀ y Mex Oy y (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 can be expressed as ß 2006 by Taylor & Francis Group, LLC. Chemical composition % C 0.22 Si 0.30 Mn 0.66 P S 0.018 0.011 Al B Cr 0.049 <0.0005 0.56 1000 Cu 0.18 Mo 0.44 N 0.020 Austenitizing temp. 10508C Holding time 15 min 800 71 Temperature, 8C F A 71 69 18 Ac3 Ac1e Ac1b 71 29 29 P 46 Ni 0.15 3 600 26 B 28 Ms 3 20 48 400 M 70 79 69 3 50 200 a) 500 0 0.1 485 1 425 300 340 10 b) 270 240 102 215 103 145 180 165 155 104 105 106 Time, s FIGURE 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.) c Chemical composition % Si 0.30 0.88 Mn 0.66 P 0.018 S 0.011 Al B Cr 0.049 <0.0005 0.56 1000 Cu 0.18 Mo 0.44 N 0.020 Ni 0.15 Austenitizing temp. 830 C Holding time 15 min Accm 800 Ac1a Ac1b Temperature, C 100 100 K A+K P 600 38 90 5 62 70 B 400 200 23 9 95 80 Ms M RA30 RA30 0 0.1 1 RA30 890 880 10 RA30 895 102 30 25 RA3 890 885 870 420 103 390 365 104 330 245 220 105 106 Time, s FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. 900 Air cooling rates, C/s 800 Temperature,8C AC1 1% 1% 700 600 0.6 F+A 500 0.8 B+F+A Ms 400 300 3 1.4 33 19 7 41 23 17 56 (a) (b) 200 Water quenching 1 10 100 Time, s 1000 FIGURE 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) various air cooling rates to approximately Ac1 followed by water quenching. (From E.A. Loria, Met. Technol., 1977, pp. 490–492.) DG0 ¼ GR À GA for 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 creating enthalpy. 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 free creating enthalpy may be calculated as follows: DG ¼ DG0 À 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 values of DG0 are negative with an increasing trend at higher temperatures. In the case of silver, DG0 ¼ 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 was taken as the basis. At higher temperatures the disintegration pressure of Ag2O becomes higher 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 on the surface, the oxide layer that is formed separates the two reaction partners, i.e., the metal and the oxygen. This oxide layer, which is material-specific, becomes thicker with time. There are several formulas expressing the dependence of the oxide layer thickness on time. For higher temperatures a parabolic law is usually used: y¼ ß 2006 by Taylor & Francis Group, LLC. pffiffiffiffi A1 t (6:22) Free standard creating enthalpy ΔG0, KJ/mol 200 O Ag 2 0 4Ag + O2 →2 uO → 2C 2 + O2 4Cu 2NiO O→ 2Ni + 2 –200 –400 –600 2Zn –800 Ti + ZnO 4 Cr + O 2 →2 3 + O2 O2 → 2 Cr 2O 3 →3 TiO 2 2 Al 2O 3 →3 4 Al + O 2 MgO –1000 3 →2 + O2 2Mg –1200 0 500 1000 1500 Temperature T, 8C 2000 FIGURE 6.23 Temperature dependence of the free standard creating enthalpy of some oxidation ¨ reactions 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. This parabolic law is valid for oxidation processes when the rate of oxidation depends on the diffusion 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 and oxygen are not separated, build-up of the layer follows a linear law: y ¼ 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 single valency is involved, e.g., FeO. If more valencies are involved, the oxide layer consists of sublayers 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, created during a 5-h annealing at 10008C (18328F), is shown. Relevant processes during development and build-up of the layer are schematically shown in Figure 6.24b. 6.1.4.1 Scaling of Steel When metallic parts are heated above 5608C (10408F) (this is the temperature at which the ¨ creation of wustite or FeO begins), after creation of the first part of the layer in the starting phase, the reaction follows by diffusion of Fe2þ ions from the steel toward the outside and the diffusion of oxygen ions at the scale–metal interface toward the inside. As time passes, the linear 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 of the steel. Different alloying elements, having different diffusion abilities, have different influences on the oxidation process and the build-up of scale. The chemical composition of the original material on the surface is subject to changes. ß 2006 by Taylor & Francis Group, LLC. (a) (b) Fe2O3 Wüstite FeO Iron Fe Magnetite Fe3O4 Hematite Fe2O3 Fe++ Oxygen O2 O− − Fe++ Fe+++ e− Fe FeO a Fe3O4 e− b e− c d FIGURE 6.24 Oxidation of pure iron. (a) Oxide layer and (b) processes during the build-up of the oxide layer; a, transition of Fe2þ þ 2eÀ from the metal into FeO; b, creation of FeO; c, creation of Fe3O4 and ¨ Fe2O3; 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 into three 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 the ¨ richest oxide compound wustite (FeO), e.g., Ni and Co. After saturation of the basic ¨stite. The metal with oxygen, the outer oxidation of iron begins with the creation of wu alloying elements become richer at the scale–metal 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. Because of the creation of internal oxides of alloying elements, a diffusion barrier builds up against 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 wu W). No inner oxidation takes place. The alloying elements become richer in the basic metal at the scale–metal interface. Depth of the oxide layer, mm/year Figure 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 high oxidation resistance of steels may be achieved by Cr additions of 6–30% by mass. 32 24 800 C 16 700 C 8 600 C 0 0 2 4 6 Cr content, % 8 FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. 6.1.5 DECARBURIZATION Under conditions that cause the oxidation of iron, the oxidation of carbon is also to be expected. Decarburization of a metal is based on the oxidation at the surface of carbon that is dissolved in the metal lattice. It should be noted that, depending on the carbon potential of the surrounding atmosphere, decarburization can take place independently of scaling. However, in heat treatment processes iron and carbon usually oxidize simultaneously. During the oxidation 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., when the equilibrium pressures of the carbon oxides are high enough to break the scale layer or when 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 surface 2. Carbon exchange at the gas–metal interface 3. Diffusion of carbon within the metal Generally the diffusion of carbon within the metal is the most important factor in controlling the rate of decarburization, which after a short starting period follows a parabolic time law. When a mild steel is heated below 9108C (16708F), a surface layer of ferrite is formed that acts as a barrier to carbon transport owing to the very low solubility of carbon in ferrite. Above 9108C (16708F) the steel remains austenitic throughout, and decarburization becomes severe. The model used to represent decarburization in the fully austenitic condition is shown in Figure 6.26. The steel surface is continually oxidized to form a scale, while the carbon is oxidized to form the gases CO and CO2. The scale is assumed to be permeable to these gases, which escape to the atmosphere. The carbon content at a scale–metal interface is assumed to be in equilibrium with the oxygen potential of the scale, which at that position corresponds to the equilibrium between ¨ iron and wustite. The carbon concentration profile in the metal varies from the low surface concentration to the original carbon content within the metal, as shown in Figure 6.26. In using this model, distances are measured from the original metal surface; the instantaneous scale–metal interface lies at the position x ¼ X at time t. This means that scaling has consumed a thickness X of metal during time t. Steel Scale C Original metal surface C0 Cs O X x FIGURE 6.26 Model for decarburization in fully austenitic condition. (From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970.) ß 2006 by Taylor & Francis Group, LLC. To calculate the depth of decarburization, the distribution of carbon in the metal is calculated by solving Fick’s second law for the relevant boundary condition: dC d2 C ¼D 2 dt dx for x > X (6:24) C ¼ C0 , x > 0; t ¼ 0 (6:25) C ¼ Cs , (6:26) x ¼ X; t > 0 Equation 6.25 indicates that initially the carbon concentration was uniform throughout the specimen; Equation 6.26 indicates that the carbon concentration at the metal–scale interface is constant (in equilibrium with the scale). It is assumed that decarburization does not extend to the center of the specimen and that the diffusion coefficient of carbon in austenite is independent of composition; enhanced diffusion down grain boundaries is neglected. Under these conditions the solution at a constant temperature (for which the diffusion coefficient is valid) reads pffiffiffi C0 À C erfc (x=2 2Dt) ¼ (6:27) C0 À Cs erfc (kc =2D)1=2 where C0 is the original carbon content of the metal, Cs is the carbon concentration at the metal–scale interface, D is the carbon diffusion coefficient, t is time, erfc ¼ 1 À erf (here erf is the error function), and kc is the corrosion constant of the metal (kc ¼ X2/2t). Equation 6.27 provides the carbon content within the metal for x > X as a function of time and position. The value of kc for the relevant steel in the relevant atmosphere is expressed as kc ¼ 0:571 exp(À 43,238=RT ) cm2 =s (6:28) Although the variation of the diffusion coefficient of carbon in austenite was ignored in solving Equation 6.24, it was found that the best agreement between calculated and measured carbon 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, which reads D(C ¼ 0) ¼ 0:246 exp(À34,900=RT ) cm2 =s (6:29) A comparison of measured decarburization depths with values calculated using these data showed that with 12 measurements of isothermal treatments between 1,050 and 1,2508C (1922 and 22828F) for time between 900 and 10,800 s, the mean prediction was 97% of the measured value [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 carbon content. To ascertain the effect of scaling rate on decarburization, it may seem logical to try to reduce decarburization by reducing the oxidizing potential of the atmosphere. This is a fallacy, as the carbon concentration at the metal–scale interface is constant in equilibrium with iron oxide as long as scale is present. However, the scaling rate can be affected by changing the atmosphere, and this will affect the observed depth of decarburization. ß 2006 by Taylor & Francis Group, LLC. C, % 0.90 0.925 C0 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.01 0.03 kc cm2/s 0. 4.1 10–8 4.1 10–7 0.05 ×in 0.07 0.09 FIGURE 6.27 Effect of scaling rate on decarburization of an 0.85% C steel after 1.5 h at 10508C. The position where the carbon profile cuts the x-axis indicates the position of the scale–metal interface for the 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 a 0.85% C steel heated for 1.5 h at 10508C (19228C). The carbon profiles are plotted relative to the original metal surface, while the point at which a carbon profile cuts the x-axis indicates the position of the scale–metal interface for the related conditions. The curves refer to kc values of 0, 4.1  10À8, and 4.1  10À7 cm2/s. The depth of decarburization is determined by the position at which C ¼ 0.925C0 (see the horizontal line drawn in Figure 6.27). From Figure 6.27 it appears that with increased values of kc the scale–metal interface on the x-axis shifts progressively toward the inside of the metal, while the depth of decarburization (which is the horizontal distance between the intersection of a carbon profile curve with the horizontal 0.925C0 line and its intersection with the x-axis) is found to decrease as kc increases. These results are shown in Table 6.3. This reveals an interesting situation where, by reducing the oxidation rate, the depth of decarburization is increased yet less metal is wasted. When scaling and decarburization take place simultaneously, decarburization is prevented during the starting phase of scaling. It takes place substantially only after the equilibrium pressures of CO and CO2 increase at increased temperatures and the adhesion strength of the scale (because of its increased thickness) diminishes or the scale becomes porous. 6.1.5.1 The Effect of Alloying Elements on Decarburization Alloying elements may affect decarburization due to their effect on TABLE 6.3 Effect of Scaling Rate on Decarburization Total Depth of Metal Affected 2 Scaling Rate k (cm /s) 0 4.1  10À8 4.1  10À7 Depth of Decarburized Layer cm in. cm in. 0.119 0.130 0.150 0.047 0.051 0.059 0.119 0.109 0.084 0.047 0.043 0.033 Source: From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970. ß 2006 by Taylor & Francis Group, LLC. 1. 2. 3. 4. The The The The ferrite–austenite transformation temperature activity of carbon in solution diffusion coefficient of carbon in solution scaling characteristics of iron Although this subject is a complex one, even when considered only qualitatively, the following statement is generally valid. Decarburization increases with (1) increased rate of carbon diffusion, (2) increased carbon activity, and (3) increased ferrite–austenite transformation temperature. A complication arises due to the fact that during scaling alloying elements tend to concentrate either in the scale or in the metal at the scale–metal interface. In special cases, when strong carbide-forming elements are involved, decarburization may also be influenced by the rate of dissolution of carbides in the matrix. When the alloying elements are less valuable than iron, the possibility of oxidation arises; an external oxide layer may also be formed under circumstances that are normally protective to iron. If either type of oxide formation occurs, the concentration of the alloying element in solution is reduced at the metal surface, 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 rate must 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 the effects of a few common alloying elements may be: Nickel will concentrate at the scale–metal interface and, although the scaling rate may not be greatly affected, the solubility of carbon in the surface layers may be reduced, thus restricting 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 to its effects on carbon activity and the diffusion coefficient. Since manganese is denuded in the 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. This should lead to deeper observed decarburization. Silicon also increases the activity of carbon and therefore increases the tendency of carbon to diffuse out to the scale–metal interface. Thus the general effect expected of silicon is to increase decarburization. Chromium concentrates in the scale, forming spinel, depending on its concentration. In general, scaling rates are reduced. The formation of stable carbides introduces the possibility of a slow carbide decomposition step into the mechanism. At the usual reheating 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 in solution, thus tending to reduce the rate of migration to the surface. There are therefore two conflicting factors. The lower scaling rate would tend to increase the observed decarburization, whereas the reduction of carbon activity would tend to reduce it. The later factor may be expected to predominate and reduce decarburization. 6.1.5.2 Definitions and Measurement of Decarburization The strength of a steel depends on the presence of carbides in its structure; therefore, loss of carbon from the surface softens and weakens the surface layers. In such a case the wear resistance is obviously decreased, and in many circumstances there can be a serious drop in fatigue resistance. ß 2006 by Taylor & Francis Group, LLC. To avoid the real risk of failure or inferior performance of engineering components, it is essential to minimize decarburization at all stages in the processing of steel. This requires inspection and the laying down of specifications for decarburization in various components and semiproducts. The decarburization limits in such specifications must be related to the function of the component and must enable checking by the use of agreed-upon preferably standardized measuring techniques. There are several different definitions of decarburization. A rigorous definition is that the depth of decarburization is the thickness of the layer in which the carbon content is less than that of the core, i.e., the distance from the surface to a boundary at which the carbon content of the core is reached. This boundary corresponds to the asymptote of the graph of carbon content vs. distance from the surface and is therefore somewhat diffuse and difficult to locate with precision. Thus the depth of decarburization according to this definition is difficult to measure reproducibly. A functional definition is that the depth of decarburization is the thickness of the layer in which loss of carbon has a significant effect on properties that affect the functioning of the final component. The limit of this layer can be expressed as a carbon content or a hardness level. Finally, a practical definition of the depth of decarburization is that it is the thickness of the layer in which the structure differs significantly from that of the core. 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 clearly distinguished under the microscope. The depth of complete decarburization is the thickness of 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 the ferrite 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 the inner boundary of the core, is the total depth of decarburization—the sum of the depths of complete 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 that such techniques have to meet are: 1. Ability to measure a clearly defined depth of decarburization, e.g., compatibility with the functional definition of the depth of decarburization 2. Reproducibility of measurement 3. Ease and convenience of measurement Optical metallography is the most useful and convenient method. A cross section of the component or sample around the periphery is examined, and the depth of decarburization is measured from the surface to the practical boundaries of complete and partial decarburization. This method is suitable for ferrite–pearlite structures only. For the metallographic examination of high-speed steels, a method has been established that depends on color staining by means of etching in alcoholic nitric acid (Nital). A polished cross section of annealed high-speed steel is etched in 4% Nital. During the first 30 s in the etchant the specimen surface progresses through a gray color to a purplish-blue, which changes suddenly after about 60 s to a blue-green. Where the functional definition of decarburization calls for the development of the full hardness in the surface layers, the practical boundary is the 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 neutral atmosphere 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 to place the boundary. The specimen is held at that temperature for about 5 s and is then ß 2006 by Taylor & Francis Group, LLC. water-quenched. During this short arrest the decarburized zone, which has an Ms temperature above the temperature of the bath, will partly transform to martensite; the core will remain austenitic. As soon as martensite has formed in the decarburized zone, martensite needles will begin 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 zone and the core, and the boundary can be located with considerable accuracy at any desired carbon content below the original carbon content of the core. Figure 6.28 shows specimens taken from the same hot-rolled rod that have been arrest quenched at different temperatures to place the decarburization boundary at different carbon contents. 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 is compatible 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 accurate determinations 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 area for the hardness survey. A graph of hardness vs. distance from the surface is plotted, and the deviation from the core hardness can be detected. Chemical analysis of successive surface layers is the classical referee method for measuring decarburization. The sample has to be large enough to permit accurate chemical analysis, and yet each surface layer must be fairly thin in order to give an adequate number of points on a graph of carbon content vs. distance from the surface. The graph of carbon content against distance can be used to indicate the first deviation from the core composition or to locate any decarburization boundary. Complete decarburization is not very easy to locate on this graph, because the carbon content of ferrite is too low for very accurate 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 because 0.4 Arrest temp. 3308C 3408C Carbon, % 0.3 3508C 3808C 0.2 4408C 0.1 0 0 0.2 0.4 0.6 0.8 Depth of decarburization, mm FIGURE 6.28 Depth of decarburization to various carbon contents, established by the arrest-quench method. (From ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970.) ß 2006 by Taylor & Francis Group, LLC. the sample size required is much smaller than for other methods. The only limitation is the need to place the spark accurately on a flat area parallel to the original surface and at least 15 mm in diameter. Successive layers have to be exposed by grinding, because the maximum depth measured in one exposure is limited to about 500 mm. 6.1.6 RESIDUAL STRESSES, DIMENSIONAL CHANGES, AND DISTORTION Residual 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 and the resultant moment produced by residual stresses must be zero. Residual stresses are classified, 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 the material, i.e., across several grains. Internal forces resulting from these stresses are in equilibrium with respect to any cross section, and mechanical moments resulting from these stresses are in equilibrium with respect to any axis. Any intervention in the equilibrium of forces and moments of a volume element containing such residual stresses will change the element’s macroscopic dimensions. Residual stresses of the second kind are those homogeneous across microscopically small areas (one grain or subgrain region) and are in equilibrium across a sufficient number of grains. Macroscopic changes in the dimensions of a volume element possessing these stresses may become apparent only if distinct disturbances of this equilibrium occur. Residual stresses of the third kind are those inhomogeneous across microscopically small areas (within several atomic distances of single grains) and are in equilibrium across subgrain regions. No macroscopic changes of the dimensions of the stressed material will result when such equilibria are disturbed. Residual stresses of the first kind are called macroresidual stresses, and those of the second and third kinds are called microresidual stresses. Typical residual stresses of the third kind are stresses connected with dislocations and other lattice defects. An example of residual stresses of the second kind are stresses within grains 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 characterized by the technological processes by which they originate. The main groups of residual stresses are: Casting residual stresses Forming residual stresses Working-out residual stresses Heat treatment residual stresses Joining residual stresses Coating residual stresses In 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 their superposition in a two-phase material after quenching. (RS I–III denote residual stresses of the first to third kinds, respectively.) Estimation of residual stresses in a workpiece is very important because they represent a preloading of the material. There is always a linear superposition of internal (residual) and external stresses, and the resulting stress affects the strength of the material and its deformation behavior. ß 2006 by Taylor & Francis Group, LLC. Phase A Grain ol x x Phase B Cut x − x s RS I s RS II s RS III + + + – – – s RS FIGURE 6.29 All three kinds of residual stresses in a two-phase material after quenching and their ˇˇ ´ superposition (shown schematically). (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) In the case of a dynamic loading on a component, the residual stresses act as a constant preloading. 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 surface but also on the distribution of stresses across the section. Figure 6.30 shows schematically two cases 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 residual stress (compressive) on the surface rapidly decreases below the surface, while in case II the residual stress, although smaller at the surface, decreases more slowly below the surface. The component can withstand the applied load only when the curve c þ b representing the sum of the external and residual stresses does not intersect the fatigue strength (straight line a). In case I, in spite of higher compressive residual stresses at the surface, at a distance below the surface the sum of external and residual stresses is higher than the fatigue strength, and a crack can be expected to form at this point. In case II, although the compressive residual stress is lower at the surface, its distribution below the surface is more favorable and the sum of 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 their multiaxis nature. In practice, the estimation of the sum of external and residual stresses is complicated by the difficulty of determining the direction of the stresses at the critical point of the workpiece. ß 2006 by Taylor & Francis Group, LLC. s+ s+ Case I c a c+b c O b Case II O Distance from surface b a c+b Distance from surface s s FIGURE 6.30 Schematic presentation of superposition of the external load and residual stresses at a ¨ fatigue test. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB ¨ Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) 6.1.6.1 Thermal Stresses in the Case of Ideal Linear-Elastic Deformation Behavior When a metallic body is heated or cooled, as soon as a temperature difference between the surface 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 and hence the greatest residual stresses. Let us therefore discuss thermal stresses due to local and temporal differences in shrinking during quenching of ideal linear-elastic cylinders in which no 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 thermal shrinking behavior, in turn cause locally and temporally different thermal strains and hence 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 a cylinder contracts more rapidly than its core. As a result, the surface zones of the cylinder are subject 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, these stresses are counterbalanced by longitudinal, tangential, and radial compressive stresses within the core of the cylinder. s sr σ s t l l sr sr FIGURE 6.31 Thermal stresses in the surface zone and core of an ideal linear-elastic cylinder during ˇˇ ´ rapid cooling (quenching). (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) ß 2006 by Taylor & Francis Group, LLC. T Core Surface t max ΔT t20 log t ΔTmax s sh l log t Surface 0 log t Core FIGURE 6.32 Top to bottom: time–temperature history, temperature difference between surface and core, and development of longitudinal stresses during transformation-free quenching of an ideal linearˇˇ ´ 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 temperature–time history at the very surface and at the core of the cylinder, the temperature difference between surface and core, and the development of longitudinal stresses during quenching of an ideal linear-elastic cylinder. The largest temperature difference DTmax is attained at t ¼ tmax, where the slopes of temperature–time curves are identical for the core and the surface. Obviously, the surface reaches its maximum thermal stress before t ¼ tmax; the core, however, reaches its maximum later than t ¼ tmax. The magnitude of the developed longitudinal stresses depends on cylinder diameter, as shown in Figure 6.33 for cylinder diameters of 30, 50, and 100 mm, when the cylinders were quenched from 8008C (14728F) in water at 208C (688F). Because the maximum temperature difference between surface and core occurs later for the larger diameter cylinders, the maximum stresses also occur later for larger diameters. The longitudinal surface stress maximum always occurs at t < tmax, whereas those of the core occurs later than tmax. At t < 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 temperature gradients are established, which cause large compressive stresses in the core. Upon reaching the temperature balance at 208C (688F) (t ¼ t20), the ideal linear-elastic cylinders are free of residual stresses. 6.1.6.2 Transformational Stresses Let us consider the development of pure transformational stresses in a material whose coefficient of thermal expansion is zero. Furthermore, assume that if in the course of quenching the martensite start temperature Ms is passed, complete martensitic transformation occurs, with corresponding volume increase. ß 2006 by Taylor & Francis Group, LLC. 800 D = 100 mm Longitudinal stress, N/mm2 600 50 400 Surface 30 200 0 –200 100 50 Core 30 –400 –600 2-10–1 4 6 8 1 2 4 6 8 10 2 2 4 6 10 Time t , s FIGURE 6.33 Dependence of longitudinal stresses at surface and core of ideal linear-elastic cylinders on their diameters, when quenched in water from 800 to 208C. Calculated for unalloyed steel with medium ˇˇ ´ carbon content. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) The temperature–time curves for the surface and core of a cylinder of such a material are shown in Figure 6.34. After passing the Ms temperature at time t ¼ t1, as a consequence of transformation-induced volume increase, compressive transformational stresses develop at the surface. These stresses within the surface zone must be compensated for by tensile transformational stresses within the core of the cylinder. The magnitudes of both stresses increase in the course of further surface cooling. T Core Surface Ms t1 s tr l t2 t 20 log t Core + 0 log t − Surface FIGURE 6.34 Temperature–time history and development of longitudinal transformation stresses, when ˇˇ ´ 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.) ß 2006 by Taylor & Francis Group, LLC. s l sh + s l tr Core Surface 0 Shrinking ti Transformation log t Core Surface s l sh + sl tr Surface 0 ti t 20 log t Core FIGURE 6.35 Combined thermal (shrinking) and transformation stresses during quenching of an ideal ˇˇ ´ linear-elastic cylinder that transforms from austenite to martensite. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) When the core temperature reaches Ms at time t ¼ t2, a transformation-induced volume increase occurs in the core, which leads to a reduction of the tensile stresses present there. The surface compressive stresses are correspondingly reduced. After reaching temperature equalization at t ¼ t20, the same amounts of martensite are present across the whole cylinder, so that finally a residual stress-free state is established. If, however, different amounts of martensite are formed within distinct areas, also under the idealized assumptions made here, some transformational residual stresses will remain. In addition to the longitudinal stresses, tangential and radial residual stresses are caused by structural transformation. Within the surface zone, tangential compressive and radial tensile stresses are to be expected, while in the core all components should be tensile stresses. When thermal (shrinking) and transformational stresses act simultaneously during quenching 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 graph shows the time dependence of the longitudinal components of thermal and transformational stresses at surface and core. The lower graph shows the time dependence of the total stress after the formal superposition of the two. The initiation of martensitic transformation immediately 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 ¼ t20 the tensile 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. 6.1.6.3 Residual Stresses When Quenching Cylinders with Real Elastic–Plastic Deformation Behavior In real practice there is no ideal linear-elastic deformation behavior as assumed above. The yield strength (Ry) of metallic materials, which limits the elastic deformation range, is strongly temperature-dependent and decreases with increasing temperature. ß 2006 by Taylor & Francis Group, LLC. 600 Rm 16MnCr5 17CrNiMo6 R y, R m, N/mm2 400 Ry 200 0 200 400 600 Temperature T, C 800 1000 FIGURE 6.36 Yield strength (Ry) and tensile strength (Rm) of the steels DIN 16MnCr5 and DIN ˇˇ ´ 17CrNiMo6 as a function of temperature. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) At any temperature, plastic deformations will develop when stresses surpass the corresponding yield strength. The ultimate tensile strength, which limits the uniaxial loading capacity of the material, is also temperature-dependent as shown in Figure 6.36 for two low-alloy steels. During quenching of a cylinder, biaxial longitudinal and tangential stresses develop in its surface zone, whereas triaxial longitudinal, tangential, and radial stresses develop in the cylinder core. Plastic deformations can occur only if the local equivalent stresses equal or exceed the yield strength of the material at the corresponding temperature. Equivalent stresses can be calculated according to various hypotheses. Assuming the validity of the van Mises criterion, the equivalent stress of a triaxial stress state, given by the principal stresses s1, s2, s3, is 1 eq ¼ pffiffiffi [(1 À 2 )2 þ (2 À 3 )2 þ (3 À 1 )2 ]1=2 2 (6:30) During quenching of a cylinder in its surface zone, s1 ¼ sl and s2 ¼ st, while in its core s1 ¼ sl, s2 ¼ st, and s3 ¼ sr. The condition for the onset of plastic deformation will be fulfilled when seq ¼ Ry. The local shrinking and transformational stress components and consequently the equivalent stress (seq) depend on temperature, cooling conditions, geometry, and the mechanical and thermal properties of the material, and the yield strength (Ry) depends on temperature and the structure of the material. The temperature dependence of the yield strength is obviously of particular importance for the stresses that result upon quenching. Figure 6.37 shows the temperature–time history and development of yield strength for surface and core of a cylinder during quenching. Figure 6.37a depicts the case of transformation-free cooling, and Figure 6.37b is valid for cooling with martensitic transformation. To determine the occurrence of plastic deformations at any instant, the local yield strength must be compared with the local equivalent stress. Because plastic deformations never occur homogeneously over the whole cross section of the cylinder, residual stresses always remain after temperature equalizations. Plastic deformations can be caused by either thermal (shrinking) stresses or transformational stresses or by a combination of the two. ß 2006 by Taylor & Francis Group, LLC. T, Ry T, Ry Core Core Surface Surface surface core Ry Ry surface Ry M1 Iog t t max (a) MS core Ry t1 (b) t2 Iog t FIGURE 6.37 Temperature–time history and development of yield strength for surface and core during ˇˇ ´ 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.) 6.1.6.3.1 Thermal (Shrinking) Residual Stresses Figure 6.38 shows the cooling curves for the surface and core of a cylinder during quenching without martensitic transformation and the temperature- (and time-) dependent yield strengths, 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 the core temperature (Figure 6.38a). As a result, longitudinal tensile stresses develop at the surface and compressive stresses develop at the core. If they were elastically accommodated, T, Ry Core Surface surface core Ry Ry (a) log t t max σ sh, R l y surface Surface 0 Ry t max Core (b) log t core Ry σ sh l Surface t max 0 (c) log t Core FIGURE 6.38 Longitudinal thermal (shrinking) residual stresses when quenching a cylinder. (From ˇˇ ´ B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) ß 2006 by Taylor & Francis Group, LLC. their development would be as shown in Figure 6.38b. However, because of the temperature dependence of the yield strengths for surface and core, neither the surface nor the core can withstand these stresses without plastic deformation, and so the surface zone is plastically extended and the core is plastically compressed. After the time t ¼ tmax, the temperature of the core decreases faster than that of the surface, leading to a reduction of the magnitudes of shrinking stresses in both regions. However, the stress values of core and surface reach zero at different instants, as they can no longer coexist at the same time in a stress-free state because of 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 are opposed 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 end of quenching, and hence thermal (shrinking) residual stresses remain that are compressive at the surface and tensile in the core, as depicted in Figure 6.38c. 6.1.6.3.2 Transformational Residual Stresses Figure 6.39 shows cooling curves for surface and core when quenching a cylinder that, upon cooling below the Ms temperature, transforms completely to martensite. For simplicity, it is assumed that no thermal (shrinking) stresses occur. Figure 6.39b shows the yield strengths for surface and core, showing their strong increase with the onset of martensitic transformation. T, Ry Core Surface Ms core Ry surface Ry Mf (a) s ltr, Ry t1 t2 log t core Ry Core 0 t1 t2 log t Surface (b) surface Ry s ltr Core 0 (c) t1 t2 log t Surface ˇˇ ´ 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.) ß 2006 by Taylor & Francis Group, LLC. The surface of the cylinder starts to transform to martensite at t ¼ t1. At that time the volume 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 for by tensile stresses at the core. From Figure 6.39b it can be concluded that both areas plastically deform. In the course of further cooling, the tensile stressed core reaches Ms at t ¼ t2. The immediate volume increase reduces both the tensile stresses of the core and the compressive 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. Upon further cooling the still existing volume incompatibilities between surface and core create transformational stresses of opposite sign to those that are produced by the plastic deformations. After reaching temperature equalization, compressive residual stresses remain in the core 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 under local tensile or compressive stresses may enhance the local strains. 6.1.6.3.3 Hardening Residual Stresses When austenitized steel cylinders are quenched to room temperature, both thermal (shrinking) and transformational stresses develop, causing hardening residual stresses, which cannot be described by simply superimposing the shrinking and transformational stresses. Of fundamental importance is the fact that any local martensitic transformation coupled with a volume increase always shifts the existing stress (irrespective of its sign) to more negative values. As a reaction, for reasons of equilibrium, the unaffected material zones react with positive stress changes. Structural transformations that occur in tensile-stressed material regions are therefore inevitable to reduce the stresses, while transformations that take place in compressively stressed zones always enhance the (negative) values of the stresses. Consequently, because the thermal (shrinking) stresses of core and surface change sign in the course of cooling during the time interval tc,0Àts,0 as depicted in Figure 6.40a, the positions of the initiation time of transformation at the surface (ts,i) and in the core (tc,i) relative to this time interval are of key importance for the hardening residual stresses that will remain at the end of quenching. The average time that passes before the quenching stresses invert is t0 ¼ (1=2)(ts,0 þ tc,0 ) (6:31) Because for full-hardening steel cylinders the time for surface transformation (ts,i) always occurs earlier than the time for core transformation (tc,i), it is appropriate to distinguish between 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 time and remaining longitudinal residual stress distributions across the section of cylinder ß 2006 by Taylor & Francis Group, LLC. Steel heat treatment sl s l RS Surface to 0 t c.0 t so log t Core (a) sl t 0 < t s, i < t c, i 1 Surface 0 s l RS log t Core 1 (b) sl s l RS Surface t 0 ≈ t s, i < t c, i 0 2 2 log t Core (c) sl s l RS Surface 3 3 t s, i < t c, i ≈ t 0 0 log t Core (d) sl s l RS Surface 4 4 t s, i < t c, i < t 0 0 log t Core (e) (Core) 0 (Surface) 0.5 AIA 1.0 FIGURE 6.40 Different possibilities of generation and development of hardening residual stresses (b–e) compared to pure thermal (shrinking) residual stresses (a), when quenching a cylinder with real elastic– ˇˇ ´ plastic deformation behavior. (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, Springer-Verlag, New York, 1992.) specimens with real elastic–plastic deformation behavior after complete temperature equalization at the end of the quenching process. Figure 6.40a shows a transformation-free quenching, and Figure 6.40b–Figure 6.40e demonstrate the combined effects of thermal (shrinking) and transformational processes. The numbers 1–4 depict the initiation of transformation at the surface, while 1’–4’ represent that of the core. Figure 6.40b illustrates the case when both surface and core transform after t0. At the end of 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 transforms slightly before t0 and the core transforms later. At the end of this cooling process, both core ß 2006 by Taylor & Francis Group, LLC. Temperature and surface remain under compressive residual stresses, while the regions in between are subjected to tensile residual stresses. Figure 6.40d illustrates the case when the surface transforms before t0 and the core at about t0. At the end of this cooling process, tensile surface residual stresses and compressive core residual stresses remain. Figure 6.40e illustrates the case when both surface and core transform before t0. In this case 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 during further cooling. Martensitic transformation in the core takes place when tensile stresses are acting there. This again causes stress inversions in the surface zone and in the core. At the end of this cooling process, tensile stresses at the surface and compressive stresses in the core remain. When full hardening of equal-sized cylinders with different Ms temperatures is compared with respect to residual stress distributions, one finds that cylinders made of steels with low Ms temperatures show tensile surface residual stresses, whereas cylinders made of steel with high Ms temperatures give compressive surface residual stresses, as schematically illustrated in Figure 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 and the 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 Ms temperature. When high shrinking stresses and high Ms values act together, no secondary stress inversion occurs during further cooling, and ultimately compressive residual stresses remain within the surface zone. On the basis of the preceding discussion, the whole range of expected hardening residual stress distributions in quenched steel cylinders can be divided into three main groups, as Ms,1 Ms,2 Ms,3 Surface stresses 0 log t Ms increasing 3 2 0 log t 1 FIGURE 6.41 Influence of different Ms temperatures on the development of surface residual stresses. ˇˇ ´ (From B. Liscic, H.M. Tensi, and W. Luty (Eds.), Theory and Technology of Quenching, SpringerVerlag, New York, 1992.) ß 2006 by Taylor & Francis Group, LLC. s l RS s l RS s l RS Transformation Under Under compression tension in the in the surface core 0 Transformation Under compression in the surface 0 0 Shrinking type c Transition type s 0 0.5 Transformation Under Under tension compression in the in the surface core c 1.0 0 Transformation type s 0.5 1.0 Ratio of cross section c 0 s 0.5 1.0 Cylinder diameters in mm for residual stresses of Steel Quenching process 850 C Ck 45 850 C 20 C H2O 60 C oil Shrinking type Transition type 100···30 50 15 30 Transformation type 5 10 ˇˇ ´ 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 under existing stress states will affect the residual stress distribution. It should be emphasized that the residual stress distributions that are created during quenching of cylinders with different diameters but made of the same steel can be shifted from the transformation type to the shrinking type with increasing cylinder diameter as well as with the higher quenching intensity, i.e., higher cooling rates. Some cylinder diameter and quenchants are specified for the unalloyed steel DIN Ck45 where the basic residual stress types occur. The above statements and specifically the principle that local stresses are shifted to more negative values due to transformation-induced volume increase also hold for all nonmartensitic transformations that are accompanied by volume changes. In the individual case, the effect of volume changes on the final residual stress state depends on when the transformations start at the core and surface relative to time t0. 6.1.6.4 Dimensional Changes and Distortion during Hardening and Tempering As a consequence of thermal (shrinking) stresses and transformational stresses, changes occur in both the shape and size of workpieces during hardening and tempering. Because there are many factors that influence dimensional changes and distortion, the most difficult problem in practice is to predict the amount of dimensional changes and distortion. It is likely that computer modeling of the quenching process, which can account for the influence of all relevant factors, will in the future enable more precise prediction. Let us therefore discuss only some basic mechanisms of dimensional changes and distortion during hardening and tempering. 6.1.6.4.1 Influence of Thermal (Shrinking) Stresses Because of thermal (shrinking) stresses during quenching, generally all bodies whose shape is different from a sphere tend by deformation to assume a spherical shape, which offers the ß 2006 by Taylor & Francis Group, LLC. least resistance during deformation. This means, at the practical level, that bodies that have the shape of a cube will assume spherically distorted sides, bodies with the shape of a prism will become thicker and shorter, and plates will shrink in area and become thicker. These deformations are greater with greater temperature differences between the surface and the core, i.e., with higher quenching intensity (which also corresponds to bigger differences between the austenitizing temperature and the temperature of the quenchant); with greater cross-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 an austenitic steel, in which the martensitic transformation can be disregarded. Figure 6.43 shows, according to Frehser and Lowitzer [15], the effect of different quenching intensities on dimensional changes and distortion of plates made of low-carbon steel (0.10% C) after quenching in water, oil, molten salt bath, and air. In Figure 6.43a the plate is solid, and in Figure 6.43b the plate has an inner square hole of 100  100 mm. The outer full lines denote the original size of each plate. To illustrate the dimensional changes more clearly, they have been drawn to a larger scale (see the 0.4-mm scale). From this figure it is evident that the more drastic the quench, the greater are the dimensional changes and distortion. Figure 6.44 shows that a greater difference between the austenitizing temperature and the temperature of the quenchant causes greater dimensional changes and distortion. Figure 6.45 shows the effect of the high-temperature strength of the material. The steel having the greatest high-temperature strength (18/8 steel) exhibits the highest dimensional stability. ∧ 0.4 mm change in size = ∧ 0.04 mm = 20 Ÿ 100 (b) 200 (a) 920 C/water 920 C/oil 920 C/molten bath at 220 C 920 C/air FIGURE 6.43 Dimensional changes and distortion of plates made from low-carbon steel (0.10% C) after cooling in water, oil, molten salt bath, and air. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. 60 200 60 920 C/oil 800 C/oil 20 20 ∧0.04 mm = FIGURE 6.44 Effect of difference between the austenitizing temperature and the temperature of the quenchant 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.) 6.1.6.4.2 Influence of Transformation Stresses During heating and cooling, steels pass through various structural transformations accompanied by volume changes. These changes are usually studied by using a dilatometer and are registered as changes in length of the specimen, as shown, for example, in Figure 6.46 for eutectoid steel. During heating a continuous increase in length occurs up to Ac1, where the steel shrinks as it transforms to austenite. After the austenite formation is completed, the length increases again. However, the expansion coefficient for austenite is not the same as the expansion coefficient for ferrite. 17% Cr steel 920 C/water 18/8 steel 920 C/water 200 200 0.1% C steel 920 C/water ∧ 40 40 Dimensional change scale: = 0.04 mm FIGURE 6.45 Dimensional changes and distortion after quenching steel plates of different composition from 9208C in water. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. Change in length Ac1 T3 T2 T1 + 0 − −200 Ms 0 200 400 600 Temperature, C 800 1000 FIGURE 6.46 Dilatometer curves showing change in length during heating and rapid cooling of a eutectoid (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 Ms temperature, the volume increases and the length of the specimen therefore increases. After cooling to room temperature, most martensitic steels contain some retained austenite, the amount of which increases with increased carbon content, with higher austenitizing temperature, and with the amount of some alloying elements dissolved during austenitization. The larger the quantity of retained austenite contained in the steel after hardening, the smaller the increase in volume and in length of the specimen. Various structural constituents have different densities and hence different values of specific volume, as shown in Table 6.4. The amount of carbon dissolved in austenite, in martensite, or in different carbides has a relatively strong effect on the specific volume as the formulas for calculating specific volume in this table indicate. When calculating the changes in volume that take place during the transformations of different structural phases, the carbon content must be taken into account, as shown in Table 6.5. TABLE 6.4 Specific Volume of Phases Present in Carbon Tool Steels Phase or Phase Mixture Austenite Martensite Ferrite Cementite e-Carbide Graphite Ferrite þ cementite Low carbon content martensite e-carbide Ferrite þ e-carbide Range of Carbon (%) 0–2 0–2 0–0.02 6.7 + 0.2 8.5 + 0.7 100 0–2 0.25–2 0–2 Calculated Specific Volume at 208C (cm3/g) 0.1212 þ 0.0033  (%C) 0.1271 þ 0.0025  (%C) 0.1271 0.130 + 0.001 0.140 + 0.002 0.451 0.271 þ 0.0005  (%C) 0.1277 þ 0.0015  (%C – 0.25) 0.1271 þ 0.0015  (%C) Source: From K.E. Thelning, Steel and its Heat Treatment, 2nd ed., Butterworths, London, 1984. ß 2006 by Taylor & Francis Group, LLC. TABLE 6.5 Changes in Volume during Transformation to Different Phases Transformation Spheroidized pearlite ! austenite Austenite ! martensite Spheroidized pearlite ! martensite Austenite ! lower bainite Spheroidized pearlite ! lower bainite Austenite ! upper bainite Spheroidized pearlite ! upper bainite Change in Volume (%) À4.64 þ 2.21  (%C) 4.64 À 0.53  (%C) 1.68  (%C) 4.64 À 1.43  (%C) 0.78  (%C) 4.64 À 2.21  (%C) 0 Source: 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 of carbon dissolved therein, and using the data from Table 6.5, one can calculate the changes in volume that occur during hardening. If the steel contains undissolved cementite, this volume has to be deducted during the calculation. The following equation should be used: DV 100 À Vc À Va Va  1:68C þ ( À 4:64 þ 2:21C ) ¼ 100 100 V (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 À Vc À Va is the amount of martensite in vol%, and C is the carbon dissolved in austenite and martensite, respectively, in % by weight. The increase in volume during martensitic transformation depends not only on the carbon content 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-alloy oil-hardening steels, while the high-alloy ledeburitic Cr alloy steels show the least volume increase 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. Because the retained austenite producing volume contraction (compared to the original volume) counteracts the volume increase caused by martensitic transformation, the austenitizing temperature may influence the volume changes during hardening. It should also be noted that engineering steels are not isotropic materials (because of the rolling process they have undergone), which means that the linear change occurring during hardening will not be the same in the direction of rolling as in the direction perpendicular to it. 6.1.6.4.3 Dimensional Changes during Tempering During tempering, relaxation as well as structural transformations occur, which change the volume of the hardened steel and its state of stress. Martensite decomposes to form ferrite and cementite, which implies that there is a continuous decrease in volume. The continuous decomposition 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 of structural constituents on the volume changes during tempering of a hardened steel. The dashed curves represent increases in volume during different tempering stages. The retained austenite, which in carbon steels and low-alloy steels is transformed to bainite in the second stage of tempering at about 3008C (5728F), results in an increase in volume. ß 2006 by Taylor & Francis Group, LLC. Water-hardening steels e.g., DIN C100 Oil-hardening steels e.g., DIN 90MnV8 Air-hardening steels e.g., DIN X210Cr12 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 Change in vol% 0.8 1.0 FIGURE 6.47 Volume changes of different steels during hardening when martensite is formed across the ¨ whole section. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB ¨ Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) When high-alloy tool steels are tempered at 500–6008C (932–11128F), very finely distributed carbides are precipitated. This gives rise to a stress condition that results in increased hardness and greater volume. Simultaneously with the precipitation of carbides the alloy content of the matrix is reduced, which implies that the Ms point of the retained austenite will be raised to higher temperatures. During cooldown from the tempering, the retained austenite will 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 tempering temperature. For low-alloy steels (105WCr6, see curve 1 of Figure 6.49), one can easily recognize the particular tempering stages. At low tempering temperatures (first tempering stage), a volume contraction takes place as a consequence of e-carbide precipitation. At higher tempering temperatures (second tempering stage), transformation of the retained Increase in volume Decomposition of martensite to ferrite and cementite Retained austenite to martensite Retained austenite to bainite 0 100 200 Carbide precipitation 300 400 500 600 700 800 Tempering temperature, C FIGURE 6.48 Schematic presentation of the effect of changes of structural constituents on volume changes during tempering of hardened steel. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. Change in length, % 0.15 0.10 0.05 0 –0.05 –0.10 –0.15 0 2 3 1 100 200 300 400 6 500 5 4 1: 2: 3: 4: 5: 6: D IN D IN D IN D IN D IN D IN 105WC r6 40C rMOV21.14 210C rW46 X 100C rMoV 5.1 50N i C r13 165C rMoV 46 600 Tempering temperature, C FIGURE 6.49 Change in length of different steels during tempering as a function of tempering tem¨ perature. (Designation of steels according to DIN.) (From H.J. Eckstein (Ed.), Technologie der Warme¨ behandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) austenite again causes a certain volume increase, and in the third tempering stage the progressive decomposition of martensite leads to the volume decrease. For high-alloy tool steels (e.g., 210CrW46, curve 3 of Figure 6.49), a stabilization of austenite is evident, so that the effect of the volume increase (due to austenite–bainite or austenite–martensite 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 found after tempering. It should be noted that the changes in length shown in Figure 6.49 represent only the order of magnitude of the expected changes, because the actual value depends in each case on the specific heat treatment conditions. The austenitizing temperature, which determines the amount of carbon dissolved and the amount of retained austenite, has a strong influence on expected volume changes. 6.2 6.2.1 ANNEALING PROCESSES STRESS-RELIEF ANNEALING Stress-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 a workpiece without intentionally changing its structure and mechanical properties. Residual stresses in a workpiece may be caused by 1. Thermal factors (e.g., thermal stresses caused by temperature gradients within the workpiece 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 simultaneous existence of thermal and transformational stresses (e.g., during the solidification of liquid metals, hot forming, hardening, or welding). Thermal stresses are always directly proportional to the existing temperature gradient, which further depends on the cross-sectional size and on the heating or cooling rate. In workpieces made of steel, for the above reasons, local residual stresses may amount to between about 10 N/mm2 and values close to the yield strength at room temperature. The consequences of residual stresses may include 1. 2. 3. 4. Dimensional changes and warpage of the workpiece Formation of macroscopic and microscopic cracks Asymmetric rotation of shafts Impairment of the fatigue strength of engineering components ß 2006 by Taylor & Francis Group, LLC. Residual stresses in a workpiece can be reduced only by a plastic deformation in the microstructure. This requires that the yield strength of the material be lowered below the value of the residual stresses. The more the yield strength is lowered, the greater the plastic deformation 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 increasing temperature, 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. For plain carbon and low-alloy steels this temperature is usually between 450 and 6508C (842 and 12008F), whereas for hot-working tool steels and high-speed steels it is between 600 and 7508C (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-relief annealing should be left with a machining allowance sufficient to compensate for any warping resulting from stress relief. When dealing with hardened and tempered steel, the temperature of stress-relief annealing should be about 258C (778F) below that used for tempering. If the tempering temperature was quite low, after stress-relief annealing quite a high level of residual stresses will remain. In some other cases, for instance with a gray iron, the maximum temperature of the stress-relief annealing should be limited because of possible strength loss. Therefore gray iron must not be stress-relief annealed above 5508C (10228F). In the heat treatment of metals, quenching or rapid cooling is the cause of the greatest residual stresses. A high level of residual stress is generally to be expected with workpieces that have a large cross section, are quenched at a high cooling rate, and are made of a steel of low hardenability. In such a case high-temperature gradients will arise on the one side, and on the other side structural transformations will occur at different points of the cross section at different 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 yield strength of the material. Because of this fact, steels that have a high yield strength at elevated temperatures can withstand higher levels of residual stress than those that have a low yield strength at elevated temperatures. 50 40 800 A 30 20 600 Rm 10 Elongation, % Yield strength and ultimate tensile strength, MPa 1000 0 400 sso 200 ssu 0 –200 –100 0 100 200 300 400 500 600 Temperature, C FIGURE 6.50 Change in some mechanical properties of low-carbon unalloyed steel with increasing temperature, according to Christen. A, Elongation; Rm, ultimate tensile strength; sso, upper yield ¨ strength; ssu, lower yield strength. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung von ¨ Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) ß 2006 by Taylor & Francis Group, LLC. Increase of the yield strength, N/mm2 120 Mo 100 80 V 60 Ti Mn 40 Cu 20 Cr 0 300 Ni 350 400 450 Temperature, C 500 550 FIGURE 6.51 Increase in yield strength at elevated temperatures when 0.5% of each alloying element ¨ indicated is added to an unalloyed steel. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigung¨ stechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) The level of yield strength at elevated temperatures depends on the alloying elements in the steel. Figure 6.51 shows the increase in yield strength at temperatures of 300–5508C (572– 10228F) when 0.5% of each element was added to an unalloyed steel. It can be seen from this diagram that additions of Mo and V are most effective in increasing the yield strength at elevated temperatures. To reduce residual stresses in a workpiece by stress-relief annealing, a temperature must be reached above the temperature corresponding to the yield strength that is adequate to the maximum of the residual stresses present. In other words, every level of residual stress in a workpiece corresponds to a yield strength that in turn depends on temperature. In addition to temperature, soaking time also has an influence on the effect of stress-relief annealing, i.e., on the reduction of residual stresses, as shown in Figure 6.52. The relation between temperature and soaking time during stress-relief annealing can be described by Hollomon’s parameter: P ¼ T (C þ log t) (6:33) where P is Hollomon’s parameter (heat treatment processes with the same Hollomon parameter value have the same effect), C is the Hollomon–Jaffe constant, T is temperature (K), and t is time (h). The Hollomon–Jaffe constant can be calculated as C ¼ 21:3 À (5:8  % carbon) (6:34) Figure 6.53 shows (according to Larson–Miller method) calculated values of the yield strength at elevated temperatures (for 0.2% strain) for three grades of alloyed structural steels for hardening and tempering (designations according to DIN). Using this diagram, the abscissa of which represents the actual Hollomon parameter P, knowing the temperature and time of the stress-relief annealing, one can read off the level of residual stresses that will remain in the workpiece after this annealing process, i.e., the level up to which the residual stresses will be reduced by this stress-relief annealing. If, for instance, for DIN 24CrMoV5.5 steel, a ß 2006 by Taylor & Francis Group, LLC. 10 Reduction of residual stresses, % 20 30 40 1h 10 h 24 h 48 h 50 60 70 80 90 100 200 300 400 500 600 Temperature, C 700 FIGURE 6.52 Effect of soaking time (at different temperatures) of stress-relief annealing on the ¨ reduction 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.) 400 200 Cr 30 V5 Ni Mo 100 80 .11 60 50 40 rM 7.4 5.5 oV Mo iCr a b 20 C 24 30 28N Yield strength or minimum residual stress after stress-relief annealing, N/mm2 P = T−(20+log t)−10−3 h 1015 Holding time, t (a) (b) 0.1 5 10 20 16 17 18 19 20 21 Hollomon's parameter P 22 23 550 600 650 700 550 600 550 650 700 600 650 550 600 650 Temperature T, C 700 700 FIGURE 6.53 Yield strength at elevated temperatures (for 0.2% strain) calculated according to the Larson–Miller method for three grades of alloyed structural steels for hardening and tempering (designations according to DIN). (a) Calculated values and (b) experimentally obtained values. (From ¨ ¨ G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 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. Higher temperatures and longer times of annealing may reduce residual stresses to lower levels, as can be seen from Figure 6.53. As in all heat treatment processes where Hollomon’s parameter is involved, selection of a higher temperature may dramatically shorten the soaking time and contribute substantially to the economy of the annealing process. Dealing with structural steels for hardening and tempering, the stress-relief process and the tempering process can be performed simultaneously as one operation, because Hollomon’s parameter is also applicable to tempering. In such a case the stress-relief diagram may be used in combination with the tempering diagram to optimize both the hardness and the level of reduced 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 internal stresses arise. New stresses that may be induced during cooling depend on the cooling rate, on the cross-sectional size of the workpiece, and on the composition of the steel. Figure 6.54 shows the effect of cooling rate and cross-sectional diameter of forgings made of a CrMoNiV steel on the level of tangential residual stresses after stress-relief annealing. A general conclusion about stress-relief annealing is the following: In the temperature range 450–6508C (842–12008F), the yield strength of unalloyed and low-alloyed steels is lowered 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 be considerable. While unalloyed and low-alloy steels with Ni, Mn, and Cr after stress-relief annealing above 5008C (9328F) may get the residual stresses reduced to a low level, steels alloyed with Mo or Mo þ 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 at elevated temperature. 6.2.2 NORMALIZING m m m 60 0 m m. 80 0m 100 80 =1 000 mm 120 Dia Tangential residual stresses, N/mm2 Normalizing or normalizing annealing is a heat treatment process consisting of austenitizing at temperatures of 30–808C (86–1768F) above the Ac3 transformation temperature (for 60 0 40 m m 0m 20 40 20 0 0 10 20 30 40 50 60 Average cooling rate to 400, C/h 80 FIGURE 6.54 Tangential residual stresses in a CrMoNiV alloy steel depending on the cooling rate and ¨ cross-section diameter. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, ¨ Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. Temperature, C 1000 b Ac3 c 800 Ac1 a 600 d 400 200 0 Time FIGURE 6.55 Time–temperature regime of normalizing. a, Heating; b, holding at austenitizing tem¨ perature; c, air cooling; d, air or furnace cooling. (From G. Spur and T. Stoferle (Eds.), Handbuch der ¨ Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) hypoeutectoid steels) followed by slow cooling (usually in air), the aim of which is to obtain a fine-grained, uniformly distributed, ferrite–pearlite structure. Normalizing is applied mainly to unalloyed and low-alloy hypoeutectoid steels. For hypereutectoid steels normalizing is performed only in special cases, and for these steels the austenitizing temperature is 30–808C (86–1768F) above the Ac1 transformation temperature. Figure 6.55 shows the thermal cycle of a normalizing process, and Figure 6.56 shows the range of austenitizing temperatures for normalizing unalloyed steels depending on their carbon 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 a result of heating to a high temperature, e.g., for forging or welding. Figure 6.57 shows the effect of grain refining by normalizing a carbon steel of 0.5% C. Such grain refinement and 1200 1147 C E 1100 γ Temperature, C 1000 G 900 γ + Fe3C 800 α +γ 723 C 700 P 0 0.4 Pearlite a + Pearlite 600 500 S K Pearlite + Fe3C 0.8 1.2 1.6 Carbon content, % 2.0 2.4 FIGURE 6.56 Range of austenitizing temperatures for normalizing unalloyed steels depending on their carbon content. (Temperature range above the line S–E is used for dissolution of secondary carbides.) ¨ a, ferrite; g, austenite; Fe3C, cementite. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigung¨ stechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 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 500Â. (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 the mechanical properties of the workpiece or (previous to hardening) to obtain better and more uniform results after hardening. In some cases, normalizing is applied for better machinability of low-carbon steels. A special need for normalizing exists with steel castings because, due to slow cooling after casting, a coarse-grained structure develops that usually contains needlelike ferrite (Wid¨ mannstatten’s structure), as shown in Figure 6.58. A normalizing treatment at 780–9508C (1436–17428F) (depending on chemical composition) removes this undesirable structure of unalloyed and alloyed steel castings having 0.3–0.6% C. After hot rolling, the structure of steel is usually oriented in the rolling direction, as shown in Figure 6.59. In such a case, of course, mechanical properties differ between the rolling direction and the direction perpendicular to it. To remove the oriented structure and obtain the 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, a heterogeneous 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 fur Grundstoffindustrie, Leipzig, 1987.) ß 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 100Â. (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 the newly formed ferrite–pearlite structure during normalizing treatment can be explained with the following mechanism. At normalizing, the steel is subjected first to a a ! g (ferrite– pearlite to austenite) transformation, and after the holding time at austenitizing temperature, to a recurring g ! a (austenite to ferrite–pearlite) transformation. The effect of normalizing depends on both austenitization and cooling from the austenitizing temperature. During austenitizing a far-reaching dissolution of carbides is aimed at, but this process competes with the growth of austenite grains after complete carbide dissolution, which is not desirable. Besides the carbide dissolution, the degree of homogenization within the austenite matrix is important for obtaining a new arrangement of ferrite and pearlite constituents in the structure after normalizing. Both dissolution and homogenizing are time- and temperaturedependent diffusion processes that are slower when the diffusion paths are longer (higher local differences in carbon concentration) and the diffusion rates are smaller (e.g., increasing amounts of alloying elements). Therefore, especially with alloyed steels, lower austenitizing temperatures and longer holding times for normalizing give advantages taking into account the austenite grain growth. As shown in Figure 6.60, high austenitizing temperatures result in a coarse-grained austenite structure, which yields a coarse structure after normalizing. Holding time at austenitizing temperature may be calculated using the empirical formula t ¼ 60 þ 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 cooling from the austenitizing temperature, first a preeutectoid precipitation of ferrite takes place. With a lower cooling rate, the precipitation of ferrite increases along the austenite grain boundaries. For the desired uniform distribution of ferrite and pearlite after normalizing, however, a possibly simultaneous formation of ferrite and pearlite is necessary. Steels having carbon contents between 0.35 and 0.55% C especially tend to develop nonuniform ferrite distributions as shown in Figure 6.61. The structure in this figure indicates overly slow cooling 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, with steels having carbon contents between 0.2 and 0.5%, formation of an undesirable needlelike ¨ ferrite (oriented at austenite grain boundaries), the so-called Widmannstatten’s structure, may result as shown in Figure 6.62. Formation of pearlite follows only after complete ß 2006 by Taylor & Francis Group, LLC. T2 > T1 > A1 T2 Austenite T1 A1 Pearlite Heating Cooling FIGURE 6.60 Schematic presentation of the influence of austenitizing temperature on the grain size of ¨ the structure of a eutectoid steel after normalizing. (From H.J. Eckstein (Ed.), Technologie der Warme¨ behandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) precipitation of ferrite by transformation of the remaining austenite structure at temperature Ar1. It starts first at the boundaries of ferrite and austenite and spreads to the interior of the austenite grains. The greater the number of the pearlitic regions formed, the more mutually hindered the pearlite grains are in their growth, and consequently the finer the grains of the normalized structure. The influence of alloying elements on the austenite to ferrite and pearlite 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 range corresponding to the transformation behavior of the steel in question that results in a pure ferrite–pearlite structure. If, for round bars of different diameters cooled in air, the cooling curves in the core have been experimentally measured and recorded, then by using the appropriate CCT diagram for the steel grade in question, it is possible to predict the structure and hardness after normalizing. To superimpose the recorded cooling curves onto the CCT diagram, the time–temperature 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 desired ferrite–pearlite structure in the core of all investigated bars of different diameters cooled in FIGURE 6.61 Nonuniform distribution of ferrite and pearlite as a consequence of unfavorable temperature control during normalizing of unalloyed DIN C35 steel. Magnification 100Â. (From G. Spur ¨ ¨ and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.62 Formation of needlelike ferrite at grain boundaries after normalizing of the unalloyed ¨ steel DIN C35, because of too fast a cooling rate. Magnification 500Â. (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 cooled in the same way in air will transform to martensite and bainite. In this case, to obtain a desired 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 the temperature at which the formation of pearlite is finished (%6008C (%11008F)). 6.2.3 ISOTHERMAL ANNEALING Hypoeutectoid low-carbon steels for carburizing as well as medium-carbon structural steels for hardening and tempering are often isothermally annealed, for best machinability, because 1000 Hardness HV 900 Ac3 700 15 Austenite 10 3 85 mm 20 3 300 000 600 300 75 2 15 150 Bainite Ms 30 400 80 55 =1 1 35 10 500 85 Ac1 45 40 30 Ferrite 60 70 Pearlite m. dia 600 20 10 Temperature, C 800 Martensite 200 100 722 0 Q1 702 654 576 438 348 101 1 Time, s 278 244 102 1 2 228 213 103 4 8 15 min 174 104 105 60 12 8 16 24 4 h FIGURE 6.63 CCT diagram of the unalloyed steel DIN Ck45 (austenitizing temperature 8508C), with superimposed 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, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. 900 Ac3 800 Ac1 700 100 A P 35 93 m. mm 400 000 600 =1 300 75 30 150 500 10 Temperature, C D ia 600 53 5 B 75 300 Ms 20 200 M 100 796 Hardness HV 0 870 101 1 Time, s 102 1 796 786 782 103 101 753 772 743 104 102 min 454 363 370 105 103 285 106 s 104 FIGURE 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 in ¨ ¨ air. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) a well-differentiated, nontextured ferrite–pearlite structure is the optimum structure for machinability of these steels. If low-carbon steels are soft annealed, they give long shavings when turned and a bad surface appearance (sometimes called ‘‘smearing’’ or ‘‘tearing’’) because of the accumulation of the material on the tool’s cutting edge. On the other hand, nonannealed workpieces, having harder structural constituents like bainite, result in heavy wear of the cutting edge when machined. An isothermally annealed structure should have the following characteristics: 1. 2. 3. 4. 5. High proportion of ferrite Uniformly distributed pearlite grains Fine lamellar pearlite grains Short pearlite lamellae Coarse ferrite grains Figure 6.65 shows the structure of a thin-wall die forging made of low-alloy steel for carburizing (DIN 16MnCr5) after a normalizing anneal (Figure 6.65a) and after an isothermal annealing process (Figure 6.65b). The desired ferrite–pearlite structure originates during an isothermal annealing, the principle of which is explained by Figure 6.66. This figure shows an IT diagram of a low-alloy steel for carburizing (DIN 15CrNi6) with superimposed cooling curves for different cooling rates at continuous cooling. The slowest cooling rate of 3 K/min relates to a furnace cooling, and the fastest cooling rate of 3000 K/min relates to a quenching process. From the diagram in Figure 6.66 it can be clearly seen that bainite formation can be avoided 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 to avoid both the formation of bainite and the formation of a textured structure (see the open arrow in Figure 6.66), and this is the isothermal annealing process, which consists of ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.65 Structure of a forging made of low-carbon steel for carburizing (DIN 16MnCr5) (a) after ¨ normalizing and (b) after isothermal annealing. Magnification 200Â. (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 temperature–time diagram of an isothermal annealing is given in Figure 6.67. The metallurgical mechanism of a good isothermally annealed structure depends on the austenitizing conditions as well as on the temperature and time of the isothermal transformation and on cooling from the austenitizing temperature to the isothermal transformation temperature. The austenitizing temperature and time should be high enough to completely dissolve all carbides, to homogenize the austenite matrix, to stabilize the austenite structure, and achieve a coarse-grained ferrite–pearlite structure after cooling. The undesired textured structure originates by preeutectoid ferrite precipitation along stretched phases acting as germs, for instance manganese sulfides, carbon segregations, or aluminum nitride precipitations. These phases 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, and aluminum as possible, and during austenitizing a complete dissolution of nitride precipitations and carbides should be achieved. Therefore the austenitizing temperature is adequately high, i.e., about 1008C (2128F) above Ac3, and the holding times are usually about 2 h. Field of textured structure 1000 3 K/min 30 Temperature, C 300 P 3000 F A Isothermal annealing B 500 M 320 400 0 10−2 10−1 1 10 Time, min 250 170 HV 102 103 FIGURE 6.66 The principle of isothermal annealing. TTT diagram of the low-alloy steel for carburizing ¨ ¨ DIN 15CrNi6. (From J. Wunning, Harterei-Tech. Mitt. 32:43–49, 1977, pp. 43–49 [in German].) ß 2006 by Taylor & Francis Group, LLC. Temperature, C 1000 800 Ac3 Ac1 600 400 200 0 Time ¨ FIGURE 6.67 Temperature–time 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 minimum cooling rate between the austenitizing temperature (%9508C (%17508F)) and the isothermal transformation temperature (%6508C (12008F)). Thus, about 3008C (5728F) decrease should pass through at a minimum cooling rate of 20–40 K/min. This means that the whole batch of treated workpieces should be cooled from about 9508C (17508F) to about 6508C (12008F) in less than 10 min. During this cooling process an undercooling below the chosen isothermal transformation 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 a good structure after isothermal annealing, all measures that reduce the carbon diffusion rate or 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 DIN 16MnCr5. It can be seen that cooling too slowly from the austenitizing temperature to the transformation temperature results in an undesirable textured structure of ferrite and pearlite, and if during this cooling process an undercooling takes place (i.e., the transformation temperature has been chosen too low) before the pearlite formation, then bainite will be present in the structure, which is not allowed. Big companies usually have internal standards to estimate the allowable degree of texturing of 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 question may be determined by means of the appropriate IT diagram. Figure 6.70 shows such a diagram for 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 ferrite–pearlite; correct annealing (center). Textured ferrite–pearlite structure; too slow cooling from the austenitizing to the transformation temperature (right). Ferrite – pearlite þ bainite; ¨ ¨ undercooling before pearlite transformation. (From J. Wunning, Harterei-Tech. Mitt. 32:43–49, 1977, pp. 43–49 [in German].) ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.69 Internal standard of the German company Edelstahlwerke Buderus A.G.-Wetzlar for estimation of the allowable degree of texturing of the structure after isothermal annealing. Magnifica¨ ¨ tion 100Â. (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 the shortest time to start the transformation is achieved. Below this temperature, longer times are again necessary to start the transformation. In the range of the pearlite nose temperature, fine lamellar 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 and 10768F), while for alloyed steels they are between 640 and 6808C (1184 and 12568F). The optimum isothermal annealing temperature is 10–208C (50–688F) above the pearlite nose temperature. The necessary transformation time depends on the alloying elements in the steel. In the practice of isothermal annealing the holding time at the transformation temperature includes an adequate reserve because of compositional tolerances in different steel heats. Usually for low-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 isothermally annealed, the biggest problem is to realize sufficiently fast cooling from the austenitizing ß 2006 by Taylor & Francis Group, LLC. 900 Ac3 Start of ferrite transformation Temperature, C 880 A c1 700 Start of pearlite transformation Austenite 600 84 91 81 33 500 400 Ms 93 Pearlite 95 End of transformation Start of transformation 35 300 Bainite 31 Martensite 200 Hardness HRC 100 0 1 Hardness HRB 46 10 Time, s 102 1 2 103 4 8 min 15 104 105 106 60 1 2 4 8 24 h 1 2 35 days 10 FIGURE 6.70 Isothermal transformation (IT) diagram of the steel DIN 17CrNiMo6. Austenitizing ¨ temperature 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 austenitizing temperature and the temperature of the cooling medium, and the heat transfer coefficient between 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 close above the Ac1 temperature, with subsequent slow cooling. The microstructure of steel before soft annealing is either ferrite–pearlite (hypoeutectoid steels), pearlite (eutectoid steels), or cementite–pearlite (hypereutectoid steels). Sometimes a previously hardened structure exists before soft annealing. The aim of soft annealing is to produce a soft structure by changing all hard constituents like pearlite, bainite, and martensite (especially in steels with carbon contents above 0.5% and in tool steels) into a structure of spheroidized carbides in a ferritic matrix. Figure 6.71 shows the structure with spheroidized carbides (a) after soft annealing of a medium-carbon low-alloy steel and (b) after soft annealing of a high-speed steel. Such a soft structure 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 the strength and hardness of the material should be as low as possible, for good machinability medium strength or hardness values are required. Therefore, for instance, when ball bearing steels 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 the sequence of operations is hot working, soft annealing, cold forming, hardening, and tempering. The required degree of spheroidization (i.e., 80–90% of globular cementite or carbides) is sometimes specified. To evaluate the structure after soft annealing, there are sometimes internal standards, for a particular steel grade, showing the percentage of achieved globular ß 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 at 720–7408C and (b) a high-speed steel annealed at 8208C. Magnification 500Â. (From G. Spur and ¨ ¨ T. 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 of spheroidization is expressed in this case as percentage of remaining lamellar pearlite. The physical mechanism of soft annealing is based on the coagulation of cementite particles within the ferrite matrix, for which the diffusion of carbon is decisive. Globular cementite within the ferritic matrix is the structure having the lowest energy content of all structures in the iron–carbon system. The carbon diffusion depends on temperature, time, and the kind and amount of alloying elements in the steel. The solubility of carbon in ferrite, which is 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 that it 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 dependent also 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 can be formally divided into two stages. At first the cementite lamellae assume a knucklebone shape, as shown in Figure 6.73. As annealing continues, the lamellae form globules at their ends and, 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 carbide particles, which disappear. In both stages, the rate of this process is controlled by diffusion. The thicker the cementite lamellae, the more energy necessary for this process. A fine lamellar pearlite structure may more easily be transformed to a globular form. In establishing the process parameters for a soft (spheroidizing) annealing, a distinction should be drawn among hypoeutectoid carbon steels, hypereutectoid carbon steels, and alloyed steels. In any case the value of the relevant Ac1 temperature must be known. It can be taken from the relevant IT or CCT diagram or calculated according to the formula Ac1 ¼ 739 À 22(% C) þ 2(% Si) À 7(% Mn) þ 14(% Cr) þ 13(% Mo) þ 13(% Ni) þ 20(% V), [ C] (6:36) The temperature range for soft annealing of unalloyed carbon steels may be taken from the iron–carbon diagram as shown in Figure 6.74. The holding time at the selected temperature is approximately 1 min/mm of the workpiece cross section. For alloyed steels, the soft annealing temperature may be calculated according to the empirical formula ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.72 Internal standard of the German company Edelstahlwerke Buderus A.G.-Wetzlar for evaluation of the degree of spheroidization after soft annealing of grade DIN 100Cr6 steel. Magnification 500Â. Amount of lamellar pearlite remaining 1, 0%; 2, 8%; 3, 20%; 4, 35%; 5, 60%; 6, 80%. (From ¨ ¨ G. 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 spheroids during soft annealing. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. 1050 E Temperature, C 1000 950 900 850 G 800 O 750 K P 700 S 650 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 Carbon content, wt% FIGURE 6.74 Temperature range for soft annealing of unalloyed steels having carbon contents of ¨ ¨ 0.6–1.35% C. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) T ¼ 705 þ 20(% Si À % Mn þ % Cr À % Mo À % Ni þ % W) þ 100(% V) [ C] (6:37) (a) (b) (c) Temperature, C Temperature, C Temperature, C This 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 amounts of alloying elements, only these indicated maximum values are to be taken into account. Figure 6.75 shows possible temperature–time regimes for soft annealing. The swinging regime (Figure 6.75c) is used to accelerate the transformation of cementite lamellae to globular form. Increasing the temperature above Ac1 facilitates the dissolution of cementite lamellae. At subsequent 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 been established that gives the amount of globular cementite compared to the total amount of 800 700 Ac1 600 500 400 800 700 Ac1 600 500 400 800 700 Ac1 600 500 400 FIGURE 6.75 Temperature–time regimes at soft annealing. (a) Annealing at 208C below Ac1, for unalloyed steels and for alloyed steels with bainitic or martensitic starting structure; (b) annealing at 108C above Ac1 (start) and decreasing temperature to 308C below Ac1 for alloyed steels; (c) swinging ¨ annealing +58C around Ac1 for hypereutectoid steels. (From G. Spur and T. Stoferle (Eds.), Handbuch ¨ der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. Annealing temperature, K 1000 Degree of spheroidization “e” 0.40 960 0.50 0.60 0.80 0.95 920 0.20 880 840 800 0.1 0.2 0.4 0.6 0.8 1 2 4 6 8 10 Annealing time, h FIGURE 6.76 Time–temperature diagram for soft annealing of the unalloyed steel DIN C35 (previously ¨ deformed 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 ¼ 1 means that 100% of the globular cementite (i.e., no lamellar cementite) has remained. Because the degree of spheroidization depends on the time and temperature of the soft annealing process, diagrams may be established that correlate 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 ¼ 0.8), has considerable influence on ultimate tensile strength, yield strength, and elongation, as shown in Figure 6.77 for the unalloyed eutectoid steel DIN C75. The hardness after soft annealing depends on the time and temperature of 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 as shown in Figure 6.79, from which it can be seen that decreasing tensile strength and increasing the 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 20–25 K/h (furnace cooling) 850 28 25 750 650 Re 22 19 550 16 450 A (L0 = 80 mm) Elongation, % Tensile strength (Rm) Yield strength (Re), MPa Rm A 350 13 10 0 20 40 60 80 100 Degree of spheroidization, % 250 FIGURE 6.77 Change of ultimate tensile strength, yield strength, and elongation with increasing spheroidization of an unalloyed eutectoid steel, DIN C75. (From H.J. Eckstein (Ed.), Technologie der ¨ ¨ Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) ß 2006 by Taylor & Francis Group, LLC. Hardness, HRB 130 110 90 600 C 625 C 650 C 700 C 675 C 70 0 50 100 150 200 Time, h FIGURE 6.78 Hardness of an unalloyed steel with 0.89% C after soft annealing, depending on the ¨ spheroidization time and temperature. (From H.J. Eckstein (Ed.), Technologie der Warmebehandlung ¨ von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) For medium-alloy steels up to 6308C (11668F), with a cooling rate of 15–20 K/h (furnace cooling) For high-alloy steels up to 6008C (11128F), with a cooling rate of 10–15 K/h (furnace cooling) Further cooling below the temperatures indicated is usually performed in air 6.2.5 RECRYSTALLIZATION ANNEALING Turning speed (v60), m/min Recrystallization annealing is an annealing process at temperatures above the recrystallization temperature of the cold-worked material, without phase transformation, that aims at regeneration of properties and changes in the structure that exists after a cold-forming process 200 150 100 50 a b c 500 600 700 800 900 Tensile strength Rm, N/mm2 1000 FIGURE 6.79 Influence of the ultimate tensile strength and degree of spheroidization on machinability of steels for carburizing and structural steels for hardening and tempering, expressed as 1 h turning speed (v60) in m/min. (a) Spheroidization degree less than 30%; (b) spheroidization degree between 40 ¨ and 60%; (c) spheroidization degree greater than 70%. (From G. Spur and T. Stoferle (Eds.), Handbuch ¨ der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.80 Low-carbon steel with 0.05% C (a) after cold working with 20% reduction (hardness 135 HV) and (b) after subsequent recrystallization annealing at 7508C (hardness 75 HV). Magnification 200Â. (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 a cold-forming process and subsequent recrystallization annealing must possess good coldforming ability. These materials include soft unalloyed steels, microalloyed steels for deep drawing, microalloyed high-strength steels, unalloyed and alloyed carbon steels, stainless steels, and soft magnetic steels. The prerequisite to recrystallization on annealing is that the degree of deformation during cold working has been large enough to produce the required number of defects in the crystals to 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 the energy applied for deformation is transformed into heat, but a relatively small part (less than 5%) of it remains accumulated in the material due to the formation of crystal lattice defects. It is 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 a crystal defect associated with internal stresses, the increase in the dislocation density causes the accumulation of internal stresses (i.e., of internal energy) and thereby increases the free enthalpy. Such a thermodynamically unstable material condition tends, at increased temperatures, to decrease the free enthalpy by rearranging and demolishing lattice defects. The greater the plastic deformation in a cold-forming process, the greater the strengthening of the material, which is characterized by an increase in tensile strength and yield strength and a decrease in elongation as shown in Figure 6.81. The material becomes harder and more 800 700 600 500 400 300 0 10 20 30 40 50 60 60 600 50 Elongation, % 700 Yield strength, N/mm2 Tensile strength, N/mm2 900 500 400 300 30 20 10 200 100 40 0 10 20 30 40 0 0 10 20 30 40 50 60 Degree of deformation at cold working, % FIGURE 6.81 Strengthening of a low-carbon steel by the cold-rolling process. (From G. Spur and ¨ferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ¨ T. Sto ß 2006 by Taylor & Francis Group, LLC. brittle, so that in some cases a further step in the forming process cannot be applied without a recrystallization annealing. Also the anisotropy of the material, i.e., the dependence of mechanical properties on the direction of the cold-forming process, can be annulated by recrystallization annealing, by bringing the oriented grains that are deformed in one direction back to the original globular form. Thermic activation, i.e., increasing the temperature at recrystallization annealing, can be used to reestablish the original structure (before cold working) with the original density of dislocations, which results in decreased hardness and strength and increased ductility and formability. The recrystallization annealing process includes the following phenomena: grain recovery, polygonization, recrystallization, and grain growth. 6.2.5.1 Grain Recovery Grain recovery is a process of tempering a cold-worked metallic structure at low temperatures (150–3508C (300–6628F)) without causing any discernible changes in the microstructure. It results only in decreasing the internal stresses without substantially decreasing the strength of the material. However, during this process characteristic changes occur in the electrical resistance and its temperature coefficient of the cold-worked material. The activation energy needed for grain recovery depends on the degree of cold working. The higher the degree (i.e., the greater the deformation), the less the activation energy required. The temperature of grain recovery correlates with the recrystallization temperature of the same material according to the formula TGR ¼ TR À 300 [ C] (6:38) 6.2.5.2 Polygonization Polygonization of a cold-worked structure is the creation of a new polygonal arrangement of edge dislocations in the metallic crystal lattice that takes place at temperatures close above the grain recovery temperature. As shown in Figure 6.82, in such a case the applied thermal energy is sufficient to rearrange the edge dislocations. In this case the originally bent sliding planes take a polygonal shape, forming segments within a grain called subgrains. The angles between subgrains are very small (about 18). As a consequence of a substantial energy discharge by discharge of internal stresses, material strength is decreased. Polygonization takes place primarily in heavily cold-worked structures, especially in ferritic matrices, below the recrystallization temperature. (a) a (b) b b FIGURE 6.82 Schematic presentation of polygonization. Arrangement of edge dislocations and sliding planes (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, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. Hardness, HV10 250 a b 200 150 100 250 350 450 550 650 750 Temperature, C FIGURE 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.) 6.2.5.3 Recrystallization and Grain Growth The process of recrystallization begins when the recrystallization temperature is overstepped. The recrystallization temperature of a material is the temperature at which the formation of new grains begins within a cold-worked microstructure, as shown in Figure 6.83. From this figure one 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 strength decrease substantially while ductility increases. In practice, the recrystallization temperature TR is often considered the temperature of a cold-worked material at which recrystallization is completed after 1 h of annealing. There is a correlation between the recrystallization temperature (TR) and the melting temperature (TM) of the material, which reads TR ¼ 0:4TM (6:39) Figure 6.84 shows that this correlation holds for practically all pure metals if both temperatures TR and TM are taken in deg. K. The recrystallization temperature can be Recrystallization temperature TR, C 1500 W 1250 Re Ma 1000 750 Cr Pd 500 Be Ni Fe 250 Mg Al Pb 0 Ta Cb Ti Pt Au Ag Cu Zn Cd –250 0 500 1000 1500 2000 2500 3000 3500 Melting temperature TM, C FIGURE 6.84 Correlation between the recrystallization temperature and the melting temperature for ¨ ¨ pure metals. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. influenced by the degree of deformation during cold working, the heating rate, and the starting 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), starting slowly, reaching a maximum rate, and finishing slowly. The nuclei from which new grains grow are situated preferably at the grain boundaries of compressed cold-worked grains. New grains grow from these nuclei until they meet up with other grains. Recrystallization brings about the formation and movement of large-angle grain boundaries. Figure 6.85 is a schematic presentation of new grain formation and growth during the recrystallization process as a function of annealing time. As time passes, the new grains, starting from nuclei, grow unhindered within the cold-worked grains. Simultaneously, new nuclei are formed. At the movement of large-angle grain boundaries, new grains consume the previously 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, as well as the condition of the lattice defects in it, differ substantially from those of the previous structure. The recrystallization process itself can be hindered by precipitations, dispersions, and a second phase. The most important technological parameters of recrystallization annealing that influence the rate of recrystallization and the material properties after recrystallization are: 1. Material-dependent parameters—the chemical composition and the starting structure (including the degree of deformation) 2. Process-dependent parameters—annealing temperature, annealing time, and heating and cooling rates The course of a recrystallization process can be presented in an isothermal time–temperature–recrystallization 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 annealing time. The lower the degree of deformation at cold working, the higher the required t1 t2 > t1 t3 > t2 t4 > t3 t5 > t4 t6 > t5 FIGURE 6.85 Schematic presentation of new grain formation and growth during the recrystallization ¨ process as a function of annealing time t. (From G. Spur and T. Stoferle (Eds.), Handbuch der ¨ Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. Temperature End of recrystallization Beginning of recrystallization Holding time (logarithmic) FIGURE 6.86 An isothermal time–temperature–recrystallization diagram. (From H.J. Eckstein (Ed.), ¨ ¨ Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1987.) Recrystallization temperature, C recrystallization temperature, as shown in Figure 6.87. The higher the heating rate, the higher the recrystallization temperature. It can be concluded from Figure 6.86 that with substantially longer annealing times, a full recrystallization can be achieved at relatively low temperatures. The degree of deformation at cold working has a very important influence on the size of newly formed grains during recrystallization. If the cold working is carried out with a very low degree of deformation but without sufficient strengthening of the material to enable the process of recrystallization, a decrease in stresses can still be achieved by movement of the deformed grain boundaries. In this case grains with low dislocation density grow (because there are only a few nuclei) and a coarse-grained structure develops as shown in Figure 6.88. Consequently, there is a critical degree of deformation at cold working that at subsequent recrystallization annealing leads to sudden grain growth, as shown in Figure 6.89 for a low-carbon steel. With an increase in the carbon content of the steel, this critical degree of deformation shifts from about 8 to 20% of deformation at cold working. 1200 1100 α 1000 900 800 700 γ 600 500 α 400 0 10 20 30 40 50 60 70 80 90 Degree of deformation at cold working, % FIGURE 6.87 Recrystallization temperature of a- and g-iron as a function of the degree of deformation ¨ at cold working. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, ¨ Warmebehandeln, Carl Hanser, Munich, 1987.) ß 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 500Â. (From G. Spur and ¨ ¨ T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) 6.3 HARDENING BY FORMATION OF MARTENSITE 6.3.1 AUSTENITIZING Austenitizing 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 the metallurgical and technological aspects of austenitizing. 6.3.1.1 Metallurgical Aspects of Austenitizing Grain size The way austenite is formed when a certain steel is heated depends very much on the steel’s starting microstructure. Let us take as an example an unalloyed eutectoid steel with 0.8% C and follow the process of its austenitization using the schemes shown in Figure 6.90. At room temperature 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 ferrite lattice. The higher the temperature, the greater this tendency is. Upon heating, on reaching the Ac1 temperature (7238C (13338F)), the transformation of ferrite into austenite (g-Fe) 0 10 20 30 40 50 60 70 80 90 100 Degree of deformation at cold working, % FIGURE 6.89 Grain growth in the range of the critical degree of deformation (at 10%) for a steel with ¨ 0.06% C. Recrystallization temperature, 7008C. (From G. Spur and T. Stoferle (Eds.), Handbuch der ¨ Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. α-Fe α-Fe α-Fe Fe3C γ-Fe Austenite (a) (b) (c) (d) (e) Austenite (f) FIGURE 6.90 Transformation of a pearlitic structure to austenite when heating an unalloyed eutectoid steel of 0.8% C. starts immediately adjacent to the cementite plates (see Figure 6.90b). After that the cementite plates 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 formation of 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 before all the cementite has been dissolved. This means that after all the ferrite has transformed into austenite, 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 of FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. a2 a1 a1 2 2 a2 a1 1 2 a2 3 3 Austenite a3 (a) Carbide 1 a3 (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 of the cementite persists in the form of spheroidized particles (the small dark spots in the white areas). They dissolve only with longer holding times at temperature. Once these cementite particles completely dissolve, the structure consists of only one phase—austenite (see Figure 6.90f). 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 carbon concentration is high, while in other spots far from cementite plates it is low. Equalizing of the carbon concentration proceeds gradually by diffusion, resulting in a homogeneous austenite structure at the end of this process. The holding time at austenitizing temperature necessary for this process is called the homogenization time. During pearlite– austenite transformation, several austenite grains are formed from one pearlite grain, i.e., the newly formed austenite is fine-grained. Nucleation sites for austenite formation depend on the starting microstructure as shown 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 pearlitic structures it starts primarily at the intersection of pearlite colonies but also at cementite lamellae. 6.3.1.1.1 Kinetics of Transformation during Austenitizing Figure 6.93 shows the volume percent of austenite formed from pearlite in eutectoid steel as a function of time at a constant austenitizing temperature. From the beginning of austenitization Volume of austenite, % 100 75 50 25 0 0 5 10 15 Time, s 20 25 30 FIGURE 6.93 Volume percent austenite formed from pearlite in eutectoid steel as a function of time at a constant austenitizing temperature. (From G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International, Materials Park, OH, 1990.) ß 2006 by Taylor & Francis Group, LLC. a certain incubation time is necessary to form the first nuclei, and then the process proceeds at a more rapid rate as more nuclei develop and grow. At higher temperatures the diffusion rate increases and austenite forms more rapidly, as shown in Figure 6.94. The duration of austenitizing process depends on the austenitizing temperature and the steel composition. The influence of time at austenitization can best be explained by the diagrammatic 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 austenitizing temperature of 7308C (13468F) is maintained (after a rapid heating to this temperature), the transformation 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 is in 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 been transformed and the structure consists of ferrite, austenite, and cementite. About 1 min later 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 to dissolve them completely. Figure 6.95e and Figure 6.95f apply to a hypereutectoid steel containing 1.2% C. If this steel is austenitized at 8108C (14908F), the pearlite starts to transform in about 2 s, and in about 5 s the structure consists only of austenite and cementite. It is not possible for the cementite to be completely dissolved at this temperature. To achieve complete solution of the cementite, 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 degree of carbide dissolution and acceptable grain size, taking into account that the grain growth increases with higher austenitizing temperatures and longer holding times. Since the amount of carbide is different for different types of steel, the holding time (from the metallurgical point of view) depends on the grade of steel. However, carbide dissolution and the holding time are dependent not only on the austenitizing temperature but also the rate of heating to 100 Austenite, % 80 730 C 751 C 60 40 20 0 1 10 100 1000 Time, s FIGURE 6.94 Effect of austenitizing temperature on the rate of austenite formation from pearlite in a eutectoid steel. (From G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International, Materials Park, OH, 1990.) ß 2006 by Taylor & Francis Group, LLC. Temperature C 900 A3 Acm 800 C 900 800 A A+C A+ P A1 700 700 600 600 P 500 500 0 0.2 0.4 0.6 0.8 1.0 1.2 % C 10−1 1 (a) (b) Temperature C 900 A3 Acm C 900 800 800 10 102 103 104 105 s A1 700 600 500 F+ P +A 500 10−1 1 (d) F+A + C 700 600 A A+C 0 0.2 0.4 0.5 0.8 1.0 1.2 % C (c) Temperature C 900 A3 Acm 10 102 103 104 105 s C 900 800 800 F+ P A A+C A + P+ A1 700 600 C 700 600 P+C 500 0 0.2 0.4 0.5 0.8 1.0 1.2 % C (e) 500 10–1 1 (f) 10 102 103 104 105 s FIGURE 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 and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) this temperature. Varying the rate of heating to this temperature will have an effect on the rate of transformation and dissolution of the constituents. The influence of the role of heating (and correspondingly of the holding time) on carbide dissolution, grain growth, and hardness after hardening for various grades of steel has been studied in detail and published in Refs. [18,19]. These time–temperature–austenitizing diagrams (Zeit-Temperatur-Austenitisierung Schaubilder in German) have been produced either as isothermal 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 continuous heating diagrams (the steel specimens were heated continuously at different heating rates). ß 2006 by Taylor & Francis Group, LLC. Consequently, isothermal diagrams may be read only along the isotherms, and the continuous heating diagrams may be read only along the heating rate lines. Figure 6.96 shows an isothermal type of time–temperature–austenitizing diagram of grade DIN 50CrV4 steel. From this type of diagram one can read off, for instance, that if the steel is held 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 every austenitizing process is carried out at a specified heating rate. Figure 6.97 shows a time– temperature–austenitizing 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 to 24008C/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 pearlite and ferrite would have been transformed to inhomogeneous austenite. At a heating rate 1300 1200 Homogenuous austenite Temperature, C 1100 1000 ACC 900 Austenite + carbide AC3 830 800 AC1 Ferrite + pearlite + austenite AC2 Ferrite + pearlite Heating rate to hardening temp. 130 C/s 700 0.01 0.1 1 10 102 103 Time, s FIGURE 6.96 Isothermal time–temperature–austenitizing 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 zur ¨ ¨ Warmebehandlung der Stahle, Vol. 4, Zeit-Temperatur-Austenitisierung-Schaubilder, Part 2, Verlag ¨ Stahleisen, Dusseldorf, 1976 [in German].) ß 2006 by Taylor & Francis Group, LLC. Heating rate, C/s 2400 1000 300 100 30 10 3 1 0.22 0.05 1300 1200 1100 Temperature, C Homogeneous austenite 1000 900 Ac3 Inhomogeneous austenite 800 Ferrite + pearlite Ac1 Austenite Ac2 Ferrite + pearlite 700 10−1 1 10 102 103 104 105 Time, s FIGURE 6.97 Time–temperature–austenitizing diagram for continuous heating of the steel grade DIN Ck45 (0.49% C, 0.26% Si, 0.74% Mn). (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].) of 108C/s (508F/s) the pearlite and ferrite would have been transformed to inhomogeneous austenite 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 and Ac3 transformation temperatures with increasing heating rates. This is especially important when short-time heating processes like induction hardening or laser beam hardening, with heating rates ranging to about 10008C/s (18328F/s), are applied for surface hardening. In such a 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 steel in question, for example, the conventional hardening temperature would be in the range of 830–8508C (1526–15628F), but for induction or laser beam hardening processes the hardening temperatures required are between 950 and 10008C (1742 and 18328F). When heating at a rate of 10008C/s (18328F/s) to the austenitizing temperature of 10008C (18328F), only 1 s is necessary, and the above-mentioned short heating time processes operate in approximately this time range. As Figure 6.97 shows, much higher temperatures are necessary to achieve the ß 2006 by Taylor & Francis Group, LLC. homogeneous austenite structure. In such a case one is, of course, concerned with the grain growth. Figure 6.98 shows the grain growth (according to American Society for Testing and Materials [ASTM]) when grade DIN Ck45 steel is continuously heated at different heating rates 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 various temperatures. It shows, for example, that maximum hardness would be achieved upon austenitizing the steel at 8508C (15628F) for about 900 s (or heating at a heating rate of 18C/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 also dependent on the initial structure of the steel. This is illustrated in Figure 6.100. Figure 6.100a shows that a structure of spheroidized cementite (after soft annealing) of the hypoeutectoid DIN Cf53 carbon steel will attain the maximum hardness of 770 HV when heated at a rate of 18C/s (33.88F/s) to 8758C (16098F) (holding time 855 s or 14 min). The hardened and Heating rate C/s 2400 1000 300 100 30 10 3 1 0.22 0.05 1300 4 to 3 1200 1100 8 1000 9 10 11 to 10 Ac3 900 800 Grain size (ASTM): Temperature, C 4 6 Ac1 700 10−1 1 10 102 103 104 105 Time, s FIGURE 6.98 Time–temperature–austenitizing diagram for continuous heating showing the grain ¨ growth of steel grade DIN Ck45. (From J. Orlich, A. Rose, and P. Wiest (Eds.), Atlas zur Warmebe¨ ¨ handlung der Stahle, Vol. 3, Zeit-Temperatur-Austenitisierung-Schaubilder, Verlag Stahleisen, Dusseldorf, 1973 [in German].) ß 2006 by Taylor & Francis Group, LLC. Heating rate C/s 2400 1000 300 100 30 10 3 1 0.22 0.05 1300 1200 780 800 1000 840 Ac3 900 840 840 820 800 780 800 Ac1 Hardness after quenching (HV): Temperature, C 1100 700 0.1 1 10 102 103 104 105 Time, s FIGURE 6.99 Time–temperature–austenitizing diagram for continuous heating showing the achievable hardness after hardening 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].) 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 the rate of 10008C/s (18328F/s) (holding time less than 1 s). For this reason, when short-time heating processes are used, the best results are achieved with hardened and tempered initial structures. For eutectoid and hypereutectoid steel grades, which after quenching develop substantial amounts of retained austenite, the attainment of maximum hardness after quenching is more complicated. 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 maximum hardness of 900 HV after quenching is attained on heating to a very narrow temperature range, and furthermore this temperature range is displaced toward higher temperatures as the heating rate is increased. If this steel is quenched from temperatures that exceed the optimum range, the resulting hardness is reduced owing the presence of an increasing amount of retained austenite. ß 2006 by Taylor & Francis Group, LLC. Hardness after quenching, HV1 900 Rate of heating in C/s 800 700 1 10 100 1000 600 700 800 900 Hardness after quenching, HV1 (a) 1000 1100 Temperature, C 1200 1300 1200 1300 900 Rate of heating in C/s 800 1 10 100 1000 700 600 700 (b) 800 900 1000 1100 Temperature, C FIGURE 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, a holding time of 5–15 min after they have reached the hardening temperature is quite enough to make certain that there has been sufficient carbide dissolution. For medium-alloy structural steels this holding time is about 15–25 min. For low-alloy tool steels, it is between 10 and 30 min; and for high-alloy Cr steels, between 10 min and 1 h. 6.3.1.2 Technological Aspects of Austenitizing In heating metallic objects to their austenitizing (hardening) temperature, there are two kinds of heating rates to be distinguished: those that are technically possible and those that are technologically allowed. The technically possible heating rate is the heating rate the heating equipment could realize in actual use. It depends on 1. The installed heating capacity of the equipment 2. The heat transfer medium (gas, liquid, vacuum) 3. The temperature difference between the heat source and the surface of the heated objects (workpieces put in a hot or cold furnace) 4. The mass and shape of the workpiece (the ratio between its volume and superficial area) 5. The number of workpieces in a batch and their loading arrangement ß 2006 by Taylor & Francis Group, LLC. Hardness after quenching, HV1 1000 Rate of heating in C/s 900 800 700 1 10 100 1000 600 500 700 800 900 1000 1100 Temperature, C 1200 1300 FIGURE 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 Heat Treatment, 2nd ed., Butterworths, London, 1984.) The technologically allowed heating rate is the maximum heating rate that can be applied in actual circumstances, taking into account the fact that thermal stresses that develop within the 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 temperature gradients can arise between the surface and the core of the workpiece. This heating rate depends on 1. The mass and shape of the workpiece (the ratio between its volume and superficial area) 2. The chemical composition of the material 3. The initial microstructure When workpieces of heavy sections or of complicated shapes are heated, temperatures between 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 are developed that overstep the strength of the material (which is substantially lower at higher temperatures), cracks will result. If the heating rate is too high through the transformation temperature range (between Ac1 and Ac3), warping may occur because of volume change of the structure lattice. The tendency of a steel to crack during heating depends on its chemical composition. Carbon content has the 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 following empirical formula termed the C equivalent (Cekv): Cekv ¼ C þ Mn Cr Mo Ni V Si À 0:5 Ti W Al þ þ þ þþ þþþ 5 4 3 10 5 5 5 10 10 (6:40) where the element symbols represent wt% content. This formula is valid up to the following maximum values of alloying elements. ß 2006 by Taylor & Francis Group, LLC. C 0.9% Mn 1.1% Cr 1.8% Mo 0.5% Ni 5.0% V Si Ti W Al 0.25% 1.8% 0.5% 2.0% 2.0% The values of the alloying elements actually present are put into the formula in wt%. If the amount of an alloying element exceeds the limit given above, then the indicated maximum value should be put into the formula. The higher the calculated Cekv value, the greater the sensitivity of the steel to cracking. For instance, Cekv 0.4: The steel is not sensitive to cracking (it may be heated quite rapidly). Cekv ¼ 0.4–0.7: The steel is medium sensitive to cracking. Cekv ! 0.7: The steel is very sensitive to cracking (when heating up a preheating operation should be included). The initial microstructure also has some influence on the technologically allowed heating rate. A steel with a homogeneous microstructure of low hardness may be heated more rapidly than a steel of high hardness with inhomogeneous microstructure. The thermal gradients and consequently the thermal stresses developed when heating to austenitizing temperature can usually be diminished by preheating the workpiece to temperature lying close below the transformation temperature Ac1 and holding it there until temperature equalizes throughout the cross section. The theoretical time–temperature diagram of the austenitizing process is shown in Figure 6.102. Practically, however, there is no such strict distinction between the heating and soaking TC Co Surfa ce re Ta 1 1—preheating 2—heating up 3—heating through (thermal soaking) 4—structure homogenizing (metallurgical soaking) Ta —austenitizing temperature ta —austenitizing time th 3 2 ta 4 FIGURE 6.102 Austenitizing process (theoretically). ß 2006 by Taylor & Francis Group, LLC. Temperature F C 1832 1000 I II 1652 900 III 1472 800 1292 700 1112 600 932 500 ∅1 × 4 in. I ∅ 25 × 100 mm ∅2 × 7 in. II ∅ 50 × 175 mm ∅4 × 8 in. III ∅ 100 × 200 mm 752 400 572 300 392 200 212 100 32 0 02 4 6 8 10 12 14 16 18 20 22 24 26 28 min Heating-up time FIGURE 6.103 Time–temperature curves for steel bars of different diameters heated in a salt bath at 10008C. Full line, measured temperature at surface; dashed line, measured temperature at center. (From K.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 the preset temperature considerably earlier than the center, the closer the temperature of the steel approaches the preset temperature, the smaller the temperature difference between surface and core, as shown in Figure 6.103. It can therefore be assumed that when the surface has reached the preset temperature, part of the soaking time (depending on the cross-sectional size) has already been accomplished. Certainly, one has to be aware of the corner effect— corners, sharp edges, and thin sections reach the preset temperature much earlier than the core of the workpiece. The most important parameters of every austenitizing process are: 1. The austenitizing temperature 2. The heat-up and soak time at austenitizing temperature For each grade of steel there is an optimum austenitizing (hardening) temperature range. This temperature range is chosen so as to give maximum hardness after quenching and maintain a fine-grained microstructure. It can be determined experimentally as shown in Figure 6.104 and Figure 6.105. From Figure 6.104 it is clear that the lowest possible hardening temperature for the steel in question is 8508C (15628F). A lower hardening temperature would result in the formation of bainite and even pearlite with inadequate hardness. When the hardening temperature is increased (see Figure 6.105), the grain size and the amount of retained austenite increase. At 920 and 9708C (1688 and 17788F) the retained austenite 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 850– 8808C (1562–16168F). The optimum hardening temperature range for unalloyed steels can be determined from the iron–carbon equilibrium diagram according to the carbon content of the steel. This range is 30–508C (86–1228F) above the Ac3 temperature for hypoeutectoid steels and 30–508C (86–1228F) above Ac1 for hypereutectoid steels, as shown in Figure 6.106. Because the curve S–E in this diagram denotes also the maximum solubility of carbon ß 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 from hardening temperatures between 800 and 8508C. Dimensions of test pieces: 30-mm diameter  100 mm. Magnification 400Â. (a) Hardening temperature 8008C, hardness 55 HRC; (b) hardening temperature 8258C, 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, the more carbon can be dissolved in austenite. For alloyed steels the optimum austenitizing (hardening) temperature range depends on the chemical composition, because different alloying elements shift the A1 temperature to either higher or lower temperatures. For ß 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 from hardening temperatures between 870 and 9708C. Dimensions of test pieces: 30-mm diameter  100 mm. Magnification 400Â. (a) Hardening temperature 8708C, hardness 62.5 HRC, retained austenite 12%; (b) hardening temperature 9208C, hardness 62 HRC, retained austenite 20%.; (c) hardening temperature 9708C, 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 range have to be consulted. It should also be mentioned that increasing the austenitizing temperature causes the following effects. ß 2006 by Taylor & Francis Group, LLC. Iron−carbon equilibrium diagram TC E 1148 G γ + Fe3C0 0 α+ γ A Austenite cm γ 912 A3 727 A1 α' + pearlite Pearlite S Pearlite + Fe3C 0.77 2.11 C, wt% The range of optimun hardening temperatures FIGURE 6.106 Optimum hardening temperature range for unalloyed steels, depending on the carbon content. 1. It increases the hardenability of the steel because of the greater amount of carbide going into solution and the increased grain size. 2. It lowers the martensite start temperature (Ms). Owing to the more complete carbide dissolution, 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 to pearlite or bainite starts. This is expressed as a shift in the start of transformation curves in an IT diagram to later times. 4. It increases the amount of retained austenite after quenching due to stabilization of the austenite, which at higher temperatures is more saturated with carbon from dissolved carbides. Heat-up and soak time at austenitizing temperature is a very important parameter for bulk 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. Until recently there was no reliable, objective method for accurately predicting heat-up and soak times for heat treatment cycles that took into account all workpiece characteristics, variations in furnace design, and load arrangement. Current determinations of heat-up and soak time are based on either a very conservative and general rule (e.g., 1 h/in. of cross section) or some empirical 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 from starting (room) temperature to the preset temperature in its core. The main factors that influence heat-up and soak time are diagrammed in Figure 6.107. On the basis of experiments with 26 specimens (cylinders, round plates, and rings of various dimensions) made of unalloyed and low-alloy structural steels, Jost et al. [20] found from core temperature measurements that the heat-up and soak time depends substantially on the geometry of the heated workpiece and its mass. They found the heat-up and soak time to be 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 heated chamber furnace of 8 kW capacity and 240  240  400 mm working space, to the hardening ß 2006 by Taylor & Francis Group, LLC. Actual conditions Workpiece Furnace Type Working space Heating mode and installed capacity Heat transfermedium Temperature distribution Temperature Number of workpieces Loading arrangement Trays Shape size Surface area Mass Heat conductivity Heat-up and soak time FIGURE 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):298–301, 1976 [in German].) temperature, 8708C (15988F)), they found that the heat-up and soak time (t) can be calculated using the equation t ¼ 0:42(m=A)À3:7 (6:41) The regression coefficients 0.42 and 3.7 are, of course, valid for their experimental conditions only. Comparison with their experimentally obtained results (see the points in Figure 6.108) showed a standard deviation of s2 ¼ 1.4 min2, or s ¼ +1.2 min, indicating that this way of predicting 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 soak times according to the general expression t ¼ a(m=A) þ b (6:42) provided that for a given situation the straight line of regression and relevant values of the regression coefficients a and b are fixed by means of some preliminary experiments. It should be stressed, however, that the described results of this investigation are valid for single workpieces only. In another investigation [21], a method enabling heat treaters to accurately determine the heat-up and soak times for different loads treated in batch-type indirect fired furnaces was 50 40 t, min 30 20 10 0 30 40 50 60 70 80 m/A , Kg/m2 90 100 110 120 FIGURE 6.108 Dependence of the heat-up and soak time on the mass/surface area ratio, (m/A). (From S. Jost, H. Langer, D. Pietsch, and P. Uhlig, Fertigungstech. Betr. 26(5):298–301, 1976 [in German].) ß 2006 by Taylor & Francis Group, LLC. developed. To develop the method, a statistical and experimental investigation of load temperature conditions was performed. A computer-aided mathematical model of heat and mass 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, load arrangements, workpiece shapes, and thermal properties. The treated loads were divided into several groups in terms of workpiece allocation and aerodynamic patterns of the furnace atmosphere, 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 915–1680 mm, width 610–1420 mm, and height 610– 1270 mm. The furnaces were equipped with four burners firing into the trident burner tubes located on the side walls, with a circulating fan located on top of the furnace as shown in Figure 6.110. The thermocouples were located in different parts of the load (measuring always the surface temperature of the workpieces)—on the top and bottom, in the core, at the corners, and on the surfaces facing radiant tubes—to 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 very complicated and are characterized by nonlinear three-dimensional radiation and convection and by nonlinear heat conduction within the workpieces. In this case, the mathematical (a) Monolayer, horizontally oriented, ordered loads Packed Spaced (b) Monolayer, horizontally oriented, random loads (c) Multilayer ordered and random loads Packed (d) Spaced Bulk Vertically oriented loads FIGURE 6.109 Load characterization. (a) Monolayer, horizontally oriented, ordered loads; (b) monolayer, horizontally oriented, random loads; (c) multilayer ordered and random loads. (d) Vertically oriented loads. (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, Proceedings of the International Heat Treatment Conference on Equipment and Processes, April, 18–20, 1994, Schaumburg, IL, pp. 55–61.) ß 2006 by Taylor & Francis Group, LLC. Roof fan Radiant tubes q qwc wr qtr qtc qtr qtc qwr qpc qwc qgc qpr qgc qpr Load qpc Radiation from the radiant tubes Convection from the radiant tubes Radiation from the walls Convection from the walls Convection from the furnace gases Radiation between parts Conduction through the parts FIGURE 6.110 Heat transfer in the used furnace and 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, Proceedings of the International Heat Treatment Conference on Equipment and Processes, April, 18–20, 1994, Schaumburg, IL, pp. 55–61.) model to describe the heat and mass exchange is a system of integral and differential nonlinear equations. 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, number of trays in the basket, workpiece characteristic size Type of load (according to load characterization, see Figure 6.109) Type of steel (carbon, alloyed, high-alloy) Load thermal properties Load and furnace emissivities Temperature conditions (initial furnace and load temperature) Fan characteristic curve parameters Composition of protective atmosphere As an example, maximum and minimum steel part temperatures for a test (heating of shafts) together with the calculated data are shown in Figure 6.111. The experimental data show that the temperature curve of the load thermocouple usually reaches the set furnace temperature well within the soak time requirements. The experimentally determined soak time is seen to be considerably shorter than the soak time defined by the heat treater. It was found that the discrepancy between soak times determined from the test data and calculations does not exceed 8%, which is acceptable for workshop practice. The developed computer model was used for simulation of temperature conditions for different load configurations, and a generalized formula and set of graphs were developed. The generalized equation for the soak time determination is ts ¼ 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 ß 2006 by Taylor & Francis Group, LLC. Temperature, C 900 800 700 600 500 400 300 0 15 30 45 60 75 Time, min 90 105 120 135 Furnace 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, and M.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, 18–20, 1994, Schaumburg, IL, pp. 55–61.) require different graphs. The correction factor k depends on the type of steel. The generalized equation (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 a straightforward way of determining the soak time without the use of charts while allowing for a quick and accurate soak time calculation. 6.3.2 QUENCHING INTENSITY MEASUREMENT AND EVALUATION BASED ON HEAT FLUX DENSITY In 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 ˇˇ ´ concept, which expresses quenching intensity by a single number, the main idea of Liscic was to express the quenching intensity by continuous change of relevant thermodynamic functions during the whole quenching process. Instead of recording only one cooling curve (as in laboratory-designed tests) in the center of a small (usually 1/2 in.) cylindrical specimen, the 300 N=3 Soak time (tsb), min 250 N=2 N = Number of trays N=4 200 N=1 150 N>4 100 50 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 Load 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, Proceedings of the International Heat Treatment Conference on Equipment and Processes, April, 18–20, 1994, Schaumburg, IL, pp. 55–61.) ß 2006 by Taylor & Francis Group, LLC. heat flux density at the surface of a standard-size probe becomes the main feature in measuring, recording, and evaluating the quenching intensity. The first substantial difference between using the small laboratory specimen and using the probe applied in the method described below is that when quenching, for example, in an oil quenching bath, because of its small mass and small heat capacity the former will cool down in 15–30 s, whereas the latter will take 500–600 s to cool to the bath temperature, allowing the entire quenching process of real components to be followed. This workshop-designed method is applicable to 1. All kinds of quenchants (water and brine, oils, polymer solutions, salt baths, fluid beds, circulated gases) 2. A variety of quenching conditions (different bath temperatures, different agitation rates, 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 quenching parameters (specific character of the quenchant, its temperature, and mode and degree of agitation). This method . . . . Enables a real comparison of the quenching intensity among different quenchants, different quenching conditions, and different quenching techniques Provides an unambiguous conclusion as to which of two quenchants will give a greater depth of hardening (in the case of the same workpiece and same steel grade) and enables the exact calculation of cooling curves for an arbitrary point on a round bar cross section of a specified diameter, to predict the resulting microstructure and hardness (an exception is the case of delayed quenching, where the cooling rate is discontinuously changed; for an explanation see Ref. [23]) Furnishes information about thermal stresses and possible superposition of structural transformation stresses that will occur during a quenching process Provides the basis for automatic control of the quenching intensity during the quenching process The method itself, known in the literature as the temperature gradient method, is based on the known physical rule that the heat flux at the surface of a body is directly proportional to the temperature gradient at the surface multiplied by the thermal conductivity of the body material: q¼l dT dx (6:44) where q is the heat flux density (W/m2) (i.e., quantity of heat transferred through a surface unit 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 the surface (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 surface reproducibly measures the real surface temperature of the probe (Tn), which is important to register all the phenomena that are taking place on the surface during quenching. The intermediate thermocouple (Ti) measures the temperature at a point 1.5 mm below the surface. The readings of Tn and Ti enable the heat treater to easily calculate the temperature gradient near the surface of the probe at each moment of cooling. The central thermocouple ß 2006 by Taylor & Francis Group, LLC. Tn 200 Tc Ti 100 N5 1.5 Diam. 50 All dimensions in mm ˇˇ ´ 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 from the core and provides at every moment the temperature difference between the surface and the core, 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 10À5 s; the fastest transient temperature 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 tip before each measurement (self-renewable thermocouple). 4. The body of the probe, made of an austenitic stainless steel, does not change in structure during the heating and quenching process, nor does it evolve or absorb heat because of structural transformation. 5. The size of the probe and its mass ensure sufficient heat capacity and symmetrical radial 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 diameters greater than 50 mm. ß 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 software enables the storage of the temperature–time data for all three thermocouples and the calculation and graphical display of relevant functions. The software package consists of three modules: Module I. TEMP-GRAD (temperature gradient method) Module II. HEAT-TRANSF (calculation of heat transfer coefficient and cooling curves) 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 (Figure 6.114a 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 delayed quenching with a discontinuous change in cooling rate, because in case A the time when maximum 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 extracted amount 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 extracted heat was only 5 MJ/m2 and by 50 s this value was only 20 MJ/m2. However, immediately after that in the period between 50 and 100 s, in case A the extracted amount of heat increased from 50 to only 55 MJ/m2, whereas in case B it increased from 20 to 86 MJ/m2. Both of the Ð calculated integral ( q dt) curves, designated with the open square symbols in Figure 6.114e and Figure 6.115e, have been calculated as the area below the heat flux density vs. time curves, designated similarly in Figure 6.114b and Figure 6.115b. That is, they represent the heat extracted only through the surface region between the point 1.5 mm below the surface and the surface itself. Comparing the time required to decrease the heat flux density from its maximum to a low value of, e.g., 100 kW/m2 (see Figure 6.114b and Figure 6.115b), one can see that in case A 45 s is 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 with delayed burst of the thick polymer film. Discontinuous change in cooling rate is inherent to this quenching regime. In this respect it is interesting to analyze the cooling rate vs. surface temperature diagrams of Figure 6.114f and 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 the center (D), in case B the maximum cooling rate at 1.5 mm below surface (during a certain short period between 350 and 3008C (662 and 5728F) surface temperature) is higher than the maximum cooling rate at the surface itself. This can also be seen in Figure 6.115a, which shows that the slope of the cooling curve Ti between 500 and 3008C (932 and 8428F) is greater than the slope of the cooling curve for the very surface (Tn). This is another experimental proof that in delayed quenching cooling rates below the workpiece surface can be higher than at the surface itself. Another analysis, with respect to thermal stresses during quenching (on which the residual stresses and possible distortion depend), is possible by comparing Figure 6.114d and Figure 6.115d. This comparison shows that quenching in a PAG polymer solution of high ß 2006 by Taylor & Francis Group, LLC. concentration (case B), compared to oil quenching (case A), resulted in 27% lower maximum temperature 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 mm below the surface, (D), contributing to lower distortion than in oil quenching. Whereas with oil quenching the maximum temperature difference between the center and the point 1.5 mm below the surface (D) is higher than the maximum temperature difference between the point 1000 900 Tc Ti 700 600 Measured Temperature T, C 800 Tn 500 400 300 200 100 0 1 1n 2i 10 20 50 100 200 500 1000 Time t, s 3c 1c-n Calculated (b) 1 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0 0 1c-n 2 2i-n 3c-i 5 10 20 tmax 50 100 200 500 1000 Time t, s Calculated Heat flux density q, kW/m2 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0 Heat flux density q, kW/m2 3 Tn = surface temperature, T i = temperature 1.5 mm below the surface, Tc = temperature in the center of the probe. (a) (c) 2 100 200 300 400 500 600 700 800 900 10001100 1200 Temperature Tn, C 2i-n 3c-i ˇˇ ´ 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 temperature vs. time, T ¼ f(t); (b) calculated heat flux density vs. time, q ¼ f(t); (c) calculated heat flux density vs. surface temperature, q ¼ f(Tn). ß 2006 by Taylor & Francis Group, LLC. 500 400 350 300 Calculated Temperature differences T, C 450 250 200 150 100 50 0 (d) 1 1c-n 2 2i-n 5 10 20 50 100 200 500 1000 Time t, s 10 20 50 100 200 500 1000 Time t, s 3c-i 100 90 80 Calculated 60 50 40 t0 q t dt, (MJ/m2) 70 30 20 10 0 1 2 1c-n (e) 2i-n 5 3c-i 50 45 35 30 25 20 Calculated Cooling rate dT, K/s dt 40 15 10 5 0 (f) 1n 0 100 200 300 400 500 600 700 800 900 10001100 1200 Temperature Tn, C 2i 3c FIGUREÐ 6.114 (Continued ) (d) Calculated temperature differences vs. time, DT ¼ f(t). (e) Calculated integral q dt ¼ heat extracted vs. time. (f) Calculated cooling rates vs. surface temperature dT/dt ¼ f(Tn). 1.5 mm below the surface and the surface itself. (&), In the case of delayed quenching (case B), the maximum temperature difference between the point 1.5 mm below the surface and the surface itself (& in Figure 6.115d) is slightly higher than the maximum temperature defference between the center and the point 1.5 mm below the surface (D), which is reached about 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 temperature difference between the center and surface (*) occurs 20 s after immersion, when the surface temperature is 4508C (8428F) (see Figure 6.114a), i.e., above the temperature of the Ms point. In ß 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 has already fallen to 3608C (6808F) (see Figure 6.115a). In this respect, dealing with steels that have a high Ms temperature, water-based polymer solutions always run a higher risk of overlapping thermal stresses with those created due to austenite-to-martensite transformation. 1000 900 Tc Ti Tn 700 Measured Temperature T, C 800 600 500 400 300 200 100 0 1 1n 1800 1600 1400 1200 1000 800 600 400 200 0 1 (b) Heat flux density q, kW/m2 Calculated 3000 2800 2600 2400 2200 2000 1c-n 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 0 0 (c) 1c-n 2 2i-n 5 3c-i 10 20 30 100 200 500 1000 Time t, s Tqmax Calculated Heat flux density q, kW/m2 (a) 2 5 10 20 50 100 200 300 1000 Time t, s 2i 3c Tn = 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 1200 Temperature Tn , C 2i-n 3c-i Tqmax ˇˇ ´ 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/s agitation rate. (a) Measured and recorded temperature vs. time data, T ¼ f(t). (b) Calculated heat flux density vs. time, q ¼ f(t). (c) Calculated heat flux density vs. surface temperature, q ¼ f(Tn). ß 2006 by Taylor & Francis Group, LLC. Temperature differences ΔT, C 500 450 400 Calculated 350 300 250 200 150 100 50 0 (d) 1 2 5 10 20 50 100 200 300 1000 Time t, s 1c-n 2i-n 3c-i 200 180 Calculated 140 120 100 t0 tx ∫ q dt, MJ/m2 460 80 60 40 20 0 (e) 1 2 5 1c-n 2i-n 3c-i 10 20 50 100 200 300 1000 Time t, s 25 15 10 5 0 (f) Calculated dT Cooling rate dt , K/S 20 0 1n 100 200 300 400 500 600 700 800 900 1000 1100 1200 2i 3c Temperature Tn, C FIGURE 6.115 (Continued ) (d) Calculated temperature differences vs. time, DT ¼ f(t). (e) Calculated Ð integral q dt ¼ heat extracted vs. time. (f) Calculated cooling rates vs. surface temperature, dT/dt ¼ f(Tn). The probability of crack formation can be seen at a glance by comparing the surface temperature of the probe at the moment the maximum heat flux density occurs (Tqmax). As seen 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 the value of Tqmax, the higher is the risk of crack formation, especially with steel grades having high Ms temperature. ß 2006 by Taylor & Francis Group, LLC. When direct immersion quenching is involved with continuous cooling (not delayed quenching with discontinuous cooling), the depth of hardening, when comparing two quenchÐ ing processes, is determined as follows: The larger the values of qmax and q dt and the shorter the time tqmax,, the greater will be the depth of hardening. Module II of the software package, HEAT-TRANSF, makes it possible (based on the input 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 present 1. The heat transfer coefficient between the probe’s surface and the surrounding fluid vs. time, a ¼ f(t) (Figure 6.116a) 2. The heat transfer coefficient between the probe’s surface and the surrounding fluid vs. surface temperature, a ¼ f(Tn) (Figure 6.116b) Using the calculated values of a, the software program enables the calculation of cooling curves at any arbitrary point of the round bar cross section of different diameters, as shown in Figure 6.117a and Figure 6.117b. The Module III of the software package, CCT-DIAGR, is used to predict the resulting microstructure 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 CCT diagrams of their own choice. The program enables the user to superimpose every calculated cooling curve on the CCT diagram of the steel in question. From the superimposed cooling curves shown 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/m2K 1800 1600 1400 1200 1000 800 600 400 200 Heat transfer coefficient a , W/m2K (a) 0 0.01 0.1 1 10 100 Time, s 1000 1800 1600 1400 1200 1000 800 600 400 200 (b) 0 0 100 200 300 400 500 600 700 800 900 Temperature, C FIGURE 6.116 Heat transfer coefficient (a) vs. time and (b) vs. surface temperature when quenching the ˇˇ ´ Liscic-NANMAC probe (50-mm diameter  200 mm) in a 208C mineral oil bath without agitation. ß 2006 by Taylor & Francis Group, LLC. 900 800 Temperature, 8C 700 600 500 400 300 200 100 (a) 0 0.01 0.1 1 10 Time, s 100 1000 0.1 1 10 Time, s 100 1000 900 800 Temperature, C 700 600 500 400 300 200 100 (b) 0 0.01 FIGURE 6.117 Comparison of measured (- - -) and calculated (—) cooling curves for the center of a 50-mm diameter bar quenched in (a) mineral oil at 208C, without agitation and (b) 25% PAG polymer solution, 408C bath temperature, and 0.8 m/s agitation rate. C Si Mn 0.38 Chemical composition 0.23 0.64 P S Cr Cu 0.019 0.013 0.99 Mo 0.17 0.16 Ni V 0.08 <0.01 1000 AISI 4140 900 Austenitizing temp. 850 C 3/4 R C Temperature, C 800 30 700 S F 2 600 12 7 P 5 A 40 40 60 70 Ac3 Ac1 60 10 200 500 B 400 2 Ms 3 75 300 85 75 M 200 100 58 0 1 10 53 52 102 34 28 27 230 103 220 104 105 106 Time, s FIGURE 6.118 CCT diagram of AISI 4140 steel with superimposed calculated cooling curves for surface (S), three-quarter radius (3/4R) and center (C) of a round bar of 50-mm diameter. ß 2006 by Taylor & Francis Group, LLC. If for a round bar cross section of the chosen diameter the cooling curves are calculated at three 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 values after quenching at those points and to obtain the hardness distribution curve displayed graphically on the computer screen. In the case of delayed quenching with discontinuous change of cooling rate, the prediction of structural transformations and hardness values after quenching from an ordinary CCT diagram is not correct because the incubation time consumed (at any point of the cross section) before the cooling rate abruptly changes has not been taken into account. For a detailed explanation see Shimizu and Tamura [11]. 6.3.3 RETAINED AUSTENITE AND CRYOGENIC TREATMENT The martensite start (Ms) and martensite finish (Mf) temperatures for unalloyed steels depend on their carbon content, as shown in Figure 6.119. As can be seen from this diagram, when steels of more than 0.65% C are quenched the austenite-to-martensite transformation does not 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 room temperature, a portion of austenite will remain untransformed; this is referred to as retained austenite. The greater the amount of carbon in the steel, the greater the amount of retained austenite after quenching, as shown in Figure 6.120c. Retained austenite, which is a softer constituent of the structure, decreases the steel’s hardness after quenching. If present in amounts of more than 10%, a substantial decrease in the hardness of the quenched steel may result (see curve a of Figure 6.120a). When quenching hypereutectoid steels from the usual hardening temperature (Figure 6.120b) i.e., from the g þ Fe3C region, the same hardness will result independently of carbon content (curve b in Figure 6.120a), because the hardness of martensite depends only on the carbon dissolved in austenite (g), which further depends (according to the solubility limit, line S–E) on the hardening temperature. The structure of hardened hypereutectoid steels therefore consists of martensite þ Fe3C þ retained austenite. When quenching hypereutectoid steels from the region of pure austenite (g), i.e., from above the Acm temperature (which is not usual), the structure after hardening consists only of martensite and retained austenite, and the hardness decreases with carbon content as shown 600 Temperature, 8C 500 400 Ms 300 200 Mf 100 20 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 C, wt% FIGURE 6.119 Martensite start (Ms) and martensite finish (Mf) temperatures vs. carbon content in unalloyed steels. ß 2006 by Taylor & Francis Group, LLC. c 1000 Hardness, HV10 b 800 600 (a) Quenched from γ region to 0 C 400 a (b) Quenched from γ + Fe3C region to 0 C (c) After transformation to 100% martensite 200 0.0 0.2 0.4 0.6 0.8 1.2 1.0 1.4 Temperature, C (a) 1000 Usual hardening temp. E γ 900 A cm γ + Fe3C γ+α 800 700 S 600 Retained austenite, vol% (b) 40 30 20 10 0 0.0 (c) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Carbon content, % FIGURE 6.120 (a) Hardness of carbon (unalloyed) steels depending on carbon content and austenitizing temperature; (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 Verlag ¨ fur Grundstoffindustrie, Leipzig, 1987.) by curve a of Figure 6.120a. If the retained austenite is transformed (e.g., by subsequent cryogenic 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 to the temperature range corresponding to the first tempering stage, the retained austenite is stabilized, which implies that it has become more difficult to transform when subjected to cryogenic treatment. The stabilization of retained austenite is assumed to be due to the dissolution, at the arrest temperature, of the martensite nuclei formed during cooling from the austenitizing temperature. When martempering is performed, i.e., the quenching process is interrupted somewhere around the Ms temperature, a similar stabilization of retained austenite occurs. When the cooling to room temperature is then continued, the same effect, in principle, results as that obtained by the subzero cryogenic treatment with respect to the transformation of retained austenite to martensite. The initial amount of retained austenite (after quenching) is dependent to a very large extent on the austenitizing (hardening) temperature. The higher the hardening temperature, the greater the amount of retained austenite, but greater amounts of retained austenite may also be transformed to martensite by subzero cryogenic treatment for the same stabilizing temperature and same stabilizing time as shown in Figure 6.121. The stabilizing effect ß 2006 by Taylor & Francis Group, LLC. 120 x x 80 x 5% transt. 40 x x 40% 30 20 10 0 101 Stablizing temperature, C (a) 102 103 104 Stablizing time, min x 120 x x 80 x 20% x 40 x 70% transt. 0 101 102 30 103 x 104 30% x 80 40 50 Stablizing time, min (b) 120 60 40 x 50 x 40 x 70% transt. 0 (c) 101 102 103 60 104 Stablizing time, min FIGURE 6.121 Influence of stabilizing temperature and time on the amount of retained austenite that transforms on being subzero treated at À1808C 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 austenite after 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 carried out within 5 min after the temperature of the steel has reached 208C (688F), about 70% of the retained austenite will be transformed. If 40 min is allowed to pass before the subzero treatment, 60% will be transformed, and after 50 h holding at 208C (688F) only 30% of the retained 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 only 10 min before subzero treatment, only 30% of the retained austenite will be transformed. In order to transform the greatest possible amount of retained austenite, the subzero cryogenic treatment should be performed immediately after quenching before tempering. The question of whether the retained austenite in the structure is always detrimental or whether 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 carbon content in the case, the problem of retained austenite is always a real one, especially with steels containing nickel. The higher alloy nickel steels, such as types AISI 4620, 4820, and ß 2006 by Taylor & Francis Group, LLC. 2400 Load, kg 2000 1600 50% retained 0% retained austenite austenite 16MnCr5 (1.13Mn, 1.00Cr) 14NiCr14 (3.67Ni, 0.78Cr) 20MoCr4 (0.49Cr, 0.5Mo) 1200 800 400 104 Ck15 (Carbon steel) 105 106 107 Cycles FIGURE 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. 3:25–28, 1994.) 9310, are particularly likely to have retained austenite in their microstructures after heat treatment because nickel acts as an austenite stabilizer. Tests performed by M. Shea (cited in Ref. [24]), showed marked improvement in tensile bending strain values when retained austenite was present in the 20–30% range for AISI 8620 and 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 be accommodated 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 large quantities of retained austenite (in the range of 50%) improve contact fatigue of carburized and case-hardened steels. In another publication [24], several grades of carburized and casehardened steels were compared (both before and after subzero treatment), and a clear improvement 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 10–20%, and in some instances up to 25%, of retained austenite is not objectionable for most applications and may be beneficial. On the other hand, retained austenite can be detrimental, causing premature wear of sliding on the components’ surface or of sliding and rolling of gear teeth, because it is a softer constituent of the microstructure. The presence of retained austenite in cases of carburized and case-hardened components that are to be subsequently ground is definitely detrimental because under certain grinding conditions it causes severe grinding burns and cracking. The susceptibility of carburized and casehardened components to cracking during grinding becomes greater, the greater the amount of retained austenite, this amount further depending on the steel grade and carburizing temperature as shown in Figure 6.123. 6.3.3.1 Transforming the Retained Austenite When steels are tempered, retained austenite decomposes to bainite during the second tempering stage (230–2808C (446–5368F)). For high-alloy chromium steels, hot-work steels, and high-speed steels, the range of decomposition of retained austenite is shifted toward higher 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 ß 2006 by Taylor & Francis Group, LLC. 100 Carburizing time 3 h 60 Carburizing temp. 50 18CrNi8 15CrNi6 X 20MnCr5 16MnCr5 20NiCrMo6 70 20 MoCr4 in C: 1000 X 40 950 30 X Retained austenite, vol% 80 X 90 X 20 10 900 0 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Cr content, wt% FIGURE 6.123 Influence of the Cr content of low-alloy steels for carburizing on the amount of retained austenite at carburizing. The amount of retained austenite was determined metallographically 0.05 mm ¨ below the specimen’s 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 is cooling down from the tempering temperature. Subzero cryogenic treatment may be applied to transform the retained austenite to martensite, substantially lowering its amount, sometimes to as little as about 1 vol%, which cannot be detected metallographically but only by x-ray diffraction. Decreasing the amount of retained austenite achieves 1. Increase in hardness and consequently in wear resistance 2. More dimensional stability in the finished part (smaller change in dimensions due to structural volume change in use) 3. Less susceptibility to the development of cracks at grinding Figure 6.124 shows a heat treatment cycle that includes subzero treatment. The most important parameters of the treatment are (1) the temperature below 08C (328F) that should Temperature Austenitization Quenching Tempering Time Subzero treatment FIGURE 6.124 A heat treatment cycle including subzero treatment. ß 2006 by Taylor & Francis Group, LLC. be attained and (2) the cooling capacity of the equipment. In some cases, temperatures of À80 to À1008C (À112 to À1488F) are sufficient, but for other steels, especially high-alloy ones, lower temperatures of À1408C (À2808F) or even À1808C (À2928F) are necessary. Holding time at low temperature is unimportant, because the transformation of retained austenite to martensite does not depend on time, but only on the temperature to which the metal has been cooled. Only that portion of the retained austenite will be transformed to martensite that corresponds to the cooling temperature realized. Further transformation will take place only if 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 at most À758C (À1038F) or À788C (À1088F) and is used for small quantities and small mass of parts. 2. Circulating air that has been cooled in a heat exchanger. This low-temperature cascade system (Figure 6.125) cools the parts put in a basket with air circulated by a fan. The air flows from top to bottom, extracting the heat of the parts, exiting through a grate at the bottom of the basket and flowing further through the heat exchanger, which is cooled by two or four compressors. Such metal-treating freezers have been built with a capacity to cool a mass of 270–680 kg of parts to À858C (À1218F) in about 2 h. 3. Evaporation of liquid nitrogen. For subzero treatment of relatively small quantities of parts down to À1808C (À2928F), equipment such as that shown in Figure 6.126 is used. The parts to be cooled are put in the working space 1, and the liquid nitrogen is in the container 4. Because of heat coming through the walls, the pressure in the container 4 increases. Using this pressure, when the valve 6 is opened, liquid nitrogen is injected into the working space, where it evaporates instantly. A fan 7 circulates the evaporated nitrogen through the working space, taking the heat out of parts and lowering their temperature. The amount of the injected liquid nitrogen and consequently the temperature of cooling can be controlled by adjusting the valve 6. The overpressure that develops in the working space because of constant nitrogen evaporation is let out through the exhaust valve 8. A temperature of À1808C (À2928F) can be reached in less than 10 min. 4. In a container connected to a cryogenerator. This method enables subzero treatment of large quantities of parts to as low as À1908C (À3108F). The cryogenerator powered FIGURE 6.125 Low-temperature cascade system for subzero cooling by circulating air that has been cooled in a heat exchanger. ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.126 Subzero treatment equipment with evaporation of liquid nitrogen. by an electric motor works on the principle of the Stirling cycle. By continuous circulation of air the working space with parts is gradually cooled to desired temperature. Figure 6.127 shows the cooling curve from room temperature to À1808C (À2928F) and the natural reheating curve from À180 to 08C (À292 to 328F) for the empty container of 100 dm3, connected to a cryogenerator. It can be seen that after 1 h of cooling a temperature of À1208C (À1848F) has already been reached, but an additional 1.5 h is necessary to reach À1808C (À2928F). The natural reheating from À180 to 08C (À292 to 328F), as shown, takes much longer (about 20 h). 6.4 6.4.1 HARDENING AND TEMPERING OF STRUCTURAL STEELS MECHANICAL PROPERTIES REQUIRED A combined heat treatment process consisting of hardening plus tempering (to temperatures between 450 and 6808C (842 and 12568F)) applied to structural steels (in German called ¨ Vergutung) 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 to C 25 K 298 0 273 Hea ti n g curve cu rve C oo l i n g –50 –100 223 173 –150 123 –180 04 8 12 16 20 93 Heating hours 3 2 Cooling hours 1 FIGURE 6.127 Cooling curve and natural heating back curve of an empty container connected to a cryogenerator. ß 2006 by Taylor & Francis Group, LLC. Strength sN < sV < sH Hardened H Ductility eN eV > eH d,y Toughness zN < Z V > ZH ak, sB, d s Hardened and tempered Strength V Normalized N ss ne gh u To ss ne h ug To Brittleness N V Ductility e FIGURE 6.128 Schematic presentation of ductility, toughness, and brittleness. (From E. Just, VDI-Ber. 256:125–140, 1976 [in German].) withstand dynamic loading or impact. The aim of hardening and tempering structural steels will be better understood if one has a clear notion of the difference between toughness and ductility. Ductility is the property denoting the deformability of a material and is measured in a tensile test as elongation (A in %) and reduction of area (Z in %). It is a one-dimensional property. Toughness of a material, however, is a two-dimensional property because it is an integral (or product) of strength and ductility, as schematically shown in Figure 6.128. Steels of the same ductility but different strength levels can differ in toughness. As Figure 6.128 shows, 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 in separate tests as impact toughness (ak, J/cm2) or as fracture toughness (KIC, N/mm3/2). The lower the ductility of a material, the more brittle it is. Total brittleness accordingly denotes zero ductility of the material. The aim of the hardening and tempering process can also be explained by means of the stress–strain diagram schematically shown in Figure 6.129. As hardened, a steel has high yield strength but low ductility, and a small area below the stress–strain curve (curve 2) indicates low toughness. As-hardened and tempered (curve 3) steel has higher yield strength than in its normalized condition but also much higher ductility than in its hardened condition. The greatest area below the stress–strain curve indicates a substantial increase in toughness compared to either normalized or hardened conditions. For a certain steel grade, the relation between mechanical properties and the tempering temperature can be read off from a diagram as shown in Figure 6.130 for the steel DIN 20CrNiMo2 (0.15% C, 0.20% Si, 0.88% Mn, 0.53% Cr, 0.50% Mo, 0.86% Ni). It can be clearly seen from the lower part of this diagram how the impact toughness increases when the steel is tempered to a temperature above 5508C (10228F). Such diagrams enable precise optimization of the strength level and toughness by selection of the proper tempering temperature. ß 2006 by Taylor & Francis Group, LLC. 1. Normalized 2. Hardened Hardened 3. and tempered 2 Stress (s), N/mm2 3 1 Strain (e) FIGURE 6.129 Stress–strain 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 the microstructure after hardening and tempering. Maximum toughness values are obtained when tempering a structure that after quenching consists of fine-grained martensite (having a grain size of ASTM ! 6) (see Figure 6.131). How different microstructures after different heat treatment processes influence the impact toughness of 3.5% Ni steel at low temperatures Rm 900 90 Rp0.2 80 800 70 700 Z Contraction z, % 100 1000 Impact toughness ak, J/cm−2 Elongation, % , N/mm2 Tensile strength Rm Yield strength Rp0.2, N/mm2 1100 60 600 500 30 20 10 200 ISO-V specimens ak(−20 C) 100 50 ak(−50 C) 0 650 700 500 550 600 Tempering temperature, C FIGURE 6.130 Hardening and tempering diagram of DIN 20CrNiMo2 steel. Hardening temperature 9508C; 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, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. FIGURE 6.131 Microstructure of DIN 34CrNiMo6 steel after hardening and tempering. Tempered ¨ fine-grained martensite. Magnification 500Â. (From G. Spur and T. Stoferle (Eds.), Handbuch der ¨ Fertigungstechnik, 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 in impact toughness begins) is of special interest. The lower the transition temperature, the higher the toughness. Certainly, when hardening workpieces of big cross section, not only martensite 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 surface toward the core of the workpiece. So after tempering, besides tempered martensite, other structural constituents having lower toughness are present. Figure 6.133 shows the relationships among transition temperature, yield strength, and microstructure. For high strength values especially, the superiority of fine-grained martensite structure with respect to toughness is evident. From a series of tests with hardened and tempered steels with about 0.4% C, Figure 6.134 shows a general relation between the structural constituents and the properties characterizing ductility (elongation and reduction of area) and impact toughness, respectively, for different levels of yield strength. It is clear that tempered martensite always gives the best ductility and toughness. Impact energy (ISO-V) specimens 120 a 90 b c 60 d 30 0 –150 –100 –50 0 Temperature, 8C 50 FIGURE 6.132 Influence of different microstructure and respective heat treatments on the impact toughness at low temperatures (ISO notch specimens) of a 3.5% Ni alloyed steel. a, Hardened by quenching in water and tempered; b, normalized and tempered; c, normalized only; d, hardened ¨ by quenching in water only. (From G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, ¨ Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. B % % M/ 25 % B M/ 50 % B % 75 50 l M/ 75 B % 0% 25 10 30 50 % F+ P/5 % 0% 10 F/70 B % % F/9 B 10 0% 0% B Bu P F+ 0% 10 F−P 50 B GS 20 0 RT G S ASTM 3 GS Transition temperature, C 100 −50 M ASTM 7 −100 ISO-V longitudinal −150 200 400 600 1000 800 1200 1400 Yield strength Rp, N /mm2 FIGURE 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). (From ¨ ¨ G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) Elongation, % 25 M 20 B 15 F+P 10 5 (a) 0 70 M Contraction Z, % 60 B 50 40 F+P 30 20 10 (b) Impact energy (20 C) 0 250 (c) 200 M 150 100 50 0 300 B F+P 500 700 900 1100 1300 Yield strength Rp, N/mm2 FIGURE 6.134 (a) Elongation; (b) reduction of area; and (c) impact toughness of hardened and tempered 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 6–7. Impact toughness: ISO notch ¨ specimens. Testing direction: longitudinal. (From G. Spur and T. Stoferle (Eds.), Handbuch der Ferti¨ gungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987.) ß 2006 by Taylor & Francis Group, LLC. 175 ] Impact energy (20 C) 150 Rm = 700 N/mm2 125 100 75 850 1000 1200 1400 50 25 0 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Carbon content, % FIGURE 6.135 Impact toughness as a function of tensile strength and carbon content for the structure ¨ of tempered martensite. Grain size: ASTM 6–7. (From G. Spur and T. Stoferle (Eds.), Handbuch der ¨ Fertigungstechnik, 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 in Figure 6.135, of steels for hardening and tempering, those with 0.2–0.3% C have the best impact toughness. When testing the impact toughness of a steel, one should be aware that toughness is usually higher in the longitudinal direction (rolling direction) than in the transverse direction. That is because some phases or nonmetallic inclusions that are present in every steel (carbides, oxides, and sulfides) are stretched during rolling in the longitudinal direction. In this way a textured structure originates that has lower impact toughness in the transverse direction than in the longitudinal direction. As a measure of this effect, the factor of isotropy (the ratio of transverse impact toughness to longitudinal impact toughness) is sometimes used. The great influence of the microstructure after hardening (before tempering) on the impact toughness of a steel is evident from Figure 6.136. Appearance of preeutectoid Percentage of structure before tempering Impact toughness, J/cm–2 200 98 M 150 50 M 50 B 100 80 M 20 F + P 50 0 –120 50 M 50 F + P –80 –40 0 40 80 Temperature, C FIGURE 6.136 Influence of the microstructure after hardening (before tempering) on the impact ¨ ¨ toughness of DIN 16MnCr5 steel. (From H.J. Spies, G. Munch, and A. Prewitz, Neue Hutte 8(22):443–445, 1977 [in German].) ß 2006 by Taylor & Francis Group, LLC. ferrite or ferrite and pearlite in the structure results in a substantial decrease in the impact toughness. When selecting a structural steel for hardening and tempering, in order to better adapt the mechanical properties to the requirements of the treated parts, the expected microstructure must be considered. To be able to reproducibly influence mechanical properties, one should know the relationships among the heat treatment regime, microstructure, and resulting mechanical properties. Unalloyed steels for hardening and tempering, because of their low hardenability, exhibit a high degree of section sensitivity with respect to hardness distribution after hardening as shown in Figure 6.137. After quenching a bar specimen of 30-mm diameter of the steel in question in conventional quenching oil, a hardness of only 40 HRC was achieved at the surface. When specimens of the same diameter were quenched in fast quenching oil, the hardness was 45 HRC; when quenched in 10% aqua-quench solution the hardness was 56 HRC; 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 is used, for large cross-sectional sizes, the depth of hardening will be small and the core will remain unhardened. Because of the second conclusion, when selecting a structural steel grade for hardening and tempering, its hardenability must always be adapted to the workpiece’s cross-sectional size and the required strength level. Figure 6.139 shows the preferred fields for the application of some common steel grades for hardening and tempering according to the actual bar diameter and the strength level required. This recommendation is based on the assumption that 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) and 60 ∅10 Hardnes, HRC 50 ∅20 ∅30 40 30 ∅40 ∅50 20 0 25 20 Surface 15 10 5 0 Center 5 10 15 20 25 mm Surface FIGURE 6.137 Hardness distribution (measured) on the cross section of bars of different diameters made of unalloyed steel (0.52% C, 0.24% Si, 0.90% Mn, 0.06% Cr) quenched in conventional hardening oil from 8608C. (From K.E. Thelning, Steel and Its Heat Treatment, 2nd ed., Butterworths, London, 1984.) ß 2006 by Taylor & Francis Group, LLC. 60 ∅10 ∅20 Hardness, HRC 50 ∅30 40 ∅40 ∅50 30 20 0 25 20 15 10 5 Surface 0 5 10 15 Center 20 25 mm Surface FIGURE 6.138 Hardness distribution (measured) on the cross section of bars of 10–50-mm diameters made of unalloyed steel (0.52% C, 0.24% Si, 0.90% Mn, 0.06% Cr) quenched from 8608C in water containing 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) are required. 6.4.2 TECHNOLOGY OF THE HARDENING AND TEMPERING PROCESS Hardening, which is the first operation of the hardening and tempering process, will yield a martensitic structure (provided a correct austenitization and quenching with a cooling rate greater than the critical rate for the steel in question have been realized), the hardness of which depends on the dissolved carbon content, according to the empirical formula pffiffiffiffi (6:45) H100 % mart % 60 C þ 20 where C is the carbon content in wt%. Tensile strength, N/mm2 1500 1400 30CrNiMo8 1300 1200 42CrMo4 37Cr4 900 800 37Cr4 30CrNiMo8 Ck45 30CrNiMo8 42CrMo4 37Cr4 Ck45 700 600 30CrNiMo8 42CrMo4 1100 1000 30CrNiMo8 42CrMo4 42CrMo4 37Cr4 Ck45 Ck22 Ck22 500 to 16 mm >16 to 40 mm >40 to 100 mm >100 to 160 mm >160 to 250 mm Bar diameter FIGURE 6.139 Applicability of steel grades for hardening and tempering according to required strength level and bar diameter. Steel designations according to DIN. (From German standard DIN 17200.) ß 2006 by Taylor & Francis Group, LLC. The required critical cooling rate for unalloyed steels is about 2508C/s (4828F/s). Alloyed steels have lower critical cooling rates or a higher hardenability, which means that the same quenching conditions yield a greater depth of hardening. As a measure of hardenability, the ideal critical diameter D1 (see Chapter 5) can be applied. The actual depth of hardening, however, is influenced by, in addition to the alloying elements, the austenitizing temperature (especially for steels containing carbides difficult to dissolve) and quenching conditions. Consequently, two steels having the same D1 value may give different depths of hardening. For a designer, therefore, information based only on the percentage of martensite does not seem practical, because even for the same D1 values the designer might get different depths of hardening. Besides, microstructures having the same amount of martensite do not always give the same hardness. The hardness of martensite depends on dissolved carbon content and may be calculated for 50% martensite according to the empirical formula pffiffiffiffi H50% mart % 44 C þ 14 for C < 0:7 % (6:46) More practical information for the designer, about expected depth of hardening, may be obtained 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 equivalent distance (E in mm) on the relevant Jominy curve. According to Just [25], this correlation for the surface of round bars reads Esurf ¼ R1=2 À 3 [mm] for R < 50 mm; E < 30 mm (3=4)H 3=4 (6:47) and for the core of round bars: Ecore ¼ R [mm] 2H 1 = 4 (6:48) After calculating the equivalent Jominy distance E, one can read off the hardness from the relevant Jominy curve. Figure 6.140 shows this correlation as a diagram (for radii from 0 to 50 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 on how 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 properties after subsequent tempering. To check the quality of hardening achieved, the degree of hardening (S ) can be used. It is the ratio between the achieved (measured) hardness and the maximum hardness attainable with the steel in question: S¼ H H ¼ pffiffiffiffi Hmax 60 C þ 20 (6:49) where H is the actual hardness measured at a specified point of the cross section, in HRC, and Hmax is the maximum attainable hardness in HRC. The degree of hardening is valid, of course, 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 of S > 0.7 (corresponding to about 50% martensite) are satisfactory. ß 2006 by Taylor & Francis Group, LLC. 40 R Es = 34 4 30 H 3 −3 H 0.3 0.4 Jominy distance E, mm 20 10 1 2 Surface 0 40 Ec = 2 R 4 H 0.3 0.4 1 2 H 30 20 10 Core 0 0 10 20 30 40 Radius R, mm 50 FIGURE 6.140 Correlation among radius of round bars (quenched by immersion), quenching intensity H and equivalent Jominy distance for the surface and core of the bars. (From E. Just, VDI-Ber. 256:125– 140, 1976 [in German].) When hardening and tempering structural steels, the value of hardness after hardening and tempering is usually specified. The required degree of hardening can also be expressed as a function of the hardness after hardening and tempering (Ht): S! 1 1 þ 8eÀHt =8 (6:50) Figure 6.141 shows the minimum values of the required degree of hardening as a function of hardness after hardening and tempering, limiting the allowed area. By specifying the required degree of hardening, one can avoid the risk of an incorrect hardening and tempering. It is known that too low a value of hardness after hardening (not enough 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 to the required value, the toughness and other mechanical properties important for dynamically stressed parts will be insufficient because of an inadequate microstructure. Such a risk can be avoided by specifying the minimum degree of hardness for the critical point of the cross section, which can easily be checked after hardening. In selecting sufficiently severe quenchants to obtain a high percentage of martensite and great depth of hardening, one has to be aware of the risk of cracking. Hardening cracks are dependent 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 quenching conditions favor the formation of cracks) 3. The steel grade itself (the lower the Ms temperature of the steel, the greater the risk of cracking) ß 2006 by Taylor & Francis Group, LLC. S= Hh 20 + 60√C Allowed region 0.8 0.6 S> 1 1 + 8e−Ht /8 0.4 Not allowed region 0 40 10 20 30 Hardness after tempering Ht 1700 1500 1300 0.0 1000 0.2 800 Degree of hardening S 1.0 N/mm2 50 HRC FIGURE 6.141 Minimum values of the degree of hardening required as a function of hardness after hardening and tempering. (From E. Just, VDI-Ber. 256:125–140, 1976 [in German].) The Ms temperature can be calculated using the formula Ms ¼ 548 À 440 C À 14Si À 26Mn À 11Cr À 9Mo À 14Ni þ 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 on the risk of cracking. Tempering, which is the second important operation, decreases high hardnesses more than low hardnesses, as can be seen in Figure 6.142. This figure shows the Jominy curve of the steel DIN 28Cr4 (0.24–0.31% C, 0.15–0.36% Si, 0.62–0.78% Mn, 0.75– 1.07% 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 has 60 Hh = 20 + 60 √C Hardness, HRC 50 Steel DIN 28Cr4 Hardness after Jominy test Tempered to 500 C 60 min Tempered to 600 C 60 min 40 30 20 10 0 0 10 20 30 40 50 60 Jominy distance, mm FIGURE 6.142 Influence of tempering temperature on level of hardness at various Jominy distances. (From E. Just, VDI-Ber. 256:125–140, 1976 [in German].) ß 2006 by Taylor & Francis Group, LLC. been decreased much more by tempering than low hardness values at greater distances from the quenched end. With respect to the cross section of hardened real components, this effect means 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 (in the range from about 320 to 7208C (608 to 13288F)) and a logarithmic function of tempering time, according to the following formula [25], which is valid for a 100% martensite structure: Ht ¼ 102 À 5:7  10À3 [Tt (12 þ 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 respect to resulting hardness; however, very short or very long tempering times do not yield optimum toughness. To obtain the optimum toughness for chromium steels, the tempering times should be between 1 and 5 h. There is a firm relationship between the hardness after tempering and the hardness after hardening. Spies et al. [26] have, by using multiple linear regression, quantified the influence of hardness after hardening, chemical composition, and tempering temperature on hardness after tempering and developed the formula HB ¼ 2:84Hh þ 75(% C) À 0:78(% Si) þ 14:24(% Mn) þ 14:77(% Cr) þ 128:22(% Mo) À 54:0(% V) À 0:55T t þ 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: Hh C Si Mn Cr Tt 20–65 HRC 0.20–0.54% 0.17–1.40% 0.50–1.90% 0.03–1.20% 500–6508C (932–12028F) According to the German standard DIN 17021, an average relation between the hardness after hardening (Hh) and the hardness after hardening and tempering (Ht) reads Hh ¼ (Tt =167 À 1:2)Ht À 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 time of 1 h. Because, as already mentioned, high hardnesses decrease at tempering much more than low hardnesses, the prediction is more precise if the degree of hardening (S) is accounted for. Hh ¼ 8 þ (Ht À 8) exp [S(Tt =917)6 ] [HRC] where S is the degree of hardening, S < and Tt is tempering temperature (K). ß 2006 by Taylor & Francis Group, LLC. (6:55) 650 600 Hardness after hardening Hh, HRC T 8C 60 500 400 S=1 500 50 650 600 500 400 S = 0.7 40 30 20 10 0 0 10 20 30 40 50 60 Hardness after tempering Ht, HRC FIGURE 6.143 Relationship among hardness after hardening, degree of hardening, tempering temperature, and hardness after tempering. (From E. Just, VDI-Ber. 256:125–140, 1976 [in German].) Figure 6.143 shows a diagram from which it is possible to predict at a glance the hardness required after hardening for a desired hardness after tempering, taking into account the actual tempering temperature and the necessary degree of hardening. It is also possible to calculate the necessary tempering temperature for a specified hardness after hardening and tempering when chemical composition and the degree of hardening are known: pffiffiffi Tt ¼ 647[S(60 c þ 20)=Ht À 0:9]1=4 À 3:45SHt þ (537 À 561S )(%g C) þ 505S(%g V) þ 219S(% Mo) þ 75S(% Cr) þ 66S (% Si) À 51(C) (6:56) where Ht is hardness after hardening and tempering [HRC]; S is the degree of hardening, S 1.0; 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 tempering temperature may be determined if the ultimate tensile strength or yield strength required after hardening and tempering is known. Figure 6.144a and Figure 6.144b show such diagrams 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 DETERMINATION OF PROCESS PARAMETERS Increasingly, modern heat treatment equipment incorporates microprocessors for automatic control of temperature–time cycles, protective or reactive atmosphere, material handling, and, to some extent, quenching operations. On the other hand, determination of the process parameters necessary to achieve the heat-treated properties required is normally based on empirical results. For routine often-repeated heat treatment processes (e.g., carburizing, hardening, and tempering), computer programs can be written to establish treatment parameters provided that adequate data are available on workpiece geometry, material properties desired after heat treatment, steel grade used, and the actual heat treatment equipment itself. The aims of such an approach are to optimize the heat treatment operation by saving time and energy and to maintain close tolerances on the material properties imparted. The basic prerequisite is the ß 2006 by Taylor & Francis Group, LLC. Water quenched from 830−850 C Diam. 25 mm N/mm2 kp/mm2 Diam. 50 mm kp/mm2 Diam. 100 mm kp/mm 1370 140 140 140 1180 120 120 120 980 100 100 100 780 80 590 40 200 20 0 (a) Z 60 390 80 Rm 80 Rm 60 Z Rp0.2 Z 40 A5 0 500 550 600 650 700 Rm 60 Rp0.2 40 20 A5 0 500 550 600 650 700 Rp0.2 20 A5 0 500 550 600 650 700 Tempering temperature, C N/mm2 Oil quenched from 840–870 C Diam. 25 mm Diam. 50 mm kp/mm2 kp/mm2 1370 140 140 1180 120 120 980 100 100 780 80 80 Rm 590 390 Z 60 Rp0.2 40 60 40 A5 200 0 20 20 0 500 550 600 650 700 (b) Rm Z Rp0.2 A5 0 500 550 600 650 700 Tempering temperature, C FIGURE 6.144 Tempering diagrams for the unalloyed steel DIN C45 when quenched (a) from 830 to 8508C in water, for bar diameters 25, 50, and 100 mm and (b) from 840 to 8708C in oil, for bar diameters 25 and 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 prediction of relevant metallurgical and physical phenomena. ˇˇ ´ As described by Liscic and Filetin [27], the process of hardening and tempering has been divided into operations of austenitization (A), quenching (Q), and tempering (T) as shown in Figure 6.145. Within the austenitization operation, the phases are (1) preheating, (2) heating to 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 advantages compared with earlier methods: 1. Planning of the process and preparation of the technological documentation are incomparably faster and simpler. ß 2006 by Taylor & Francis Group, LLC. Temperature Surface Core Hardening 1 2 3 A Tempering 4 5 Q 6 7 T Time FIGURE 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 time– temperature profiles can be determined more precisely. 3. The professional level of the technological documentation is always high and consistent, 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 some other equipment for the same process. 5. If modern heat treatment equipment with microprocessor control is available, the resulting treatment parameters can be distributed directly (online) to all units where the process will be performed and controlled automatically. The general scheme of computer-aided planning of the hardening and tempering process is shown in Figure 6.146. Use of a computer for this purpose requires Preparatory stage Database on steel characteristics Production Calculations / subroutines Feedback information Input data: on steel grade (heat) Computer-aided process planning Parameters for A, Q, T On the workpiece On required mechanical properties On equipment Process operation A Q Technological documentation and temperature−time diagram Performance of hardening and tempering Checking of resultant properties T Feedback information Equipment characteristics Data on quenching conditions Feedback information on resultant properties FIGURE 6.146 Scheme of computer-aided planning of the hardening and tempering process. (From ˇˇ ´ B. Liscic and T. Filetin, Heat Treat. Met. 3:62–66, 1987.) ß 2006 by Taylor & Francis Group, LLC. 1. A database on characteristics of the steel grades treated 2. A database on the equipment employed (especially data on quenching severities available) 3. Subprograms stored in the computer memory for the necessary calculation of parameters The input data in a particular case are: 1. Data on the steel grade in question 2. Data on the workpiece (shape, critical cross section, surface condition, number of pieces 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 percentage of 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 grade or heat: chemical composition; carbon equivalent; austenitizing temperature; time for homogenization of the structure; Ms temperature; Jominy hardenability curve; holding time at tempering temperature; and susceptibility to temper brittleness. To determine the parameters of the hardening and tempering process, the following relationships must be known and stored in the computer memory in the form of adequate mathematical equations: 1. The effect of shape and cross-sectional size of the workpiece on the time necessary for heating 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), a preheating 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 a and b were experimentally determined for the equipment used. 2. The influence of steel grade, cross-sectional size, and actual quenching conditions on the depth of hardening. This is necessary for selection of optimum quenching conditions (quenching medium, temperature, and agitation rate) to satisfy the required degree of hardening. The method by which this selection was performed is described in Chapter 5 (see Section 5.6.4). 3. The relationship between hardness after tempering and tempering temperature for the steel in question. The necessary tempering temperature (Tt) was calculated by means of the formula ! ln (Hh À 8)=(Ht À 8) 1=6 À273 [8C] Tt ¼ 917 S (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 390 and 6608C (734 and 12208F). The time necessary for heating up to the tempering temperature and for temperature equalization 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 at tempering temperature was taken from the database for the steel grade in question. ß 2006 by Taylor & Francis Group, LLC. Technological documentation for the process: hardening and tempering Workpiece: bolt Dimensions, mm: 40 dia. x 120 Steel grade: BS 708A37 Requirements: R = 900 N/mm2 Hardening grade = .84 Point on cross section: 1/4R Technological parameters Operation Equipment and/or media Temperature ( C) Time (min) Preheating Chamber furnace air atmosphere 450 62 Austenitization Chamber furnace air atmosphere 840 47 Quenching Oil-agitation 1 m/s 20 10 Tempering Pit furnace-circulating air 627 152 In air 20 140 Cooling after tempering Total time = 411 min 900 T, C 800 700 600 500 400 300 200 100 1 2 3 4 5 6 7 8 h FIGURE 6.147 An example of the computer-generated parameters and time–temperature cycle for ˇˇ ´ hardening and tempering a 40-mm diameter bolt made of BS 708A37 (En 19B) steel. (From B. Liscic and T. Filetin, Heat Treat. Met. 3:62–66, 1987.) Cooling from the tempering temperature is carried out in air or inert gas in all cases where the steel is not prone to temper brittleness. If it is susceptible, faster cooling in oil or in air blast in necessary. By using the described subprograms, the input data are processed by means of stored equations into the following output data or process parameters: 1. Temperature and time of preheating 2. Temperature and time for austenitization 3. Quenching conditions (quenching medium, its temperature, its agitation rate; time required for complete cooling of the workpiece) ß 2006 by Taylor & Francis Group, LLC. TABLE 6.6 Comparison of Required and Measured Hardness at the Center of Cylinders Made of Steel Grade BS 708A37 (En 19B), after Hardening and Tempering under Computer-Calculated Conditions Dimensions (mm) Required Values Length 120 200 320 Diameter Hardening Grade (S) Ultimate Tensile Strengtha (N/mm2) 0.95 0.84 0.65 Measured Values Hardness after Tempering (HRC) 1240 1000 850 30 50 80 Tempering Temperature (8C) Ultimate Tensile Strength (N/mm2) Hardness after Tempering (HRC) 519 583 621 1210 1050 900 37.3 32.6 27.7 38 31 24.5 a Calculated from the hardness (DIN 50150). ˇˇ ´ Source: From B. Liscic and T. Filetin, Heat Treat. Met. 3:62–66 (1987). 4. Temperature and time for tempering 5. Mode of cooling from tempering temperature to room temperature Figure 6.147 shows an example of computer-generated documentation for hardening and tempering 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 bars of 30-, 50-, and 80-mm diameters made of grade BS 708A37 steel with the measured hardness after hardening and tempering under computer-calculated conditions. 6.5 AUSTEMPERING The austempering process (see Figure 6.148) consists of austenitization, quenching into a hot bath maintained between 260 and 4508C (500 and 8428F), holding at this temperature until the transformation of austenite to bainite is complete, and cooling to room temperature at T T Ta Ta Temperature, C Temperature, C Ac3 Ttr (a) Ac3 F P T tr B Ms (b) Time, s t Time, s log t FIGURE 6.148 Scheme of an austempering process (a) in time–temperature diagram and (b) in isothermal transformation (IT) diagram. ß 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 final structure (bainite) is obtained gradually during the isothermal transformation of austenite 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 process Dealing with austempering one should use the IT diagram of the steel in question to optimize 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 cracking 2. Increased ductility and toughness, especially in the range of high strength (hardness) values between 50 and 55 HRC Reduced distortion and less possibility of cracking are the result of lower thermal stresses, as well as lower transformational stresses compared to conventional hardening. Although at austempering there are also temperature differences between the surface and core of the workpiece, during quenching, these differences are substantially smaller, as shown in Figure 6.149, because the difference between the austenitizing temperature and the temperature of the quenching bath is much smaller (for 200–4008C (392–7528F)) than in conventional hardening. 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 different points 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 effects contribute to much lower risk of cracking and distortion, thereby minimizing the production of scraped parts and additional cost for straightening or grinding to repair the distortion. Temperature The greatest temperature difference between surface and core at I The greatest temperature difference between surface and core at II C S S = surface C = core C S Time I Conventional hardening and tempering II Austempering FIGURE 6.149 Temperature differences between surface and core of the workpiece in conventional ¨ hardening and in austempering. (From K.H. Illgner, Fachber. Huttenpraxis Metallweiterverarb. 17(4):281–288, 1979 [in German].) ß 2006 by Taylor & Francis Group, LLC. Increased ductility and toughness as well as increased bendability and fatigue life are the strongest reasons to apply austempering instead of hardening and tempering. Figure 6.150 shows the relation of impact toughness and Brinell hardness (HB) of a Cr–Mn–Si steel after conventional hardening and tempering and after austempering, as a function of tempering temperature and austempering temperature, respectively. The most important difference is that a good combination of hardness and toughness after conventional hardening and tempering is possible only at high tempering temperatures, which means low hardness, whereas at austempering a good combination of hardness and impact toughness may be achieved at high hardness values. Another comparison of impact toughness of a carbon steel after hardening and tempering and after austempering, as a function of hardness, is shown in Figure 6.151. It is evident that austempering yields much better impact toughness, especially at high hardness, around 50 HRC. It is necessary to emphasize that high toughness after austempering is possible only under conditions of complete transformation of austenite to bainite. Table 6.7 shows a comparison of some mechanical properties of austempered and of hardened and tempered bars 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, and fatigue life. Figure 6.152 shows the fatigue diagram of DIN 30SiMnCr4 steel after conventional hardening and tempering and after austempering. The increase in fatigue resistance values is especially remarkable for notched specimens. Regarding bendability, Figure 6.153, from an early work of Davenport [30], shows the results 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 suitability of a steel for austempering is limited first of all with minimum incubation time (the distance of the transformation start curve from the ordinate). Another limitation may be the very long transformation time. Figure 6.154 shows the transformation characteristics of four AISI grades of steel in relation to their suitability for austempering. The AISI 1080 steel has only limited suitability for austempering (i.e., may be used only for very thin cross sections) 550 Hardness Hardening and tempering Hardness Hardness HB 500 14 12 10 450 8 400 350 300 250 Impact toughness 6 4 Impact toughness Impact toughness, kg/cm2 Austempering 600 2 0 200 250 300 350 400 300 400 500 550 600 Austempering temperature, C Tempering temperature, C FIGURE 6.150 Impact toughness and hardness (HB) of five heats of a Cr–Mn–Si steel after conventional hardening and tempering and after austempering, as a function of tempering temperature and austempering temperature, respectively. (From F.W. Eysell, Z. TZ Prakt. Metallbearb. 66:94–99, 1972 [in German].) ß 2006 by Taylor & Francis Group, LLC. Rod diam. 0.180 in. C 0.74% Mn 0.37% Si 0.145% 0.039% S P 0.044% Energy absorbed in breaking impact strength, ft-lb 50 Each plotted point represents the average of several tests 40 30 Quench and temper method 20 Direct method austempered 10 0 40 45 50 55 60 Rockwell C hardness 65 FIGURE 6.151 Comparison of impact toughness of a carbon steel after conventional hardening and tempering and after austempering, as a function of hardness. (From G. Krauss, Steels: Heat Treatment and Processing Principles, ASM International, Materials Park, OH, 1990.) because the pearlite reaction starts too soon near 5408C (10048F). The AISI 5140 steel is well suited to austempering. It is impossible to austemper the AISI 1034 steel because of the extremely fast pearlite reaction at 540–5958C (1004–11038F). The AISI 9261 steel is not suited to austempering because the reaction to form bainite is too slow (too long a transformation time) at 260–4008C (500–7528F). The austempering process is limited to sections that can be cooled at a sufficient rate to prevent transformation to pearlite during quenching to the austempering bath temperature. Maximum section thickness is therefore important in determining whether or not a part can be successfully austempered. For eutectoid or hypereutectoid carbon steels like AISI 1080, a section thickness of about 5 mm is the maximum that can be austempered to a fully bainitic structure. Unalloyed steels of lower carbon content are restricted to a proportionately lesser thickness (except those containing boron). With increasing alloy content, heavier sections can be austempered, in some alloy steels up to 25 mm cross section. When some pearlite is permissible 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 austempered parts made of various steels. Process parameters for the austempering process are: 1. Austenitizing temperature and time 2. Quenching intensity when cooling from the austenitizing temperature to the austempering bath temperature 3. Temperature of transformation, i.e., the austempering bath temperature 4. Holding time at austempering temperature The austenitizing temperature and time (as in any hardening process) are responsible for carbide dissolution and homogenizing of the structure, which has a substantial influence on the impact toughness of treated parts. ß 2006 by Taylor & Francis Group, LLC. TABLE 6.7 Comparison of Some Mechanical Properties of Austempered and of Hardened and Tempered Bars Made of AISI 1090 Steel Propertya Austempered at 4008C (7508F)b Quenched and Temperedc 1,415 (205) 1,020 (148) 11.5 30 415 105,000e 1,380 (200) 895 (130) 6.0 10.2 388 58,600f Tensile strength, MPa (ksi) Yield strength, MPa (ksi) Elongation, % Reduction of area, % Hardness, HB Fatigue cyclesd a Average values. Six tests. c Two tests. d Fatigue specimens 21 mm (0.812 in.) in diameter. e Seven tests: range 69,050–137,000. f Eight tests: range 43,120–95,220. Source: From ASM Handbook, 9th ed., Vol. 4, Heat Treating, ASM International, Materials Park, OH, 1991, p. 155. b Quenching must be severe enough to avoid any pearlite formation on cooling from the austenitizing temperature to the temperature of the austempering bath. Molten nitrite–nitrate salts 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 water to a hot salt bath, care must be taken to prevent spattering. The higher the temperature of the salt bath, the less water should be added. Because of evaporation, the amount of water added must be controlled. Bending fatigue strength s±, N/mm2 720 680 640 Smooth specimens 600 560 Austempered sB = 1260 N/mm2 Hardened + tempered sB = 1380 N/mm2 Hardened + tempered sB = 1260 N/mm2 520 480 440 400 Notched specimens 360 320 280 2 3 45 6 2 3 45 7 10 Number of cycles 7 107 FIGURE 6.152 Fatigue of DIN 30SiMnCr4 steel after conventional hardening and tempering and after austempering. (From F.W. Eysell, Z. TZ Prakt. Metallbearb. 66:94–99, 1972 [in German].) ß 2006 by Taylor & Francis Group, LLC. Hardness: Rockwell C 50 Austempered Quenched and tempered FIGURE 6.153 Carbon steel wire (0.78% C, 0.58% Mn) of 4.6-mm diameter, (left) austempered and (right) hardened and tempered to 50 HRC and bent under comparable conditions. (From E.S. Davenport, Steel, March 29, 1937.) 800 800 500 400 F+C A+F+C A 300 Ms 200 100 0 1 min 1080 −1 10 1 10 (a) 10 min 2 10 Time, s 1 day 10 h 1h 3 4 10 10 600 A+F 300 MI 200 5140 10−1 Ms 200 1 min 1034 10−1 1 10 10 min 102 Time, s 62 1 103 1 day 10 h 1h 104 Ae3 Ae1 700 10 102 Time, s 103 104 28 34 38 37 31 37 41 50 55 F+C 500 A+F+C 400 A 300 Ms 200 100 52 0 105 (d) 1 min 9261 1 105 A 600 Temperature, C F+C 300 0 (c) 12 25 25 27 35 43 A+F+C 100 F+C A+F+C Ms (b) Ae1 500 400 A 400 0 10 Hardness, HRC Temperature, C 600 A+F 500 800 Ae3 13 24 31 26 30 37 44 50 51 53 100 66 5 800 700 Ae1 10 102 10 min 103 Time, s 1 day 10 h 1h 104 Hardness, HRC Temperature, C 600 Ae3 A 700 Temperature, C 11 32 38 40 40 41 43 50 55 57 Hardness, HRC Ae1 700 Hardness, HRC Transformation temperature, i.e., austempering bath temperature, is one of the two most important parameters as it directly influences the strength (hardness) level of the treated parts. The higher the austempering temperature, the lower the strength (hardness) of the austempered parts. The bainitic region can be divided according to austempering temperature into 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) is softer and tougher, whereas the structure of lower bainite (needlelike, with small carbides under 608 within the needles) is harder and more brittle. 65 105 FIGURE 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.) ß 2006 by Taylor & Francis Group, LLC. TABLE 6.8 Section Sizes and Hardness Values of Austempered Parts of Various Steel Grades Section Size Steel nm 1050 1065 1066 1084 1086 1090 1090e 1095 1350 4063 4150 4365 5140 5160e 8750 50100 b 3 5c 7c 6c 13c 5c 20c 4c 16c 16c 13c 25c 3b 26c 3b 8c Salt Temperature in. Ms Temperaturea 8C b 0.125 0.187c 0.281c 0.218c 0.516c 0.187c 0.820c 0.148c 0.625c 0.625c 0.500c 1.000 0.125b 1.035c 0.125b 0.312c 8F 8C 8F Hardness (HRC) 345 —d —d —d —d —d 315f —d —d —d —d —d 345 315f 315 —d 655 —d —d —d —d —d 600f —d —d —d —d —d 655 600f 600 —d 320 275 260 200 215 — — 210g 235 245 285 210 330 255 285 — 610 525 500 395 420 — — 410g 450 475 545 410 630 490 545 — 41–47 53–56 53–66 55–58 55–58 57–60 44.5 (avg.) 57–60 53–56 53–56 52 max 54 max 43–48 46.7 (avg.) 47–48 57–60 a Calculated. Sheet thickness. c Diameter of section. d Salt temperature adjusted to give maximum hardness and 100% bainite. e Modified austempering; microstructure contained pearlite as well as bainite. f Salt with water additions. g Experimental value. b Source: 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 with austempering temperature, the temperature of the austempering bath must be kept within close tolerance (+68C (+438F)). The holding time at austempering temperature should be sufficient to allow complete transformation. 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 of treatment but it is not harmful to the mechanical properties of austempered parts. REFERENCES ¨ ¨ 1. G. Spur and T. Stoferle (Eds.), Handbuch der Fertigungstechnik, Vol. 4/2, Warmebehandeln, Carl Hanser, Munich, 1987. ¨ 2. H.J. Eckstein (Ed.), Technologie der Warmebehandlung von Stahl, 2nd ed., VEB Deutscher Verlag ¨ fur Grundstoffindustrie, Leipzig, 1987. ˇ´ ˇˇ ´ 3. B. Liscic, S. Svaic, and T. Filetin, Workshop designed system for quenching intensity evaluation and calculation of heat transfer data, Proceedings of the First International Conference on Quenching and Control of Distortion, Chicago, IL, September 1992, pp. 17–26. ß 2006 by Taylor & Francis Group, LLC. ¨ 4. A. Rose and W. Strassburg, Anwendung des Zeit-Temperatur-Umwandlungs-Schaubildes fur ¨ ¨ ¨ kontinuierliche Abkuhlung auf Fragen der Warmebehandlung, Archiv. Eisenhuttenwes. 24(11/ 12):505–514 (1953) (in German). ¨ 5. H.P. Hougardy, Die Darstellung des Umwandlungsverhaltens von Stahlen in den ZTU-Schaubil¨ dern, Harterei-Tech. Mitt. 33(2):63–70 (1978) (in German). ¨ 6. E.Scheil, Arch. Eisenhuttenwes. 8:565–567 (1934/1935) (in German). ¨ ¨ 7. F. Wever and A. Rose (Eds.), Atals zur Warmebehandlung der Stahle, Vols. I and II, Verlag ¨ Stahleisen, Dusseldorf, 1954/56/58. ¨ ¨ 8. W. Peter and H. Hassdenteufel, Aussagefahigkeit der Stirnabschreckprufung und der Zeit-Temper¨ ¨ ¨ atur-Umwandlungsschaubildes fur das Ergebnis der Hartung von Rundstaben, Stahl Eisen 87(8):455–457 (1967) (in German). 9. M. Atkins, Atlas of Continuous Transformation Diagrams for Engineering Steels, British Steel Corporation, BSC Billet, Bar and Rod Product, Sheffield, U.K., 1977. 10. E.A. Loria, Transformation behaviour on air cooling steel in A3–A1 temperature range, Met. Technol., 490–492 (1977). 11. N. Shimizu and I. Tamura, Effect of discontinuous change in cooling rate during continuous cooling on pearlitic transformation behaviour of steel, Trans. ISIJ 17:469–476 (1977). 12. ISI, Decarburization, ISI Publication 133, Gresham Press, Old Woking, Surrey, England, 1970. ˇˇ ´ 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 tool steels, Stahl Eisen. 77(18):1221–1233 (1957) (in German). ¨nning, Verfahrenstechnik des Isothermglu ¨hens, Harterei-Tech. Mitt. 32:43–49 (1977) (in German). ¨ 16. J. Wu 17. 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, Zeit¨ Temperatur-Austenitisierung-Schaubilder Verlag Stahleisen, Dusseldorf, 1973 (in German). ¨ ¨ 19. J. Orlich and H.J. Pietrzenivk (Eds.), Atlas zur Warmebehandlung der Stahle, Vol. 4, Zeit-Temper¨ atur-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 der ¨ Warmebehandlung von Stahl, Fertigungstech. Betr. 26(5):298–301 (1976) (in German). 21. M.A. Aronov, J.F. Wallace, and M.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 18–20, 1994, Schaumburg, IL, pp. 55–61. 22. N.I. Kobasko, Teplovie procesi pri zakalke stali, Mettaloved. Termiceskaja Obrab. Metalov. 3:2–6 (1968). ˇˇ ´ 23. B. Liscic and G.E. Totten, Controllable delayed quenching, Proceedings of the International Conference on Equipment and Processes, April 18–20, 1994, Schaumburg, IL, pp. 253–262. 24. J. Parrish, Retained austenite: new look at an old debate, Adv. Mater. Process. 3:25–28 (1994). ¨ ¨ 25. E. Just, Verguten-Werkstoffbeeinflussung durch Harten und Anlassen, VDI-Ber. 256:125–140 (1976) (in German). ¨glichkeiten der Optimierung der Auswahl vergu ¨nch, and A. Prewitz, Mo ¨tbarer 26. H.J. Spies, G. Mu ¨ ¨ ¨tbarkeit, Neue Hutte 8(22):443–445 (1977) (in German). ¨ Baustahle durch Berechnung der Hart-und Vergu ˇˇ ´ 27. B. Liscic and T. Filetin, Computer-aided determination of the process parameters for hardening and tempering structural steels, Heat Treat. Met. 3:62–66 (1987). ¨ ¨ 28. K.H. Illgner, Qualitats-und Kostenvorteile mit Zwischenstufenvergutungsanlagen im Vergleich zu ¨ ¨ normalen Vergutungsanlagen, Fachber. Huttenpraxis Metallweiterverarb. 17(4):281–288 (1979) (in German). ¨ 29. F.W. Eysell, Die Zwischenstufenvergutung und ihre betriebliche Anwendung, Z. TZ Prakt. Metallbearb. 66:94–99 (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. ß 2006 by Taylor & Francis Group, LLC. 7 Heat Treatment with Gaseous Atmospheres Johann Grosch CONTENTS 7.1 7.2 7.3 General Introduction ................................................................................................. 415 Fundamentals in Common ........................................................................................ 417 Carburizing ................................................................................................................ 422 7.3.1 Introduction..................................................................................................... 422 7.3.2 Carburizing and Decarburizing with Gases..................................................... 422 7.3.2.1 Gas Equilibria ................................................................................... 423 7.3.2.2 Kinetics of Carburizing ..................................................................... 426 7.3.2.3 Control of Carburizing...................................................................... 428 7.3.2.4 Carbonitriding................................................................................... 431 7.3.3 Hardenability and Microstructures ................................................................. 432 7.4 Reactions with Hydrogen and with Oxygen .............................................................. 440 7.5 Nitriding and Nitrocarburizing.................................................................................. 446 7.5.1 Introduction..................................................................................................... 446 7.5.2 Structural Data and Microstructures .............................................................. 448 7.5.2.1 Structural Data ................................................................................. 448 7.5.2.2 Microstructures of Nitrided Iron ...................................................... 450 7.5.2.3 Microstructures of Nitrided and Nitrocarburized Steels................... 452 7.5.2.4 Microstructural Specialties................................................................ 456 7.5.3 Nitriding and Nitrocarburizing Processes ....................................................... 457 7.5.3.1 Nitriding............................................................................................ 457 7.5.3.2 Nitrocarburizing................................................................................ 460 7.5.3.3 Processing Effects on the Nitriding and Nitrocarburizing Results ... 461 7.6 Properties of Carburized and Nitrided or Nitrocarburized Components .................. 463 References .......................................................................................................................... 469