Skip Nav Destination

## Abstract

The existence of austenite and ferrite, along with carbon alloying, is fundamental in the heat treatment of steel. In view of the importance of structure and its formation to heat treatment, this chapter describes the various microstructures that form in steels, the various factors that determine the formation of microstructures during heat treatment processing of steel, and some of the characteristic properties of each of the microstructures. The discussion also covers the constitution of iron during heat treatment and the phases of heat-treated steel with elaborated information on iron phase transformation, hysteresis in heating and cooling, ferrite and austenite as two crystal structures of solid iron, and the diffusion coefficient of carbon.

## Introduction

The responsiveness of steel from heat treatment is due to some important properties of iron and the metallurgical effects of carbon in iron. Fundamentally, all steels are mixtures or, more properly, alloys of iron with a small amount of carbon (along with varying amounts of other alloying elements such as manganese, chromium, nickel, and molybdenum). One important effect is the size of carbon atoms relative to that of iron atoms. The carbon atom is only 1/30 the size of the iron atom, and carbon atoms are sufficiently small to fit between the interstices of the larger iron atoms. Other atoms small enough to fit in the interstitial regions of solid iron are hydrogen, nitrogen, and boron. In general, interstitial atoms can easily diffuse—jumping from one interstitial site to another—unlike larger atoms (which can only jump by “substitution” into the vacancies within a crystal lattice). This, along with the effect of temperature on diffusion, makes the mobility of carbon responsive during solid-state heating.

The other important metallurgical phenomenon is the allotropy of iron, which means that the iron atoms can arrange themselves into more than one crystalline form or phase. At room temperature, for example, iron atoms arrange themselves into a body-centered cubic (bcc) crystal structure—called ferrite or alpha (α) iron. At higher temperatures, the iron atoms form a face-centered cubic (fcc) crystal structure—called austenite or gamma (γ) iron. The existence of these two phases, along with carbon alloying, is fundamental in the heat treatment of steel.

In addition to the allotropy of iron, the strengthening of steel by heat treatment also depends on the difference in solubility of carbon in ferrite and austenite. Solid solubility is a measure of how much solute can be dissolved (or incorporated) into the host lattice, as described in Chapter 1, “Structure of Metals and Alloys,” in this book. Iron ferrite and austenite have different spacing between iron atoms, and the atoms in austenite are more widely spaced than in ferrite (Fig. 1). This allows austenite to accommodate more carbon atoms in the interstitial regions of the crystal lattice. Carbon is dissolved in the octahedral interstitial sites between iron atoms in ferrite and austenite. The interstitial sites for carbon in ferrite are much smaller than those in austenite; therefore, the solubility of carbon in ferrite is significantly lower than in austenite. Temperature influences the extent of solubility, because higher temperatures expand the host lattice and thus provide a better opportunity for the solute to fit in the lattice. However, carbon is almost insoluble in alpha iron—ranging from only 0.008 wt% near room temperature to a maximum solubility limit of 0.02 wt% at 727 °C (1340 °F) (Fig. 2).

Fig. 1

Crystal structure and lattice spacing of iron atoms with (a) body-centered cubic and (b) face-centered cubic crystal structures. Source: Ref 1

Fig. 1

Crystal structure and lattice spacing of iron atoms with (a) body-centered cubic and (b) face-centered cubic crystal structures. Source: Ref 1

Fig. 2

Iron-rich side of iron-carbon equilibrium phase diagram, showing extent of ferrite phase field and decrease of carbon solubility with decreasing temperature. See also Fig. 27 in Chapter 1.

Fig. 2

Iron-rich side of iron-carbon equilibrium phase diagram, showing extent of ferrite phase field and decrease of carbon solubility with decreasing temperature. See also Fig. 27 in Chapter 1.

When the solubility of carbon in either austenite or ferrite is exceeded, not all the carbon atoms can be accommodated within the interstitial sites between the iron atoms. In this case, the excess carbon atoms may combine to form graphite or, more typical for steels, an iron-carbide compound (Fe3C) referred to as cementite or θ-carbide. Cementite has an orthorhombic crystal that can accommodate more carbon atoms in its crystal structure. For each atom of carbon in the compound, there are three atoms of iron, giving an atomic composition of 25 at. % C. The corresponding weight percent carbon in cementite turns out to be 6.7.

The orthorhombic lattice of cementite is a bit more complex than either the bcc structure of ferrite or the fcc structure of austenite. Cementite also is not completely stable, because it ultimately decomposes into carbon (graphite) over time (dashed line in Fig. 2). However, cementite is sufficiently stable to be considered as a near-equilibrium phase (solid line in Fig. 2) that occurs when carbon levels exceed the solubility limit in iron. The morphology and distribution of cementite also can be manipulated by heat treatment. The compound cementite has higher strength and lower ductility than either ferrite or austenite and, depending on its morphology and distribution, contributes in a variety of ways to the strengthening, deformation, and fracture of steels.

Cementite is very hard, ranging in hardness from 800 to 1400 HV, depending on the substitution of elements for iron. Heat treatment also can alter the amount, shape, and distribution of the hard cementite particles in the microstructure of steel. For example, the hard cementite phase can be dissolved into a single iron-carbon phase when the iron phase becomes austenitic. This process, known as austenitizing, dissolves cementite, because carbon is much more soluble in austenite, with a maximum solubility of approximately 2 wt% at 1150 °C (2100 °F). This is 2 orders of magnitude greater than the maximum solubility of carbon in alpha (α) ferrite. Thus, austenitizing is often used as the starting point to create a single-phase solid solution. Then, by cooling from the austenite region, a two-phase formation of ferrite and cementite can be controlled for strengthening. This allows plain carbon steel to obtain a wide range of properties after heat treatment.

Carbon also has two other important effects on the constitution of iron during heat treatment. Carbon lowers the temperature for complete austenitization (Fig. 3). This allows for complete dissolution of cementite at lower temperatures and the subsequent manipulation of carbide formation during cooling. Moreover, when austenitized steel is rapidly cooled (quenched), it produces a different mechanism of phase transformation. During rapid quenching from austenite to ferrite, there is not enough time for the excess carbon atoms to diffuse and form cementite along with the bcc ferrite. Therefore, some (or all) of the carbon atoms become trapped in the ferrite lattice, causing the composition to rise well above the 0.02% solubility limit of carbon in ferrite. This causes lattice distortion, so much so that the distorted bcc lattice rapidly transforms into a new metastable phase called martensite. Martensite does not appear as a phase on the iron-carbon equilibrium phase diagram because it is a metastable (nonequilibrium) phase that occurs from rapid cooling.

Fig. 3

Eutectoid region of the iron-carbon phase diagram and critical transformation temperature for cementite formation from austenite

Fig. 3

Eutectoid region of the iron-carbon phase diagram and critical transformation temperature for cementite formation from austenite

The unit cell of the martensite crystal is a body-centered tetragonal (bct) crystal structure, which is similar to the bcc unit cell, except that one of its edges (called the c-axis) is longer than the other two axes (Fig. 4c). The distorted form of the bct is a supersaturated condition that accommodates the excess carbon. The bct structure also occupies a larger atomic volume than ferrite and austenite, as summarized in Table 1 for different microstructural components as a function of carbon content. The density of martensite thus is lower than ferrite (and also austenite, which is denser than ferrite). The resulting expansion gives martensite its high hardness and is the basis for strengthening steels by heat treatment. Figure 4 also is an illustration of an interstitial carbon atom placed in an austenitic and martensitic cell that helps visualize how the crystal can accommodate supersaturated carbon from rapid cooling. In addition, even though austenite is denser than ferrite, there are fewer interstitial regions in ferrite than can accommodate interstitial carbon.

Table 1
Atomic volumes of selected microstructural constituents of ferrous alloys
PhaseApparent atomic volume, Å3
Ferrite11.789
Cementite12.769
Ferrite + carbides11.786 + 0.163 C(a)
Pearlite11.916
Austenite11.401 + 0.329 C(a)
Martensite11.789 + 0.370 C(a)
PhaseApparent atomic volume, Å3
Ferrite11.789
Cementite12.769
Ferrite + carbides11.786 + 0.163 C(a)
Pearlite11.916
Austenite11.401 + 0.329 C(a)
Martensite11.789 + 0.370 C(a)
(a)

Carbon percent.

Source: Ref 2

Fig. 4

Structure of (a) ferrite (body-centered cubic), austenite (face-centered cubic), and (c) martensite (body-centered tetragonal)

Fig. 4

Structure of (a) ferrite (body-centered cubic), austenite (face-centered cubic), and (c) martensite (body-centered tetragonal)

## Constitution of Iron

The atoms in solids typically arrange themselves into a unique crystal structure under equilibrium conditions, but some elements are allotropic. That is, their structure transforms from one crystal structure to another with changes in temperature and pressure, where each crystal structure is a distinctively separate solid-state phase. For example, both iron and carbon have a number of allotropic forms. Carbon can exist as diamond, soot, graphite, and the more recently discovered form of fullerenes. However, the allotropy of carbon is not a significant variable as an alloying element in iron. In contrast, the allotropy of iron is of fundamental importance in the heat treating of steel. Iron is an allotropic element that changes its structure at several temperatures known as transformation temperatures (Fig. 5).

Fig. 5

Equilibrium transformation temperatures of pure iron. fcc, face-centered cubic; bcc, body-centered cubic

Fig. 5

Equilibrium transformation temperatures of pure iron. fcc, face-centered cubic; bcc, body-centered cubic

### Iron Phase Transformation

The process by which iron (or any material) changes from one atomic arrangement to another when heated or cooled is called a phase transformation. Figure 5 illustrates changes in the phases of pure iron during very slow (near-equilibrium) heating or cooling. During a phase transformation, the temperature stays constant during heating (or cooling) until the phase transformation of iron is complete. This is the same behavior as the temperature plateau during the phase changes of pure metal during melting or solidification. The so-called critical temperatures of the iron phase transformation are assigned the letter A, derived from the French word arrêt, which stands for the arrest in temperature during heating or cooling through the transformation temperature. The letter A also is followed by either the letter c or r to indicate transformation by either heating or cooling, respectively. The use of the letter c for heating is derived from the French word chauffant, meaning warming. If cooling conditions apply, the critical temperature is designated as Ar, with the letter r being derived from the French word refroidissant for cooling.

Below the melting point of 1540 °C (2800 °F), there are three temperature plateaus when solid iron undergoes a phase change. Consider first the process of solidification as liquid iron cools from its melting point of 1540 °C (2800 °F). It begins to freeze, with no further drop in temperature until it transforms itself completely into a solid form of iron referred to as delta iron or delta ferrite. Ferrite has a bcc crystal structure. Delta ferrite is the high-temperature bcc phase of iron. After solidification is complete, the temperature drops at a uniform rate until the temperature of 1394 °C (2541 °F) is reached. This temperature marks the beginning of a transformation of the bcc delta iron into an fcc crystal phase, called austenite or gamma iron. The temperature stays constant until the transformation is complete, that is, until all of the iron has an austenitic (fcc) phase structure.

Further cooling of the gamma (fcc) iron continues at a uniform rate until the temperature reaches 912 °C (1674 °F). This is the transformation temperature when gamma iron begins the transformation into a nonmagnetic form of iron with a bcc crystal lattice. The temperature holds steady during cooling until all the iron atoms are completely transformed into a bcc crystal lattice. Again, this low-temperature bcc phase of iron is referred to as alpha iron or alpha ferrite. Finally, a similar cooling plateau occurs at 769 °C (1416 °F), which is the transformation temperature when the nonmagnetic form of alpha iron changes into a magnetic form of alpha iron. This is the Curie temperature. In ferromagnetic materials that are below their Curie temperature, the magnetic moments of adjacent atoms are parallel to each other, such that all of the individual magnetic moments are aligned in one direction.

These phase changes, which can be done very slowly, are called equilibrium transformations, meaning that sufficient time is needed for the metal to reach equilibrium for a given phase change. Enough energy, referred to as latent heat, must be added or released to complete an equilibrium transformation. Experimental latent-heat values at the transformation temperatures of pure iron are given in Table 2 for the phase transformations. In the case of the Curie temperature (TC), the additional energy needed to disorient the magnetic dipoles in iron is described by a sharp increase in specific heat (Fig. 6).

Table 2
Latent heats of phase transformations for pure iron
TransformationTemperature, KTemperature, °CLatent heat, kJ/kg
Alpha ferrite (α) to austenite (γ)118591216
Austenite (γ) to delta ferrite (δ)1667139415
Fusion (liquid to solid)18811538247 ± 7
TransformationTemperature, KTemperature, °CLatent heat, kJ/kg
Alpha ferrite (α) to austenite (γ)118591216
Austenite (γ) to delta ferrite (δ)1667139415
Fusion (liquid to solid)18811538247 ± 7

Source: Ref 3

Fig. 6

Specific heat of iron from 0 to 3200 K. TC is the Curie temperature. Source: Ref 3

Fig. 6

Specific heat of iron from 0 to 3200 K. TC is the Curie temperature. Source: Ref 3

During an equilibrium transformation, the temperature remains constant until the phase change is complete for the entire material. This temperature plateau or arrest is roughly indicated by the small steps at the transformation temperatures indicated in Fig. 6. All equilibrium transformations are based on the movement of atoms by diffusion, which occurs by pronounced thermal agitation of atoms or molecules. Thus, all equilibrium transformations are classified as thermal or diffusive (reconstructive) transformations, because phase growth or decomposition is activated by the thermal (kinetic) energy of the atoms in the solid. Slow changes under near-equilibrium conditions are also reversible; the very same changes can take place in reverse order. That is, when iron or steel is subjected to slow heating from room temperature, alpha iron first becomes nonmagnetic alpha iron and then becomes gamma iron on further heating.

#### Hysteresis in Heating and Cooling

Under conditions of very slow heating or cooling under near-equilibrium conditions, the transformation temperatures for heating and cooling are the same. However, heating rates in commercial practice usually exceed those in controlled laboratory experiments, and a higher rate of heating or cooling can change the transformation temperature. For example, when the heating rate is high, the Ac transformation temperatures will be higher than those in Fig. 6. Likewise, on slow cooling in commercial practice, transformation changes occur at temperatures a few degrees below the Ar transformation temperatures in Fig. 6. Also, the faster the heating or cooling rate, the greater the gap between the Ac and Ar points. Going one step further, transformation temperatures during cooling can be suppressed several hundreds of degrees by rapid cooling or quenching. The rate of cooling and heating can be a key factor in many heat treatments.

#### Ferrite and Austenite in Pure Iron

Ferrite and austenite are the two crystal structures of solid iron under equilibrium conditions. Ferrite has atoms at each of the four corners of the unit cell (Fig. 1, 4) with one atom in the center. The edge length (or lattice parameter) of alpha ferrite is approximately 2.87 Å at 20 °C (70 °F) and increases to approximately 2.9 Å at 910 °C (1670 °F). In contrast, the lattice parameter of the austenitic unit cell is on the order of approximately 3.57 Å at the transformation temperature of 912 °C (1674 °F). This provides greater interatomic space for the greater solubility of carbon in austenite, as compared to ferrite.

An fcc crystal also is a close-packed structure, which means that the atoms are packed together with a minimum total volume. Therefore, austenite can pack more atoms into a given volume than ferrite. One unit cell of the bcc structure consists of two complete atoms, calculated from the one atom in the center of the cell, plus the four corners with one-quarter of each corner atom within the cube of the unit cell. Like ferrite, austenite also has atoms at the four corners of the unit cell. However, the fcc lattice also has six additional atoms of each face of the unit cell (Fig. 1, 4), where one-half of each face-centered atom is within the cube of the unit cell. Thus, the fcc unit cell is equivalent to four complete atoms (1/2 of each atom on the six faces, plus 1/4 of each atom on the four corners). The packing results in a higher density of fcc compared to that of a bcc lattice at high temperatures (Table 3). A plot of volume per atoms also indicates a sharp contraction when alpha ferrite transforms into austenite (Fig. 7).

Table 3
Density of austenite, alpha ferrite, and delta ferrite at selected temperatures
PhaseTemperatureDensity, g/cm3
°C°F
Alpha ferrite20687.870
Alpha ferrite91016707.47
Austenite91216737.694
Austenite139025347.66
Delta ferrite139425417.406
PhaseTemperatureDensity, g/cm3
°C°F
Alpha ferrite20687.870
Alpha ferrite91016707.47
Austenite91216737.694
Austenite139025347.66
Delta ferrite139425417.406
Fig. 7

Volume per atom for iron. Source: Ref 4

Fig. 7

Volume per atom for iron. Source: Ref 4

#### Diffusion Coefficient of Carbon in Iron

As noted, carbon readily diffuses as an interstitial atom. The activation energy for diffusion of carbon in iron is small (Table 4), and the diffusion coefficient is larger than that of typical substitutional elements (Fig. 8), where the diffusion coefficient (D) is a function of temperature according to an Arrhenius equation, such that:
$D=D0exp⁡[−QRT]$
where D0 is the frequency factor (in units of cm2/s), Q is the activation energy (kJ/mol), T is absolute temperature (K), and R is the gas constant (8.31 J/mol · K).
Table 4
Activation energies for diffusion of selected elements in iron
Diffusing elementDiffusing throughDiffusion activation energy (Q), cal/molDiffusion frequency factor (D0), cm2/s
CarbonFerrite (α iron)18,1000.0079
CarbonAustenite (γ iron)33,8000.21
NickelAustenite (γ iron)66,0000.5
ManganeseAustenite (γ iron)67,0000.35
ChromiumFerrite (α iron)82,00030,000
ChromiumAustenite (γ iron)97,00018,000
Diffusing elementDiffusing throughDiffusion activation energy (Q), cal/molDiffusion frequency factor (D0), cm2/s
CarbonFerrite (α iron)18,1000.0079
CarbonAustenite (γ iron)33,8000.21
NickelAustenite (γ iron)66,0000.5
ManganeseAustenite (γ iron)67,0000.35
ChromiumFerrite (α iron)82,00030,000
ChromiumAustenite (γ iron)97,00018,000

Source: Ref 1

Fig. 8

Diffusion coefficients (D) of interstitial elements (hydrogen, carbon, nitrogen) compared with substitutional elements in alpha iron. Adapted from: Ref 5

Fig. 8

Diffusion coefficients (D) of interstitial elements (hydrogen, carbon, nitrogen) compared with substitutional elements in alpha iron. Adapted from: Ref 5

Typical values for D0 are given in Table 5, and the temperature dependence of D is shown for a number of material systems in Fig. 9. The change of the diffusion coefficient of carbon as the concentration of carbon changes in iron at 930 °C (1700 °F) is shown in Fig. 10. The activation energy, Q, reflects the energy required to move an atom over a barrier from one lattice site to another; the barrier is associated with the requirement that the atom must vibrate with a sufficient amplitude to break the nearest neighboring bonds to move to a new location.

Table 5
Representative data for diffusion of carbon in ferrite and austenite
Diffusing speciesSolvent metalD0, m2/sActivation energy, kJ/molCalculated values
TemperatureD, m2/s
°C°F
FeαFe (body-centered cubic)2.8 × 10–42515009303.0 × 10–21
90016501.8 × 10–15
FeγFe (face-centered cubic)5.0 × 10–528490016501.1 × 10–17
110020107.7 × 10–16
CαFe6.2 × 10–7805009302.4 × 10–12
90016501.7 × 10–10
CγFe2.3 × 10–514890016505.9 × 10–12
110020105.3 × 10–11
Diffusing speciesSolvent metalD0, m2/sActivation energy, kJ/molCalculated values
TemperatureD, m2/s
°C°F
FeαFe (body-centered cubic)2.8 × 10–42515009303.0 × 10–21
90016501.8 × 10–15
FeγFe (face-centered cubic)5.0 × 10–528490016501.1 × 10–17
110020107.7 × 10–16
CαFe6.2 × 10–7805009302.4 × 10–12
90016501.7 × 10–10
CγFe2.3 × 10–514890016505.9 × 10–12
110020105.3 × 10–11

Source: Ref 6

Fig. 9

Arrhenius plot of diffusivity of various metal systems. bcc = body-centered cubic; hcp = hexagonal close-packed; fcc = face-centered cubic. Adapted from: Ref 7

Fig. 9

Arrhenius plot of diffusivity of various metal systems. bcc = body-centered cubic; hcp = hexagonal close-packed; fcc = face-centered cubic. Adapted from: Ref 7

Fig. 10

Variation of diffusion coefficient with carbon concentration. Source: Ref 8

Fig. 10

Variation of diffusion coefficient with carbon concentration. Source: Ref 8

## Phases of Heat-Treated Steel

The heat treatment of steel is based on the physical metallurgical principles that relate to processing, properties, and structure (Ref 9). In heat treatment, the processing is most often entirely thermal and modifies only structure. Thermomechanical treatments, which modify component shape and structure, also are important processing approaches that fall into the domain of heat treatment. Scientific principles link the processing parameters to structure and properties and are increasingly necessary for proper application of the equipment and instrumentation now available for control of heat treatment processes. Examples of scientific efforts that directly support the technology of heat treatment include characterization of mechanisms of phase transformations that produce desired structures and properties of heat treated parts; determination of phase transformation and annealing kinetics that establish processing times, temperatures, and cooling rates for heat treatments; and evaluation of mechanisms of deformation and fracture of the structures produced by heat treatment.

In view of the importance of structure and its formation to heat treatment, the purpose of this section is to describe the various microstructures that form in steels, the various factors that determine the formation of microstructures during heat treatment processing of steel, and some of the characteristic properties of each of the microstructures. Structure-sensitive properties such as strength, ductility, and toughness establish the ease of manufacturing, service performance, and limitations to service conditions of heat-treated steels. The descriptions of the microstructures and principles presented here should be considered only introductory.

### Iron-Carbon Phase Diagram

The microstructures that result from heat treatment of steel are composed of one or more phases in which the atoms of iron, carbon, and other elements in steel are associated. Phase diagrams assume equilibrium, that is, that the carbon and iron have had sufficient time to distribute themselves in the various phases as shown. Sometimes, equilibrium is difficult to achieve, especially in steels that contain elements that diffuse only sluggishly, and, in fact, certain heat treatments such as hardening are designed to prevent formation of equilibrium structures. Thus, the fact that equilibrium may not be achieved, together with the shift of the phase-field boundaries by alloying elements, place limitations on the direct use of the iron-carbon phase diagram.

As noted in Chapter 1, steels and cast irons contain, in addition to iron and carbon, many other elements that shift the boundaries of the phase fields in the iron-carbon diagram. Some alloying elements such as manganese and nickel are austenite stabilizers and extend the temperature range over which austenite is stable. Elements such as chromium and molybdenum are ferrite stabilizers and restrict the ranges of austenite stability. Therefore, care must be taken in the direct use of the iron-carbon diagram to predict phase relationships in commercial alloys that contain elements in addition to iron and carbon. Nevertheless, the iron-carbon diagram is an important reference for understanding the relationships between the structure and heat treatment of steels, and, subject to the aforementioned limitations, it is used in this chapter to illustrate the basis for microstructural formation in steels as well as iron-carbon alloys.

Figure 2 and 3 show two portions of the iron-carbon phase diagram (see also Fig. 28 in Chapter 1, “Structure of Metals and Alloys,” in this book). As noted, the temperatures at the boundaries of the various phase fields are frequently referred to as critical temperatures. Because the critical temperatures are often identified by changes in slope or thermal “arrests” in heating and cooling curves, they are given the designation A. If equilibrium conditions are applicable, the designations Ae1, Ae3, and Aeem, or simply A1, A3, and Acm, are used, as shown in Fig. 2 and 3 for different phase boundaries.

If heating conditions (which raise the critical temperatures relative to equilibrium) apply, Ac1, Ac2, and Accm are used; c is derived from the French word chauffant. If cooling conditions (which lower the critical temperature relative to equilibrium) apply, the designations Ar1, Ar3, and Arcm are used; r is derived from the French word refroidissant. There is hysteresis in the transformation temperatures because continuous heating and cooling leave insufficient time to accomplish the diffusion-controlled phase transformations at the true equilibrium temperatures.

The symbols used to designate the different critical temperatures are summarized in Table 6. The A1 transformation temperature (whether Ae1, Ac1, or Ar1) is referred to as the lower critical temperature, while the A3 transformation temperature (whether Ae3, Ac3, or Ar3) is referred to as the upper critical temperature. As noted, carbon lowers the A3 transformation temperature.

Table 6
Definitions of critical transformation temperatures in steel
 Ae1 The critical temperature when some austenite begins to form under conditions of thermal equilibrium (i.e., constant temperature) Ac1 The critical temperature when some austenite begins to form during heating, with c derived from the French chauffant Ar1 The temperature when all austenite has decomposed into ferrite or a ferrite-cementite mix during cooling, with r derived from the French refroidissant Ae2 Curie temperature (770 °C) for high-temperature demagnetization of ferrite Ae3 The upper critical temperatures when all the ferrite phase has completely transformed into austenite under equilibrium conditions Ac3 The temperature at which transformation of ferrite to austenite is completed during heating Ar3 The upper critical temperatures when a fully austenitic microstructure begins to transform to ferrite during cooling Aeem In hypereutectoid steel, the critical temperature under equilibrium conditions between the phase region of an austenite-carbon solid solution and the two-phase region of austenite with some cementite (Fe3C) Accm In hypereutectoid steel, the temperature during heating when all cementite decomposes and all the carbon is dissolved in the austenitic lattice Arcm In hypereutectoid steel, the temperature when cementite begins to form (precipitate) during cooling of an austenite-carbon solid solution Arr The temperature at which delta ferrite transforms to austenite during cooling Ms The temperature at which transformation of austenite to martensite starts during cooling Mr The temperature at which martensite formation finishes during cooling
 Ae1 The critical temperature when some austenite begins to form under conditions of thermal equilibrium (i.e., constant temperature) Ac1 The critical temperature when some austenite begins to form during heating, with c derived from the French chauffant Ar1 The temperature when all austenite has decomposed into ferrite or a ferrite-cementite mix during cooling, with r derived from the French refroidissant Ae2 Curie temperature (770 °C) for high-temperature demagnetization of ferrite Ae3 The upper critical temperatures when all the ferrite phase has completely transformed into austenite under equilibrium conditions Ac3 The temperature at which transformation of ferrite to austenite is completed during heating Ar3 The upper critical temperatures when a fully austenitic microstructure begins to transform to ferrite during cooling Aeem In hypereutectoid steel, the critical temperature under equilibrium conditions between the phase region of an austenite-carbon solid solution and the two-phase region of austenite with some cementite (Fe3C) Accm In hypereutectoid steel, the temperature during heating when all cementite decomposes and all the carbon is dissolved in the austenitic lattice Arcm In hypereutectoid steel, the temperature when cementite begins to form (precipitate) during cooling of an austenite-carbon solid solution Arr The temperature at which delta ferrite transforms to austenite during cooling Ms The temperature at which transformation of austenite to martensite starts during cooling Mr The temperature at which martensite formation finishes during cooling

Note: All of these changes, except the formation of martensite, occur at lower temperatures during cooling than during heating and depend on the rate of change of temperature.

### Austenite

As described in Chapter 1, austenite or gamma iron has an fcc crystal form that is stable at high temperatures with a maximum carbon solubility limit of just over 2 wt% (Fig. 3). The single-phase austenite field dominates the iron-carbon diagram at high temperatures. In all low-alloy steels, therefore, it is possible to produce a single-phase austenite microstructure. This characteristic is important during heat treatment, as cooling from the single-phase austenite field makes possible a wide variety of heat treatments based on transformation of the austenite, as described further in this and subsequent chapters.

The austenite grain size has an important influence on the final microstructure and properties that depend on the application. Coarse austenite grains increase hardenability, although in practice, alloying is used to control hardenability, because finer austenite is usually preferred to refine the final microstructure, given that it enhances strength and toughness. Austenite grain size is expected to increase with time or temperature, and alloying elements in solution or in the form of precipitates may retard grain growth by means of solute drag or precipitate pinning effects. The pinning precipitates in many steels are aluminum nitrides (AlN), which provide effective grain refinement at low temperatures in the austenite phase field during heat treatments such as normalizing and carburizing. Aluminum is added to many steels to remove oxygen from the liquid prior to casting and combines with solute nitrogen to form aluminum nitrides during cooling or reheating. The aluminum fine-grained practice is therefore available to suppress grain growth in many steels during subsequent reaustenitizing.

Because the austenite grain size is important, it is often desirable to measure the prior austenite grain size. The term prior austenite grain size refers to the austenite grains that existed at high temperature and are no longer present, having transformed to a different microstructure at room temperature. Evidence for the location of the prior austenite grain boundaries can be readily obtained in some microstructures using careful metallographic techniques (and specialized etchants). Some structures are more evident than others. For example, prior austenite grains are more apparent in fully martensitic microstructures, because the martensite plates or packets grow within a single austenite grain and do not cross the austenite boundaries. In low-carbon ferritic microstructures, it can be very difficult to bring out the prior austenite microstructure.

The austenitic phase also offers good ductility due to the available slip system in fcc structures (see Chapter 1). The stable austenitic phase at high temperatures enables hot working of even low-alloy steels. The single-phase austenite, without the obstacles that second phases present to dislocation motion and without the sites for fracture initiation that second-phase particles offer, deforms and recrystallizes readily so that substantial reductions in section size by hot rolling or forging may be accomplished. Traditionally, hot deformation is performed in the upper temperature range of the austenite field. Hot deformation of austenite at lower temperatures, or even in the two-phase ferrite-austenite field (controlled rolling), is used in the thermomechanical processing of steels along with the addition of small amounts of alloying elements (microalloying) such as niobium and vanadium, which precipitate as fine alloy carbonitrides at low temperatures. The low-temperature deformation and/or precipitation retard or prevent austenite recrystallization and grain growth and therefore produce finer austenite grains and that subsequently transform into finer transformation products during cooling after hot deformation.

Some highly alloyed steels with nickel or manganese also have stable austenitic phases at low temperatures, a condition that provides good ductility and toughness even at cryogenic temperatures. Ferrite or alpha iron, however, with a bcc structure is highly formable at room temperature, but dislocation motion in the bcc structure becomes severely restricted at low temperatures. This well-known effect of all bcc metals is described by the ductile-to-brittle transition temperature, which refers to the region in which the toughness drops, and the fracture mechanisms and features change from ductile to brittle.

Ferrite or alpha iron has a bcc form or phase of iron that is stable at low temperatures. Microstructures in low-carbon steels that consist largely of polycrystalline ferrite are highly formable at room temperature; dislocations move readily on the many slip systems of the bcc structure (see Chapter 1). However, at low temperatures, dislocation motion in the bcc structure is severely restricted. As a result, ferrite grains fracture in a brittle manner with little plastic deformation at low temperatures.

### Ferrite

Ferrite in steels is a solid solution of iron containing some interstitial carbon and (depending on the steel) other elements (such as silicon, chromium, manganese, molybdenum, and/or nickel) that replace the iron atoms in the bcc solid solution. Interstitial alloying elements also have a much stronger influence on the properties of iron than substitutional alloying elements. Even a small addition of carbon, less than its solubility limit, in pure iron substantially increases the room-temperature yield strength of the material (Fig. 11). On the other hand, the substitutional solid-solution elements silicon, copper, manganese, molybdenum, nickel, aluminum, and chromium are far less effective as ferrite strengtheners (Fig. 12). The strong effects of the interstitial elements carbon and nitrogen, as well as phosphorus, are clearly evident. The strength of ferritic steels also is determined by grain size—that is, a finer grain size results in a higher yield strength. Thus, control of grain size through thermomechanical treatments, heat treatments, and/or alloying is vital to the control of strength and toughness of most steels.

Fig. 11

Effect of small carbon additions on room-temperature yield strength of iron. Source: Ref 10

Fig. 11

Effect of small carbon additions on room-temperature yield strength of iron. Source: Ref 10

Fig. 12

Effect of alloying elements on yield strength. Source: Ref 10

Fig. 12

Effect of alloying elements on yield strength. Source: Ref 10

A wide variety of steels use the properties of ferrite, but only a few commercial steels are completely ferritic. Interstitial carbon diffuses easily through the bcc lattice, and ferritic steels, even low-carbon steels, typically have some cementite (Fe3C). Depending on carbon content and cooling rate, several different microstructures of ferrite and cementite can occur (see also the “Proeutectoid Ferrite and Cementite” section in this chapter). In low-carbon steels, the morphology of ferrite can vary from equiaxed grains to needlelike or acicular forms. For example, Fig. 13 (Ref 11) illustrates the effect cooling rates on microstructures with increasing cooling rates during continuous cooling of an Fe-0.01%C alloy. Continuous cooling arrest temperatures are clearly tied to various microstructures and transformation curves. When a critical cooling rate is attained for a given product of austenite decomposition, the formation of that product is suppressed, and transformation shifts to another product requiring transformation mechanisms with reduced dependence on diffusion.

Fig. 13

Transformation start temperatures and associated transformation curves as a function of cooling rate for various austenite transformation products in an Fe-0.01%C alloy. Source: Ref 11

Fig. 13

Transformation start temperatures and associated transformation curves as a function of cooling rate for various austenite transformation products in an Fe-0.01%C alloy. Source: Ref 11

If, for some reason, cooling is too rapid for cementite formation, carbon also can be trapped in the interstitial sites and contributes to various aging phenomena unique to ferritic steels (Ref 11, 12). One process is associated with segregation of carbon atoms to dislocations and grain boundaries and is referred to as strain aging. The other process is associated with precipitation of fine carbide particles either on dislocations or in the ferrite matrix and is referred to as quench aging (Ref 12). Figure 14 shows an example of fine dendritic cementite particles that have formed by quench aging on dislocations in the ferrite of a low-carbon steel. Both strain aging and quench aging effectively pin dislocations and are responsible for the discontinuous yielding of low-carbon steels with largely ferritic microstructures.

Fig. 14

Transmission electron micrograph showing cementite precipitated on dislocations in 0.08C-0.63Mn steel aged 115 h at 97 °C (207 °F). Courtesy of J.E. Indacochea

Fig. 14

Transmission electron micrograph showing cementite precipitated on dislocations in 0.08C-0.63Mn steel aged 115 h at 97 °C (207 °F). Courtesy of J.E. Indacochea

### Pearlite and Bainite

The eutectoid region of the iron-carbon equilibrium diagram (Fig. 3) is at 0.77 wt% C, where three equilibrium phases (austenite with dissolved carbon, ferrite with dissolved carbon, and cementite) can coexist at a temperature of 727 °C (1340 °F). If austenite, in an iron-carbon alloy containing 0.77 wt% C, is cooled below 727 °C (1340 °F), then it must transform to ferrite and cementite. This transformation results in a microstructure referred to as pearlite, where a unique parallel array of ferrite and cementite lamellae develop (Fig. 15). Many other alloys have eutectoids with similar lamellar microstructures of their particular phase constituents, but the iron-carbon eutectoid is the most widely recognized. Note that pearlite is not the only morphology of eutectoid decomposition. With rapid cooling rates, the ferrite-cementite formation can result in lath (or platelike) morphologies called bainite.

Fig. 15

Pearlite colonies of a plain carbon UNS G10800 steel showing colonies of pearlite. 4% picral etch. Original magnification: 200×. The cementite appears dark and the ferrite light. Source: Ref 13

Fig. 15

Pearlite colonies of a plain carbon UNS G10800 steel showing colonies of pearlite. 4% picral etch. Original magnification: 200×. The cementite appears dark and the ferrite light. Source: Ref 13

Pearlite in a eutectoid steel is nucleated at austenite grain boundaries and grows as spherical-shaped colonies or nodules into the austenite (Fig. 16). The pearlite nodules nucleate at prior austenite grain boundaries and triple points to minimize the free energy needed for the transformation. Each nodule contains subunits (colonies) of cementite and ferrite lamellae; each colony has a specific orientation in relation to the parent austenite grain (see Ref 14 for more details). Pearlite formation is heavily dependent on the long-range diffusion of carbon into the growing cementite lamellae of the pearlite. Also, iron atoms must rearrange themselves by short-range diffusion from the fcc structure of austenite to their arrangements in the crystal structures of ferrite and cementite at the interface of the growing pearlite colonies.

Fig. 16

Relationship of pearlite lamellae, colonies, and nodules to prior austenite grains. Source: Ref 14

Fig. 16

Relationship of pearlite lamellae, colonies, and nodules to prior austenite grains. Source: Ref 14

The rate of transport of carbon and iron atoms is temperature dependent and increases exponentially with increasing temperature. At temperatures just below the 727 °C (1340 °F) eutectoid temperature (Fig. 3), the thermodynamic driving force (i.e., the decrease in free energy per unit volume when austenite is replaced by pearlite) is low in order to offset the increase in energy in the eutectoid reactions associated with pearlite colony and austenite interfaces and the ferrite-cementite interfaces within the pearlite. As a result, the nucleation rate of colonies is low, and the spacing of cementite lamellae within the colonies is large. The coarse interlamellar spacing increases the diffusion distance for carbon and causes a low rate of growth for those colonies that manage to nucleate. Thus, pearlite transformation at temperatures close to the eutectoid temperatures is sluggish, and the pearlite microstructure that forms is relatively coarse. With increased undercooling, the thermodynamic driving force increases, the nucleation rate of pearlite colonies increases, interlamellar spacings decrease, and the growth rate of colonies increases. As a result of the latter changes, the transformation of austenite to pearlite accelerates with decreasing temperature.

The increase in nucleation rate with greater undercooling below the eutectoid temperature results in the classic C-shaped curves of an isothermal transformation diagram, where a steel is quenched from austenite and held at given temperature below the eutectoid temperature for a set period of time, and then quenched again to room temperature. For example, Fig. 17 shows an isothermal transformation diagram for a eutectoid steel. The diagram shows the beginning and end of the eutectoid transformation of austenite to pearlite for specimens cooled from the single-phase austenite field and held isothermally at temperatures between A1 and 540 °C (1000 °F). The acceleration of the transformation with decreasing temperature is apparent.

Fig. 17

Isothermal transformation diagram for 1080 steel containing 0.79% C and 0.76% Mn. Austenitized at 900 °C (1650 °F); ASTM grain size No. 6. Source: Ref 15

Fig. 17

Isothermal transformation diagram for 1080 steel containing 0.79% C and 0.76% Mn. Austenitized at 900 °C (1650 °F); ASTM grain size No. 6. Source: Ref 15

At temperatures below 540 °C (1000 °F), the diffusion of iron atoms is reduced to the extent that they can no longer be readily transferred even the very short distance across the pearliteaustenite interface. Therefore, the mechanism for the change in crystal structure from austenite to ferrite changes from diffusion to shear. Instead of an atom-by-atom transfer across an interface, large numbers of iron atoms shear or move cooperatively to form lath- or plate-shaped crystals of ferrite. Carbon diffusion and cementite formation must still occur because of the low solubility of carbon in the bcc ferrite, but the cementite forms as separate particles rather than as continuous lamellae, as in pearlite. The microstructure produced by both shear and diffusion is termed bainite, after Edgar C. Bain, who did much pioneering work in the characterization of austenite transformation and hardenability of steels (Ref 16).

Two forms of bainite develop in steels. One is termed upper bainite because it forms at relatively high temperatures, just below the range of pearlite formation. Upper bainite forms in patches containing many parallel laths of ferrite. Carbon is rejected from the ferrite and concentrates to form relatively coarse cementite particles between the ferrite laths. Figures 18(a) and 18(c) are light optical micrographs of upper bainite in two steels. Figure 19(a) is a micrograph from 4150 steel with patches of upper bainite formed by partial transformation of the austenite at 460 °C (860 °F). The austenite that did not transform at 460 °C (860 °F) formed martensite (light background phase) on quenching to room temperature. The general morphology of upper bainite is shown in Fig. 19(a), but the ferrite laths and cementite particles are too fine to be resolvable in the light micrograph.

Fig. 18

Microstructure of (a) upper bainite and (b) lower bainite in a Cr-Mo-V rotor steel. 2% nital + 4% picral etch. Original magnification: 500×. (c) S5 tool steel austenitized, isothermally transformed (partially) at 540 °C (1000 °F) for 8 h, and water quenched to form upper bainite (dark); balance of austenite formed martensite. 4% picral + 2% nital. Original magnification: 1000×. (d) S5 tool steel austenitized, isothermally transformed at 400 °C (750 °F) for 1 h, and air cooled to form lower bainite. 37 to 38 HRC. 4% picral + 2% nital. Original magnification: 1000×

Fig. 18

Microstructure of (a) upper bainite and (b) lower bainite in a Cr-Mo-V rotor steel. 2% nital + 4% picral etch. Original magnification: 500×. (c) S5 tool steel austenitized, isothermally transformed (partially) at 540 °C (1000 °F) for 8 h, and water quenched to form upper bainite (dark); balance of austenite formed martensite. 4% picral + 2% nital. Original magnification: 1000×. (d) S5 tool steel austenitized, isothermally transformed at 400 °C (750 °F) for 1 h, and air cooled to form lower bainite. 37 to 38 HRC. 4% picral + 2% nital. Original magnification: 1000×

Fig. 19

Light micrograph showing patches of (a) upper bainite formed in 4150 steel partially transformed at 460 °C (860 °F) and (b) lower bainite (dark plates) in 4150 steel (nital etch).

Fig. 19

Light micrograph showing patches of (a) upper bainite formed in 4150 steel partially transformed at 460 °C (860 °F) and (b) lower bainite (dark plates) in 4150 steel (nital etch).

The other type of bainite is called lower bainite because it forms at lower temperatures than does upper bainite. The ferrite takes a plate morphology, and the cementite is present as very fine particles within the ferrite plates (Fig. 18b and 18d). Figure 19(b) shows lower bainite that has formed in a 4150 steel. The bainite plates are at angles with respect to each other, giving an acicular or needlelike appearance to the microstructure rather than the blocky or feathery appearance of upper bainite. Again, the very fine carbide particles in the bainite plates are not resolvable in the light micrograph.

### Proeutectoid Ferrite and Cementite

As described in Chapter 1, steels may contain either less carbon (hypoeutectoid steels) or more carbon (hypereutectoid steels) than the eutectoid composition. In these cases, cooling of the single-phase austenite first changes into a heterogeneous two-phase field before the eutectoid temperature is reached. In hypoeutectoid steels, a type of ferrite called proeutectoid ferrite forms before the eutectoid-reaction temperature is reached (Fig. 20a). In hypereutectoid steels, the two-phase structure during cooling is composed of austenite and cementite, and so the cementite forming prior to the eutectoid reaction is called proeutectoid cementite (Fig. 20b)

Fig. 20

Formation of (a) proeutectoid ferrite in hypoeutectoid steel and (b) proeutectoid cementite in hypereutectoid steel

Fig. 20

Formation of (a) proeutectoid ferrite in hypoeutectoid steel and (b) proeutectoid cementite in hypereutectoid steel

When hypoeutectoid steel is austenitized and then slowly cooled below the A3 critical temperature, proeutectoid ferrite nucleate is formed in the austenite grain boundaries. As the ferrite grains grow, carbon is rejected into the austenite grain. Eventually, the carbon concentration is sufficient for pearlite formation, and the balance of the microstructure is transformed to pearlite. The microstructures of two hypoeutectoid steels are shown in Fig. 21. They have a mixed ferrite-pearlite microstructure, with the amount of pearlite (dark) depending on the carbon content. Most of the pearlite colonies appear uniformly black because the light is scattered by the lamellar structures, which are too closely spaced to be resolvable in light micrographs. Ferritic-pearlitic steels are common for a host of structural applications. These steels are relatively inexpensive and are produced in large quantities with a wide range of properties. In most ferrite-pearlite steels, the carbon content and the grain size determine the microstructure and resulting properties.

Fig. 21

Microstructure of typical ferrite-pearlite structural steels at two different carbon contents. (a) 0.10% C. (b) 0.25% C. 2% nital + 4% picral etch. Original magnification: 200×. Source: Ref 10

Fig. 21

Microstructure of typical ferrite-pearlite structural steels at two different carbon contents. (a) 0.10% C. (b) 0.25% C. 2% nital + 4% picral etch. Original magnification: 200×. Source: Ref 10

The growth of proeutectoid ferrite depends on the rejection of carbon atoms into the austenite and the transfer of iron atoms across the ferrite and austenite interface from the fcc to the bcc structure. The latter process depends on the degree of coherency or disorder in atom arrangement at the interface. Alloying elements also can affect the partitioning and interface structure during the formation of proeutectoid ferrite. Under some conditions, substitutional alloying elements are incorporated into the ferrite structure if they are ferrite stabilizers or are rejected from the ferrite if they are austenite stabilizers. More details on proeutectoid ferrite is in Ref 17. Generally under conditions of slow cooling, the proeutectoid ferrite grows uniformly into austenite and an equiaxed ferrite grain structure. However, if the austenite in hypereutectoid steels is rapidly cooled, the transfer of iron atoms across ferrite and austenite interfaces is restricted, and the diffusion-controlled growth of ferrite is replaced by a shear mechanism. As a result, a plate-shaped morphology of ferrite, frequently referred to as acicular or Widmanstätten ferrite, develops in rapidly cooled low-carbon steels. Substitutional alloying elements such as manganese tend to retard the formation of equiaxed ferrite grains and promote acicular ferrite formation.

In hypereutectoid steels, proeutectoid cementite nucleates and grows on austenite grain boundaries during cooling from the austenite phase field (Fig. 20b). Figure 22 shows a network of proeutectoid cementite that has formed on austenite grain boundaries of a hypereutectoid steel. Initial proeutectoid cementite growth appears to depend only on diffusion of carbon and therefore can proceed very rapidly. In alloy steels, later stages of cementite growth require the partitioning of substitutional alloying elements (such as chromium) and therefore are very sluggish. The very rapid initial growth of proeutectoid cementite may occur even during oil quenching for hardening and is associated with the intergranular fracture often observed in high-carbon steel quenched from temperatures above Acm.

Fig. 22

Microstructure of 1.2%C-Fe alloy showing cementite outlining the prior austenite grain boundaries and cementite needles in the grains of pearlite. The grain-boundary cementite is called proeutectoid cementite. This microstructure represents a hypereutectoid steel. 4% picral etch. Original magnification: 200×. Source: Ref 13

Fig. 22

Microstructure of 1.2%C-Fe alloy showing cementite outlining the prior austenite grain boundaries and cementite needles in the grains of pearlite. The grain-boundary cementite is called proeutectoid cementite. This microstructure represents a hypereutectoid steel. 4% picral etch. Original magnification: 200×. Source: Ref 13

In view of the brittleness that continuous networks of proeutectoid cementite impart, hypereutectoid steels are reheated intercritically into the austenite and cementite two-phase field for annealing (if maximum ductility and machinability are desired) or for hardening (if wear and fatigue resistance are required). During the intercritical heating, proeutectoid cementite networks as well as the lamellae of cementite in pearlite partially dissolve and spheroidize. For example, consider a 1095 steel (0.95 wt% C) received from a steel mill. If this steel is heated to 760 °C (1400 °F), the temperature-composition point is shown in Fig. 23 as an open circle with the horizontal arrowed line passing through it. Because the temperature-composition point lies in the shaded two-phase region labeled γ + Cm, this steel must consist of a mixture of austenite having composition O (0.85% C) and cementite of composition P (6.7% C). The schematic diagram illustrates what the microstructure would look like in the two-phase (intercritical) region of the phase diagram. The cementite appears as small, spherically shaped grains distributed fairly randomly over the austenite grains.

Fig. 23

Extension of the iron-carbon phase diagram to alloys illustrating intercritical heating to spheroidized cementite in a hypereutectoid steel. Source: Ref 18

Fig. 23

Extension of the iron-carbon phase diagram to alloys illustrating intercritical heating to spheroidized cementite in a hypereutectoid steel. Source: Ref 18

After quenching, the spheroidized cementite particles are then dispersed in a matrix of martensite. This improves toughness, because fracture is initiated at the fine spherical carbide particles, thus promoting a transgranular fracture morphology rather than an intergranular fracture due to proeutectoid cementite along prior austenite grain boundaries. With spheroidized cementite particles in hypereutectoid steels, toughness is related to the spacing of the particles (Ref 19).

### Martensite

Martensite is the phase formed in steels by a diffusionless, shear transformation of austenite and is the base structure for hardened steels. Martensite is not shown on the iron-carbon diagram because it does not form under equilibrium conditions; generally, rapid cooling to temperatures well below A1 is required to form martensite. As expected from the iron-carbon diagram, martensite eventually decomposes to a mixture of ferrite and cementite if heated below A1.

Shear or the displacive, cooperative movement of many atoms has already been mentioned as a mechanism by which bainite and acicular proeutectoid ferrite form. The formation of the latter structures, however, occurs under conditions such that carbon diffusion accompanies the formation of bcc ferrite. When martensite forms, even the carbon atoms cannot diffuse. Thus, the carbon atoms are trapped in the octahedral interstitial sites, creating a supersaturated ferrite with a body-centered tetragonal (bct) crystal structure. The higher the concentration of carbon atoms, the greater the tetragonality (Fig. 24; Ref 20).

Fig. 24

Effect of carbon on the tetragonality of martensite

Fig. 24

Effect of carbon on the tetragonality of martensite

Figure 25 shows schematically a martensite plate that has formed in austenite adjacent to a free surface. The martensite surface is tilted by the shear transformation, and the austenite plane along which the martensite forms is termed the habit plane. To accomplish the shape deformation shown, not only must the fcc austenite lattice transform to the bct lattice of martensite, but also the martensite crystal once formed must accommodate itself to the constraints of the surrounding bulk austenite and the restrictions imposed by the plane-strain deformation parallel to the habit plane. This accommodation is accomplished by slip or twinning of the martensite plate, and as a result, martensite in steels contains a high residual density of dislocations and/or fine twins.

Fig. 25

Diagram of martensite crystal, showing shear and surface tilting. Adapted from: Ref 21

Fig. 25

Diagram of martensite crystal, showing shear and surface tilting. Adapted from: Ref 21

The martensitic transformation is characterized by athermal kinetics; that is, the amount of martensite formed is independent of time and is a function only of the amount of undercooling below the martensite-start temperature (MS), the temperature at which martensite starts to form on cooling in a given steel. The following equation has been developed (Ref 22) for estimating the volume fraction of martensite, f, formed by quenching to any temperature, TQ:
$f=1−e−0.011[Ms−TQ]$

Thus, if the MS of a given steel is known, the amount of martensite formed on quenching to any temperature below MS can be established. A plot of martensite percentage versus the amount of undercooling below the MS is in Fig. 26.

Fig. 26

Extent of martensite formation as a function of undercooling below the martensite-start temperature. Source: Ref 23

Fig. 26

Extent of martensite formation as a function of undercooling below the martensite-start temperature. Source: Ref 23

The MS temperature is a function of the carbon and alloying-element content of a steel, and several relationships have been developed to relate MS to composition. Table 7 is a summary of typically used calculations for critical temperatures and martensite-start temperatures, although various martensite-start formulas have been developed over the years. The following formula is commonly used as a function of alloying given in weight percent (wt%):
$MS(°F)=930−600×C−60×Mn−20×Si−50×Cr−30×Ni−20×Mo−20×W$
Table 7
Typical formulas for calculating transformation temperatures of low-alloy steels
FormulaReference
Ae1 (°F) ~ 1333 – 25 × Mn + 40 × Si + 42 × Cr – 26 × Ni37
Ae3 (°F) ~ 1570 – 323 × C – 25 × Mn + 80 × Si – 3 × Cr – 32 × Ni37
Ac1 (°C) ~ 723 – 10.7 × Mn + 29.1 × Si + 16.9 × Cr – 16.9 × Ni + 290 × As + 6.38 × W38
Ac3 (°C) ~ 910 – 203 × ($C$) + 44.7 × Si – 15.2 × Ni + 31.5 × Mo + 104 × V + 13.1 × W38
MS (°F) ~ 930 – 600 × C – 60 × Mn – 20 × Si –50 × Cr – 30 × Ni – 20 × Mo – 20 × W39
M10 (°F) ~ MS – 1840
M50 (°F) ~ MS – 8540
M90 (°F) ~ MS – 18540
Mf (°F) ~ MS – 38740
Bs (°F) ~ 1526 – 486 × C – 162 × Mn – 126 × Cr – 67 × Ni – 149 × Mo40
B50 (°F) ~ Bs – 10840
Bf (°F) ~ Bs – 21640
FormulaReference
Ae1 (°F) ~ 1333 – 25 × Mn + 40 × Si + 42 × Cr – 26 × Ni37
Ae3 (°F) ~ 1570 – 323 × C – 25 × Mn + 80 × Si – 3 × Cr – 32 × Ni37
Ac1 (°C) ~ 723 – 10.7 × Mn + 29.1 × Si + 16.9 × Cr – 16.9 × Ni + 290 × As + 6.38 × W38
Ac3 (°C) ~ 910 – 203 × ($C$) + 44.7 × Si – 15.2 × Ni + 31.5 × Mo + 104 × V + 13.1 × W38
MS (°F) ~ 930 – 600 × C – 60 × Mn – 20 × Si –50 × Cr – 30 × Ni – 20 × Mo – 20 × W39
M10 (°F) ~ MS – 1840
M50 (°F) ~ MS – 8540
M90 (°F) ~ MS – 18540
Mf (°F) ~ MS – 38740
Bs (°F) ~ 1526 – 486 × C – 162 × Mn – 126 × Cr – 67 × Ni – 149 × Mo40
B50 (°F) ~ Bs – 10840
Bf (°F) ~ Bs – 21640

Figure 27 shows MS as a function of carbon content (Ref 23). The decrease in MS with increasing carbon content is related to the increased shear resistance produced by increasing amounts of carbon in solid solution in the austenite. An important consequence of low MS temperature, according to the aforementioned equation, is the reduced amount of martensite that forms on cooling to room temperature. Therefore, large volume fractions of austenite may be retained in high-carbon steels.

Fig. 27

Effect of carbon content on martensite-start (MS) temperature in steels. Composition ranges of lath and plate martensite in iron-carbon alloys are also shown. Source: Ref 24

Fig. 27

Effect of carbon content on martensite-start (MS) temperature in steels. Composition ranges of lath and plate martensite in iron-carbon alloys are also shown. Source: Ref 24

Figure 27 shows the two types of martensite that form in carbon steels. The two categories are based on the morphology and microstructural characteristics of the martensite (Ref 9, 23, and 25). The lath morphology forms in low- and medium-carbon steels and consists of regions or packets where many fine laths or board-shaped crystals are arranged parallel to one another. The habit plane of the laths is close to but not exactly {111}. The width of most of the laths is less than 0.5 μm, that is, below the resolution of the light microscope, and therefore the microstructure appears very uniform, with only the largest laths resolvable. Figure 28(a) is a micrograph of lath martensite in a low-alloy steel. Electron microscopy is required to show that the fine structure of lath martensite consists of a high density of tangled dislocations and that retained austenite is present as thin films between the martensite laths (Ref 27).

Fig. 28

Light micrographs of morphologies of martensite. (a) Lath martensite in low-carbon steel (0.03C-2.0Mn, wt%) at original magnification: 100×. (b) Plate martensite in matrix of retained austenite in a high-carbon (1.2 wt% C) steel at 1000×. (c) Mixed morphology of lath martensite with some plate martensite (P) in a medium-carbon (0.57 wt% C) steel at original magnification: 1000×. All 2% nital etch. Source: Ref 26

Fig. 28

Light micrographs of morphologies of martensite. (a) Lath martensite in low-carbon steel (0.03C-2.0Mn, wt%) at original magnification: 100×. (b) Plate martensite in matrix of retained austenite in a high-carbon (1.2 wt% C) steel at 1000×. (c) Mixed morphology of lath martensite with some plate martensite (P) in a medium-carbon (0.57 wt% C) steel at original magnification: 1000×. All 2% nital etch. Source: Ref 26

The plate morphology of martensite (Fig. 28b) forms in high-carbon steels and consists of martensite plates that form at angles with respect to each other. Consistent with the low MS of this alloy, a large amount of retained austenite is present. The fine structure of plate martensite consists of thin twins, approximately 10 nm thick, and/or dislocation arrays typical of low-temperature plastic deformation. The impingement of nonparallel plates during development of a martensite microstructure sometimes causes microcracks to form in the martensite (Ref 28). Examples of microcracks are shown in the large plate of Fig. 29. The density of microcracks in plate martensite is reduced by formation of martensite in fine-grained austenite, by lowering the carbon concentration of the austenite by intercritical austenitizing (thereby developing a more parallel martensite morphology and less impingement), and by tempering.

Fig. 29

Light micrograph (aqueous 10% sodium bisulfide etch) showing plate martensite and retained austenite in an Fe-1.39C alloy. Source: Ref 29

Fig. 29

Light micrograph (aqueous 10% sodium bisulfide etch) showing plate martensite and retained austenite in an Fe-1.39C alloy. Source: Ref 29

The carbon range in which a mixed morphology of lath and plate martensite forms is sensitive to alloy content and is not well known. Even in the range of carbon contents where lath martensite forms, there is a gradual decrease in the definition of packets with increasing carbon content (Ref 30).

#### Martensite Hardness and Hardenability

As-quenched martensite is very hard and brittle, typically requiring tempering for some degree of softening. The hardness of martensite depends solely on the carbon content (Table 8). Alloying does not change the hardness of martensite. However, alloying can slow the kinetics of pearlite formation and thus promote the formation of martensite at slower cooling rates. If martensite can form at slower cooling rates, then the depth of hardening is increased. The ability of steel to be hardened to greater depths during quenching is referred to as hardenability (see also Chapter 4, “Hardness and Hardenability of Steels,” in this book).

Table 8
Effect of carbon concentration and martensite content on the as-quenched hardness of steel
C, %Hardness, HRC, with extent of martensite, %
9995908050
0.1038.532.930.727.826.2
0.1239.534.532.329.327.3
0.1440.636.133.930.828.4
0.1641.837.635.332.329.5
0.1842.939.136.833.730.7
0.2044.240.538.235.031.8
0.2245.441.939.636.333.0
0.23464240.537.534
0.2446.643.240.937.634.2
0.2647.944.542.238.835.3
0.2849.145.843.440.036.4
0.3050.347.044.641.237.5
0.3251.548.245.842.338.5
0.335248.546.54339
0.3452.749.346.943.439.5
0.3653.950.447.944.440.5
0.3855.051.449.045.441.5
0.4056.152.450.046.442.4
0.4257.153.450.947.343.4
0.4357.253.5514844
0.4458.154.351.848.244.3
0.4659.155.252.749.045.1
0.4860.056.053.549.846.0
0.5060.956.854.350.646.8
0.5261.757.555.051.347.7
0.5462.558.255.752.048.5
0.5663.258.956.352.649.3
0.5863.859.557.053.250.0
0.6064.360.057.553.850.7
C, %Hardness, HRC, with extent of martensite, %
9995908050
0.1038.532.930.727.826.2
0.1239.534.532.329.327.3
0.1440.636.133.930.828.4
0.1641.837.635.332.329.5
0.1842.939.136.833.730.7
0.2044.240.538.235.031.8
0.2245.441.939.636.333.0
0.23464240.537.534
0.2446.643.240.937.634.2
0.2647.944.542.238.835.3
0.2849.145.843.440.036.4
0.3050.347.044.641.237.5
0.3251.548.245.842.338.5
0.335248.546.54339
0.3452.749.346.943.439.5
0.3653.950.447.944.440.5
0.3855.051.449.045.441.5
0.4056.152.450.046.442.4
0.4257.153.450.947.343.4
0.4357.253.5514844
0.4458.154.351.848.244.3
0.4659.155.252.749.045.1
0.4860.056.053.549.846.0
0.5060.956.854.350.646.8
0.5261.757.555.051.347.7
0.5462.558.255.752.048.5
0.5663.258.956.352.649.3
0.5863.859.557.053.250.0
0.6064.360.057.553.850.7

Hardenability of steel is usually determined by the Jominy end-quench test, where a bar of standard dimension is austenitized and quenched at one end. This results in different cooling rates along the length of the bar and thus different hardness and percentages of martensite along the bar (see Chapter 4). Positions along the Jominy bar are equivalent to cooling rates. Positions on the Jominy bar also can be equated to an equivalent bar diameter (see Chapter 4).

### Tempered Martensite

As-quenched martensite is supersaturated with carbon, has a very high interfacial energy per unit volume associated with the fine laths or plates of the martensitic microstructure, contains a high density of dislocations that store considerable strain energy, and may coexist with retained austenite. As a result of these characteristics, martensitic microstructures are quite unstable and decompose when heated. A practical benefit of the decomposition is increased toughness, and for this reason, almost all hardened steels are heated to some temperature below Ac, a heat treatment process that is referred to as tempering.

A wide range of microstructures may be produced by tempering of martensite. Carbon atoms rearrange themselves into various configurations and structures within the martensite crystals even at temperatures well below 100 °C (212 °F) (Ref 31). Tempering between 100 °C and Ac1 produces various types of carbide-particle dispersions as well as major changes in the matrix martensite. The reactions that produce the carbides have long been recognized and are classified as stages of tempering: T1, T2, and so on. It has been suggested that the reactions that depend on very short-range rearrangement of carbon atoms in the as-quenched martensite prior to carbide formation be classified as aging reactions, as distinguished from the carbide-forming reactions (Ref 32, 33).

Table 9 lists the various reactions and microstructural changes that may be developed by tempering steel. The aging and tempering classifications serve primarily to mark microstructures that form on the way to equilibrium, which will ultimately become a microstructure that consists of spheroidized carbide particles dispersed in a matrix of equiaxed ferrite grains. Many of the reactions or microstructural states require further characterization, some occur concurrently, and others may yet be discovered. Thus, investigations and assessment in tempering reactions are a continuing area of interest (e.g., Ref 34, 35, 36). The actions are controlled by diffusion of carbon, iron, and/or alloying elements, and therefore, steel composition, time, and temperature determine where a given tempering treatment stops in the sequence of structural changes indicated in Table 8.

Table 9
Tempering reactions in steel
Temperature rangeReaction and symbol (if designated)Comments
°C°F
−40 to 100−40 to 212Clustering of 2 to 4 carbon atoms on octahedral sites of martensite (A1); segregation of carbon atoms to dislocations and boundariesClustering is associated with diffuse spikes around fundamental electron diffraction spots of martensite.
20 to 10070 to 212Modulated clusters of carbon atoms on (102) martensite planes (A2)Identified by satellite spots around electron diffraction spots of martensite
60 to 80140 to 175Long-period ordered phase with ordered carbon atoms (A3)Identified by superstructure spots in electron diffraction patterns
100 to 200212 to 390Precipitation of transition carbide as aligned 2 nm diameter particles (T1)Recent work identifies carbides as eta (orthorhombic, Fe2C); earlier studies identified the carbides as epsilon (hexagonal, Fe2.4C).
200 to 350390 to 660Transformation of retained austenite to ferrite and cementite (T2)Associated with tempered-martensite embrittlement in low- and medium-carbon steels
250 to 700480 to 1290Formation of ferrite and cementite; eventual development of well-spheroidized carbides in a matrix of equiaxed ferrite grains (T3)This stage now appears to be initiated by chi-carbide formation in high-carbon Fe-C alloys.
500 to 700930 to 1290Formation of alloy carbides in Cr-, Mo-, V-, and W-containing steels. The mix and composition of the carbides may change significantly with time (T4).The alloy carbides produce secondary hardening and pronounced retardation of softening during tempering or longtime service exposure at approximately 500 °C (930 °F).
350 to 550660 to 1020Segregation and cosegregation of impurity and substitutional alloying elementsResponsible for temper embrittlement
Temperature rangeReaction and symbol (if designated)Comments
°C°F
−40 to 100−40 to 212Clustering of 2 to 4 carbon atoms on octahedral sites of martensite (A1); segregation of carbon atoms to dislocations and boundariesClustering is associated with diffuse spikes around fundamental electron diffraction spots of martensite.
20 to 10070 to 212Modulated clusters of carbon atoms on (102) martensite planes (A2)Identified by satellite spots around electron diffraction spots of martensite
60 to 80140 to 175Long-period ordered phase with ordered carbon atoms (A3)Identified by superstructure spots in electron diffraction patterns
100 to 200212 to 390Precipitation of transition carbide as aligned 2 nm diameter particles (T1)Recent work identifies carbides as eta (orthorhombic, Fe2C); earlier studies identified the carbides as epsilon (hexagonal, Fe2.4C).
200 to 350390 to 660Transformation of retained austenite to ferrite and cementite (T2)Associated with tempered-martensite embrittlement in low- and medium-carbon steels
250 to 700480 to 1290Formation of ferrite and cementite; eventual development of well-spheroidized carbides in a matrix of equiaxed ferrite grains (T3)This stage now appears to be initiated by chi-carbide formation in high-carbon Fe-C alloys.
500 to 700930 to 1290Formation of alloy carbides in Cr-, Mo-, V-, and W-containing steels. The mix and composition of the carbides may change significantly with time (T4).The alloy carbides produce secondary hardening and pronounced retardation of softening during tempering or longtime service exposure at approximately 500 °C (930 °F).
350 to 550660 to 1020Segregation and cosegregation of impurity and substitutional alloying elementsResponsible for temper embrittlement

Source: Ref 33

Significant increases in toughness are achieved by tempering at temperatures above 150 °C (300 °F). In general, subject to the development of various embrittlement phenomena, as tempering temperature increases, toughness increases and hardness decreases. Therefore, in applications where high hardness must be retained, tempering is performed at relatively low temperatures, usually between 150 and 200 °C (300 and 390 °F). Very fine carbide particles precipitate from the supersaturated martensite as a result of low-temperature tempering (Ref 37). The carbides are not cementite but rather transition carbides. Transition carbides include epsilon-carbide with a hexagonal structure and eta-carbide with an orthorhombic structure. Both epsilon-carbide and eta-carbide have carbon contents substantially higher than that of cementite.

Steels tempered to develop the fine transition carbides show a modest but significant increase in toughness. The hardness, however, remains high because of the extremely fine carbide dispersion and the retention of much of the dislocation substructure introduced by the martensitic transformation.

In steels tempered between 200 and 350 °C (390 and 660 °F), the transition carbide is replaced by cementite or chi(χ)-carbide, and retained austenite transforms to ferrite and cementite. The chi-carbide is a complex carbide with a monoclinic structure that forms in tempered high-carbon martensites and is eventually replaced by cementite. Chi-carbides are coarser than the transition carbides present at the interfaces of the martensite plates as well as within the plates (Ref 38).

Tempering between 200 and 350 °C (390 and 660 °F) also leads to transformation of retained austenite (see Chapter 7, “Hardening and Tempering of Steels,” in this book). The retained austenite is stable throughout the tempering-temperature range in which the transition carbide forms but begins to transform at temperatures above 200 °C (390 °F). Austenite in medium-carbon steels is retained between martensite laths and, when it transforms on tempering, produces relatively coarse plates of interlath cementite (Ref 27).

The coarse carbides produced by replacement of the transition carbides and transformation of the retained austenite, together with a limited recovery of the dislocation substructure of the martensite, reduce impact toughness. This decrease in impact toughness produced by tempering in the range of 250 to 400 °C (480 to 750 °F) is referred to as tempered martensite embrittlement, as described in Chapter 7, “Hardening and Tempering of Steels,” in this book.

Tempering at temperatures above 400 °C (750 °F) produces substantial coarsening of the microstructure. Not only do the cementite particles coarsen and spheroidize, but also the martensitic matrix is significantly altered. The laths are almost dislocation-free and are now ferrite because all of the carbon has completely precipitated as carbides. The reduction in dislocation density is driven by the reduction of the strain energy that accompanies the elimination of the dislocations and is accomplished by various recovery mechanisms.

As tempering temperature increases above 400 °C (750 °F), hardness and strength drop rapidly and toughness improves significantly. In alloy steels, the development of fine alloy carbide dispersions offsets the softening that accompanies the changing dislocation substructure and coarsening of the lath and cementite structure. In fact, if the alloy carbide dispersions are sufficiently fine and dense, an increase in hardness may develop. This increase in hardness due to alloy carbide precipitation high in the tempering-temperature range is referred to as secondary hardening.

As noted, toughness increases significantly with increasing tempering temperature. However, if impurities such as phosphorus, antimony, and tin are present in a steel, these elements may segregate to grain boundaries and/or carbide-matrix interfaces and cause large reductions in impact toughness (Ref 39). This phenomenon develops during tempering in, or slow cooling through, the temperature range 350 to 550 °C (660 to 1020 °F) and is referred to as temper embrittlement. Research on embrittlement was active in the 1950s because of failures in turbine generator rotors (Ref 40), with many studies on the effects of alloying and impurity elements (Ref 39, 41).

In general, temper embrittlement forms a C-shaped curve in time-temperature plots. At high temperatures, the kinetics of impurity diffusion to grain boundaries are rapid, but the tendency to segregate is low because the matrix solubility for the element increases with temperature. Hence, embrittlement occurs rapidly but to a small degree. At low temperatures, the tendency to segregate is high, but the diffusion kinetics are not rapid enough to reach maximum embrittlement. The optimum combination of thermodynamic and kinetic factors favoring embrittlement occurs at some intermediate temperature, called the “knee” of the C-curve. For many commercial steels of interest, the knee may occur in the temperature range from 455 to 510 °C (850 to 950 °F) but can be shifted up or down depending on the composition, grain size, and microstructure of the steel.

## References

1.
K.

Thelning
,
Steel and Its Heat Treatment
,
Butterworths
,
1967
2.
A.

Bavaro
,
Heat Treatments and Deformation
,
Trait. Therm.
,
Vol 240
,
1990
, p
37
41
3.
R.L.

Orr
and
J.

Chipman
,
Trans. AIME
,
Vol 239
,
1967
, p
630
4.
H.

Stuart
and
N.

Ridley
,
J. Iron Steel Inst.
,
Vol 204
,
1966
, p
71
5.
E.

Hornbogen
,
Werkstoffe
, 2nd ed.,
Springer-Verlag
,
1979
6.
E.A.

Brandes
and
G.B.

Brook
,
Smithell’s Metal Reference Book
, 7th ed.,
Butterworth-Heinemann
,
1992
7.
L.H.

Van Vlack
,
Elements of Materials Science and Engineering
, 4th ed.,
,
1980
8.
A.G.

Guy
,
Elements of Physical Metallurgy
, 2nd ed.,
,
1959
9.
G.

Krauss
, Physical Metallurgy and Steel Heat Treatment,
Metals Handbook Desk Edition
,
American Society for Metals
,
1985
, p
28
-
1
to
28
-
10
10.
B.L.

Bramfitt
, Effects of Composition, Processing, and Structure on Properties of Irons and Steels, Materials Selection and Design,
Vol 20
,
ASM Handbook
,
ASM International
,
1997
11.
G.

Krauss
, Ferritic Microstructures,
Steels: Processing, Structure, and Performance
, 2nd ed.,
ASM International
,
2015
, p
113
131
,
12.
G.

Krauss
, Ferritic Microstructures,
Steels: Processing, Structure, and Performance
2nd ed.,
ASM International
,
2015
, p
213
232
13.
B.L.

Bramfitt
and
S.J.

Lawrence
, Metallography and Microstructures of Carbon and Low-Alloy Steels,
Metallography and Microstructures
,
Vol 9
,
ASM Handbook
,
ASM International
,
2004
, p
608
626
14.
Brian Newbury
, Eutectoid Structures in Invariant Transformation Structures,
Metallography and Microstructures
,
Vol 9
,
ASM Handbook
,
ASM International
,
2004
, p.
152
164
,
15.
Isothermal Transformation and Cooling Transformation Diagrams
,
American Society for Metals
,
1977
, p
28
16.
H.W.

Paxton
and
J.B.

Austin
,
Historical Account of the Contribution of E.C. Bain
,
Metall. Trans.
,
Vol 13
,
1972
, p
1035
1042
17.
G.

Krauss
, Pearlite, Ferrite, and Cementite,
Steels: Processing, Structure, and Performance
,
ASM International
,
2015
, p
39
62
,
18.
J.D.

Verhoeven
,
Steel Metallurgy for the Non-Metallurgist
,
ASM International
,
2007
19.
G.

Krauss
, The Relationship of Microstructure to Fracture Morphology and Toughness of Hardened Hypereutectoid Steels,
Microstructure and Residual Stress Effects on the Properties of Case Hardened Steels
,
TMS-AIME
,
1984
20.
C.S.

Roberts
,
Effect of Carbon on the Volume Fractions and Lattice Parameters of Retained Austenite and Martensite
,
Trans. AIME
,
Vol 197
,
1953
, p
203
204
21.
B.A.

Bibby
and
J.W.

Christian
,
The Crystallography of Martensite Transformations
,
JISI
,
Vol 197
,
1961
, p
122
131
22.
D.P.

Koistinen
and
R.E.

Marburger
,
A General Equation Prescribing the Extent of the Austenite-Martensite Transformation in Pure Iron-Carbon Alloys and Plain Carbon Steels
,
Acta Metall.
,
Vol 7
,
1959
, p
59
60
23.
G.

Krauss
,
Principles of Heat Treatment of Steel
,
American Society for Metals
,
1980
24.
A.J.

DeArdo
,
G.A.

Ratz
, and
P.J.

Wray
, Ed.,
Thermomechanical Processing of Microalloyed Austenite
,
TMS-AIME
,
1982
25.
A. R.

Marder
and
G.

Krauss
,
The Morphology of Martensite in Iron-Carbon Alloys
,
Trans. ASM
,
Vol 60
,
1967
, p
651
660
26.
D.

Aliya
and
S.

Lampman
, Physical Metallurgy Concepts in Interpretation of Microstructures,
Metallography and Microstructures
,
Vol 9
,
ASM Handbook
,
ASM International
,
2004
, p.
44
70
27.
G.

Thomas
,
Retained Austenite and Tempered Martensite Embrittlement
,
Metall. Trans. A
,
Vol 9
,
1978
, p
439
450
28.
A.R.

Marder
and
A.O.

Benscoter
,
Micro-cracking in Fe-C Acicular Martensite
,
Trans. ASM
,
Vol 61
,
1968
, p
293
299
29.
A. R.

Marder
,
A.O.

Benscoter
, and
G.

Krauss
,
Microcracking Sensitivity in Fe-C Plate Martensite
,
Metall. Trans.
,
Vol 1
,
1970
, p
1545
1549
30.
T.

Maki
,
K.

Tsuzaki
, and
I.

Tamura
,
The Morphology of Microstructure Composed of Lath Martensite in Steels
,
Trans. Iron Steel Inst. Jpn.
,
Vol 20
,
1980
, p
207
214
31.
Winchell Symposium on Tempering of Steel
,
Metall. Trans. A
,
Vol 14
,
1983
, p
985
1146
32.
G.B.

Olson
and
M.

Cohen
,
Early Stages of Aging and Tempering of Ferrous Martensites
,
Metall. Trans. A
,
Vol 14
,
1983
, p
1057
1065
33.
G.

Krauss
, Tempering and Structural Change in Ferrous Martensites,
Phase Transformations in Ferrous Alloys
, in
Phase Transformations in Ferrous Alloys
,
A.R.

Marder
and
J.I.

Goldstein
, Ed.,
The Metallurgical Society, TMS-AIME
,
1984
34.
G.

Krauss
, Tempering of Martensite in Carbon Steels,
Phase Transformations in Steels
,
,
2012
, p
126
150
.
35.
G.

Krauss
,
Tempering of Lath Martensite in Low and Medium Carbon Steels: Assessment and Challenges
,
Steel Research International
,
Vol 88
(
No. 10
),
2017
, p
1700038
.
36.
G.A.

Thomas
,
J. G.

Speer
,
D. K.

Matlock
,
G.

Krauss
, and
R. E.

Hackenberg
, Time-Temperature Equivalence in Martensite Tempering,
Proceedings of the International Conference on Martensitic Transformations (ICOMAT) 2008
,
G.B.

Olson
,
D.S.

Lieberman
,
A.

Saxena
, Ed.
Wiley
,
2013
, p
595
600
37.
D.L.

Williamson
,
K.

Nakazawa
, and
G.

Krauss
,
A Study of the Early Stages of Tempering in an Fe-1.22%C Alloy
,
Metall. Trans. A
,
Vol 10
,
1979
, p
1351
1363
38.
C.-B.

Ma
,
T.

Ando
,
D.L.

Williamson
, and
G.

Krauss
,
Chi-Carbide in Tempered High Carbon Martensite
,
Metall. Trans. A
,
Vol 14
,
1983
, p
1033
1045
39.
C.J.

McMahon
Jr.
, Temper Brittleness—An Interpretive Review,
Temper Embrittlement in Steel
, STP 407,
ASTM
,
1968
, p
127
167
40.
J.

Comon
,
P.F.

Martin
, and
P.G.

Bastien
, Statistical Study of Factors Influencing Impact Strength of Turbine Generator Rotors—Influence of Temper Embrittlement, “Temper Embrittlement in Steel,” STP 407,
ASTM International
,
1968
, p
74
89
41.
C.J.

McMahon
Jr.
,
Temper Embrittlement of Steels: Remaining Issues
,
Mater. Sci. Forum
,
Vol 46
,
1989
, p
61
76
[1]

Portions adapted from G. Krauss, Physical Metallurgy and Steel Heat Treatment, Metals Handbook Desk Edition, American Society for Metals, 1985; T. Ericsson, Principles of Heat Treating of Steels, Heat Treating, Vol 4, ASM Handbook, ASM International, 1991; and J. Dossett and H. Boyer, Practical Heat Treating, 2nd ed., ASM International, 2006|

## Send Email

Recipient(s) will receive an email with a link to 'Practical Heat TreatingBasic Principles > Chapter 2: Fundamentals of Steel Heat Treatment[1]' and will not need an account to access the content.

Subject: Practical Heat TreatingBasic Principles > Chapter 2: Fundamentals of Steel Heat Treatment[1]

(Optional message may have a maximum of 1000 characters.)

×
Close Modal