The structure of martensite obtained by quenching plays a critical role in imparting strength and toughness to steel.
However, due to variations in the type, composition and heat treatment conditions of the steel, the morphology, internal fine structure and microcrack susceptibility of tempered martensite can vary significantly.
These changes have a profound impact on the mechanical properties of martensite.
Therefore, it is imperative to have an in-depth knowledge of the morphological characteristics of martensite and understand the various factors that influence its morphology.
1. Morphology of martensite
The morphology and fine structure of martensite have been extensively studied using thin-film transmission electron microscopy.
Research has revealed that although the morphology of martensite in steel can be diverse, its characteristics can typically be divided into the following categories:
1. Lattice martensite
Lath martensite is a common martensite structure that forms in low- to medium-carbon steel, maraging steel, stainless steel, and other iron-based alloys.
Figure 1 illustrates the typical structure of lathed martensite in mild steel.
Fig. 1 Low Carbon Alloy Steel 100X Martensite Strip (0.03% C, 2% Mn)
The microstructure of certain steels is made up of numerous groups of laths, which is why it is called lath martensite.
In some cases, the lath is not easily exposed or etched and instead appears blocky, leading to its alternative name, blocky martensite.
Because the primary substructure of this type of martensite is dislocation, it is commonly called dislocation martensite.
Cluster martensite is composed of several groups of laths, with each group of laths consisting of multiple strips of approximately equal size, arranged approximately parallel to each other in a specific direction.
Figure 2 highlights the high-density displacement within the laths that is characteristic of lathed martensite.
Fig. 2 Transmission microstructure of low carbon alloy steel thin film (0.03% C, 2% Mn) 20000X
Additionally, phase transformation twins may exist within the slats, but they are typically localized and are not present in significant quantities, nor are they the primary form of fine structure.
The crystal orientation relationship between lath martensite and its parent austenite is typically called the Kurdjumov-Sachs (KS) relationship, with the habit plane being (111)γ.
However, in the case of 18-8 stainless steel, the usual plane of lathed martensite is (225)γ.
Figure 3 illustrates the crystallographic characteristics of the microstructure of lathed martensite, as determined by research.
Fig. 3 Schematic diagram of the crystallographic characteristics of the microstructure of lathed martensite
A large area composed of bundles of martensite slats arranged in parallel is called a slat group and is denoted by A.
A single grain of primary austenite can contain several groups of laths, typically ranging from 3 to 5.
Each group of strips can be divided into multiple parallel regions, as shown in the figure.
In some cases, when certain solutions are used for etching, only the boundary of the lath group is visible, resulting in a blocky appearance of the microstructure, hence the name blocky martensite.
When color recording techniques are employed, such as 100cc HCl + 5g CaCl 2 + 100cc CH 3 CH Solution, black and white tones can be observed within the slat group.
Regions with the same tone correspond to martensite slats with the same orientation and are called homotropic beams.
According to the Kurdjumov-Sachs (KS) orientation relationship, martensite can exhibit 24 different orientations in the original austenite, including six orientations that can generate parallel lathed martensite (see Figure 4).
Fig. 4 Martensite (111) in steel γ Possible orientation during in-plane forming
An isopathic beam refers to a bundle of slats that have been transformed from one of the slats.
Several parallel collinear beams combine to form a group of bands.
Some researchers suggest that within a group of slats, only two groups can alternate their positions.
Therefore, a slat group is typically composed of two groups of aligned slat beams that alternate with each other and may also alternate with each other at large angle grain boundaries. However, there are cases in which the group of slats is mainly composed of a single type of homotropic beam, as illustrated in C in Figure 3.
An aligned beam consists of strips arranged in parallel, as depicted at D in Figure 3.
This scenario can be observed through electron microscopy, as shown in Figure 5.
Fig. 5 Some microstructures in the isotropic beam of lathed martensite in Fe-0.2% C alloy (transmission electron micrograph)
According to the research results on Fe-0.2%C alloy, the strip width distribution is a lognormal distribution as shown in Fig.
Fig. 6 Distribution of film and replica technology strips
As seen in the figure, the width of the slat with the highest frequency of occurrence varies from 0.15 to 0.20 μm, and the distribution curve is strongly skewed towards the smaller sized slats. However, a small proportion of slats are 1 to 2 μm wide.
Figure 7 illustrates that larger slats are often distributed throughout the slat bundle, which is a key feature of the slat bundle microstructure.
Fig. 7 Microstructure of lathed martensite in Fe-0.2% C alloy (transmission electron micrograph)
Experimental results indicate that changing the austenitizing temperature changes the austenite grain size, but has minimal impact on the lath width distribution.
However, the size of the lath group increases as the austenite grain size increases, while the ratio between the two remains approximately constant. Thus, the number of lath groups generated in an austenitic grain normally remains unchanged.
Thin-film electron microscopy measurements show that the area of the lath boundary in unit volume of martensite is approximately 65,000 cm²/cm³.
The area of the small-angle crystal boundaries in the slat bundle is about 5 times that of the large-angle crystal boundaries.
In the Fe-Cr-Ni alloy based on 18-8 stainless steel, both lath martensite and ε'-martensite (close-packed hexagonal lattice) can be generated, resulting in a microstructure that differs significantly from that of the Fe-C alloy, as shown in Figure 8.
Fig. 8 Microstructure of Fe-15% Cr-12 and Ni alloy martensite (Ms = – 90°) (aqua regia, glycerin corrosion)
The structure does not contain groups of slats or symppositional beams; instead, it is created as a group of thin slats surrounding a sheet of ε'-martensite (as shown in the parallel strips in the figure).
However, the electron microscopic structure of this lath martensite is identical to that found in Fe-C and Fe-Ni alloys.
2. Flaked martensite
Another typical martensite structure in iron series alloys is lamellar martensite, which is commonly found in high- and medium-carbon tempered steels and high NiFeNi alloys.
The typical structure of lamellar martensite in high-carbon steel is shown in Fig.
Fig. 9 T12A 400X steel superheated quenching structure (heated to 1000 ℃, water quenched)
This specific type of martensite is known by several names, such as lenticular martensite, due to its biconvex lens-like shape. It is also referred to as acicular martensite or bamboo leaf martensite, because when observed under a microscope intersecting with the grinding surface of the sample, it appears as needle-shaped or bamboo leaf-shaped structures.
The substructure of lamellar martensite is mainly composed of twins and is therefore also called twin martensite. The microstructure of lamellar martensite is characterized by the fact that the lamellae are not parallel to each other.
When an austenitic grain with uniform composition is cooled to a temperature slightly lower than Ms, the first martensite formed will course through the entire austenitic grain and divide it into two halves. This limits the size of the martensite formed later, resulting in varying sizes of lamellar martensite. As illustrated in Figure 10, the martensite flakes formed later tend to be smaller.
Fig. 10 Microstructure of lamellar martensite
The size of the flakes depends almost entirely on the grain size of the austenite.
Flaky martensite can often be seen with obvious middle ridge (see Fig. 11).
Fig. 11 Flaked martensite (with obvious intermediate ridge, T12 steel is carburized at 1200 ℃ for 5 hours and quenched at 180 ℃)
Currently, the rule for the formation of medium ridges is not well defined.
The habit plane of lamellar martensite is (225) γ or (259) γ. The orientation relationship with the parent phase is the Kurdjumov-Sachs (KS) relationship or the Xishan relationship.
As shown in Figure 12, the martensite contains numerous fine lines that are Luan transform crystals, while the thin banded ribs in the central part of the joint are midridges.
Fig. 12 TEM structure of lamellar martensite
The existence of the Lüders transformation crystal is an important characteristic of lamellar martensite.
The Lüders crystal spacing is approximately 50 Å and generally does not extend to the martensite boundary.
The edge of the sheet presents a complex matrix of dislocations, which are generally believed to be screw dislocations arranged regularly in the (111) α´ direction.
The transformation of Lüders crystal to lamellar martensite is generally a (112)α´ Lüders crystal.
However, in the Fe-1.82% C alloy (c/a=1.08), a (110) Lüders crystal will mix with a (112)α´ Lüders crystal.
Depending on the internal substructure of lamellar martensite, it can be divided into twin transformation area (middle part) centered on the middle crest and twin free area (in the surrounding part of the lamella there are dislocations).
The proportion of dual zones varies with league composition.
In Fe-Ni alloys, the higher the Ni content (the lower the Ms point), the larger the double zone.
According to research on Fe-Ni-C alloy, even for an alloy with the same composition, the double zone proportion increases with decreasing Ms point (as caused by changing austenitizing temperature).
However, the density of the transformation twins barely changes, and the thickness of the twins remains at about 50 Å.
Lath martensite and lamellar martensite are the two most basic martensite morphologies in steel and alloy.
Its morphological and crystallographic characteristics are listed in Table 1.
Table 1 Types and characteristics of martensite in iron-carbon alloys
| Characteristics | Slat martensite | lamellar martensite | |
| Usual surface | （111）γ | （225）γ | （259）γ |
| orientation relationship | KS Relationship (111) γ lll（110） α ´【110】 γ 【111】 α.' | KS Relationship (111) γ lll（110） α ´【110】 γ 【111】 α.' | Relationship Xishan (111) yll (110) α.' 【211】 γ ll【110】 α.' |
| Formation temperature | M>350℃ | M≈200~100℃ | M.<100℃ |
| Alloy composition %C | <0.3 | 1~1.4 | 1.4~2 |
| Closed at 0.3~1 | |||
| Histomorphology | The slats are generally arranged in parallel groups from the austenite grain boundary to the grain interior, and the slat width is generally 0.1 ~ 0.2μ, length less than 10μ. An austenitic grain contains several groups of laths. There are small-angle grain boundaries between slat bodies and large-angle grain boundaries between slat groups. | The convex blade of the lens (or needle, bamboo leaf) is slightly thicker in the middle, the primary is thicker and longer and passes through the austenite grains, while the secondary is smaller. Between the primary lamellae and the austenite grain boundary, the angle between the lamellae is large and they collide with each other to form microcracks. | On the same left, there is an intermediate ridge in the center of the slice, and thin slices with a zigzag distribution are common between the two primary slices. |
| Substructure | Dislocation (entanglement) network, the dislocation density increases with carbon content, generally (0.3 ~ 0.9) × A small amount of thin twins can sometimes be seen at 1012 cm/cm3. | Thin twins with a width of about 50｜ form the Lie transform and twin regions with the middle ridge as the center. As the M point decreases, the transformation twin region increases and the edge of the sheet is a complex dislocation matrix. The twin plane is (112) α ※, the twin direction is (11I) α ´ | |
| Training process | Cooling nucleation, new martensite sheets (slats) are produced only during cooling | ||
| The growth speed is slow and a slat is formed in about 10-4s | The growth speed is high and a leaf is formed in about 10-7s | ||
| There is no “explosive” transformation, and the cooling transformation rate is about 1%/℃ within less than 50% of the transformation amount | When M<0 ℃, an “explosive” transformation occurs, and the new martensite sheet does not produce evenly with the drop in temperature, but because of the self-triggering effect, it forms into groups (“Z” shape) continuously and massively in a very small temperature range, accompanied by a temperature rise of 20~30℃ | ||
3. Other morphology of martensite
3.1 Butterfly martensite
In Fe Ni alloys or Fe Ni C alloys, when martensite is formed within a certain temperature range, martensite with special morphology will appear, as shown in Fig.
Fig. 13 Microstructure of Prato Martensite
The three-dimensional shape of this martensite is a slender rod and its section is butterfly-shaped, so it is called butterfly martensite.
Butterfly martensite was found to form in Fe-31% Ni or Fe-29% Ni-0.26% C alloy in the temperature range of 0 to -60 ℃.
Electron microscope studies have confirmed that its internal substructure comprises high-density dislocations, with no visible twins.
The crystallographic relationship to the parental phase generally adheres to the KS relationship. Butterfly martensite mainly forms between 0 and -20 ℃, coexisting with lamellar martensite between -20 and -60 ℃.
It can be seen that for the two alloy systems mentioned above, the formation temperature range of butterfly martensite lies between the formation temperature range of laminated martensite and lamellar martensite.
The junction of two wings of butterfly martensite is very similar to the middle ridge of lamellar martensite. It is assumed that the martensite (probably twinning) growing from here to both sides along different orientations will be butterfly shaped.
The joint part of butterfly martensite is similar to the joint part of two pieces of martensite formed by an explosion, but it does not contain any twin structures, which is different from sheet martensite.
From the point of view of internal structure and microstructure, butterfly martensite is similar to lath martensite, but does not occur in rows.
At present, many aspects of butterfly martensite are still unclear. However, its morphology and properties fall between lath martensite and lamellar martensite, making it an interesting topic to explore.
3.2 Flaky martensite
This martensite was discovered in a Fe-Ni-C alloy that exhibits an exceptionally low Ms point. It appears as a very thin band in three-dimensional shape, with the bands intersecting and exhibiting twists, branches, and other unique shapes, as depicted in Figure 14c.
Fig. 14 Fe-Ni-C alloy cooled to the Ms point
Microstructure of martensite formed at the same temperature
The electron microscopic structure of this martensite is shown in Fig.
Fig. 15 Electron microscopic structure of lamellar martensite (Fe-31%, Ni0.23% C, Ms=- 190 ℃, cooled to – 196 ℃)
The material under examination is a complete Luan martensite composed of (112) α´ Luan crystals without a central ridge, which distinguishes it from lamellar martensite.
It was observed that the morphology of martensite in the Fe-Ni-C system changes from lenticular to lamellar as the formation temperature decreases.
In the Fe-Ni-C alloy with carbon content of approximately 0.25% and Ms = -66 ℃, the structure is explosive lamellar martensite, as depicted in Figure 14a.
As Ms decreases to -150℃, a small amount of lamellar martensite begins to appear, as shown in Figure 14b.
At the point where Ms drops to -171 ℃, the entire structure is composed of lamellar martensite (see Figure 14c).
It was found that the transition temperature from lens sheet to thin sheet increases with increasing carbon content.
When the carbon content reaches 0.8%, the formation zone of lamellar martensite is below -100 ℃.
As the transformation temperature decreases, during the transformation of lamellar martensite, there is not only continuous formation of new martensite sheets, but also thickening of old martensite sheets.
Thickening of old martensite sheets is not visible in lamellar martensite.
3.3 ε' Martensite
All the martensites mentioned above have a body-centered cubic (α') or body-centered square structure.
In alloys with low stacking fault energy in austenite, a dense hexagonal lattice ε' martensite can also form.
This type of martensite is predominant in alloys with high Mn-Fe-C content.
However, 18-8 stainless steel represented by Fe-Cr-Ni alloys often coexists with α'-martensite.
ε' martensite is also thin, as shown in Figure 16.
Widmanstatten formation is observed along the (111)γ surface, with a substructure characterized by numerous stacking faults.
Fig. 16 Martensite Microstructure of Fe-16.4% Mn Alloy (Nitrate Alcohol Corrosion)
2. Relationship between chemical composition and martensitic morphology and internal substructure of alloys
The presence of alloying elements in steel has a crucial impact on the shape of martensite.
A common example is that the shape of martensite in Fe-C and Fe-Ni alloys changes from laths to flakes as the alloy content increases. For example, in the Fe-C alloy, below 0.3% carbon, martensite is lath-shaped, while above 1% carbon it becomes flake-shaped. In the range of 0.3% to 1.0% carbon, both forms of martensite can be present.
However, different sources may show inconsistent concentrations that trigger the transition from slatted martensite to lamellar martensite. This variability is linked to the effect of quenching rate, with a higher quenching rate leading to a lower minimum carbon concentration required for double martensite formation.
Figure 17 illustrates the impact of carbon content on the type of martensite, the Ms point and the amount of austenite retained in Fe-C alloys.
Fig. 17 Effect of carbon content on point Ms, lath martensite content and retained austenite content (carbon steel hardened at room temperature)
The figure demonstrates that steel with a carbon content of less than 0.4% contains almost no retained austenite.
As the carbon content increases, the Ms point decreases while the amount of Luan crystalline martensite and retained austenite increases.
Table 2 summarizes the relationship between martensite morphology and the composition of binary iron alloys.
Table 2 Martensite Morphology of Binary Fe Alloys
|
League system |
Slat martensite |
lamellar martensite |
Martensite |
|||||
|
League composition (%) |
Point M (℃) |
League composition (%) |
Point M (℃) |
League composition (5%) |
||||
|
Extended Y Zone |
Fe-C Fe-N Fe-Ni Fe-Pt Fe-Mn Fe-Ru Faith-Go Fe-Cu Fe-Co |
<1.0 <0.7 <29 <20.5 <14.5 7.5~19 20~48 2~6 0~1 1~24 |
700~200 700~350 700~25 700~400 700~150 600 ~ 200 550~40 – 700~620 620~800 |
0.6~1.95 0.7~2.5 29~24 24.6 – – – – – – |
500~40 350~100 25~195 -30 – – – – |
– – – – 14.5~27 11~17 35~53 – – – |
||
|
Reduced Y area |
Fe-Cr Fe-Mo Fe-Sn Fe-V Some |
<10 <1.94 <1.3 <0.5 <0.3 |
700~260 700~180 |
– – – – – |
– – – – – |
– – – – – |
||
The table demonstrates that all alloying elements in the γ zone are transformed into lath martensite.
As the concentration of alloying elements in the expanded P zone increases, the overall Ms point decreases significantly, accompanied by a change in martensite morphology.
For example, in binary alloys such as Fe-C, Fe-N, Fe-Ni, Fe-Pt and others, the morphology of martensite transforms from lath to flake with increasing alloying element content.
However, the addition of Mn, Ru and Ir can greatly reduce the stacking fault energy of austenite, resulting in a change in morphology from lath martensite to ε´ martensite with an increase in alloying element content in binary alloys of iron.
Fe-Cu and Fe-Co alloys are exceptions among the expanded γ-zone elements.
Although Cu is part of the Y-zone expanding element, the small amount of solid solution in Fe leads to a relatively stable Ms point and therefore shows the same tendency as the Y-zone shrinkage alloys.
Fe-Co alloy is unique compared to other alloys. With increasing Co content, the Ms point increases, making it a special case.
In general, there are several types of alloying elements in steel, but if a third element is added to the Fe-C or Fe-Ni alloy, a small amount will not significantly change the morphology of martensite compared to the binary alloy.
As mentioned previously, Fe-Ni-C alloys can form lath, butterfly, lens sheet and thin sheet martensite. The relationship between the formation temperature of these four forms of martensite and the carbon content and Ms point is shown in Figure 18.
Fig. 18 Relationship between martensite morphology, carbon content and Ms point of the Fe-Ni-C alloy
The figure shows that the formation temperature of lenticular and lamellar martensite increases as the carbon content increases.
The figure also highlights the formation area of butterfly martensite with a hatched area.
Table 3 summarizes the relationship between the morphology, substructure and crystallographic characteristics of martensite in iron-based alloys.
Table 3 Characteristics of Fe System Martensite
| Usual surface | orientation relationship | Martensite morphology | Second type of shear | Martensite substructure | Point M. | Austenite Failure Energy | Steel Grade |
| （111） （225）（259） |
K.S. K.S. Xishan |
Clapboard | Double slip | Dislocation | High average low |
Low low or medium high |
Low carbon copper, high Mn steel, low Ni steel; high and medium carbon steel, stainless steel, medium Ni steel; high Ni steel, high carbon steel |
In steel, martensite with carbon content less than 0.20% is generally considered to have a body-centered cubic lattice structure. Martensite with carbon content greater than 0.20% is considered to have a body-centered tetragonal lattice structure.
It is commonly believed that low-carbon steel body-centered cubic martensite is equivalent to dislocation martensite, while body-centered tetragonal martensite is equivalent to high-carbon twin martensite. However, in Fe-Ni alloys, double martensite can also have a body-centered cubic structure.
As a result, the relationship between crystal structure and substructure remains uncertain.
3. Factors affecting martensite morphology and substructure
The above discussion covers the law of change in martensite morphology due to a change in alloy composition.
Currently, there is much debate about the factors impacting this change and there is no clear consensus.
It is widely accepted that morphological changes are essentially changes in substructure, and common perspectives include:
1. Mrs. Dot
Supporters of this point of view believe that the morphology of martensite depends on the temperature Ms.
They state that in Fe-C alloys, an increase in carbon content results in a decrease in temperature Ms.
At temperatures below a certain range (300-320°C), it becomes easier to form transformation twins and resultant lamellar martensite.
Table 4 describes the relationship between martensite morphology, carbon steel crystalline characteristics and carbon content and temperature Ms.
Table 4 Relationship between martensite morphology and crystallographic characteristics of carbon steel and carbon content and Ms point of the steel
| Carbon content (%) | Crystal structure | Mentoring relationship | Usual surface | Point M (℃) | Martensite morphology |
| <0.3 | Cubic or square centered body | CS Relationship | （111） | >350 | Slat Martensite |
| 0.3~1.0 | Centroid square | CS Relationship | Strip (111), sheet (225) | 350~200 | Mixed martensite |
| 1.0~1.4 | Centroid square | CS Relationship | （225） | <200 | Flaked martensite with partial twins and dislocations in the substructure |
| 1.4~1.8 | Body · Square Heart | Xishan relationship | （259） | <100 | Typical lamellar martensite with obvious middle ridge and “Z-shaped” arrangement |
The transformation of martensite morphology from lath to flake with a decrease in the Ms point can be explained as follows:
Table 4 demonstrates a correlation between habit surface and martensite morphology. The formation temperature of low-carbon martensite is generally believed to be high, with the (111)γ plane as the usual plane due to its large shear. At these high temperatures, sliding is easier to occur than twinning and there are fewer (111) γ crystal systems in the face-centered cubic lattice, resulting in a limited number of initial orientations for martensite formation, leading to the formation of martensite grouped within the same austenite.
As the point temperature Ms decreases, twinning becomes easier to occur than sliding, and the habit plane changes to (225) γ or (259) γ. This change results in an increase in the number of crystal systems and initial orientations for martensite formation, leading to the formation of Li crystalline lamellar martensite with adjacent sheets not parallel to each other within the same austenite.
It has been established that the formation of martensite at high temperature cannot result in double lamellar martensite, even if the austenite is significantly strengthened. The Ms point in Fe-Ni-C alloys can be changed by changing the austenitization temperature, allowing different Ms points to be obtained within the same alloy.
When the cooling temperature is slightly lower than the corresponding Ms point, the change in martensite morphology from butterfly shape to sheet shape can be observed. Furthermore, the decrease in the formation temperature leads to an increase in the twin transformation zone.
The morphology of strain-induced martensite formed in the same alloy at various temperatures above the Ms point was also studied, revealing that the morphology of martensite changes with the change in the strain temperature (i.e., the formation temperature of strain-induced martensite). . These results confirm that the morphology of martensite and the internal structure of this type of alloy are exclusively related to the Ms point.
Furthermore, under high pressure and decreasing Ms point, the occurrence of transformation twins becomes more likely, leading to a change in martensite morphology from lath to sheet, as shown in Fig. 19. This experimental evidence supports the importance of the point Ms.
Fig. 19 Effect of 4000 MPa pressure on the Ms point and on the martensitic substructure of the ferromagnetic alloy
In the actual formation process, multiple martensites are produced consecutively at varying temperatures between the Ms and Mf points.
The temperature at which each martensite crystal forms is unique, therefore the internal structure and morphology of each martensite crystal are also distinct.
Therefore, it is more correct to state that the formation temperature, and not the Ms point, affects the morphology and internal structure of martensite.
2. The stacking fault energy of austenite
According to Kelly et al., they propose a hypothesis that states that the lower the austenite stacking fault energy, the more challenging it becomes to produce the transformation into bainite crystals and the more likely the formation of lath martensite.
Both 18-8 stainless steel and the Fe-8% Cr-1.1% C alloy exhibit low stacking fault energies. At the temperature of liquid nitrogen, dislocation martensite is formed. This phenomenon is difficult to explain using Ms Point's hypothesis, but it can be explained by this hypothesis.
Furthermore, in Fe-30~33% Ni alloy lamellar martensite, the twin transformation zone increases as the Ni content increases. As Ni is known to increase the stacking fault energy of austenite, this experimental phenomenon supports the hypothesis.
It is worth mentioning that this experimental phenomenon can also be explained by the Ms Point theory, as Ni decreases the Ms Point.
3. Strength of austenite and martensite
Recently, Davis and Magee proposed a hypothesis regarding the relationship between austenite strength and martensite morphology. They used an alloying method to change the strength of austenite and studied the resulting changes in martensite morphology.
The results revealed that the morphology of martensite changes based on the austenite yield strength at point Ms, which is approximately 206MPa. Above this limit, lamellar martensite with a usual plane of {259} γ forms. Below this limit, lath martensite with a habit plane of {111} γ or lamellar martensite with a habit plane of {225} γ forms.
As a result, Davis and Magee believe that the strength of austenite is the main factor affecting martensite morphology. They also further investigated the strength of martensite. When the strength of austenite is less than 206 MPa, if the strength of the resulting martensite is high, it forms as {225}γ martensite. If the strength of martensite is low, {111}γ martensite is formed.
This hypothesis can be applied to explain morphological changes resulting from changes in alloy composition or point Ms, particularly the transformation of {111}γ to {225}γ in FeNi alloys and {111}γ to {225}γ in {259 }γ in Fe-C alloys.
Furthermore, the hypothesis provides a clear understanding of the formation of {225}γ martensite, which has not been well defined in the past. It is formed when weak austenite transforms into strong martensite.
Although carbon has limited effects on strengthening austenite, it has a significant impact on strengthening martensite. {225} γ martensite occurs mainly in alloy systems with high carbon content.
This hypothesis is based on the following:
If relaxation of the transformation stress in martensite occurs only through twinned deformation, the resulting martensite will have the {259} γ habit plane.
When relaxation of the transformation stress is carried out partially in austenite through the sliding mode and partially in martensite through the twinning mode, the martensite will have the {225} γ habit plane.
If martensite also undergoes sliding mode, the habit plane will be {111} γ.
Experimental results suggest that this hypothesis is partially correct, but more research is still needed in the future.
It should be noted that the strength of austenite and martensite as described in this hypothesis is closely related to several factors such as alloy composition, type, Ms point, austenite stacking fault energy and others. Therefore, this hypothesis cannot be considered isolated.
4. Critical shear stress of martensite slip and double strain
This hypothesis emphasizes that the internal structure of martensite is mainly determined by the deformation mode during transformation, which is mainly controlled by the critical shear stress of slip or twin.
Figure 20 illustrates the effect of the critical shear stress of martensite slip or twin and the temperature of Ms and Mf on the formation of martensite morphology.
Fig. 20 Schematic diagram of the influence of critical shear stress and temperature Ms Mf on martensite morphology caused by martensite sliding or twinning
The arrows in the figure represent the potential directions of movement of the corresponding lines, which are caused by changes in the alloy composition. The movement of the lines leads to the intersecting movement of the twin slip curves.
From the figure, it can be seen that for low carbon steel (where the Ms and Mf points are both high), the critical shear stress required for sliding is lower than that required for twinning, resulting in the formation of lathed martensite with high dislocation density. On the other hand, for high-carbon steels (where the Ms and Mf points are both low), the critical shear stress required for twinning is small, resulting in the formation of lamellar martensite with a large number of twins.
In the case of medium carbon content, the points Ms and Mf are as shown in the figure. During martensitic transformation, lath martensite forms first, followed by lamellar martensite. This results in a mixed structure of both types of martensite.
Although this view appears fundamentally correct, the factors that cause changes in shear stress and how alloy composition or point Ms influence the critical shear stress for martensitic sliding or twinning are still unclear.
Some believe that increasing the transformation driving force leads to the transformation into lamellar martensite. For Fe-C alloys, the driving force limit for the change in martensite morphology is 1318 J/mol, and for Fe-Ni alloys it ranges from 1255 to 1464 J/mol. Others believe that the increase in C and N content in martensite, causing ordering, is closely related to the morphological transformation.
4. Formation of lamellar martensite microcracks in Fe-C alloy
When high carbon steel is hardened, it is susceptible to the formation of microcracks in the martensite.
Previously, it was thought that these microcracks were the result of microstress caused by volume expansion during martensitic transformation.
However, recent metallographic observations have revealed that the formation of microcracks is actually due to the collision of growing martensite, as illustrated in Figure 21.
Figure 21. Schematic diagram of microcracks formed by the collision of two Fe-C martensite sheets. (Section AA represents the cross section of a martensite sheet, which has diffused into two martensite sheets.)
Martensite formation occurs quickly. When martensite sheets collide with each other or with an austenite grain boundary, a significant stress field is generated due to the impact.
Because high-carbon martensite is extremely brittle and cannot be relieved by sliding or double deformation, it is prone to forming impact cracks.
This inherent defect increases the brittleness of high carbon martensite steel.
Under the influence of other stressors such as thermal stress and structural stress, microcracks will transform into macrocracks.
The presence of microcracks will also significantly reduce the fatigue life of the components.
Microcracks in Fe-C alloy lamellar martensite often occur at the junction of several radial martensite needles or within the martensite needles, as illustrated in Figure 22.
Fig. 22 Optical microscopic characteristics of microcracks in Fe-1.39%C alloy martensite
The sensitivity of microcrack formation in martensite is generally expressed in terms of the area of microcracks per unit volume of martensite (Sv).
Experimental evidence suggests that the sensitivity of martensite to microcrack formation is influenced by several factors, including:
1. Effect of Quenching Cooling Temperature
As the quenching cooling temperature decreases, the amount of austenite retained (represented by γR) in the quenched steel structure decreases, resulting in an increase in the amount of martensite and in the sensitivity to the formation of microcracks, as illustrated in Figure 23.
Fig. 23 Relationship between Fe-C martensite formation microcrack sensitivity and quenching temperature (1.39% C, heated at 1200 ℃ for 1 hour)
2. Effect of martensite transformation amount
Figure 24 illustrates the relationship between the amount of martensite transformation and the susceptibility to microcrack formation.
Fig. 24 The relationship between the microcrack sensitivity (SV) of martensite formation in Fe-1.86% C alloy and the average volume (V) of each piece of martensite, the number of martensite sheets in unit volume (NV ) and the martensite transformation:
According to the figure, the sensitivity to the formation of microcracks (Sv) increases with the increase in the martensite transformation variable, however, when the transformation fraction (f) exceeds 0.27, Sv does not continue to increase.
Although the number of martensite per unit volume (Nv) increases, the size of the martensite sheet formed, represented by the average volume (V) of a piece of martensite, decreases due to the continuous splitting of austenite.
Thus, the size of the martensite sheet (V) may have a critical value that affects the sensitivity (Sv) to microcrack formation. If V exceeds this critical value, the sensitivity to microcrack formation (Sv) increases with increasing transformation fraction.
In conclusion, crack formation is predominantly determined by the size of the martensite plates. Although the total number and area of cracks may increase with increasing martensite transformation variable, the large martensite flakes formed in the early stage result in the formation of most cracks during the initial stages of transformation.
3. Effect of martensite sheet length
Experience shows that as the length of the martensite sheet increases (i.e., the maximum sheet size increases), the susceptibility of the martensite to microcrack formation also increases, as illustrated in Figure 25.
Fig. 25 Relationship between the sensitivity of microcrack formation and the length of the martensite sheet (the number next to the dot is the martensite content%)
Long martensite sheets are more susceptible to impact than other martensite sheets due to their size. Furthermore, they tend to intersect with austenite grains, increasing the likelihood of encountering grain boundaries.
Experiments have shown that microcracks are predominantly formed in coarse martensite, while fine martensite rarely results in the formation of microcracks.
As a result, there is likely to be a critical martensite size for microcracks to occur in the martensite. Likewise, if the composition of austenite is relatively uniform, there will be a critical austenite grain size below which microcracks will not occur.
The idea that fine grains of austenite can reduce microcracks in hardened high-carbon steels has been implemented in production. However, it is still unclear whether the sensitivity to microcracks depends on the size of the martensite sheet itself or on the stress field generated by the growth of critical-sized martensite sheets.
4. Effect of austenite grain size
In the case of homogeneous austenite, the length of the martensite sheets formed in the initial phase is linked to the size of the austenite grains. Coarse grains of austenite result in the formation of coarse martensite, which is more prone to microcrack formation.
The experimental results, as shown in Figure 26, support this idea. The results indicate that steel with a high carbon content is more prone to cracking when quenched at higher temperatures.
Therefore, it is generally recommended to select a lower quenching temperature for quenching high carbon steel.
Fig. 26 Effect of carbon steel austenite grain size (1.22% C) on field microcrack sensitivity
5. Effect of carbon content on martensite
The effect of carbon content on the formation of microcracks in martensite is demonstrated in Figure 27.
Fig. 27 Effect of carbon content in martensite on sensitivity to microcracks
It can be seen in Figure 27 that the probability of microcrack formation increases as the carbon content in martensite increases.
However, if the carbon content in austenite is greater than 1.4%, the susceptibility to microcrack formation decreases. This is related to the usual plane of the crystal during martensitic transformation.
When the carbon content in steel exceeds 1.4%, the shape of martensite changes. The sheets become thicker and shorter, the angle between the martensite sheets becomes smaller, and the impact force and tension are reduced. As a result, the sensitivity to microcrack formation decreases.
Table 5 shows that the sensitivity to microcrack formation in 1.39% carbon steel decreases significantly with decreasing carbon content in martensite. Data are presented for a grain size of 3.
|
Temperature A1~Aw (℃) |
Carbon content in martensite (%) |
Retained austenite (%) |
Carbide amount (%) |
Sensitivity to the formation of microcracks S. (mm-1) |
|
1010 910 871 857 834 799 768 732 |
1.39 1:30 p.m. 1.21 1.18 1.05 1.01 0.92 0.83 |
33.5 22 15 13 12 8 9 6 |
3.9 6 6.5 12 15 17.5 20 |
18 17 13 9 10 4.5 1.5 0.15 |
Metallographic analysis indicates that the reduction in sensitivity to microcracks is associated with the presence of more parallel growing martensite in the microstructure.
Lath martensite has high plasticity and toughness, and the risk of mutual impact is reduced due to the parallel growth of lath martensite, leading to low sensitivity to microcracks.
As mentioned previously, high carbon steel is susceptible to cracking due to its coarse austenite grain structure and the high carbon content in martensite. To mitigate this, the production process tends to utilize lower heating temperatures and shorter holding times to lower the carbon content in the martensite and obtain finer grains.
In general, hypereutectoid steels, which undergo incomplete quenching, produce cryptocrystalline martensite, which is less prone to microcracks. This is why they have excellent general properties.
