Effect of Solution Treatment Temperature on Microstructure and Properties of Fe-0.72Mn-3.7Al-0.53C Low-Density Cast Steel

Abstract

In the present research, the microstructure and mechanical properties of low-density Fe-0.72Mn-3.7Al-0.53C steel were investigated after solution treatment at 900 °C, 1000 °C, 1110 °C and 1200 °C for 1 h. The density of steel is about 7.0 g·cm⁻³ due to the addition of a higher content of aluminum elements. The microstructure was characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM), and the mechanical behavior was analyzed by room temperature tensile testing. The results show that the microstructure of the steel is ferrite and martensite after solution treatment, and that martensite can be divided into dislocation martensite and twinned martensite according to different substructures. Part of the martensite grows in a mirror-symmetrical manner in order to adjust the strain energy that increases with the system undercooling to form twinned martensite. After solution treatment at different temperatures, the tensile strength and elongation of the steel increased and then decreased with the increase of the solution treatment temperature, and the tensile strength could reach 928.92 MPa, while maintaining excellent toughness and elongation at 5.89%.

1. Introduction

In recent years, in order to better cope with environmental degradation and resource scarcity and to achieve the goal of energy conservation and emission reduction, Fe-Mn-Al-C system low-density steels have received extensive attention from researchers because of their high strength and low density, and they have great potential for development in the automotive, marine and aerospace industries [1,2,3,4,5,6]. In addition, low-density steel also has attractive properties such as high energy absorption behavior, high strength, good fatigue properties and good high temperature antioxidants. Gutierrez-Urrutia et al. [6] showed that Fe-30.5Mn-2.1Al-1.2C had a strong plastic product of 88 G Pa·% after solid solution treatment at 1100 °C for 2 h. The reduction in density of low-density steels in the Fe-Mn-Al-C system is due to the addition of light elements, which changes the lattice parameter of the steel, while the lower atomic masses also lead to a reduction in density [7,8,9]; for example, the addition of 12% aluminum reduces the density of iron by 17%, where the effect of lattice expansion is 10% and the effect of atomic mass reduction is 7% [5]. Low-density steels are subject to various deformation mechanisms, including transformation-induced plasticity (TRIP), twinning-induced plasticity (TWIP), shear band-induced plasticity (SIP) and microband-induced plasticity (MBIP) [10,11,12], which results in good strength and plasticity in Fe-Mn-Al-C low-density steels.

The addition of a large amount of Al elements shrinks the γ (Austenite) region and expands the α (Ferrite) region, promoting the formation of ferrite; δ-ferrite will be accompanied by α-ferrite generation, which leads to a decrease in toughness after the subsequent solution treatment [13,14] and makes δ-ferrite be able to exist at room temperature. Now that δ-ferrite breaks the continuity of the austenite matrix, cracks will appear, making it difficult for the TRIP and TWIP effect to work, which results in the reduction of plasticity [15,16]. According to chen et al. [4], in order to eliminate the adverse effect of δ-ferrite on plasticity, the solution treatment temperature should be higher than 1000 °C and no more than 1300 °C.

However, the Fe-Mn-Al-C system of low-density steels has been studied in rolled steel, and little research has been conducted on low-density cast steels. Guo et al. [17] studied the microstructure of Fe-15Mn-10Al-5Ni-0.8C cast steel and established the austenite grain growth kinetic model. Wang et al. [18] investigated the heat deformation behavior and dynamic recrystallization behavior of as-cast Fe-15Mn-15Al-5Ni-1C, established the constitutive equation and recrystallization prediction model for low-density steels with Ni addition, and characterized its microstructures. Not only that, among the Fe-Mn-Al-C system low-density steels studied so far, the research objects are mostly austenitic low-density steels with a high content of both Al and Mn elements, and there are fewer studies on low-density steels with a low Al content. In addition, solution treatment could improve the homogeneity of the microstructure of low-density steel so as to improve the strength and plasticity of low-density steel. For this reason, optimizing the parameters of solution treatment is crucial to obtain excellent mechanical properties of low-density cast steels [19,20].

Based on the above discussion, in this experiment, low-alloy Fe-0.72Mn-3.7Al-0.53C low-density cast steel (experimental steel) on the basis of high-strength, low-alloy cast steel was developed to investigate the effect of solution treatment of cast low-density at 900~1200 °C on the phase transformation behavior, microstructure and mechanical properties.

2. Materials and Methods

The experimental steel Fe-0.72Mn-3.7Al-0.53C was melted in a vacuum induction furnace, and the chemical composition of the experimental steel is shown in

Table 1. Chemical composition of the experimental steel (wt.%).

C	Si	Mn	Mo	Cr	Ni	V	Ti	Nb	Al	P, S	Fe
0.53	0.88	0.72	0.85	0.83	1.14	0.053	0.046	0.028	3.72	0.006	bal

. The samples of 10 mm × 10 mm × 10 mm was prepared by wire electro discharge machined (EDM) cutting. The as-cast samples were processed with solution treatment at 900 °C, 1000 °C, 1110 °C and 1200 °C for 1 h and then water-quenched to room temperature.

The microstructure of Fe-0.72Mn-3.7Al-0.53C was observed by a 20 kV JEOL IT-500A (JEOL, Tokyo, Japan) scanning electron microscope, which was equipped with an energy dispersive spectrometer (EDS). The etching agent used was 4% nitric acid alcohol. Samples for transmission electron microscopy Tecnai F-20 (FEI, Hillsboro, TX, USA) were prepared by mechanical polishing to 100 μm followed by ion milling. Tensile samples of gauge length 35 mm, diameter 8 mm, clamping width 19.38 m and total length 80 mm, were tested at the UTM-5105 universal electronic testing machine (Sansi Company, Shenzhen, China), with a tensile rate of 0.5 mm/min and the entrance force of 100 N. Vickers hardness was measured by DUH-211 microhardness tester (SHIMADZU, Kyoto, Japan) used a 100 gm load and dwell time of 10 s. The density of the experimental steel was measured by a densitometry based on the Archimedes Principle, and the density was about 7.0 g·cm⁻³.

3. Results and Discussion

3.1. Microstructure

Figure 1 shows the scan electronic microscopy of Fe-0.72Mn-3.7Al-0.53C low-density cast steel after solution treatment at different temperatures. It was observed that the microstructure consisted of ferrite and martensite. The ferrite existed stably as a matrix, and the martensite was transformed from austenite during cooling. After solution treatment at different temperatures, the morphologies of martensite in the experimental steels were also different. The martensite was not parallel to each other but arranged at 60° or 120°. After solution treatment at 900 °C and 1000 °C, the content of lath martensite in the austenite of the experimental steel was less, and the content of austenite was more. Along with the increase in temperature came an increase in the content of martensite, when the temperature rises to 1100 °C and 1200 °C; almost all of the austenite transformed into martensite, and the microstructure of the experimental steel was a mixed structure of lath martensite and plate martensite. According to the different substructures, the martensite obtained from austenite transformation can be divided into dislocation martensite and twinning martensite after the solution treatment [21,22].

The difference in the martensite morphology is mainly caused by the strain energy. When the content of carbon is low, the bcc lattice of the martensite, similar to the austenite lattice spacing and Fe atomic spacing, is prone to phase transformation, and the interfacial energy is lower after solution treatment at a high temperature [23]. With the increase of the content of carbon, however, the lattice difference between martensite and austenite increases, which leads to an increase of volume strain energy and interfacial energy, and the increase of interfacial energy promotes the nucleation of martensite, resulting in plate martensite after solution treatment at 1100 °C and 1200 °C, as shown in Figure 1c,d.

At the same time, in the process of martensitic transformation, the Ms point temperature decreases with the increase of the content of carbon [24], and the cooling rate increases during the continuous cooling process; the transformation temperature of the supercooled austenite decreases [25], the lattice change brings about the distortion energy of the martensite continues to increase, and the morphology of the martensite will also change, from lath to plate. Not only that, if the orientation of martensite also shifts during the nucleation and growth of martensite, the system will adjust the strain energy by mirror-symmetrical growth, which leads to the transformation of the martensite substructure from dislocation to twinning. This also explains the difference of martensite morphology after solution treatment at different temperatures. The martensitic transformation process (γ_fcc→ε_hcp) inevitably leads to volume expansion, and the newly formed martensite and prior austenite are, respectively, subjected to compressive stress and tensile stress, and the stress field generated by martensitic transformation leads to an increase in strain energy. Moreover, adding a large number of alloying elements to the experimental steel reduces the Ms point, and the volumetric strain energy also becomes relatively large.

From the perspective of energy, the system always follows the principle of energy reduction. The formation of a new phase will inevitably generate stress around the parent phase, and the phase transformation was impeded by the lattice distortion generated by the stress field. In the process of transforming supercooled austenite into martensite, martensite will grow along a certain crystal plane and direction of austenite to reduce the strain energy, which leads to the formation of martensite in martensite. Substructures of dislocations and twins are formed during growth.

Figure 2 shows the TEM bright field images of the experimental steel after 1100 °C and 1200 °C solution treatment for 1 h. The martensite in the experimental steel after solution treatment at 1100 °C and 1200 °C is a mixture of lath martensite and plate martensite, which result in the excellent mechanical properties of these two groups of experimental steels. The twins and dislocations are substructures of plate martensite and lath martensite, respectively. Generally speaking, twins are formed in two ways: the first is deformation twinning caused by external stress, and the second is phase transformation twinning formed by phase transformation. The biggest difference between the two is whether the original structure has been changed. Among them, the original crystal structure is not changed when the deformation twins are formed by stress, but the crystal structure has changed due to the self-organization of the system during the phase transformation twinning process [20,23,25]. However, it can be seen from Figure 2a,b that the substructure of plate martensite in the experimental steels after solution treatment at 1100 °C and 1200 °C contains not only twins, but also a large number of dislocations, which is due to the fact that the nucleation position of plate martensite is at the austenite grain boundary after solution treatment at a high temperature and that a large number of dislocations, which piled up at the grain boundary, provide a site for the nucleation and growth of martensite, which makes the martensite tend to grow along the crystal plane, resulting in martensite nucleation and growth [26]. There are not only twins, but also a large number of dislocations in the martensite grains after solution treatment at 1100 °C and 1200° C.

Figure 3 shows the TEM bright-field image and SAED pattern of twinned martensite. In Figure 3b, pattern [$\overline{1}$10]_t is mirror symmetric around the common normal of the (112) plane, which can be considered as rotated 70.5° counterclockwise around [110]. Figure 3c is the SAED with [11$\overline{1}$] zone axis. In Figure 3a, the laths with different lining degrees of martensite are twin-grown because the darker slats are parallel to the incident electron beam, so two overlapping single-crystal diffraction patterns of [$\overline{1}$10] were obtained, indexed as [$\overline{1}$10]_α and [$\overline{1}$10]_t. It is a typical feature of selected electron diffraction of bcc (body centered cubic) {112} <111> twins and matrix, since the incident electrons are parallel to <110> [27]. However, an extra diffraction spot appears at n/3 of (112), which is impossible for bcc {112} <111> twins [27]. This is supported by Figure 3c, as the single-crystal diffraction pattern of bcc Fe was obtained from the same angle when the sample was tilted parallel to the electron beam at [$11\overline{1}$].

Furthermore, when the electron beam is parallel to [$11\overline{1}$], the lath boundary reaches the edge position, where the boundary plane between the medium twins of the experimental steel is determined as (011). On the contrary, the bcc {112} <111> twin boundary of (112) reaches the edge when the electron beam is parallel to [$\overline{1}$10]. From crystallography, it is possible that when the electron beam is parallel to [$\overline{1}$10], (011) may project a band on the screen (($\overline{1}$10) plane), which makes the bulk boundary appear thicker; this has been observed in many lath martensite examples with different carbon contents [28,29]. Since both (011) and (112) were intersected with ($\overline{1}$10) via [$\overline{1}$11], they are easily mistaken for the feature of the (112) twin plane, and conversely, the mirror-symmetric fine grains are mistaken for bcc {112} <111> twins [28]. The extra diffraction spots commonly observed in the electron diffraction pattern of martensite can be attributed to the double diffraction of the twinned martensite variants rather than the bcc {112}<111> twins [30].

Figure 4 shows the bright field image at the grain boundaries between martensite and ferrite after solution treatment at 1100 °C and 1200 °C, respectively, and it can be clearly observed that a high density of dislocations occurs at the grain boundaries after solution treatment. Plenty of studies have pointed out that the martensitic transformation is a non-diffusion-type transformation, and the transformation speed is very fast [21,22]. The semi-coherent interface composed of dislocations separates the martensite from the austenite, so the growth of martensite is not only affected by the driving force, but the defects in the crystal also play an important role. Lots of dislocations at the austenite grain boundaries provide conditions for martensite growth; that is, once martensite nucleates, the time required for martensite growth is extremely short, so martensite and austenite grow. There must be a semi-coherent interface between the twin crystals so as to ensure its rapid growth; not only that, when martensite is sheared, it needs to rely on the slip of dislocations to generate new interfaces. The above factors led to the need for a large number of dislocations at the austenite grain boundaries to meet the growth conditions of martensite.

3.2. Mechanical Properties

The mechanical properties of Fe-0.72Mn-3.7Al-0.53C low-density cast steel after solution treatment are shown in

Table 2. Mechanical properties of the experimental steel after solution treatment at different temperatures.

Temperature (°C)	Time (h)	TS (Mpa)	TE (%)	TS × TE (GPa·%)
900	1	834.27	4.84	4.04
1000	1	836.74	4.98	4.17
1100	1	928.92	5.89	5.47
1200	1	848.62	5.24	4.45

. The mechanical properties reached the peak value at 1100 °C with a tensile strength of 928.92 Mpa, an elongation of 5.89% and TS × TE of 5.47 GPa·%. After solution treatment at 900 °C, the experimental steel had poorer comprehensive mechanical properties with a tensile strength of 834.27 Mpa, an elongation of 5.89% and TS × TE of 4.04 GPa·%.

Figure 5 shows the stress–strain curves of the experimental steel after solution treatment at different temperatures and the relationship between the mechanical properties of the experimental steel and the solution treatment temperature. The tensile strength and elongation of the experimental steel increased as the solution temperature increased from 900~1100 °C, reaching a peak value at 1100 °C; the tensile strength was 928.92 Mpa, and the elongation was 5.89%, which was due to the production of a large amount of martensite in the austenite grains. After solution treatment at 1200 °C, the tensile strength and elongation decreased to 848.62 Mpa and 4.45%, respectively.

The relative movement of dislocations leads to the plastic deformation of the material when it under stress, and the hindered movement of dislocations will increase the strength. Grain refinement strengthening and precipitation strengthening are the main strengthening methods for high strength low-alloys steels [1,3]. However, after the solution treatment, the precipitates in the ferrite dissolved as the temperature increased, and almost all of the precipitates in the ferrite were dissolved after solution treatment at 1200 °C. When the cooling rate was high, there was not enough time for carbides and nitrides to precipitate, or the size of the carbides and nitrides was small (Figure 1b,c). They are hardly hindered and bent when the dislocation lines cut through these carbides and nitrides, which led to the lower tensile strength of the experimental steel. Meanwhile, due to the continuous formation of ferrite during the cooling process and the sufficient diffusion of carbon, Mn and other elements into the austenite to form martensite, during the stretching process, the ferrite structure with low strength yielded first. This resulted in a lower strength of the experimental steels. In the meantime, due to the high cooling rate, the volume expansion accompanied by martensite transformation led to the formation of a large number of mobile dislocations in the ferrite. Therefore, even when the stress level is low, these dislocation sources were also activated, and the continuous yielding behavior was observed. Additionally, due to the existence of a large number of twinned martensite, which was formed as the increase of solution treatment temperature, the uncoordinated deformation of twinned martensite and ferrite in tensile engineering will lead to the cleavage fracture of ferrite, and this brittle fracture method will lead to a great reduction of the elongation of the experimental steel [31]. Moreover, the austenite grains grow as the solution temperature increases, and the strength of the experimental steel also decreases, which leads to a decrease in the strength and elongation of the experimental steel after solution treatment at 1200 °C.

In addition, a high density of dislocations was piled up at the prior austenite grain boundaries after solution treatment at 1100 °C and 1200 °C, and the interaction of dislocations produced a large number of slip systems. When external stress is applied, these slips start to reduce the internal stress, but the existence of twinned martensite at the austenite grain boundary prevents the relative sliding of these slip systems; secondly, the twinned martensite, as a hard and brittle phase, formed the time-to-interaction interaction, resulting in the presence of microcracks within the grains, both of which resulted in a decrease in the toughness of the experimental steel.

Figure 6 shows the work-hardening curves after solution treatment at different temperatures. Overall, the work-hardening rate of the experimental steel changed in three stages, that is, three stages of rapid decrease, continuous stabilization and continued decrease.

The rapid decline of the work hardening rate of the experimental steel in the first stage is related to the deformation of the ferrite [32,33]. The work hardening of the dislocations in the ferrite is the main source of the work hardening rate in the first stage. The initial deformation stage also corresponds to the continuous yielding characteristics of the stress–strain curve of the experimental steel. In the first stage, the initial work hardening rate of the experimental steel after solution treatment at 1200 °C reached its peak value, which is because δ-ferrite appeared in the matrix of the experimental steel after solution treatment at a higher temperature [34]. Moreover, as the solution treatment temperature increased to 1100 °C, the stage of rapid decrease of the work hardening rate became shorter, which indicated that with the increase of solution temperature, the elastic deformation stage of the experimental steel was shorter, and the plastic deformation stage was longer.

The second stage is the continuous and stable stage of the work hardening rate. With the increase of strain, the work hardening rate of the four groups of experimental steels was relatively stable at about 15,000 Mpa. This stage was the coordinated deformation stage of martensite and ferrite. As a hard and brittle phase, even with increasing strain, martensitic deformation was small, so the work hardening rate was stable in a lower range. The true strains of the experimental steels after solution treatment at 1100 °C and 1200 °C were larger at the end of this stage, which corresponds to the better elongation of these two groups of experimental steels [34,35].

The third stage is the continuous decline stage of work hardening. With the continuous increase of strain, the work hardening rate decreases to the maximum value of the corresponding true stress–strain curve when fracture occurs.

The microhardness of the experimental steel after solution treatment at different temperatures is shown in

Table 3. Microhardness after solution treatment at different temperatures.

Temperature (°C)	Time (h)	Microhardness (HV)
		Ferrite	Martensite
900	1	106.6	415.52
1000	1	117.1	421.16
1100	1	120.4	464.24
1200	1	143.2	454.95

. The microhardness of ferrite and martensite of the experimental steels did not change much, with a highest ferrite hardness of 143.2 HV and a highest martensite hardness of 464.24 HV. However, there was a huge distinction between the microhardness of the martensite and ferrite.

Figure 7 shows the microhardness of the experimental steel after solution treatment at different temperatures. Between 900 °C and 1100 °C, the microhardness of the experimental steel increased with the increase of temperature, and the hardness of the experimental steel decreased slightly from 1100 °C and 1200 °C.

4. Conclusions

In the present research, Fe-0.72Mn-3.7Al-0.53C low-alloy, low-density cast steel is used as the research object. After solution treatment at different temperatures for 1 h, the phase evolution, microstructure and mechanical properties of Fe-0.72Mn-3.7Al-0.53C low-alloy, low-density cast steel were studied. The results are as follows:

The microstructure of Fe-0.72Mn-3.7Al-0.53C low-density cast steel after solution treatment is ferrite and martensite. Part of the martensite grows in a mirror-symmetrical manner in order to adjust the strain energy that increases with the system undercooling to form twinned martensite after solution treatment at 1100 °C and 1200 °C.

In the range of 900 to 1200 °C, the tensile strength and elongation of Fe-0.72Mn-3.7Al-0.53C low-density cast steel increase first and then decrease with the increase of solution temperature. The tensile strength and the elongation reach a peak value at 1100 °C, which are 928.92 Mpa and 5.89%. The production of a large amount of martensite in the austenite grains leads to increase in tensile strength and elongation, and for solution treatment at 1200 °C, grain coarsening will adversely affect the tensile strength and elongation of the steel.

The extra diffraction spots commonly observed in the electron diffraction pattern of martensite can be attributed to the double diffraction of the twinned martensite vari-ants rather than the bcc {112}<111> twins.

The Fe-0.72Mn-3.7Al-0.53C low-density cast steel shows the typical characteristics of tertiary work hardening during the deformation process. The work hardening of the dislocations in the ferrite is the main source of the work hardening rate in low strain, and as strain increases, martensite and ferrite are coordinate deformed until fracture occurs.