Hot ductility behavior of a Fe–0.3C–9Mn–2Al medium Mn steel

The need for energy efficiency and lightweight trend in the automobile industry has led to the development and industrialization of various types of advanced high-strength steels (AHSS) over the past decades [1–5]. However, the basic properties of newly developed steels should be clarified before their wide application. Hot ductility is a key property that guarantees the continuous casting of steels. Understanding this property is beneficial for hot rolling. Hot ductility has been studied extensively for low C–Mn and C–Mn microalloyed steels to avoid transverse cracking [6–7]. The transformation of ferrite films and the precipitation of V(C,N) decrease hot ductility [8]. Mejía et al. [9] concluded that precipitation and inclusions cause poor ductility at 650–800^oC for microalloyed AHSS. Lan et al. [6] investigated the hot ductility of Fe–Mn–C TWIP steel and observed interdendritic brittle cracks without any dimple. The microsegregation of Mn and C elements and the retardation of dynamic recrystallization (DRX) affect the hot ductility of TWIP steel.

Medium Mn steel is a promising candidate for the industrial application of third-generation AHSS because of its excellent strength and ductility [10]. Recently, relevant studies have focused on chemical design, microstructure characterization, mechanical behaviors, and hydrogen embrittlement [11–14]. The mechanical properties for forming or service have been studied extensively, and some industrial issues should be also paid attention. Hot ductility should be considered to provide guidance for continuous casting or hot rolling, which is the front procedure of the whole manufacturing line. For medium Mn steel, Mn, Al, and Si elements are usually added during melting. The high amount of alloying elements may lead to problems during continuous casting. Some pilot production revealed that large edge cracks, transverse cracks, and interdendritic cracks occur during the straightening operation of medium Mn steel [10]. The addition of Al increases A_c3 (austenite finished transformation temperature), and the formation of ferrite film likely occurs during secondary cooling and straightening. Mn segregation is also inevitable for medium Mn steel. Mn-segregated bands deteriorate the mechanical properties [15]. The wide ferrite–austenite intercritical temperature range is significant for medium Mn steel, and the properties at the intercritical temperature range need to be clarified [16–17]. Currently, the continuous casting for medium Mn steel remains a huge challenge, and the processing parameters are uncertain. Studies rarely focused on the hot ductility of medium Mn steel.

The present study aims to understand the hot ductility of medium Mn steel. This study may serve as a guidance to avoid the formation of cracks and optimize the continuous casting conditions for medium Mn steel. The fracture behavior and mechanism of hot ductility evolution are clarified.

A 25 kg ingot was prepared by vacuum induction melting. The nominal chemical composition of the investigated medium Mn steel is Fe–9Mn–2Al–0.3C, and the measured chemical composition is given in Table 1. The real alloying elements of Mn, Al, and C are within the expected contents. The contents of P and S are 0.0075wt% and 0.0088wt%, respectively. The N and O contents are negligible below 50 ppm. The cast medium Mn steel features a dendrite structure. Cylindrical specimens were machined from the ingot for hot tensile tests. As shown in Fig. 1, the dimensions of the cylindrical specimens were 121.5 mm in length and 6 mm in diameter. A Gleeble3800 thermo-mechanical simulator was used to conduct hot tensile tests. Hot tensile experiments were conducted in accordance with the GB/T 228.2–2015 standard. Fig. 2 illustrates the thermo-mechanical curves of the hot tensile tests. For medium Mn steel, the initial microstructure characterization was performed. Specimens were heated to 1300°C and held for 90 s to simulate the initial state during continuous casting. Then, the specimens were cooled to 600–1300°C and held for 30 s. The specimens were strained until failure at a constant strain rate of 4 × 10⁻³ s⁻¹. Finally, the specimens were quenched to ambient temperature via an Ar stream system. Three repeated specimens were conducted for each condition to improve accuracy.

Figure 1.  Dimensions of the hot tensile specimen.

The reduction of area (RA) at the fracture was measured to quantify hot ductility. The RA can be expressed as follows:

where RA, S₀, and S₁ are the reduction of area, original cross section area, and fracture cross section area, respectively.

The fracture surface was observed under a Zeiss Merlin Compact field emission scanning electron microscope (FE-SEM). The cross-sectional fracture microstructures were also observed. The prior austenite grain sizes at the fracture, transition, and edge areas were quantified in accordance with the GB6394–2002 standard. The specimens were etched in a water bath via picric acid to reveal the austenite grain. The specimens were re-etched by 4vol% nital solution for microstructure observation. Microstructure observation was conducted under a ZEISS Imager M2m optical microscope and FE-SEM. Thermodynamic calculations of the studied Fe–0.3C–9Mn–2Al steel were conducted using Thermo-Calc software (2017a) with database TCFE9. The property diagram of each phase was presented.

Figure 2.  Schematic of thermo-mechanical curves of the hot tensile test. The insert shows the machining of the specimen from ingot.

The stress–strain curves of the investigated medium Mn steel are shown in Fig. 3. The tensile strength or the peak stress decreases with the increase in test temperature, which is a common phenomenon during the hot tensile test. The tensile strength is much higher than the yield strength in the specimens tested below 750°C, and the stress–strain curves demonstrate strong strain hardening behavior after yielding. Obvious yield point elongation or Lüders strain is found in the specimens tested at 600–720°C. Discontinuous yielding is not common during the hot tensile test, and the occurrence of yield point elongation will be discussed with microstructure evolution. The stress–strain curves of the specimens tested at 750–1300°C demonstrate a continuous yielding behavior, and no yielding point elongation could be observed. A stress plateau stage exists after reaching a peak stress. The stress–strain curves demonstrate a stress softening trend with the increase in test temperature. This behavior may be related to the activation of DRX. DRX could lead to dynamic stress softening during the flow curves, and the increase in test temperature contributes to DRX.

Figure 3.  Stress–strain curves of the investigated medium Mn steel at different temperatures.

The hot ductility is quantified by the data of RA. Fig. 4 shows the changes in RA value and peak stress at different temperatures. The peak stress shows an opposite change trend with the test temperature. The RA reaches the maximum value of 89% when tested at 1000°C. The RA value decreases when the test temperature increases up to 1050°C and then sharply decreases when the temperature exceeds 1200°C. When test temperature is lower than 1000°C, the RA shows a decreasing trend with the decrease in test temperature. A relatively rapid decrease in the RA could be observed between 800 and 900°C. Then, the RAs decrease slowly at 720 and 750°C. For specimens tested at 600, 650, or 700°C, the RA decreases rapidly and the minimum value is lower than 65% when tested at 650°C. The morphologies of the specimens after the hot tensile test are shown in Fig. 5. The macro morphologies clearly show the deformation region during the hot tensile test. Obvious necking can be found in the specimens tested between 600 and 1050°C. The macro morphologies show similar results to the hot ductility curves. Three types of the brittle zone are usually referred during the temperature range of continuous casting [6,18]. The first brittle zone is the brittleness caused by the solidification from liquid. For the investigated steel, the sharp decrease in RA above 1050°C results from the first brittle zone. The second brittle zone usually occurs when the strain rate is higher than 10⁻² s⁻¹. The third brittle zone is related to the ferrite film from prior austenite for carbon steels, and the control of the third brittle zone is the most important for crack propagation. The decrease in RA below 720°C may be due to the third brittle zone. Previous studies proposed that the critical value to estimate the crack sensitivity of steel continuous casting is about 40%–60% [19]. The cracking sensitivity may not be the same level for different types of steels. The hot ductility of the investigated steel seems moderate. However, cracking still occurs near the fracture surfaces.

Figure 4.  Changes in RA and peak stress at different temperatures.

Figure 5.  Morphologies of the specimens after the hot tensile test. The insert shows the observed cross section.

Fig. 6 shows the macro fracture morphology of the specimens tested at different temperatures. A small portion of the melt region exists at the edge area of each specimen, which is a common phenomenon during the hot tensile test [20]. Some river pattern and intergranular facets can be observed (black arrow) from the magnified view of specimen tested at 600°C (Fig. 7(a)). Small and shallow dimples also exist at the fracture surface (red arrow). The mixed brittle–ductile fracture feature is reflected by the coexistence of cleavage step, intergranular facet, and dimple at the surface. Liu et al. [21] pointed that hot ductility is related to the fracture type. The fracture surfaces are filled with large and deep dimples in the specimens tested at 720–1000°C, which can be observed directly from Figs. 6(d)–6(i). Fig. 7(b) depicts the fracture surface of the specimen at a higher magnification. The average width of the dimples is 163 μm. Overall, the average width of the dimples increases with the increase in test temperature. As compared with the results in Section 3.2, the change in RA is somewhat in proportion with the average width of the dimples. Fig. 8 shows the changes in the RA and average width of dimples at different temperatures. Specimens with large and deep dimples show high RA, which correspond to high hot ductility. With large and deep dimples, the interval of voids is also large. Defects can grow after nucleation. Therefore, crack propagation can be hindered in a certain degree. RA can be increased during the hot tensile test. When the test temperature exceeds 1050°C, the size of the dimples and the RA decrease.

Figure 6.  Macro fracture morphology of the specimens tested at different temperatures: (a) 600°C; (b) 650°C; (c) 700°C; (d) 720°C; (e) 750°C; (f) 800°C; (g) 900°C; (h) 1000°C; (i) 1050°C.

Figure 7.  Magnified view of the fracture surface tested at (a) 600°C and (b) 900°C.

Figure 8.  Changes in RA and average width of dimples at different temperatures.

The cross section after the hot tensile test is also observed near the fracture surface. During the hot tensile test, the deformation is mainly conducted at austenite state. The microstructure is worth observing. Figs. 9(a)–9(c) show the prior austenite grain size of the specimens (test temperature of 600°C) at the fracture, transition, and edge areas, respectively. The grain at the fracture area features an elongated morphology. For elongated grains, the average length and width are used to describe the grain size. The grain size is measured as 237 μm in length and 33 μm in width. The lath characterization reflects typical deformation grains without strain softening or recrystallization. Almost equiaxed grains can be observed at the transition and edge areas, and the grain size is as large as around 100 μm. The equivalent diameter is suitable for describing the equiaxed grains. The grains at the fracture in the specimens tested at 650–1050°C are shown in Figs. 9(d)–9(i). Elongated grains are obvious in the specimens tested at 650–720°C, whereas small equiaxed recrystallization grains are observed in the specimens tested at or above 750°C.

Figure 9.  Morphology of the cross section after hot tensile test at different areas and temperatures: (a) 600°C, fracture; (b) 600°C, transition; (c) 600°C, edge; (d) 650°C, fracture; (e) 700°C, fracture; (f) 720°C, fracture; (g) 750°C, fracture; (h) 1000°C, fracture; (i) 1050°C, fracture.

Table 2 summarizes the prior austenite grain size data of the specimens at different observed areas and different test temperatures. The change in grain at the fracture is obvious. The elongated grain is large in the specimens tested below 720°C. The finest grain size is found in the specimens tested at 750–1000°C, and the recrystallized grain shows an equiaxed shape. Almost all the grains at the transition and edge areas show an equiaxed shape. No obvious difference in grain size can be found between the transition and edge areas. During the hot tensile test, the deformation occurs near the fracture area. Dynamic recrystallization is activated with the deformation for specimens above 750°C. Therefore, fine equiaxed recrystallization grains can be obtained. Increasing the test temperature up to 1050°C leads to grain growth. The grain size at the fracture shows an increasing trend above 1000°C. At the transition and edge areas, almost no deformation is activated during hot deformation. No obvious change in grain size is found.

Thermal equilibrium simulation is conducted to understand the evolution of hot ductility. The phase fractions diagram of the Fe–0.3C–9Mn–2Al steel is shown in Fig. 10. After solidification, single austenite exists at the high temperature range. Austenite transforms into ferrite at around 762°C according to the thermodynamic calculation. The austenite–ferrite transformation remains in a wide temperature range, and the austenite–ferrite intercritical temperature range is crucial for medium Mn steel. The austenite–ferrite transformation starting (A₃) and finishing (A₁) temperatures are 765 and 514°C, respectively, according to the dilation experiment test. The results of the experiment and the thermodynamic calculation are similar. Cementite and other carbides precipitate when the temperature decreases below 650°C. Ferrite films and precipitations decrease hot ductility [22–23].

Figure 10.  Thermal equilibrium phase fraction diagram of the Fe–0.3C–9Mn–2Al steel.

Fig. 11 shows the microstructure of the specimen tested at 720°C. The martensite matrix is transformed from prior austenite during cooling after the tensile test. A small portion of ferrite also exists besides the martensite matrix. Large voids can be observed near the fracture surface. The zoom view shows that voids are prone to occur near the ferrite films. The ferrite films, especially those at the grain boundary, can lead to strain concentration during hot tensile test because they are softer than the austenite [24]. The soft ferrite is elongated by the strain concentration and its slide along the austenite–ferrite interface [25]. Thus, voids nucleate near the strain concentration, and the voids evolve into the source of fracturing. More ferrite is transformed from austenite with decreasing temperature. Higher fraction of ferrite brings more austenite–ferrite interface [21]. The hot ductility shows a rapid decreasing trend at 600–700°C. On the one hand, cementite precipitation decreases ductility [26]. On the other hand, the test temperature is lower than the dynamic recrystallization temperature [27–28]. Dynamic recrystallization cannot be activated during hot tensile test. These two aspects lead to the poor hot ductility at 600 and 650°C. In addition, yield point elongation occurs when hot tensile is tested at 600–720°C. In general, the strain–stress curves during hot tensile test show a smooth moving behavior with strain hardening, stress plateau, or softening [6]. Yield point elongation is a plastic instability phenomenon that leads to the occurrence of Lüders bands [29–30]. Plastic instability at ambient temperature restricts the application of medium Mn steel. In addition, it negatively affects the forming properties and the surface quality. Yield point elongation is related to the pinning effect between substitutional C atoms and mobile dislocations at the ferrite [31–32]. Discontinuous yielding occurs at the wide austenite–ferrite intercritical temperature range during hot tensile test. For the current medium Mn steel, the ferrite might be responsible for the discontinuous yielding behavior at high temperature, and further direct explanation will be provided.

Figure 11.  Microstructure (a) and magnified view (b) of the fractured specimen tested at 720°C.

The hot ductility of an Fe–0.3C–9Mn–2Al medium Mn steel is studied. The fracture behavior during hot tensile test is clarified. The following conclusions are drawn.

(1) The hot ductility decreases with decreasing temperature from 1000°C. The RA decreases sharply, and the value is lower than 65% for specimens tested at 650°C.

(2) The fracture belongs to ductile failure for specimens tested between 720 and 1000°C. Large and deep dimples can delay crack propagation. The change in the average width of dimples is in positive proportion with the change in RA. The specimens tested at 600–700°C show a mixed brittle–ductile fracture feature.

(3) The wide austenite–ferrite intercritical temperature range is crucial for the hot ductility of medium Mn steel. The formation of ferrite film on austenite grain boundaries lead to the strain concentration. Voids occur near the strain concentration, and the voids evolve into the source of fracturing. Yield point elongation occurs at the austenite–ferrite intercritical temperature range during the hot tensile test.

This work was financially supported by the Fundamental Research Funds for the Central Universities, China (Nos. FRF-TP-18-039A1 and FRF-IDRY-19-013) and the China Postdoctoral Science Foundation (No. 2019M650482).