The high-temperature mechanical properties of twinning-induced plasticity (TWIP) steel with 0.05 % C, 25 % Mn, 3 % Al, 3 % Si have been investigated using the GLEEBLE 3500 machine. The result shows that the zero ductility temperature and the zero strength temperature of the TWIP steel are measured at 1,225 °C and 1,275 °C, respectively. The brittleness temperature interval I is from 1,200 °C to the melting point, and the brittleness temperature interval III is from 650 °C to 800 °C. The tensile fracture has been examined using the scanning electron microscope, optical microscope and electron backscatter diffraction to determine the fracture mechanisms. The result shows that the twin is not the main influencing factor of the high-temperature plasticity of TWIP steel. Instead, the degree of dynamic recrystallization determines its high-temperature plasticity. A small number of AlN particles are found near the fractures, but these particles are so coarse, therefore, have no influence on the brittle fracture, and ferrite transformation and work hardening are the main reasons that cause the brittle fracture.
Twinning-induced plasticity (TWIP) steels are single austenite at room temperature, which are ideal automotive steels for a new generation with high strength, high plasticity and high strain hardening [1, 2].When Fe-Mn-Si-Al TWIP steel contains 25 % Mn, more than 3 % Al, and 2–3 % Si, stable austenitic organization and suitable stacking fault energy will be obtained at low temperature, which are essential for the arise of its TWIP effect [3, 4]. The TWIP steel is well known for its excellent property also due to intense strain hardening up to large values of strain , and the product of elongation and tensile strength is more than 60,000 MPa %. In addition, TWIP steel also has a high capacity of energy absorption and no fracture appearance transition temperature (FATT) .
The TWIP steel is still in its early stage of commercial production, and most domestic studies on TWIP steel are yet at the laboratory stage [7, 8, 9]. To the authors’ knowledge, studies on high-temperature mechanical properties of as-cast Fe-25Mn-3Si-3Al TWIP steel produced by steel plant have not yet been reported in the literature. In this article, the TWIP steel is produced by electroslag refining (ESR) process from a steel plant. Since ESR process owns ingot clean and compact effects, therefore, the quality of this steel is better than the laboratory produced ones. Among all kinds of defects of as-cast steel, cracks take account of more than 50 % , and high-temperature mechanical properties of steels have a strong relationship with the formation of the crack. In this article, the high-temperature mechanical properties of low-carbon Fe-Mn-Si-Al TWIP steel have been investigated using the GLEEBLE 3500 machine, and methods such as scanning electron microscope (SEM), electron backscatter diffraction (EBSD) are employed to analyze fracture mechanisms. This research can provide theoretical basis for the parameter control of the TWIP steel in relating aspects of continuous casting and hot working in the future.
The process of induction furnace–argon oxygen decarburization (AOD) smelting–molding–ESR has adopted to produce low-carbon Fe-Mn-Si-Al TWIP steel in a steel plant, and the chemical composition of the steel used in this study is given in Table 1. This study selected a cylindrical casting ingot of Φ1,080 mm after ESR. Cylindrical specimens with a length of 120 mm and a diameter of 10 mm were cut parallel to the axis at the 1/2 length of the ingot, which were in the columnar crystal area. The thread and thread pitch of the end of specimens are 12 and 1.5 mm, respectively. In addition, the specimen must be manufactured with superior surface finish and a small range of tolerance. The pattern and figures of the specimen are shown in Figure 1.
In this work, high-temperature tensile test is conducted utilizing the GLEEBLE 3500 testing machine. The reduction in area (ψ) and the tensile strength (Rm) with the change of temperature can be obtained through data processing in the experiment. Based on that, brittleness temperature intervals could be determined. Additionally, the zero ductility temperature (ZDT) and zero strength temperature (ZST) of the TWIP steel, which are the corresponding values of temperature when ψ and Rm are zero [11, 12], could also be calculated using figures described above.
In this experiment, the machine system was filled with Ar gas after being evacuated. The specimens were clamped horizontally and the effective length of heating was 20 mm. Since the deformation simulation of each part in the process of continuous casting production took 1×10−3/s as the representative strain rate , the tests were conducted at the same rate. The specimens were reheated to 1,250 °C at a rate of 10 °C/s for soaking for 3 min and finally cooled or heated with a rate of 5 °C/s to different test temperatures between 650 °C and 1,275 °C, at which they were maintained for 1 min to stabilize the temperature before the tensile tests. Tensile tests were conducted every 50 °C from 650 °C to 1,200 °C and every 25 °C from 1,200 °C to 1,275 °C. The specimens were quickly cooled down to room temperature with compressed air flow after rupture. The sketch map of temperature and deformation control during tensile tests is shown in Figure 2. Temperature was measured using a platinum–rhodium thermocouple spot-welded on the surface in the middle of the sample.
The fracture morphologies were observed utilizing the ZEISS ULTRA 55 SEM, and inclusions were analyzed by means of SEM-EDS. Some fractures of typical samples were cut into two parts along the drawing direction and etched with 4 % nital after grinding and polishing. Then, by means of an optical microscope, the microstructures near the fractures were analyzed. The deformed specimens were then cut into slice with a length of 7 mm and a thickness of 2 mm along the axis of extension. The electrolytic polishing was applied to the slices with the solution of 20 % perchloric acid and alcohol. After that, the specimens were studied using SEM-EBSD.
Variation of maximum tensile stress at different temperatures between 650 °C and 1,275 °C for the specimens is displayed in Figure 3. Rm equals to zero when the tensile test temperature is 1,275 °C. Therefore, the ZST is about 1,275 °C. It has been found that the tensile strength of the specimens increases gradually as temperature decreases. When the temperature is respectively high, Rm value rises slightly as the decrease of the temperature, with 4.8 MPa at 1,250 °C and 36.3 MPa at 1,050 °C. However, when the temperature drops down 1,000 °C, the tensile strength increases sharply. It is worth noting that the specimen owns the strongest tensile strength (472.5 MPa) in this experiment at 650 °C.
The essence of metal plastic deformation is the movement of dislocation . With the decrease of temperature, the force between the atoms is enhanced, and the atomic energy is reduced and the diffusion ability is weakened, which all inhibit the dislocation motion. Therefore, the tensile strength of the specimens increases as the temperature decreases.
The hot plasticity values at different temperatures are presented in Figure 4.
As shown in Figure 4, the reduction in area drops down to zero when the temperature reaches 1,225 °C, 1,250 °C and 1,275 °C, which indicates that ZDT is about 1,225 °C. The plasticity of TWIP steel increases rapidly as temperature decreases, and ψ value is 70.8 % at 1,200 °C, and the TWIP steel has very good plasticity from 900 °C to 1,000 °C, with the ψ value more than 79 % and a maximum value of 87.9 %; When the temperature is below 900 °C, the plasticity of TWIP steel reduces as temperature decreases, and the minimum ψ value is 14.2 % at 750 °C. With the temperature further decreasing, the plasticity of TWIP steel improves slightly and the ψ value is around 20 %.
From the results, the ZST and ZDT of the low-carbon TWIP steel are measured at 1,275 °C and 1,225 °C, respectively, which are lower than other steels. When the ZST and ZDT are lower, the steel’s crack sensitivity is higher, therefore, the TWIP steel has a high crack sensitivity. On the other hand, the brittleness temperature interval at the solidification front of steels is from ZST to ZDT [15, 16], and the range of temperature interval (ΔT=ZST–ZDT) is also a measure about internal crack tendency of steels at solidification front. The ΔT of TWIP steel in this research is about 50 °C, which is wider compared with other steels. It also indicates that this TWIP steel has a stronger high-temperature crack sensitivity.
A research  points out that the crack of steel is not easy to appear when ψ value is greater than 60 %, and the crack sensitivity will greatly increase when ψ value is less than 40 % [18, 19]. Therefore, in this paper, the brittle temperature range of the TWIP steel is also judged as the temperature domain with reduction in area of fracture less than 40 %. Thus, the brittleness temperature interval I is from 1,200 °C to the melting point, and the brittleness temperature interval III is from 650 °C to 800 °C.
The true stress–strain curves for the specimens at different temperatures are displayed in Figure 5. At 750 °C, the stress enhances with the climb of the strain, but the growth rate declines gradually, which is shown in Figure 5(a). During the deformation process, the work hardening rate reduces continuously, but the work hardening effect in this procedure is always greater than the softening influence of dynamic recovery. The dynamic recrystallization (DRX) curve reflects great difference compared with the stress-strain curve of the dynamic recovery. The softening influence in the DRX procedure could be greater than the work hardening effect when strain variables are defined at certain values. It can be found that DRX of the TWIP steel occurs obviously when temperature is or higher than 800 °C. The process of DRX is completed by the formation and growth of recrystallization crystal nucleus. The increase rate in dislocation density is relatively slow since the experiment is conducted at lower strain rate and high temperature. Therefore, there must be work hardening again after DRX that the recrystallization nucleation can be conducted once again. The softening of the DRX is out of sync with work hardening, which makes DRX going on discontinuously, and thus jitter and peaks appear in the stress–strain curve, which is shown in Figure 5(b) and 5(c).
The fracture morphology in Figure 6(a) shows an intergranular fracture shape like rock candy at 1,250 °C, which may be caused by overburning. In Figure 6(b), the fracture of the steel is changed from the manner of intergranular brittle fracture to the manner of intergranular brittle fracture mixed with dimple fracture at 1,100 °C. The reduction in area increases dramatically, and ψ value is 51.7 %. The fracture morphology is a small amount of large size dimples and plane of no feature when the temperature drops to 900 °C in Figure 6(c). The typical dimple fracture appears when ψ value is 87.9 %, which implies the best plasticity. When the temperature is below 800 °C, Figure 6(d), 6(e) and 6(f) clearly reveal that intergranular cracking and transgranular cracking are both present on the tensile fractures, which are typical brittle fractures. Additionally, the ψ values of these specimens are below 20 %.
Figure 7 shows the microstructures of the TWIP steel near the tensile fracture at different temperatures. Figure 7(a) and 7(b) shows clearly that DRX phenomena in the steel are rather distinct when temperature is higher than 800 °C, and the area of DRX at 1,000 °C is larger than that at 800 °C. However, in brittleness temperature interval III, it can be found from Figure 7(c) that phenomenon of DRX and phase change is not obvious. There is much acicular austenite in original austenite grain, and some cracks along acicular constituent. Usually, the austenitic organization is granular. However, the acicular austenite can be found during the austenitizing courses of low-carbon steel at low speed . Si has a strong promoting effect for the formation of acicular austenite, but DXR will hinder the formation of acicular austenite .
The real grain size near the tensile fracture was measured by means of EBSD. Surprisingly, the reduction in area and reciprocal of average grain size varies with temperature at the same trend from 700 °C to 1,000 °C, as shown in Figure 8. Therefore, it can be concluded that high-temperature plasticity of the TWIP steel is closely connected with the degree of recrystallization.
From the experiment, more deformation twins of TWIP steel appear with the increase of the plastic deformation. Twins play a role in refining TWIP matrix and diverting the crystal orientation, and thus increase in the ratio of plastic deformation. Meanwhile, the plasticity of TWIP steel is improved. Twinning deformation is the main deformation manner of TWIP steel . However, none of the deformation twins is found at the tensile fracture in the plasticity temperature interval. Instead, a few annealing twins have been found at the fracture, as shown in Figure 9, where the red lines represent the twin boundaries. When the temperature increases, the stacking fault energy enhances, so the formation of deformation twins is limited. Qin et al.  investigated the deformation mechanism of TWIP steel and found that when the temperature is higher than 600 °C, there is no deformation twin. Therefore, it can be concluded that twins are not the determined factor of the outstanding plasticity of the TWIP steel appearing in high-temperature tensile test.
The process of DRX is conducted in the good plasticity temperature interval, and the essence of grain growth is the grain boundary migration in crystal structure. Therefore, the grain boundary migration occurs continuously with the process of DRX. During this procedure, the rate of grain boundary migration exceeds the rate of grain boundary sliding in this temperature interval; therefore, the formed micro-cracks cannot aggregate and enlarge on the grain boundary, which means that they are all restricted within the grain. But the stress concentration at the tip exacerbates the growth of cracks and further breaks up the grains and connecting with each other, and eventually results in the fracture of the specimen. The best plasticity temperature interval of test steel is 900 °C~1,000 °C, and in this interval the temperature is beneficial for grain boundary migration. Thus, the embrittlement is greatly retarded, and specimens show transgranular cracking with good plasticity. Therefore, the process of DRX plays a decisive role in the formation of good plasticity.
Fine AlN particles not only act as a source of stress concentration to form micropores, but also take pinning effect on grain boundaries, so that the process of DRX is suppressed and the plasticity of steel is deteriorating. In this study, transmission electron microscope (TEM) is utilized to analyze nanoprecipitations, but fine AlN particles are not found on austenite grain boundaries in the brittleness temperature interval III. Su et al.  thought that the very high Al addition favors coarse AlN particles. Zhuang et al.  studied the nonmetallic inclusions of the same TWIP steel and found that AlN could precipitate and grow up in liquid TWIP metal. Thus, AlN particles in the test steel are bigger than others and the sizes are between 7 and 15 μm, as shown in Figure 10, which are found near the tensile fracture with SEM-EDS. The chemical composition of AlN particles in Figure 10 is given in Table 2. The effect of bulky AlN particles is different from that of fine AlN particles, and they have hardly any pinning effect on grain boundaries. The DRX process occurs obviously at 800 °C and 900 °C, which shows that small number of large AlN particles cannot pin grain boundaries effectively, which leads to the conclusion that AlN particles in the studied TWIP steel is not the cause of the embrittlement in the brittleness temperature interval III.
|Number||Chemical composition/mass %|
Dai et al.  investigated Fe-XMn-3Si-3Al TWIP steel and found that the transformation induced plasticity (TRIP) effect did not occur when the content of Mn reached 25 %, and after deformation this TWIP steel was still single austenite at room temperature. However, in the experiment of this paper, a small amount of ferrite is found on the fracture when the tensile test is carried out at the temperature from 650 °C to 800 °C. In Figure 11, the white color represents the austenite, red represents ferrite and blue represents low-angle grain boundary. The contents of ferrite are about 1 % in both samples, and the content of ferrite was so little that the phase change was difficult to be detected accurately in x-ray diffraction (XRD) experiment. As of now, no report has been made about ferrite transformation in Fe-25Mn-3Si-3Al TWIP steel in the high-temperature tensile test. There are many low-angle boundaries because of the formation of acicular austenite at 700 °C, as shown in Figure 11(a), and some ferrite in low-angle boundaries is more fine and discrete than that at 800 °C.
In the temperature interval from 650 °C to 800 °C, deformation increases the dislocation density and exaggerates the carbon diffusion. Therefore, the driving force of phase transformation from the austenite to ferrite is improved, so the precipitation of fine ferrite is induced on both grain boundary and inside. The ferrite weakens the strength in the precipitation area, and the smaller the precipitates are, the easier the stress concentration is. On the other hand, TWIP steel has strong strain-hardening characterization , which lead to high tensile strength of the tested TWIP steel. The minimum tensile strength reflected in the temperature interval mentioned above is 200 MPa, which is shown in Figure 3. The result of the factors is that fractures generate and grow along weakening area, causing brittle cracking. Therefore, the strong strain hardening and deformation-induced ferrite transformation are the main causes, which make the TWIP steel brittle fracture in the brittleness temperature interval III.
The deformation-induced ferrite transformation is a process of dynamic nucleation and metadynamic phase transformation at a definite temperature , so the ferrite transformation has not been found in the test of heat treatment and deformation at room temperature reported in the past studies of this TWIP steel. Yang et al.  found that optimization of deformation temperature and deformation rate could control deformation-induced ferrite transformation effectively, but the cooling process was not a decisive influence for ferrite precipitation. Therefore, the ferrite precipitation can be effectively controlled by application reasonable process and the plasticity of this TWIP steel can be improved in brittleness temperature interval III.
The main results and conclusions from the work are summarized as follows:
The ZST of low-carbon Fe-Mn-Si-Al TWIP steel is about 1,275 °C and the ZDT is about 1,225 °C. The ZST and ZST are lower than other steels, and the range of temperature interval (ΔT=ZST–ZDT) is wide at solidification front, so this TWIP steel has a higher crack sensitivity.
The brittleness temperature interval I of low-carbon Fe-Mn-Si-Al TWIP steel is from 1,200 °C to the melting point, and the brittleness temperature interval III is from 650 °C to 800 °C. The continuous straightening process of continuous casting and hot working process should avoid the brittleness temperature interval.
In the good plasticity temperature interval, the twin is not the reason of the high plasticity of the low-carbon Fe-Mn-Si-Al TWIP steel. The DRX process is the decisive factor for the formation of good plasticity, and the high temperature plasticity of the TWIP steel has a close relationship with the degree of DRX.
In the brittleness temperature interval III of low-carbon Fe-Mn-Si-Al TWIP steel, the small number of AlN particles are not the cause of embrittlement, while the strong strain hardening and deformation-induced ferrite transformation on the grain boundary are the main causes of the TWIP steel brittle fracture. The existence of intragranular ferrite may play a role in promoting transgranular brittle cracking. However, further work is required to prove this conclusion.
Funding source: National Natural Science Foundation of China
Award Identifier / Grant number: 51350110515
Award Identifier / Grant number: 51574022
 C. Scott, S. Allain, M. Faral and N. Guelton, Metall. Res. Tech., 103 (2006) 293–302. Search in Google Scholar
 O. Grässel, L. Krüger, G. Frommeyer and L.W. Meyer, Int. J. Plast., 16 (2000) 1391–1409. Search in Google Scholar
 O. Grässel, G. Frommeyer, C. Derder and H. Hofmann, J. Phys. IV France, 7 (1997) 383–388. Search in Google Scholar
 O. Grässel and G. Frommeyer, Mater. Sci. Technol., 14 (1998) 1213–1217. Search in Google Scholar
 O. Bouaziz, S. Allain and Y. Estrin, Scripta Mater., 62 (2010) 713–715. Search in Google Scholar
 Y.N. Wang, Q. Li, Y.L. He, R.Y. Fu and L. Li, Heat Treat., 21 (2006) 25–29. Search in Google Scholar
 F.C. Ma, Bao Steel Technol. Res., 6 (2008) 62–66. Search in Google Scholar
 P. La, Z.W. Du, Y. Ji and J.Q. Zhang, China Metall., 24 (2014) 7–16. Search in Google Scholar
 J.H. Liu, C.L. Zhuang, S.Q. Li, Q.L. Zhang and C. Lei, The National Steel Academic Conference Proceedings, Xi’an (2014). Search in Google Scholar
 K.K. Cai and Z.J. Dang, J. Univ. Sci. Technol. Beijing, 15 (1993) 52. Search in Google Scholar
 T. Nakagawa, T. Umeda and J. Murata, ISIJ Int., 35 (1995) 723–729. Search in Google Scholar
 C.H. Yu, M. Suzuki, H. Shibata and T. Emi, ISIJ Int., 36 (1996) 159–162. Search in Google Scholar
 X.H. Wang, Ferrous Metallurgy-Steelmaking, Higher Education Press, Beijing (2007). Search in Google Scholar
 Z.C. Liu, Metallography and Heat Treatment, Chemical Industry Press, Beijing (2008). Search in Google Scholar
 X.D. Kang, H.C. Di, X.M. Zhang, G.D. Wang and X.H. Liu, Res. Iron Steel, 3 (2003) 32–34. Search in Google Scholar
 J.F. Xu, S.P. He, X.P. Jiang, T. Wu and Q. Wang, ISIJ Int., 53 (2013) 1812–1817. Search in Google Scholar
 H.G. Suzuki, S. Nishimura and J. Imamura, Tetsu-to-Hagane, 67 (1981) 1180–1189. Search in Google Scholar
 X.H. Wang, G.S. Zhu, H.X. Yu and W.J. Wang, J. Univ. Sci. Technol. Beijing, 27 (2005) 545–548. Search in Google Scholar
 B. Mintz, ISIJ Int., 39 (1999) 833–855. Search in Google Scholar
 Y.P. Kang, Metal Solid-State Phase Transformation and Application, Chemical Industry Press, Beijing (2010). Search in Google Scholar
 S.C. Liu, G.Z. Li, S.S. Wang and Z.F. Qi, Trans. Met. Heat Treat., 13 (1992) 40–44. Search in Google Scholar
 Z.L. Mi, D. Tang, H.T. Jiang, Y.J. Dai, H.Q. Wang, S.S. Li and J. Wang, J. Univ. Sci. Technol. Beijing, 29 (2007) 1021–1023. Search in Google Scholar
 X.M. Qin, L.Q. Chen, H.S. Di and W. Deng, Acta Metall. Sinica, 47 (2011) 1117–1122. Search in Google Scholar
 H. Su, W.D. Gunawadarna, A. Tuling and B. Mintz, Mater. Sci. Technol., 23 (2007) 1357–1365. Search in Google Scholar
 C.L. Zhuang, J.H. Liu, Z.L. Mi, H.T. Jiang, D. Tang and G.X. Wang, Steel Res. Int., 85 (2014) 1432–1439. Search in Google Scholar
 Y.J. Dai, D. Tang, Z.L. Mi, H.T. Jiang and J.C. Lu, Mater. Eng., 7 (2009) 39–42. Search in Google Scholar
 A. Khosravifarda, A.S. Hamadab, M.M. Moshksara, R. Ebrahim, D.A. Porte and L.P. Karjalainen, Mater. Sci. Eng. A, 582 (2013) 15–21. Search in Google Scholar
 H.S. Zhao, Bao Steel Technol., 1 (2012) 48–53. Search in Google Scholar
 Z.M. Yang, Y. Zhao, Y.Z. Wang, Y.M. Che and Y.W. Ma, Acta Metall. Sinica, 36 (2000) 1061–1066. Search in Google Scholar
© 2017 Walter de Gruyter GmbH, Berlin/Boston
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.