Characterization and Control of Secondary Phase Precipitation of Nb–V–Ti Microalloyed Steel during Continuous Casting Process

Abstract

The strand surface and subsurface cracks could be prevented through the control of the strand surface microstructure, which correlates with the precipitation behavior of carbonitrides in the microalloyed steel. In this study, the carbonitride precipitation behavior was characterized in-situ with a high-temperature confocal laser scanning microscope, and the effects of cooling rate on the morphology and distribution of precipitates was investigated. The results show that the carbonitride precipitation process is usually accompanied by the formation of dark particles due to the volume expansion of the solute depletion region. The evolution of dark particles suggests that carbonitrides mainly precipitate between 910°C and 1085°C, and the “fast-growing region” ranges from 910°C to 960°C. As the cooling rate increases, the size and volume fraction of carbonitrides decrease. Meanwhile, the nucleation location changes from grain boundary to grain interior. To quantify the pinning capacity of the carbonitrides, the pinning force factor σ is defined according to the classical Gladman equation, which is a power function of the cooling rate, namely σ=0.43−0.1(1+2.76Vc^3.48)⁻¹. Combining the above research, a new secondary cooling method is proposed and has been applied on an actual caster, which improves the crack resistance of the bloom surface microstructures.

1. Introduction

Secondary phase particles can enhance the strength of microalloyed steel through fine grained strengthening as well as precipitate strengthening. However, when the precipitation behavior of secondary phase particles cannot be controlled properly in the continuous casting process, a large amount of chain-like carbonitrides can precipitate along austenite grain boundaries, leading to the increase of the crack sensitivity of the strand, and even cause the surface and subsurface cracks.^1,2,3,4,5) To design suitable continuous casting parameters, the carbonitride precipitation behavior during the solidification process should be investigated systematically.

Cooling rate could influence the category, dimension and morphology of the precipitates, which consequently has an influence on the microstructure and mechanical properties of the steel.^6,7,8) Many investigations have been made on the relationship between the cooling rate and the precipitation behavior of the secondary phase in low alloy steels. Chen et al.⁹⁾ found that the size of the austenite grains and TiC in the Ti–Mo steel gradually decreased with increasing undercooling. Moreover, the nucleation location of TiC can be changed from grain boundary to grain interior. According to the research from Luo et al.,¹⁰⁾ carbides in M42 high-speed steel grew along the grain boundary in a network distribution under low cooling rate (<1°C/s), whereas the nucleation and growth of carbides was inhibited under high cooling rate (1–5°C/s). Similar conclusions have been drawn in the microalloyed steels. Dou et al.^11,12) and Ma et al.¹³⁾ studied the effects of cooling rate on the secondary phase precipitation and proeutectoid phase transformation in microalloyed steel containing Nb, V and Ti. The results indicated that the intensive cooling can promote the dispersion of the second phase particles inside austenite grains and strengthen the strand surface microstructure. This phenomenon has also been confirmed in the earlier studies.^14,15)

Generally, the precipitation nose temperature of carbonitride ranges from 800–1000°C.^16,17) The temperature of continuous casting strand out of the mold is higher than 1000°C. Hence, the precipitation behavior of carbonitrides is mainly affected by the cooling condition in the secondary cooling zones (SCZ). Production practices show that surface cracks cannot be effectively prevented by adjusting the secondary cooling intensity to keep the strand surface temperature away from the low ductility temperature zone during bending or straightening process.^12,18,19,20) Therefore, Kato et al.¹⁹⁾ proposed the surface structure control (SSC) cooling method to control the slab surface microstructure and secondary phase precipitation behavior of microalloyed steel. Nevertheless, some detailed operating parameters, such as the cooling rate and temperature range, are still unclear for the reasonable adoption in production.²⁰⁾ Therefore, it is necessary to develop a secondary cooling method matching with the solidification characteristics of the steel to reduce the surface cracks.

In this paper, a high-temperature confocal laser scanning microscope (HTCLSM) was used to characterize the precipitation behavior of the secondary phase. Besides, Field emission scanning electron microscopy (FESEM) were employed to observe the size and volume fraction of the second phase particles at different cooling rates. Ultimately, the temperature range and cooling rate for controlling the carbonitride precipitation during the continuous casting process were obtained, and a new cooling method was proposed to improve the crack resistance of bloom surface microstructures. It is encouraging that the subsurface cracks of the continuous casting bloom were well controlled through plant trials.

2. Materials and Methods

A HTCLSM was employed for an in-situ observation of the solidification of SG02 steel. SG02 steel belongs to non-quenched and tempered steel, which is strengthened by adding V, Ti and Nb microalloyed elements into carbon manganese steel. Chemical composition of the steel is listed in Table 1. Metallurgical specimens used in this study were extracted from the columnar crystal zone of a defect-free bloom with a section size of 220 mm×220 mm, as shown in Fig. 1. Specimen was machined into a cylinder with a size of φ5 mm×3 mm, and then put into an alumina crucible for experiment. The center segregation and porosity of the bloom were avoided during sampling, and the as-cast microstructure of each specimen was similar.

Table 1. Chemical composition of the SG02 steel.

Composition	C	Si	Mn	P	S	N	V	Nb	Ti
Mass fraction (%)	0.43	0.45	1.41	0.01	0.013	0.012	0.07	0.017	0.015

Fig. 1.

The schematic diagram of sampling location. (Online version in color.)

Specimens were polished to remove the oxide layer, which were further cleaned in an ultrasonic cleaning instrument for 10 min. Subsequently, specimens were put into the heating furnace of HTCLSM, and then the experiments were carried out under the thermal regime, as shown in Fig. 2. Specimens were heated from room temperature to 1480°C at a heating rate of 5°C/s and then held isothermally for 15 minutes to dissolve the carbonitrides. After that, the specimens were cooled to 25°C at five different cooling rates of 0.1°C/s, 0.5°C/s, 1°C/s, 3°C/s, and 5°C/s to observe the effects of cooling rate on the precipitation behavior of the carbonitrides. During the experiments, specimens were heated and cooled in an alumina crucible under an ultra-high purity inert atmosphere (99.999% Ar), and the temperature was measured by the thermocouple placed at the crucible holder. Details of the principle and the method of operation are the same as those described by Dou et al.,¹²⁾ Griesser et al.²¹⁾ and Kimura et al.²²⁾

Fig. 2.

Schematic illustration of the heat cycle in the experiment.

3. Results and Discussion

To control the precipitation behavior of carbonitrides in continuous casting process, detailed operating parameters such as the precipitation temperature and reasonable cooling rate should be clarified. Therefore, the precipitation behavior of carbonitrides at different temperatures and cooling rates should be investigated.

3.1. In-situ Characterization of Secondary Phase Precipitation

Thermodynamic or kinetic calculations are commonly applied to obtain the precipitation temperature of the carbonitrides. However, due to the difficulty of obtaining the phase transition parameters accurately, the calculated results often deviate from the actual results. In contrast, in-situ characterization of carbonitride precipitation is the most direct and effective method. To simulate the secondary cooling condition during the continuous casting process of the microalloyed steel, a HTCLSM was employed to investigate the carbonitide precipitation behavior at the cooling rate of 0.1°C/s.

The microstructural evolution of the SG02 steel specimen is displayed in Fig. 3. In-situ observation results illustrate that a few dark particles appear near the austenite grain boundary when the temperature decreases from 1480°C to 1085°C (Fig. 3(a)). As is found by Slater et al.²³⁾ and other scholars,^12,24) the dark particles are the direct signs of the carbonitrides. Segregation of solute atoms usually occurs at the austenite grain boundary, which accelerates the precipitation of the carbonitrides and dark particles. However, the internal morphology of grain boundaries cannot be clearly observed due to the HTCLSM imaging and experimental conditions. Similar experimental phenomena were reported in previous in-situ observations.^9,22,25,26) Therefore, dark particles were first found near the austenite grain boundary. Upon further cooling, dark particles increase rapidly between 1085 and 910°C. Although nanoscale carbonitrides cannot be captured by HTCLSM, larger size dark particles can be easily identified. After that, the proeutectoid phase transformation occurs between 723°C and 665°C, which can be seen in the Figs. 3(d)–2(f). The surface temperature of the SG02 steel bloom is generally above 900°C at the straightening point, so γ→α phase transformation is not the major reason for the crack formation in the continuous casting process.

Fig. 3.

In-situ observation of the microstructural evolution by HTCLSM (Cooling rate is 0.1°C/s; (a) 1085°C; (b) 1030°C; (c) 910°C; (d) 723°C; (e) 690°C; (f) 665°C). (Online version in color.)

Thermo-Calc software is used to calculate the phase equilibrium diagram of the SG02 steel, as shown in Fig. 4. It is found that Nb(C,N) starts to precipitate at 1146°C, which is close to the temperature observed in-situ. Although the TiN precipitates as early as 1432°C, it is more likely to precipitate at grain boundary at high temperature. Parker et al.¹⁶⁾ and Xie et al.¹⁷⁾ obtained the PTT curve of carbonitride precipitation in Nb-containing steel through kinetic calculation. The results indicated that the initial precipitation temperature of carbonitrides was about 1100°C, and the fastest precipitation temperature was around 900–950°C, which were consistent with the in-situ observation by HTCLSM. Liu²⁷⁾ also discovered some convex dark particles on the surface of the Ti microalloyed steel via in situ observation, and found that beneath the dark particle was Ti(C,N). Hence, it can be inferred that carbonitride precipitation induces the formation of dark particles. The schematic illustration of dark particle formation process is summarized in Fig. 5.

Fig. 4.

Phase equilibrium diagram of the SG02 steel calculated using Thermo-Calc software. (Online version in color.)

Fig. 5.

Schematic illustration of the dark particle formation. (Online version in color.)

In stage 1, microalloying elements begin to converge with carbon and nitrogen at the early stage of carbonitride formation, which promotes the formation of the depletion region of solute atoms. Due to the existence of concentration gradient, volume expansion occurs near the depletion region and small bulges are formed.^28,29)

In stage 2, the precipitation behavior of carbonitrides promotes the formation of dark particles. Multiple small bulges around the carbonitrides stack together to form larger bulges that can be captured by HTCLSM.

The second phase particles are more likely to precipitate at austenite grain boundaries, at where the depletion region of carbon makes the austenite unstable, enhancing γ→α phase transformation and the formation of ferrite.^30,31,32) Scholars generally believe that carbonitrides as inoculant particles promote the nucleation of proeutectoid ferrites.^11,13)

To investigate the relation between carbonitride precipitation and temperature, the number of dark particles is regressed and given in Fig. 6. The Image Pro Plus (IPP) software was used to calculate the number of dark particles within austenite internal 100 μm² area at different temperatures. Correspondingly, the number within 1 mm² area can be approximately evaluated. According to the statistical results, the quantitative relationship between the number of dark particles and temperature was obtained by nonlinear regression fitting. It can be discovered that dark particles begin to appear at about 1085°C, and the number is only 1.7×10³/mm². When the temperature is between 960 and 910°C, the “fast-growing region” of the dark particles appears, and the number increases sharply from 1.4×10⁴/mm² to 8.8×10⁴/mm², accounting for about 78% of all the dark particles. The discovery of the “fast-growing region” is helpful to accurately control the precipitation behavior of the secondary phase during continuous casting process. When the temperature ranges from 910°C to 800°C, the number of the dark particles increases slowly. Carbonitride precipitation belongs to typical diffusion phase transition. The phase transition velocity increases with the undercooling near the equilibrium temperature. However, when the undercooling increases to a certain extent, the phase transition velocity slows down due to the decrease of atomic diffusion ability. Therefore, the existence of “fast-growing region” conforms to the secondary phase nucleation kinetics. According to the equilibrium calculation (Fig. 4), Nb(C,N) and V(C,N) begin to precipitate at 1146°C and 991°C, respectively. With the decrease of temperature, the number of carbonitrides increase rapidly in a short time. Upon further cooling, the number increases slowly until about 800°C. Therefore, the quantitative relationship obtained by in situ observation is in agreement with the equilibrium calculation.

Fig. 6.

Number of dark particles at different temperatures. (Online version in color.)

Through the nonlinear regression of the statistical data, the quantitative relation between the number of dark particles and temperature could be expressed by Eq. (1). According to the classical Avrami equation (Eq. (2)), the law of phase transformation against time is that the degree of phase transformation increases slowly at the beginning, and then increases rapidly after a certain period of time, finally tends to be flat. Therefore, Eq. (1) proposed in this paper conforms to the Avrami equation. In summary, when the strand surface temperature is between 1085°C and 910°C, especially in the“fast-growing region” of 960–910°C, it is the best time to control the precipitation of carbonitides in the continuous casting process.   

$$N=3\mathrm{\hspace{0.17em} }462.6+90\mathrm{\hspace{0.17em} }207.9{(1+e^{(0.084T-78.786)})}^{-1}$$(1)

  

$$X=1-e^{-B{t}^n}$$(2)

Where, N is the number of dark particles per 0.01 mm²; T is the temperature, °C; X is the volume fraction, %; t is the time, s; n and B are constants with respect to time.

3.2. Effects of Cooling Rate on the Morphology of Precipitates

Cooling rate determines the nucleation position and morphology of the second phase particles. Figure 7 shows the morphology of precipitates under different cooling rates observed by FESEM in backscattering mode. At a low cooling rates of 0.1°C/s or 0.5°C/s, chain-like carbonitrides precipitate from the austenite grain boundaries, and their grain size is relatively larger. The precipitation of carbonitrides at the austenite grain boundary could be attributed to the fact that the austenite grain boundary is beneficial for the rapid enrichment of the micro-alloying elements. Since FESEM has higher magnification and resolution than HTCLSM, the distribution of carbonitrides at the grain boundary can be clearly observed. When the cooling rate is between 1°C/s and 3°C/s, the number of precipitates decreases, and part of precipitates form a cluster-like distribution. When the cooling rate is higher than 5°C/s, no chain-like precipitates could be found, and their number as well as grain size are significantly reduced. According to the diffusion law, the diffusion coefficient of solute elements decreases exponentially with the increase of cooling rate. Therefore, most solute elements solubilize in steel matrix instead of forming carbonitrides at a high cooling rate.

Fig. 7.

Morphology of precipitates at different cooling rates, (a) 0.1°C/s; (b) 0.5°C/s; (c) 1°C/s; (d) 3°C/s; (e) 5°C/s; (f) energy-dispersive spectrum analysis. (Online version in color.)

3.3. Effects of Cooling Rate on the Size and Volume Fraction of Precipitates

Figure 8 shows the average size and volume fraction of carbonitride particles under different cooling rates. The volume fraction of precipitates is calculated out by McCall-boyd method,³³⁾ as shown in Eq. (3). McCall-boyd method was first proposed to measure the volume fraction of tiny ThO₂ particles in cobalt alloy. After that, it is used to evaluate the volume fraction of second phase particles in microalloyed steel.³⁴⁾ This method requires measuring the diameter of each particle on a scanning electron micrograph and calculating the number of particles. In this study, 20 observed fields of each cooling rate were counted by using IPP software, and then the average volume fraction of precipitates was calculated by Eq. (3).

Fig. 8.

Average size and volume fraction of precipitation particles under different cooling rates. (Online version in color.)

According to statistics, when the cooling rate increases from 0.1°C/s to 5°C/s, the average size of carbonitrides decreases from 229 nm to 61 nm, and the volume fraction decreases from 0.7% to 0.14%. In summary, the increased cooling rate facilitates the formation of smaller carbonitrides and promotes their dispersion inside the austenite grains.   

$$f=\left[\frac{1.4\pi }{6}\right]\times \left[\frac{N{D}^2}{A}\right]$$(3)

Where, D is the average diameter of carbonitrides, μm; N is the number of carbonitrides; A is the area of field observed, μm².

The cooling rate affects the size and volume fraction of the secondary phase and determines its pinning force. Therefore, it is necessary to investigate the quantitative relation between the cooling rate and pinning force, so as to provide reference for the secondary cooling optimization of continuous casting. According to Suzuki et al.,³⁵⁾ carbonitrides with large size and small volume fraction have weak pinning force on grain boundaries. This conclusion is also consistent with the Gladman equation,³⁶⁾ shown in Eq. (4).   

$$D_c=\frac{\pi d}{6f}\left(\frac{3}{2}-\frac{2}{Z}\right)$$(4)

Where, D_c is the theoretical grain size, nm; d is the average size, nm; f is the volume fraction, %; Z is the ratio of the maximum grain size to the average size, generally 3.7.

Obviously, the pinning force of carbonitrides is primally affected by d/f, which is defined as the pinning force factor σ in this study. σ presents the average diameter of the secondary phase particles per unit volume, μm/%. The value of σ should be as large as possible to obtain a smaller pinning force. Figure 9 shows the relation of the cooling rate to the average pinning force factor σ. When the cooling rate is less than 1°C/s, σ increases quickly with the increase of cooling rate. Nevertheless, σ increases slowly when the cooling rate is higher than 1°C/s. Combining the analysis of the HTCLSM specimens with the FESEM (Fig. 7), it is revealed that the cooling rate should be higher than 1°C/s in order to effectively control the secondary phase precipitation in SG02 steel.

Fig. 9.

The pinning force factor σ at different cooling rates. (Online version in color.)

Equation (5) illustrates the quantitative relation between the cooling rate V_c and the average pinning force factor σ.   

$$\sigma =0.43-0.1{(1+2.76V{c}^{3.48})}^{-1}$$(5)

To verify the quantitative relation, the average pinning force factor of the carbonitrides on the bloom surface center of SG02 steel is obtained by statistics. Moreover, a solidification and heat transfer mathematical model is established to calculate the average cooling rate of the SCZ. As shown in Fig. 9, the statistical result accords with the fitting curve. Furthermore, it also indicates that carbonitrides have enough time to nucleate and grow under the mild cooling condition of secondary cooling, and the pinning effect of carbonitrides is enhanced. Therefore, it is necessary to develop a new cooling method matching the solidification characteristics of microalloyed steel so as to reduce the cracks.

4. A New Cooling Strategy/Technology and Plant Trials

During the continuous casting process of SG02 steel, mild cooling was adopted to avoid the low ductility zone at the straightening point. Unfortunately, there were still serious cracks under the bloom subsurface. To find the crack formation reason, optical microscope (OM) and scanning electron microscope (SEM) equipped with energy dispersive spectrometer (EDS) were employed to observe the micromorphology and element distribution of subsurface cracks. Figure 10 shows the OM images of crack distribution of SG02 steel bloom. Usually, the cracks extend along the proto-austenite grain boundaries and their length is less than 1 mm. These small cracks connect each other as a net and seriously damage the hot ductility of steel. Moreover, the enrichments of V, Nb and Ti elements near the cracks can be seen in the SEM image (Fig. 11) although it is difficult to find the fine precipitates by SEM. Hence, it can be inferred that carbonitrides, as the stress concentration source, reduce the binding force of grain boundaries and lead to the formation of the subsurface cracks.

Fig. 10.

OM morphology of the subsurface cracks of steel bloom. (Online version in color.)

Fig. 11.

Element distribution in a subsurface crack. (Online version in color.)

The temperature range (1085–910°C) and reasonable cooling rate (≥1°C/s) for controlling the precipitation of secondary phase particles are obtained, and a new secondary cooling strategy and technology is proposed to reduce the subsurface cracks of the bloom, as shown in Fig. 12. The core idea of this method is to implement intensive cooling on the bloom below the mold, so that the carbonitrides on the bloom surface can precipitate rapidly in the SCZ, and the surface microstructure of the bloom can be strengthened. The SG02 steel continuous casting is taken as an example to illustrate the detailed workflow.

Fig. 12.

Workflow of a new cooling strategy and technology for the microalloy steel.

A solidification and heat transfer model considering the actual water distribution in the SCZ is established to calculate the thermal behavior of SG02 steel bloom as well as the cooling rate of the SCZ. The descriptions of the modeling method and the calculation of cooling rate were provided in the previous works of our team.^12,37,38) The heat transfer model was verified by comparing the measured temperatures with the calculated ones. The surface temperature of the bloom was measured using an infrared radiation pyrometer. During the measurement, the pyrometer was perpendicular to the surface center of the inner arc and peak values were adopted as the local temperature, so as to reduce the influence of spraying water and oxide scale on the measured results. Figure 13 shows the temperature profiles of the bloom under different cooling patterns. The temperature of the bloom surface center in the SCZ is between 1125°C and 1022°C under the actual production conditions. Obviously, the bloom surface experienced a comparatively longer time under high temperature and there is enough time for carbonitrides to nucleate and precipitate from the matrix. The average cooling rate of each segment in the SCZ is 2.05°C/s, 0.61°C/s, 0.50°C/s and 0.31°C/s, respectively, and only the cooling rate of SCZ-Isegment is higher than 1°C/s.

Fig. 13.

The comparison of the bloom surface temperatures with the original and the new secondary cooling patterns. (Online version in color.)

It is an effective method to prevent the precipitation and growth of the secondary phase by conducting intensive cooling on the bloom below the mold. When the water flowrate in the SCZ-Isegment remains constant and the water flowrate in the SCZ-IIsegment increases to 2.8 times than that the original water flowrate (36.7 L/min), the temperature of the bloom surface center reduces greatly to 908°C at the end of SCZ-IIsegment (Fig. 13), indicating that most of carbonitride particles have precipitated. The cooling rate of the SCZ-IIsegment increases from 0.61°C/s to 1.52°C/s. As the corner of the bloom involves two-dimensional cooling, the solidification behavior shows characteristics of relatively lower temperature and higher cooling rate,^12,39) which provides more favorable condition for inhibiting the precipitation of carbonitrides. As shown in Fig. 13, the temperature of bloom corner after leaving the mold is 721°C, indicating that all the carbonitrides have precipitated in the mold. Theoretically, the larger the water flowrate, the more dispersive to particle dispersion. However, it should also be noted that excessive water flowrate will not only reduce the temperature of the bloom at the straightening point and make it fall into the low ductility temperature zone, but also increase burden for the water supply system of continuous casting. Therefore, a series of plant trials were conducted to reduce the subsurface cracks of the bloom by increasing 2.8 times of the original water flowrate of the SCZ-IIsegment. During the process of plant trials, the water flowrate of other segments (SCZ-I, SCZ-III, SCZ-IV) as well as other parameters such as casting speed, superheating remain constant.

After the implementation of the scheme, the number of chain-like precipitates on bloom surface decreases, and the secondary phase particles tend to disperse in steel matrix, as shown in Fig. 14. The surface and subsurface cracks of SG02 steel blooms were small and usually covered by thick iron oxide scale. To evaluate the plant trials, fixed-length blooms produced before and after the plant trials were put into a large pickling tank for erosion. After carefully inspection, the cracks of SG02 steel blooms disappeared by adopting the new cooling process. Moreover, the qualified ratio of the rolled products increases from 65% to 91%.

Fig. 14.

Morphology of precipitates under (a) mild cooling and (b) intensive cooling. (Online version in color.)

5. Conclusion

In this study, carbonitride precipitation behavior was characterized in-situ by HTCLSM, and the effects of cooling rate on the morphology, distribution and the pinning force factor σ of carbonitrides were investigated by FESEM. Based on these, a new cooling method for the continuous casting process of SG02 microalloyed steel is proposed. The following conclusion can be drawn.

(1) The carbonitride precipitation process is usually accompanied by the emergence of dark particles, which is caused by the expansion of the depletion region near the secondary phase particles. The number of dark particles is inversely exponentially proportional to temperature, namely N=3462.6+90207.9(1+e^{(0.084T−78.786)})⁻¹. The growth pattern of precipitates conforms to the Avrami equation. Carbonitrides mainly precipitate between 1085°C and 910°C, and the “fast-growing region” ranges from 910°C to 960°C.

(2) With the increase of cooling rate, the size and volume fraction of the carbonitrides decrease, and the nucleation location changes from grain boundary to grain interior. The pinning force factor σ is defined according to classical Gladman equation, which is a power function of the cooling rate, namely σ=0.43−0.1(1+2.76Vc^3.48)⁻¹. For SG02 steel, the cooling rate should be higher than 1°C/s in order to effectively weaken the pinning effects of the precipitates.

(3) Intensive cooling is carried out by increasing the water flowrate of the SCZ-IIsegment to 2.8 times than that the original water flowrate, so as to strengthen the surface microstructure of the SG02 steel bloom. The plant trials show that the cracks disappeared after applying this new secondary cooling pattern.

Acknowledgments

The present work was financially supported by the National Natural Science Foundation of China (No. U21A20112), Innovative & Entrepreneurial Talent Project in Jiangsu province, China (No. 2016A426) and the Nanjing Iron & Steel Co., Ltd., China (No. IGAB20120007). Special thanks are extended to our cooperating company for facilitating the industrial trials and applications.

References

• 1)   P.  Gong,  E.  Palmiere and  W. M.  Rainforth: Acta Mater., 97 (2015), 392. https://doi.org/10.1016/j.actamat.2015.06.057
• 2)   K. A.  Annan,  C. W.  Siyasiya and  W. E.  Stumpf: ISIJ Int., 58 (2018), 333. https://doi.org/10.2355/isijinternational.ISIJINT-2017-464
• 3)   T. P.  Qu,  J.  Tian,  K. L.  Chen,  Z.  Xu and  D. Y.  Wang: Ironmaking Steelmaking, 46 (2019), 353. https://doi.org/10.1080/03019233.2017.1396722
• 4)   T.  Senuma: ISIJ Int., 41 (2001), 520. https://doi.org/10.2355/isijinternational.41.520
• 5)   J. S.  Qing,  L.  Wang,  K.  Dou,  B.  Wang and  Q.  Liu: High Temp. Mater. Process., 35 (2016), 575. https://doi.org/10.1515/htmp-2014-0240
• 6)   D. J.  Mun,  E. J.  Shin,  Y. W.  Choi,  J. S.  Lee and  Y. M.  Koo: Mater. Sci. Eng. A, 545 (2012), 214. https://doi.org/10.1016/j.msea.2012.03.047
• 7)   K. C.  Cho,  D. J.  Mun,  J. Y.  Kim,  J. K.  Park,  J. S.  Lee and  Y. M.  Koo: Metall. Mater. Trans. A, 41 (2010), 1421. https://doi.org/10.1007/s11661-010-0211-9
• 8)   K. R.  Carpenter,  R.  Dippenaar and  C. R.  Killmore: Metall. Mater. Trans. A, 40 (2009), 573. https://doi.org/10.1007/s11661-008-9749-1
• 9)   T.  Chen,  C.  Ji and  M. Y.  Zhu: J. Alloy. Compd., 823 (2020), 153650. https://doi.org/10.1016/j.jallcom.2020.153650
• 10)   Y.  Luo,  H.  Guo and  J.  Guo: Ironmaking Steelmaking, 46 (2019), 698. https://doi.org/10.1080/03019233.2018.1461593
• 11)   K.  Dou,  L. T.  Meng,  Q.  Liu,  B.  Liu and  Y. H.  Huang: Met. Mater. Int., 22 (2016), 349. https://doi.org/10.1007/s12540-016-2676-6
• 12)   K.  Dou and  Q.  Liu: Metall. Mater. Trans. A, 51 (2020), 3945. https://doi.org/10.1007/s11661-020-05819-9
• 13)   F.  Ma,  G.  Wen and  W.  Wang: Steel Res. Int., 84 (2013), 370. https://doi.org/10.1002/srin.201200161
• 14)   O.  Comineli,  R.  Abushosha and  B.  Mintz: Mater. Sci. Technol., 15 (1999), 1058. https://doi.org/10.1179/026708399101506788
• 15)   S.  Clark,  V.  Janik,  A.  Rijkenberg and  S.  Sridhar: Mater. Charact., 115 (2016), 83. https://doi.org/10.1016/j.matchar.2016.03.021
• 16)   J. S.  Park,  Y. S.  Ha,  S. J.  Lee and  Y. K.  Lee: Metall. Mater. Trans. A, 40 (2009), 560. https://doi.org/10.1007/s11661-008-9758-0
• 17)   S. S.  Xie,  J. D.  Lee,  U. S.  Yoon and  C. H.  Yim: ISIJ Int., 42 (2002), 708. https://doi.org/10.2355/isijinternational.42.708
• 18)   N.  Baba,  K.  Ohta,  Y.  Ito and  T.  Kato: Rev. Metall. Paris, 103 (2006), 174. https://doi.org/10.1051/metal:2006163
• 19)   T.  Kato,  Y.  Ito,  M.  Kawamoto,  A.  Yamanaka and  T.  Watanabe: ISIJ Int., 43 (2003), 1742. https://doi.org/10.2355/isijinternational.43.1742
• 20)   C.  Du,  J.  Zhang,  J.  Wen,  Y.  Li and  P.  Lan: Ironmaking Steelmaking, 43 (2016), 331. https://doi.org/10.1179/1743281215Y.0000000044
• 21)   S.  Griesser,  M.  Reid,  R.  Pierer,  C.  Bernhard and  R.  Dippenaar: Steel Res. Int., 85 (2014), 1257. https://doi.org/10.1002/srin.201300024
• 22)   S.  Kimura,  K.  Nakajima,  S.  Mizoguchi and  H.  Hasegawa: Metall. Mater. Trans. A, 33 (2002), 427. https://doi.org/10.1007/s11661-002-0103-8
• 23)   C.  Slater,  K.  Hechu and  S.  Sridhar: Mater. Charact., 126 (2017), 144. https://doi.org/10.1016/j.matchar.2017.02.025
• 24)   S.  Griesser and  R.  Dippenaar: ISIJ Int., 54 (2014), 533. https://doi.org/10.2355/isijinternational.54.533
• 25)   Z.  Liu,  Y.  Kobayashi,  J.  Yang,  K.  Nagai and  M.  Kuwabara: ISIJ Int., 46 (2006), 847. https://doi.org/10.2355/isijinternational.46.847
• 26)   T.  Liu,  M.  Long,  D.  Chen,  Y.  Huang,  J.  Yang,  H.  Duan,  L.  Gui and  P.  Xu: Metall. Mater. Trans. B, 51 (2020), 338. https://doi.org/10.1007/s11663-019-01758-y
• 27)   T.  Liu: Ph.D. thesis, Chongqing University, (2020), http://cdmd.cnki.com.cn/Article/CDMD-10611-1021541544.htm, (accessed 2022-06-18).
• 28)   G.  Shigesato,  M.  Sugiyama,  S.  Aihara,  R.  Uemori and  Y.  Tomita: Tetsu-to-Hagané, 87 (2001), 93 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.87.2_93
• 29)   L.  Xing,  X.  Fan,  M.  Wang,  L.  Zhao and  Y.  Bao: Metall. Mater. Trans. B, 52 (2021), 3208. https://doi.org/10.1007/s11663-021-02248-w
• 30)   L.  Chen,  K.  Matsuura,  M.  Ohno and  D.  Sato: ISIJ Int., 52 (2012), 1841. https://doi.org/10.2355/isijinternational.52.1841
• 31)   V. M.  Schastlivtsev,  D. A.  Mirzayev,  S. E.  Karzunov and  I. L.  Yakovleva: ISIJ Int., 35 (1995), 955. https://doi.org/10.2355/isijinternational.35.955
• 32)   T. A.  Kop,  Y. V.  Leeuwen,  J.  Sietsma and  S. V. D.  Zwaag: ISIJ Int., 40 (2000), 713. https://doi.org/10.2355/isijinternational.40.713
• 33)   R. J.  Seher and  G. N.  Maniar: Metallography, 5 (1972), 409. https://doi.org/10.1016/0026-0800(72)90029-8
• 34)   G.  Qian,  G.  Cheng and  Z.  Hou: ISIJ Int., 54 (2014), 1611. https://doi.org/10.2355/isijinternational.54.1611
• 35)   H. G.  Suzuki,  S.  Nishimura,  J.  Imamura and  Y.  Nakamura: Trans. Iron Steel Inst. Jpn., 24 (1984), 169. https://doi.org/10.2355/isijinternational1966.24.169
• 36)   T.  Gladman: Proc. R. Soc. Lond. A, 294 (1966), 298. https://doi.org/10.1098/rspa.1966.0208
• 37)   Y. S.  Han,  W.  Yan,  J. S.  Zhang,  W. Q.  Chen,  J.  Chen and  Q.  Liu: ISIJ Int., 60 (2020), 106. https://doi.org/10.2355/isijinternational.ISIJINT-2019-409
• 38)   Y. S.  Han,  W.  Yan,  J. S.  Zhang,  J.  Chen,  W. Q.  Chen and  Q.  Liu: J. Iron Steel Res. Int., 28 (2021), 160. https://doi.org/10.1007/s42243-020-00412-1
• 39)   M. R. R. I.  Shamsi and  S. K.  Ajmani: Steel Res. Int., 81 (2010), 132. https://doi.org/10.1002/srin.200900103