Solidification Structure, Non-metallic Inclusions and Hot

ISIJ International, Vol. 59 (2019), No. 5

© 2019 ISIJ 872

ISIJ International, Vol. 59 (2019), No. 5, pp. 872–879

* Corresponding author: E-mail: [email protected]: https://doi.org/10.2355/isijinternational.ISIJINT-2018-694

Solidification Structure, Non-metallic Inclusions and Hot Ductility of Continuously Cast High Manganese TWIP Steel Slab

Yu-nan WANG*

Steelmaking Research Department, Research Institute, Baoshan Iron and Steel Co., Ltd., Shanghai, 201999 P. R. China.

(Received on October 25, 2018; accepted on December 17, 2018)

In the current study, the solidification structure, non-metallic inclusions and hot ductility of continuously cast high manganese TWIP steel slab have been investigated and the inclusion formation behavior have been revealed by FactSage (CRCT-ThermFact Inc., Montréal, Canada). The area ratio of equiaxed grain zone of the TWIP steel slab is 0.18. Two main types of inclusions in the TWIP steel slab are single AlN particle and AlN+MnS aggregates. It is found that MgAl2O4 and AlN particles can precipitate in the initial solidifica-tion stage, which can act as heterogeneous nuclei of other inclusions. In the high temperature tension test, the reduction of area (RA) of the TWIP steel slab samples are higher than 40 pct in the temperature range from 873 K to 1 473 K (600°C to 1 200°C). Brittle fractures are observed in the fracture surface of the TWIP steel slab samples with dimples. Contents of manganese, carbon, sulfur and phosphorus, strain rate, and dynamic recrystallization (DRX) are factors influencing the hot ductility of TWIP steel slab.

KEY WORDS: solidification structure; non-metallic inclusion; hot ductility; continuous casting; high manga-nese TWIP steel.

1. Introduction

In recent years, the demand for high quality automotive steel has been increasing.1) The development directions of the automobile are the reductions of weight, fuel consump-tion, emissions, and the improvement of safety, and thus put forward high requirements for automotive steels that account for about 70 pct of the weight of automobile.2,3) Twin induced plasticity (TWIP) steels, which show good mechanical properties, such as high strength, excellent duc-tility, and high energy absorption capacities, are the most attractive automobile structure steels and have received researchers’ extensive concerns.4–6) Typical TWIP steel systems are Fe–Mn–C, Fe–Mn–Al–C, and Fe–Mn–Al–Si.7)

Currently, the solidification structures, hot ductility and inclusions of high manganese TWIP steels produced by ingot-casting processes have been investigated. In our previous research, the solidification structures, the thermal properties and phase transformation characteristics of high manganese TWIP steel ingots have been investigated.7) P. Lan et al. have analyzed the solidification microstructure, solidification defects, thermophysical properties and hot ductility of Fe-22Mn-0.7C TWIP steel ingot.8–12) S. E. Kang et al.13–18) have investigated the influences of composition and AlN on the hot ductility of high aluminum TWIP steels, which were produced by 50 kg vacuum induction furnace. I. Mejía,19) A. E. Salas-Reyes20) and A. S. Hamada21) have studied the hot ductility behavior of high-manganese aus-

tenitic TWIP steels. The steels were fabricated in induction furnaces.

Also, the effects of cooling conditions and aluminum contents on the evolution of non-metallic inclusions in high manganese TWIP steel ingots have been indicated by our group.22,23) In addition, G. Gigacher et al.24) have investi-gated endogenous inclusions in some high-alloy TRIP and TWIP steels which were melted in an induction-heat furnace with a capacity of 25 kg. J. H. Park et al.25) have investi-gated the effects of aluminum and manganese contents on the size, composition, and three-dimensional morphologies of inclusions formed in Fe-xMn-yAl (x = 10 and 20 mass pct, y=1, 3, and 6 mass pct) steels which produced by a high-frequency induction furnace. Zhuang et al.6) have investigated endogenous inclusions formed in Fe-25Mn-3Si-3Al TWIP steels in laboratorial ingot, mold casting after AOD steelmaking and electroslag remelting (ESR) process at industrial plant, respectively. Von Schweinichen et al.1) have revealed the effects of such casting parameters as superheat, pouring rate, hot top, and stirring conditions on the solidification and cleanliness of low carbon-alloyed and high manganese-alloyed ingots. H. Liu26,27) have performed the character of non-metallic inclusions in Fe–Mn–Si–Al TWIP steels during argon oxygen decarburization-electro-slag remelting-forging (AOD-ESR-forging) process.

Unfortunately, the solidification characteristics and non-metallic inclusions of continuously cast high manganese TWIP steel slab have seldom been reported. The solidifica-tion characteristics and non-metallic inclusions have great influences on the continuous casting process and quality of casting products. In this paper, the solidification structure,


© 2019 ISIJ873

hot ductility and non-metallic inclusions of continuously cast TWIP steel slab were investigated based on experi-ments and thermodynamic calculations.

2. Experimental

2.1. High Manganese TWIP Steel PreparationIn the present work, a high manganese TWIP steel was

produced by the route of combined blown converter → ladle furnace (LF) → RH degassing → continuous casting in steelmaking plant of Baosteel. After desulphurization, the molten iron was decarbonized by oxidization in com-bined blown converter. During the tapping process, pure aluminum was first added into the molten steel, and then part of the electrolytic manganese was added. After tap-ping, aluminous slag was added to the surface of the ladle slag to deoxidize it. During the LF treatment process, an aluminum deoxidizer was first added, and then the electro-lytic manganese was added in batches to reach the target of the manganese content in the molten steel. During the RH process, the evacuation time was greater than 21 min and the degassing time was 8 min. All the alloying elements in the molten steel were added to the target values during the RH process. The steel casting was carried out using a vertical-bending caster. The tundish superheat was 50°C. The cast-ing speed was 0.8 m/min. The mold cooling intensity was high and the secondary cooling intensity was low. The soft reduction was used. The size of continuous casting slab was 230 × 1 450 mm.

2.2. Chemical Analysis20 mm × 20 mm × 10 mm sample was cut from the

position of 1/4 of width and thickness of the slab for ana-lyzing the chemical compositions of the steel. The oxygen in the sample was determined by the pulse heating inert gas fusion-infra-red absorption method. The sulfur in the sample was determined by the infrared absorption method after combustion in an induction furnace. The nitrogen in the sample was determined by the thermal conductimetric method after fusion in a current of inert gas. The contents of magnesium and calcium in the sample were determined by the inductively coupled plasma-atomic emission spectrom-etry (ICP-AES). The contents of carbon, silicon, manganese, phosphorus, niobium, titanium, aluminum and boron in the sample were determined by method for spark discharge atomic emission spectrometric analysis. The chemical com-position of the TWIP steel slab is listed in Table 1.

2.3. Characterization of Solidification Structure andNon-metallic Inclusions

The solidification structure was observed from the cross section of the TWIP steel slab. After being polished with a grinder, the cross section of the slab was soaked in 33 volume pct dilute hydrochloric acid for approximately 7 minutes at 343 K (70°C). Then the columnar and equiaxed

grains were observed on macro examination. The proportion of equiaxed grains was determined by area measurement.28)

The steel samples for inclusion detection were cut at the position of 1/4 of width and thickness. After grinding and polishing, inclusion analysis was performed using a scan-ning electron microscope (SEM, EVO MA 25, Carl Zeiss, Oberkochen, Germany) attached with an automated EDS (energy-dispersive spectroscopy; Oxford Instruments, UK) programmed to detect non-metallic inclusions of an area of 10 × 10 mm2. The number of the view fields were 508. The program provides an EDS mapping of inclusions regarding elemental analysis and is set at 15 kV, 15 mA, and 10 mm working distance. The number densities of inclusions per unit area (NA), the size distributions, and the volume ratios of inclusions in the steels were obtained by the inclusion automatic statistics results.

2.4. Thermodynamic AnalysisIn the present study, the FactSage thermodynamic soft-

ware29,30) was used for the thermodynamic calculations for the inclusion formation in high manganese TWIP steels. The FactPS database, FToxid database, and FSstel database were employed for the steels, respectively. The variation of different inclusion phases, the liquidus and solidus were calculated from FactSage 7.2 thermodynamic software by using the “equilibrium cooling” option.

2.5. Hot Tensile TestA Gleeble 3800 thermo-mechanical simulator was used

to measure the hot ductility of the samples which was cut from columnar grain zone. The sketch of sampling loca-tion is shown in Fig. 1. The hot tensile test was performed under vacuum and the sizes of samples are ϕ10 × 120 mm. The ends of the samples were threaded so that they can be screwed into the grips of the tensile test machine. In the hot tensile tests, the samples were reheated to 1 523 K (1 250°C) at a rate of 10 K s −1 for soaking for 5 minutes, to dissolve any precipitates present and finally cooled at a rate of 3 K s −1 to the test temperature between 873 K and

Table 1. Chemical compositions (in mass pct) of the high manganese TWIP steel slab.

C Si Mn P S Nb Ti Al B Mg O N

0.65 0.066 15.69 0.010 0.0009 0.02 0.004 1.6 0.0008 0.0002 0.0004 0.0064

Fig. 1. Sampling location of the TWIP steel slab.


© 2019 ISIJ 874

1 473 K (600°C and 1 200°C), at which they were main-tained for 2 minutes to stabilize the temperature filed before the tensile tests. Schematic of heating cycle of hot tensile test is shown in Fig. 2. Temperature was measured using a chromel-alumel thermocouple spot-welded on the surface in the middle of the sample. The reduction in area (RA pct) was measured to estimate the hot ductility of the steels.

3. Results and Discussion

3.1. SolidificationStructureoftheTWIPSteelSlabFigure 3(a) shows the solidification structure of the left

third of the cross section of the slab. The size, shape, and orientation of the grains reveal three distinct zones: the central region of coarse randomly oriented equiaxed grains; the middle region of columnar grains, normally elongated to near the slab surface; the outer surface zone, correspond-ing to very fine chill crystals, is present at the extreme skin.

The details of the selected four regions are shown in Figs. 3(b)–3(d) and 3(e). The thickness of outer surface zones with very fine grains is about 2.5–4 mm in region (b), (c) and (d). As shown in Fig. 3(e), the equiaxed grain size is about 5–7 mm.

As shown in Fig. 3(a), the length of the columnar grains of inner arc slab is about 110–120 mm, while that of outer arc slab is about 80–90 mm. Since the slab through the vertical-bending caster, which was influenced by gravity, the growth of columnar grains of outer arc was inhibited and the length of the columnar grains of inner arc was longer. The area of equiaxed grain zones is small in the slab, and the proportion of equiaxed grain zone is 0.18. Our previ-ous research7) have shown that the thermal conductivity of the high manganese steel (17.07 mass pct) is much smaller than that of the low manganese steel (0.028 mass pct). Low thermal conductivity of steel can cause large temperature gradient of primary solidified shell, which promotes the growth of columnar grains. Thus, the proportion of equiaxed grain zone is small in the TWIP steel slab.

3.2. Thermodynamic CalculationsThe variation of different inclusion phases is presented Fig. 2. Schematic of heating cycle of hot tensile test.

Fig. 3. Solidification structure of the TWIP steel slab: (a) left third of the cross section; (b), (c) and (d) columnar grains details; (e) equiaxed grains details.


© 2019 ISIJ875

in Fig. 4. Table 2 shows the liquidus and solidus of the TWIP steel slab and the formation temperatures of inclu-sions. In the slab, the oxide phase of MgAl2O4 is formed at 1 718 K (1 445°C). The sulfide phases of MnS and MgS are formed at 1 614 K and 1 598 K (1 341°C and 1 325°C), respectively. The nitride phase of AlN is formed at 1 713 K (1 440°C). The liquidus and solidus of the TWIP steel slab are 1 724 K and 1 609 K (1 451°C and 1 336°C), respec-tively. Obviously, AlN and MgAl2O4 are formed during solidification, and MnS and MgS are formed in the solid phase. One can see the major nitride is AlN and the major sulfide is MnS.

3.3. Characteristic of Inclusions in the TWIP Steel SlabOver 5000 inclusions were analyzed using a SEM

equipped with an automated EDS in the slab. The chemi-cal compositions, morphologies, and sizes of inclusions are obtained.

3.3.1. Type and Morphologies of InclusionsIn this study, the non-metallic inclusions formed in the

TWIP steel slab are classified as one of the following six types: (1) single Al2O3 particle; (2) MgAl2O4 spinel inclu-sions; (3) single AlN particle; (4) single MnS particle; (5) AlN+MnS, AlN was associated with MnS aggregate; (6) MgAl2O4+MnS+MgS, MgAl2O4 core with MnS+MgS layer; The calculation results of inclusion types are in good agreement with the experimental observation results.

The number densities of inclusions formed in the TWIP steel slab is shown in Fig. 5. Figure 5 shows that the main types of inclusions in the slab are single AlN particle and AlN+MnS complex inclusion, the number densities of them

are higher than 10/mm2.The typical morphologies of inclusions in the TWIP steel

slab are shown in Fig. 6. The irregular polygon single Al2O3 particle about 5 μm in size could be found in the slab (Fig. 6(a)). The near spherical single MnS particles about 3 μm in size could also be found (Fig. 6(c)). There are two main types of AlN inclusions: (1) single AlN particle, present in rectangular (Fig. 6(b)). (2) AlN+MnS complex inclusion, AlN associated with MnS aggregate (Fig. 6(d)), with a size of 5 μm. The MgAl2O4+MnS+MgS complex inclusion presents in rectangular (Fig. 6(e)), with a size between 10 and 15 μm.

In the calculation results, the formation temperatures of AlN and MgAl2O4 are high, they can act as the heteroge-neous nuclei of other particles. The formation temperatures of MnS and MgS are lower than those of AlN and MgAl2O4, they were often formed on the AlN and MgAl2O4 particles. The calculation results can explain structures of complex inclusions well.

3.3.2. Size and Volume Ratio of InclusionsThe inclusion size distribution of the TWIP steel slab is

shown in Fig. 7. In the slab, there are many inclusions less than 5 μm, which frequency is 78.9 pct. The frequency of inclusions which sizes are larger than 10 μm is 3.8 pct. The average size, number density, and volume ratio of inclu-sions are 3.34 μm, 57/mm2, and 7.2×10 −5, respectively. The volume ratio of inclusions in the slab can be calculated by Dehoff’s equations31) expressed by Eqs. (1) through (3).

NN

dV

a� �2

�................................ (1)

1 1 1

d n di� �� ............................... (2)

f d NV V� ��6

3 .............................. (3)

Where, NV is the number of inclusions per unit volume in

Fig. 4. Evolution of inclusions, calculated from thermodynamic databases.

Table 2. The liquidus and solidus of the high manganese TWIP steel slab and the formation temperatures of inclusions.

Liquidus SolidusFormation Temperatures of Inclusions

MgAl2O4 AlN MnS MgS

1 724 K (1 451°C)

1 609 K (1 451°C)

1 718 K (1 445°C)

1 713 K (1 440°C)

1 614 K (1 341°C)

1 598 K (1 325°C)

Fig. 5. Number densities of inclusions observed in the high man-ganese TWIP steel slab.


© 2019 ISIJ 876

Fig. 6. Morphologies of typical inclusions in the high manganese TWIP steel slab: (a) single Al2O3 particle; (b) single AlN particle; (c) single MnS particle; (d) AlN+MnS complex inclusion; (e) MgAl2O4+MnS+MgS complex inclusion.


© 2019 ISIJ877

Fig. 7. Inclusion size distribution in the high manganese TWIP steel slab.

specimen (m −3), Na is the number of inclusions per unit area in specimen (m −2), di is the apparent particle size of mean of ith inclusion among n inclusions (m), d is the harmonic mean of inclusion particle size (m), and fV is the volume ratio of inclusions.

Figure 8 shows the average size of each inclusion in the TWIP steel slab. It is shown that there are small inclusions whose sizes are less than 2 μm, such as single Al2O3 and AlN particles. Also, there are some large inclusions whose sizes are greater than 4 μm, such as single MgO particle, MgAl2O4+MnS+MgS, MgAl2O4+AlN+MnS+MgS and AlN+TiN(NbN)+MnS+MgS complex inclusions. Figure 9 shows the volume ratio of each inclusion in the TWIP steel slab. Figure 9 shows the volume ratio of each inclusion in the TWIP steel slab. The volume ratios of AlN+MnS complex inclusion and MgAl2O4+MnS+MgS complex inclusion are large, which are 2.5×10 −5 and 2.4×10 −5, respectively. However, the volume ratios of single incu-sions, such as AlN, MnS, Al2O3 and MgAl2O4 are small, which are lower than 6×10 −6. The result indicates that MnS and MgS are more likely to precipitate on AlN and MgAl2O4 instead of being precipitated separately.

3.4. Analysis of Hot Ductility and Fracture MechanismFigure 10 reveals the hot ductility curve of the TWIP

steel slab, along with the experiment results from previ-ous researches.7,17,18,21) The chemical composition of TWIP steels from references are shown in Table 3. Hot ductility curve of the TWIP steel slab shows fluctuation in the RA from 48 pct to 74 pct in the temperature range from 873 K to 1 173 K (600°C to 900°C). With increasing temperature from 1 173 K to 1 473 K (900°C to 1 200°C), the variation of RA is insignificant and the RA is in the range from 44 pct to 49 pct. It is shown that the RA of the TWIP steel slab is higher than 40 pct throughout the temperature range under examination.

It is worthy to notice that RA of the TWIP steel slab is similar with that of as cast TWIP steel with the composi-tion of Fe-17Mn-0.6C-2.1Al, as shown in Fig. 10. Kang et al.17,18) revealed that the ductility of as cast TWIP steels with the composition of Fe-18Mn-0.6C-1.5Al are lower

than that of the TWIP steel slab. Lan et al.10) indicated that the inevitable solute microsegregation manganese and carbon, along with the microporosity in as cast matrix result

Fig. 8. Average size of each inclusion in the high manganese TWIP steel slab.

Fig. 9. Volume ratio of each inclusion in the high manganese TWIP steel slab.

Fig. 10. Hot ductility curve of the high manganese TWIP steel slab, together with results from references 7, 17, 18, 21).


© 2019 ISIJ 878

in the homogeneity decrease of TWIP steel, leading to the ductility loss of the matrix, especially the dendrite bound-aries zone. According to the investigation by Kang et al.17) and Yang et al.,7) the formation of sulfur and phosphorus compound with low melting point leads to the weakness of hot ductility. In a recent work by Lv et al.,32) the RA of high manganese steel started to decrease when the contents of sulfur and phosphorus are higher than 0.006 mass pct respectively. Thus, high contents of manganese, sulfur and phosphorus in the Fe-18Mn-0.6C-1.5Al TWIP steels would cause the deterioration of ductility. Hamada et al.21) revealed that the ductility of as rolled TWIP steel with the composi-tion of Fe-16Mn-0.3C-1.5Al is higher than that of the TWIP steel slab. It is illustrated by Wray33) that a fine-grain struc-ture can promote the early onset and fast rate of dynamic

Table 3. Chemical composition (in mass pct) of TWIP steels from references.

Steel C Si Mn P S Al N Reference

TWIP slab 0.65 0.066 15.69 0.010 0.0009 1.6 0.0064

17Mn-0.6C-2.1Al as cast 0.57 – 17.07 0.007 <0.0005 2.1 0.0043 Yang et al. 7)

18Mn-0.6C-1.5Al as cast 0.615 0.20 18.30 0.022 0.0032 1.51 0.0080 Kang et al. 17)

18Mn-0.6C-1.5Al as cast 0.61 – 18.09 0.009 0.0062 1.53 0.0095 Kang et al. 18)

16Mn-0.3C-1.5Al as rolled 0.29 – 16.40 – – 1.54 – Hamada et al. 21)

Fig. 11. Fracture surfaces (SEM) for the high manganese TWIP steel slab tested at different temperatures: (a) 1 073 K (800°C); (b) 1 123 K (850°C).

recrystallization (DRX) in steel. Sah et al.,34) Crowther and Mintz,35) and Fernandez et al.36) also obtained the similar conclusion. Mintz and Abushosha37) reported that DRX and the movement of grain boundaries resist the voids to link up and thereby leads to high RA values. It is supposed that the grain refinement of the TWIP steel by rolling gives rise to higher RA than that from the TWIP steel slab with coarser grains. Besides, the high RA of TWIP steel in Hamada’s experiment can also be explained by the high tensile strain rate (1 s −1) and the low carbon content.10,37)

Fracture surfaces of the TWIP steel slab at different tem-peratures obtained by SEM are shown in Fig. 11. At 1 073 K and 1 123 K (800°C and 850°C), the fracture surface shows hybrid-fracture characteristics of brittle and dimple, as shown in Figs. 11(a) and 11(b). Since the as-cast TWIP steel has no phase transformation during solidification and no DRX occurs in the high temperature zone during hot tensile test,7) the hot ductility at different temperatures is not much different and the fracture surface characteristics are similar.

4. Conclusions

(1) The characteristics of the solidification structure of the high manganese TWIP steel slab was analyzed. The length of the columnar grains of inner arc slab is about 110–120 mm, while that of outer arc slab is about 80–90 mm. The diameter of equiaxed grains is about 5–7 mm. The proportion of equiaxed grain zones of the slab was 0.18.

(2) The inclusions formed in the high manganese TWIP steels can be classified as one of the following six types: (1) single Al2O3 particle; (2) MgAl2O4 spinel inclu-sions; (3) single AlN particle; (4) single MnS particle; (5) AlN+MnS, AlN was associated with MnS aggregate; (6) MgAl2O4+MnS+MgS, MgAl2O4 core with MnS+MgS layer.

(3) The ductility of the high manganese TWIP steel slab are higher than 40 pct in the temperature range from 873 K to 1 473 K (600°C to 1 200°C). Brittle fractures are observed in the surface of the slab samples with dimples. Contents of manganese, carbon, sulfur and phosphorus, strain rate, and DRX are factors influencing the hot ductility of the TWIP steel slab.

REFERENCES

1) P. Von Schweinichen, Z. Chen, D. Senk and A. Lob: Metall. Mater. Trans. A, 44 (2013), 5416.

2) W. F. Powers: Adv. Mater. Process., 157 (2000), 38.3) D. Cornette, P. Cugy, A. Hildenbrand, M. Bouzekri and G. Lovato:

Rev. Metall., 102 (2005), 905.4) O. Grässel, L. Krüger, G. Frommeyer and L. Meyer: Int. J. Plast., 16

(2000), 1391.


© 2019 ISIJ879

5) H. Idrissi, K. Renard, L. Ryelandt, D. Schryvers and P. J. Jacques: Acta Mater., 58 (2010), 2464.

6) C. Zhuang, J. Liu, Z. Mi, H. Jiang, D. Tang and G. Wang: Steel Res. Int., 85 (2014), 1432.

7) J. Yang, Y. Wang, X. Ruan, R. Wang, K. Zhu, Z. Fan, Y. Wang, C. Li and X. Jiang: Metall. Mater. Trans. B, 46 (2015), 1365.

8) P. Lan, L. Song, C. Du and J. Zhang: Mater. Sci. Technol., 30 (2014), 1297.

9) P. Lan, H. Tang and J. Zhang: Metall. Mater. Trans. A, 47 (2016), 2964.

10) P. Lan, H. Tang and J. Zhang: Mater. Sci. Eng. A, 660 (2016), 127.11) P. Lan and J. Zhang: Steel Res. Int., 87 (2015), 250.12) P. Lan, H. Tang, Y. Ji, C. Du and J. Zhang: Chin. J. Eng., 38 (2016),

795.13) S. E. Kang, J. R. Banerjee, A. Tuling and B. Mintz: Mater. Sci.

Technol., 30 (2014), 1328.14) S. E. Kang, A. Tuling, I. Lau, J. R. Banerjee and B. Mintz: Mater.

Sci. Technol., 27 (2011), 909.15) S. E. Kang, J. R. Banerjee, A. S. Tuling and B. Mintz: Mater. Sci.

Technol., 30 (2014), 486.16) S. E. Kang, J. R. Banerjee, E. M. Maina and B. Mintz: Mater. Sci.

Technol., 29 (2013), 1225.17) S. E. Kang, J. R. Banerjee and B. Mintz: Mater. Sci. Technol., 28

(2012), 589.18) S. E. Kang, A. Tuling, J. R. Banerjee, W. D. Gunawardana and B.

Mintz: Mater. Sci. Technol., 27 (2011), 95.19) I. Mejía, A. E. Salas-Reyes, A. Bedolla-Jacuinde, J. Calvo and J. M.

Cabrera: Mater. Sci. Eng. A, 616 (2014), 229.20) A. E. Salas-Reyes, I. Mejía, A. Bedolla-Jacuinde, A. Boulaajaj, J.

Calvo and J. M. Cabrera: Mater. Sci. Eng. A, 611 (2014), 77.21) A. S. Hamada and L. P. Karjalainen: Mater. Sci. Eng. A, 528 (2011),

1819.

22) Y. Wang, J. Yang, X. Xin, R. Wang and L. Xu: Metall. Mater. Trans. B, 47 (2016), 1378.

23) Y. Wang, J. Yang, R. Wang, X. Xin and L. Xu: Metall. Mater. Trans. B, 47 (2016), 1697.

24) G. Gigacher, W. Krieger, P. R. Scheller and C. Thomser: Steel Res. Int., 76 (2005), 644.

25) J. H. Park, D. J. Kim and J. M. Dong: Metall. Mater. Trans. A, 43 (2012), 2316.

26) H. Liu, J. Liu, S. Michelic, F. Wei, C. Zhuang, Z. Han and S. Li: Ironmaking Steelmaking, 43 (2016), 171.

27) H. Liu, J. Liu, S. Michelic, S. Shen, X. Su, B. Wu and H. Ding: Steel Res. Int., 87 (2016), 1723.

28) Y. H. Xiong, P. J. Li, A. M. Yang, W. D. Yan, D. B. Zeng and L. Liu: Acta Metall. Sin., 38 (2002), 529.

29) C. W. Bale, P. Chartrand, S. A. Degterov, G. Eriksson, K. Hack, R. Ben Mahfoud, J. Melancon, D. Pelton and S. Peterson: Calphad, 26 (2002), 189.

30) C. W. Bale, E. Belisle, P. Chartrand, S. A. Decterov, G. Eriksson, K. Hack, I.-H. Jung, Y.-B. Kang, J. Melancon, A. D. Pelton, C. Robelin and S. Peterson: Calphad, 33 (2009), 295.

31) R. T. DeHoff and F. N. Rhines: Quantitative Microscopy, McGraw-Hill, New York, (1968), 158.

32) B. Lv, F. C. Zhang, M. Li, L. H. Qian and T. S. Wang: Mater. Sci. Eng. A, 527 (2010), 5648.

33) P. J. Wray: Metall. Trans. A, 15 (1984), 2009.34) J. P. Sah, G. J. Richardson and C. M. Sellars: Met. Sci., 8 (1974),

325.35) D. N. Crowther and B. Mintz: Mater. Sci. Technol., 2 (1986), 951.36) A. I. Fernández, P. Uranga, B. López and J. M. Rodriguez-Ibabe:

Mater. Sci. Eng. A, 361 (2003), 367.37) B. Mintz and D. N. Crowther: Int. Mater. Rev., 55 (2010), 168.38) B. Mintz, S. Yue and J. J. Jonas: Int. Mater. Rev., 36 (1991), 187.

Documents

Solidification Structure, Non-metallic Inclusions and Hot