Issue
Metall. Res. Technol.
Volume 119, Number 2, 2022
Article Number 209
Number of page(s) 9
DOI https://doi.org/10.1051/metal/2022021
Published online 11 March 2022

© EDP Sciences, 2022

1 Introduction

The control of non-metallic inclusions in steelmaking process is essential due to the increasing demand of high quality steel products. Metallurgists try to produce steel with less and smaller inclusions to reduce nozzle clogging during the process of continuous casting and to improve mechanical properties of plate products. The concept of oxide metallurgy, i.e. inclusion engineering, has attracted the interest of steel industries for the development of high-quality steels, especially high heat input welding shipbuilding steels [1,2], where the microstructure is crucial to final mechanical properties. Non-metallic inclusions can play as nucleation particle for the formation of acicular ferrite during cooling, which increases the toughness of the coarse grain inheat affected zone (HAZ) of the weld. Due to the controll of composition and size of inclusions during the steelmaking process, the product quality can be greatly improved [3].

Among all inclusions, finely dispersed Ti-oxides, Mg-oxides, nitrides and rare earth oxides and sulfides are potential nucleation particles to refine the grain of steel [4]. To use these inclusions to improve the steel performance, a good control of their size and distribution is required [5]. There were a number of studies [612] to modify the inclusions by the addition of Ti to steel. Some researchers [11,13] concluded that an increased Ti containing inclusion number density in the steel leads to a finer solidification structure. The cooling rate affects the inclusion number density and inclusion size greatly. With the cooling rate increasing, the inclusion number density is increased and the size becomes smaller [10,13]. During the steel solidification process, the number and size of TiN precipitated is crucial in the inclusion metallurgy [14]. Grain boundary pinning with fine TiN precipitates is one of the inclusions that have been used to reduce the austenite grain size to increase the austenite grain boundary area [15], thereby improving the impact toughness of low-carbon micro-alloyed steels [16].

The premise of positively utilizing inclusions is that the particles have a sufficiently small size and homogenously dispenses in steel matrix. It is therefore necessary to understand the growth behavior of inclusions in the steelmaking process. During deoxidation, the added deoxidant reacts with the dissolved oxygen in the molten steel and nuclei of non-metallic inclusions are formed as long as the critical supersaturation for nucleation is reached. After the nucleation, inclusion growth is mainly governed by (i) diffusion and Ostwald ripening and (ii) collision and subsequent coagulation induced by Brownian movement, Stokes rising and turbulent flow of the molten steel [1719]. After the growth by diffusion mechanism, the growth can mainly be attributed to Ostwald ripening [18,20,21]. Suito et al. [18,2225] have systematically investigated the growth behavior of different inclusions [2628] in Fe-10% Ni alloy steel. These inclusions were forgenerated by pure metals (e.g. Al, Ti, Mg, Zr and Ce) and alloys (e.g. Ca-Al, Si-Mn and Ti-Mg) deoxidation. However, the complex deoxidation products in the Si-Mn-Al-Ti system have not yet been studied. SiMn, Al and 70TiFe deoxidation are widely used and complex inclusions are often generated in steelmaking practices. The growth behavior of the inclusions at the beginning of the deoxidation process is particularly important due to the fact that it determines the size and shape of inclusions.

In the present work, the inclusion growth for the complex deoxidation product of SiMn-Al-Ti has been studied. SiMn alloy was added as the first deoxidant followed by 70TiFe addition. The changes in size, composition and shape of the Al-Ti inclusions as function of time were studied by two-dimensional (2D) and three-dimensional (3D) characterization methods. The effect of Al impurity in the 70TiFe alloy on the inclusion characteristics is also investigated in this work. Subesequently, the mechanism of inclusion growth is discussed.

2 Experimental

The steel specimens were prepared using a vertical tube furnace (Fig. 1). A mixture of high purity electrolytic iron (99.99%) and reagent-grade graphite powder were charged into a graphite crucible for preparation of the Fe-C alloy. This mixture was subsequently melted at 1823 K for 4 h under argon gas atmosphere at a flow rate of 200 ml/min. The prepared Fe-C alloy is used in the deoxidation experiment in Al2O3 crucible. The deoxidation is done by adding a silico-manganese (SiMn) alloy followed by addition of an Al-containing 70TiFe alloy (Tab. 1.) The composition of the steel and the alloys are shown in Tables 2 and 3. A steel sample was taken into the molten bath by dipping a sampler and rapidly withdrawing it and quenching in water. Immediately after adding the SiMn alloy, a sample was taken. Subsequent samplings were done at respectively 30 s, 1 min, 2 min and 5 min after the addition of the 70TiFe alloy.

Inclusions were extracted from the steel matrix by dissolving a steel sample of about 0.3 g in a HCl (1:1) solution at 353 K (80 °C), resulting in complete dissolution of the steel matrix. The solution is then filtered on a 0.2 μm filtration membrane to separate the insoluble fraction from the solution. The composition and morphology of inclusions on the filter were then analyzed by the scanning electron microscope (SEM XL-30 FEG) equipped with the energy dispersive spectrometer (EDS).

The two-dimensional inclusion size and composition was determined by SEM and by automatic inclusion analysis (AIA) on the electron probe microanalysis (EPMA JXA-8530F). The effective minimum detectable inclusion area was around 0.5 µm2. It was analyzed for at least 100 inclusions and 1 mm2 of steel matrix area.

The inclusion phase diagram for the Fe-Al-Ti-O system was calculated with the Thermo-Calc Version 2019a software. The data used for this calculation is shown in Tables 4 and 5.

thumbnail Fig. 1

Experimental set-up and procedure (a) vertical tube furnace used to perform the deoxidation experiments; (b) the heating, alloying and sampling procedure (the time and red arrows show interval time for the experiment process); (c) steel samples taking in the experiment.

Table 1

The weight of materials used in the experiment (g).

Table 2

Chemical composition of the designed steel specimen in this work (wt.%).

Table 3

Composition of the alloys used for the steel deoxidation (wt.%).

Table 4

Reactions in the Fe-Al-Ti-O system with standard free energy changes [2932].

3 Results and discussion

3.1 Al in 70TiFe alloy

Due to the fact that the 70TiFe alloy is produced through the aluminothermic method, it contains 4 wt.% Al (Tab. 3). As Al is a strong reducing element, it has an important influence on the formation of inclusions. If Al exists in the form of inclusions in this alloy, these inclusions can end up as harmful inclusions in the steel. On the other hand if Al is present as metallic Al, the aluminium oxidation reaction occurs after the addition, which affects the formation behaviour of the Ti containing inclusions. The nature of Al in the alloy was therefore analyzed.

The microstructure and composition of the phases in the 70TiFe alloy are shown in Table 6 and Figure 2 respectively. Al is found in solid solution in the matrix and the Ti-rich phase. As no aluminum oxide inclusions were found in the alloy, it is concluded that Al in the added 70TiFe alloy exists in the form of elemental Al (dissolved Al).

Table 5

Interaction coefficients [32,33].

Table 6

Composition of the phases in the 70TiFe alloy (wt.%).

thumbnail Fig. 2

Section of 70TiFe alloy.

3.2 Chemical composition of the inclusions in the steel

The evolution of size and composition of the inclusions during the deoxidation was characterized by SEM and EPMA on cross-sectioned and polished samples (Fig. 3). The size of the inclusions increases with the time after deoxidation/alloying. 10 min after SiMn addition, the steel sample contains spherical inclusions with a diameter less than 2.5 µm (Fig. 3a). These are SiO2-MnO-Al2O3 containing inclusions. The Al2O3 is coming from the alumina crucible as this is the only source of Al or Al2O3 in this time period.

Once the 70TiFe alloy is added, the composition of the inclusions changes to the Al-Ti-N-O system. 30 seconds after adding the 70TiFe alloy, the core of the inclusions is composed of AlxTiyO precipitates which are surrounded by TiN precipitates (Fig. 3b). Due to the presence of Al in the 70TiFe alloy, the Al content in the inclusions has increased. The small amount of Si and Mn in the inclusions indicates that some of the Si and Mn in the inclusions have not been completely reduced. 1 min after the addition of 70TiFe, the size of the AlxTiyO inclusions has not further increased (Fig. 3c). 2 min after adding the 70TiFe alloy, TiN has mainly precipitated around Al2O3 inclusions rather than having AlxTiyO inclusions as the core (Fig. 3d). And also at 5 min after adding the 70TiFe alloy, TiN is usually wrapped around the Al2O3 inclusions (Fig. 3e) 35 min after adding the 70TiFe alloy, the steel sample was furnace-cooled. The inclusions, especially the TiN inclusions, became larger, up to 35 µm (Fig. 3f).

This experiment shows that the composition of the inclusions remains in the Al-Ti-N-O system once the 70TiFe alloy is added, but that the core of the TiN precipitation changes from AlxTiyO precipitates to Al2O3 precipitates.

The chemical composition of the inclusions acting as a function of time after deoxidation is plotted in Figure 4. The content of SiO2 and MnO in the inclusions decreases immediately after the 70TiFe addition. There are no SiO2 and MnO containing inclusions observed 1 min after adding 70TiFe. The Al2O3 content in the inclusions increases after deoxidation by 70TiFe. The Al in the 70TiFe alloy reduces the SiO2 and MnO in the inclusions, retards the formation of TiOx and leads to the formation of Al2O3 inclusions. The Ti content in the inclusions decreases as a result of the Ti oxide reduction by Al. On the other hand, when the sample is cooled slowly in the furnace 35 min after deoxidation, the Ti content in the inclusions is increased, whereas the Al2O3 content is decreased. This is due to the more TiN inclusions formation as there was sufficient time for Ti and N to react under the furnace cooling condition (cooling rate: 5 °C/min) compared with the quenched samples (cooling rate: 1500 °C/min–9000 °C/min).

For steels containing ∼0.020% to 0.045% Ti and ∼0.005% to 0.020% Al, the stable phases are Al2TiO5 and Al2O3 according to the stability diagram of the oxides in the Fe-Al-Ti-O system at 1600 °C (Fig. 5). Only at low Al contents (e.g. steels containing ∼0.002% to 0.007% Al), Al2TiO5 can be formed which contains ∼0.020% to 0.045% Ti. Al2O3 inclusions generates when the content of Al in steel decreases. This is why after short addition time, Al2TiO5 is observed, whereas for longer times after Ti deoxidation, more Al2O3 particles are observed.

thumbnail Fig. 3

Morphology and composition of the inclusions in the steel with different alloying time (a: 10 min after adding SiMn alloy; b to e: 30 s, 1 min, 2 min, 5 min respectively after adding FeTi70 alloy; f: furnace-cooled sample).

thumbnail Fig. 4

Chemical composition of the inclusions and Ti content present in inclusions with increasing time after the alloy addition.

thumbnail Fig. 5

Stability phase diagram of oxides in the liquid Fe-Al-Ti-O system at 1600 °C.

3.3 Morphology of the inclusions

The three-dimensional morphology of the inclusions was analyzed by extracting them from the samples quenched at 30 s, 1 min, 2 min and 5 min after adding the 70TiFe alloy. Figures 6a and 6b shows the composition and morphology of the inclusions at 30 s. The inclusions are spherical or quasi-spherical and are smaller than 2 µm. When the inclusions contain more Al2O3, the morphology of the inclusions are more like to spherical. TiN precipitates are angular and they precipitate onto the AlxTiyO spherical particles. TiN precipitation therefore modifies the inclusion morphology.

Several AlxTiyO inclusions larger than 5 µm in size can be found at 1 min after adding the 70TiFe alloy (Figs. 6c and 6d). This suggests that the AlxTiyO inclusions grow fast. The shape of the inclusions is spherical or quasi-spherical. When less TiN particles precipitate onto the AlxTiyO, the shape is also closer to a sphere.

After deoxidation for 2 min (Figs. 6e and 6f), the size of the inclusions is larger than that for 1 min. A number of inclusions are more than 3 µm and in particular Al2O3-rich inclusions can be larger than 10 µm. The shape of the Al2O3-rich inclusions also changes from spherical to irregular shapes with facetted planes.

After deoxidation for 5 min, the number of large irregular inclusions increases. Al2O3-rich inclusions larger than 10 µm in size can be found in this case (see Fig. 6h). Angular TiN precipitates are observed on these inclusions. In addition, TiN inclusions also tend to grow into large-sized inclusions (the changes in inclusion size with deoxidation time in Fig. 7).

The precipitation of angular TiN on the oxide inclusions changes the inclusion shape factor. The circularity (CFk) paramenter is used to valuate the effect of TiN inclusions on the shape. The calculation method of the circularity CFk is expressed as equation (1).(1)Pk and Ak are the perimeter and area of inclusion k, respectively. They are measured by the Image J software. The smaller CFk, the more prominent is the sharp corner of the inclusions. The average CFk decreases from 0.73 to 0.30 with increasing alloying time (Fig. 8). After Fe-Ti deoxidation from 30 s to 2 min, the decrease of the average CFk is slow, while the decrease from 2 min to 5 min is relatively large. This result corresponds to the three-dimensional morphology of the inclusions and one of the reasons is that more irregular TiN particles precipitate onto the Al2O3 surface with longer oxidation time.

thumbnail Fig. 6

Morphology and element mapping of the inclusions at different time (a: spherical AlxTiyO particles 30 s; b: quasi-sphericals AlxTiyO particle with TiN precipitates attached on the particle surface 30 s; c: spherical AlxTiyO particles at 1 min; d: quasi-sphericals AlxTiyO particle with TiN precipitates attached on the particle surface at 1 min; e: quasi-spherical AlxTiyO particles at 2 min; f: quasi-spherical Al2O3-rich particles with TiN precipitates attached on the particle surface at 2 min; g: Angular TiN particles at 5 min; h: sharp corners Al2O3-rich particles at 5 min).

thumbnail Fig. 7

Average size of inclusions with time.

3.4 Growth of the inclusions

Figure 7 shows the average size of the inclusions acting as a function of time. In the period of 30 s to 5 min after adding the 70TiFe alloy, the average inclusion size increases rapidly up to 1.9 µm. Between 5 and 35 min, the growth rate of the inclusions is slower than that of the previous period. While the number of inclusions with a size below 1.9 µm linearly decreases with time. However, the volume density of inclusions with a larger size (>2 μm) increases with time (Fig. 9).

According to the work of Zhang and Pluschkell [34] and Xuan et al. [35], the coalescence process between inclusions happens in the molten steel due to the following multiple collisions mechanisms: (1) Brownian collisions, ; (2) Stokes collisions, ; and (3) turbulent collisions, . The respective collision volume values of , and can be evaluated by using equations (2) to (5), respectively [3641]:(2) (3) (4) (5)where k is the Boltzmann constant (= 1.3807 × 10−23 J/K), T is the temperature (= 1873 K), µ is the dynamic viscosity of the steel (= 0.006 kg/m · s) [36] and g is the gravitational acceleration (= 9.81 m/s2). ρf and ρox are the densities of the liquid steel (=7100 kg/m3) and inclusions (=3950 kg/m3), respectively. The parameter ϵ is the turbulent energy dissipation rate (= 0.01 m2/s3) [42], and αt is the agglomeration coefficient, A121 is the Haymaker constant in the liquid steel (= 14.3 × 10−19 J) [43], ri and rj are the radius of the two colliding inclusions.

According to the result of Zhang [43], if the size of the inclusions is within the nanoscale range, the growth is governed by Brownian collision and Ostwald ripening, especially at the period before 30 s after deoxidant addition. The inclusion average diameter is 0.84 to 1.9 microns according to Figure 7 from 30 s to 5 min. At the period of 30 s∼5 min after deoxidation, when ri is 0.42 µm and rj is 0.95 µm for the radius of the two colliding inclusions, the collision volume values , and are 1.38 × 10−17 m3/s, 5.97 × 10−17 m3/s and 6.68 × 10−16 m3/s, respectively. For the Al-Ti-O-N inclusions, the growth of the inclusions is smaller than for pure Al2O3 inclusions and also much smaller than the calculated result according to the research of Xuan [34]. Nevertheless, the mechanism that affects the growth of inclusions is the same. Based on the results of this calculation, the growth of the inclusions is mainly controlled by turbulent collisions, where the value of turbulent collision volume is one order of magnitudelarger than that of Brownian and Stokes collisions. This result is consistent with the results of other researchers [37].

thumbnail Fig. 8

Shape circularity factor of inclusions with increased alloying time.

thumbnail Fig. 9

The evolution of the volume density of inclusions within different size ranges in the steels with deoxidation time.

4 Conclusions

The formation process and growth of Al-Ti-O-N inclusions are studied in this work. The deoxidation is done by silico-manganese followed by addition of Al-containing 70TiFe alloy. The evolution of the size, composition and shape of the inclusions with the time after 70TiFe addition was studied. The effect of Al in the FeTi70 alloy on the inclusion characteristics is also investigated in this work. The findings are:

  • After adding 70TiFe alloy, the inclusions changed from Si-Mn-Al inclusions to N-Ti-Al-O inclusions. The Al2O3 content in the inclusions increases till 5 min after 70TiFe addition and the Ti content in the inclusions decreases accordingly. When the the sample was cooled in the furnace 35 min after adding the 70TiFe alloy, more Ti was observed in the inclusions due to the precipiation of TiN.

  • The average size of the inclusions changes with time. The number of inclusions less than 2 µm linearly decreases with time. The growth of inclusions is controlled by the turbulent collision mechanism after 30 s.

  • The three-dimensional morphology of the inclusions after adding 70TiFe alloy was investigated by extracting from the quenched steel samples. Due to the precipiation of angular TiN precipitates on the Al2O3-containing inclusions, the average circularity decreases with increased alloying time.

Conflict of interest

The author declares no conflict of interest exists with the publication of this manuscript or its data.

References

  1. S. Shinichi, I. Katsuyuki, A. Toshikazu, High tensile strength steel plates with excellent HAZ toughness for shipbuilding − JFE EWEL technology for excellent quality in HAZ of high heat input welded joints. JFE technical report 5, 24 (2005) [Google Scholar]
  2. L. Zhang, T. Kannengiesser, Austenite grain growth and microstructure control in simulated heat affected zones of microalloyed HSLA steel, Mater. Sci. Eng. A 613, 326 (2014) [CrossRef] [Google Scholar]
  3. M. Kivio, L. Holappa, T. Iung, Addition of dispersoid titanium oxide inclusions in steel and their influence on grain refinement, Metall. Mater. Trans. B 41B, 1194 (2010) [CrossRef] [Google Scholar]
  4. Ø. Grong, C. van der Eijk, G. Tranell et al., Microstructure control of steels through dispersoid metallurgy using novel grain refining alloys, ISIJ Int. 46, 824 (2006) [CrossRef] [Google Scholar]
  5. C. Vandereijk, Ø. Grong, F. Haakonsen et al., Progress in the development and use of grain refiner based on cerium sulfide or titanium compound for carbon steel, ISIJ Int. 49, 1046 (2009) [CrossRef] [Google Scholar]
  6. J.-S. Byun, J.-H. Shim, Y.W. Cho et al., Non-metallic inclusion and intragranular nucleation of ferrite in Ti-killed C-Mn steel, Acta Mater. 51, 1593 (2003) [CrossRef] [Google Scholar]
  7. J.-H. Shim et al., Nucleation of intragranular ferrite at Ti2O3 particle in low carbon steel, Acta Mater. 47, 2751 (1999) [CrossRef] [Google Scholar]
  8. J.-H. Shim et al., Ferrite nucleation potency of non-metallic inclusions in medium carbon steels, Acta Mater. 49, 2115 (2001) [CrossRef] [Google Scholar]
  9. H.-H. Jin et al., Formation of intragranular acicular ferrite grains in a Ti-containing low carbon steel, ISIJ Int. 43, 1111 (2003) [CrossRef] [Google Scholar]
  10. H. Goto et al., Effect of cooling rate on composition of oxides precipitated during solidification of steels, ISIJ Int. 35, 708 (1995) [CrossRef] [Google Scholar]
  11. N. Kikuchi et al., Effect of Ti de-oxidation on solidification and post-solidification microstructure in low carbon high manganese steel, ISIJ Int. 47, 1255 (2007) [CrossRef] [Google Scholar]
  12. N. Kikuchi et al., Micro-structure refinement in low carbon high manganese steels through Ti-deoxidation: austenite grain growth and decomposition, ISIJ Int. 49, 1036 (2009) [CrossRef] [Google Scholar]
  13. N. Kikuchi et al., Micro-structure refinement in low carbon high manganese steels through Ti-deoxidation − inclusion precipitation and solidification structure, ISIJ Int. 48, 934 (2008) [CrossRef] [Google Scholar]
  14. J. Kunze et al., Precipitation of titanium nitride in low-alloyed steel during cooling and deformation, Steel Res. 68, 441 (1997) [CrossRef] [Google Scholar]
  15. H.S. Kim, C.-H. Chang, H.G. Lee, Evolution of inclusions and resultant microstructural change with Mg addition in Mn/Si/Ti deoxidized steels, Scr. Mater. 53, 1253 (2005) [CrossRef] [Google Scholar]
  16. W. Yan, Y.Y. Shan, K. Yang, Effect of TiN inclusions on the impact toughness of low-carbon microalloyed steels, Metall. Mater. Trans. A 37A, 2156 (2006) [Google Scholar]
  17. M. Wakoh, T. Sawai, S. Mizoguchi, Effect of S content on the MnS precipitation in steel with oxide nuclei, ISIJ Int. 36, 1014 (1996) [CrossRef] [Google Scholar]
  18. H. Ohta, H. Suito, Characteristics of particle size distribution of deoxidation products with Mg, Zr, Al, Ca, Si/Mn and Mg/Al in Fe-10 mass% Ni Alloy, ISIJ Int. 46, 14 (2006) [CrossRef] [Google Scholar]
  19. H. Lei, K. Nakajima, J.C. He, Mathematical model for nucleation, ostwald ripening and growth of inclusion in molten steel, ISIJ Int. 50, 1735 (2010) [CrossRef] [Google Scholar]
  20. H. Suito, R. Inoue, Thermodynamics on control of inclusions composition in ultra-clean steels, ISIJ Int. 36, 528 (1996) [CrossRef] [Google Scholar]
  21. H. Ohta, H. Suito, Dispersion behavior of MgO, ZrO2, Al2O3, CaO-Al2O3 and MnO-SiO2 deoxidation particles during solidification of Fe-10mass percent Ni alloy, ISIJ Int. 46, 22 (2006) [CrossRef] [Google Scholar]
  22. A.V. Karasev, H. Suito, Characteristics of fine oxide particles produced by Ti/M (MMg and Zr) complex deoxidation in Fe-10 mass %Ni alloy, ISIJ Int. 48, 1507 (2008) [CrossRef] [Google Scholar]
  23. H. Suito, H. Ohta, Characteristics of particle size distribution in early stage of deoxidation, ISIJ Int. 46, 33 (2006) [CrossRef] [Google Scholar]
  24. G.V. Pervushin, H. Suito, Effect of primary deoxidation products of Al2O3, ZrO2, Ce2O3 and MgO on TiN precipitation in Fe-10 mass %Ni alloy, ISIJ Int. 41, 748 (2001) [CrossRef] [Google Scholar]
  25. A. Karasev, H. Suito, Quantitative evaluation of inclusion in deoxidation of Fe-10 mass pct Ni alloy with Si, Ti, Al, Zr, and Ce, Metall. Mater. Trans. B 30B, 249 (1999) [CrossRef] [Google Scholar]
  26. Y. Jin, Z.Z. Liu, R. Takata, Nucleation and growth of alumina inclusion in early stages of deoxidation: numerical modeling, ISIJ Int. 50, 371 (2010) [CrossRef] [Google Scholar]
  27. W.T. Lou, M.Y. Zhu, Numerical simulation of gas and liquid two-phase flow in gas-stirred systems based on Euler-Euler approach, Metall. Mater. Trans. B 44B, 762 (2013) [CrossRef] [Google Scholar]
  28. J.H. Peng, D.Y. Su, C.X. Wang, Combined effect of aluminum content and layer structure on the oxidation performance of Ti1-xAlxN based coatings, J. Mater. Sci. Technol. 30, 803 (2014) [CrossRef] [Google Scholar]
  29. H. Ito, M. Hino, S. Ban-ya, Deoxidation equilibrium of magnesium in liquid iron, Tetsu-to-Hagane 83, 773 (1997) [CrossRef] [Google Scholar]
  30. W-Y. Cha, T. Miki, Y. Sasaki et al., Identification of titanium oxide phases equilibrated with liquid Fe-Ti alloy based on EBSD analysis, CAMP-ISIJ 19, 760 (2006) [Google Scholar]
  31. E.T. Tuikdogan, Physical Chemistry of High Temperature Technology, Academic Press, New York, 1980, 5, 22 [Google Scholar]
  32. The Japan Society for the Promotion of Science, The 19th Committee on the Steelmaking: Steelmaking Data Sourcebook, Gordon and Breach Science Publishers, New York, 1998, 279 [Google Scholar]
  33. G. Yuanchang, W. Changzhen, Interaction coefficients in Fe-C-Ti-i (i = Si, Cr, AI, Ni) systems, Metall. Trans. B 21B, 543 (1990) [CrossRef] [Google Scholar]
  34. L.F. Zhang, W. Pluschkell, Nucleation and growth kinetics of inclusions during liquid steel deoxidation, Ironmaking Steelmaking 30, 106 (2003) [CrossRef] [Google Scholar]
  35. C.J. Xuan, A.V. Karasev, P.G. Jonsson, Evaluation of agglomeration mechanisms of non-metallic inclusions and cluster characteristics produced by Ti/Al complex deoxidation in Fe-10 mass % Ni alloy, ISIJ Int. 56, 1204 (2016) [CrossRef] [Google Scholar]
  36. Y. Miki et al., Model of inclusion removal during RH degassing of steel, Iron Steelmaker 24, 31 (1997) [Google Scholar]
  37. L.F. Zhang, S. Taniguchi, K. Cai, Fluid flow and inclusion removal in continuous casting Tundish, Metall. Mater. Trans. B 31B, 253 (2000) [CrossRef] [Google Scholar]
  38. H. Lei, G. Xu, J. He, Magnetic field, flow field and inclusion collision growth in a continuous caster with EMBR, Chem. Eng. Technol. 30, 1650 (2007) [CrossRef] [Google Scholar]
  39. H. Lei, J. He, A dynamic model of alumina inclusion collision growth in the continuous caster, J. Non-Cryst. Solids 352, 3772 (2006) [CrossRef] [Google Scholar]
  40. L.F. Zhang, W. Pluschkell, Nucleation and growth kinetics of inclusions during liquid steel deoxidation, Ironmaking Steelmaking 30, 106 (2003) [CrossRef] [Google Scholar]
  41. J. Zhang, H. Lee, Numerical modeling of nucleation and growth of inclusions in molten steel based on mean processing parameters, ISIJ Int. 44, 1629 (2004) [CrossRef] [Google Scholar]
  42. K. Nakanishi, J. Szekely, Deoxidation kinetics in a turbulent flow field, Trans. Iron Steel Inst. Jpn. 15, 522 (1975) [CrossRef] [Google Scholar]
  43. L.F. Zhang, Nucleation, growth, transport, and entrapment of inclusions during steel casting, JOM 65, 1138 (2013) [CrossRef] [Google Scholar]

Cite this article as: Rensheng Chu, Zhanjun Li, Annelies Malfliet, Bart Blanpain, Muxing Guo, Growth mechanism of Al-Ti-O inclusions in steelmaking process, Metall. Res. Technol. 119, 209 (2022)

All Tables

Table 1

The weight of materials used in the experiment (g).

Table 2

Chemical composition of the designed steel specimen in this work (wt.%).

Table 3

Composition of the alloys used for the steel deoxidation (wt.%).

Table 4

Reactions in the Fe-Al-Ti-O system with standard free energy changes [2932].

Table 5

Interaction coefficients [32,33].

Table 6

Composition of the phases in the 70TiFe alloy (wt.%).

All Figures

thumbnail Fig. 1

Experimental set-up and procedure (a) vertical tube furnace used to perform the deoxidation experiments; (b) the heating, alloying and sampling procedure (the time and red arrows show interval time for the experiment process); (c) steel samples taking in the experiment.

In the text
thumbnail Fig. 2

Section of 70TiFe alloy.

In the text
thumbnail Fig. 3

Morphology and composition of the inclusions in the steel with different alloying time (a: 10 min after adding SiMn alloy; b to e: 30 s, 1 min, 2 min, 5 min respectively after adding FeTi70 alloy; f: furnace-cooled sample).

In the text
thumbnail Fig. 4

Chemical composition of the inclusions and Ti content present in inclusions with increasing time after the alloy addition.

In the text
thumbnail Fig. 5

Stability phase diagram of oxides in the liquid Fe-Al-Ti-O system at 1600 °C.

In the text
thumbnail Fig. 6

Morphology and element mapping of the inclusions at different time (a: spherical AlxTiyO particles 30 s; b: quasi-sphericals AlxTiyO particle with TiN precipitates attached on the particle surface 30 s; c: spherical AlxTiyO particles at 1 min; d: quasi-sphericals AlxTiyO particle with TiN precipitates attached on the particle surface at 1 min; e: quasi-spherical AlxTiyO particles at 2 min; f: quasi-spherical Al2O3-rich particles with TiN precipitates attached on the particle surface at 2 min; g: Angular TiN particles at 5 min; h: sharp corners Al2O3-rich particles at 5 min).

In the text
thumbnail Fig. 7

Average size of inclusions with time.

In the text
thumbnail Fig. 8

Shape circularity factor of inclusions with increased alloying time.

In the text
thumbnail Fig. 9

The evolution of the volume density of inclusions within different size ranges in the steels with deoxidation time.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.