Home Laboratory Study on Oxide Inclusions in High-Strength Low-Alloyed Steel Refined by Slag with Basicity 2–5
Article Open Access

Laboratory Study on Oxide Inclusions in High-Strength Low-Alloyed Steel Refined by Slag with Basicity 2–5

  • Huixiang Yu EMAIL logo , Xinhua Wang , Jing Zhang and Wanjun Wang
Published/Copyright: January 10, 2015

Abstract

Non-metallic inclusions in high-strength low-alloyed steel refined by slag with basicity (B) 2–5 and Al2O3 content 20%–30% (in mass percent) were investigated by slag–metal equilibrium experiments in laboratory and thermodynamic calculations. Most inclusions in equilibrium were quasi-spherical CaO-MgO-Al2O3-SiO2 system and the sizes were less than 5 µm. The average content ratio, CaO/Al2O3 and CaO/SiO2 of the inclusions, increased with slag basicity increasing when Al2O3 content in slag was around 25%, MgO/Al2O3 and CaO/Al2O3 increased with Al2O3 content in slag decreasing when slag basicity was around 3.3. The MgO/Al2O3 of the inclusions was influenced mainly by Al2O3 content in slag and CaO/Al2O3 was influenced mainly by slag basicity. To make the inclusions out of relative lower melting region (<1,773 K), the values of CaO/Al2O3, MgO/Al2O3 and (CaO+MgO)/Al2O3 of the inclusions should be enhanced by increasing basicity and decreasing Al2O3 content in slag properly.

PACS number: 81.05.Bx

Introduction

High-strength low-alloyed (HSLA) steel plates are widely used in shipbuilding and offshore platform and are strictly required for strength, low-temperature toughness and ductility properties. HSLA steels are usually Al-killed and refined by slag of high basicity (B=%CaO/%SiO2, in mass percent) and containing high Al2O3 content [14]. As a result, inclusions of MgO-Al2O3-CaO, CaO-Al2O3 and CaO-CaS-Al2O3 system are usually formed including calcium aluminates with low melting temperature, such as 12CaO·7Al2O3 and CaO·Al2O3 [5, 6]. Inclusions with low melting point can be elongated to stringers during subsequent rolling, which are the main reason that deteriorate the properties of ductility, anti-hydrogen-induced cracking and anti-stress corrosion cracking of the steel plates [7, 8]. The detrimental effect of inclusions becomes more apparent with increasing strength level of HSLA steel.

Wang et al. and Son et al. studied the transformation of inclusions in steel during aluminum deoxidation process as well as the refining process with high basicity and Al2O3-bearing slags [57, 9, 10]. During these processes, MgO and CaO in slag and ladle refractory can be reduced by the dissolved aluminum in the molten steel, resulting in the pick-up of [Mg] and [Ca] in the steel to several mass ppm, so that deoxidation products of Al2O3 gradually change to inclusions of MgO-Al2O3, MgO-Al2O3-CaO or CaO-Al2O3.

Researches on the effect of refining slag on inclusions in equilibrium have been reported. Slags refined with <2 basicity and <20% Al2O3 were reported by Wang et al. and Jiang et al. [1113]. Chen and Jiang et al. investigated the effect of high basicity and Al2O3 content (B=8, Al2O3≥35%), and studied inclusions in 0.6%C-1.5%Si-0.8%Mn steel refined by slags with basicity 3.5 and containing 25% Al2O3 as well as those in 0.4%C-0.25%Si-0.65%Mn-1.1%Cr-0.2Mo steel under slag with basicity 5 and containing 25% Al2O3 [1416]. They reported that inclusions in equilibrium with these slags were mainly MnO-SiO2-Al2O3-CaO and CaO-MgO-Al2O3-SiO2 with low melting point; inclusions in the steel refined with 3.5 or 5 basicity and 25% Al2O3 slag contained more MgO and had relatively higher melting temperature.

Very few researches were reported that studied inclusions in HSLA steel refined by 2–5 basicity slag, and whether calcium aluminate inclusions with low melting point can be effectively lowered by the refining slag with relative lower basicity and Al2O3 content was rarely reported. For this reason, the current paper investigates inclusions in HSLA steel refined by the slag with 2–5 basicity and 20%–30% Al2O3 using slag–metal equilibrium experiments.

Experiment

A vertical Si-Mo-heated high-temperature furnace was used for the equilibrium experiments between the metal and slag. Two hundred grams steel (0.045 mass% C, 0.19 mass% Si, 1.8 mass% Mn) and 40 g slag were melted together in MgO crucibles at 1,873 K under argon-protective atmosphere. During the experiments, 99.999% purity Ar gas was introduced into the reaction tube at a constant flow rate of 1.5 L/min. The slag was made by mixing reagent-grade CaO, Al2O3, SiO2 and MgO oxide powder in proportion together. The slag composition before experiments is shown in Table 1. Ninety minutes was needed to reach slag–metal equilibrium under 1,873 K, so that all experiments were performed for 90 minutes under 1,873 K without stirring to float inclusions out and establish equilibrium, and then the crucible was taken out from the furnace and quenched rapidly in water.

The contents of silicon, manganese and phosphorus in steel sample were analyzed by inductively coupled plasma atomic emission spectroscopy (ICP-AES), and carbon and sulfur by infrared absorption method. The contents of CaO, MgO, Al2O3 and MnO in slag sample were analyzed by ICP-AES, FeO by titration method and SiO2 by gravimetric method. Thirty-five points on the cross-sectioned plane of each steel sample were randomly chosen to detect inclusions by scanning electron microscope and energy-dispersive x-ray spectrometer (EDS) to get statistical information of inclusions including morphology, size and chemical composition.

The effect of slag basicity and Al2O3 content in the slag on inclusions was investigated. As shown in Table 1, the basicity levels of 2, 3.5 and 5 with 25% Al2O3 were chosen to study the effect of basicity. The Al2O3 levels of 20%, 25% and 30% with 3.5 slag basicity were chosen to study the effect of Al2O3 content. MgO was added to the slag at a saturated level that was predicted by thermodynamic software Thermo-Calc.

Table 1

Original chemical composition of slags, mass%

Slag No.B (CaO/SiO2)Al2O3MgO
122511
23.5257
35256
43.5206
53.5308

Before equilibrium experiments, pre-equilibrium experiments were performed to decide the time to reach slag–steel equilibrium for the slag with 3.5 basicity, and containing 25% Al2O3 and saturated MgO under 1,873 K. In previous investigations [9, 1416], 90–120 minutes was always considered as the holding time for the equilibrium experiments between steel and top slag. In the present work, three holding times of 90, 90 and 120 minutes were chosen to decide the time for equilibrium and study the repeatability of equilibrium experiments under the current conditions.

Results

Pre-equilibrium experiments

The chemical composition of steel and slag samples of the three heats, with holding times of 90, 90 and 120 minutes, respectively, after equilibrium is shown in Tables 2 and 3, showing similar composition and low sulfur content for the three heats.

Table 2

Chemical composition of steel samples, mass%

Heat No.CSiMnPS
10.0440.181.770.00700.0004
20.0410.181.780.00800.0003
30.0400.171.780.00700.0003
Table 3

Chemical composition of slag samples, mass%

Heat No.CaOMgOSiO2Al2O3MnOFeOS%CaO/%SiO2
149.638.7415.2624.80.19<0.10.0153.25
250.218.6615.39250.18<0.10.0163.26
350.119.0415.3125.030.18<0.10.0173.27

The EDS detection indicated that inclusions in steel for the three heats mainly contained CaO, Al2O3, MgO and approximately 11% SiO2 as shown in Figure 1. It can be seen that the composition distribution of inclusions of the three heats had little difference.

Figure 1 Composition distribution of inclusions in steel samples of pre-equilibrium experiments. (a) 90 min holding; (b) 90 min holding; (c) 120 min holding
Figure 1

Composition distribution of inclusions in steel samples of pre-equilibrium experiments. (a) 90 min holding; (b) 90 min holding; (c) 120 min holding

Since the composition of steel, slag and inclusions for the three heats was approximately identical, the equilibrium can be reached after 90 minutes under the present experimental conditions, and the results showed good repeatability. Thus, the reaction time for the following experiments was set to 90 minutes.

Equilibrium experiments

The chemical composition of steel samples after equilibrium under the five slag conditions is shown in Table 4 and the chemical composition of the slag samples after equilibrium is listed in Table 5. Low contents of MnO and FeO suggest that the reduced atmosphere in the reaction tube was controlled well. For the slag conditions No. 1, 2 and 3 (Table 1), the basicity was 1.93, 3.26 and 4.54, respectively, after equilibrium, and Al2O3 content was similarly 24%–26%. For the slag conditions No. 4, 2 and 5 (Table 1), the Al2O3 content in the slag was about 21%, 25% and 30%, respectively, after equilibrium, and slag basicity after equilibrium was fixed at approximately 3.3.

Table 4

Chemical composition of steel samples, mass%

CompositionCSiMnPS
Concentration0.041–0.0480.17–0.191.76–1.820.0065–0.00820.0003–0.0010
Table 5

Chemical composition of slag samples, mass%

Slag No.B (CaO/SiO2)Al2O3MgOMnOFeO
11.9324.4813.650.34<0.1
23.26258.660.18<0.1
34.5426.129.560.16<0.1
43.2721.229.760.2<0.1
53.3130.0410.190.17<0.1

The typical morphology and corresponding EDS of inclusions in the steel samples equilibrated with slag No. 1–5 are shown in Figure 2(a)–(e), respectively, indicating spherical shape and <5 µm size. The boundaries of inclusions under slag No. 3 and 4 conditions are less smooth than those under slag No. 1, 2 and 5 conditions.

Figure 2 Typical morphology and EDS of inclusions detected in steel samples under slag No. 1–5. (a) Slag No. 1; (b) slag No. 2; (c) slag No. 3; (d) slag No. 4; (e) slag No. 5
Figure 2

Typical morphology and EDS of inclusions detected in steel samples under slag No. 1–5. (a) Slag No. 1; (b) slag No. 2; (c) slag No. 3; (d) slag No. 4; (e) slag No. 5

Most inclusions in steel samples after equilibrium were CaO-MgO-Al2O3-SiO2 system, and SiO2 content in inclusions was relative stable for each slag condition but varied with slag basicity, as shown in Figure 3, where the liquid regions of different temperature calculated using thermodynamic software Factsage are also plotted. For slag No. 1, 2 and 3 which had different basicity but identical Al2O3 content, the average SiO2 content in inclusions in equilibrium was 22%, 11% and 9%, respectively. The dots in Figure 3 represent the composition of separate inclusions and the stars are the average values. The curved regions stand for the liquid zones of different temperature in Celsius degree.

Figure 3 Composition distribution of inclusions under different basicity slag conditions. (a) Slag No. 1 (B: 1.93); (b) slag No. 2 (B: 3.26); (c) slag No. 3 (B: 4.54)
Figure 3

Composition distribution of inclusions under different basicity slag conditions. (a) Slag No. 1 (B: 1.93); (b) slag No. 2 (B: 3.26); (c) slag No. 3 (B: 4.54)

With the decrease of SiO2 content in CaO-MgO-Al2O3-SiO2 system, the liquid zone (<1,773 K) moved toward the direction of (CaO + MgO)/Al2O3 and decreases. The SiO2 content in inclusions decreased with the rise of slag basicity and (CaO + MgO) content increased. The inclusions were mainly liquid (<1,773 K) when the slag basicity was 1.93 or 3.26, while some inclusions entered higher temperature region (≥1,773 K) when the slag basicity was 4.54.

For slags No. 4, 2 and 5 which had different Al2O3 content but identical basicity of 3.3, the average SiO2 content in inclusions in equilibrium was approximately 11% as shown in Figure 4.

Figure 4 Composition distribution of inclusions under different Al2O3 slags. (a) Slag No. 4 (Al2O3: 21.22%); (b) slag No. 2 (Al2O3: 25%); (c) slag No. 5 (Al2O3: 30.04%)
Figure 4

Composition distribution of inclusions under different Al2O3 slags. (a) Slag No. 4 (Al2O3: 21.22%); (b) slag No. 2 (Al2O3: 25%); (c) slag No. 5 (Al2O3: 30.04%)

The inclusions under slag No. 4 condition were significantly different from those under slag No. 2 and 5. Most inclusions were liquid (<1,773 K) when Al2O3 in slag was about 25% or 30%, while many inclusions were solid (≥1,773 K) when Al2O3 was about 21%.

Discussion

Effect of slag basicity on inclusions

The ratios of components’ average content in inclusions, CaO/Al2O3, MgO/Al2O3 and CaO/SiO2, under different basicity slag conditions are shown in Figure 5. With the rise of slag basicity but identical Al2O3 content, the CaO/SiO2 of inclusions increased sharply with values of 1.39, 3.89 and 5.77, CaO/Al2O3 increased with values of 0.93, 1.08 and 1.42 and MgO/Al2O3 showed no trend with values of 0.24, 0.21 and 0.27, respectively. Among the three slag conditions, the inclusions equilibrated with slag No. 3 with basicity 4.54 had the highest value of CaO/Al2O3 and MgO/Al2O3.

Figure 5 The relationship between average content ratio of inclusions and slag basicity
Figure 5

The relationship between average content ratio of inclusions and slag basicity

Wang et al. [9] also investigated the effect of slag basicity on inclusions by slag–metal reactions in laboratory; the steel used in experiments was high-strength alloying steel (0.35C-0.24Si-0.62Mn-1.13Cr-0.23Mo). The average CaO content in inclusions increased and Al2O3 content decreased, and the ratio CaO/Al2O3 increased with increasing (CaO+MgO)/SiO2 in slag, while the ratio MgO/Al2O3 in inclusions showed no trend, as shown in Figure 6.

Figure 6 Influences of [Al] and slag basicity on composition of the inclusions [9]
Figure 6

Influences of [Al] and slag basicity on composition of the inclusions [9]

Effect of Al2O3 content in slag on inclusions

The ratios CaO/Al2O3, MgO/Al2O3 and CaO/SiO2 of inclusions under the slag conditions containing different Al2O3 content are shown in Figure 7. With decreasing Al2O3 content in slag but identical slag basicity of 3.3, the MgO/Al2O3 of inclusions increased greatly with values of 0.15, 0.21 and 0.57, CaO/Al2O3 increased a little with values of 1.05, 1.08 and 1.13 and CaO/SiO2 showed no trend with values of 3.55, 3.89 and 3.21, respectively. Among the three slag conditions, the inclusions equilibrated with slag No. 4 containing about 21% Al2O3 had the highest value of CaO/Al2O3 and MgO/Al2O3.

Figure 7 The relationship between average content ratio of inclusions and Al2O3 content in slag
Figure 7

The relationship between average content ratio of inclusions and Al2O3 content in slag

The effect of Al2O3 content in slag on inclusions obtained in the present work agrees with the research of Wang et al. [9]. The inclusions in high-strength alloying steels in equilibrium with slag A (6.5 basicity and 41% Al2O3) and slag B (5 basicity and 23% Al2O3) were investigated by slag–metal reactions in laboratory. The Al2O3 content in inclusions equilibrated with slag A was more than that with slag B, MgO content was less and the ratio MgO/Al2O3 in inclusions was lower, as shown in Figure 8.

Figure 8 Effect of slag on composition distribution of the inclusions [9]. (a) Slag A; (b) slag B
Figure 8

Effect of slag on composition distribution of the inclusions [9]. (a) Slag A; (b) slag B

The relationship between log (XMgO/XAl2O3) of inclusions and log (aMgO/aAl2O3) of slags at 1,873 K is shown in Figure 9, where Xi is the mole fraction of i in inclusions and ai is the activity of i in slags calculated using thermodynamic software Thermo-Calc. The calculated conditions were as follows: temperature (T), 1,873 K; pressure (P), 101,325 Pa; B, 3.3; MgO, saturated; Al2O3, 21%, 25% and 30%. The log (XMgO/XAl2O3) of inclusions exhibited a linear relation to log (aMgO/aAl2O3) of slags with the slope close to unity. These phenomena could reveal that the slag/metal/inclusion system investigated in the present work was in thermodynamic equilibrium. The results were identical to that of Jiang et al. [15] and Park et al. [17], as shown in Figure 10.

Figure 9 Composition of inclusions, log (XMgO/XAl2O3) as a function of log (aMgO/aAl2O3) of slag at 1,873 K
Figure 9

Composition of inclusions, log (XMgO/XAl2O3) as a function of log (aMgO/aAl2O3) of slag at 1,873 K

Figure 10 Composition of inclusions, log (XMgO/XAl2O3) as a function of log (aMgO/aAl2O3) at 1,823 K [17]
Figure 10

Composition of inclusions, log (XMgO/XAl2O3) as a function of log (aMgO/aAl2O3) at 1,823 K [17]

The method to increase the melting temperature of inclusions

The liquid zone (<1,773 K) moved toward the direction of (CaO + MgO)/Al2O3 decreasing with SiO2 content in CaO-MgO-Al2O3-SiO2 system decreasing, as shown in Figure 3. For the present work, the SiO2 content in inclusions was mainly influenced by slag basicity. For example, the basicity of slag No. 1–5 after equilibrium was 1.93, 3.26, 4.54, 3.27 and 3.31, and the average SiO2 content in inclusions was approximately 22%, 11%, 9%, 11% and 11%, respectively. The CaO/Al2O3 in inclusions was more than 1 when slag basicity was ≥3.26 as shown in Figure 5. For HSLA steel, the slag basicity should not be low to make high efficient desulfurization. So to make the inclusions out of the liquid region (<1,773 K), the inclusions should move toward the direction of (CaO+MgO)/Al2O3 increasing, that is to say, the ratios CaO/Al2O3, MgO/Al2O3 and (CaO+MgO)/Al2O3 of inclusions should be increased.

The ratios CaO/Al2O3, MgO/Al2O3 and (CaO+MgO)/Al2O3 of inclusions under the five slag conditions are shown in Figure 11. The inclusions in equilibrium with slag No. 4 had the highest (CaO+MgO)/Al2O3 and MgO/Al2O3 and the second highest CaO/Al2O3, then lots of inclusions were solid (≥1,773 K). The inclusions in equilibrium with slag No. 3 had the second highest (CaO+MgO)/Al2O3 and MgO/Al2O3 and the highest CaO/Al2O3, then part of inclusions were solid (≥1,773 K). The inclusions in equilibrium with slag No. 12 and 5 had relatively lower (CaO+MgO)/Al2O3, MgO/Al2O3 and CaO/Al2O3, then most inclusions were liquid (<1,773 K).

Figure 11 The average content ratio of inclusions under the five slag conditions
Figure 11

The average content ratio of inclusions under the five slag conditions

As shown in Figures 5, 7 and 11, the CaO/Al2O3 of inclusions increased with the rise of slag basicity and the MgO/Al2O3 increased with Al2O3 in slag decreasing, and the CaO/Al2O3 was mainly influenced by slag basicity and the MgO/Al2O3 was influenced mainly by Al2O3 content in slag. So the values of CaO/Al2O3, MgO/Al2O3 and (CaO+MgO)/Al2O3 of inclusions should be enhanced by increasing basicity and decreasing Al2O3 content in slag properly to make the inclusions solid (≥1,773 K).

Conclusions

Laboratory-scale experiments as well as thermodynamic calculations have been done to investigate the effect of refining slags (B: 2–5, Al2O3: 20%–30%) on oxide inclusions in HSLA steel. The following conclusions were obtained:

  1. Oxide inclusions in equilibrium with the slags were mainly quasi-spherical CaO-MgO-Al2O3-SiO2 system with size less than 5 µm.

  2. The average content ratios, CaO/Al2O3 and CaO/SiO2 of inclusions, increased with the rise of slag basicity when Al2O3 content in slag was around 25%. The MgO/Al2O3 and CaO/Al2O3 of inclusions increased with Al2O3 content in slag decreasing when basicity was around 3.3. The MgO/Al2O3 of inclusions was influenced mainly by Al2O3 content in slag and the CaO/Al2O3 was influenced mainly by slag basicity.

  3. The inclusions under slag No. 3 and 4 conditions had relatively higher values of (CaO+MgO)/Al2O3, MgO/Al2O3, and CaO/Al2O3. As a result, lots of inclusions were solid (≥1,773 K). While the inclusions under slag No. 1, 2 and 5 conditions had relatively lower values of (CaO + MgO)/Al2O3, MgO/Al2O3 and CaO/Al2O3, most inclusions were liquid (<1,773 K).

  4. To make the inclusions out of relative lower melting region (<1,773 K), the ratios, CaO/Al2O3, MgO/Al2O3 and (CaO + MgO)/Al2O3 of inclusions, should be enhanced by increasing basicity and decreasing Al2O3 content in slag properly.

Funding statement: Funding The authors are grateful to the team of National Basic Research Program of China (No. 2010CB630806) and State Key Laboratory of Advanced Metallurgy (USTB) (No. 41603015) for financial support.

References

1. WangXH. Proceedings of national steelmaking-continuous casting technology meeting, July 30–August 1, 2008, Hangzhou, The Chinese Society for Metals, 2008.Search in Google Scholar

2. OguraY, KikuchiY, HasegawaT, et al. Tetsu-to-Hagane1986;72:130915.10.2355/tetsutohagane1955.72.9_1309Search in Google Scholar

3. BannenbergN, BergmannB, GayeH. Steel Res1992;63:4317.10.1002/srin.199201737Search in Google Scholar

4. ChenWJ, ChenJS, ZhengJZ. CSM 2009 annual meeting proceedings, November 11–13, 2009, Beijing, The Chinese Society for Metals, 2009.Search in Google Scholar

5. SonJH, JungIH, JungSM, et al. ISIJ Int2008;48:154251.10.2355/isijinternational.48.1542Search in Google Scholar

6. LiQ, WangXH, LiHB, et al. J Iron Steel Res Int2011;18:5636.10.1016/S1006-706X(11)60004-6Search in Google Scholar

7. WangXH, LiXG, LiQ, et al. Steel Res Int2014;85:15563.10.1002/srin.201300044Search in Google Scholar

8. ZhuGS, LiHB, LiBH, et al. Proceedings of the 5th international congress on the science and technology of steelmaking, October 1–3, 2012, Dresden, ENGICOM GmbH, 2012.Search in Google Scholar

9. WangXH, ChenB, JiangM, et al. Proceedings of the 4th international congress on the science and technology of steelmaking, October 6–8, 2008, Gifu, The Iron and Steel Institute of Japan, 2008.Search in Google Scholar

10. YuHX, WangXH, ZhangJ, et al. J Iron Steel Res Int2011;18:611.10.1016/S1006-706X(12)60002-8Search in Google Scholar

11. WangLF, ZhangJM, WangXH, et al. J Univ Sci Technol B2004;26:269.Search in Google Scholar

12. JiangZH, LiY, JiangMF. CSM 2003 annual meeting proceedings, October 28–30, 2003, Beijing, The Chinese Society for Metals, 2003.Search in Google Scholar

13. ChenSH, JiangM, HeXF, et al. J Miner Metall Mater2012;19:4908.10.1007/s12613-012-0585-3Search in Google Scholar

14. ChenB. Investigation on liquid steel–slag equilibrium and the non-metallic inclusions in alloying structural steels [Dissertation for Doctor degree]. Beijing: University of Science and Technology Beijing, 2008.Search in Google Scholar

15. JiangM, WangXH, ChenB, et al. ISIJ Int2008;48:88590.10.2355/isijinternational.48.885Search in Google Scholar

16. WangXH, JiangM, ChenB, et al. Sci China Technol Sci2012;55:186372.10.1007/s11431-012-4874-xSearch in Google Scholar

17. ParkJH, KimDS. Metall Mater Trans B2005;36B:495502.10.1007/s11663-005-0041-0Search in Google Scholar

Received: 2014-7-1
Accepted: 2014-10-9
Published Online: 2015-1-10
Published in Print: 2015-11-1

©2015 by De Gruyter

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.

Downloaded on 12.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/htmp-2014-0116/html
Scroll to top button