Journal of Iron and Steel Research, International, 2014, 21 (Supplement 1)
Oxide Inclusions in Ferromanganese and Its Influence on the Quality of Clean Steels Pei-wei HAN1,
Shao-jun CHU1,
Ping MEI2, Yi-fei LIN2
(1. School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, Beijing 100083, China; 2. Jiangsu Jiangnan Ferroalloy Co., Ltd., Changzhou 213119, Jiangsu, China) Abstract: Low and medium carbon ferromanganese produced by oxygen decarburization process and electric silicothermic process was briefly introduced, and the quality of products by these two processes was analyzed. Results showed that the total oxygen content in medium carbon ferromanganese by electric silicothermic process in China, which ranged from 0.039% to 0.171%, was between those of the common and refined products by oxygen decarburization process outside of China. The increments of total oxygen content in liquid steel were estimated when ferromanganese was added for the purpose of Mn element adjustment at the end of smelting. Refined low and medium carbon ferromanganese, which had low total oxygen content, was recommended for composition adjustment of clean steels during final stage of a heat. It is possible that the inclusions in the ferromanganese alloy greatly influenced the quality of clean steel indirectly by affecting the amount, size and composition of inclusions in steel. Key words: low carbon ferromanganese; medium carbon ferromanganese; oxide inclusions; clean steels
in China. The aim of present investigation is to introduce domestic product quality of medium carbon ferromanganese, and the possible problems incurred by ferroalloy addition are discussed when they are applied for clean steel production.
Inclusion control has attracted more attention recently, as more rigorous requirements on steel cleanliness are continually emphasized with the modern industry development and technology progress. Recently, researches on inclusions in steel have mainly concentrated on the following two aspects [1]. One is inclusion removal and inclusion modification during steel refining process. The other is the influence of factors such as lining refractory material, mold flux and reoxidation. Ferroalloy addition is one of the important steps during secondary refining to deoxidize the steel or alloy. Limited by the purity of the ferroalloy, it is unavoidable to introduce some impurities such as inclusions into the steel, affecting the quality of the steel. However, there is very little information available concerning inclusions in ferroalloy and its influence on steel cleanliness during secondary steelmaking. Although many works have been done on the ferrosilicon [2, 3] , ferrochromium [4, 5] , ferromanganese [6-9] and some other ferroalloys [10], these work was not enough to figure out the effect of inclusion in ferroalloys. Especially, the inclusion strongly depended on the metallurgical process, but the information about the inclusions in the ferroalloys products produced in China was rarely reported. For example, the total oxygen content, an important composition for the steel production, is not even involved in existing ferroalloy national standards
1 Detection Method for Inclusions in Ferromanganese
The morphology of inclusions in ferroalloys such as size, shape, and composition was usually determined by metallographic method combined with scanning electron microscopy (SEM) [5, 6]. Although convenient, the information about quantity and threedimensional morphology were still difficult to obtain with metallographic method. Bulk sample electrolysis was employed to study the morphology of inclusions in ferroalloy in this work, which was commonly used to detect macro inclusions in steel. The electrolyte for ferromanganese sample electrolysis contained 0.8%–1.0% potassium bromide, 1.2%–1.5% ascorbic acid, 0.2 g/L sodium sulfite, 0.3% glycerol, 3%–4% sodium citrate and 100 g/L ammonium sulfate. In the meanwhile, the morphology and composition of inclusions in ferroalloy were analyzed by SEM and X-ray diffraction (XRD) analysis. The present investigation focused on detection of oxide inclusions in ferroalloy. The amount of oxide inclusions in ferroalloy could be determined indirectly
Foundation Item: Item Sponsored by National Natural Science Foundation of China (51274030) Biography: Pei-wei HAN, Doctor Candidate; E-mail:
[email protected]; Corresponding Author: Shao-jun CHU, Doctor, Professor; E-mail:
[email protected] 23
Oxide Inclusions in Ferromanganese and Its Influence on the Quality of Clean Steels
by T.O. (the total oxygen content). Rod samples were preferred for oxygen analysis instead of powder samples, because the ferroalloy components were susceptible to oxidation.
where, [Si] is dissolved silicon in the silicomanganese ferroalloy, (MnO) was manganese oxide in the manganese ore. Due to the difference of process, T.O. in LC and MC FeMn was fairly different, as shown in Table 1. T.O. in LC FeMn by oxygen decarburization process was as high as 1.11% in common product, or as low as 0.04% in refined products. And they were 0.89% and 0.02% in MC FeMn, respectively. While, T.O. in MC FeMn by electric silicothermic process in China ranged from 0.039 wt.% to 0.17 1wt.%, which was between the common and refined products by oxygen decarburization process outside of China. There were no actual products of refined LC FeMn and MC FeMn in China, and T.O. in laboratory products varied between 0.011% and 0.015%. Oxygen decarburization process was chosen to produce LC and MC FeMn in ferroalloy plants outside of China. In order to meet the requirement of steel plant, their products were divided into the common and the refined, depending upon the differences between oxygen content in the ferroalloy. Currently, LC and MC FeMn provided to domestic steel plant were produced with the electric silicothermic process, T.O. in which was between the common and refined products by oxygen decarburization process outside of China. As shown in Table 1, fluctuations of T.O. between different heats were fairly large, indicating that different process resulted in the difference in amount and distribution of oxide inclusions in MC FeMn. The SEM photo of oxide inclusions in MC FeMn was given in Fig. 1, in which corresponding composition of point A is 48 at.% O, 12 at.% Si, 40 at.% Mn, and corresponding composition of point B was 64 at.% O, 31 at.% Ca, 5 at.% Mn. The amount and size grading of inclusions
2 Quality Analysis of Ferromanganese Produced by Different Metallurgical Process
Oxygen decarburization process and shaking ladle-electric silicothermic process were the main production methods of LC FeMn (low carbon ferromanganese) and MC FeMn (medium carbon ferromanganese). In the oxygen decarburization process, molten high carbon ferromanganese produced in submerged arc furnace or blast furnace as raw materials was added into oxygen basic converter. Oxygen was blown into the molten ferromanganese through oxygen lance, and carbon was oxidized into carbon monoxoide. The desirable concentration according to the requirement of LC and MC, could be achieved by controlling the amount of blowing oxygen. The reaction principle was described as, 2[C]+O2=2CO (1) where, [C] is carbon dissolved in the molten ferromanganese. In the shaking ladle-electric silicothermic process, silicomanganese ferroalloy with low carbon content as intermediate ferroalloy was firstly produced in submerged arc furnace. The silicon in the ferroalloy acted as the reducing agent, which reduced the manganese oxide in the manganese ore, and the manganese content in the ferroalloy increased. When the silicon content decreased and met the standard requirement, MC and LC FeMn was obtained. The reaction could be expressed as, [Si]+2(MnO)+2(CaO)=(2CaO•SiO2)+2Mn (2)
Table 1 Chemical composition of low and medium carbon ferromanganese Ferromanganese process Oxygen
Chemical composition/%
Product Common
decarburization process
Refined
Electric
Common
Mn
Si
C
P
S
N
O
MC
81.10
0.12
0.77
0.17
0.004
0.07
0.89
LC
79.60
0.01
0.34
0.17
0.004
0.14
1.11
MC
81.00
0.39
1.38
0.16
0.003
0.07
0.02
LC
81.40
0.32
0.48
0.16
0.003
0.04
0.04
MC
78.39
0.99
1.40
0.18
0.010
-
silicothermic process
Refined
MC
82.25
0.20
1.43
24
0.18
0.010
-
0.039 - 0.171 0.011 - 0.015
Note
Ref. [6]
Product of a certain ferroalloy plant in China Laboratory product
Oxide Inclusions in Ferromanganese and Its Influence on the Quality of Clean Steels
shape, while those in alloys with the silicon content from 0.3% to 0.5% only had rhombic MnO[6, 9]. And at silicon content above 0.5%, the inclusions of complex compounds containing Mn, Si, S and O, rather than pure manganese oxide formed.
3 Discussion on Influence of Ferromanganese on Liquid Steel Composition Adjustment
Table 3[1, 3, 10, 11] shows typical steel cleanliness requirements for various steel grades on the amount of impurity elements content and inclusions, whose T.O. were all less than 20×10-6. The process of clean steels production included two different proceses: primary smelting process and refined process. When ferroalloy was used as deoxidizer or composition adjustment, oxide inclusions in ferroalloy might have a bad effect on quality of clean steels. Products of LC FeMn and MC FeMn were divided to the common and the refined outside of China, depending on the steelmaking process and quality requirements of various steel grades. In China, there were no such concepts and actual products of refined LC FeMn and MC FeMn. T.O. was not involved in existing ferroalloy national standards. For this reason, it was deserved to research on whether quality of clean steels was affected by the common LC FeMn and MC FeMn with electric silicothermic process.
Fig. 1 SEM photo of oxide inclusions in MC FeMn
in corresponding MC FeMn obtained by bulk sample electrolysis are shown in Table 2. Compared with product of oxygen decarburization process, the common oxide inclusions in domestic MC FeMn were composed of MnO and SiO2, which were probably tephroite (2MnO•SiO2). The morphology of inclusions mostly was long strip or square, and the size varied from 30 to 150 μm. The complex inclusions regularly accumulated with CaO and MnO inclusions, as shown in Fig. 1. MnO inclusions in MC FeMn of common product by oxygen decarburization process appeared as a crystalline rhombic shape or a dendritic morphology, and their size varied between 3 to 180 μm, as shown in Fig. 2[6]. It was reported MnO inclusions in ferromanganese with silicon content less than 0.3% showed both dendritic and rhombic
Table 2 Analysis of inclusions in medium carbon ferromanganese by bulk sample electrolysis Electrolyte
Weight loss/kg
Common used for steel electrolysis As described in section 1
Amount of inclusions mg mg/10 kg
Size grading of inclusions/mg <80 μm 80~140 μm 140~300 μm >300 μm
0.42
0.30
7.11
-
0.30
-
-
0.34
2.90
85.80
-
2.90
-
-
(a) Dendrite-shaped MnO inclusion; (b) Rhombic-Shaped MnO inclusion. Fig. 2 MnO inclusions in MC FeMn by oxygen decarburization process (area scanning)
25
Oxide Inclusions in Ferromanganese and Its Influence on the Quality of Clean Steels Table 3 Cleanliness requirement for various steel grades Steel grade
Maximum impurity content/10-6
Critical inclusion size/μm
Deep drawing steel
[C]≤20, [N]≤20, T.O.≤20
100
DI can steel
[C]≤30, [N]≤ 30, T.O.≤20
20
Pipeline steel
[S]≤10, T.O.≤15
Inclusion shape control
Tire cord steel
[H]≤2, [N]≤40, [Al]≤10, T.O.≤15
20
Bearing steel
T.O.≤10, [Ti]≤15
15
Heavy plate steel
[H]≤2, [N]≤30–40, T.O.≤20
Single inclusion, 13; cluster, 200
As shown in Table 4, refined ferromanganese was recommended to control the Mn concentration in the range from 0.1% to 0.5% (1.25 kg–6.25 kg ferromanganese/per ton steel) and the increment of T.O. less than 1×10-6. It was possible that the fine pure MnO inclusions in ferromanganese remained in steel due to the melting point of pure MnO (1844 ºC) above the liquid steel temperature, and a low floatation and removal rate[7]. It was also reported that ferromanganese grades did not significantly influence the amount of inclusions in steel, but affected the size and composition of inclusions[8]. Therefore, steel quality was affected by the inclusions in ferromanganese, when added as a final adjustment immediately before casting. Table 5[11] shows the cleanliness changes of pipeline steels at each process step. The amount of macro inclusions changed so much after the LF treatment. It was deserved to determine whether it had a relationship with composition adjustment by adding ferromanganese, except factors such as lining refractory material
Pipeline steel was taken as an example, to evaluate the increments of total oxygen content respect to Mn element adjustment with ferromanganese addition during secondary refining. Pipeline steels were used to transport crude oil or natural gas, and had a rigorous requirement on steel quality, as the composition and microstructure were “ultra-pure”, “ultra-homogenized”, “ultra-fine”. Oxide inclusion in pipeline steels was one of the sources of HIC (hydrogen induced cracking) and SSCC (sulfide stress corrosion cracking). In order to avoid oxide inclusions with size larger than 50 μm, T.O. was controlled less than 0.0015%. The process route of pipeline steels production in a certain plant was described as, blast furnace hot metal→hot metal desulfurization pretreatment → 250 t combined blown converter → ladle blowing Ar gas → RH vacuum treatment → LF (ladle furnace) treatment → continuous casting. Components were required to be in the narrow range of w [C]±0.015%, w [Mn]±0.15%. Composition adjustment was achieved during the LF treatment.
Table 4 Increment of total oxygen content of X70 pipeline steel respect to Mn element adjustment Increment of T.O. in steel/10-6 Steel grade
X70
[Mn]/%
T.O./10-6
1.4–1.7
≤5
Mn element adjustment/%
Ferromanganese by
Ferromanganese by
oxygen decarburization
electric silicothermic
Manganese
process
metal
process Common
Refined
Common
Refined
0.50
55.63
1.25
10.69
0.94
3.00
0.40
44.50
1.00
8.55
0.75
2.40
0.30
33.38
0.75
6.41
0.56
1.80
0.20
22.25
0.50
4.28
0.38
1.20
0.10
11.13
0.25
2.14
0.19
0.60
0.05
5.56
0.13
1.67
0.09
0.30
Note: 1) T.O. of ferromanganese was taken from Table 1, manganese content of ferromanganese and manganese metal was assumed 80% and 100% respectively, recovery was assumed 100%; 2) T.O. of manganese metal was assumed 0.060% depending on the saturated solubility of oxygen in pure liquid manganese at 1600 °C[12, 13].
26
Oxide Inclusions in Ferromanganese and Its Influence on the Quality of Clean Steels Table 5 Cleanliness changes of pipeline steel at each process step in plant Number of micro
Amount of macro
inclusions/mm-2
inclusions/(mg•(10 kg)-1)
20.8
7.28
31.52
After RH treatment
16.5
4.74
10.87
After LF treatment
17.3
2.46
32.27
Tundish
21.0
3.78
36.56
Billet
14.7
2.79
1.13
Process step
T.O./10-6
After blowing Ar
and mold flux contamination. However, it was no doubt that the establishment of clean steels process system could hardly do without clean ferroalloys as a guarantee. Currently, production model and part of the quality standards of ferroalloy did not meet the need of new steel developments in China.
E-Publishing, Inc., Kunming, 2004, pp. 12-51. [2] O. Wijk, V. Brabil, ISIJ Int. 36 (1996) S132-S135. [3] K. V. Grigorovich, S. S. Shibaev, I. V. Kostenko, in: Asmo Vartiainen (Eds.), 12th International Ferroalloys Congress, Outotec Oyj, Helsinki, 2010, pp.929-934. [4] M. I. Gasik, A. I. Panchenko, A. S. Salnikov, Metallurgical and Mining Industry 3 (2011) 1-9.
4 Conclusions
[5] T. Sjökvist, Influence of Ferrochromium and Ferromanga-
The oxygen content in medium carbon ferromanganese by electric silicothermic process in China, which ranged from 0.039 wt.% to 0.171 wt.%, was between the common and refined products by oxygen decarburization process outside of China. Refined low and medium carbon ferromanganese with total oxygen content less than 0.04% was recommended to adjust the composition during the refining process to avoid the possible increment of total oxygen content. It was possible that the amount, size and composition of inclusions in steel were affected by oxide inclusions in ferromanganese, resulting in the deterioration of steel quality.
nese Additions on Inclusion Characteristics in Steel, Royal Institute of Technology, Stockholm, 2011. [6] T. Sjökvist, P. Jönsson, Ö. Grong, Metall. Mater. Trans. A 32 (2001) 1049-1056. [7] T. Sjökvist, P. Jönsson, in: ISS-AIME (Eds.), 57th Electric Furnace Conference Proceedings, ISS-AIME, Pittsburgh, 1999, pp.383-393. [8] T. Sjökvist, P. Jönsson, Ö. Grong, P. Cowx, Ironmak. Steelmak. 30 (2010) 73-80. [9] M. Balbi, G. Silva, R. Roja, La Fonderia Italiana 24 (1975) 350-352. [10] M. M. Pande, M. Guo, X. Guo, D. Geysen, S. Devisscher, B. Blanpain, P. Wollants, Ironmak. Steelmak. 37 (2010) 502-511. [11] Z. P. Gao, Secondary Refining Coursebook, Metallurgical Industry Press, Beijing, 2011.
References:
[12] Y. N. Pivovarov, V. Y. Dashevskii, Russian Metallurgy
[1] L. F. Zhang, X. H. Wang, in: CSM Steelmaking Committee
2006 (2006) 286-290.
(Eds.), 13th CSM Annual Steelmaking Conference Proceedings,
[13] K. T. Jacob, Metall. Mater. Trans. B 12 (1981) 675-678.
27