Cleanliness of Low Carbon AluminumKilled Steels during

Cleanliness of Low Carbon Aluminum-Killed Steelsduring Secondary Refining Processes

Wen Yang, Xinhua Wang, Lifeng Zhang,� Qinglin Shan, and Xuefeng Liu

Efficient secondary refining process is necessary for massive and stable production of low

carbon aluminum-killed (LCAK) steels. Plant trials were performed to investigate the

cleanliness of steels. Characteristics of composition, cleanliness, and inclusions of LCAK

steel during different secondary refining processes, including LF, CAS, RH-LIT, and RH,

were studied and compared. The results showed that CAS, RH-LIT, and RH processes had

better control of low carbon, silicon, and nitrogen than LF process. High cleanliness of

LCAK steels could be achieved by all the mentioned refining processes. It was concluded

that the total oxygen (T.O.) should be <35 ppm to reduce the amount of inclusions and

reach the level of clean steels. The removal rate of inclusions during RH-LIT and RH

processes was much higher than that of LF and CAS. The T.O. content and the amounts of

inclusions during CAS, RH-LIT, and RH could be quickly decreased to a low value within

10min. The results showed that CAS and RH-LIT as well as RH refining processes can

produce LCAK steels that meet the requirements of high efficient, low cost, clean, and

stable production, while LF is more suitable for the heats with poor control of end point of

BOF, and for the process with calcium treatment to control sulfur content and lower the

clogging of Submerged Entry Nozzle during thin slab continuous casting.

1. Introduction

Low carbon aluminum-killed (LCAK) steels are well used

for production of cold rolled sheets used for automobiles,

household appliances, etc. It requires that LCAK steel

should have low carbon, low silicon, and low nitrogen to

obtain good forming property, and the total oxygen (T.O.)

and the size of inclusions of LCAK steels should be strictly

controlled to guarantee high surface quality of the product.

The secondary refining processes of LCAK steels grades

include LT (Ladle Treatment by Gas Stirring), CAS

(Composition Adjustment by Sealed Argon Bubbling),

RH (Ruhrstahl–Heraeus), and LF (Ladle Furnace treat-

ment). Besides improving steel quality, secondary refining

process is the important step to link up steelmaking and

continuous casting. The LT gas stirring treatment was

developed approximately in 1951.[1] This process is simple

and low cost,[2] however, slags are needed to be deoxidized

to prevent the reoxidation of molten steel and prevent the

increase of inclusion amount. Moreover, the molten steel is

easy to be reoxidized by air when the stirring is too strong.

As the tight quality requirement of the steel, LT is hardly

used alone or little used nowadays. The CAS refining proc-

ess was developed in 1975 by Nippon Steel[3] and expanded

gradually at the end of 1970s. Comparing the LT process,

the CAS process utilizes a snorkel, which hinders the

explosion of the molten steel to air and slag and therefore

lowers the reoxidation of the molten steel. Thus, the stir-

ring intensity can be strengthened to promote the removal

of inclusions, leading to the short refining time and rela-

tively high cleanliness of steel and high yield of alloy. RH

refining process was developed by the two companies

of Ruhrstahl and Heraeus, so called RH, in 1959.[4] The

advantage of RH is degassing and removal of inclusions

utilizing high vacuum and circulation flow. Due to the

using of vacuum chamber, the RH process can well avoid

the reoxidation of molten steel from air and slag, and the

stirring of RH can be very strong. Multiple functions of CAS

and RH were expanded in the 1970s and 1980s, such as

CAS-OB, RH-OB, RH-KTB, RH-PB for additional function

W. Yang, Prof. X. Wang, Prof. L. ZhangState Key Laboratory of Advanced Metallurgy and School ofMetallurgical and Ecological Engineering, University of Science andTechnology Beijing, Beijing 100083, ChinaEmail: [email protected]. ShanSteelmaking Department of Shougang Jingtang United Iron & SteelCo., Ltd, Tangshan 063200, ChinaX. LiuSchool of Materials Science and Engineering, University of Scienceand Technology Beijing, Beijing 100083, China

DOI: 10.1002/srin.201200213

� 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim steel research int. 84 (2013) No. 5 473

FULLPAPER

转载

http://www.paper.edu.cn中国科技论文在线

of reheating, decarburization, and desulfurization.[5]

Besides, the RH light treatment (RH-LIT) was proposed

in 1975 to reduce the burden of steelmaking furnace.[5]

Different with RH route in which deoxidation is performed

during BOF tapping, in RH-LIT route there is no deoxida-

tion performance during tapping, and the first stage of RH-

LIT refining process is decarburization by the dissolved

oxygen in vaccum condition, following by deoxidation with

aluminum addition, which leads to a lower dissolved oxy-

gen before deoxidation and less alumina generated after

deoxidation. As a result, RH-LIT process allows relatively

high carbon and low dissolved oxygen at the end point of

BOF. LF refining process was developed in 1971,[5,6] the

main characteristic is steel heating by electric arc, thus it

can melt ferroalloy by electric arc heating, adjust compo-

sition, and temperature, perform desulfurization and deox-

idation by a reducing slag, and control the cleanliness and

inclusions in the molten steel. The typical advantages of

the LF refining process include high cleanliness of steel,

modification of inclusions, and stable temperature, thus, it

is widely used nowadays although it was developed later

than other refining processes.

There have been many studies on the cleanliness of

LCAK steel at different refining processes,[7–23] the T.O.

content in LCAK steels summarized in Table 1.[7–14]

From these studies, it is known that the high cleanliness

of LCAK steel can be achieved by all the refining processes

discussed above. In recent years, the LCAK steel pro-

duction faces many challenges, including the require-

ments of high quality of the steel product and low

production costs, less environment pollution, and low

energy consumption. Thus, high efficient, low cost, and

stable production of high quality steel is necessary for the

sustainable development of steel enterprises. However, it is

lack of comparison between different refining processes of

LCAK steel, leading to considerable divergences on the

choice of refining process for LCAK steel. Accordingly, in

the current paper, characteristics of steel composition,

cleanliness, and inclusions of each refining process includ-

ing LF, CAS, RH, and RH-LIT were studied and compared

to provide a basis for the choice of refining processes.

2. Experimental and Analysis Methodology

Industrial trials for the production of LCAK steel were

performed. The production routes are ‘‘KR (to remove

sulfur from the hot metal)! 300 t BOF!LFþCa treat-

ment/CAS/RH/RH-LIT! continuous casting.’’ The LF

refining with Ca treatment process is for the production

of SPHC steel while the other refining processes are for the

production of DC01 steel. The compositions of SPHC and

DC01 steel are listed in Table 2.

In this work, calcium treatment was assumed to be a

part of LF refining process. In the LF refining route, during

tapping of BOF, a certain amount of aluminum was added

into the molten steel to decrease the dissolved oxygen to

several ppm. Synthetic slag was added into the ladle during

tapping to reduce the FeO and MnO in the slag. During LF

refining, high basicity slag was used and aluminum was

added to improve deoxidation. Steel samples were taken

before LF refining, at the earlier stage, the middle stage, the

end of LF refining, after Ca treatment, and after soft

blowing.

In CAS refining route, during tapping of BOF, a certain

amount of Al and synthetic slag were added into the molten

steel as well. After tapping, slag modifier was added into

the slag to reduce the slag oxidizability. At CAS station, the

snorkel was submerged into the molten steel after the slag

surface was blown open to a bright face, followed by a

strong stirring by argon gas to promote the removal of

inclusions. The temperature was decreased by adding steel

scraps and be raised by adding aluminum and blowing

oxygen. Steel samples were taken before CAS refining,

during process of argon blowing with time interval of

approximately 7 min, and after argon stop.

In the RH refining route, the end point of BOF was

controlled under common practice, deoxidation, and

alloying were performed during tapping of BOF. During

the RH refining, a large argon gas flow rate was employed

to promote the recirculation and removal of inclusions

from the molten steel. While in the RH-LIT route, the

carbon content at the end point of BOF was relatively

higher than the normal RH process, and no deoxidation

or alloying were performed during the tapping of BOF.

Decarburization was firstly performed during RH-LIT

process under a certain vacuum degree. After decarburiza-

tion, aluminum was added for deoxidation. Then, the

generated inclusions were removed by flow transport

under excellent dynamic conditions. Steel samples were

taken after deoxidation. Since the temperature drop was

large during RH process, aluminum addition, and oxygen

blowing were performed to increase the temperature of

the molten steel.

The schematic operation and sampling scheme of

each secondary refining process were shown in Figure 1.

Temperature of the molten steel was measured several

times during the refining process of each heat. The dis-

solved oxygen content was measured by a zirconia sensor

during RH and RH-LIT refining processes and at the begin-

ning of other refining processes.

The contents of C, S, and T.O. of the steel samples were

analyzed using infrared analysis. The concentration of

silicon was analyzed using ICP-AES and the nitrogen con-

tent was analyzed using the thermal conductance method.

Inclusions on the cross-section of each steel sample

were detected using the automated SEM/EDS inclusion

analysis (ASPEX). ASPEX is a computer-controlled scan-

ning electron microscope that is designed for the auto-

mated imaging and elemental analysis of a wide spectrum

of surfaces and particulates. The system provides a fully

integrated SEM and EDX platform for addressing the

microscale visualization needs. It can simultaneously

474 steel research int. 84 (2013) No. 5 � 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

FULLPAPER

中国科技论文在线 http://www.paper.edu.cn

Steel works Refining Ladle Tundish Slab Year Ref.

America

Inland No.4 BOF shop RH-OB 60–80 8–40 2003 [7]

Dofasco, Canada EAF, LMF 14 2003 [7]

BOF, LMF 23 2003 [7]

Europe

Dillinger, Germany Tank Degas 10–15 2000 [7]

Sidmar, Belgium 37 2000 [7]

RH 20–56 2002 [8]

Corus strip Products, Port Talbot Works, UK CAS-OB, or RH 10–15 2001 [7]

Teesside Technology Center, Corus, UK 6 2002 [7]

Voest-Alpine Stahl GmbH, Linz, Austria RH-OB 10–36 25 2003 [7]

China

Panzhihua Iron and Steel Co. RH-MFB 20–24 2000 [7]

Ar stirring 15.2 2000 [7]

Ma’anshan Iron and Steel ASEA-SKF 18–24 2000 [7]

BaoSteel CAS 73–100 38–53 14–17 2000 [9]

WISCO, No.2 Works RH (Pressure vessel steel 28–34 24–26 12–19 2000 [7]

WISCO, No.3 Works RH 16 20 15 2002 [7]

Anshan Iron and Steel ANS 35–56 33–35 25 2003 [10]

Lianyuan Iron and Steel LFþCa treatment 30–36 34–42 37–43 2006 [11]

BaoSteel, Meishan works LFþCa treatment 34 50 16 2006 [12]

Qian’an Iron and Steel CAS 19–24 17–20 11–15 2007 [13]

Tangshan Iron and Steel LFþCa treatment 16–39 19–46 12–38 2009 [14]

Table 1. Total oxygen content (ppm) reported during various processing steps for LCAK steel after the year of 2000.


FULLPAPER


Steel C Si� Mn P� S� Alt N�

SPHC 0.05 0.03 0.2 0.015 0.012 0.015–0.05 0.006

DC01 0.03 0.03 0.2 0.015 0.012 0.015–0.05 0.006

Table 2. Composition of SPHC and DC01 steel (%).

www.steel-research.de



SPHC 0.05 0.03 0.2 0.015 0.012 0.015–0.05 0.006

DC01 0.03 0.03 0.2 0.015 0.012 0.015–0.05 0.006




Figure 1. Key operation and sampling scheme of each refining process (Sam., sampling; meas., measuring; add., adding).


SPHC 0.05 0.03 0.2 0.015 0.012 0.015–0.05 0.006

DC01 0.03 0.03 0.2 0.015 0.012 0.015–0.05 0.006




FULLPAPER


detect the size, shape, area, number, location, and elemen-

tal composition of inclusions in steel samples. The working

magnification was set at �225 and the minimum detect-

able inclusion was �1.0 mm in the current study.

A statistical parameter, area fraction, is defined as the

following formula to characterize the amount of non-

metallic inclusions.

AF ¼ Ainclusion

Atotal(1)

where, AF is the area fraction of inclusions, ppm; Ainclusion

is the total area of detected inclusions, mm2; and Atotal is the

sample detection area, mm2.

Since an inclusion with irregular shape especially a

cluster might be cut into two or more particles on one

cross section, AF is more accurate than the number density

to describe the amount of the irregular shape inclusions

since the area contains the information of both number

and size.

In the current study, in order to reveal the three-dimen-

sional morphology of inclusions in the steel produced by

CAS and RH refining processes, inclusions were extracted

from steel samples.[24,25] After the surface of the steel

sample being polished, about 3 g of each steel sample

was dissolved in hydrochloric acid solution (1:1 in volume)

at 708C. After the iron matrix was completely dissolved,

deionized water was added to dilute the solution. Then, the

inclusions were filtered out using a polycarbonate filter

paper propelled by a vacuum pump. The pore size of the

filter was 0.2 mm in order to capture most of the inclusions.

As alumina-based inclusions are insoluble to HCl acid,

they will remain on the filter paper. The extracted

inclusions were then observed using SEM after depositing

carbon powder over the filter paper under vacuum

conditions.

3. Results and Discussions

The content of carbon, silicon, sulfur, T.O., nitrogen, and

inclusions in steel samples taken during steel refining

processes were evaluated.

3.1. Carbon Content

Figure 2 shows the carbon content in steel during

secondary refining processes. Since the carbon content

of SPHC is a little higher than that of DC01, the carbon

content in steel during LF refining process is higher than

other processes. This figure shows that the carbon content

during LF, CAS, and RH was slightly changed during

refining. While at the initial stage of RH-LIT process

the carbon content decreased significantly to <0.01%

due to the decarburization, after the addition of carburant,

the carbon content increased to the target value. Since

decarburization can be performed in RH-LIT process

route, the carbon content at the end point of BOF can

be controlled in a higher value and wider range than the

normal RH process, leading to an easier control to

reach the target value at the end point of BOF. While in

other refining routes, since no decarburization was per-

formed and might occur carbon pick-up during refining, it

is required that the carbon content in steel should be

controlled to a low value and in a narrow range at the

end point of BOF, which increases the difficulty of BOF

control and increases the damage possibility of the

BOF refractory.

3.2. Silicon Content

To obtain good forming property of LCAK sheets, [Si] in

LCAK steel should be <0.03%. The variation of [Si] in each

refining process as well as the comparison of [Si] content

and [Si] pick-up after different refining processes are

shown in Figure 3, indicating that the order from high

to low [Si] content by different refining processes is:

LF>CAS, RH>RH-LIT, while the order from large to small

silicon pick-up during refining is: LF>CAS, RH, and RH-

LIT.

The reaction inducing the increasing of silicon is

3ðSiO2Þ þ 4½Al� ¼ 2ðAl2O3Þ þ 3½Si� (1)

The larger [Si] pick-up in LF refining process is due to the

active reaction between slag and molten steel. During the

LF refining process, the activity of SiO2 in slag is improved

by decreasing the oxidizability of the slag. Besides, the

stirring of steel caused by argon blowing provides a good

dynamic condition to the reaction.

0 10 20 30 40 50 600.00

0.01

0.02

0.03

0.04

0.05

0.06

Car

bon

cont

ent (

%)

Time of refining (min)

LF RH-LIT CAS RH

Carburant additionin RH-LIT

Figure 2. Variation of carbon content during refining processes.



FULLPAPER


3.3. Desulfurization

The casting speed of the thin slab continuous casting is

higher than that of the traditional continuous casting proc-

ess. Low sulfur content is required to avoid slab breakout

during casting. Besides, calcium treatment is usually used

to modify inclusions from solid to liquid to prevent the

submerged entry nozzle from being clogged, and low sulfur

content is needed to improve the efficiency and stability of

the calcium treatment. Thus, the desulfurization during

different refining processes is discussed as well.

The [S] content in steel and the efficiency of desulfur-

ization of each secondary refining process are shown

in Figure 4, which indicates that the LF refining process

has the best desulfurization efficiency and with the lowest

[S] content in steel after refining.

0 10 20 30 40 50

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035 LF CAS RH-LIT RH


[Si]

in s

teel

(%)

Deoxidation in RH-LIT

LF CAS RH-LIT RH0.000

0.005

0.010

0.015

0.020

0.025

0.030

[Si]

and

[Si]

pick

-up

afte

r ref

inin

g (%

)

[Si] after refining [Si] pick-up

(a) (b)

Figure 3. Variation of [Si] during refining processes (a), and comparisons of [Si] content and [Si] pick-up after different refiningprocesses (b).

0 10 20 30 40 50 600.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014 LF RH-LIT CAS RH

[S] i

n st

eel (

%)

Time of refining (min) LF CAS RH-LIT RH0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

[S] c

onte

nt a

fter r

efin

ing

(%)

(a) (b)

LF CAS RH-LIT RH0

10

20

30

40

50

60

70

Effic

ienc

y of

des

ulfu

rizat

ion

(%)

(c)

Figure 4. Variation of [S] content during refining processes (a), [S] content in steel after secondary refining (b), and efficiency ofdesulfurization (c) of each refining process.



FULLPAPER


The desulfurization reaction is expressed by[26]

3ðCaOÞ þ 3½S� þ 2½Al� ¼ 3ðCaSÞ þ ðAl2O3Þ (2)

According to Equation 2, in order to increase the efficiency

of desulfurization, the activities of CaO and [Al] should be

higher. Since the steel grade decides the [Al] content in

steel, the desulfurization mainly depends on the activity of

CaO in slag. Because a high basicity slag was used in LF

refining process in the current study, the activity of CaO

was high, leading to the highest desulfurization efficiency.

On the other hand, LF process is designed to desulfurize

because the heating method is more suitable for the use of

high basicity slag.

3.4. Temperature Control

Larger temperature drop rate requires more temperature

compensation to ensure castability of the molten steel,

leading to more cost and possible worse steel cleanliness.

Thus, small temperature drop rate is usually respected.

In the current study, the temperature drop rate is

defined as the average drop rate in the range from the

beginning of refining to the first temperature rising oper-

ation except for the LF refining process, in which the

parameter is defined as average temperature drop in the

refining time except the time period of slag addition, elec-

trical arc heating, and soft blowing. Figure 5 presents the

variation of temperature of molten steel during refining

processes and the average temperature drop rate of each

refining process. The arrows in Figure 5a show the first

temperature rising operations during refining processes. It

is indicated that the order of temperature drop from large

to small during different refining processes is: RH>RH-

LIT>LF and CAS. The large temperature drop rate during

RH and RH-LIT refining which is owing to the strong argon

blowing and the larger circulation rate of the molten steel

will increase the burden of BOF for higher temperature

requirement.

The heating methods of the refining processes are

different from each other. The attribute of LF refining

process is electric arc heating of steel, thus the temperature

control is much easier and more accurate, which is good

for discharging the burden for the furnace, slagging during

the refining, and keeping suitable temperature of the melt

to caster, but it may induce carbon pick-up and nitrogen

pick-up. While the temperature of the molten steel during

CAS or RH refining is raised by aluminum addition and

oxygen blowing, which increases the T.O. and the amount

of inclusions in steel.

3.5. Total Oxygen and Nitrogen

In most refining processes except RH-LIT, deoxidation

were performed by adding aluminum during tapping of

BOF, thus the oxygen content at the end point of

BOF is assumed to be the dissolved oxygen before

deoxidation. Figure 6 shows the contents of [C] and [O]

before deoxidation in CAS, RH, and RH-LIT process, and at

0 10 20 30 40 501560

1580

1600

1620

1640

1660

1680 LF CAS RH-LIT RH

Tem

pera

ture

of m

olte

n st

eel (

)


slag

LF CAS RH-LIT RH0

1

2

3

4

5

6

7

Ave

rage

tem

p. d

rop

rate

(°C

/min

)

(a) (b)

Figure 5. Variation of temperature (a) and the average temperature drop rate (b) of each refining process.

0.00 0.01 0.02 0.03 0.04 0.05 0.060

400

800

1200

1600

2000 CAS RH RH-LIT at end

point of BOF RH-LIT after

decarburization C-O equilibrium

)mpp( tnetnoc ]

O[

[C] content (%)

1225

645

Figure 6. Contents of [C] and [O] of steel in different refiningprocesses.



FULLPAPER


the end point of BOF in RH-LIT process is shown in the

figure as well. The dissolved oxygen content increased with

the decreasing of carbon at the end of BOF, and smaller

carbon content induced more deviation of dissolved oxy-

gen from its equilibrium value, which is calculated at the

pressure of 1 atm. In RH-LIT refining route, the [O] content

before deoxidation was smallest,�215 ppm. In RH refining

route, the value was 645 ppm higher than that of RH-LIT.

The [O] content before deoxidation was highest in CAS

route, �1225 ppm higher than that in RH-LIT. The high

oxygen content at the end point of BOF might be due to the

excessive oxidation of melt by oxygen blowing. Lower

carbon content induces higher oxygen content and easier

excessive oxidation. Thus, it is harmful to pursue too low

carbon content at the end point of BOF.

Figure 7 shows the variation of T.O. and N contents

during each secondary refining process. At the earlier stage

of refining processes of CAS and RH, the content of

T.O. decreased significantly, with a decreasing speed of

�4 ppm min�1, as shown in Figure 8a. During LF refining

0 10 20 30 40 50-9

-6

-3

0

3

6

Heating of RH-LIT

LF CAS RH-LIT RH


)nim/

mpp( H

R/SA

C/FL fo td/.O.Td-

Heating of RH

-50

-40

-10

0

10

20

30 -dT.O./dt of R

H-LIT (ppm

/min)

LF CAS RH-LIT RH0

10

20

30

40

50

60

T.O

. afte

r sec

onda

ry re

finin

g (p

pm)

(a) (b)

Figure 8. The decreasing rate of T.O., �dT.O./dt of each refining process (a) and T.O. contents (b) after secondary refining.

0 10 20 30 40 500

20

40

60

80

100

LF T.O. N

T.O

. and

N in

ste

el (p

pm)


Ca treatment

0 10 20 30 40 500

10

20

30

40

50

60

70

T.O

. and

N in

ste

el (p

pm)


CAS T.O. N

Argon blowing

Al addition

(a) (b)

(c) (d)

16 20 24 28 32 36 400

30

60

90

120

180

200

220


RH-LIT T.O. N

T.O

. in

stee

l (pp

m)

Heating by Aladdition and O

2 blowing

0

4

8

12

16

20

N in steel (ppm

)0 10 20 30 40 50

0

20

40

60

80

T.O

. and

N in

ste

el (p

pm)


RH T.O. N

Heating by Al addition and O2 blowing

Figure 7. Variation of T.O. and N during refining processes, a) LF, b) CAS, c) RH-LIT, and d) RH.



FULLPAPER


process, the T.O. decreased gradually, while during CAS

and RH refining, after a fast decreasing at the first 8–9 min,

the T.O. content kept almost constant. During RH-LIT and

RH refining processes, the heating of steel by the addition

of aluminum and oxygen blowing caused a sharp increase

of T.O. At the end of each refining process, the T.O.

decreased to a low value – 33 ppm by LF, 34 ppm by

CAS, 45 ppm by RH-LIT, and 28 ppm by RH, as shown

in Figure 8b, indicating that high cleanliness of steel can

be achieved by all the refining processes discussed.

However, T.O. during RH-LIT was relatively higher than

that of other refining processes due to the shorter net time

for the removal of inclusions.

Figure 9 shows the nitrogen content and nitrogen pick-

up for each refining process. It is indicated that the order

of nitrogen content from high to low is: LF>CAS and

RH>RH-LIT, and the nitrogen pick-up was largest during

LF refining process, which was as high as 12 ppm and

mainly occurred during Ca treatment, as shown in

Figure 7a, which might be attributed to the air absorption

caused by the boiling of molten steel on the near-surface of

the molten pool at time of Ca–Fe wire injection that was

induced by the high steam pressure of calcium.

3.6. Inclusions in Steel during Refining

Inclusions in LCAK steel after deoxidation are mainly

alumina-based. Large alumina-based inclusions have a

detrimental effect on the castability of the molten

steel[27–29] and the surface quality of the steel sheet[30–32].

Currently, large alumina-based inclusions are either modi-

fied into low melting ones by calcium treatment,[9,33,34] or

removed by flow transport,[35,36] or become defects of the

steel product.[26]

Some aspects of inclusion characteristics, including

composition, morphology, amount, size, and remove rate

in each refining process, were analyzed.

3.6.1. Composition and Morphology

Figure 10 shows the composition of inclusions at the end

of each refining process, detected by ASPEX. The curve in

the figure represents the boundary of liquid inclusion

region. In LF route, after calcium treatment the oxide

inclusions transferred to CaO–MgO–Al2O3 (Figure 10a),

and many of them were located in the liquid region, which

is good for preventing nozzle clogging and improving the

castability of the steel. However, Figure 10b represents that

after calcium treatment a vast amount of CaS generated,

most of them were precipitaed on the oxides, this would

decrease the modification effect of alumina inclusions.

Thus, lower sulfur content is required to improve the

inclusion modification. At the end of CAS, RH-LIT, and

RH refining processes, since there was no calcium treat-

ment, inclusions were solid alumina, which would induce

nozzle clogging during casting. Thus, on the point of view

of the castability of the molten steel, LF refining (including

calcium treatment) is better than CAS, RH-LIT, and RH.

The transformation of inclusions during LF refining

process was owing to the reaction of calcium treatment.

High basicity reductive slag was used during LF refining,

the reaction between the slag and the molten steel changed

the composition of inclusions, especially after calcium

addition. Inclusions were modified to liquid ones mainly

according to the following reaction.

7ðAl2O3Þ þ 12½Ca� þ 12½O� ¼ ð12CaO� 7Al2O3Þ (3)

As mentioned above, oxide inclusions after LF refining was

mainly MgO–CaO–Al2O3 that could be partially dissolved

by hydrochloric acid, while after CAS and RH refining

inclusions were mainly alumina that would not be dis-

solved by hydrochloric acid. Accordingly, the morphology

of inclusions after LF refining was directly observed by

SEM, and inclusions after CAS and RH refining were firstly

extracted using HCl acid and then observed by SEM to

reveal the three-dimensional morphology of alumina

inclusions.

Figure 11 shows the morphology of inclusions at the

end of each refining process. Since inclusions after RH

refining were the same as those after RH-LIT refining,

inclusions only after RH-LIT refining are shown here.

After LF refining, inclusions were transformed to spherical,

and most of them were small. CaS was precipitated on the

gap part of the spheres shown in Figure 11a. However, after

CAS and RH-LIT refining, there were still many irregular

inclusions, and large clusters and aggregates were

observed, which were very detrimental to the subsequent

casting process. Besides clusters and aggregates, many

small spherical and polyhedral particles were observed

as well in the steel samples after CAS and RH refining.

3.6.2. Amount and Removal Rate

The amount of inclusions is characterized by their area

fraction (AF). Figure 12 shows the variation of AF of

inclusions during each secondary refining process.

LF CAS RH-LIT RH0

5

10

15

20

25

30

35

40N

and

N p

ick-

up a

fter

refin

ing

(ppm

) N after refining N pick-up

Figure 9. N content and N pick-up during each refiningprocess.



FULLPAPER


Figure 10. Composition of inclusions at the end of each refining process: a,b) LF, c) CAS, d) RH-LIT, e) RH.



FULLPAPER


During LF refining process, before calcium treatment, the

AF of inclusions decreased, while after calcium treatment,

the amount of inclusions increased sharply, which was

caused by the massive generation of new inclusions especi-

ally sulfide (as shown in Figure 10b) and slight reoxidation

of molten steel (as shown in Figure 7a) during calcium

treatment. During other refining processes, the variation of

AF was consistent with the tendency of T.O. At the early

stage of refining process (�10 min), the AF of inclusions

decreased significantly, and then decreased slowly. During

RH-LIT and RH refining processes, the heating of molten

steel by adding aluminum and blowing oxygen extremely

increased the AF of inclusions due to the massive gener-

ation of new alumina inclusions.

The relationship between the AF of alumina-based

inclusions and the T.O. is shown in Figure 13. It is clearly

indicated that the AF of alumina-based inclusions

increased with the increasing of T.O. in steel, and it

increased slowly when the T.O. was<35 ppm but increased

exponentially after the T.O. was over 35 ppm. Thus, the

T.O. in steel should be controlled to<35 ppm to reduce the

amount of inclusions. As discussed by Zhang,[22,37] the T.O.

only represents the inclusions smaller than 50 mm. How to

control large inclusions that are mainly exogenous ones

Figure 11. Morphology of inclusions after refining processes a) LF, b) CAS, and c) RH-LIT.



FULLPAPER


from slag and lining refractory materials is another import-

ant task of steelmakers.

The AF of inclusions at the end of each refining process

is shown in Figure 14, where LF(�Ca) means before

calcium treatment and LF(þCa) means after calcium treat-

ment. The AF of inclusions was smallest by refining of

RH-LIT or RH, �37 ppm, around 52 ppm at the end of

CAS refining and before calcium treatment in LF refining

process. However, after calcium treatment, the AF was

sharply increased to �232 ppm owing to the massive

generation of CaS and slight reoxidation of the steel during

the practice of calcium treatment.

The minimum AF of inclusions characterizes the

removal ability of inclusions from the molten steel, while

10 20 30 40 50 600

50

100

150

200

250

300


AF

of in

clus

ions

(ppm

)

Ca treatment

LF

0 10 20 30 40 500

50

100

150

200

250

300

AF

of in

clus

ions

(ppm

)


CAS

Argon blowing

16 20 24 28 32 36 40 44

0

200

400

600

800

1000

AF

of in

clus

ions

(ppm

)


Heating by adding Aland blowing O2

RH-LIT

0 10 20 30 40 50

0

100

200

300

400

500

600

700

AF

of in

clus

ions

(ppm

)


RH

Heating by adding Al and blowing O2

Figure 12. AF of inclusions during each secondary refining process.

0 20 40 60 80 1000

100

200

300

400

AF

of in

clus

ions

(ppm

)

T.O. content in steel (ppm)

Figure 13. Relationship between the AF of inclusions and theT.O. of the steel.

LF(- Ca) LF(+ Ca) CAS RH-LIT RH0

20

40

60200

220

240

AF

of in

clus

ions

(ppm

)

Figure 14. AF of inclusions at the end of each secondaryrefining.



FULLPAPER


the removal efficiency of inclusions is characterized by

average removal rate of the AF at the fast decreasing stage

of AF, defined as Equation 4.

hIR ¼½AF�in � ½AF�low

DtIR(4)

where, hIR is the removal rate of inclusions in AF, ppm/

min; [AF]in is the initial AF of inclusions, ppm; [AF]low is the

AF of inclusions at the end of fast decreasing stage shown

in Figure 12, ppm; DtIR is the time of the fast decreasing

stage, min. The schematic diagram is shown in Figure 15.

The calculated removal rate of inclusions during each

refining process is shown in Figure 16, indicating that the

removal rate of inclusions in AF in RH-LIT and RH was far

larger than that in LF and CAS owing to the quick circu-

lation flow of molten steel. The very low removal rate

during LF refining was due to the relatively small bottom

argon blowing rate. Another reason is that the initial AF of

inclusions considered in the calculation (at 14 min) might

not be included in the actual initial fast decreasing stage

that is generally during the initial 10 min of the refining

process.

3.6.3. Size Distribution

Figure 17 shows the average size and size distribution of

inclusions during refining processes. The number density

(ND) of inclusions is defined as the number of inclusions

per unit detection area. The tendency of average size of the

inclusions was similar to the AF of inclusions. During LF

refining, before calcium treatment, the size of inclusions

decreased gradually because of the removal of >2 mm

inclusions. Immediately after calcium treatment, the aver-

age size decreased to only 1.53 mm owing to the massive

generation of 1–2 mm inclusions by the calcium treatment.

As the time increasing during soft blowing, inclusions

aggregated and their average size increased. During CAS

refining, during argon blowing process, the average size

decreased, while after argon blowing was off, the average

size increased firstly due to the aggregation of small

inclusions and then decreased due to the floating removal

of large inclusions. During RH-LIT refining, at the early

stage of steel heating, many large alumina generated which

led to a large average size. At the later stage of heating,

it quickly deceased due to the fast removal of large

inclusions. During RH refining, at the early stage, it

decreased significantly due to the removal of inclusions,

and increased obviously during the heating stage for the

same reason as that during RH-LIT.

The average size of inclusions at the end of each refining

process is shown in Figure 18, indicating that the average

size of inclusions after LF refining was smallest due to the

modification by slag–steel reaction and calcium treatment.

The actual size of inclusion in CAS or RH refining may be

larger than the measured one because of the dominant

irregular and clustered alumina inclusions.

3.7. Total Refining Time

The secondary refining process is a connection link

between the preceding steelmaking and the following

casting, thus besides functions mentioned previously,

the secondary refining process is also a regulation of the

production tempo. To improve the production efficiency

and reduce manufacturing costs, high speed continuous

casting is expected, which requires shorter secondary

refining time.

The refining time in the current work is defined as the

time from start to the end of treatment, e.g., for LF and CAS

refining it is the time from the start of to the end of argon

blowing, and for RH-LIT and RH refining it is the time from

the start to the end of vacuum. The refining time of each

process is shown in Figure 19, indicating that the refining

time of LF was longest and that of CAS was shortest. Here,

the LF refining process included calcium treatment. RH-

LIT process contained a decarburization stage and a steel

heating process, leading to a relatively long refining time.

Although there was no decarburization stage during RH

process, however, there were two steel heating processes,

LF CAS RH-LIT RH0

10

20

30

40

50

60

70

Rem

oval

rate

of i

nclu

sion

sin

AF

(ppm

/min

)

Figure 16. Removal rate of inclusions during each refiningprocess.

Figure 15. Schematic of removal rate of inclusions.



FULLPAPER


Figure 17. Average size and size distribution of inclusions during each refining process.



FULLPAPER


which induced much longer time than expected.

Therefore, the temperature rising operation should be

shortened as far as possible to lower the total refining time.

This requires a compact and reasonable plants layout

and more scientific management of the manufacturing

process.

3.8. Material Consumption

Low cost manufacturing is necessary and urgent nowadays

for the environmentally friendly and sustainable develop-

ment of steel industries. The key of low cost manufacturing

is to produce high quality products with the least consump-

tion of resource raw materials. In the current study, material

consumption of each refining process was analyzed.

Material consumption mainly includes the consump-

tion of: alloys, slags, argon gas, oxygen gas, power, and so

on. For the current study, the steel grade was similar and

the consumption of Mn–Fe for alloying was almost the

same. The slags added during the tapping of BOF were

nearly the identical except that the slag modifier added

after tapping in RH-LIT route was less. Thus, the materials

with large consumption mainly includes aluminum metal

for deoxidation and alloying such as Al–Fe, Al gravels, other

alloy such as Ca–Fe, slags added during refining process

such as synthetic slag and fluorite, argon gas for stirring,

and materials for temperature rising such as electrical

power during LF refining and aluminum and oxygen

during RH-LIT and RH refining.

Table 3 shows the consumption of the main auxiliary

materials for each refining process, the power for vacuum

pumps in RH-LIT and RH was not considered. It indicates

that the material consumption during LF refining process

was much larger than those during other processes.

The comparison on the aspects analyzed above among

the four secondary refining processes is summarized

in Table 4, indicating that CAS and RH-LIT as well as

RH can well meet the requirements of high efficient, low

cost, and stable production of LCAK steels, while LF is

LF CAS RH-LIT RH0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0A

vera

ge s

ize

of in

clus

ions

(μ)

Figure 18. Average size of inclusions at the end of each refiningprocess.

LF CAS RH-LIT RH0

10

20

30

40

50

60

70

Tota

l ref

inin

g tim

e (m

in)

Figure 19. Refining time of each secondary refining process.

Materials LF CAS RH-LIT RH

Al–Fe (kg t�1-steel) 3.45 6.94 0 5.55

Al grain (kg t�1-steel) 1.29 0.30 1.11 0.70

Ca–Fe (kg t�1-steel) 3.82 0 0 0

Synthetic slag (kg t�1-steel) 8.77 0 0 0

Fluorite (kg t�1-steel) 4.62 0 0 0

Argon (Nm3 t�1-steel) 0.091 0.088 0.408 0.387

Electricity for heating (kWh t�1-steel) 23.37 0 0 0

Al grain for heating (kg t�1-steel) 0 0 1.34 0.87

Oxygen for heating (m3 t�1-steel) 0 0 1.08 0.91

Table 3. Main materials consumption of each secondary refining process.



FULLPAPER


more suitable for the case with a low level of refining

technology, or for the case with poor control of end points

of BOF steelmaking, or for thin slab continuous casting

process.

4. Conclusions

Characteristics of the composition, cleanliness, and

inclusions of LCAK steels by different secondary refining

processes were studied based on the plant trials, and then a

comprehensive comparison was performed. The following

conclusions were obtained:

(1) The processes of CAS, RH-LIT, and RH could keep low

values of carbon, silicon, and nitrogen which are

required for the production of LCAK steels, while

higher pick-up of silicon and nitrogen were found

for LF refining process.

(2) High cleanliness of LCAK steels could be achieved by

all the mentioned refining processes. The T.O., e.g., was

33 ppm for LF, 34 ppm for CAS, 45 ppm for RH-LIT, and

28 ppm for RH. The T.O. should be <35 ppm to reduce

the amount of inclusions.

(3) The amount of inclusions characterized by the AF after

each refining was: 52 ppm before calcium treatment

and 232 ppm after calcium treatment for LF process,

52 ppm for CAS, 37 ppm for both RH-LIT and RH proc-

esses. The higher amount of inclusions after calcium

treatment was due to the massive generation of CaS

and slight reoxidation during calcium treatment.

(4) The removal rate of inclusions during RH-LIT and RH

was much higher than during LF and CAS processes.

The T.O. content and the amount of inclusions during

CAS, RH-LIT, and RH could be quickly decreased to a

low value within 10 min refining.

(5) According to the comprehensive comparison, it is

known that CAS and RH-LIT as well as RH can well

meet the requirements of high efficient, low cost, and

Aspects Items LF CAS RH-LIT RH

Composition and

temperature control

Resist [C] fluctuation at the end of BOF D D * D

Resist temperature fluctuation at the end of BOF * D D D

Decarburization � � * D

Desulfurization * D � �

Avoid Si pick-up � * * *

Temperature drop * * � �

Temperature compensation * D D D

Cleanliness [O] before deoxidation � � * D

T.O. after refining * * D *

[N] after refining � D * D

Avoid [N] pick-up D * * *

Inclusions Modification * � � �

Amount � D * *

Average size * D D D

Removal rate � D * *

Castability of steel * D D D

Efficiency and cost Refining time � * D D

Production cost � D * D

Construction cost D * � �

Note: *, Good; D, Fair; �, Poor.

Table 4. Comparison of different refining processes.



FULLPAPER


stable production of LCAK steels, while LF is more

suitable for the case with low level of refining technol-

ogy, or the case with poor control of end points of BOF,

or for thin slab continuous casting process.

Acknowledgments

The authors are grateful for support from the National

Science Foundation China (grant no. 51274034), the

High Quality Steel Consortium at University of Science

and Technology Beijing (China), and Shougang Jingtang

United Iron & Steel Co. (China) for industrial trials.

Received: August 21, 2012;

Published online: December 5, 2012

Keywords: cleanliness; inclusion; LCAK steel; secondary

refining

References

[1] E. Spire, AIME Electric Furnace Steel Proc. 1951, 9, 75.

[2] J. Szekely, H. J. Wang, K. M. Kiser, Metall. Trans. B

(Process Metall.) 1976, 7B, (2) 287.

[3] H. Haraguchi, K. Okobira, U. Muneo, Tetsu Hagane

1975, 61(4), S135.

[4] H. Thielmann, H. Maas, Stahl Eisen 1959, 79(5), 276.

[5] H. Kajioka, Ladle Refining-Challenges to Mass

Production Systems of Various Grade and High

Quality Steels, Metallurgical Industry Press, Beijing

2002.

[6] H. Ushiyama, G. Yuasa, T. Yajima, Ironmaking

Steelmaking 1978, 5(3), 121.

[7] L-f. Zhang, J. Iron Steel Res. Int. 2006, 13(4), 1.

[8] R. Dekkers, B. Blanpain, P. Wollants, ISSTech 2003

Conf. Proc. 2003, 2, 197.

[9] G. Shi, L. Zhang, Y. Zheng, J. Zhi, Iron Steel

(in Chinese) 2000, 35(3), 12.

[10] X. Yuan, H. Li, X. Wang, Iron Steel (in Chinese) 2003,

38(11), 15.

[11] Q. Peng, Y. Li, C. Yang, B. Chen, Steelmaking

(in Chinese) 2007, 23(3), 45.

[12] C. Ren, Research on the Cleanliness of Continuous

Casting Slabs of Low Carbon Al-Killed Steel, Thesis,

University of Science and Technology, Beijing, 2007.

[13] T. Zhang, Y. Bao, H. Cui, F. Yue, Y. Jin, Special Steel

(in Chinese) 2008, 29(2), 60.

[14] W. Yang, J. Cao, X-h. Wang, Z-r. Xu, J. Yang, J. Iron

Steel Res. Int. 2011, 18(9), 6.

[15] S. Yang, L. Zhang, J. Li, K. Peaslee, Metall. Mater.

Trans. B 2012, 43B(4), 731.

[16] L. Zhang, J. Iron Steel Res. Int. 2006, 13(4), 01.

[17] L. Zhang, X. Wang, Shandong Metall. (in Chinese)

2005, 27(1), 1.

[18] L. Zhang, X. Wang, Shandong Metall. (in Chinese)

2005, 27(2), 1.

[19] L. Zhang, J. Iron Steel Res. Int. 2006, 13(3), 1.

[20] J. Wei, Z. Tian, L. Zhang, K. Cai, Y. Zhou, Inclusions

in the Low Carbon Al-killed Steel Produced by a

CSP Thin Slab Casting Process at Handan Steel, in

Proccedings of AISTech 2005 Iron & Steel Technology

Conference and Exposition, Vol II, AIST, Warrandale,

PA (May 9–12, 2005, Charlotte, NC), 2005, pp. 585–

592.

[21] L. Zhang, B. G. Thomas, K. Cai, L. Zhu, J. Cui,

ISSTech 2003 Conf. Proc. 2003, 2, 141.

[22] L. Zhang, B. G. Thomas, ISIJ Int. 2003, 43(3), 271.

[23] L. Zhang, B. G. Thomas, X. Wang, K. Cai, Evaluation

and control of steel cleanliness – Review, 85th

Steelmaking Conference Proceedings, March 10,

2002 – March 13, 2002, (Nashville, TN, United

states), Iron and Steel Society, 85, 2002, 431–452.

[24] S. Li, L. Zhang, X. Zuo, Observation of Inclusions in

the Steel Using Partial Acid Extraction and SEM in

Proccedings of Materials Science and Technology

(MS&T), AIST, Warrandale, PA, (Pittsburgh) 2008,

pp. 1259–1269.

[25] L. Zhang, S. Li, J. Wang, X. Zuo, Iron Steel (Chinese)

2009, 44(3), 75.

[26] G. Ye, P. Jonsson, T. Lund, ISIJ Int. 1996, 36 (Suppl),

S105.

[27] B. Wang, Y. Pan, L. Tian, K. Hou, W. Wang,

Steelmaking (in Chinese) 2008, 24(6), 41.

[28] R. Dekkers, Ph.D. Thesis Thesis, Katholieke

University, Leuven, 2002.

[29] R. Dekkers, B. Blanpain, P. Wollants, F. Haers,

B. Gommers, C. Vercruyssen, Steel Res. 2003, 74(6),

351.

[30] G. Zhu, H. Yu, W. Wang, X. Wang, Iron Steel (in

Chinese) 2004, 39(4), 54.

[31] W. Gao, Adv. Mater. Res. 2012, 402, 221.

[32] P. Rocabois, J. Pontoire, ISSTech 2003 Conf. Proc.

2003, 2, 995.

[33] Y. Higuchi, M. Numata, S. Fukagawa, K. Shinme, ISIJ

Int. 1996, 36 (Suppl), S151.

[34] N. Verma, P. Pistorius, R. Fruehan, M. Potter,

M. Lind, S. Story, Metall. Mater. Trans. B 2011,

42(4), 720.

[35] W. Zhu, D. Jin, B. Li, Iron Steel (in Chinese) 1991,

26(2), 22.

[36] Z. Xue, Y. Wang, L. Wang, Z. Li, J. Zhang, Acta

Metall. Sin. (in Chinese) 2003, 39(4), 431.

[37] L. Sun, L. Zhang, J. Li, A. Dong, Y. Chen, S. Yang,

Iron Steel Technol. 2012, 9(4), 20.



FULLPAPER


Documents

Cleanliness of Low Carbon AluminumKilled Steels during