See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/286460324

Mechanisms of inclusion evolution and intra-granular acicular ferrite

formation in steels containing rare earth elements

Article in Acta Metallurgica Sinica (English Letters) · June 2012

CITATIONS READS

21 438

3 authors, including:

Xiaoxuan Deng
Shougang Institute of Technology
39 PUBLICATIONS 235 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Fast decarburzation in RH degasser View project

Fluid flow in the mold View project

All content following this page was uploaded by Xiaoxuan Deng on 19 December 2015.

The user has requested enhancement of the downloaded file.

Available online at www.amse.org.cn

Acta Metall. Sin.(Engl. Lett.)Vol.25 No.3 pp241-248 June 2012

Mechanisms of inclusion evolution and intra-granular acicular

ferrite formation in steels containing rare earth elements
Xiaoxuan DENG ∗ , Min JIANG and Xinhua WANG
School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, Beijing
100083, China
Manuscript received 1 December 2011; in revised form 23 February 2012

Inclusion characteristic and microstructure of rare earth (RE) elements containing

steel were evaluated with scanning electron microscopy with energy dispersive spec-
troscopy (SEM-EDS), element-mapping, optical microscopy (OM), and automated
feature analysis (AFA) option equipped with ASPEX PSEM. Factsage was used to
calculate the equilibrium inclusion composition. Based on the calculation, an inclu-
sion evolution mechanism was proposed. Furthermore, line scanning analysis was used
to elucidate the intra-granular acicular ferrite (IAF) nucleation mechanism. The re-
sult showed that two different inclusions exist in sample steel: (Mn-Al-Si-Ti-La-Ce-O)
+MnS complex inclusion and isolated MnS inclusion. Almost all nucleation sites for
IAF are complex inclusions, while single MnS inclusion cannot induce IAF. A possible
formation mechanism of complex inclusion is proposed based on calculated results
using Factsage, which agrees well with experimental results. A Mn-depletion zone
(MDZ) which exists adjacent to the (Mn-Al-Si-Ti-La-Ce-O) +MnS complex inclusion
can account for the IAF formation. However, the low volume fraction (1.49×10−7 )
of effective inclusion may result in only 10% (volume fraction) IAF.
KEY WORDS Intra-granular acicular ferrite; RE containing complex inclusion;
Factsage; Inclusion precipitation; Mn-depletion zone

1 Introduction

Non-metallic inclusions are generally believed to be detrimental to the mechanical prop-

erties of steels. However, fine inclusions can act as heterogeneous nucleation sites of IAF,
which has aroused worldwide interests in recent twenty years. IAF, first discovered by
Smith et al [1] in weld metals, has a chaotic crystallographic orientation, which can refine
the austenite grains and retard the propagation of cleavage crack in steel. So IAF can pro-
vide excellent combination of strength and toughness in steel, especially for heat affected
zone (HAZ) and weld metals.
Much attention had been paid to investigate the nucleation mechanism of IAF on differ-
ent inclusions, such as low mismatch strain mechanism [2] , Mn-depletion mechanism [3−6] ,
thermal strain difference mechanism [7,8] and so on. Among them, Mn-depletion mecha-
nism is the most promising mechanism which arouses worldwide concern. It is generally
∗
Corresponding author. PhD candidate; Tel: +86 10 82370978.

E-mail address: [email protected] (Xiaoxuan DENG)

· 242 ·

known that Mn is an austenite-stabilized element. Therefore, a depletion of Mn in steel

near precipitated MnS inclusions can lead to a reduction of austenite zone and thus pro-
mote the nucleation of ferrite in a MDZ during the austenite to ferrite transformation.
Different scholars reported that MDZ exists near MnO-SiO2 with MnS precipitate on it [3] ,
cation-vacancy-type inclusions, such as Ti2 O3 [4] , Mn titanatesl [5] and (Ti-Mn-O)+(Al-Si-
Mn-O)+MnS complex inclusion [6] . Until now, as far as the authors know, whether MDZ
exists or not in steel matrix near complex inclusion containing rare earth (RE) metals is
unknown. So investigation on that topic should be done.
RE elements, with a strong affinity to oxygen and sulfur, were widely applied to spher-
odizing inclusions (such as MnS) to avoid the anisotropy of mechanical properties in final
rolling products. Thewlis et al [9] found that fine CeS inclusions, with high number density
embedded in steel matrix can alter the ferrite nucleation from grain boundary to inclusion
sites and thus enhance the IAF transformation. Recently the authors [10] reported that
Ce2 O3 with a low misfit value with ferrite can act as the nucleation sites for IAF under fast
cooling rate. However, the inclusion characteristic and microstructure of furnace-cooled
RE containing sample and the inclusion formation evolution have not been discussed yet.
In the present work, an attempt has been made to exploit the inclusion characteristics
such as the type, size distribution and number density in laboratory furnace-cooling spec-
imen. Factsage (version 6.2) was used to calculate the equilibrium inclusion composition.
Based on the calculated result, a possible inclusion evolution mechanism was put forward.
Then IAF nucleation mechanism focusing on Mn-depletion mechanism has been discussed
based on experimental results.

2 Experimental

The sample steel was prepared in vertical electric resistance furnace with MoSi2 heating
bars and Al2 O3 reaction tube. The reaction tube was evacuated by flowing argon gas flow
(1.5 L/min) during the whole process. When the iron melt was heated to 1600 ◦ C, Fe-C
alloy was added to make 0.1% carbon, electrolyte Si and Mn were added to make 0.25%
Si and 1.0 % Mn. Then RE metal was added to the melt. After that, the furnace power
was turned off and crucible with the melt for the sample was slowly cooled down in the
furnace. Chemical composition (wt pct) of prepared steel is C 0.074, Si 0.22, Mn 1.00, P
0.0072, S 0.0094, Ti 0.0011, Al 0.0017, La 0.0013, Ce 0.0026, N 0.0059 and T.[O] 0.0041.
Acid-soluble Al, P, Mn and Si content in steel were analyzed by ICP-AES method. C, S
and O contents were analyzed by infrared absorption method.
In order to clarify the inclusions in steel samples, 10 mm×10 mm×10 mm cubic samples
were cut from sample steel. The samples were mechanically mirror polished using diamond
powder emersion and then subjected to examination using ASPEX Instruments PSEM
microscope equipped with EDS. AFA option in this equipment was used to evaluate the
inclusions in steel sample automatically. With this option, predefined area was scanned
and every particle larger than 0.5 µm is evaluated and quantified. The information such
as size, chemical composition, area, diameter and its classification were recorded to a file,
which could transfer to Excel file later. The detected element in present EDS analysis
included: Mg, Al, Si, S, Ti, Mn, La and Ce. However, this method cannot detect the
multi-phase particles. So these inclusions were detected manually. The total scanning area
 · 243 ·

Fig. 1 SEM morphologies and EDS spectra of typical inclusions in sample steel: (a) complex inclu-
sion; (b) isolated MnS

is 14 mm2 .
In order to illustrate the evolution mechanism of inclusion, Factsage 6.2 was used to
calculate the equilibrium inclusion composition as the temperature decreased from 1600 to
1000 ◦ C. FSstel and FToxid database were chosen for the calculation and the result was
exported to Excel file for graphing. The phases considered in present calculation are liquid
steel, molten oxides, δ-ferrite, austenite, α-ferrite, MnS, MnO, SiO2 , AlCeO3 , AlLaO3 ,
MnAl2 O4 and Mn3 Al3 Si3 O12 . The molten oxides (slag phase) consist of MnO, SiO2 , MnS
and Ti2 O3 .
A real color confocal microscope (Lasertec Optelics C130) was used to evaluate the
microstructure of steel sample. And the fraction of IAF was quantified based on point
counting method [11] using a 25 point grid with 1000 points in total.

3 Results

3.1 Inclusion characteristics

Two types of inclusion exist in sample steel: (1) complex oxides (Mn-Al-Si-Ti-La-Ce-
O) with MnS phase and (2) isolated MnS inclusion, which are shown in Figs. 1a and 1b,
respectively. The complex inclusion is almost spherical while single MnS inclusion appears
to be angular .According to this feature, it can be easily inferred that the MnS phase in the
type (1) inclusion precipitates and grows on liquid complex oxides, which consist a small
amount of sulfur, and the type (2) inclusion forms in solid phase.
· 244 ·
Number density of inclusions in sample
steel for different size group is shown in
Fig. 2. It shows a typical bell-shape curve
(shown in red dash line in Fig. 2), which is
also reported in other literature [12,13] . The
peak of number density for complex inclu-
sion is in range of 1–1.5 µm, with the value
of 4.8 mm−2 while for isolated MnS inclu-
sion is in range of 3–3.5 µm with the value
of 0.98 mm−2 .
Inclusion in steel matrix is 3D distrib-
ution, so spatial distribution of inclusion,
such as volume fraction is also a very Fig. 2 Size distribution of inclusions in sample
important character of inclusion. Oh et steel
al [14] reported that volume fraction is more
effective in IAF formation than their type and size distribution. Volume fraction can be
calculated by Dehoff et al [15] equation:
π π n
dV = dH = Pn (1)
2 2 (1/di )
i=1

NV = NA /dV (2)
π
fV = (dV )3 NV (3)
6
where NV is number of inclusions per unit volume; NA number of inclusions per unit area;
di apparent diameter of ith inclusion among n inclusions; dH harmonic mean of inclusions;
dV average diameter of inclusions; fV volume fraction of inclusions.
Volume fractions of isolated MnS and complex inclusion are calculated by Eqs. (1)–
(3), as is presented in Table 1. The value is 1.49×10−7 and 7.02×10−8 , respectively. Oh
et al [14] reported that the minimum volume fraction of inclusion for stable IAF formation
is 1.0×10−4 , so the small volume fraction of inclusion in present study is far from enough
for IAF nucleation.

Table 1 Volume fraction of inclusion detected in steel sample

Inclusion type dV /mm NA /mm−2 NV /mm−3 fV
Isolated MnS 6.16 7.50 1.22 1.49×10−7
Complex inclusion 2.74 17.78 6.47 7.02×10−8

3.2 As-cast microstructure and nucleation sites for IAF

Figure 3 shows the microstructures of sample steel with different magnifications, which
consist of large amount of polygonal ferrite (PF), a little Perlite (P) and IAF with volume
fraction of only 10%. Obviouly, this small amount of IAF structure restricts the application
of the steel when high toughness is needed.
Although only small volume fraction of IAF exists in steel, there still exist many nucle-
ating sites for IAF structure. A typical interlocking ferrite laths nucleated on the inclusion
 · 245 ·

Fig. 3 As-cast microstructures of sample steel

Fig. 4 Effective (a, b) and non-effective (c, d) nucleating sites for IAF

surface, as shown in Figs. 4a and 4b. These ferrite laths have a chaotic crystallographic
orientation, which can retard the propagation path for a cleavage crack. Unfortunately, PF
dominate the microstructure, as shown in Fig. 3. Careful investigation on the nucleation
sites has shown that nearly all nucleation sites for IAF are complex inclusion. Single MnS
inclusion cannot induce IAF in present study (see Figs. 4c and 4d), which can be explained
by thermal strain difference mechanism, and depicted in detail elsewhere [7] . Obviously,
obtaining large amount of complex inclusion is critical to control the microstructure.
To make clear the different phases in complex inclusion mentioned above, further in-
vestigation, for example, element mapping is conducted to show the phase distribution
· 246 ·

Fig. 5 Element mapping pattern of typical inclusion nucleating IAF

in this complex inclusion. Many complex inclusions are chosen for this element mapping,
and phase distribution of these selected inclusions is similar. Figure 5 shows the element
mapping pattern of typical inclusion nucleating IAF.

4 Discussion

4.1 Evolution mechanism of complex inclusion

According to the discussion in section 3.2, (Mn-Al-Si-Ti-La-Ce-O) +MnS complex in-
clusion is beneficial for IAF formation. So a clear elucidation of formation mechanism of
this inclusion during solidification will facilitate the control of this beneficial inclusion and
thus promote the IAF formation.
Figure 6 shows the calculated evolu-
tion of the equilibrium inclusion composition
as the temperature drops from liquid steel
(1600 ◦ C) to solid steel (1000 ◦ C). Possible
formation mechanism of the complex inclu-
sion is as follows:
(1) When electrolyte Si was added, solid
SiO2 formed immediately, as Eq. (4) indi-
cated.

[Si] + 2[O] = SiO2(inclusion) (4)

Similarly, Mn was added to the melt sub- Fig. 6 Evolution mechanism of inclusion dur-
sequently, Eq. (5) occurs. ing solidification process

[Mn] + [O] = MnO(inclusion) (5)

MnO cannot exist in the solid phase, because reaction (6) occurs easily. So relatively
high MnO and SiO2 contents exist in liquid phase between 1500–1550 ◦ C, as illustrated in
Eq. (6).
SiO2(inclusion) + 2MnO(inclusion) = (2MnO · SiO2 ) (6)
 · 247 ·

Then RE metals are added to the melt, Eq. (7) occurs. So REAlO3 exists even in liquid
phase, as shown in Fig. 6.
[RE] + [Al] + 3[O] = REAlO3(inclusion) (7)
At the same time, the existing Ti also reacts with oxygen in steel, as shown in Eq. (8).
[Ti] + x [O] = TiOx (8)
To sum up, SiO2 + (Ce, La)AlO3 +TiOx solid phase, which may form with the collision
of small solid particles (SiO2 , (Ce, La)AlO3 , and TiOx ) in the inclusion core was most likely
formed during initial Si/Mn/RE deoxidation. A liquid phase (mainly MnO-SiO2 -Ti2 O3 )
forms around the solid core inclusion in liquid steel at approximate 1500 ◦ C. Added Mn
can also contact with S to form MnS solid phase, which is a reason to explain why so many
isolated MnS inclusion in sample steel.
(2) When temperature decreases, MnS phase precipitates on the surface of the spherical
liquid inclusion in the embedded form. The precipitation of MnS phase on inclusion during
solidification has two forms [12] : MnS precipitates in embedded form with oxide inclusion
and around oxide core wrapping it. Wakoh et al [16] reported that when the manganese
silicate exists in the multi-phase inclusions, MnS would precipitate on it in embedded form,
which agrees well with present result. When MnS phase forms, Mn in steel matrix will
diffuse into the inclusion boundary, forming a MDZ which will enhance the driving force
of IAF.
(3) As temperature decreased further, the remaining liquid inclusion continued to so-
lidify as TiOx +SiO2 and (Mn-Si-Al)Ox phases.

4.2 Formation mechanism of IAF

As discussed in section 4.1, proposed mechanism of inclusion formation suggests that
MDZ exist around the complex inclusion. To give a direct evidence of MDZ, line scan
analysis were employed to scan Mn content in the inclusion and its adjacent matrix. A
typical IAF nucleation inclusion (see Fig. 7a) is chosen for this line scan analysis. It is
clearly showed a MDZ exists in steel substrate near the inclusion, as shown in Fig. 7b.

(b)

100
Counts / a.u.

0
0 10 20

80
70
60
50
40
2 3 4
Distance / m

Fig. 7 Line scanning analysis of Mn along the nucleation site of inclusion and surrounding steel
matrix: (a) SEM image; (b) line scan pattern
 · 248 ·

Mn is an austenite-stabilized element. Therefore, a depletion of Mn in steel near pre-

cipitated MnS inclusions can lead to a reduction of austenite zone and thus promote the
nucleation of ferrite in a MDZ during austenite to α-ferrite transformation. The slow cool-
ing in this experiment may facilitate the formation of MDZ, which is also reported by Kim
et al [17] . However, because of low cooling rate, the number density and volume fraction of
this effective complex inclusion is too small [12] , which account for low volume fraction of
IAF .So proper cooling rate should be pursued to keep the balance between the inclusion
effectiveness for IAF and volume fraction of inclusion.

5 Conclusions

(1) Inclusion can be classified into two categories: (Mn-Al-Si-Ti-La-Ce-O) +MnS com-
plex inclusion and isolated MnS inclusion in present furnace-cooled steel sample. Almost all
nucleation sites for IAF are complex inclusions, while single MnS inclusion cannot induce
IAF.
(2) A possible formation mechanism of complex inclusion is proposed based on cal-
culated results using Factsage. The experiment result agrees with the calculated results
satisfactorily.
(3) A MDZ exists adjacent to the (Mn-Al-Si-Ti-La-Ce-O)+MnS complex inclusion.
This MDZ can be the driving force for IAF formation. However, the low volume fraction
(1.49×10−7 ) of effective inclusion may result in only 10% (volume fraction) IAF. Proper
cooling rate should be pursued to keep the balance between the inclusion effectiveness for
IAF and volume fraction of inclusion.

Acknowledgements—This work was supported by the National Natural Science Foundation of China
(No. 2010CB30806).

REFERENCES
[1] Y.E. Smith, A.P. Coldren and R.L. Cryderman, Climax Molybdenum Company Ltd (Tokyo, 1972) p.119
[2] B.L. Bramfitt, Metall Trans 1(7) (1970) 1987
[3] M. Wakoh, T. Sawai and S. Mizoguchi, Tetsu-to-Hagané 78 (11) (1992) 73
[4] J.M. Gregg and H.K.D.H. Bhadeshia, Acta Mater 45(2) (1997) 739
[5] C. Eijk, O. Grong and J. Walmsley, Mater Sci Technol 16(1) (2000) 55
[6] Z.Y. Hu, C.W. Yang, M. Jiang, G.W. Yang, W.J. Wang and X.H. Wang, Acta Metall Sin 47(8) (2011)
971 (in Chinese)
[7] D. Brooksbank and K.W. Andrews, J Iron Steel Inst 210(4) (1972) 247
[8] D.S. Sarma, A.V. Karasev and P.G. Jonsson, ISIJ Int 49(7) (2009) 1063
[9] G. Thewlis, Mater Sci Technol 22(2) (2006) 153
[10] X.X. Deng, X.H. Wang, M. Jiang, Z.Y. Hu, X.J. Shao and W.J. Wang, J Univ Sci Technol Beijing
(2012) in Press
[11] G.F. Vander Voort, Metallography, Principles and Practice (McGraw-Hill, New York, 1984) p.427
[12] H.S. Kim, H.G. Lee and K.S. Oh, ISIJ Int 42(12) (2002) 1404
[13] H. Goto, K. Miyazawa, K. Yamaguchi, S. Ogibayashi and K. Tanaka, ISIJ Int 34(5) (1994) 414
[14] Y. Oh, S. Lee, J. Byun, J. Shim and Y. Cho, Mater Trans JIM 41(12) (2000) 1663
[15] R.T. Dehoff and R.N. Rhines, Quantitative Microscopy (McGraw-Hill, New York, 1968) p.168
[16] M. Wakoh, T. Sawai and S. Mizoguchi, Tetsu-to-Hagan 82(7) (1996) 593
[17] H.S. Kim, H.G. Lee and K.S. Oh, Metall Mater Trans A 32A(6) (2001) 1519

View publication stats