Formation Mechanisms of Cracks Formed During Hot Rolling
of Free-Machining Steel Billets
YONGJIN KIM, HYUNMIN KIM, SANG YONG SHIN, KIHO RHEE, SANG BOG AHN,
DUK LAK LEE, NACK J. KIM, and SUNGHAK LEE
In this study, cracks formed in the edge side of Bi-S–based free-machining steel billets during
hot rolling were analyzed in detail, and their formation mechanisms were clarified in relation
with microstructure. Particular emphasis was placed on roles of bands of pearlites or C- and
Mn-rich regions and complex iron oxides present in the edge side. Pearlite bands in the cracked
region were considerably bent to the surface, while those in the noncracked region were parallel
to the surface. This was because the alignment direction of pearlite bands was irregularly
deviated up to 45 deg from the normal direction parallel to the surface, while the billet was
rolled and rotated at 90 deg in the same direction between rolling passes. On the edge side,
where pearlite bands were bent, iron oxides intruded deeply into the interior along pearlite
bands, which worked as stress concentration sites during hot rolling and, consequently, main
causes of the crack initiation in the rolled billet. On the surface of the wire rod rolled from the
cracked billet, a few scabs were found when some protrusions were folded during hot rolling. In
order to prevent the cracking in billets and scab formation in wire rods, (1) the increase of
rolling passes and the decrease of reduction ratio for homogeneous rolling of billets and (2) the
reduction in sulfur content for minimizing the formation and intrusion of complex iron oxides
were suggested.
DOI: 10.1007/s11661-011-0934-2
Ó The Minerals, Metals & Materials Society and ASM International 2011
I.
INTRODUCTION
FREE-MACHINING steels used for machining
automotive or precision machinery parts contain nonmetallic inclusions such as MnS or metallic inclusions
such as Pb, Bi, and Sn in order to improve machinability.[1–3] These inclusions enhance the machinability by
reducing the force required for machining and by
promoting the initiation and propagation of voids or
cracks at interfaces between inclusions and steel matrix.
They also extend the life of machining tools as they
work as lubricants at interfaces between tools and
chips.[4] Since nonmetallic and metallic inclusions should
be homogeneously distributed inside the matrix to have
excellent machinability, studies on kind, size, fraction,
and distribution of inclusions have been actively conducted.[1,2,4,5] Inclusions generally form brittle interfaces
with the steel matrix and work as crack initiation sites.[6–10]
YONGJIN KIM and HYUNMIN KIM, Research Assistants, are
with the Center for Advanced Aerospace Materials, Pohang University
of Science and Technology, Pohang 790-784, Korea. SANG YONG
SHIN, Research Assistant Professor, is with the Department of
Materials Science and Engineering, Pohang University of Science and
Technology. Contact e-mail: [email protected] KIHO RHEE,
Senior Researcher, SANG BOG AHN, Senior Principal Researcher,
and DUK LAK LEE, Group Leader, are with the Wire Rod Research
Group, Technical Research Laboratories, POSCO, Pohang 790-785,
Korea. NACK J. KIM and SUNGHAK LEE, Professors, are jointly
appointed with the Graduate Institute of Ferrous Technology and also
with the Center for Advanced Aerospace Materials, Pohang University
of Science and Technology.
Manuscript submitted March 23, 2011.
Article published online September 11, 2011
882—VOLUME 43A, MARCH 2012
Because of the existence of brittle inclusions and
interfaces, free-machining steels hardly have sufficient
plastic deformability and, thus, cracking sometimes
occurs during hot rolling of free-machining steel billets
or wire rods.[11–14]
Free-machining steels containing Pb and S were
widely used up to now, and needs for their replacement
were strongly raised because they contain a harmful
element of Pb for human bodies.[15–17] In the European
Union and Japan, the use of Pb in free-machining steels
was recently restricted,[18] but the development and
commercialization of alternative free-machining steels
such as Bi-S–based free-machining steels was delayed.
This is because the hot rolling of Bi-S–based steels is
more difficult than that of Pb-S–based steels as the hotrolling cracking occurs more frequently. It was reported
that the reason for the hot-rolling cracking in freemachining steels was associated with the segregation of
low-melting-temperature elements such as Bi, which
could be formed during the steel-making process.[7–9] In
fact, low-melting-temperature elements can play a role
in initiating cracks as they are melted and segregated
during hot rolling. However, free-machining steels
contain a number of nonmetallic inclusions of MnS,
which can also affect the cracking as they are brittle and
heavily elongated along the rolling direction.
There are several research reports on the numerical
analysis of hot rollability of free-machining steels.
Farrugia[11] analyzed stress distributions of hot-rolled
plates with varying tensile triaxiality ratio, principal
stress ratio, and principal strain ratio to investigate
various physical conditions occurring during hot rolling.
METALLURGICAL AND MATERIALS TRANSACTIONS A
Foster et al.[19] conducted modeling works for analyzing
stress concentrations of inclusions such as MnS under
an applied load. However, since most of these modeling
studies generally treat only MnS inclusions by using
physical parameters, they have limitations for explaining
the cracking behavior of hot-rolled free-machining steel
billets composed of band structures of ferrites, pearlites,
and MnS inclusions. In fact, microstructural parameters
including pearlite band structure, grain size, and surface
oxide layer might directly or indirectly influence the
cracking because they are varied with the alloy compositions and rolling conditions. Thus, the proper control
of the hot-rolling conditions and the resultant microstructures is essentially needed for preventing hot-rolling
cracking, but very few studies on the systematical
explanation of the cracking phenomenon of hot-rolled
steel billets have been conducted.[20] Therefore, the
formation mechanisms of the cracking and the need for
presenting ways to prevent cracking require imminent
attention in view of the productivity of wire rods, labor
cost, energy reduction, and quality improvement.
In the present study, the microstructural analysis was
conducted on the cracking phenomenon occurring
during hot rolling of a Bi-S–based free-machining steel
billet to elucidate the crack formation mechanism.
Based on the analysis results, the correlation between
microstructural factors such as nonmetallic or metallic
inclusions and pearlite band structures and hot-rolling
conditions was investigated by the detailed microstructural examination of the cracked region, and methods to
prevent or minimize the cracking were suggested.
1523 K (1250 °C) for 5 hours in a furnace, and was
rolled to make billets of 160 9 160 mm in cross-sectional area. During hot rolling, the billet was rotated at
90 deg in the same direction between rolling passes.
After the rolling, cracks of several millimeters in length
were formed on the surface of the edge side of the billet,
whereas no cracks were found on the surface of the flat
side. This billet was maintained at 1473 K (1200 °C) for
3 hours and was rolled again to make wire rods of
10 mm in diameter. On the surface of the wire rod,
defects such as scabs were formed.
The fabricated billet and 10-mm-diameter wire rod
were cross sectioned in parallel to the rolling direction,
polished, and etched in 2 pct nital solution. The center
and surface regions were observed by an optical
microscope and a scanning electron microscope (SEM
model JSM-6330F, JEOL*). Five optical micrographs at
*JEOL is a trademark of Japan Electron Optics Ltd., Tokyo.
least were taken at various magnifications, from which
volume fractions of ferrites, pearlites, and MnS inclusions and ferrite grain sizes were measured by an image
analyzer (model SigmaScan Pro version 4.0, Jandel
Scientific Co., Erkrath, Germany). In order to examine
cracks or scabs formed on the surface of the billet or
wire rod, the surface was cleaned in an ultrasonic cleaner
and was observed by an SEM.
III.
RESULTS
A. Microstructure of the Rolled Billet
II.
EXPERIMENTAL
The chemical composition of the Bi-S–based freemachining steel billet was (0.06 to 0.07)C-(1.2 to
1.5)Mn-(0.3 to 0.5)S-(0.05 to 0.10)Bi-(0.1 to 0.2)Sn
(wt pct). A bloom of 300 9 400 9 6400 mm in size
made by continuous casting was homogenized at
In the rolled billet, cracks were formed in the edge
side, whereas no cracks were found in the flat side.
Figures 1(a) and (b) are optical micrographs of the
interior region of 10 mm distant from the flat side
surface. This billet consists of ferrites, pearlites, and
MnS inclusions. Volume fractions of ferrites, pearlites,
Fig. 1—(a) Optical and (b) SEM micrographs of the billet, showing a typical ferrite-pearlite microstructure containing MnS inclusions. Nital
etched.
METALLURGICAL AND MATERIALS TRANSACTIONS A
VOLUME 43A, MARCH 2012—883
Table I.
Volume Fractions of Ferrite, Pearlite, and MnS Inclusion, Ferrite Grain Size, and Vickers Hardness
in the Free-Machining Steel Billet and Wire Rod
Volume Fraction (Pct)
Steel
Billet
Wire Rod
Ferrite
Pearlite
MnS
Ferrite Grain
Size (lm)
bal
bal
4.6 ± 0.29
4.3 ± 0.78
1.9 ± 0.21
1.9 ± 0.26
55 ± 0.9
18 ± 1.4
Vickers
Hardness
(VHN)
107 ± 3.2
128 ± 3.1
Fig. 2—SEM micrographs of (a) the surface of the billet edge region and (b) the cross-sectional area (sectioned parallel to the rolling direction)
of the surface crack observed in (a). Oxides are infiltrated from the crack tip into the interior of the billet. Not etched.
and MnS inclusions measured from the micrographs are
summarized in Table I. The ferrite grain size is about
50 lm, and pearlites show the band structure aligned
along the rolling direction (Figure 1(a)). As MnS
inclusions were deformed by rolling forces during hot
rolling, they are slightly elongated along the rolling
direction (Figure 1(b)).
Figures 2(a) and (b) are SEM micrographs of the
surface of the billet edge region and the cross-sectional
area (sectioned parallel to the rolling direction) of the
surface crack. The crack is initiated in the edge region
and is opened wide to develop into a large and deep
crack (Figure 2(a)). The cross-sectional area of this
crack (Figure 2(b)) indicates that the depth and width of
the crack are 3 to 4 mm. The crack surface is covered
with oxides, some of which are infiltrated from the crack
tip into the interior.
Figure 3 shows a montage of optical micrographs of
the cross-sectional area of the crack. Most of the pearlite
bands are aligned parallel to the surface, but some bands
884—VOLUME 43A, MARCH 2012
near the crack region tend to be slightly bent, as
indicated by dotted red lines. After the cracked region
was sectioned perpendicular to the rolling direction,
the cross-sectional area was observed, as shown in
Figures 4(a) and (b). In the area far from the crack
region, pearlite bands are roughly aligned parallel to the
surface. However, in the area near the crack region, they
are seriously bent up to 45 deg from the surface, as
indicated by a dotted yellow rectangle in Figure 4(a).
Figure 4(b) is the magnified micrograph of the dotted
yellow rectangular region in Figure 4(a) and clearly
shows the slanted pearlite bands, as indicated by dotted
red arrows.
B. Microstructure of the Rolled Wire Rod
The billet containing cracks was hot rolled to make a
10-mm-diameter wire rod. On the surface of the rolled
wire rod, a surface scab is formed, as shown in
Figure 5(a). The size of the scab is about 1 to 2 mm,
METALLURGICAL AND MATERIALS TRANSACTIONS A
Fig. 3—Montage of optical micrographs of the cross-sectional area of the surface crack observed in Fig. 2(b). Nital etched.
and there is a depressed region around the scab.
Figures 5(b) and (c) are optical micrographs of the
cross-sectional area of the scab. A yellow dotted line in
Figure 5(a) shows the cross-sectional direction. A crack
of 4 to 5 mm in length is formed underneath the scab,
and oxides are deeply infiltrated into the crack
(Figure 5(b)). In the depressed region (red rectangular
area in Figure 5(b)) dug by the scab, pearlite bands are
aligned parallel to the inclined surface (red line in
Figure 5(c)). As in the billet, there are ferrites, pearlites,
and MnS inclusions, and their volume fractions are
shown in Table I. As pearlite bands and MnS inclusions
were deformed by continued rolling processes, they are
more severely elongated along the rolling direction. The
ferrite grain size is finer than that of the billet.
Optical micrographs of the areas beyond and beneath
the crack (dotted red rectangular areas in Figure 5(b)) are
shown in Figures 6(a) and (b). The ferrite grain size in the
area beyond the crack is finer than that beneath the crack.
In the area beneath the crack, pearlite bands are well
developed along the rolling direction, whereas pearlite
bands are not clearly formed in the area beyond the crack.
The hardness in the areas beyond and beneath the crack
was measured by a Vickers hardness tester under a 300 g
load. The hardness values were averaged after measuring
ten hardness values for each area. The hardness of the area
beneath the crack is 136 ± 1.7 VHN and is similar to that
(138 ± 0.6 VHN) of the normal area (the center region
without cracks of the wire rod). The hardness of the area
beyond the crack is 144 ± 2.1 VHN and is higher than
that of the normal area, as grains are refined by the
METALLURGICAL AND MATERIALS TRANSACTIONS A
deformation-induced dynamic recrystallization during
hot rolling.[21–24]
IV.
DISCUSSION
A. Crack Formation in the Edge Side of the Rolled Billet
Large and deep cracks are observed in the edge side of
the rolled billet, as shown in Figures 2(a) and (b). MnS
inclusions are thought to be one of most probable causes
of the cracking occurring during hot rolling of freemachining steels. They are deformed with the matrix
microstructure during hot rolling, but their deformation
is not coherently proceeded because of the existence of
interfaces between MnS and the matrix. When the
deformed microstructures are observed, voids or microcracks are formed at MnS/matrix interfaces, as MnS/
matrix interfaces are opened or MnS inclusions are
cracked themselves.[6,10,14] However, these opened interfaces or cracked MnS inclusions hardly act as initiation
sites for surface cracks observed in the edge side, as
shown in Figures 3 and 4(a). According to the detailed
observation of the cracked region (Figure 4(b)), pearlite
bands are seriously bent up to 45 deg away from the
normal rolling direction. This phenomenon of bent
pearlite bands is interesting, because it is scarcely
reported in conventionally rolled steel plates, sheets, or
wire rods.[25] This also implies that the crack formation
in the edge side is closely related with the directionality
of the alignment of pearlite bands.[26,27] In order to
VOLUME 43A, MARCH 2012—885
Fig. 4—(a) Montage of optical micrographs of the cross-sectional area (sectioned perpendicular to the rolling direction) of the surface crack.
(b) High-magnification optical micrograph of the dotted rectangular area in (a). Dotted lines indicate the alignment direction of pearlite bands.
Nital etched.
confirm this relationship, the high-temperature oxidation test, which can show the formation of oxides and
their intrusion into the interior, was conducted on the
billet specimens whose pearlite bands are aligned parallel or vertical to the surface. A number of oxides were
formed after the specimens were held at 1273 K
(1000 °C) for 2 hours, and the extent of the intrusion
of oxides layer into the interior was investigated.
After the high-temperature oxidation test of the billet
specimen whose pearlite bands are aligned parallel to the
surface, optical microstructures of the surface region are
shown in Figures 7(a) through (c). Before the test, the
specimen surface is smooth, and the alignment of
pearlite bands is parallel to the surface (Figure 7(a)).
After the test, the specimen surface becomes roughened,
and oxides are formed down to the depth of 50 to
100 lm (Figures 7(b) and (c)). Here, the oxide intrusion
depth is indicated by a yellow arrow in Figure 7(b).
Most of the oxides are intruded along grain boundaries,
and a few pearlites are found in the surface region
because of the decarburization due to the high-temperature exposure.[28]
The high-temperature oxidation test results of the
billet specimen whose pearlite bands are aligned vertical
to the surface are shown in Figures 8(a) through (c).
Before the test, pearlite bands are aligned vertical to the
886—VOLUME 43A, MARCH 2012
surface (Figure 8(a)). After the test, the specimen
surface becomes quite roughened, and oxides are formed
further down to the depth of 150 to 200 lm, as indicated
by a yellow arrow in Figure 8(b). The oxide intrusion
depth in this specimen is larger than that of the specimen
having pearlite bands aligned parallel to the surface.
Oxides are intruded deeply along grain boundaries as
pearlite bands might accelerate the oxidation.
According to the preceding high-temperature oxidation data, the depth of the oxide formation in the surface
region is varied with the directionality of pearlite bands.
Particularly, in the specimen having pearlite bands
aligned vertical to the surface, the oxide intrusion takes
place severely along pearlite bands.[27] This implies that
the selective oxidation occurs deeply along pearlite
bands when pearlite bands are aligned vertical to the
surface, and that the directionality of pearlite bands
greatly affects the oxidation behavior, although the
oxidation amount is similar in both specimens. However, the selective oxidation at high temperatures cannot
be explained by pearlite bands, because the hot rolling
of Bi-S–based free-machining steel billets was carried
out above the Ac3 temperature, where pearlite bands did
not exist. Hwang et al.[26] and Offerman et al.[29]
reported that the formation of pearlite bands was
closely related with the band-shaped C- and Mn-rich
METALLURGICAL AND MATERIALS TRANSACTIONS A
Fig. 5—(a) SEM micrograph of a surface scab of a wire rod rolled from the billet and (b) and (c) optical micrographs of the cross-sectional area
(sectioned parallel to the rolling direction) of the surface scab observed in (a). (c) High-magnification optical micrograph of the rectangular area
in (b). Nital etched.
Fig. 6—(a) and (b) Optical micrographs of the dotted rectangular areas in Fig. 5(b). Nital etched.
regions and C- and Mn-depleted regions formed by
segregation of C and Mn during continuous casting, and
that pearlites were aligned in a band shape along these
band-shaped C- and Mn-rich regions to form pearlite
bands. This implies that the directionality of bandshaped C- and Mn-rich regions during hot rolling
readily expects the directionality of pearlite bands after
hot rolling.[30,31] Thus, the selective oxidation high
temperatures during hot rolling can be explained by
METALLURGICAL AND MATERIALS TRANSACTIONS A
band structures of C- and Mn-rich regions, instead of
pearlite bands. The oxide intrusion or selective oxidation at high temperatures can work as stress concentration sites during hot rolling and, consequently, causes of
the crack initiation in the surface region of the rolled
billet. In order to solve this problem, the formation
process of band structures is needed to be understood.
Reasons for the bending of band structures from the
surface in the billet can be explained as follows. The
VOLUME 43A, MARCH 2012—887
Fig. 7—Optical micrographs of the billet surface (a) before and (b) and (c) after the oxidation test at 1273 K (1000 °C) for 2 h. Pearlite bands
are aligned parallel to the billet surface.
billet is rolled and rotated at 90 deg in the same
direction between rolling passes. Band structures are
aligned parallel to the surface in the flat side, which is
well contacted with rolls because the homogeneous
rolling occurs in this flat surface area. In the surface
region of the edge side, which is not sufficiently
contacted with rolls, on the contrary, band structures
are often bent, and their alignment direction is irregularly deviated from the normal direction parallel to the
surface. These band structures can be seriously bent up
to 45 deg away from the normal rolling direction, which
acts as a major cause of the intrusion of oxides from the
surface to the interior and the subsequent crack formation during hot rolling. When seriously bent bands are
exposed to a high-temperature atmosphere, oxides
intrude deeply into the interior, and intruded oxides
work as crack initiation sites by contact, friction, or
collision with rolls, thereby resulting in the formation of
cracks on the billet surface.[26,32,33] Figure 9 shows a
schematic diagram of the bending process of band
structures during hot rolling of billets rotated at 90 deg
in the same direction between rolling passes.
The formation of oxides is also affected by chemical
compositions. Only a few pearlites are formed in the
888—VOLUME 43A, MARCH 2012
surface region because of the decarburization due to the
high-temperature exposure, which can work as a cause
of the irregular formation of pearlite bands.
Figures 10(a) and (b) are SEM micrograph and energy
dispersive spectroscopy (EDS) spectrum of an oxide
formed inside the crack near the billet edge region.
These data indicate that the oxide is a complex iron
oxide containing carbon and sulfur. In free-machining
steels inevitably containing a considerable amount of
sulfur for improving machinability, sulfur-containing
complex iron oxides are well formed during hot rolling.
Hayashi et al.[34] reported that sulfur significantly raised
the spallation of oxide scales and that the nonuniform
distribution of oxide scales could work as an origin of
the formation of various surface defects of steel plates.
The high content of sulfur can lead to the detrimental
formation of oxides, the fast intrusion of oxides into the
interior, and the acceleration of cracking. In order to
reduce this detrimental effect of sulfur, a considerable
amount of Mn is generally added into free-machining
steels. It is known that Mn reduces the activity of sulfur
and prevents the formation of FeS, which is not sound
in steels, by forming MnS.[34] When the oxidation occurs
at high temperatures, Mn readily reacts with oxygen to
METALLURGICAL AND MATERIALS TRANSACTIONS A
Fig. 8—Optical micrographs of the billet surface (a) before and (b) and (c) after the oxidation test at 1273 K (1000 °C) for 2 h. Pearlite bands
are aligned vertical to the billet surface.
Fig. 9—Schematic diagram showing the bending process of band structures during hot rolling of billets rotated at 90 deg in the same direction
between rolling passes.
METALLURGICAL AND MATERIALS TRANSACTIONS A
VOLUME 43A, MARCH 2012—889
Fig. 10—(a) SEM micrograph and (b) EDS spectrum of an oxide formed inside the crack near the billet edge region.
Fig. 11—Schematic diagram showing the folding process of surface protrusions to form surface scabs during hot rolling of a wire rod.
form MnO, instead of MnS, because of its high affinity
with oxygen. At this time, sulfur can react with iron to
form FeS and exists at interfaces between subsequently
formed iron oxide layers of FeO, Fe2O3, and Fe3O4. The
liquid phase is also formed by eutectic reactions of FeFeS or Fe-FeO-FeS, while FeS is inevitably formed,
which can promote crack initiation. Therefore, it can be
concluded that cracking in the billet edge region is
associated with the bending of band structures and the
deep intrusion of sulfur-containing complex oxides. In
order to prevent or minimize cracking, the homogeneous rolling of billets, good contact of the billet surface
with rolls, and reduction of sulfur in the composition of
free-machining steels are suggested.
B. Scab Formation in the Rolled Wire Rod
When the billet containing cracks was hot rolled to
make a 10-mm-diameter wire rod, surface scabs were
formed, as shown in Figure 5(a). The subsequent rolling
of the billet makes ferrites, pearlites, and MnS inclusions
elongated along the rolling direction and reduces the
ferrite grain size (Figure 5(c) and Table I). The detailed
observation (Figures 6(a) and (b)) of the cracked region
beneath the surface scab indicates that the formation of
the surface scab is closely related with the already
cracked billet. Large and deep surface cracks formed in
the billet (Figures 2(a) and (b)) can be intruded or
extruded during hot rolling, and the surface of the rolled
wire rod becomes roughened. The protrusions of the
roughened surface can be folded during hot rolling to
form surface scabs. This folding process of surface
protrusions to form surface scabs during hot rolling of
wire rods is schematically shown in Figure 11.
890—VOLUME 43A, MARCH 2012
The localized oxidation and subsequent crack formation can occur seriously under shear stress states. When
the wire rod is rolled between very fast rolling passes,
considerably high shear stresses are operated at pearlite
bands and grain boundaries; thus, the crack formation
can be accelerated by the stress concentration at pearlite
bands and grain boundaries.[35] In order to prevent or
minimize the cracking of the wire rod, therefore, the
homogenization and optimization of the rolling process
is needed to reduce the stress concentration, while
surface cracks formed in the billet should be prevented.
As mentioned previously, reasons and mechanisms
for the formation of cracks or scabs in free-machining
steel billets or wire rods were investigated, from which
the fabrication conditions of defect-free billets or wire
rods can be suggested as follows. First, the hot rolling
conditions of billets have to be homogenized and
optimized for minimizing the cracking in the edge side
of the rolled billet. For example, the number of rolling
passes is to be increased, while the reduction ratio of
each pass is to be decreased, in order to homogeneously
distribute the rolling force. Second, the content of sulfur
has to be decreased to reduce the oxide formation and
intrusion into the interior. The addition of Bi and Si is
desirable, instead of the reduction in the sulfur content,
in the alloy designing of machining-free steels.
Figures 12(a) through (c) show optical and SEM
micrographs of the surface and cross-sectional area of
the billet (cross-sectional area: 160 9 160 mm) and wire
rod (diameter: 8.5 mm) fabricated by the modified
composition ((0.06 to 0.07)C-(1.2 to 1.5)Mn-(0.25 to
0.45)S-(0.05 to 0.10)Bi-(0.1 to 0.2)Sn(wt pct)) and rolling
conditions. Even in the edge side of the billet, pearlite
bands are aligned parallel to the surface without bent
METALLURGICAL AND MATERIALS TRANSACTIONS A
Fig. 12—Optical and SEM micrographs of the (a) billet and (b) and (c) wire rod fabricated by the modified composition ((0.06 to 0.07)C-(1.2 to
1.5)Mn-(0.25 to 0.45)S-(0.05 to 0.10)Bi-(0.1 to 0.2)Sn(wt pct)) and rolling conditions.
pearlite bands (Figure 12(a)). The wire rod fabricated
from the rolling of this billet does not include surface
defects such as scabs (Figure 12(b)). The oxide layer is
homogeneously formed on the surface, which is considerably smooth and good (Figures 12(b) and (c)). These
results indicate that surface defect problems frequently
occurring during hot rolling of billets or wire rods are
successfully solved by modifying the chemical composition and rolling conditions and by using facilities on the
actual production lines. The machining test was conducted on this wire rod under conventional machining
conditions (cutting speed: 100 m/min, feed rate: 0.2 mm/
rev, cutting depth: 1 mm, and cutting condition: wet) by
using a lathe (model PUMA 240M, Doosan Infracore
Co., Ltd., Seoul, Korea). The machining test results of the
wire rod were similar to those of the wire rod of a Pb-based
free-machining steel whose chemical composition and
fabrication condition were similar, although the test
results are not provided here. This indicates that the
present Bi-S–based free-machining steel billet and wire
rod are sound and show excellent machinability.
V.
CONCLUSIONS
In the present study, cracks formed on the edge side
of Bi-S–based free-machining steel billets during hot
METALLURGICAL AND MATERIALS TRANSACTIONS A
rolling were analyzed, and their formation mechanisms
were clarified in relation with microstructure.
1. Large and wide cracks (depth and width: 3 to
4 mm) were observed in the edge side of the rolled
billet, whereas no cracks were found in the flat side.
While the billet was rolled and rotated at 90 deg in
the same direction between rolling passes, the alignment direction of pearlite bands in the edge side
was irregularly deviated up to 45 deg from the normal direction parallel to the surface, which acted as
a major cause of the crack formation during hot
rolling.
2. The high-temperature oxidation test was conducted
on the billet specimens whose pearlite bands were
aligned parallel or vertical to the surface. The depth
of the oxide formation in the surface region was
varied with the directionality of pearlite bands or
band structures of C- and Mn-rich regions. Particularly in the specimen having pearlite bands aligned
vertical to the surface, oxides intruded deeply into
the interior along band structures. This oxide intrusion worked as stress concentration sites during hot
rolling and, consequently, causes of crack initiation
in the rolled billet.
3. When the billet containing cracks was hot rolled to
make a wire rod, surface scabs were formed. The
size of the scab was about 1 to 2 mm, and there
VOLUME 43A, MARCH 2012—891
was a depressed region around the scab. The
detailed observation of the cracked region beneath
the surface scab indicated that the scab formation
was closely related with the already cracked billet.
Large and deep cracks formed in the billet were
intruded or extruded during hot rolling, and the protrusions were folded during hot rolling to form scabs.
4. In order to prevent the cracking or scab forming in
free-machining steel billets or wire rods, (a) the
number of rolling passes was increased, while the
reduction ratio of each pass was decreased, to
homogeneously distribute the rolling force; and (b)
the content of sulfur was decreased to reduce the
oxide formation and intrusion. Using these modified composition and rolling conditions, crack- or
scab-free billets or wire rods could be successfully
fabricated.
ACKNOWLEDGMENTS
This work was supported by POSCO under Contract No. 20108008. The authors thank Dr. Yu Hwan
Lee, Technical Research Laboratories, POSCO, for his
help with experimental analysis.
REFERENCES
1. T. Akasawa, H. Sakurai, M. Nakamura, T. Tanaka, and K.
Takano: J. Mater. Process. Technol., 2003, vols. 143–144, pp. 66–
71.
2. H. Yaguchi: J. Appl. Metalwork., 1986, vol. 4, pp. 214–25.
3. H. Yaguchi: Mater. Sci. Technol., 1989, vol. 5, pp. 255–67.
4. K. Kishi and H. Eda: Wear, 1976, vol. 38, pp. 29–42.
5. A.D. Foster, J. Lin, D.C.J. Farrugia, and T.A. Dean: J. Strain
Anal. Eng. Des., 2007, vol. 42, pp. 227–35.
6. T.J. Baker, F.P.L. Kavishe, and J. Wilson: Mater. Sci. Technol.,
1986, vol. 2, pp. 576–82.
7. S.K. Cho: J. Kor. Inst. Metall. Mater., 1996, vol. 34, pp. 822–29.
8. S. Dinda and W.R. Warke: Mater. Sci. Eng. A, 1976, vol. A24,
pp. 199–208.
9. J.C. Lynn, W.R. Warke, and P. Gordon: Mater. Sci. Eng. A, 1975,
vol. A18, pp. 51–62.
10. W. Roberts, B. Lehtinen, and K.E. Easterling: Acta Metall., 1976,
vol. 24, pp. 745–58.
892—VOLUME 43A, MARCH 2012
11. D.C.J. Farrugia: J. Mater. Process. Technol., 2006, vol. 177,
pp. 486–92.
12. A. Foster, J. Lin, D. Farrugia, and T. Dean: J. Mater. Process.
Technol., 2006, vol. 177, pp. 497–500.
13. J.L. Nazabal, J.J. Urcola, and M. Fuentes: Metallography, 1984,
vol. 17, pp. 439–54.
14. G. Sridhar, S.K. Das, and N.K. Mukhopadhyay: Eng. Fail. Anal.,
1999, vol. 6, pp. 155–72.
15. M. Hashimura, K. Miyanishi, and A. Mizuno: Nippon Steel Tech.
Rep., 2007, vol. 96, pp. 45–49.
16. M. Toshiyuki, T. Kunikazu, and S. Tetsuo: JFE Tech. Rep., 2010,
vol. 15, pp. 10–16.
17. S. Yamamoto, S. Takamori, Y. Osawa, and A. Sato: Nippon
Kinzoku Gakkaishi, 2001, vol. 65, pp. 614–20.
18. T. Fukui: Honda Environmental Annual Report, Honda Motor Co.,
Ltd, Tokyo, 2007.
19. A.D. Foster, J. Lin, D.C.J. Farrugia, and T.A. Dean: J. Strain
Anal. Eng. Des., 2007, vol. 42, pp. 227–35.
20. H.H. Bok, M.G. Lee, H.D. Kim, and M.B. Moon: Metall. Mater.
Int., 2010, vol. 16, pp. 185–95.
21. S.I. Kim, Y. Lee, and B.L. Jang: Mater. Sci. Eng. A, 2003,
vol. A357, pp. 235–39.
22. A. Najafi-Zadeh, J.J. Jonas, and S. Yue: Metall. Mater. Trans. A,
1992, vol. 23A, pp. 2607–17.
23. Z.Q. Sun, W.Y. Yang, J.J. Qi, and A.M. Hu: Mater. Sci. Eng. A,
2002, vol. A334, pp. 201–06.
24. B.W. Choi, D.H. Seo, and J. Jang: Metall. Mater. Int., 2009,
vol. 15, pp. 373–78.
25. D.B. Park, J.W. Lee, Y.S. Lee, K.T. Park, and W.J. Nam: Metall.
Mater. Int., 2009, vol. 15, pp. 197–202.
26. B. Hwang, H. Lee, Y. Kim, and S. Lee: Mater. Sci. Eng. A, 2005,
vol. A402, pp. 177–87.
27. D. Clover, B. Kinsella, B. Pejcic, and R. De Marco: J. Appl.
Electrochem., 2005, vol. 35, pp. 139–49.
28. J.I. Hwang, T.G. Jhee, Y.K. Kim, and T.Y. Hwang: Metall.
Mater. Int., 2010, vol. 16, pp. 693–99.
29. S.E. Offerman, N.H. van Dijk, M.T. Rekveldt, J. Sietsma, and S.
van der Zwaag: Mater. Sci. Technol., 2002, vol. 18, pp. 297–303.
30. P.E.J. Rivera-Diaz-Del-Castillo, S. van der Zwaag, and J. Sietsma:
Metall. Mater. Trans. A, 2004, vol. 35A, pp. 425–33.
31. W. Xu, P. Rivera-Diaz-Del-Castillo, and S. van der Zwaag: ISIJ
Int., 2005, vol. 45, pp. 380–87.
32. H.R. Le, M.P.F. Sutcliffe, P.Z. Wang, and G.T. Burstein: Acta
Mater., 2004, vol. 52, pp. 911–20.
33. M.F. Frolish, M. Krzyzanowski, W.M. Rainforth, and J.H. Beynon:
J. Mater. Process. Technol., 2006, vol. 177, pp. 36–40.
34. S. Hayashi, T. Sekimoto, K. Honda, T. Kinoshita, K. Tanaka, K.
Ushioda, T. Narita, and S. Ukai: ISIJ Int., 2009, vol. 49, pp. 1938–
44.
35. H. Nisitani, T. Fukuda, K. Fujimura, and H. Horio: Proc. 4th Int.
Offshore and Polar Eng. Conf., Osaka, Japan, Apr. 10–15, 1994,
The International Society of Offshore and Polar Engineers,
Golden, CO, 1994.
METALLURGICAL AND MATERIALS TRANSACTIONS A