Materials Transactions, Vol. 44, No. 8 (2003) pp.

1630 to 1635
#2003 The Japan Institute of Metals
EXPRESS REGULAR ARTICLE

Spheroidization of Low Carbon Steel Processed by Equal Channel Angular Pressing

Dong Hyuk Shin1 , Soo Yeon Han1; * , Kyung-Tae Park2 , Yong-Seog Kim3 and Young-Nam Paik4
1
Department of Metallurgy and Materials Science, Hanyang University, Ansan, Kyunggi-Do 425-791, Korea
2
Department of Advanced Materials Science and Engineering, Taejon National University of Technology, Taejeon 305-719, Korea
3
Department of Materials Science and Engineering, Hongik University, Seoul 121-791, Korea
4
College of Mechanical Engineering and Industrial System Engineering, Kyung Hee University, Yongin 449-701, Korea

Spheroidization behavior of cementite in a low carbon steel processed by the equal channel angular pressing technique was investigated.
The effects of annealing temperature and time, and accumulated strain on the morphology of cementite and mechanical properties were studied.
The results indicated that the application of the severe plastic deformation can improve the kinetics of spheroidization significantly. In this study,
the enhanced spheroidization kinetics was discussed in terms of carbon dissolution from cementites and defects induced in cementites by the
severe plastic deformation. In addition, the softening of the steel after the spheroidization treatment was evaluated.

(Received April 1, 2003; Accepted June 19, 2003)

Keywords: equal channel angular pressing, spheroidization, low carbon steel, pearlite, cementite

1. Introduction nonstoichiometry, as well as kinks and ledges to the

Spheroidization treatment of cementites in steels is used to
increase their formability during wire drawing of high
strength wire rods and heading of high strength bolts at
ambient temperatures. A significant research effort has been
directed to investigate the mechanisms and kinetics of
spheroidization process.1,2) Based on the investigations, the
spheroidization treatment is conducted either by intercritical
or by subcritical treatment.3,4) The intercritical treatment
refers to a heat treatment scheme that consists of heating to a
temperature region between the upper and lower critical
temperatures (between A1 and A3 ) prior to the spheroidizing
treatment below the lower critical temperature. The subcrit-
ical treatment, on the other hand, denotes the heating of the
steel below the lower critical temperature (A1 ). Although the
kinetics of cementite spheroidization depends on the chem-
ical compositions of steel and heat treating schedule, time
required for the spheroidization usually takes from 10 to 24
hours. Thus the spheroidization treatment is by far the most
time consuming steps in manufacturing the wire rod and bolt.
Driving force for the cementite spheroidization is the
reduction in surface free energy of the system. The
spheroidization process, i.e. transformation of cementites
from lamellar morphology to spherical shapes, decreases the
interface area between cementite and ferrite phases, reducing
surface free energy of the system. The energy reduction
associated with the spheroidization, in general, is not large
enough to drive the process at a high rate, making the process
time consuming. One of the effective methods noted to
increase the spheroidization speed is to introduce defects in
the cementite via severe plastic deformation. Hono et al.5)
showed that the cementites in a near eutectic steel spher-
oidizes more readily after a severe drawing. Effective strain
applied to the sample during the wire drawing was
approximately 1.0. This increased speed of spheroidization
has been attributed to the dissolution of carbon in ferrite
matrix and the introduction of defects such dislocations,
*Graduated Student, Hanyang University.
cementite during the plastic deformation.6,7)
Recently, Shin et al.8–12) applied equal channel angular
(ECA) pressing on low carbon steels for grain refinement.
The ECA pressing induces shear deformation in the sample
by passing through an angular channel of equal in cross
section. As the cross sectional area of the sample does not
change during the ECA pressing, a very high accumulated
strain such as 4–12 could be imparted to the sample by
repetition of the pressing. Each pass of the ECA pressing
induces effective strain approximately 1.0 to the sample. It
was noted during the investigation that temperature and time
of the cementite spheroidization may be reduced significantly
in the ECA pressed steels.9,12) The spheroidization in ECA
pressed sample occurred fully when heated at 873 K only for
1 hour. Larger strains imparted to the sample during ECA
pressing must have increased the kinetics of the spheroidiza-
tion.
In this study, a more detailed study on the spheroidization
of ECA pressed carbon steel was conducted. The ECA
pressed steels processed up to 12 passes were heated for 1 to
72 hours at temperatures ranging from 723 to 973 K.
Morphology of the cementites after the heat treatment was
observed using scanning electron microscopy (SEM) and
transmission electron microscopy (TEM) to study the
progress of the spheroidization. In addition, tensile properties
of the steels with the spheroidization treatment were
evaluated.
2. Experimental Procedure
For this experiment, a low carbon steel (Fe–0.15%C–
0.25%Si–1.1%Mn (mass%)) in a form of bar was used. The
sample received was in a normalizing heat treatment
condition at 1223 K for 30 minutes followed by air cooling.
The average diameter of grains of the as-received sample was
approximately 30 mm. The bar sample was machined into a
rod of 18 mm in diameter and 120 mm in length for the ECA
pressing experiment. Prior to ECA pressing, the rod was
preheated in a furnace at 623 K for 10 minutes for uniform
 Spheroidization of Low Carbon Steel Processed by Equal Channel Angular Pressing
plunger
sample f
Fig. 1 A schematic illustration of the ECAP die used in the present
investigation.
heating and inserted into a die of ECA press as schematically
illustrated in Fig. 1. The inner contact angle and the arc of
curvature at the out point of contact between the channels of
the die were 90 and 20 respectively. The die geometry was
designed to yield an effective strain of about 1 per each pass.
The die was maintained at the same preheating temperature
and its inner surface was coated with graphite lubricant in
order to reduce the friction between the sample and die wall.
The pressing speed of upper punch was at 2 mm/s.
During ECA pressing, the sample was rotated 180 around
its longitudinal axis between the passes. This pattern of the
pressing is commonly designated as route C.13) Route C was
selected since it restores the shape of the original segment at
each even number pass and thereby nearly equiaxed ultrafine
grain structure can be obtained. The ECA pressed samples
were annealed for spheroidization in a temperature range
from 753 to 973 K for 1 to 72 hours. The microstructures of
both steels thus produced were examined using an optical
microscopy, TEM and SEM. Room temperature tensile
properties of the steels were measured with the initial strain
rate of 1:33  10 3 s 1 by an Instron machine.
3. Experimental Results and Discussion
Figure 2(a) shows the microstructure of carbon steel used
in this study. The pearlite colony distributed uniformly
throughout the sample and its area fraction was approx-
imately 15%. The ferrite grains and pearlite colonies were
equiaxed in morphology and size of the both phases was
30 mm. A TEM micrograph of the pearlite colony is also
shown in Fig. 2(b). The micrograph revealed a well-aligned
lamellar morphology cementites in the colony.
After the sample was ECA pressed for 4 passes, morphol-
ogy of the cementites were changed significantly as shown in
1631
Fig. 2 Microstructures of as-received 0.15%C steel: (a) Optical micro-
graph showing ferrite and pearlite; (b) TEM micrograph showing lamellar
structure in pearlite.
Fig. 3. The cementites became wavy (Fig. 3(a)) and severed
(Fig. 3(b)) from the severe plastic deformation imposed by
the ECA pressing. The difference in morphology of cemen-
tites deformed is believed to be caused by the orientation of
the cementite in relation to the shear deformation direction of
the sample. When the cementite is aligned perpendicular to
the shear deformation plane, the thinned and severed
morphology might be resulted. On the other hand, if aligned
with some angle, they might become wavy as shown in the
figure. Both of these morphologies were observed on the
cross section of the sample in transverse direction. In a cold
drawn wire, the wavy cementite is observed only on the cross
section in transverse and the severed one on cross section in
longitudinal direction.1)
One of the interesting things to note from the micrographs
is that the cementite deformed without fracturing even after
such a severe deformation. Previous studies indicated that the
cementite fractured due to its lack of slip systems under such
a heavy plastic deformation.14) The compressive stress
component of the pressing might have some role in
preventing the fracturing of the cementite. In addition,
dislocation density at the ferrite in the colony was signifi-
cantly higher than that in the ferrite grains, as reported in
previous studies.15)
The ECA pressed samples were annealed at 753 K for
various times for spheroidization (Fig. 4). When the sample
was annealed for 1 hour, the morphology of the cementite
remained similar to that of the as-ECA pressed sample. As
the holding time was increased to 24 hours, the cementite
1632 D. H. Shin, S. Y. Han, K.-T. Park, Y.-S. Kim and Y.-N. Paik

started to spheroidizes in a process like the Reyleigh

Fig. 3 TEM micrographs of pearlite in 4 pass steel samples: (a) curled and
wavy cementite lamellar plates; (b) severely necked cementite fragments
parallel to each other.
instability of a liquid jet.16) The defects such as kinks and
ledges on the surface cementite caused by the severe plastic
deformation might have caused the instability in the lamellar
structure of the cementite. After annealing for 72 hours, most
of the cementites became spheroidized with an aspect ratio
close to 1 (Fig. 4(c)). The average size of the cementites was
approximately 0.06 mm in diameter. The spheroidized ce-
mentite size is significantly smaller than that obtained
through conventional spheroidization treatment, 0.1–1 mm.17)
Effect of temperature on the cementite spheroidization in
ECA pressed steel was investigated by annealing the sample
at 783 K or 813 K for 1 hour (Fig. 5). As the temperature is
increased, the spheroidization rate was observed to increase.
The microstructure of the sample annealed at 783 K for 1
hour appeared to be similar to that annealed 753 K for 24
hours (Fig. 4(b)). When the sample was annealed at 813 K,
the cementites became fully spheroidized like the sample
annealed 753 K for 72 hours (Fig. 4(c)). The average size of
the spheroidized cementites was approximately 0.05 mm.
Although the spheroidization occurred mainly at the sites of
former colony as noted in Figs. 5(b) and (d), a small fraction
of the cementites formed at the boundaries of submicron
grains formed by the ECA pressing. The white arrows in Fig.
5(d) indicate such cementites. For these cementites to form,
there should be some carbon dissolved over its solubility
limit and diffuse along the grain boundaries. This phenom-
enon is an indirect evidence of carbon dissolution into ferrite
matrix by severe deformation as reported by Hono et al.6) in a
carbon steel after severe plastic deformation. During the

Fig. 4 TEM micrographs of pearlite in 4 pass steel samples annealed at 753 K for various annealing time: (a) 1 h; (b) 24 h; (c) 72 h.
 Spheroidization of Low Carbon Steel Processed by Equal Channel Angular Pressing
Fig. 5 TEM and SEM micrographs of pearlite in 4 pass steel samples annealed for 1 h: (a), (b) 783 K; (c), (d) 813 K.
annealing treatment, the grain size of ferrite remained close
to 0.3 mm in diameter, which might be due to a pinning effect
of such grain boundary cementites.
As the annealing temperature is further increased to 873 K
and 973 K, the spheroidized cementites and ferrite grains
were observed to coarsen (Fig. 6). The annealing time was
also 1 hour. A SEM micrograph of the sample annealed at
873 K shows that the cementite coarsened to 0:08 mm and
the ferrite grains became recrystallized and coarsened to
7 mm in diameter. The coarsening of the cementite was
more significant at grain boundaries of ferrite phase. The size
was increased to 0.27 mm. As the annealing temperature is
increased further to 973 K, the cementite in the ferrite matrix
was coarsened more to 0.21 mm and those at grain boundaries
were measured to be 0.38 mm. One of the interesting
characteristics with sample annealed at 973 K is that the
spheroidized cementite was somewhat dispersed throughout
the ferrite matrix. In other words, the spheroidization of the
cementite was not localized to the former pearlite colony
area. Higher annealing temperature might have promoted the
diffusion of carbon to ferrite matrix and precipitation of the
cementite. Under the two annealing conditions, the ferrite
matrix was recrystallized completely. With the annealing
conditions in proceeding sections, samples were either
polygonized or partially recrystallized.
The effect of accumulated strain on the spheroidization
behavior was studied by increasing number of ECA passes on
the steel. Figure 7 shows TEM micrographs of samples after
8 and 12 passes of ECA pressing at 723 K. The samples were
examined without any additional annealing treatment. A
1633
partial spheroidization of the cementite was noted with the
sample after 8 passes and full spheroidization after 12 passes.
The increase in the accumulated strain to 12 has reduced
the spheroidization temperature significantly down to 723 K.
This indicates that the large strain imparted on the sample
could reduce the temperature of spheroidization by 300 K.
In addition, the spheroidization time was reduced approx-
imately to 12 minutes. Each of the ECA pressing takes about
1 minute. These results shows that the defects such as
dislocations as well as carbon dissolution introduced by
severe deformation imposed on steels could increase the
kinetics of spheroidization significantly.
The samples annealed at 873 K and 973 K for 1 hour were
tested for their tensile properties along with as-received and
as-ECA pressed samples (Fig. 8). The annealed samples were
selected since the matrix is completely recrystallized. The
yield strength and fracture elongation of the as-received
sample was 320 MPa and 30%, respectively. The yield
strength of the as-ECA pressed sample was increased to
900 MPa, but the fracture elongation was reduced to
10%, due to work hardening and grain refining effects.
Annealing at 873 K reduced the yield strength to 450 MPa
and recovered the elongation to 28%. The strength of the
sample was higher by 45% and elongation was similar
compared with the as-received sample. The grain refining as
noted in Fig. 6(a) should have contributed to the increased
strength. The annealing treatment at 973 K, on the other hand,
reduced the yield strength of the sample lower than that of as-
received sample to 310 MPa. In addition, the fracture
elongation was improved to 37%. The spheroidization
1634 D. H. Shin, S. Y. Han, K.-T. Park, Y.-S. Kim and Y.-N. Paik

Fig. 6 SEM micrographs of microstructural change of ferrite and pearlite Fig. 7 TEM micrographs of pearlite in pressed steel samples: (a) 8 passes;
regions in 4 pass steel samples annealed at various temperature for 1 h: (b) 12 passes.
(a) 873 K; (c) 973 K.

(1) The ECA pressing resulted in wavy and severed

1000 cementites in the 0.15%C carbon steel. The cementites,
0.15C steel however, were not fractured during the pressing.
As-received (2) The spheroidization speed of ECA pressed steel
800 As-ECAPed
873K, 1hr annealed increased with annealing temperature and time. When
973K, 1hr annealed
Stress, σ / MPa

annealed at 753 K, the spheroidization occurred after 72

600 hours. When the temperature was increased to 783 K, it
took about one hour. Further increase in annealing
temperature resulted in coarsening of ferrite grains and
400
cementites.
(3) The spheroidization occurred mainly at former pearlite
200 colony sites when the annealing temperature is lower
than 873 K. When the temperature was increased to
973 K, the spheroidized cementites formed not only at
0 former pearlite colony sites but also inside ferrite
0 5 10 15 20 25 30 35 40
grains.
Strain, % (4) Increase in accumulated strains in the ECA pressed
Fig. 8 Stress-strain curves of pressed and annealed steel.
steel was observed to decrease the spheroidization
temperature and time. As the steel was pressed up to 12
passes, the spheroidization was observed during the
treatment of the sample at 973 K improved the formability of pressing. The pressing temperature was 723 K and total
the sample over the as-received sample. This beneficial effect pressing time was approximately 12 minutes.
is mainly due to the smaller size of the spheroidized
cementites formed in the sample. Acknowledgments

4. Conclusions This work was sopported by Korea Ministry of Science

and Technology through ‘21st Century New Frontier
The investigation of spheroidization behavior of cementite Research and Development Program’ and ‘National
in ECA pressed steel led to following conclusions: Research Laboratory Program’.
 Spheroidization of Low Carbon Steel Processed by Equal Channel Angular Pressing
REFERENCES
1) J. Languillaume, G. Kapelski and B. Baudelet: Acta Mater. 45 (1997)
1201.
2) J. Languillaume, G. Kapelski and B. Baudelet: Mater. Lett. 33 (1997)
241.
3) I. E. Dolzhenkov and I. N. Lotsmanova: Met. Sci. Heat Treat. 17 (1975)
211.
4) L. Heron: Metallurgia 80 (1969) 53.
5) K. Hono, M. Ohnuma, M. Murayama, S. Nishida, A. Yoshie and T.
Takahashi: Scrip. Mater. 44 (2001) 977.
6) M. H. Hong, W. T. Reynolds Jr., T. Tauri and K. Hono: Metall. Mater.
Trans. A30 (1999) 717.
7) F. Danoix, D. Julien, X. Sauvage and J. Copreaux: Mater. Sci. Eng.
A250 (1998) 8.
8) D. H. Shin, B. C. Kim, Y. S. Kim and K. T. Park: Acta Mater. 48 (2000)
1635
2247.
9) D. H. Shin, B. C. Kim, K. T. Park and W. Y. Choo: Acta Mater. 48
(2000) 3245.
10) D. H. Shin, I. Kim, J. Kim and K. T. Park: Acta Mater. 49 (2001) 1285.
11) D. H. Shin, Y. S. Kim and E. J. Lavernia: Acta Mater. 49 (2001) 2387.
12) K. T. Park, Y. S. Kim and D. H. Shin: Metall. Mater. Trans. A32 (2001)
2073.
13) Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto and T. G. Langdon: Acta
Mater. 46 (1998) 3317.
14) A. Inoue, T. Ogura and T. Masumoto: Trans. JIM 17 (1977) 143.
15) J. D. Embury and J. P. Hirth: Acta Mater. 42 (1994) 2051.
16) S. Kou: Transport Phenomena and Materials Processing, (John Wiley
and Sons Inc., New York, 1996) 387.
17) R. E. Reed-Hill and R. Abbaschian: Physical Metallurgy Principles,
(PWS Publishing Co., Boston, 1994) 591.