DOI 10.1515/htmp-2013-0094 High Temp. Mater. Proc.

2014; 33(6): 553 – 561

Haitao Liu*, Yanchong Yu, Weiqing Chen, Qingxian Wang and Guangshun Wang

Effect of Mn/S Ratio on the Hot Ductility

of Eco-friendly Bi-S based Free Cutting Steel
Abstract: The hot ductility of eco-friendly Bi-S based free based free cutting steel has been developed, in which lead
cutting steels with different Mn/S ratios was studied using is replaced by bismuth [1–4]. The Bi-S based steels have
a Gleeble-1500 thermal–mechanical simulator. The hot similar excellent machinability compared to the Pb-S
ductility of the steel was found to depend on the Mn/S based steels because bismuth and lead have similar phys-
ratio, and the Mn/S ratio of the steel should be greater ical properties [3–6].
than 3.5 for hot rolling of billets without crack develop- Unfortunately, eco-friendly Bi-S based free cutting
ment. The low Mn/S ratio would inhibit the occurrence of steels are not yet widely used for commercial purpose
the dynamic recrystallization and cause the formation of because their hot rolling is more difficult than that of
the low melting point sulfide Fe-rich (Fe,Mn)S as second- the Pb-S based free cutting steels [7–12]. The steels were
ary phases, which could obviously reduce the strength of reported to have relatively low ductility to restrict the
the grain boundary and resulted in the formation of cracks hot working temperature range, which could induce the
along the grain boundary. The higher the Mn/S ratio in the cracking during hot rolling [9–12]. The poor hot workabil-
steel, the lower the Fe content in the Fe-rich (Fe,Mn)S ity has generally been associated with the Mn/S ratio, but
phases. When the Mn/S ratio in the steel was high enough, few studies have been done about the effect of Mn/S ratio
the sulfide phases in the steel were mainly MnS as primary on the hot ductility of the steels with high sulfur content
inclusions and the low melting point sulfide phases could [13–14]. In this paper, a critical Mn/S ratio for the eco-
be effectively avoided forming. While the Mn/S ratio could friendly Bi-S based free cutting steel has been established
influence the hot ductility of the steel over the whole tem- by hot tensile test, which is in good agreement with the
perature range of 900–1200 °C, the segregation of bismuth theoretical deduction results from literature [15], and the
along grain boundary could be harmful to the hot ductility mechanism of the effect of Mn/S ratio on the hot ductility
in addition to the lower Mn/S ratio for the temperature of the steel has also been investigated in relation to the
was no more than 1050 °C. microstructural changes.

Keywords: free-cutting steel, hot ductility, Mn/S ratio,

bismuth, sulfide phases 2 Experimental procedures
PACS® (2010). 81.40.-z The tested steels with different Mn/S ratios were melted in
a vacuum induction furnace at a temperature of 1580 °C.
Pure Bi and ferromanganese alloys lines on a pure iron
*Corresponding author: Haitao Liu: State Key Laboratory of
Advanced Metallurgy, and School of Metallurgical and Ecological
wire was inserted into the liquid steel by a feeding device
Engineering, University of Science and Technology Beijing, Beijing of the furnace after melting, then refined for 2 min, when
100083, China. E-mail: liuhaitao1204@126.com the temperature and composition were uniform, the
Yanchong Yu, Weiqing Chen: State Key Laboratory of Advanced molten steel was cast to a 6.5 kg ingot with a pouring tem-
Metallurgy, and School of Metallurgical and Ecological Engineering, perature 1560 °C. The ratio of Mn/S was set to the varing
University of Science and Technology Beijing, Beijing 100083,
range from 1 to 4 in order to find the critical value for the
China
Qingxian Wang, Guangshun Wang: Qingdao Iron and Steel Group Bi-S based free cutting steel to improve the hot ductility
Co., Ltd., Qingdao 266043, China of these steels. The chemical compositions are shown in
Table 1.
These ingots were first reheated at 1200 °C for 2 h in
a heating furnace, and then, an air forging hammer was
1 Introduction used to forge them into wire rods with a diameter of
15 mm. During the forging process, the temperature was
Since the use of the environmentally detrimental Pb-S always >1050 °C, and finally, the wire rods were air cooled
based free cutting steel was limited, an eco-friendly Bi-S to room temperature. Specimens with a dimension of
554 H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel

Table 1: Chemical compositions of tested steels, mass%

Steels C Si Mn S P Bi Mn/S ratio

1 0.07 0.06 0.41 0.34 0.054 0.10 1.21

2 0.08 0.09 0.92 0.31 0.055 0.17 2.97
3 0.08 0.02 1.28 0.35 0.068 0.15 3.66
4 0.07 0.01 1.25 0.31 0.050 0.17 4.03

10 mm diameter and 120 mm length for hot tensile testing

were prepared from these wire rods.
The hot tensile test was performed with a Gleeble-
1500 thermal-mechanical simulator under argon protec-
tion atmosphere, using a constant strain rate of 10−2 s−1 in
the range of 850–1200 °C with an interval of 50 °C. The Fig. 1: Hot ductility curves of all tested steels
specimens were first heated to 1350 °C with a heating rate
of 10 °C s−1 and held for 5 min, then cooled to each testing
content of steel 4, because bismuth in the steel also had a
temperature with a cooling rate of 3 °C s−1 and held for
detrimental influence on the hot ductility mainly in this
1 min before the tensile testing. After rupture, the samples
temperature range.
were immediately quenched by water spraying to main-
Gonzalo and Oscar et al. suggested that there was a
tain the microstructure at the testing temperature, and the
critical Mn/S ratio, which could be considered as a border
reduction in area (RA) was measured to evaluate their hot
between good and bad results of hot ductility of steels,
ductility.
and this critical value depended on the S content of the
The fracture surfaces of these steels were observed
steel [15]. A comparison between the experimental Mn/S
through a scanning electron microscope (SEM). The longi-
ratio in this work and the theoretically deduced critical
tudinal sections close to the point of fracture were pre-
Mn/S ratio by Gonzalo and Oscar et al. is shown in Fig. 2.
pared and etched with a picral solution, and then, the
It can be seen that the agreement between the experi­
­microstructures and the inclusions were detected by a
mental results in this work and the theoretical deduction
combination of the SEM in backscattered electron (BSE)
by Gonzalo and Oscar et al. was well when the S content of
mode and secondary electron (SEI) mode, and the energy
the steel was in the range of 0.30–0.35 wt-%, and the value
dispersive X-ray spectroscopy (EDS).
of the critical ratio at this time was about 3.5. Therefore,

3 Results and discussion

3.1 Effect of Mn/S ratio on hot ductility

The RA of all tested steels with different Mn/S ratios in the

testing temperature range is shown in Fig. 1. The hot duc-
tility of Bi-S based free cutting steel was found to depend
on the Mn/S ratio. The steels showed the largest hot duc-
tility at the ratio of 3.66 and 4.03, while the RA was obvi-
ously increased with the increase of the Mn/S ratio for the
ratio in the steel was less than 3.66. Meanwhile, the Mn/S
ratio could influence the hot ductility of the steel over the
whole temperature range of 900–1200 °C, for example,
the hot ductility of steel 1 with the ratio of 1.21 was all very
bad over this range. The RA of steel 4 was lower than that Fig. 2: A comparison between the experimental Mn/S ratio in this
of steel 3 at 850–1000 °C even though it has a higher Mn/S work and the theoretically deduced critical Mn/S ratio by Gonzalo
ratio which might be attributed to the higher bismuth and Oscar et al. [15]
  H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel 555

Fig. 3: Fracture morphology of tensile tested samples: (a) steel 1 at 1150 °C; (b) steel 1 at 1000 °C; (c) steel 2 at 1150 °C; (d) steel 2 at 1000 °C;
(e) steel 4 at 1150 °C; (f) steel 4 at 1000 °C

the Mn/S ratio of the eco-friendly Bi-S based free cutting 3.2 Fracture morphology
steel, in which the S content is around the range of 0.30–
0.35 wt-%, should be greater than 3.5 for hot rolling of The fracture surfaces of the tensile test specimens were
billets without crack development. observed by SEM. The steels with good hot ductility all
556 H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel

e­xhibited ductile dimple fractures, but intergranular

brittle fractures when the hot ductility was poor. The
typical fracture surfaces of the tensile test specimens are
shown in the SEM photos (Fig. 3). The fractures of steel 1 at
1150 and 1000 °C both exhibited intergranular brittle
failure (Fig. 3a and b), which could indicate that the grain
boundary strength of the steel with low Mn/S ratio was
lower. Steels 2 and 4 at 1150 °C both showed a ductility
failure (Fig. 3c and e), but their fractures at 1000 °C were
mainly intergranular (Fig. 3d and f), which agrees with the
values of the hot ductility given in Fig. 1.

3.3 E
 ffect of Mn/S ratio on dynamic
recrystallization

The flow stress curves for steels 1, 2, and 4 are shown in

Fig. 4. The temperature for the onset of dynamic recrystal-
lization (DRX) can generally be obtained from the flow
stress curves by noting the first test temperature at which
load fluctuations occur in the austenite, because the oc-
currence of DRX will provide an additional softening to
offset part of the work hardening caused by deformation
[16–18]. Mintz and Mohamed have reported that DRX is
­essential for good ductility, and the ductility for high S
steel improves only at higher temperatures when DRX is
well established [17–20]. From the examination of the flow
stress curves in relation to the hot ductility curves in this
work, it can be seen that there was no DRX occurring for
steel 1 in the whole testing temperature range, but the
onset temperatures of DRX for steels 2 and 4 were 1150 and
1100 °C respectively. The Mn/S ratio had a remarkable
effect on the onset of DRX, the higher the Mn/S ratio, the
lower the onset temperature of DRX. Meanwhile, the re-
covery temperature of the hot ductility for these steels
agreed with the onset of DRX in this work. DRX, i.e. grain
boundary migration, can move grain boundaries away
from microcracks, leading to the isolation of microcracks
and preventing their coalescence at grain boundaries
[16–22].

3.4 G
 eneration of sulfide inclusions with
different Mn/S ratios Fig. 4: Flow stress–strain curves as function of tensile temperature
for the examined steels (arrows indicate dynamic recrystallization):
The typical longitudinal microstructures of the quenched (a) steel 1; (b) steel 2; (c) steel 4
fractures for steel 1 with the Mn/S ratio of 1.21 were ob-
served using SEM and EDS, and the result was shown in or FeS phases, which have a low melting point [23]. Mean-
Fig. 5. A lot of precipitated phases were observed at while, from Figs. 3a, 3b and 4a, it can also be seen that
the grain boundary. EDS analysis of these precipitated the strength of steel 1 was so poor and this result was
phases revealed that they were Fe-rich (Fe,Mn)S phases mainly caused by the weak grain boundary. In other
  H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel 557

Fig. 5: Typical longitudinal microstructures of quenched fractures for steel 1: (a) microstructure near fracture at 1200 °C; (b) cracks
magnification at 1200 °C; (c) microstructure near fracture at 1150 °C; (d) microstructure near fracture at 1100 °C

Fig. 6: Typical morphologies of the sulfide inclusions on the surface of fractures in steel 1: (a) spherical sulfides; (b) rod-like sulfides;
(c) dendritic or rod-like sulfides; (d) bead or chain-like, fish-bone eutectic, and clusters of fine sulfides
558 H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel

words, these phases with low melting point would obvi- the low Mn/S ratio in steel 1. Meanwhile, the morphology
ously reduce the strength of the grain boundary and re- of these Fe-rich (Fe,Mn)S phases could be mainly classi-
sulted in the formation of cracks along the grain boundary fied to six types as follows: spherical, rod-like, dendritic
[24]. or rod-like, bead or chain-like, fish-bone eutectic, and
As it is known, most of the sulfide inclusions usually clusters of fine sulfides [26–27]. The typical morphologies
precipitate as MnS in the enriched interdendritic liquid of them were shown in Fig. 6.
at the last stage of solidification during freezing [23–27]. Fig. 7 showed the SEM and EDS photos of the sulfide
However, the data in the phase diagram of Fe-rich Fe-Mn-S inclusions which formed in sequence in steel 1 at 1000
system also shows that an insufficient addition of Mn to S and 900 °C. From Fig. 7, it also can be seen that there was
containing steels would cause the formation of the liquid a higher Fe content in the Fe-rich sulfide phases precipi-
phase or Fe-rich sulfide phase in a peritectic or a eutectic tated later.
mode [28–30]. In addition, the segregation of sulphur in The typical longitudinal microstructures of the
the steel would promote the formation of the Fe-rich quenched fractures for steel 2 with the Mn/S ratio of 2.97
sulfide phases along the grain boundary [28–29]. There- were shown in Fig. 8. It can be seen that there were few
fore, in this work, few MnS as primary sulfide inclusions Fe-rich sulfide phases precipitated at 1150 °C (Fig. 8a), but
were formed but many Fe-rich (Fe,Mn)S as secondary many bead or chain-like Fe-rich (Fe,Mn)S phases and
sulfide phases after the primary crystallization of the Fe cracks at 1100 °C along the grain boundary (Fig. 8b) in
phase during solidification, which would be attributed to accord with its worse ductility. Meanwhile, the Fe content

Fig. 7: SEM morphology and EDS analysis of the typical sulfide inclusions formed in sequence for steel 1: (a) 1000 °C; (b) 900 °C
  H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel 559

sulfide inclusions, and a high enough Mn/S ratio could

effectively avoid the formation of the low melting point
sulfide phases. However, the hot ductility of steel 4 was
still poor when the temperature was no more than 1050 °C
although the Mn/S ratio of it was high enough. From Figs.
9a and 10b below, it can be seen that the poor hot ductil-
ity and the cracks along grain boundary in steel 4 at this
time were mainly due to the segregation of bismuth at the
grain boundary.

3.5 Grain boundary segregation of bismuth

From the observation and analysis of the longitudinal

­microstructures of the quenched fractures for steels 2 and
4 using the SEM in SEI and BSE modes. The cracks along
austenite grain boundaries accompanied by the segrega-
tion of bismuth films had been founded when the testing
temperature was no more than 1050 °C in both steels. The
typical SEM photos of the bismuth films segregated along
austenite grain boundary at 1050 °C for steel 2 and steel 4
were shown in Fig. 10.

Fig. 8: Typical longitudinal microstructures of quenched fractures for

steel 2: (a) microstructure near fracture at 1150 °C; (b) microstructure
near fracture at 1100 °C; (c) EDS of sulfide phases

in the Fe-rich (Fe,Mn)S phases was lower compared with

that of steel 1 (Fig. 8c), because there was a higher Mn/S
ratio in steel 2 and more Mn would combine with S in
steel 2.
The typical longitudinal microstructures of the
quenched fractures at 1000 °C for steel 4 with the Mn/S Fig. 9: Typical longitudinal microstructures of quenched fractures
ratio of 4.03 were shown in Fig. 5. It can be seen that the at 1000 °C for steel 4: (a) microstructure near fracture; (b) EDS of
sulfide phases in the steel were mainly MnS as primary sulfide phases
560 H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel

the Fe-rich (Fe,Mn)S as secondary sulfide phases, these

phases with low melting point would obviously reduce the
strength of the grain boundary and resulted in the cracks
along the grain boundary.
The higher the Mn/S ratio in the steel, the lower the Fe
content in the Fe-rich (Fe,Mn)S phases. When the Mn/S
ratio in the steel was high enough, the sulfide phases in
the steel were mainly MnS as primary sulfide inclusions
and the formation of low melting point sulfide phases
could be effectively avoided.
While the Mn/S ratio could influence the hot ductil-
ity of the steel over the whole temperature range of 900–
1200 °C, the segregation of bismuth along grain boundary
could be harmful to the hot ductility in addition to the
lower Mn/S ratio for the temperature was no more than
1050 °C.

Received: May 24, 2013. Accepted: July 12, 2013.

References
[1] K. Sander, J. Lohse and U. Pirntke, Heavy Metals in Vehicles
‒ Final Report, Institut für Ökologie und Politik GmbH,
Hamburg, pp. 4‒44 (2000).
[2] J. Lohse, K. Sander and M. Wirts, Heavy Metals in Vehicles II
‒ Final Report, Institut für Ökologie und Politik GmbH,
Hamburg, pp. 45‒91 (2001).
[3] R. Philip, B. Volker and E. Ingo, Steel Res. Int., 178(12), 908‒914
(2007).
Fig. 10: Typical SEM photos of the bismuth films segregated along [4] M. Somekawa, M. Kaiso and Y. Matsushima, Kobelco
austenite grain boundary (BES mode): (a) steel 2 at 1050 °C; Technology Review, 24(1), 9‒13 (2001).
(b) steel 4 at 1050 °C [5] L. Vea, SAE Technical Paper on ‘New Steel Bar Products and
Processing for Automotive Applications’, Society of Automotive
Engineers, Detroit, pp. 77‒82 (1997).
Pure Bi has a melting point of 271.3 °C, the bismuth [6] H.T. Liu and W.Q. Chen, Steel Res. Int., 83(12), 1172‒1179
films present at austenite grain boundary would exist (2012).
as liquid forms at the testing temperatures and would sig- [7] J. Klapsia, Hutnicke Listy, 58(9), 18‒19 (2003).
[8] J. Klapsia and E. Karczmarczyk, Hutnicke Listy, 55(10), 36‒38
nificantly reduce the strength of the grain boundary and
(2000).
cause grain boundary embrittlement. Therefore, the seg- [9] Y. Kim, H. Kim, S.Y. Shin, K. Rhee, S.B. Ahn, D.L. Lee, N.J. Kim
regation of bismuth along grain boundary also would be and S. Lee, Metall. Mater. Trans. A, 42, 3095‒3105 (2011).
harmful to the hot ductility of the eco-friendly Bi-S based [10] S.K. Cho, J. Korean Inst. Met. Ma., 34(7), 822‒829 (1996).
free cutting steel in addition to the lower Mn/S ratio. [11] H. Kim, M. Kang, S.Y. Shin and S. Lee, Mat. Sci. Eng. A, 568,
8‒19 (2013).
[12] Y. Kim, H. Kim, S.Y. Shin, K. Rhee, S.B. Ahn, D.L. Lee, N.J. Kim
and S. Lee, Metall. Mater. Trans. A, 43(3), 882‒892 (2012).
4 Conclusions [13] K. Oikawa, H. Ohtani, K. Ishida and T. Nishizawa, ISIJ Int., 35(4),
402‒408 (1995).
The hot ductility of the eco-friendly Bi-S based free cutting [14] L.K. Bigelow and M.C. Flemings, Metall. Mater. Trans. B, 6(2),
steel was found to depend on the Mn/S ratio, and the 275‒283 (1975).
[15] G.A. Toledo, O. Campo and E. Lainez, Steel Research, 64(6),
Mn/S ratio of the steel should be greater than 3.5 for hot
292‒299 (1993).
rolling of billets without crack development. [16] E. López-Chipres, C. Maldonado, A. Bedolla-Jacuinde,
The low Mn/S ratio could restrain the occurrence M. EI-Wahabi and J.M. Cabrera, Mat. Sci. Eng. A, 480(1), 49‒55
of dynamic recrystallization and cause the formation of (2008).
  H.T. Liu et al., Effect of Mn/S Ratio on the Hot Ductility of Eco-friendly Bi-S based Free Cutting Steel 561

[17] B. Mintz, R. Abushosha and J.J. Jonas, ISIJ Int., 32(2), 241‒249 [24] C. Zhang, E.B. Yue and S.T. Qiu, Continuous Casting, 27(6),
(1992). 6‒10 (2008).
[18] B. Mintz and Z. Mohamed, Mater. Sci. Technol., 4(10), 895‒902 [25] Z.Z. Liu, M. Kuwabara, R. Satake and T. Nagata, ISIJ Int., 49(7),
(1988). 1087‒1093 (2009).
[19] A. Cowley and B. Mintz, Mater. Sci. Technol., 20(11), 1431‒1439 [26] L.K. Bigelow and M.C. Flemings, Metall. Mater. Trans. B, 6,
(2004). 275‒283 (1975).
[20] B. Mintz, R. Abu-Shosha and M. Shaker, Mater. Sci. Technol., [27] E.T. Turkdogan, Metall. Mater. Trans. B, 6, 663‒664 (1975).
9(10), 907‒914 (1993). [28] H. Ohtani, K. Oikawa and K. Ishida, High Temp. Mat. Processes,
[21] F. Zarandi and S. Yue, Metall. Mater. Trans. A, 37(7), 2316‒2320 19(3), 197‒210 (2000).
(2006). [29] G.S. Mann and L.H. Van-Vlack, Metall. Mater. Trans. B, 7,
[22] E. López-Chipres, C. Maldonado, A. Bedolla-Jacuinde and J.M. 469‒475 (1976).
Cabrera, Mat. Sci. Eng. A, 460, 464‒470 (2007). [30] J. Miettinen and B. Hallstedt, Calphad, 22(2), 257‒273 (1998).
[23] G. Sridhar, S.K. Das and N.K. Mukhopadhyay, Eng. Fail. Anal.,
6(3), 155‒172 (1999).