Author’s Accepted Manuscript

Torsion delamination and recrystallized cementite

of heavy drawing pearlite wires after low
temperature annealing

Lichu Zhou, Feng Fang, Linfeng Wang, Huaqing

Chen, Zonghan Xie, Jianqing Jiang
www.elsevier.com/locate/msea

PII: S0921-5093(17)31651-9
DOI: https://doi.org/10.1016/j.msea.2017.12.055
Reference: MSA35897
To appear in: Materials Science & Engineering A
Received date: 18 July 2017
Revised date: 11 December 2017
Accepted date: 12 December 2017
Cite this article as: Lichu Zhou, Feng Fang, Linfeng Wang, Huaqing Chen,
Zonghan Xie and Jianqing Jiang, Torsion delamination and recrystallized
cementite of heavy drawing pearlite wires after low temperature annealing,
Materials Science & Engineering A, https://doi.org/10.1016/j.msea.2017.12.055
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Torsion delamination and recrystallized cementite of heavy drawing pearlite

wires after low temperature annealing

Lichu Zhoua, Feng Fanga*, Linfeng Wangb, Huaqing Chenb, Zonghan Xiec,d, Jianqing Jianga
a- Jiangsu Key Laboratory of Advanced Metallic Materials, Southeast University, Nanjing, 211189, China

b- Jiangyin Walsin Steel Cable Co., Ltd, Jiangyin, 214443, China

c- School of Mechanical Engineering, University of Adelaide, SA 5005, Australia

d- School of Engineering, Edith Cowan University, WA 6027, Australia

Corresponding author:
Feng FANG：
Tel.: +86 25 52090630; E-mail address: fangfeng@seu.edu.cn

Address: School of Materials Science and Engineering, Southeast University,

Jiangning District, Nanjing, 211189, China

Abstract：
Relationship between torsion delamination and cementite recrystallization in heavily
cold drawn pearlitic steel wires was investigated following low temperature annealing. It
was found that torsion delamination occurred inevitably when the wires underwent annealing
at temperatures ranging from 210℃ to 350℃ with duration from 2 min. to 60 min. The
torsion angle corresponding to uniform torsional deformation of annealed wires was dropped
about 10% as compared to the as-drawn wires. The torsional tolerance is independent of
tensile mechanical properties when subjected to annealing treatment. Low temperature
annealing didn’t alter the morphology of cementite in the wires, but instead it changed
cementite structure from disordered (i.e., near amorphous) to nano-crystalline state. After
low temperature annealing, dislocation cells formed in ferrite during torsion tests. The
recrystallized cementite impeded dislocation motion and was found to break up around
high-density dislocation zones. Micro-cracks resulting from torsion delamination were
observed near damaged cementite sites. By contrast, in as-drawn steel wires cementite and
ferrite deformed in a coherent and coordinated manner. In doing so, the wires exhibited a
greater tolerance to torsion damage.
Keyword: pearlitic steel wires; annealing; cementite; recrystallization; torsion
1. Introduction
Owing to a unique combination of high strength and good ductility, pearlitic steel
wires have been widely used in engineering applications such as suspension bridge cables,
automotive tyre cords and cutting wires[1, 2]. They are produced by cold drawing process,
during which the strength of the wires increased significantly as a result of pearlite
interlamellar spacing reduction, increased dislocation density and texture formation [3-7].
To further improve the mechanical performance of pearlitic steel wires, much effort
has been made to investigate the microstructural evolution and its effect on the mechanical
property of the wires [8-16]. By increasing drawing strain to 6.5, pearlitic steel wires with
the tensile strength of 7 GPa have been prepared [17]. In addition to pearlite lamellar
refinement and dislocation multiplication in ferrite, the structural change of cementite during
drawing deformation and its role on strain hardening have attracted much attention.
Deformation-induced cementite decomposition has been widely reported [18-21]. Cementite
platelets may also transform into nano-crystalline or even amorphous state [1, 7, 22, 23]. On
the other hand, thermal annealing at a temperature as low as 483 K could restore cementite
in the as-drawn steel wires from amorphous to nano-crystalline state[23]. However, it has
always been a challenging task to ascertain the effect of cementite dissolution and cementite
crystalline structure on the mechanical property and deformation behavior of the wires, since
the thickness of cementite lamellae was often reduced to < 5 nm[22, 24]. In addition, high
dislocation density in ferrite also influenced the observation and analysis of deformation in
the wires.[3, 5, 25].
The combination of tensile and shear stresses in load-carrying applications imparts not
only high strength, but also a good ductility to the pearlitic steel wires. Torsion test is
commonly used to measure torque required to fracture pearlitic steel wires[26, 27] and, at the
same time, assess their torsional ductility. It is understood that the torsion property of the
pearlitic steel wires are affected by many factors, such as surface cracks[28], metallurgical
inclusion and abnormally large cementite [29, 30]. Torsion delamination is the direct
manifestation of torsion-caused material deterioration. Joung et al. [31] reported that torsion
delamination occurred in steel wires, accompanied by the strength reduction after
post-deformation annealing. On the other hand, low temperature annealing treatment was
found to simultaneously enhance the tensile strength and tensile ductility of cold drawn
pearlitic steel wires [23]. Redistribution of C atoms from dissolved cementite was considered
to be responsible for the torsion delamination. However, contradictory reports about the
diffusion directions of C atoms were provided by different researchers[32, 33]. It is worth
noting that the dissolution of cementite was believed to be insignificant in the wires with a
drawing strain of 1.5 to 2.5 [32-36]. Masaki Tanaka et al. have noted vanished cementite
lamellae beneath the delamination crack and contemplated that delamination fracture might
be a shear one associated with local severe plastic deformation [37]. Some other factors;
namely, texture, residual stress and size of wire, have also been taken into account [38-42].
Many different viewpoints have made it difficult to develop a clear, decisive understanding
of the mechanism governing torsion delamination in steel wires. However, a generally
accepted opinion is that occurrence of torsion delamination is almost inevitable, when
as-drawn pearlitic steel wires underwent low temperature annealing, during which there was
almost no change in its overall microstructure [28-31, 36, 37, 42, 43]. Therefore, the core
issue remains: how low-temperature annealing affects the torsion deformation process of
heavily drawn pearlitic steel wires?
The purpose of the present study is to reveal the mechanism of torsion delamination of
heavily drawn pearlitic steel wires subjected to low-temperature annealing. Torsion
performance of the wires was assessed before and after low-temperature annealing treatment.
Torsion fracture and associated cracks were examined. To identify the origin of torsion
delamination, microstructure, dislocation configuration and cementite structure were
analyzed. Finally, the relationship between cementite recrystallization and torsion
delamination of the pearlitic steel wires were discussed.

2. Experimental Methods
Materials used in this study were hot-rolled pearlitic steel rods with a composition in

the mass fraction (%) of Fe–0.86C–0.2Cr–0.45Si–0.6Mn–＜0.015S+P. After pickling and

phosphating [44], the rods (13.5 mm in diameter) were cold drawn to 5 mm in diameter
(ε=2.0) by multi-pass drawing process. The temperature of pearlitic steel wires during
drawing was below 100℃. The annealing of cold drawn wires was carried out in oil bath at

three different temperatures; namely, 210℃，280℃，350℃ , over a period of 2 min to 60 min.

The heat treated wires were cooled in air to room temperature.

Torsion tests of steel wires were conducted using CTT500 torsion tester. Fig. 1 shows
the test equipment and sketch map for torsion tests. The length of the test wires is 35 cm.
One of the end of the wires is fixed, meanwhile another end can be rotated. Rotational speed
was set at 360°/min. Both the torque and torsional angle were recorded in real time. Four
identical samples were tested under each condition. Torsional fracture angle of the annealed
wires are almost random. Torsion tests were stopped at different torsion angle for obtaining
samples for microstructure observation. Tensile tests were conducted at room temperature
using CMT5105 universal materials tester, operating at a speed of 4 mm/min. Four identical
samples were tested under each condition to determine average tensile strength and percent
elongation (i.e., ductility). The tensile tests were performed according to the Chinese
National Standard GB/T228-2002.
For microstructural observation, after polishing along longitudinal direction, the
specimens were etched by 4% nitric acid alcohol solution and examined using FEI Siron-400
scanning electron microscope (SEM). TEM samples were prepared by GATAN 691
Precision Ion Polishing System (PIPS) after the thickness of the samples was mechanically
thinned to below 40 um. Sample was cooled by liquid nitrogen in PIPS to avoid the
temperature rise. Ion angle and energy were set to 3° and 2 Kev at the final stage of TEM
sample preparation to minimize microstructure damage. The microstructure of the pearlitic
steel wires were examined and JEOL-2100F transmission electron microscope (HRTEM).

Fig. 1 CTT500 torsion tester and the mechanism of torsion test

3. Results
3.1 Microstructure of pearlitic steel wires
Fig. 2 shows the microstructures obtained from the longitudinal section of as-drawn
and annealed pearlitic steel wires. Both of the two wires comprised a layered structure
aligned along the drawing direction. Microstructure of the pearlitic steel wire was not
changed by those low temperature annealing.

Fig. 2 SEM images of pearlitic steel wires. a) as-drawn. b) annealed at 350℃ for 30 min

Fig. 3 shows the TEM images of as-drawn and annealed pearlitic steel wires. While
lamellar structure could still be seen in both wires, the cementite structure (thinner lamellae)
seemed to have been changed after annealing. In magnified images (shown as inserts in the
bottom right), cementite platelets in as-drawn wires have uniform contrast. While in
annealed wires as shown in Fig. 3b, cementite platelets have uneven contrast. The contrast
change in cementite was caused by cementite recrystallization [32, 45, 46]. As shown in
HRTEM image in Fig. 3c, the ferrite exhibits distinct lattice fringes, suggesting that the
crystallinity of the ferrite remained unchanged during the preparation process. By contrast,
lattice fringes are not apparent in cementite. In addition, discrete bright spots did not appear
from the cementite when analyzed by the Fast Fourier Transformation (FFT). The FFT
image of the cementite shows disordered structure. In agreement with previous results, the
HRTEM result indicates that under heavy cold drawing the cementite crystal layers were
transformed into amorphous or disorder-structure [7, 22, 23, 47]. During the low temperature
annealing treatment, it is understood that the crystal state of cementite was changed from
amorphous to nano-crystalline[23]. Fig. 3d shows HRTEM image of the annealed wires, in
which alternating layers of cementite and ferrite can be seen. Lattice stripes with different
directions can be observed in cementite, suggesting that the cementite had transformed into
nano-crystals with different crystallographic orientations. Although there is slight difference
in heat treatment temperature and time between reported literature[23] and this study, the
fine structure of cementite resulting from low temperature annealing is typically composed
of interconnected nanocrystal.

Fig. 3 TEM images of pearlite steel wires. a) as-drawn. b) annealed at 280℃ for 6 min; c)
HRTEM images of as-drawn wire. d) HRTEM images of wire after 280℃ for 6 min.

3.2. Mechanical properties

Fig. 4a shows the distribution of the angle of twist of pearlitic steel wires against
annealing temperature. As shown in the diagram, the torsion fracture angle of as-drawn
wires is about 6500°. In comparison, the torsion fracture angle of annealed wires has a
relatively large range but considerably lower than as-drawn wires. The variation of fracture
angle is presumably caused by torsion delamination. Torque-torsion angle curve of the three
types wires are shown in Fig. 4b to reveal the difference in torsion performance. During
torsion tests, the torque of as-drawn pearlitic steel wires followed a smooth, slowly
increasing path after the yield point, producing excellent torsion performance. In contrast,
following the yielding, there was a rapid decline in torque for the two annealed wires. The
torque decline has been reported as a result of crack formation that resulted in torsion
delamination [29, 36]. Notably, torsion angle of uniform torsion deformation of annealed
wires were only about 1/10 of the as-drawn wires. Afterwards, the torque fluctuated with
increasing torsion angle. At that time cracks were supposed to propagate in the annealed
wire. Further analysis of delamination crack would be presented in Fig. 6 to reveal the
interactions between crack and microstructure. Finally, torsional fracture angle could not be
used to represent torsional performance of annealed wires due to a significant variation in its
value. The cracks produced at small torsion angles may be responsible for significant drop in
torsional tolerance of annealing wire. Such drop in torsion tolerance and randomness of
fracture caused by low temperature annealing has a serious implication for safety-critical
applications where catastrophic failure cannot be tolerated.
 Fig. 4 a) Torsion angle of the wires corresponding to different annealing temperatures as a
function of annealing time; b) Torsion angle-torque curve for steel wires subjected to
different annealing temperatures.

Torsion test have shown a “random” nature in the torsion performance of heavily
drawn pearlitic wire after low-temperature annealing. By contrast, the tensile mechanical
properties of those steel wires are more consistent. For an example, Fig. 5a shows tensile
strength and elongation of the wire after annealing at 350℃ as a function of annealing time.
Tensile strength of the wire firstly increases at 2min and 4min. Then the strength gradually
decreased from about 1980 MPa at 2min to 1770MPa at 60min. It is believed that the
annealing strengthening of heavily drawn pearlite wires was related with redistribution of
residual stress, carbon diffusion and cementite recrystallization [23, 25, 31-33, 45, 46, 48].
Newly produced evidences have shown that cementite recrystallization should be the key
reason for the low temperature annealing strengthening of pearlitic steel wires prepared at
drawing strain of 2 (which is similar to this study)[23]. On the other hand, the annealing
soften is believed to result from defect recovery when the microstructure of steel wires
remains to be lamellar [25, 45]. The defect recovery also increases the ductility of the steel
wires. By increasing annealing time to 60min, the elongation at maximum strength of the
wires increased gradually from 1.5% to 4.6%.
 Fig. 5b shows typical engineering stress-strain curves of pearlitic steel wires (as cold
drawn, 280℃-6 min and 350℃-30min). The curves of three samples are different to each
other. Tensile strength and elongation at max-strength of the as-drawn wire were 1950MPa
and 1.60%. The 280℃ annealed wires show a slightly higher strength and ductility
(2117MPa and 2.3%). Meanwhile the 350℃ annealed wires show a lower strength and
obviously superior ductility (1810MPa and 4.3%).
In the first place, it is necessary to consider the torsion delamination was caused by
decrease of mobile dislocation density by annealing[49]. However tensile strength of wire
after 350℃ annealing has decreased significantly when annealing time is longer than 6min.
Meanwhile tensile ductility of the 350℃-60min annealed wire was about 3 times of the
as-drawn one. However, torsion performance of the wire is still deteriorated by annealing. It
seems decrease of mobile dislocation density was not the leading factor for torsion
deterioration after annealing. Abruption of torsion performance and tensile properties should
be related to the anisotropic fiber microstructure of heavy drawn pearlite wire[50].
Meanwhile there is an important question: why would the torsion tolerance suffer from the
annealing treatment?
 Fig. 5 a) Tensile strength and percent elongation of steel wires after 350℃ annealing
as a function of annealing time; b) Engineering strain- stress curves of the wires treated at
different annealing temperatures

3.3 Torsion delamination crack

Fig. 6a and 6b shows the torsion fracture of annealed wire and as-drawn wire,
respectively. Torsion delamination fracture is shown in Fig. 6a. Some cracks can be
observed near the fracture. As shown in inserted figure in Fig. 6a, a spiral crack twined
around the wire. It is generally accepted that the spiral crack was the main crack leading to
torsion delamination[28, 36]. By contrast, the fracture surface of as-drawn wires after torsion
tests is like a mirror (Fig. 6b), which is considered to be a cleavage one [28, 36, 51]. There
was no secondary cracks identified in the as-drawn wires after torsion fracture in this study
as well as previous reports [27, 36]. Fig. 6c shows secondary cracks in annealed wires after
torsion fracture. Local magnification of the microstructure near the crack is also presented in
Fig. 6d. Pearlite lamellar has deviated from the drawing direction owing to torsion.
Moreover lots of cementite fragments were near the secondary crack which is different to
Masaki Tanaka’s observation[37]. At the same time, some microstructure nearby was still
pearlite lamellar structure which suggests torsion deformation of the annealed wire was
inhomogeneous. Fig. 6e shows a propagated crack in annealed wires. Such cracks could
potentially lead to delamination failure in annealing wires.

Fig.6 a) Torsion delamination in fractured wires; b) Fracture of as-drawn wires after

torsion tests; c) Crack in wires experiencing torsion delamination; d) Microstructure near the
crack; e) Propagated crack in wires subjected to torsion delamination failure

3.4. Microstructure after torsion

Microstructure of annealed pearlite wire after torsion fracture are examined and
presented in Fig. 7. The lamellar structure of drawn pearlite has lost its directionality along
the drawing direction as presented in Fig. 7a. It is similar to previous reports[27, 36].
Cementite fragments were near the micro-crack as presented in Fig.7b. It indicates that
partial cementite plates composed of nano-crystal cementite formed in annealing were
broken into small pieces by torsion deformation. At the same time, the inhomogeneous
deformation microstructure is worth paying attention to. Microstructure about 2um away
from the crack was still pearlite lamellar structure.

Fig. 7 SEM micrographs of annealed pearlitic steel wires after torsion fracture
a) Microstructure after torsion; b) Enlarged view of framed area in a)

Fig. 8 shows the microstructure of as-drawn wires after torsion fracture as a

comparison. Direction of pearlitic lamellae apparently deviated from the drawing direction,
as with the annealed wires after torsion fracture. However, there was no obvious fracture or
fragmentation of cementite, which is different from the annealed wires. Micro-cracks were
also not observed in the samples.

Fig. 8 SEM micrographs of cold drawn pearlitic steel wires after torsion fracture

4. Discussion
4.1 Torsion delamination and recrystallized cementite
 It is widely accepted that crack initiation at small torsion angles is considered to be the
direct reason for torsion delamination of annealed pearlitic steel wires.[28, 29, 31, 36, 37, 40]
However, the mechanism of crack initiation has been hidden under the complex
experimental phenomena. The ductility would lose due to decrease of mobile dislocation
density after low-temperature annealing. In this study, torsion deterioration occurred even
the pearlitic wire was soften conspicuously after annealing. Yang et al. considered that
undecomposed cementite caused concentration of stress and cracks or voids could be easily
formed near the cementite [52]. In this experiment, cementite, in annealed heavily drawn
wires, has been transformed into connected recrystallized cementite (as presented in Fig. 3).
Fig. 9 shows TEM micrographs of annealed pearlitic steel wires after the torsion
deformation. The microstructure of annealed wires after torsion yield is presented in Fig. 9a.
A number of high density dislocation cells were separated by cementite. Dark field image
indicates that these dislocations were blocked by crystalline cementite. Meanwhile,
cementite lamellae were bent due to stress concentration caused by these dislocations. It is
also worth pointing out that some cementite layers were even broken apart near the
high-density dislocation region. Local stress concentrations caused by pinned dislocations
have been reported to be responsible for damage observed in the lamellar structure[26, 41,
53]. Fig. 9c shows the microstructure of annealed wires after torsion fracture. More broken
cementite platelets could be observed. High density dislocations were distributed around this
broken cementite. The disappearance of cementite [37] has not been observed. It may be due
to different sampling locations in the wires. On the other hand, microstructure presented in
Fig. 9c is inhomogeneous. Away from the local severe deformation region, pearlite lamellar
structure seem not to be influenced by torsion. Therefore, recrystallized cementite plates
served as barrier against dislocation motions in ferrite, and as a result local severe
deformation of pearlite lamellae occurred during torsion. Supported by SEM observations as
presented in Fig. 6 and Fig. 7, these severe deformation regions containing fractured
cementite and high density dislocations might be responsible for the occurrence of cracks.
 Fig. 9 TEM micrographs in pearlitic steel wires after torsion fracture

A sketch illustrating the torsion delamination of heavily drawn pearlitic steel wires
after low-temperature annealing is provided in Fig. 10. All discussions should be based on
the fiber lamellar structure of heavy drawn pearlite wire which is parallel to the drawing
direction. The deformation of pearlite structure appears to be controlled by ferrite dislocation
motion and interactions between ferrite dislocations and cementite at the interface [4, 19, 24,
26]. The change of torsion performance and tensile properties is closely related to aligned
lamellar structure in heavily cold drawn pearlite wires. Torsion (shear) stress is distributed
along the circumferential direction, and is completely different from the tensile stress along
the axial direction of wires. This difference also leads to the independence between the
torsion performance and tensile properties of pearlitic steel wires. Recrystallized cementite
in annealed wire impedes dislocation like normal hard second phase. It is generally believed
that cementite have poor plastic deformation ability due to very limited dislocation slip
system [54-56]. Meanwhile the elongated cementite lamellae become a barrier that
dislocation in ferrite couldn’t bow out. Under this condition, dislocation cells formed in the
ferrite and cementite was fractured due to stress buildup as presented in Fig. 9. Stress
concentration at the fracture tip of cementite might eventually lead to the initiation of cracks
[41, 52, 53]. Observed fractured cementite in torsion annealed wire and near the torsion
micro-cracks presented in Fig.6 and 7 are good proof for this mechanism.

Fig. 10 Model of torsion delamination

4.2 Torsion performance of cold-drawn pearlite wire

It is worth noting that the torsion performance of as-drawn pearlitic steel wires is
excellent and torsion delamination does not occur.
Fig. 11 shows TEM micrographs of as-drawn pearlitic steel wires after torsion test. A
high density dislocation cluster could be observed to stretch through several ferrite lamellae
in the yielding stage as presented in Fig. 11a. Dark field TEM image of ferrite (110)
diffraction spot, presented in Fig. 11b, indicates that these dislocations sweeping through a
number of ferrite platelets can be assigned to the same slip system. After torsion fracture, the
lamellar structure in as-drawn wire remained and high dislocation zones spread across
several lamellae, in a way similar to the initial torsion yielding (as seen in Fig. 11a).
The microstructure analysis of torsion samples indicates that lamellar ferrite and
cementite in as-drawn wires deformed in a coordinated manner. Dislocations in ferrite piled
up along the boundaries with cementite led to large residual stress after heavy drawing.
These residual stresses might contribute to co-deformation of the two phase[6, 57]. However,
the beginning of the plastic deformation of cold drawn wires would be affected by the
residual stress [58]. The influence of cementite structure is worth paying attention to.
Heavily deformed cementite in drawn wires represents an extraordinary structure, despite the
exact structure of the cementite is still under dispute. If the cementite is considered to be a
transition layer between ferrite lamellar with very high carbon content [2, 7, 17, 21, 52], its
mechanical properties would be similar to ferrite, and the co-deformation was
understandable. On the other hand, supposing that the cementite with a thickness below 5nm
was Fe-C mixed amorphous structure, amorphous below than 100nm have been reported to
have plastic deformation capability[59]. A recent report pointed out that amorphous interface
between crystalline grain can diffuse the strain concentration brought by dislocation
absorption and suppress fracture [60]. Cementite, which accurate structure, deformation
mechanism and influence upon mechanical properties of pearlitic steel wire are still
controversial, is worthy of further study.
 Fig. 11 TEM micrographs of cold drawn pearlitic steel wires
a) Cold drawn wires in post-yield stage; b) Dark-Field image of a location in b); c)
after torsion fracture

Based on above experimental results and analysis, it seems torsion delamination of

heavily drawn pearlite wires would be inevitable after low-temperature annealing if
cementite recrystallized and remained lamellar structure. To avoid it, 1) the spheroidization
of cementite may be considered. However, such a treatment may produce a negative effect
on the strength of heavily drawn wire [31, 32, 42, 43]. 2) Joung et al. have reported torsion
delamination would not occur as the pearlitic steel wire having a smaller drawing strain of
1.13[31]. However, the strength of the steel wire would be limited by the smaller drawing
strain. Finding a solution to achieving a combination of high strength and torsion ductility in
heavily drawn pearlite wires remains an interesting and challenging task.

5. Conclusion
The relationship of torsion delamination and recrystallized cementite in annealed
pearlitic steel wires has been examined, with a particular focus on facture mode, micro-crack
and microstructure in torsion tested wires. The torsion delamination mechanisms of annealed
pearlitic steel wires were explored. The conclusions can be drawn as follows:
1. After low temperature annealing, the overall microstructure of cold drawn wires did
not change, but the structure of cementite became different. The crystal structure of
cementite in as-drawn wires was found to be disordered and amorphous-like. In annealed
wires cementite lamellae are composed of recrystallized interconnected nano-sized cementite
particles.
2. The torsion performance of cold drawn steel wires prepared from the drawing stain of
2.0 was excellent. Torsion delamination occurred inevitably when the wires underwent
annealing at temperatures ranging from 210℃ to 350℃ with duration from 2 to 60 min. The
maximum torsion angle of uniform torsion deformation of annealed wires was only about
1/10 of the as-drawn ones. In addition, the decoupling of tensile properties and torsion
performance was observed.
3. Cementite lamellae in annealed wires are composed of recrystallized cementite, which
impeded ferrite dislocation motion considerably. As such, dislocation cells formed in the
ferrite lamellae. Cementite lamellae were found to break up near high-density dislocation
regions. Micro crack initiation was observed near broken cementite. By contrast, cementite
and ferrite in the as-drawn wires were deformed in a coordinated manner during torsion
tests.

Acknowledgement
This work is supported by the Natural Science Foundation of China (grant no.
51371050), the Science and Technology Advancement Program of Jiangsu Province (grant
no. BA2014088) and the Key Research Project of Jiangsu Province (grant no. BE2015097)
and the Industry-University Strategic Research Fund of Jiangsu Province (BY2016076-08),
China and the study is also partly supported by the Six-Talent-Peaks program of Jiangsu
Province (2015-XCL-004) and Key Laboratory for Advanced Metallic Materials of Jiangsu
Province (BM2007204). Dr. Z. Xie acknowledges the support provided by the Australian
Research Council Discovery Projects. L. C. Zhou acknowledges the support provided by
Fundamental Research Funds for the Central Universities and the Scientific Research
Foundation of the Graduate School of Southeast University (YBJJ1674). Authors thank C.
W. Jin for assistance in TEM analysis.

References
[1] S. Goto, R. Kirchheim, T. Al-Kassab, C. Borchers, Transactions of Nonferrous Metals Society of China, 17
(2007) 1129-1138.
[2] D. Raabe, P.-P. Choi, Y. Li, A. Kostka, X. Sauvage, F. Lecouturier, K. Hono, R. Kirchheim, R. Pippan, D.
Embury, Mrs Bulletin, 35 (2010) 982-991.
[3] X. Zhang, A. Godfrey, X. Huang, N. Hansen, Q. Liu, Acta Materialia, 59 (2011) 3422-3430.
[4] X. Zhang, A. Godfrey, W. Liu, Q. Liu, Materials Science and Technology, 27 (2011) 562-567.
[5] X. Zhang, A. Godfrey, N. Hansen, X. Huang, Acta Materialia, 61 (2013) 4898-4909.
[6] J. Toribio, B. González, J.-C. Matos, Materials Transactions, 55 (2014) 93-98.
[7] C. Borchers, R. Kirchheim, Progress in Materials Science, 82 (2016) 405-444.
[8] J. Embury, R. Fisher, Acta Metallurgica, 14 (1966) 147-159.
[9] G. Langford, Metallurgical and Materials Transactions B, 1 (1970) 465-477.
[10] Y. Li, A. Kostka, P. Choi, S. Goto, D. Ponge, R. Kirchheim, D. Raabe, Acta Materialia, 84 (2015) 110-123.
[11] S. Djaziri, Y. Li, G. Nematollahi, B. Grabowski, S. Goto, C. Kirchlechner, A. Kostka, S. Doyle, J.
Neugebauer, D. Raabe, Advanced Materials, 28 (2016) 7753-7757.
[12] B.N. Jaya, S. Goto, G. Richter, C. Kirchlechner, G. Dehm, Materials Science and Engineering: A, 707
(2017) 164-171.
[13] K. Han, G. Smith, D. Edmonds, Metallurgical and Materials Transactions A, 26 (1995) 1617-1631.
[14] K. Han, D. Edmonds, G. Smith, Metallurgical and Materials Transactions A, 32 (2001) 1313-1324.
[15] E.M. Taleff, J.J. Lewandowski, B. Pourladian, JOM Journal of the Minerals, Metals and Materials Society,
54 (2002) 25-30.
[16] G.J. Shiflet, M.A. Mangan, W.G. Meng, Interface Science, 6 (1998) 133-154.
[17] Y. Li, D. Raabe, M. Herbig, P.-P. Choi, S. Goto, A. Kostka, H. Yarita, C. Borchers, R. Kirchheim, Physical
review letters, 113 (2014) 106104.
[18] W.J. Nam, C.M. Bae, S.J. Oh, S.-J. Kwon, Scripta Materialia, 42 (2000) 457-463.
[19] Y. Ivanisenko, W. Lojkowski, R. Valiev, H.-J. Fecht, Acta Materialia, 51 (2003) 5555-5570.
[20] N. Maruyama, T. Tarui, H. Tashiro, Scripta materialia, 46 (2002) 599-603.
[21] K. Hono, M. Ohnuma, M. Murayama, S. Nishida, A. Yoshie, T. Takahashi, Scripta materialia, 44 (2001)
977-983.
[22] C. Borchers, T. Al-Kassab, S. Goto, R. Kirchheim, Materials Science and Engineering: A, 502 (2009)
131-138.
[23] L. Zhou, F. Fang, X. Zhou, Y. Tu, Z. Xie, J. Jiang, Scripta Materialia, 120 (2016) 5-8.
[24] Y. Li, P. Choi, C. Borchers, S. Westerkamp, S. Goto, D. Raabe, R. Kirchheim, Acta Materialia, 59 (2011)
3965-3977.
[25] Y. Chen, G. Csiszár, J. Cizek, X. Shi, C. Borchers, Y. Li, F. Liu, R. Kirchheim, Metallurgical and Materials
Transactions A, 47 (2016) 726-738.
[26] T.-Z. Zhao, S.-H. Zhang, G.-L. Zhang, H.-W. Song, M. Cheng, Materials & Design, 59 (2014) 397-405.
[27] N. Guo, B. Luan, Q. Liu, Materials & Design, 50 (2013) 285-292.
[28] M. Zelin, Acta Materialia, 50 (2002) 4431-4447.
[29] C.M. Bae, W.L. Nam, C.S. Lee, Scripta materialia, 35 (1996) 641-646.
[30] K. Shimizu, N. Kawabe, ISIJ international, 41 (2001) 183-191.
[31] S. Joung, U. Kang, S. Hong, Y. Kim, W. Nam, Materials Science and Engineering: A, 586 (2013) 171-177.
[32] Y. Li, P. Choi, S. Goto, C. Borchers, D. Raabe, R. Kirchheim, Acta Materialia, 60 (2012) 4005-4016.
[33] J. Takahashi, M. Kosaka, K. Kawakami, T. Tarui, Acta Materialia, 60 (2012) 387-395.
[34] R.Z. Valiev, R.K. Islamgaliev, I.V. Alexandrov, Progress in materials science, 45 (2000) 103-189.
[35] J. Languillaume, G. Kapelski, B. Baudelet, Acta Materialia, 45 (1997) 1201-1212.
[36] X. Hu, L. Wang, F. Fang, Z. Ma, Z.-h. Xie, J. Jiang, Journal of Materials Science, 48 (2013) 5528-5535.
[37] M. Tanaka, H. Saito, M. Yasumaru, K. Higashida, Scripta Materialia, 112 (2016) 32-36.
[38] H.M. Baek, S.K. Hwang, H.S. Joo, Y.-T. Im, I.-H. Son, C.M. Bae, Materials & Design (1980-2015), 62
(2014) 137-148.
[39] F. Yang, C. Ma, J. Jiang, H. Feng, S. Zhai, Scripta Materialia, 59 (2008) 850-853.
[40] N. Guo, B. Song, B.-S. Wang, Q. Liu, Acta Metallurgica Sinica (English Letters), 28 (2015) 707-714.
[41] F. Fang, L. Zhou, X. Hu, X. Zhou, Y. Tu, Z. Xie, J. Jiang, Materials & Design, 79 (2015) 60-67.
[42] F. Fang, X.-j. Hu, S.-h. Chen, Z.-h. Xie, J.-q. Jiang, Materials Science and Engineering: A, 547 (2012)
51-54.
[43] D. Park, E. Kang, W. Nam, Journal of materials processing technology, 187 (2007) 178-181.
[44] F. Fang, J.-h. Jiang, S.-Y. Tan, A.-b. Ma, J.-q. Jiang, Surface and Coatings Technology, 204 (2010)
2381-2385.
[45] C. Borchers, Y. Chen, M. Deutges, S. Goto, R. Kirchheim, Philosophical Magazine Letters, 90 (2010)
581-588.
[46] N. Min, W. Li, X. Jin, X. Wang, T. Yang, C. Zhang, ACTA METALLURGICA SINICA-CHINESE
EDITION-, 42 (2006) 1009.
[47] F. Fang, Y. Zhao, P. Liu, L. Zhou, X.-j. Hu, X. Zhou, Z.-h. Xie, Materials Science and Engineering: A, 608
(2014) 11-15.
[48] J. Languillaume, G. Kapelski, B. Baudelet, Materials Letters, 33 (1997) 241-245.
[49] X. Huang, N. Hansen, N. Tsuji, Science, 312 (2006) 249-251.
[50] Y. He, S. Xiang, W. Shi, J. Liu, X. Ji, W. Yu, Materials Science and Engineering: A, 683 (2017) 153-163.
[51] A. Durgaprasad, S. Giri, S. Lenka, S. Kundu, S. Mishra, S. Chandra, R.D. Doherty, I. Samajdar,
Metallurgical & Materials Transactions A, (2017) 1-15.
[52] Y. Yang, J. Bae, C. Park, Materials Science and Engineering: A, 508 (2009) 148-155.
[53] Z. Lichu, H. Xianjun, M. Chi, Z. Xuefeng, F.F. Jianqing JIANG, Acta Metall Sin, 51 (2015) 897-903.
[54] A. Inoue, T. Ogura, T. Masumoto, Scripta Metallurgica, 11 (1977) 1-5.
[55] C. Jiang, S.G. Srinivasan, Nature, 496 (2013) 339-342.
[56] J. Gurland, Acta Metallurgica, 20 (1972) 735-741.
[57] J. Toribio, E. Ovejero, Mechanics of Time-Dependent Materials, 1 (1997) 307-319.
[58] M. Elices, Journal of Materials Science, 39 (2004) 3889-3899.
[59] D. Jang, J.R. Greer, Nature materials, 9 (2010) 215-219.
[60] A. Khalajhedayati, Z. Pan, T.J. Rupert, Nature communications, 7 (2016).