Z. H. Cai -Microstructural evolution and mechanical properties of hot-rolled 11% manganese TRIP steel.pdf

(2805 KB) Pobierz
Materials Science & Engineering A 560 (2013) 388–395
Contents lists available at
SciVerse ScienceDirect
Materials Science & Engineering A
journal homepage:
www.elsevier.com/locate/msea
Microstructural evolution and mechanical properties of hot-rolled 11%
manganese TRIP steel
Z.H. Cai, H. Ding
n
, X. Xue, Q.B. Xin
School of Materials and Metallurgy, Northeastern University, 110819 Shenyang, China
a r t i c l e i n f o
Article history:
Received 4 July 2012
Received in revised form
26 August 2012
Accepted 23 September 2012
Available online 29 September 2012
Keywords:
Hot-rolled
Heat treatment
TRIP effect
Microstructure
Mechanical stability
a b s t r a c t
Mechanical behaviors of transformation induced plasticity (TRIP) steels largely depend on the amount
and stability of austenite. In this investigation, a large volume fraction of austenite (4 65%) was
produced in a hot-rolled Fe-11Mn-3.8Al-0.18C TRIP steel by solution treatment in the temperature
range of 750–800
1C
for 1 h. The hot-rolled alloy exhibited an excellent combination of total elongation
of 35–40% and ultimate tensile strength of 880–1100 MPa and this was found to have a similar or
higher level of tensile properties compared with other TRIP steels. In the meantime, less cold-rolling
work or annealing time was required in the present work. The outstanding properties of the
experimental steel were mainly attributed to the enhanced TRIP effect due to the large fraction of
austenite. It is shown that the morphology played a more signicant role than orientation in the
stability of austenite.
&
2012 Elsevier B.V. All rights reserved.
1. Introduction
The demands for energy conservation and environmental pro-
tection have compelled the automakers to design light-weight auto
bodies with enhanced crash resistance. Transformation induced
plasticity (TRIP) steels which have attractive combinations of
strength and ductility have been regarded as the potential candi-
dates for automotive application. There are basically two types of
lightweight steels: austenite-base and ferrite-base steels. Typical
Mn and C contents in the low alloy TRIP steels are
$
1.5 wt% and
$0.2
wt%, respectively, which are required to achieve the austenite
stability
[1].
The low alloy TRIP steels referred to as the ‘‘rst
generation’’ advanced high strength steels (AHSS), comprise ferrite
matrix (55–65%) along with bainite (25–35%) and metastable
retained austenite (5–20%)
[2].
These steels demonstrate advanta-
geous balance of strength and ductility, but commercial TRIP
grades are limited mostly by tensile strength of 800 MPa
[3].
A variety of studies
[4–6]
have shown that fully austenitic steels
with high Mn (418%) content, the representative ‘‘second genera-
tion’’ AHSS, with an excellent combination of strength and ductility
have the capability to exhibit twin induced plasticity (TWIP) with
or without the assistance of TRIP effect. However, the steels are
expensive as a result of high alloy levels and have received limited
use in the commercial application.
n
Corresponding author. Tel.:
þ8613898876262
E-mail address:
dingneu@163.com (H. Ding).
The ‘‘third generation’’ AHSS ought to be less alloyed than TWIP
steels, exhibiting intermediate properties between the rst and
second generation steels. Some publications
[3,7,12]
consider steels
with medium Mn (4–10%) as the potential candidates to achieve the
performance of the 3rd generation AHSS. Some of publications
[7,8]
on medium Mn containing steels proposed that the stability of
austenite can be attributed to the signicant partitioning of Mn
between ferrite and austenite during long (hours) holding time in the
intercritical region. Miller reported austenite fraction up to 40% in
6.0% Mn alloys
[9].
Similar level of austenite was obtained by Merwin
[10]
in a 7.1% Mn alloy; and in an 8.0% Mn alloy, austenite fraction up
to 48% was reported by Kim et al.
[11].
Therefore, Mn is an important
element to enhance the volume fraction of retained austenite.
Conventionally, TRIP steels contain about 1.5 wt% silicon which
prevents the formation of cementite during the isothermal bainitic
transformation
[13].
Nevertheless, it was reported that silicon in a
concentration higher than 0.5 wt% is detrimental to surface quality
[14].
High silicon leads to poor weld ability as well
[15].
To avoid
these problems silicon can be partially replaced by aluminum
[16–18].
Aluminum, like silicon, has the same function in suppressing
cementite formation. Furthermore, Al facilitates coating ability by
forming an inhibition layer on the steel surface
[19].
In view of above
analyses, an attempt has been made to develop a Fe-11.02Mn-3.81Al-
0.18C austenite matrix TRIP steel. Mn and Al contents of the
experimental steel have been judiciously balanced to achieve ade-
quate hardenability and strengthen the austenite phase.
The present work focuses on the microstructure evolution
resulting from the heat treatment and TRIP effect. The stability of
austenite was discussed as well.
0921-5093/$ - see front matter
&
2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.msea.2012.09.083
Z.H. Cai et al. / Materials Science & Engineering A 560 (2013) 388–395
389
2. Experimental
A 40 kg experimental steel cast ingot was manufactured using
a vacuum furnace. The cast ingot was prepared at 1200
1C
for 2 h,
hot forged between 1150
1C
and 850
1C
to 100 mm
Â
30 mm bars,
and then air cooled to room temperature. Subsequently, the bars
were soaked at 1200
1C
for 2 h, and then hot rolled down to 4 mm
in thickness after eight passes within the temperature range of
1150–850
1C.
Finally, the as-hot-rolled sheets were air cooled to
ambient temperature.
A model proposed by Moor et al.
[20]
was used to predict the
amount of austenite stabilized to room temperature through the
enrichment of austenite with Mn, Al and C. It was further deduced
that there existed a temperature in the intercritical region
resulting in the maximum austenite retention at room tempera-
ture. Consequently, the model was instructive to the optimization
of alloy composition design for the experimental steel. For the
purpose of intercritical temperature range, the phase transforma-
tions were studied by means of dilatometry. The sample used for
dilatometry was solid cylindrical specimen with a diameter of
3 mm and a length of 10 mm.
In conventional TRIP steels, a two-stage heat treatment is
adopted
[21].
They are rstly subjected to intercritical annealing
(800–900
1C)
to form a mixture of ferrite and austenite, and then
followed by isothermal annealing (350–450
1C)
to stabilize aus-
tenite. However, the heat treatment was proven to be not
applicable to the experimental steel. Instead, a more convenient
two-stage heat treatment was applied to the experimental steel.
(1) solution treatment. The as-hot rolled sheets were soaked in a
high temperature furnace at the temperature of 700, 750, 800,
850, 900
1C,
corresponding to the intercritical region, for 1 h,
respectively, and then quenched in the water immediately.
(2) Tempering. The quenched samples were tempered at a
temperature of 200
1C
for 20 min in order to relieve internal
stress, and thus the ductility could be improved, the samples were
then air cooled to ambient temperature. The attribution of
tempering to the excellent mechanical properties was discussed
in the later section.
Flat tensile specimens with a width of 12.5 mm and a gauge
length of 50 mm were machined from the tempered and untem-
pered sheets with the tensile axis parallel to the prior rolling
direction. Tensile tests were carried out at room temperature using
a universal testing machine (SANSCMT5000) at a constant cross-
head speed of 3 mm min
À
1
. The samples were etched with 25%
sodium bisulte aqueous solution. Microstructures were analyzed
by optical microscope (OM), scanning electron microscope(SEM)
equipped with electron backscatter diffraction (EBSD), and trans-
mission electron microscope (TEM). The volume fraction of auste-
nite was determined by X-ray diffraction (XRD) with CuK
a
radiation using the direct comparison method
[22].
This method
used the integrated intensities of the (200)
a
and (211)
a
peaks and
those of the (220)
g
and (311)
g
peaks. The volume fraction of the
austenite V
A
was calculated as following:
[23]:
V
A
¼
1:4I
g
I
a
þ1:4I
g
where:
I
g
—integrated
intensity of austenite’s diffraction lines,
I
a
—integrated
intensity of
a
-phase’s diffraction lines.
3. Results and discussion
3.1. Composition design and intercritical temperature
The phase fractions based on an equilibrium thermodynamic
analysis
[20]
predicted by Thermo-Calc are shown in
Fig. 1(a),
and
these predicted equilibrium austenite fractions were used as
input to the model. Predictions of austenite composition were
also made using Thermo-Calc, and the predicted C, Mn, and Al
contents in austenite are shown in
Fig. 1(b).
The resulting
predicted fraction of stabilized austenite as a function of
Fig. 1.
Schematic illustration of predictive model development for austenite stabilization as a function of temperature. (a) Phase fractions, (b) C, Al and Mn content of
austenite in experimental steel, and (c) calculated retained austenite fractions.
390
Z.H. Cai et al. / Materials Science & Engineering A 560 (2013) 388–395
annealing temperature is shown in
Fig. 1(c).
A pronounced peak is
observed at approximately 710
1C,
resulting in the maximum
austenite fraction (
$
66%) retained at room temperature.
The intercritical region obtained by dilatometry was instruc-
tive to establish appropriate heat treatment schedules.
The expansion and contraction of the sample were measured as
the ratio of length change (dl) to initial length (l
0
). As shown in
Fig. 2,
after thermal expansion in the heating stage (20–1200
1C)
at the rate of 20
1C/s,
the sample was held at 1200
1C
for 3 min.
It is clear that no transformation took place during the fast
cooling, at the rate of 100
1C/s,
to the isothermal bainitic trans-
formation (IBT), and the contraction was continuing until mar-
tensite start (Ms) temperature. The temperature range between
Ac
1
(650
1C)
and Ac
3
(990
1C)
was marked, which corresponds to
the ferrite-to-austenite transformation at the intercritical anneal-
ing temperature.
3.2. Microstructure and mechanical properties
Fig. 3
shows the optical microstructures of the hot-rolled and
heat-treated samples. The microstructural constituents in the
samples are primarily martensite (M), retained austenite
(A) and a small amount of ferrite (F), and the amount of each
phase varies signicantly with different heat treatment schedules.
Fig. 3(a)
shows the microstructure of hot-rolled sample. It con-
sists of stripe-like ferrite, blocky austenite, and lath martensite.
Fig. 3(b)
and (c) shows the microstructures solution-treated at
750
1C
and 800
1C,
respectively, ne martensite laths grouping
themselves into several packets disperse in the blocky parent
austenite matrix.
Fig. 3(d)
and (e) shows the microstructures
solution-treated at 850
1C
and 900
1C,
respectively. It is clear that
austenite decreased distinctly owing to the extensive martensitic
transformation, and martensite packets introduced very signi-
cant morphology changes, getting thicker and denser with
elevated temperature.
Tensile testing revealed that the experimental steel had an
excellent combination of tensile strength and elongation under
Fig. 2.
Dilatometric curve showing intercritical temperature range.
Fig. 3.
Optical microstructure of the samples (a) As-hot-rolled and solution-treated at (b) 750
1C,
(c) 800
1C,
(d) 850
1C
and (e) 900
1C.
Fig. 4.
Comparisons of tensile properties against tempered and untempered samples as a function of temperature (a) UTS and TEL and (b) UTS
Â
TEL.
Z.H. Cai et al. / Materials Science & Engineering A 560 (2013) 388–395
391
the heat treatment (in the range of 750–800
1C)
discussed above.
The mechanical properties and the effective inuence of temper-
ing are illustrated in
Fig. 4.
Compared with the heat-treated
specimens in the range of 700–800
1C,
the initial hot-rolled
sample demonstrated ultrahigh tensile strength of 1462 MPa
and very low tensile elongation of 9.0%, while it had a similar
level of tensile properties with the samples after being solution-
treated in the range of 850–900
1C.
It can be inferred from
Fig. 5(b)
that the sizeable gap in the amount of retained austenite
led to the great difference in mechanical properties of the steel.
Fig. 4a
shows that ultimate tensile strength (UTS) increases
continuously with increasing temperature up to 850
1C,
whereas
total elongation (TEL) almost decreases with increasing tempera-
ture. Compared with
Fig. 4(a)
and (b), we nd that the TEL had a
similar trend with volume fraction of austenite. Therefore, it is
inferred that the variation of UTS and TEL were well correlated
with two main aspects: on one hand, the proportion of martensite
and retained austenite played an important role in the mechan-
ical properties. It was reported that the tensile strength was
dependent on the volume fraction of martensite
[24].
As shown in
Fig. 3,
it is obvious that with increasing temperature the amount
of retained austenite decreased while that of martensite
increased, which was further evidenced by XRD analysis (Fig.
5),
resulting in high strength and low ductility. On the other hand, as
the solution temperature increases the size of martensite packets
increased due to the decrease of the carbon content and a coarser
morphology of austenite
[25].
It is obvious that the size of
martensite packets (marked in
Fig. 3)
increased with increasing
temperature, and they subdivided the blocky austenite into
granular and lm shapes more extensively in the temperature
range of 850–900
1C.
It is well known that the retained austenite
of lm and granular shapes is much more stable than the blocky
shape
[26].
Therefore, TRIP effect hardly occurred in the samples
solution-treated at 850–900
1C,
leading to low ductility. Further-
more, it is obvious that as for TEL, the tempered samples had a
clear advantage, especially at elevated temperatures. It is
generally considered that such an improvement on TEL is related
to the decrease in internal stress and the brittleness of the sample
caused by the low temperature tempering. Owing to the temper-
ing, the mechanical properties of the experimental steel were
improved signicantly, which is further conrmed in
Fig. 4(b).
As shown in
Fig. 4(b),
it is clear that the tempered samples had
better mechanical properties, especially the sample solution-
treated at 800
1C
had the optimal mechanical properties.
This sample exhibited ultimate tensile strength of 1082 MPa,
total elongation of 34.6%, and product of strength and elongation
(PSE) of 37.4 GPa%; by contrast, the low alloy TRIP steels, at their
optimal performance, demonstrated similar or lower plasticity
with much lower tensile strength (
r850
MPa)
[19,22,27,28],
and
the medium Mn TRIP steels, as shown in
Table 1,
had the same
strength level with a lower or similar plasticity by additional
cold-rolling work or prolonged annealing. As a great differences of
microstructure and mechanical properties existed among the
samples solution-treated at the given temperatures, the present
work was focused on the sample solution-treated at 800
1C
and
then tempered (nominated as Sample 1).
TRIP effect depends on the amount of retained austenite and
its stability to transformation during deformation. It is well
known that the stability of austenite depends on various factors
such as the chemical composition (primarily the carbon content)
[29–31],
grain size
[32]
and shape
[33]
of the austenite. In order
to identify the contribution of the retained austenite to the
mechanical properties of the experimental steel, the change of
the volume fraction of retained austenite as a function of solution
temperature was evaluated.
Fig. 5
shows the XRD patterns and measured austenite frac-
tions present at room temperature of the hot-rolled and solution-
treated samples. The amount of retained austenite increased to
71.9 vol pct for the 750
1C
solution-treated sample, followed by a
slight fall for the 800
1C
condition, and then decreased signi-
cantly for the 850
1C
condition to 12.5 vol pct. Comparing
Fig. 1(c)
with
Fig. 5(b),
we nd that the experimental results
Fig. 5.
XRD patterns and measured austenite fractions present at room temperature of the hot-rolled and solution-treated samples. (a) XRD patterns and (b) measured
austenite fractions.
Table 1
Comparisons of other medium Mn TRIP steels and experimental steel.
Composition
0.15C-4Mn
0.05C-6.15Mn-1.5Si
0.08C-(4.8–6.7)Mn
0.1C -7.1Mn
0.1C-7.1Mn
Experimental steel
Initial condition
CR-annealing
CR-annealing
CR- Batch annealing(40 h)
CR-annealing for 1 week
HR-Batch annealing(80 h)
HR-Solution(1 h)
UTS (MPa)
1084/1187
1014/1213
700–850
876/ 1198
1074
1082/1201
TEL (%)
19.1/15.9
22.8/12.7
22–38
42/ 12
33.6
34.6/19.8
PSE (GPa%)
20.7/18.9
23.1/15.4
18.7–26.6
36.8/14.4
36.1
37.5/23.8
Reference
3
7
12
8
10
CR-cold rolled, HR-hot rolled, UTS-ultimate tensile strength, TEL-total elongation, PSE-product of strength and elongation.
392
Z.H. Cai et al. / Materials Science & Engineering A 560 (2013) 388–395
were in excellent agreement with the theoretical prediction and
consistent with the work by Moor et al.
[20].
It is well established
that solution temperature was the main cause of developing great
variation in the amount of retained austenite.
It has been established that martensite is better developed
with increasing the Martensite starting (M
s
) point which is
affected by cooling rate. Rapid cooling is favorable for the
formation of martensite due to high internal stress. Woehrle
et al.
[34,35]
veried that increasing cooling rate resulted in a
rising of M
s
point. Therefore, it is reasonable to deduce that the
higher the solution temperature, the more the martensite pro-
duced. The conclusion is also supposed by an equation
[36]
of
martensitic transformation kinetics.
f
¼
1Àexp½
b
ðc
1
Àc
0
ÞÀ
a
ðM
s
ÀT
q
Þ
Fig. 6.
X-ray diffraction (XRD) patterns of the undeformed and fractured Sample 1.
where f is the proportion of martensite;
b
and
a
are the
coefcients of material greater than zero; T
q
is the quenching
media temperature; c
0
and c
1
are the carbon content of austenite
before and after quenching; c
1
is decided by the time for carbon-
enrichment in austenite.
Fig. 6
shows the comparison of X-ray diffraction patterns of
the undeformed and fractured Samples 1. It is implied that the
austenite had been reduced to a small amount after tensile
failure. The volume fraction of retained austenite measured by
XRD analysis was 8.7%, which indicated that signicant TRIP
effect happened in the Sample 1 during tensile deformation.
Fig. 7
shows the microstructure of the Sample 1 before and after
tensile deformation in the vicinity of the fracture position.
The initial microstructure consists of stripe-like ferrite, equiaxed
retained austenite matrix with or without lath martensite which
in some cases partition the austenite into lms (labeled M/A in
the
Fig. 7(a).
Compared with
Fig.7(a)
and (b) shows a large
amount of martensite packets after tensile deformation. As shown
in
Fig. 7(d),
from an optical microstructure perspective, white and
black martensite laths intensively subdivided the blocky austenite
into particles and lms, which was evidenced by EBSD in the later
section. By inspection of
Fig. 7(c),
it is obvious that the black
martensite laths were formed previously during solution treat-
ment, while the white laths with higher carbon content were
produced during deformation. The martensitic transformation
during quenching occurred preferentially in the austenite with
lower carbon content, and the austenite with higher carbon
content were retained. In this case, the transformation induced
by deformation occurred in the retained austenite resulting in
white martensite laths. In conclusion, carbon increases austenite
stability effectively.
Fig. 8
shows the microstructural features of the undeformed
Sample 1. It proves that the nature of the grains in the different
shapes was austenite, as evidenced by the diffraction pattern
(inset in
Fig. 8(b)).
It is apparent that the retained austenite of
granular and lm shapes are surrounded by lath martensite
which was transformed from the original blocky austenite. There-
fore, TEM observation was conducted to further conrm the
existence of austenite in different shapes. The relationship
Fig. 7.
SEM and OM micrographs of the Sample 1. (a)–(c)Prior to tensile deformation and (b)–(d)after tensile deformation.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin