Content uploaded by Junru Li
Author content
All content in this area was uploaded by Junru Li on Jan 17, 2019
Content may be subject to copyright.
Met. Mater. Int., Vol. 23, No. 2 (2017), pp. 239~245
doi: 10.1007/s12540-017-6473-7
Influence of High Temperature Pre-Deformation on the Dissolution Rate
of Delta Ferrites in Martensitic Heat-Resistant Steels
Junru Li, Jianjun Liu, Bo Jiang, Chaolei Zhang, and Yazheng Liu*
School of Materials Science and Engineering, University of Science and Technology Beijing,
Beijing 100083, China
(received date: 6 July 2016 / accepted date: 5 September 2016)
The dissolution process of delta ferrites and the influence of high temperature pre-deformation on the dissolu-
tion rate of delta ferrites in martensitic heat-resistant steel 10Cr12Ni3Mo2VN were studied by isothermal
heating and thermal simulation experiments. The precipitation temperature of delta ferrites in experimental
steel is about 1195 °C. M23C6-type carbides incline to precipitate and coarsen at the boundaries of delta ferrites
below 930 °C, and can be rapidly dissolved by heating at 1180 °C. The percentage of delta ferrites gradually
decreases with heating time. And a Kolmogorov-Johnson-Mehl-Avrami equation was established to describe
the dissolution process of delta ferrites at 1180 °C. High temperature pre-deformation can markedly increase
the dissolution rate of delta ferrites. Pre-deformation can largely increase the interface area between delta fer-
rite and matrix and thus increase the unit-time diffusing quantities of alloying elements between delta ferrites
and matrix. In addition, high temperature pre-deformation leads to dynamic recrystallization and increases the
number of internal grain boundaries in the delta ferrites. This can also greatly increase the diffusing rate of
alloying elements. In these cases, the dissolution of delta ferrites can be promoted.
Keywords: diffusion, deformation, grain boundary, recrystallization, precipitation
1. INTRODUCTION
Generally, the optimized microstructure of martensitic heat-
resistant steels after quenching and tempering is tempered
martensite and free from delta ferrites [1-3]. However, the delta
ferrites frequently incline to form in the martensitic heat-
resistant steels because of inappropriate chemical composi-
tions and non-equilibrium solidification. The delta ferrites
are the harmful phase in the martensitic heat-resistant steels,
for instance, delta ferrites can markedly increase the ductile-
brittle transition temperature [3-4]. In addition, we found that
the delta ferrites can sharply damage the isotropy of mechanical
properties. However, once the delta ferrites form, it is difficult to
remove by conventional heat-treatment except for long-time high
temperature diffusion [4-5]. But long-time high temperature
diffusion always causes burning loss and grain coarsening [6].
In a recent study in an austenitic stainless steel [6], the
authors mentioned that thermomechanical treatment can pro-
mote the dissolution of delta ferrites, while the dynamic
recrystallization will decrease the promotion of thermome-
chanical treatment. In our study, we also found that a high
temperature pre-deformation can markedly increase the dissolu-
tion rate of delta ferrites in a typical martensitic heat-resistant
steel 10Cr12Ni3Mo2VN which inclines to form delta fer-
rites. However, we found that one of the promotion mecha-
nisms of the pre-deformation on the dissolution of delta ferrites
is precisely dynamic recrystallization. This study may be help-
ful to provide a new idea to rapidly eliminate the delta fer-
rites in the martensitic heat-resistant steels.
2. EXPERIMENTAL PROCEDURE
The test samples for this study were prepared from an
ingot which was produced by electroslag remelting of the
10Cr12Ni3Mo2VN steel. The chemical composition of the
experimental steel is shown in Table 1.
In order to investigate the influence of the pre-deformation
on the dissolution rate of the delta ferrite, the samples were
divided into two groups. The first group was subjected to an
isothermal diffusion at 1180 °C for different time using a
resistance box-type furnace. The thermodynamic calculation
results by a JMatPro software show that the delta ferrites
precipitates above about 1195 °C in the experimental steel,
and it is austenite single-phase at 1180 °C, as shown in Fig. 1.
The second group was thermal simulation tested according
to the schedule given in Fig. 2 using a computer-controlled
servo-hydraulic Gleeble 1500 testing machine.
Optical microscopy (OM), field emission scanning electron
microscopy (FESEM) and transmission electron microscopy
*Corresponding author: lyzh@ustb.edu.cn
KIM and Springer
240 Junru Li et al.
(TEM) were used to investigate the microstructure of the samples.
For the OM and FESEM investigation, the specimens after
heat treatment were first ground and polished, then etched
in a solution of 5 g CuSO4, 70 mL HCl, and 100 mL H2O. In
addition, in order to obtain the proportion of delta ferrites,
the specimens after heat treatment were first ground and pol-
ished, and then electrolytic etched in a solution of 20 g NaOH
and 100 mL H2O. By the electrolytic etch, the boundaries of
delta ferrites can be clearly displayed, while the microstruc-
ture of the matrix is not displayed. Thin film specimens for
the TEM observation were made from slices cut from the spec-
imens after heat treatment, which were mechanically ground
to a thickness below 0.05 mm and finally electropolished in
a solution of 4 pct perchloric acid in alcohol at a temperature
-30 °C. X-ray energy dispersive spectroscopy (EDS) was
used to analyze the composition of the precipitates.
3. RESULTS
3.1. Chemical composition of delta ferrites and precipitates
The OM micrographs of the experimental steel are shown
in Fig. 3. There are numerous delta ferrites in the original
microstructures. The microstructure and the chemical com-
position of the delta ferrites in the experimental steel were
analyzed before heat treatment, the result has been shown in
Table 2. The alloying elements C and N were ignored due to
that C, N is difficult to analyze precisely by EDS. The con-
Fig. 1. The thermodynamic calculation results by JMatPro software. (b): The magnifying image of (a).
Table 1. Chemical composition of 10Cr12Ni3Mo2VN steel (in mass%)
CMnSi P S CrNiMoV N Fe
0.09 0.82 0.20 0.011 0.002 11.82 2.60 1.75 0.34 0.033 Bal.
Fig. 2. The thermal simulation test schedule.
Fig. 3. The OM (a) and FESEM (b) micrographs of the experimental
steel.
Tabl e 2 . Chemical composition of delta ferrites in the experimental
steel (in mass%)
Mn Si Cr Ni Mo V Fe
0.80 0.22 10.88 1.42 2.20 0.54 Bal.
Influence of High Temperature Pre-Deformation on the Dissolution Rate of Delta Ferrites in Martensitic Heat-Resistant Steels 241
tents of Mn and Si in the delta ferrites are similar to the
average composition (shown in Table 1). The contents of Cr
and Ni in the delta ferrites are lower than the average com-
position, and the contents of Mo and V are higher.
In addition, there are numerous precipitates at the bound-
aries of delta ferrite, as shown in Fig. 3(b). And the size and
number of these precipitates at the delta ferrites boundaries
are much bigger than in the matrix. In order to determine the
type of these precipitates, the diffraction pattern and the chemi-
cal composition of these precipitates were analyzed. The EDS
analysis result shows that these precipitates are Cr, Mo, V
and Fe rich, as shown in Table 3. The carbon content of these
precipitates is ignored. The TEM micrograph and the diffraction
pattern of the precipitate are given in Fig. 4. According to
the diffraction pattern and the chemical composition, these
precipitates can be determined as M23C6-type carbides.
These carbides can be eliminated by heating for 30 min at
1180 °C, as shown in Fig. 5. According to the thermodynamic
calculation results by the JMatPro software, M23C6-type car-
bides precipitate below 930 °C, as shown in Fig. 1. The compo-
sition segregation of delta ferrites may be the reason why
M23C6-type carbides incline to form at the delta ferrite boundar-
ies. In the other hand, the delta ferrites dissolved very little after
heating for 30 min at 1180 °C, as shown in Fig. 5. However, the
chemical composition of delta ferrites changed a lot com-
pared to that before heating. The EDS analysis result shows
that the contents of Cr, Mo, V of delta ferrites are higher after
heating for 30 min at 1180 °C than before heating, as show
in Table 4. Especially the content of Cr, it is lower than the
average chemical composition before heating, while it becomes
higher than average chemical composition after heating. That
indicates that perhaps most of ferrite former elements Cr,
Mo and V diffuse to delta ferrites after M23C6-type carbides
dissolved.
According to the thermodynamic calculation results by
JMatPro software, there are few M23C6-type carbides above
930 °C. While it is easier for M23C6-type carbides to nucleate
and grow at the delta ferrite boundaries below 930 °C due to
that delta ferrites always contain more ferrite-former alloy-
ing elements Cr, Mo and V than matrix, which are precisely
the main elements of M23C6-type carbides. In addition, the
cooling rate of the ingot is always very low due to its large
size, which makes it possible for M23C6-type carbides to have
sufficient time to precipitate and grow. That is why there are
numerous M23C6-type carbides at delta ferrite boundaries in
the annealing ingot.
3.2. The high temperature dissolution of delta ferrite
The OM micrograph of the first group samples after heat-
ing for different time at 1180 °C are shown in Fig. 6. The
samples were first ground and polished, and then electro-
lytic etched. The boundaries between delta ferrites and matrix
can be clearly observed. The matrix is martensite. There are
numerous delta ferrites (about 8.8% in volume) after heat-
ing for 30 min, as the arrow pointing in Fig. 6(a). The per-
centage of delta ferrites gradually decreases with heating
time, as shown in Fig. 7. But there are still numerous delta
ferrites after heating for a long time. For instance, the per-
centage of delta ferrites is still higher than 2.5% after heat-
ing for 480 min at 1180 °C, as shown in Fig. 6(c).
According to the Kolmogorov-Johnson-Mehl-Avrami equa-
tion (KJMA equation)[7], the dissolution process of delta
ferrites can be expressed as the Eq. 1.
δt=δ0exp[-(kt)n] (1)
where δ0 is the volume percentage of delta ferrites after
heating for t0, δt is the volume percentage of delta ferrites
after heating for t0+t, t is time, and k, n are connected with
steel grade and heat treat temperature. The k and n can be
recognized as constants when the heat treat temperature and
steel grade are fixed.
Tabl e 3 . Chemical composition of precipitates at the boundaries of
delta ferrites (in mass%)
Mn Si Cr Ni Mo V Fe
0.05 0.08 31.64 0.70 5.52 1.33 Bal.
Fig. 4. The TEM micrograph and the diffraction pattern of the precip-
itates at the boundaries of delta ferrites.
Fig. 5. The OM (a) and SEM (b) micrographs of the experimental
steel after heating for 30 min at 1180 °C.
Tabl e 4 . Chemical composition of delta ferrites after heating for 30
min at 1180 °C (in mass%)
Mn Si Cr Ni Mo V Fe
0.80 0.21 12.51 1.42 2.44 0.61 Bal.
242 Junru Li et al.
The relationship between the logarithm of the phase trans-
formation rate log(-ln(δt/δ0)) and the logarithm of time log(t) is
shown in Fig. 8(a). The correlation is pretty good. According
to Fig. 8(a), the values of the constants n and k can be obtained.
The values of the constants n and k are 0.872 and 5.59 × 10-5,
respectively. The delta ferrites transformation behavior of
other stainless steels was also studied. In an austenitic stain-
less steel SUS304, the values for n and k are 0.60 and 2.2 × 10-3
at 1180 °C [7]. The value for n is slightly lower than in the
current experiments. While the value for k is significantly
higher than in the current experiments. That may be due to
that the dissolution of delta ferrites depends on the factors such
as morphology, steel grade, heat treatment temperature, and
so on [7]. Thus, the content of delta ferrites under different
heating time can be predicted by the Eq. A, as shown in Fig.
8(b).
3.3. Influence of pre-deformation on the dissolution of
delta ferrite
In the second group, the percentage of delta ferrites is about
9.3% in sample-1 which was heated for 5 min at 1180 °C
without deformation, as shown in Fig. 9(a). The delta ferrites
are banded in sample-2 after being pre-deformed by 60%,
as shown in Fig. 9(b). The content of delta ferrites in sam-
ple-2 is about 8.5%, which is slightly lower than in sample-1.
This indicates that a small number of delta ferrites have been
dissolved during the deformation. In sample-3, which was
heated for 5 min at 1180 °C after being pre-deformed by
60% at 1080 °C, the content of delta ferrites reduces to 4%,
as shown in Fig. 9(c). That indicates a large number of delta
ferrites have been dissolved during heating for 5 min at 1180 °C.
Fig. 6. The OM micrographs of samples after heating at 1180 °C for 30 min(a), 180 min(b), and 480 min(c).
Fig. 7. The percentage of delta ferrites after heating for different time
at 1180 °C.
Fig. 8. log(-ln(δt/δ0)) as a function of log(t) (a) and the predicted and experimental content of delta ferrites under different heating time at 1180 °C (b).
Influence of High Temperature Pre-Deformation on the Dissolution Rate of Delta Ferrites in Martensitic Heat-Resistant Steels 243
In addition, the deformation degree also has a remarkable
influence on the dissolution of delta ferrites. The content of delta
ferrites significantly decreases with the increase of defor-
mation, as shown in Fig. 9(c) and Fig. 10. The percentage of
delta ferrites reduces to about 2.4% after heating for 5 min
at 1180 °C with a pre-deformation of 80%. The time 75 min,
250 min and 500 min are needed to reduce the content of delta
ferrites to 7.2%, 4.0% and 2.4% respectively only by heating
at 1180 °C without a pre-deformation. Only 5 min is needed
with a pre-deformation of 40%, 60% and 80% respectively.
The results indicate that pre-deformation can remarkably
increase the dissolution rate of delta ferrite.
3.4. Influence of pre-deformation on the microstructure
of delta ferrites
In order to study the promotion mechanism of pre-defor-
mation on the dissolution of delta ferrites, the influence of
pre-deformation on the microstructure of delta ferrites were
observed by FESEM. Without pre-deformation, there is no
interior grain boundaries in delta ferrites after heating for 5
min at 1180 °C, as shown in Fig. 11(a). However, numerous
interior grain boundaries appear after being pre-deformed at
1080 °C. Besides, the interior grain sizes significantly decrease
with the increasing of pre-deformation degree, as shown in
Fig. 11(b), (c) and (d). All the samples were water cooled after
Fig. 9. The OM micrographs of sample-1 (a) sample-2 with a pre-deformation of 60% (b) and sample-3 with a pre-deformation of 60% (c).
Fig. 10. The OM micrographs of the sample-3 which with a pre-defor-
mation of 40% (a) and 80% (b), respectively.
Fig. 11. The FESEM image (perpendicular to compression direction) of sample-1 (a) and sample-2 with a pre-deformation of 40% (b), 60% (c),
and 80% (d).
244 Junru Li et al.
thermomechanical treatment, which indicates pre-deforma-
tion leads to dynamic recrystallization in delta ferrites.
4. DISCUSSION
Delta ferrites is a nonequilibrium phase in the martensitic
heat-resistant steels at room temperature [4]. The dissolution
process of delta ferrites essentially is the uniform diffusion pro-
cess of alloying elements. The dissolution rate of delta ferrites is
controlled by the diffusion rate of alloying elements [5,6].
The literature [6] mentioned that the promotion of ther-
momechanical treatment on the dissolution of delta ferrites
in an austenitic stainless steel was due to the lattice defects
induced by the thermomechanical process. Once dynamic
recrystallization occurs, the promotion effect will be elimi-
nated. Consequently, the optimum condition for dissolution
of delta ferrites can be defined by the highest deformation
temperature and strain in which dynamic recrystallization is
not pronounced. Undeniably, lattice defects will promote the
diffusion of atoms. If the lattice defects can be remained for
a relatively long time during heating process, the dissolution
rate of delta ferrites really can be markedly increased. However,
we found that dynamic recrystallization has occurred even
when the deformation temperature decreased to 920 °C. The
dynamic recrystallization means the repair of defects. In the
other hand, the deformation temperature also cannot be reduced
to too low due to that the plasticity will greatly decrease below
950 °C.
However, we found that the pre-deformation can still mark-
edly promote the dissolution of delta ferrites although the
dynamic recrystallization occurred. It is concluded that the
promotion of pre-deformation on the dissolution of delta ferrites
is based on two reasons. Firstly, pre-deformation increases
the interface area between delta ferrites and matrix. If the
shape of delta ferrites before and after pre-deformation is
respectively assumed as spherical and ellipsoidal, the inter-
face area can be increased by 100% after being deformed 80%.
The increase of interface area will increase the unit-time dif-
fusing quantities of alloying elements between delta ferrites
and matrix. Secondly, pre-deformation leads to dynamic
recrystallization and increases the quantity of internal grain
boundaries in delta ferrites. Generally the diffusion coefficient
of alloying elements at interfaces and grain boundaries is
several orders of magnitude larger than that in the interior of
grains [8]. That is attributed to the lower regularity of atomic
arrangement and the lower activation energy of atomic dif-
fusion at interfaces and grain boundaries. This law is confirmed
in this experimental: The interface of delta ferrites after
deformation is straight in sample-2 which was deformed 60%,
as shown in Fig. 12(a). Some pits appear in the position where
has a internal grain boundary in sample-3 heated for 5 min
at 1180 °C after deformation, as shown in Fig. 12(b). That pow-
erfully proves that the dissolution rate of the delta ferrites is
larger at the internal grain boundaries. Both the increase of
unit-time diffusing quantities of alloying elements between
delta ferrites and matrix and the increase of diffusion rate of
alloying elements will promote the dissolution rate of delta
ferrites.
In the other hand, the interface area and quantity of inte-
rior grain boundaries of delta ferrites will increase more with
the deformation degree. That is the reason why the dissolution
rate of delta ferrites increases with the deformation degree
5. CONCLUSIONS
From the studies of the influence of the high temperature
pre-deformation on the dissolution rate of the delta ferrites
in 10Cr12Ni3Mo2VN steel, several important results are
summarized as follows.
(1) The M23C6-type carbides incline to precipitate and coarsen
at the boundaries of delta ferrites below 930 °C, and can be
rapidly dissolved by heating at 1180 °C.
(2) High temperature pre-deformation can greatly improve
the dissolution rate of the delta ferrite. Pre-deformation can
largely increase the interface area between delta ferrite and
matrix, which increases the unit-time diffusing quantities of
alloying elements between delta ferrites and matrix. In addition,
high temperature pre-deformation leads to dynamic recrys-
Fig. 12. The FESEM image (a long the co mpression directi on) of sample-2 with a pre-deformation of 60% (a) and sample-3 with a pre-deformation of
60% (b).
Influence of High Temperature Pre-Deformation on the Dissolution Rate of Delta Ferrites in Martensitic Heat-Resistant Steels 245
tallization and increases the number of internal grain bound-
aries in the delta ferrites, which greatly increases the diffusing
rate of alloying elements. In these cases, the dissolution of
delta ferrites can be promoted.
ACKNOWLEDGEMENT
The authors are grateful to Xining Special Steel Co., Ltd.
for providing the steel investigated, and appreciate the financial
support by the National High Technology Research and Devel-
opment Program of China (863 Program, No. 2012AA03A502).
REFERENCES
1. P. D. Bilmes, M. Solari, and C. L. Llorente, Mater. Char-
act. 46, 285 (2001).
2. P. Wang, S. P. Lu, D. Z. Li, X. H. Kang, and Y. Y. Li, Acta
Metall. Sin. 44, 681 (2008).
3. D. Carrouge, H. Bhadeshia, and P. Woollin, Sci. Technol.
Weld. Joi. 9, 377 (2004).
4. P. Wang, S. P. Lu, N. M. Xiao, D. Z. Li, and Y. Y. Li, Mat.
Sci. Eng. A 527, 3210 (2010).
5. S. H. Kim, H. K. Moon, T. Kang, and C. S. Lee, Mat. Sci.
Eng. A 356, 390 (2003).
6. M. Rezayat, H. Mirzadeh, M. Namdar, and M. H. Parsa,
Metall. Mater. Trans. A 47, 641 (2016).
7. S. Fukumoto, Y. Iwasaki, H. Motomura, and Y. Fukuda, ISIJ
Int. 52, 74 (2012).
8. D. Phelan and R. Dippenaar, ISIJ Int. 44, 414 (2004).
