Materials Science and Engineering A 527 (2010) 4320–4325

Contents lists available at ScienceDirect

Materials Science and Engineering A

journal homepage: www.elsevier.com/locate/msea

Determination of critical conditions for dynamic recrystallization of a

microalloyed steel
M. Shaban, B. Eghbali ∗
Department of Materials Science Engineering, Sahand University of Technology, P.O. Box 51335-1996, Tabriz, Iran

a r t i c l e i n f o a b s t r a c t

Article history: A low carbon Nb–Ti microalloyed steel was subjected to hot torsion testing over the range of temperatures
Received 4 February 2010 from 900 to 1100 ◦ C and strain rates from 0.01 to 1 s−1 to characterize its hot deformation behavior.
Received in revised form 14 March 2010 The initiation and evolution of dynamic recrystallization were investigated by analyzing of hot flow
Accepted 22 March 2010
curves. Two important dynamic recrystallization parameters, the critical strain and the point of maximum
dynamic softening, derived from strain hardening rate- stress curves. These parameters then were used to
predict the dynamic recrystallized fraction. The results showed that the critical stress and strain increase
Keywords:
with decreasing deformation temperature and increasing strain rate. The hot deformation activation
Microalloyed steel
Torsion testing
energy of the steel investigated in the present work is 375 kJ/mol, and the expression for steady state
DRX flow stress is
Critical strain   0.14
375, 000
Peak strain SS = 0.07Z 0.14 = 0.07 · ε̇ exp
Steady state stress
RT

The volume fraction of dynamic recrystallization as a function of processing variables was established. It
was found that the model used for predicting the kinetic of dynamic recrystallization is in good agreement
with the data directly acquired from experimental flow curves.
© 2010 Elsevier B.V. All rights reserved.

1. Introduction for initiation of DRX in steel investigated. Additionally, the strain

Dynamic recrystallization (DRX) is an important mechanism
for the microstructure control during hot deformation. DRX plays
a major role in reducing the flow stress and the grain size and
is a powerful tool for controlling mechanical properties during
industrial processing [1–4]. Prediction of the critical condition for
the initiation of DRX is of considerable interest for the modeling
of industrial processes [5–8]. Several researchers have proposed
mathematical relations to predict the initiation of DRX. For exam-
ple, Ryan and McQueen suggested that the initiation of DRX can
be identified from changes in the slope of the strain hardening
rate versus flow stress curves [9]. Alternatively, Poliak and Jonas
[5–8] proposed the use of the minimum in the absolute value of the
strain hardening slope which can be calculated from strain hard-
ening rate-flow stress curves. The objective of this investigation
is to analyze the initiation of dynamic softening behavior through
DRX under various deformation conditions in a low carbon Nb–Ti
microalloyed steel. Accordingly, the method of Poliak and Jonas is
used to determine the critical strain and the peak strain required
∗ Corresponding author. Tel.: +98 412 3443801; fax: +98 412 3444333.
E-mail addresses: eghbali@sut.ac.ir, eghbali417@yahoo.com (B. Eghbali).
of maximum dynamic softening is determined by using the work
hardening rate. These parameters are used to predict the kinetic of
dynamic recrystallization.
2. Experimental procedure
The chemical composition (wt.%) of the steel investigated was
0.09C, 0.4Si, 1.55Mn, 0.008S 0.014P, 0.013Ti, 0.031Nb, 0.028Al, and
balance Fe. Torsion specimens with a gauge length of 20 mm and
diameter of 6.7 mm were machined from the as-received plates
with the longitudinal axis parallel to the rolling direction. The
deformation tests were performed on torsion equipment described
elsewhere [10]. Specimens were enclosed in quartz tube with a
positive pressure of Argon gas atmosphere to prevent decarburiza-
tion during induction heating. The temperature was monitored by
two thermocouples inserted in the drilled ends of the specimen.
The temperature accuracy was within ±5 ◦ C. Initially a continuous
cooling torsion test was carried out to measure the critical trans-
formation temperatures. The Ar1 , Ar3 and non-recrystallization
temperature (Tnr ) were found to be 750, 790, and 977 ◦ C, respec-
tively.
Thermomechanical treatment schedule is shown in Fig. 1. At
first, each specimen was heated to 1200 ◦ C and held for 2 min at

0921-5093/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.msea.2010.03.086
 M. Shaban, B. Eghbali / Materials Science and Engineering A 527 (2010) 4320–4325 4321

This precipitation is in turn responsible for the formation of pan-

caked austenite grains by retarding the occurrence of DRX [13].
Thus, it can be realized that the softening behavior during deforma-
tion at this deformation temperature results from the counteracting
effects of partial dynamic recrystallization and work hardening of
austenite.
It can be seen that at a given temperature with increasing strain
rate working hardening rate increases. Thereby the peak stress ( p ),
the peak strain (εp ), and the steady flow stress increase. At a cer-
tain strain rate, with increasing deformation temperature the rate
of dynamic softening increases. Thus the peak stress, peak strain,
and steady state stress decrease correspondingly. In other words,
the peak stress and peak strain are dependent on deformation con-
ditions.

3.2. Hot deformation characteristics

So far, several empirical equations have been proposed to

Fig. 1. A schematic diagram of thermomechanical processing schedule conducted
in the present study. determine the deformation activation energy and hot deformation
behavior of steels. The most frequently used one is as follows [14].
This equation describes the behavior of the material deformed at
temperature to fully austenitise the material. The temperature was different temperatures and strain rates:
then decreased at a rate of 1 ◦ C/s to the test temperature (900, 1000,
  Q q
and 1100 ◦ C) and held there for 2 min to homogenize the specimen
SS = AZ q = A · ε̇ exp d
(1)
temperature. The specimen was deformed isothermally at strain RT
rates of 0.01, 0.1, and 1 s−1 . All specimens were deformed up to
where  ss is the steady state stress, A is a constant depending
strain of 2. The specimens were quenched with water jet sprays
on chemical composition and initial austenite grain size, Z is the
immediately after deformation.
Zener–Hollomon parameter, ε̇ is the strain rate, Qd is the activa-
tion energy of deformation, R is the gas constant, T is the absolute
3. Results and discussion temperature and q is the power low exponent. On taking natural
logarithms of both sides of Eq. (1) the expression for steady state
3.1. Analysis of hot flow curves flow stress is as:
Qd
The effect of strain rate and deformation temperature on the ln SS = ln A + q ln ε̇ + q (2)
RT
true strain–true stress curves of the steel investigated is shown in
Fig. 2. It is seen that all flow curves exhibit a peak and then follow From Eq. (2) it can be seen that when the deformation temper-
by a gradual fall to a steady state stress which is indicative of the ature is constant, there exists a linear relation between natural
occurrence of DRX, except for the 900 ◦ C. logarithm of the steady state flow stress and the natural logarithm
In other words, dynamic recrystallization conditions were of strain rate. When the deformation temperature is constant, on
apparently not achieved in the tests performed at temperature taking partial derivatives of both sides of Eq. (2) to − ln ε̇ we have:
lower than Tnr , i.e. at 900 ◦ C. It has been reported [11–13] that  
∂ ln SS
the dynamic strain induced precipitation would take place dur- = −q (3)
−∂ ln ε̇
ing deformation in this temperature range in microalloyed steels. T

Fig. 2. True stress–true strain curves of Nb–Ti microalloyed steel under different deformation conditions.
4322 M. Shaban, B. Eghbali / Materials Science and Engineering A 527 (2010) 4320–4325

Fig. 3. The effect of strain rate (a) and temperature (b) on the steady state stress flow stress at constant temperature and strain rate.

Therefore, according to Eq. (3) the power low exponent (q) can be reaches to zero at the end of first DRX cycle. Therefore, the steady
determined from the slope of the plot of steady state stress as a state strain can be defined from –ε curves. From –ε plots, it is
function of the logarithm of strain rate at constant temperature, as inferred that at 1100 ◦ C dynamic recrystallization is complete at all
shown in Fig. 3a. This value is calculated as q = 0.14 for this steel. strain rates. However, at 900 ◦ C dynamic recrystallization is partial
When the strain rate is constant, on taking partial derivative of both due to the fact that strain hardening rate does not reach to zero at
sides of Eq. (2) to the inverse deformation temperature we obtain: the end of deformation.
 
∂ ln SS Qd
1
=q (4)
∂ R
T ε̇

According to Eq. (4), the activation energy of deformation (Qd ) can

be determined from the slope of the plot of the steady state flow
stress and the inverse temperature at constant strain rate, as shown
in Fig. 3b. This value is calculated as q = 0.14 for this steel. The value
of Qd is obtained to be about 375 kJ/mol. Therefore, under the defor-
mation conditions over a temperature range of 900–1100 ◦ C and a
strain rate range of 0.01–1 s−1 , the steady state hot flow behavior
of present steel can be expressed as:
  375, 000 0.14
SS = 0.07Z 0.14 = 0.07 · ε̇ exp (5)
RT

3.3. The initiation of DRX

For the hot flow curves given in Fig. 2, the true stress–true strain
data were used to calculate the values of the strain hardening rate
( = d/dε). The strain hardening rate values were then plotted as
a function of the flow stress (Fig. 4) for deformation temperatures
ranging from 900 to 1100 ◦ C and strain rates ranging from 0.01 to
1 s−1 . According to the approach of Poliak and Jonas [5–8] and Ryan
and McQueen [9], in the curves of – the point at which the work
hardening rate equals zero ( = 0) represents the peak stress ( p ).
The inflection point of – curves indicates the critical stress ( c )
for the initiation of DRX.
From the data given in Fig. 4, the derivative of the strain hard-
ening rate (d/d) as a function of the flow stress for various
deformation temperatures and strain rates was calculated. The
d/d vs.  plots are given in Fig. 5. The maximum points in these
plots represent the critical stress. The critical strain (εc ) can be
defined by mapping the critical stress back in to stress–strain curves
[12].

3.4. Strain for maximum dynamic softening

The work hardening rate was plotted as a function of the true

strain at different deformation temperatures and strain rates as
shown in Fig. 6 to find the strain at which the rate of dynamic soft-
ening is maximized. The minimum points in these plots represent
the strains (ε* ) at which the rate of dynamic softening is maxi-
mized [12]. It is seen that the value of ε* increases with decreasing
deformation temperature and increasing strain rate. In addition, Fig. 4. – curves of hot deformed Nb–Ti steel at different temperature and strain
after this minimum point the work hardening rate increases and rates.
 M. Shaban, B. Eghbali / Materials Science and Engineering A 527 (2010) 4320–4325 4323

Table 1
Calculated DRX parameters of Ti-Nb steel at different deformation conditions.

Temperature 900 ◦ C 1000 ◦ C 1100 ◦ C

Strain rate 0.01 0.1 1 0.01 0.1 1 0.01 0.1 1

εc 0.49 0.5 0.77 0.13 0.27 0.46 0.12 0.165 0.28

εp 0.79 0.828 1.1 0.3 0.52 0.68 0.215 0.28 0.423
ε* 1.97 1.36 2 0.42 0.746 1.242 0.28 0.44 0.7
εSS – – – 1.26 1.36 – 0.35 0.94 1.8
 c (MPa) 107 130 167 58 91 129 42 58 86
 p (MPa) 120 135 170 69 100 135 46 65 91
 * (MPa) 111 135 – 67 96 128 45 63 88
 SS (MPa) – – – 66 91 116 43 60 81

The entire calculated DRX parameters are listed in Table 1. It is critical characteristics of DRX and Z. This can be expressed as fol-
obvious that when deformation temperature remains the same the lows:
mentioned parameters increase with increasing strain rate. How-
ever, in the case of equal strain rate, these parameters increase εc = 0.00055 Z 0.189 (6)
as the deformation temperature decreases. The dependence of the εc = 0.74283 Z 0.1433 (7)
critical and peak strain and stress of the present steel in the differ-
ent deformation conditions (different Z values) are shown in Fig. 7. The dependence of the peak stress and strain on temperature and
It can be seen that the critical characteristics of DRX increase with strain rate confirms that the deformation is controlled by a ther-
increasing Z parameter. The ratio of εc /εp and  c / p are 0.64 and mally activated process [13,14].
0.9, respectively. Therefore, there is a linear relationship between

Fig. 6. Strain hardening versus strain for Nb–Ti steel deformed at different temper-
Fig. 5. (d/d)– curves of Nb–Ti steel at different temperature and strain rates. ature and strain rates.
4324 M. Shaban, B. Eghbali / Materials Science and Engineering A 527 (2010) 4320–4325

Fig. 7. The effect of deformation condition (Z parameter) on: (a) εp , εc , and (b)  p ,  c .

3.5. Dynamic recrystallization kinetics

The volume fraction of dynamic recrystalliztion could be

described with Avrami type model as follows [12]:
  ε − ε 2 
C
XDRX = 1 − exp −0.693 (8)
ε∗ − εC

where, XDRX is dynamic recrystallization fraction, ε the true strain,

εc the critical strain for the onset of DRX, and ε* the strain for the
maximum softening rate during dynamic recrystallization. Fig. 8
shows the results of this model applied to the present data. The
volume fraction of dynamic recrystalliztion as a function of strain
at different deformation temperature, with constant strain rate, is
shown. It is seen that as the strain increases the volume fraction of
dynamic recrystalliztion reaches a constant value of 1. In addition,
it is seen that the strain required for the same amount of recrystal-
lized fraction increases with decreasing deformation temperature.
At the deformation temperature of 900 ◦ C, DRX is delayed to longer
times and higher strains with respect to those for the other two
temperatures.
The onset of steady state deformation could be identified using
the plot of work hardening against strain. The point after peak
strain, at which the work hardening rate reaches to zero, corre-
sponds to the point of the onset of steady state flow [12]. In addition,
the first DRX cycle becomes complete at the onset of steady state
deformation. In Fig. 9, the work hardening rate and recrystallized
fraction are both plotted against strain at different strain rates
with temperature being constant at 1100 ◦ C. This figure confirms
the accuracy of the model used in the prediction of dynamically
recrystallized volume fraction. It is clear that the onset of steady
state deformation calculated from work hardening rate data ( = 0)
coincides with completion of dynamic recrystallization which is
predicted by the use of Avrami type model.
Fig. 9. Coincidence of predicted steady state strain with work hardening rate ( = 0)
and with the model used for calculation of recrystallized fraction (XDRX = 95–98%).
4. Conclusions
In the present study, the dynamic recrystallization behavior of
Nb–Ti microalloyed steel was investigated using hot torsion tests.
The main results can be summarized as follows:
(1) Under the deformation conditions over a temperature range of
900–1100 ◦ C and a strain rate range of 0.01–1 s−1 , the constitu-
tive equation of steady state hot flow behavior of present steel
was developed.
(2) The deformation activation energy of austenite was calculated
about 375 kJ/mol.
(3) The important parameters of DRX such as: the critical strain
for initiation of DRX (by the method of Poliak and Jonas), the
peak strain, the strain associated with the maximum rate of
dynamic softening, the steady state strain, the critical stress,
the peak stress, the stress associated with the maximum rate
of softening, and the steady state stress was determined from
true stress–true strain and work hardening rate data.
(4) The critical strain and strain of maximum softening rate was
used to predict the kinetic of dynamic recrystallization. The
Avramy type expression was successfully applied to model the
DRX kinetic.

Acknowledgements

One of the authors (B. Eghbali) expresses his gratitude to Prof.

P.D. Hodgson at Deakin University in Australia for the provision of
the hot deformation laboratory facilities used in this research.

References
Fig. 8. Predicted volume fraction of dynamic recrystallization obtained under dif-
ferent temperatures at constant strain rate of 0.1. [1] J.J. Jonas, Mater. Sci. Eng. A A184 (1994) 155–161.
 M. Shaban, B. Eghbali / Materials Science and Engineering A 527 (2010) 4320–4325
[2] H.J. McQueen, S. Yue, N.D. Ryan, E. Fry, Mater. Process. Technol. 53 (1995)
293–298.
[3] G.R. Stewart, J.J. Jonas, F. Montheillet, ISIJ Int. 44 (9) (2004) 1581–1589.
[4] F. Montheillet, J.J. Jonas, Encycl. Appl. Phys. 16 (1996) 205–211.
[5] E.I. Poliak, J.J. Jonas, ISIJ Int. 43 (5) (2003) 684–691.
[6] E.I. Poliak, J.J. Jonas, Acta Mater. 44 (1996) 127–136.
[7] E.I. Poliak, J.J. Jonas, ISIJ Int. 43 (2003) 692–700.
[8] J.J. Jonas, E.I. Poliak, Mater. Sci. Forum 426–432 (2003) 57–66.
[9] N.D. Ryan, H.J. McQueen, Can. Metall. Q 29 (1990) 147–152.
4325
[10] P.D. Hodgson, D.C. Collinson, B. Perett, Proceed. Seventh Int. Sympos. on Phys-
ical Simulation, Tsukuba, Japan, 1997, pp. 219–224.
[11] C.M. Sellars, G.J. Davies, Proceed. Int. Conf. on Hot Working, Metals Society,
London, 1980.
[12] G.R. Stewart, A.M. Elwazri, S. Yue, J.J. Jonas, Mater. Sci. Technol. 22 (5) (2006)
519–524.
[13] S. Sakui, T. Sakai, K. Takeishi, Trans. Iron Steel Inst. Jpn. 17 (1977) 719–
724.
[14] C.M. Sellars, W.J. McG Tegart, Mem. Scientif. Rev. Metall. LXIII 9 (1966) 731–736.