Content uploaded by Pablo Leão
Author content
All content in this area was uploaded by Pablo Leão on May 31, 2022
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ymst20
Materials Science and Technology
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ymst20
Recrystallisation kinetics and yield-strength
adjustment after annealing of cold-rolled
microalloyed steel
Pablo Bruno Paiva Leão, S. L. S. Medeiros, B. R. C. Saraiva, J. R. Barros Neto,
C. C. Silva, Antonio J. Ramirez & H. F. G. de Abreu
To cite this article: Pablo Bruno Paiva Leão, S. L. S. Medeiros, B. R. C. Saraiva, J. R. Barros
Neto, C. C. Silva, Antonio J. Ramirez & H. F. G. de Abreu (2022) Recrystallisation kinetics and
yield-strength adjustment after annealing of cold-rolled microalloyed steel, Materials Science and
Technology, 38:6, 363-376, DOI: 10.1080/02670836.2022.2043626
To link to this article: https://doi.org/10.1080/02670836.2022.2043626
Published online: 04 Mar 2022.
Submit your article to this journal
Article views: 135
View related articles
View Crossmark data
MATERIALS SCIENCE AND TECHNOLOGY
2022, VOL. 38, NO. 6, 363–376
https://doi.org/10.1080/02670836.2022.2043626
Recrystallisation kinetics and yield-strength adjustment after annealing of
cold-rolled microalloyed steel
Pablo Bruno Paiva Leãoa,S.L.S.Medeiros
a,B.R.C.Saraiva
a, J. R. Barros Netob,C.C.Silva
a, Antonio J. Ramirezc
andH.F.G.deAbreu
a
aDepartment of Metallurgical and Materials Engineering, Universidade Federal do Ceará, Fortaleza, Brazil; bDepartment of Materials
Engineering, Universidade Federal do Piaui, Teresina, Brazil; cDepartment of Materials Science and Engineering, The Ohio State University,
Columbus, OH, USA
ABSTRACT
After continuous annealing process (CAP) at 790°C, 85% of the coils of 50% cold-rolled low carbon
microalloyed (LCM) steel did not exhibit yield-strength (YS) on the target range, while the 70%
cold-reduced LCM coils did. In this context, the non-isothermal recrystallisation kinetics of ferrite
for the above two full-hard LCM steel were investigated using differential scanning calorimetry
and the Friedman differential isoconversional method. The recrystallisation kinetics of ferrite for
the two deformed states showed different behaviour. Regarding a fixed degree of cold-rolling
deformation, the soaking temperature was found as the manageable parameter to control YS
during CAP. Consequently, a suitable YS of the 50% cold-rolled LCM steel was achieved by setting
the soaking temperature at 773°C.
ARTICLE HISTORY
Received 14 September 2021
Revised 31 January 2022
Accepted 2 February 2022
KEYWORDS
Low carbon microalloyed
steel; method of Friedman;
differential scanning
calorimetry; activation
energy; cold-rolling;
continuous annealing
process; recrystallisation;
yield-strength
Introduction
Low carbon microalloyed (LCM) steels allow the use
of thinner strips with suitable mechanical properties
in car bodies. This achievement is a result of simul-
taneous improvement of strength and toughness. It is
based on grain renement during the hot-rolling pro-
cess controlled by both; the solubility of precipitates in
the iron matrix and the stages of the thermomechani-
cal process [1,2].Inthesubsequentcold-rollingprocess,
dislocations are introduced into the material’s crystal
lattice, and consequently, work hardening occurs. Also,
a non-uniform accumulation of dislocation takes place.
This way, internal structures within the grains, such as
subgrains, and new grain boundaries, are formed. Even-
tually, the grain boundary (GB) area increases and cold-
deformed strips retain an amount of energy propor-
tional to their degree of reduction. This stored energy
consists essentially of the energy of all dislocations and
new interfaces [3,4].
Then, the continuous annealing process (CAP) is
one of the typical processing routes applied to cold-
worked steel, mainly to restore its mechanical prop-
erties (softening). At CAP, cold-rolled strips of low-
carbon steels are often annealed at intercritical temper-
atures [5]. In this circumstance, the heating rate signif-
icantly aects the metallurgical events during anneal-
ing treatment, such as ferrite recrystallisation. At a
slow heating rate ( <10°C/s), ferrite recrystallisation
can occur non-isothermally before austenite nucle-
ationincold-workedlow-carbonsteels,andtheholding
time does not aect the ferrite microstructure [6,7]. In
contrast, ferrite recrystallisation is delayed by austen-
ite transformation when the heating rate increases
(>100°C/s). Similarly, the inuence of holding time
and the cooling rate becomes very relevant at high
heating rates [3,4,8,9].
In this context, it is known from the literature
[10,11] that dierential scanning calorimetry (DSC)
is a powerful technique for studying phase transfor-
mations in steels and alloy materials. In addition, the
DSC technique can perform non-isothermal measure-
ments [11].Underthiscircumstance,theeectiveacti-
vation energy (Ex)of a phase transformation frac-
tion can be determined by isoconversional methods
without assuming the kinetic-model function (model-
free) [12,13]. Furthermore, several works report the
application of Friedman’s dierential isoconversional
method (FDIM) [14] in dierent solid-state kinetic
studies: kinetic of material degradation, crystallisation
and phase transformation in mould uxes for steel cast-
ing, metallic glassy crystallisation, and phase transfor-
mation in steels and metal alloys [11,15,16]. However,
studies adopting isoconversional methods for eval-
uating specically recrystallisation transformation in
cold-rolled steels are scarce, mainly considering dif-
ferent degrees of cold-deformation and correlations
CONTACT Pablo Bruno Paiva Leã o pablobruno@alu.ufc.br; pbpleao@gmail.com
© 2022 Institute of Materials, Minerals and Mining.
364 P. B. P. LEÃO ET AL.
between the behaviour of Exand the initial microstruc-
ture/crystallographic orientation of the material [10].
Therefore, the current work aims to describe
the non-isothermal ferritic recrystallisation kinetics
through DSC and model-free FDIM for the LCM steel
in two distinct degrees of cold-rolled reductions (50%
and 70%), to demonstrate that the model-free FDIM is
a viable way to evaluate the experimental recrystalli-
sation kinetic data. Also, electron backscatter dirac-
tion (EBSD) analyses were used to provide crystallo-
graphic insights of both cold-worked steel conditions.
Additionally, this study is based on an industrial case,
in which 85% of the coils produced from the LCM
steel with 50% in cold-reduction had, after CAP (at
790°C), a yield-strength (YS) value below the target
range (340–420 MPa). Meanwhile, the 70% cold-rolled
coils reached the suitable YS-values with the same soak-
ing temperature. Thus, the recrystallisation kinetics
results can provide technical information to solve a pro-
cess issue. The results obtained from a collaboration
between university and industry are reported in the
following.
Materials and methods
Industrial aspects
During the hot-rolling process, the investigated LCM
steel was produced with a nishing rolling temperature
of 920°C (±2) and a coiling temperature of 590°C (±2).
Subsequently, the strips were 50% and 70% cold-rolled
reaching a nal thickness of 1.6 and 0.7 mm, respec-
tively. In addition, industrial data from CAP were eval-
uated to establish relationships among strip thickness,
stripspeed,soakingtemperatureandYS.
50% and 70% cold-rolled samples
Full-hard steel samples were collected from the cen-
tral position of both 50% and 70% cold-rolled strips
at the entrance of CAP, and their crystallographic tex-
turesandkineticswereevaluatedinthiswork.The
chemical composition of both full-hard steel conditions
obtained by optical emission spectrometer (THERMO-
ARL 4460 M) is shown in Table 1.Thesampleswere
prepared through the regular metallographic procedure
based on grinding and polishing, using sandpapers
(from #100 to #2000 grit) and diamond suspension (6,
3, and 1 µm). At that point, the samples were etched in
2% Nital solution for microstructural observations per-
formed in an optical microscope (ZEISS model Axio
Imager 2). For microtexture investigation, the sample
preparation procedure received a nal vibratory polish-
ing step with 0.04 µm colloidal silica for one hour. The
EBSD maps were acquired employing a scanning elec-
tron microscope (model Quanta 450 FEG-FEI) coupled
with an EBSD detector (Oxford Instruments, Oxford,
UK). The EBSD analysis was carried out using an oper-
ating voltage of 20 kV, a step size of 100 nm, a working
distance of 10 mm and a sample tilt angle of 70°.
Regarding recrystallisation kinetics, four rectangu-
lar samples, for each aforesaid full-hard strip, were
ground (using sandpaper of #200 grit) up to dimen-
sions of 2 (width) ×4(length)×0.5 (thickness) mm
for removing oxidecontamination. Then, the rectangu-
lar ground samples were immediately used in the DSC
technique through a NETZSCH machine (model TG-
DSC STA 449 F3 Jupiter). This analysis was executed at
four constant heating rates (10, 15, 25 and 30°C/min)
from room temperature up to 720°C in the argon gas
atmosphere. The Friedman method [14]wasusedto
calculate the eective activation energy from the DSC
thermograms.
50% cold-rolled samples after CAP
For adjusting YS after CAP of the 50% cold-rolled
strip, twenty-three tensile samples were collected on
the central position of the galvanised strip for each
assessed soaking temperature (760°C, 780°C, 790°C,
805°C and 830°C). Meanwhile, the other parame-
ters were kept as consistent as possible (heating rate
around 10°C/s, cooling rate around 15°C/s, strip’s
speed around 60–80 m/min, soaking time about 50 s,
strip’s thickness of 1.6 mm, and skin pass mill of 1.2%).
In this context, planar and normal anisotropy proper-
ties were also measured for the annealed strips in each
investigated soaking temperature from the following
equations:
Normal Anisotropy, ¯
r=(r0+2r45 +r90)/4
(1)
Planar Anisotropy, r=(r0−2r45 +r90)/2
(2)
The anisotropy parameters (r0,r45,r90)were obtained
in duplicate from tensile tests. In this case, the ten-
sile samples were orientated in 0°, 45° and 90°, con-
cerning the strip’s rolling direction. The NBR ISSO
6892-1standard was employed for the tensile test per-
formance and tensile specimen dimensions. These tests
were performed using an Instron 5582–100KN Univer-
sal Testing Machine. Furthermore, the average grain
Tab le 1. Compositions of the cold-rolled LCM steel samples (wt%).
Sample C Mn PSSiNiCrAlTiNbNFe
50% cold-rolled steel 0.049 0.548 0.018 0.007 0.005 0.005 0.012 0.056 0.03 0.001 0.0043 Bal.
70% cold-rolled steel 0.048 0.514 0.016 0.008 0.007 0.006 0.015 0.062 0.03 0.001 0.0041 Bal.
MATERIALS SCIENCE AND TECHNOLOGY 365
size was determined according to the ASTM E112 stan-
dard.
Regarding the inuence of the holding time and
soaking temperature on the amount of austenite formed
during CAP, the commercial software JMatPro®[17]
was employed. The model of JMatPro®for simulating
the reaustenitisation of steels is based on the mod-
ied Johnson-Mehl-Avrami equation [18]. The ther-
modynamic calculation was performed using, as input
data, the chemical composition of the 50% cold-rolled
steel in Table 1, the heating rate of 10°C/s, and the
microstructural condition set as normalised. Moreover,
dierent holding times (50 and 70s) and soaking tem-
peratures (760°C, 780°C, 790°C, 805°C and 830°C) were
inserted for doing many calculations. In this case, the
software assumes an initial microstructure of a mixture
of ferrite and pearlite/carbides, and that the kinetics of
reaustenitisation are limited by carbon diusion. Dur-
ing heating, it is also adopted that, regions with car-
bides rst transform into austenite, followed by ferrite.
Finally, the starting transformed austenite is inhomo-
geneous, and the further austenite homogenisation is a
function of the holding time and soaking temperature
of reaustenitisation.
Results and discussion
Microstructure and microtexture of the steel in the
cold-rolled states
Figure 1(a, b) shows the optical micrographs on the
cross-section region of the LCM steel strips in the
50% and 70% cold-rolled conditions. Notably, the fer-
ritic grains are attened in a pancake-like shape. The
70% cold-worked strip presents smaller grains (an aver-
age thickness of 1.74 µm) than in the 50% deformed
state (an average thickness of 3.33 µm). Also, cemen-
tite particles were found more spread out on the ferritic
matrix and smaller in the 70% deformed one. The aver-
age lengths of the cementite particles were 6.77 and
2.59 µm, respectively, for 50% and 70% cold-worked
steel.
The colour-coded inverse pole gure (IPF) maps
in Figure 2(a, b) show the subsurface microstructures
of the full-hard samples in the abovementioned cold-
rolledconditions.Inbothcases,itispossibletorealise
that the ferritic grains are stretched along the rolling
direction (RD), bound by GB or high angle grain
boundaries (dened by θ≥15° and shown by black
lines). Inside the ferritic grains, there is a high density
of subgrain boundaries or low angle grain boundaries
(dened by 2° ≤θ<15° and shown by white lines)
where, on the analysed area, they represented a frac-
tion of 78% and 83%, respectively, for the 50% and 70%
cold-deformed states. The black points in Figure 2(a,
b) were regions that could not reach suitable Kikuchi
patterns.
Furthermore, the orientation distribution function
(ODF) from the individual IPF maps is shown in Figure
3(a, b), and their intensities are expressed in multi-
ple random densities (mrd). In both cases, the ODFs
displayed similar bres, a partial α-bre and a γ-
bre, but with distinct component intensities. Speci-
cally, the 50% cold-rolled strip revealed a dominant α-
bre, in which 113 ≈112 <110 >was the component
in sharper intensity (6.7 mrd). This outcome may be
attributable to prior deformed austenite during the hot-
rolling process [19]. Conversely, the 70% cold-rolled
strip exhibited a dominance of γ-bre with 111 <112 >
as the strongest orientation (8.2 mrd), which can be
developed during cold-rolling reduction [20].
Moreover, from dierent techniques, it is well estab-
lished that the stored energy in γ-bre orientations
is higher than in α-bre orientations [21,22]. Partic-
ularly, it has been pointed out that, along the γ-bre,
the 111 <112 >component presents the higher stored
energy [23]. So, the 70% cold-rolled steel gives metal-
lurgical features that indicate a state of higher stored
energy than in the 50% cold-rolled one.
The activation energy (Ex) of ferrite
recrystallisation
Figure 4shows the DSC thermograms for the studied
LCM steel at dierent heating rates. Each thermogram
showed two exothermic events, which were conrmed
bytherstderivativeoftheDSCcurve.Theexothermic
peaks, whose onset and nal temperatures are outlined
by continuous lines, are in a temperature range corre-
sponding to ferrite recrystallisation of cold-rolled steel
with similar characteristics to the steel under study
[10,24,25]. The smooth peaks pointed by arrows in
Figure 4were considered recovery events. Moreover,
the recovery and recrystallisation phenomena must be
visible as exothermic peaks during heating due to the
release of stored energy accumulated during the cold-
rolling process [26]. In this regard, the kinetic evalua-
tionswillfocusontherecrystallisationpeaks,inwhich
the dashed lines represent their peak temperatures.
The recrystallisation fraction (x)from the DSC ther-
mograms as a function of temperature was reached
by evaluating the partial area above the recrystallisa-
tion peaks by using Equation 3, which must result in
sigmoidal curves for each heating rate
x(T)=AT/A(3)
where ATis the partial area between the onset and a
chosen transformation temperature and Ais the total
area from the onset up to the nal transformation tem-
perature of the exothermic peak. The recrystallisation
fraction (x),asafunctionofthetime,canbeobtained
by using the following equation:
t=(T−T0)/B(4)
366 P. B. P. LEÃO ET AL.
Figure 1. Optical microstructure (1000×of magnification) on the cross-section of (a) 50% cold-rolled LCM steel and (b) 70% cold-
rolled LCM steel.
where Bis the heating rate, T0is the temperature at
the beginning of the transformation (t=0)and Tis
the temperature at a given time tduring the recrystalli-
sation process. Figure 5depicts the degree of recrys-
tallisationasafunctionoftime.Afterthat,itispossi-
ble to calculate the instantaneous recrystallisation rate
dx/dtthroughthederivationofthecurvesinFigure
5,andthen,itcanbeplottedtheln(dx/dt) against x.
This information will be used to calculate the ferrite
recrystallisation kinetics. In this context, the eective
activation energy was estimated by using the FDIM [14]
through the following equation:
ln (dx/dt)x=−Ex/(RTx)+C(5)
where Cis a constant, Ris the gas constant, Ex,(dx/dt)x
and Txare, respectively, the eective activation energy,
the instantaneous recrystallisation rate and the tem-
perature at a relative x. So, the data ln(dx/dt) and
Tcanbecollectedforthesamexthroughtheprior
described plots x versus Tand ln(dx/dt) versus x. Then,
MATERIALS SCIENCE AND TECHNOLOGY 367
Figure 2. Colour-coded inverse pole figure (IPF) map on the subsurface region (below 1
4of the initial thickness) of (a) 50% cold-rolled
LCM steel and (b) 70% cold-rolled LCM steel.
by plotting ln(dx/dt) against 1000/Txfor each x of the
four assessed heating rates, it must give a straight line
with a slope (Ex/R), as presented in Figure 6. Finally,
basedonEquation5,theactivationenergycanbe
estimated.
Figure 7shows the dependence of the eective acti-
vation energy on the recrystallised ferritic fraction of
the investigated LCM steel in both evaluated degrees
of deformation. It can be seen that the change in the
dependence of Exis not similar between the two con-
ditions. The 50% cold-rolled steel displays an almost
unchanged performance over the x from 0.2 to 0.4. This
regime is related to the fact that it is mainly governed by
the nucleation step [27]. In contrast, the more deformed
conditions did not exhibit this constant regime, sug-
gestingthatitmayhaveafasternucleationrateduring
heating. This evidence is consistent with the metallur-
gical features found in the full-hard coils, in which the
nucleation rate of new grains depends on the initial
state of stored energy [3].
Regarding the regimes in which the Exincreases in
Figure 7, these stages may involve simultaneous nucle-
ation and crystal growth mechanisms [27]. However,
after the initial nucleation,recrystallisation may be
dominated mainly by the consumption of the deformed
grains by the growth of the new recrystallised grains.
In this context, the 70% cold-rolled steel exhibited a
lowerrateofincreaseofEx,concerningthesameregime
in the less deformed state (x≥0.6).Thisindicates
that recrystallised grain growth is slower in the 70%
cold-rolled material than in the less deformed steel
[3]. In this way, the 70% cold-rolled steel has some
aspects that can explain its slowed grain growth kinet-
ics. One of them is the smaller and more widely dis-
tributed cementite particles. They may pin the recrys-
tallised grains and thus prevent their coarsening [28].
Another is the 111 <112 >deformed grains with
higher stored energy, in which the number of nuclei
formed should be larger, and the proportion of recrys-
tallised grains should initially increase more rapidly.
Thiseventmayleadtothemeetingofthenewfree-
dislocation grain boundaries, still in small sizes. Con-
sequently, the recrystallisation rate will tend to decrease
due to pinning between the new GBs [3].
Additionally, recent studies [29,30]reportedthat
in multi-phase steels with ductile matrix and hard
second phase particles, the GB’s and matrix/second
phase particles interfaces are prone to accumulation of
strain/dislocation during plastic deformation. Based on
this, it may be considered that smaller grain size (larger
number of GB per unit area) and less agglomerated
cementite particles (more ferrite/cementite interfaces)
368 P. B. P. LEÃO ET AL.
Figure 3. Orientation distribution function (ODF) at ϕ2=45◦of (a) 50% cold-rolled LCM steel and (b) 70% cold-rolled LCM steel.
in the 70% cold-rolled steel can also contribute
to increasing the start number of sites during the
recrystallisation process. Moreover, it is known from
the literature [31]thatcarbonitrideparticlesmay
also signicantly inuence pinning grain growth in
Ti/Nb stabilised steels. These precipitates’ character-
istics and pinning eect are primarily determined by
the hot-rolling and coiling parameters, identical in the
MATERIALS SCIENCE AND TECHNOLOGY 369
Figure 4. DSC thermograms at different heating rates (10, 15, 25 and 30°C/min) of (a) 50% cold-rolled LCM steel and (b) 70% cold-
rolled LCM steel.
evaluated conditions. Therefore, this pinning eect was
considered equivalent in both cases.
CAP industrial data
Figure 8(a) displays a relationship between strip thick-
ness and its respective average production speed. The
latter represents the average velocity that the strip runs
in the CAP line. During CAP, the strip thickness varia-
tion can be considered an independent variable because
it results from a previous process parameter (degree of
reduction during the cold-rolling process). As observed
in Figure 8(a), average strip speed has an inversely
proportional dependence on the strip thickness. This
behaviour is mainly related to strip thickness and its
weight, impacting the plant’s security and operational
370 P. B. P. LEÃO ET AL.
Figure 5. Recrystallisation fraction (x) of full-hard LCM steel as a function of time (s), at different heating rates (10, 15, 25 and
30°C/min) for (a) 50% cold-rolled LCM steel and (b) 70% cold-rolled LCM steel.
capacity. Moreover, it could present a signicant speed
variation during CAP along with the same coil. So,
the strip’s speed is not entirely manageable because it
depends on the coil’s thickness and the process stability.
Furthermore, Figure 8(b) exhibits the strip speed
versus soaking temperature of CAP and both as a func-
tion of YS (in contour plot colour). For the evalua-
tion of Figure 8(b), it is essential to take into account
the relationship described in Figure 8(a). In particular,
Figure 8(b) shows YS divergence at a constant soak-
ing temperature and dierent strip speeds. In this case,
two signicant factors, with simultaneous eects, can
be recognised; the holding time and previously dis-
cussed ferritic recrystallisation kinetics. To summarise,
thinner strips (higher cold-reduction) with disadvan-
tageous kinetics for ferritic grain growth are mainly
MATERIALS SCIENCE AND TECHNOLOGY 371
Figure 6. ln(dx/dt)as a function of 1000/Txfor different recrystallisation fraction of LCM steel for (a) 50% cold-rolled LCM steeland
(b) 70% cold-rolled LCM steel.
produced in higher velocities (shorter holding time),
remaining in the higher range of YS. While, thicker
strips (lower cold-reduction) with favourable kinetics
for ferritic grain growth are manufactured in lower
velocities (longer holding time), staying in the lower
range of YS. In this way, considering the Hall-Petch [32]
relationship, the thicker strips may have a larger nal
ferritic grain size after CAP than the thinner ones.
Also, the austenite phase is formed at intercritical
temperatures and is converted into ferrite during cool-
ing [5,6,7]. It is well known that austenite mainly nucle-
ates in ferritic GBs and ferrite/Fe3Cinterfaces,andthe
diusion phenomenon controls its growth [5,6,7,33].
Therefore, it suggests that at the intercritical temper-
ature during the annealing, austenite may nucleate in
several sites with a shorter time for growing in the
372 P. B. P. LEÃO ET AL.
Figure 7. Effective activation energy for recrystallisation (Ex)of LCM steel as a function of recrystallised fraction for the investigated
cold-rolled conditions.
Figure 8. Correlation of parameter data from continuous annealing process: (a) average strip speed as a function of strip thickness
and (b) strip speed and soaking temperature as a function of yield strength.
MATERIALS SCIENCE AND TECHNOLOGY 373
thinnerstrips.Incontrast,ithasfewersitesfornucle-
ating and a longer time for growing in the thicker
strips. Thus, these strips may have dierent nal ferritic
grains, even after CAP at the same soaking temperature,
resulting in the YS variation, as in the 50% and 70%
cold-rolled steels investigated in this work.
Furthermore, it is also possible to observe in Figure
8(b) a strong eect on YS by varying the soaking
Figure 9. (a) Yield strength and ferritic grain size, (b) JMatPro®simulation of austenite formation at holding times of 50 and 70 s and
(c) anisotropic properties (planar and normal) for the 50% cold-rolled strip as a function of soaking temperature.
374 P. B. P. LEÃO ET AL.
temperature at a determined speed. It means almost
constant thickness or slight divergence in the degree
of cold-rolling deformation. Therefore, considering
the YS adjustment for a xed thickness and the link
between the strip speed and its respective thickness, it is
evident that the soaking temperature is the controllable
parameter during CAP to control YS.
YS adjustment and anisotropy properties for the
50% cold-rolled LCM steel
Considering only the 50% cold-rolled LCM steel, Figure
9shows the eect of the soaking temperature of CAP
on its YS, ferritic grain size, percentage simulated of
austenite formed during intercritical temperature and
anisotropy properties. It can be seen in Figure 9(a)
that the grain size and YS are inversely proportional,
which is consistent with the Hall-Petch relationship
[32]. Also, no deformed ferrite grains were found in the
50% cold-rolled LCM steel microstructures after CAP.
In contrast, it consisted of equiaxed ferrite grains with
some spheroidised cementite in their grain boundaries.
It suggests that the complete ferrite recrystallisation
occurred before austenite formation [6,33]. Further-
more, the normal distributions in Figure 9(a) depict
that the central position (average of 375.93 MPa) of the
YS aimed range (340–420 MPa) was achieved by using
an annealing temperature of 760°C. However, to min-
imiseeortsinthedyeduringthesubsequentstamping
process, it is essential to target the YS in the lower limit.
So, an average YS of 351.07 MPa, close to the YS min-
imumlimit,wasobtainedfor51coilswithanaverage
soaking temperature of 772.7°C.
In addition, simulation of phase transformation
using JMatPro®software revealed a strong dependence
of the soaking temperature on the amount of austen-
ite formed during intercritical annealing, as shown
in Figure 9(b). This temperature dependence may be
explained by the diusion phenomenon that controls
the growth of austenite nuclei [6,7,33]. This suggests
that the nal ferritic grain size is mainly inuenced
by the amount of austenite formed during intercritical
annealing. Furthermore, the variation in holding time
during the manufacture of the 50% cold-worked LCM
steel in CAP is short (about 50–70 s), and the amount
of austenite shown in Figure 9(b) was the same in the
range of 50–70 s.
After CAP, the investigated steel is shaped into safety
parts of car bodies (such as centre pillar reinforcement
and inner lock pillar) via the conventional/deep draw-
ingprocess.Inthissetting,theanisotropyproperties
of the 50% cold-worked LCM steel, after CAP, are pre-
sented in Figure 9(c). These properties did not show
a noticeable variation between the investigated soak-
ing temperatures. The planar anisotropy (r) values
ranged from 0.166 to −0.001, which are near zero,
meaning that the material has a low tendency for ear-
ing during the drawing process. For normal anisotropy
(¯
r), which represents the resistance of the strip for thin-
ning during plastic deformation, the ¯
r-values ranged
from 0.97 to 1.13. It is worth realising that the smaller
¯
r-value found is close to 1, which is a reference for
isotropic ow strengths in the strip. Therefore, it is rea-
sonabletoconsiderthattheformabilitypropertiesof
the 50% cold-rolled steel, after CAP, are employable for
the conventional drawing process [34,35].
Conclusions
Inthiswork,anindustrialcasestudywasconsidered.
After a similar continuous annealing process (CAP) at
790°C, the 50% and 70% cold-rolled LCM steel pre-
sented, respectively, YS below and YS on the aimed
target (340-420 MPa). Therefore, ferritic recrystallisa-
tion kinetics of the 50% and 70% cold-rolled LCM steel
were investigated via DSC and model-free FDIM. Sub-
sequently, CAP parameter data were evaluated. Finally,
grain size, YS, normal, and planar anisotropy were mea-
sured after CAP for the 50% cold-rolled LCM steel. The
conclusions are as follows:
(1) The model-free FDIM proved to be a feasible
way to predict the ferritic recrystallisation kinetics
from experimental DSC data regarding dierent
degrees of cold-rolled deformation of LMC steel.
(2) The ferritic recrystallisation kinetics proles deter-
mined via model-free FDIM showed dierent
behaviour for the two cold-rolling conditions stud-
ied. The less deformed steel exhibited slower
nucleation and faster grain growth rate, result-
inginatrendtowardsalargerrecrystallised
ferrite nal grain size. In contrast, the higher
cold-deformed material exhibited the opposite
behaviour. This comportment could be explained
by both deformed states’ initial microstructure and
crystallographic orientation.
(3) The evaluated CAP parameters showed a linear
correlation between the speed and the thickness
of the strips due to their weight. The intercritical
soaking temperature was the dominant parameter
under YS for a given thickness.
(4) The arrangement between the information from
ferrite recrystallisation kinetics and the CAP
parameter assessments contributed to understand-
ing the YS divergence of manufactured strips
in dierent thicknesses and at a given soaking
temperature.
(5) The appropriate YS for the 50% cold-rolled con-
tinuously annealed steel was achieved by adjust-
ing the soaking temperature, which signicantly
aected the ferritic nal grain size. Otherwise, the
anisotropy properties showed no signicant dier-
ence with the variation of soaking temperature.
MATERIALS SCIENCE AND TECHNOLOGY 375
Acknowledgments
The authors acknowledge the following facilities of Uni-
versidade Federal do Ceará; Laboratório de Caracteri-
zação de Materiais (LACAM), Laboratório de Fundição
(LaF), and Central Analítica UFC/CT-INFRA FINEP/Pro-
Equipamentos-CAPES/CNPq-SisNano-MCTI 2019 (Grant
442577/ 2019-2)-INCT-FUNCAP for providing support in
this work.
Disclosure statement
No potential conict of interest was repor ted by the author(s).
Funding
This study was nanced in part by the Coordenação de Aper-
feiçoamento de Pessoal de Nível Superior – Brasil (CAPES) –
Finance Code 001.
References
[1] Lesch C, Klose FB. Advanced high strength steels
(AHSS) for automotive applications – tailored proper-
ties by smart microstructural adjustments. Steel Res Int.
2017;38:1–21. DOI: 10.1002/srin.201700210.
[2] Baker TN. Titanium microalloyed steels. Ironmak
Steelmak. 2019;46:1–55. DOI: 10.1080/03019233.2018.
1446496.
[3] Humphreys J, Rohrer SG, Rollett A. Recrystallization
and related annealing phenomena recrystallization and
related annealing phenomena. Amsterdam: Elsevier;
2017.
[4] Martínez-de-Guerenu A, Jorge-Badiola D, Gutiérrez I.
Assessing the recovery and recrystallization kinetics of
cold rolled microalloyed steel through coercive eld
measurements. Mater Sci Eng A. 2017;691:42–50. DOI:
10.1016/j.msea.2017.03.033.
[5] Toji Y, Yamashita T, Nakajima K, et al. Eect of Mn
partioning during intercritical annealing on follow-
ing γ→αtransformation and resultant mechanical
properties of cold-rolled dual phase steels. ISIJ Int.
2011;51:818–825. DOI: 10.2355/isijinternational.51.818.
[6] Chbihi A, Barbier D, Germain L, et al. Interac-
tions between ferrite recrystallization and austen-
ite formation in high-strength steels. J Mater Sci.
2014;49:3608–3621. DOI: 10.1007/s10853-014-8029-2.
[7] Valdes-Tabernero MA, Celada-Casero C, Sabirov I,
et al. The eect of heating rate and soaking time
on microstructure of an advanced high strenght steel.
Mater Caract. 2019;155:109822. D OI: 10.1016/j.matchar.
2019.109822.
[8] Muljono D, Ferry M, Dunne DP. Inuence oh heating
rate on anisothermal recrystallization in low and ultra-
low carbon steels. Mater Sci Eng A. 2001;303:90–99.
DOI: 10.1016/S0921-5093(00)01882-7.
[9] Ferry M, Muljono D, Dunne DP. Recrystallization
kinetics of low and ultra low carbon steels during
high-rate annealing. ISIJ Int. 2001;41:1053–1060. DOI:
10.2355/isijinternational.41.1053.
[10] Torres RCE, Sánchez FH, González A, et al. Study of
the kinetics of the recrystallization of cold rolled low-
carbon steel. Metall Mater Trans A. 2002;33A:25–31.
DOI: 10.1007/s11661-002-0003-y.
[11] Shu Q, Klug JL, Medeiros SLS, et al. Crystallization con-
trol for uorine-free mold uxes: eect of Na2Ocontent
on non-isothermal melt crystallization kinetics. ISIJ Int.
2020;60:2425–2435. DOI: 10.2355/isijinternational.ISI
JINT-2020-132.
[12] Simon P. Isoconversional methods. J Therm Anal
Calorim. 2004;76:123–132. DOI: 10.1023/B:JTAN.000
0027811.80036.6c.
[13] Criado JM, Sánchez-Jiménez PE, Pérez-Maqueda LA.
Critical study of the isoconversional methods of kinetic
analysis. J Therm Anal Calorim. 2008;92:199–203. DOI:
10.1007/s10973-007-8763-7.
[14] Friedman HL. Kinetics of thermal degradation of char-
forming plastics from thermogravimetry. Application to
aphenolicplastic.JPolymSciC.1964;6:183–195. DOI:
10.1002/polc.5070060121.
[15] Lad KN, Savalia RT, Pratap A, et al. Isokinetic
and isoconversional study of crystallization kinet-
ics of a Zr-based metallic glass. Thermochim Acta.
2008;473:74–80. DOI:10.1016/j.tca.2008.04.011.
[16] Liu Q, Qian D, Hua L. Transformation from non-
isothermal to isothermal tempering of steel based on
isoconversional method. J Mater Sci. 2018;53:2774–
2784. DOI: 10.1007/s10853-017-1680-7.
[17] Sente Software Ltd. (2021). JMatPro®[Internet] Guild-
ford: Surrey Research Park [cited 2021 Dec 8]. Available
from: https://www.sentesoftware.co.uk/.
[18] Bourezg YI, Elad D, Azzeddine H, et al. Investiga-
tion of recrystallization kinetics in hot-rolled Mg-La
alloy using dierential scanning calorimetry technique.
Thermochim Acta. 2020;690. DOI: 10.1016/j.tca.2020.
178688.
[19] Ray RK, Jonas JJ, Guillén MP, et al. Transforma-
tion textures in steels. ISIJ Int. 1994;34:927–942. DOI:
10.2355/isijinternational.34.927.
[20] Tóth LS, Jonas JJ, Daniel D, et al. Development
offerriterollingtexturesinlow-andextralow-
carbon steels. Metall Trans. 1990;21A:2985–3000. DOI:
10.1007/BF02647219.
[21] Every R, Hatherly M. Oriented nucleation in low-
carbon steels. Texture Stress Microstruct. 1974;1:183–
194. DOI: 10.1155/TSM.1.183.
[22] Choi S-H, Cho JH. Primary recrystallization mod-
elling for interstitial free steels. Mater Sci Eng A.
2005;405:86–101. DOI: 10.1016/j.msea.2005.05.093.
[23] Rajmohan N, Hayakawa Y, Szpunar JA, et al. Neutron
diraction method for stored energy measurement in
interstitial free steel. Acta Mater. 1997;45:2485–2494.
DOI: 10.1016/S1359-6454(96)00371-0.
[24] Diehl M, Kertsch L, Traka K, et al. Site-specic quasi in
situ investigation of primary static recrystallization in a
low carbon steel. Mater Sci Eng A. 2019;755:295–306.
DOI:10.1016/j.msea.2019.02.032.
[25] Oyarzabal M, Guerenu A, Gutiérrez I. Eect of stored
energy and recovery on the overall recrystallization
kinetics of a cold rolled low carbon steel. Mater Sci
Eng A. 2008;485:200–209. DOI: 10.1016/j.msea.2007.
07.077.
[26] Alaneme KK, Okotete EA. Recrystallization mecha-
nisms and microstructure development in emerging
metallic materials: A review. J Sci-Adv Mater Dev.
2019;4:19–33. DOI: 10.1016/j.jsamd.2018.12.007.
[27] Seo M-D, Shi C-B, Wang H, et al. Non-isothermal
melt crystallization of cuspidine in CaO–SiO2–CaF2
based glasses. J Non-Cryst Solids. 2015;412:58–65. DOI:
10.1016/j.jnoncrysol.2015.01.008.
[28] Umar M, Qayyum F, Farooq MU, et al. Investigating
the eect of cementite particles size and distribution
on local stress and strain evolution in spheroidized
medium carbon steel using crystal plasticity-based
376 P. B. P. LEÃO ET AL.
numerical simulations. Steel Res Int. 2021;92:2000407.
DOI: 10.1002/srin.202000407.
[29] Qayyum F, Guk S, Prahl U. Studying the dam-
age evolution and the micro-mechanical response of
X8CrMnNi16-6-6 TRIP steel matrix and 10% zirco-
nia particle composite using a calibrated physics and
crystal-plasticity-based numerical simulation model.
Crystals (Basel). 2021;11:759–778. DOI: 10.3390/cryst
11070759.
[30] Umar M, Qayyum F, Farooq MU, et al. Quali-
tative investigation of damage initiation at meso-
scale in Spheroidized C45EC steels by using crystal
plasticity-based numerical simulations. J Compos Sci.
2021;5:222–240. DOI: 10.3390/jcs5080222.
[31] Saito G, Zhang T, Sakaguchi N, et al. In-situ observa-
tion of abnormal grain growth in a low-alloyed carbon
steel using SEM-EBSD. Mater. 2021;15:100985. DOI:
10.1016/j.mtla.2020.100985.
[32] Naik SN, Walley SM. The Hall-Petch and inverse
Hall-Petchrelationsandthehardnessofnanocrys-
talline metals. J Mater Sci. 2020;55:2661–2681. DOI:
10.1007/s10853-019-04160-w.
[33] Garcia CI, Deardo AJ. Formation of austenite in
1.5 Pct Mn steels. Metall Trans A. 1981;12:521–530.
DOI:doi.org/10.1007/BF02648551.
[34] Riva R, Mapelli C, Venturini R. Eect of coiling temper-
atureonformabilityandmechanicalpropertiesofmild
low carbon and HSLA steels processed by thin slab cast-
ing and direct rolling. ISIJ Int. 2007;47:1204–1213. DOI:
10.2355/isijinternational.47.1204.
[35] Mishra S, Darmann C. Role and control of texture
in deep-drawing steels. Int Met Rev. 1982;27:307–320.
DOI: 10.1179/imr.1982.27.1.307.
