AISTech 2019 — Proceedings of the Iron & Steel Technology Conference

6–9 May 2019, Pittsburgh, Pa., USA

DOI 10.1000.377.206

Understanding the Role of Nb in Hot-Rolled Rebars: Influence of Final Geometry and Process
Parameters

Felipe Bastos2,3, Beatriz Pereda1,2, Beatriz Lopez1,2, Jose M. Rodriguez-Ibabe1,2, Marcelo Rebellato
1
Ceit
P. M. Lardizabal 15, 20018 San Sebastian, Spain
Phone: +34 943 212 800
Email: [email protected]
2
Universidad de Navarra, Tecnun
P. M. Lardizabal 13, 20018 San Sebastian, Spain.
3
CBMM
Av. Brigadeiro Faria Lima, 4285, 9th, 04538-133 São Paulo / SP, Brazil.
4
RMS
Rua Doutor Diogo de Faria, 1202, cj. 53, 04037-004 São Paulo / SP, Brazil.

Keywords: Nb microalloyed steels, Bar hot rolling, Dynamic recrystallization, Phase transformation

INTRODUCTION

Microalloying has been widely studied in flat products but not so much in medium carbon long products. In flat steel
production, Nb microalloying is mainly used due to the large retarding effect that it exerts on recrystallization kinetics
during hot deformation, which can be due to solute drag or to strain-induced precipitation effects [1]. This results in pancaked
austenite microstructures before phase transformation, and therefore, in refined room temperature microstructures. As well
as during thermomechanical processing (TMCP), Nb can also exert some effect during phase transformation, due to
enhanced hardenability or to the formation of nanometer sized precipitates [2]. However, the role of Nb in hot rolled bars is
not so well known. In bar rolling, much higher strain and strain-rates, and lower interpass times are applied compared to flat
product rolling. This can modify the role of Nb on the austenite microstructural evolution. To investigate this, in this work,
multipass torsion tests were performed simulating hot rolling of 8 mm and 20 mm rebars. The tests were performed with
two medium carbon steels (0.24%C), one of them microalloyed with 0.03% Nb. Two different reheating temperatures, 1100ºC
and 1220ºC, were considered, and both the austenite and the room temperature microstructures obtained after phase
transformation were analyzed.
A model for the prediction of the austenite microstructural evolution during hot deformation (MicroSim-Bars©) which takes
into account the singularities of these processes has been developed. In addition, the continuous cooling transformation curves
(CCT) were determined for the Nb microalloyed steel by dilatometry at different previous austenite grain size and Nb
dissolution conditions.

MATERIALS AND EXPERIMENTAL PROCEDURE

The composition of the medium carbon steels (0.25%C) investigated is shown in Table 1. As it can be observed from the
table, one of them is a reference CMn steel, while the other one is microalloyed with 0.03%Nb.

C Table 1.
Mn Chemical
Si composition
Nb ofCr
the steels
Cuinvestigated
Ni (weightNpercent). P S
CMn 0.25 0.55 0.23 - 0.2 0.2 0.15 0.010 0.018 0.017
CMnNb 0.24 1.25 0.50 0.03 - - - 0.006 0.030 0.027

© 2019 by the Association for Iron & Steel Technology. 2001

In the multipass torsion tests, the specimens consisted of a gauge geometry of 16.5 mm in length and 7.5 mm in diameter. In
the tests, the samples were first reheated for 20 min at soaking temperatures of 1100ºC or 1220ºC. Next, deformation sequences
representative of hot rolling of rebars of 8 and 20 mm diameter were applied, based on industrial data from several producers.
The deformation temperatures, cooling rates and interpass times (tip) corresponding to the 1100ºC reheating condition are
summarized in Figure 1. In the 1220ºC sequences, all these parameters were kept constant, except for the temperatures which
were in all cases 50ºC higher. A schematic representation of the cycle is also included in the figure. As it can be observed, after
reheating, 24 and 16 deformation passes were applied for the ‡8 and 20 mm sequences, respectively. In the tests, for simplicity,
the deformation per pass (H=0.45) and the strain rate (ߝሶ=5 s-1) were kept constant. As the figure shows, during the first
deformation passes, the temperature decreased, first at 5ºC/s (pass 1 to 7) and next at 2ºC/s (pass 8 to 11). On the other hand,
during the last passes, the specimens were heated at 2ºC/s (pass 12 to 15) or 10ºC/s (9 last deformation passes of the ‡8 mm
sequence) to simulate the adiabatic heating that takes place during the last passes of industrial bar hot rolling. Decreasing
interpass times of 5, 2 and 0.5 s were applied during the different cooling and heating stages. After the tests, specimens were
water quenched to analyze the austenite microstructure after deformation and holding time to rolling exit, or cooled down at
different rates to simulate air-cooling.

Cooling 8 mm 20 mm Pass Cooling 8 mm 20 mm

Pass rate T tip T tip rate T tip T tip
(ºC/s) (ºC) (s) (ºC) (s) (ºC/s) (ºC) (s) (ºC) (s)
1 5 1080 5 1080 5 12 -2 889 2 889 2
2 5 1055 5 1055 5 13 -2 893 2 893 2
3 5 1030 5 1030 5 14 -2 897 2 897 2
4 5 1005 5 1005 5 15 -2 901 2 901 2
5 5 980 5 980 5 16 -10 905 0.5 905 5.5
6 5 955 5 955 5 17 -10 910 0.5
7 5 930 5 930 5 18 -10 915 0.5
8 2 905 2 905 2 19 -10 920 0.5
9 2 901 2 901 2 20 -10 925 0.5
10 2 897 2 897 2 21 -10 930 0.5
11 2 893 2 893 2 22 -10 935 0.5
23 -10 940 0.5
24 -10 945 3.5
Texit 980 960

Reheating temperatures: 1100ºC and 1220ºC

Specimens
Reheating, 20 min quenched after last
T start
deformation: deformation pass +
1130/1080ºC holding time to exit
Temperature

7 passes
tip=5 s Air cooling
(cooling rate of 5 ºC/s) simulations

4 passes 9 passes
tip=2 s 4 passes tip=0.5 s
(cooling rate of 2 ºC/s) tip=2 s (heating rate of
. (heating rate of 2 10 ºC/s)
H=5s-1
H per pass=0.45 ºC/s) HEATING HEATING

Time

Figure 1. Temperature and interpass times (tip) employed in the tests carried out with a reheating temperature of 1100ºC and
schematic representation of the cycle.

2002 © 2019 by the Association for Iron & Steel Technology.

A scheme of the thermal cycle applied in the dilatometry tests to construct the continuous cooling transformation (CCT) curves
for the CMnNb steel is shown in Figure 2. First, a reheating treatment was applied at 1050ºC or 1250ºC followed by water
quenching, with the aim of obtaining different amounts of Nb in solid solution. Next, the specimens were again reheated at two
temperature ranges to obtain austenite grain sizes of #6-7 μm (T conditioning 820ºC-860ºC), or of #20 μm (T conditioning
975ºC). After this, the specimens were cooled at different rates from 0.1ºC/s to 20ºC/s to room temperature. The phase fractions
were determined using the lever rule.

Reheating:
1050ºC-1250ºC,
2 hours

Temperature Austenite
conditioning: 820ºC-
860ºC, 975ºC

Cooling
Quenching rates: 0.1, 1,
2, 5, 10,
20ºC/s

Time
Figure 2. Thermal cycles applied in the dilatometry tests

Metallographic measurements of the torsion specimens were carried out in a section corresponding to 0.9 of the outer radius of
the torsion specimen, also known as the sub-surface section [3]. In the of dilatometry specimens, a region corresponding to a
maximum area fraction of nominal strain and reduced strain gradient was selected [4]. Bechet-Beaujard etching [5] was applied
for revealing the previous austenite grain boundaries in the quenched samples, while 2%Nital was used in the transformed
microstructures. The etched samples were examined via optical microscopy, and in some cases, carbon extraction replicas were
prepared and analyzed by transmission electron microscopy (TEM) to characterize their precipitation state.

RESULTS AND DISCUSSION

Stress-Strain behavior
Figure 3 shows examples of the stress-strain curves determined for both steels in the ‡8 mm rebar simulations at 1220ºC
reheating condition. The different cooling rate/interpass time ranges have also been marked in the figures. During the first
seven deformation passes, for both steels, the stress increases due to temperature decrease. However, for the Nb steel, after
approximately pass four, (T#1050ºC) a higher stress increase rate is appreciated. This is in accordance with the behavior usually
reported for Nb microalloyed steels in decreasing temperature deformation schedules. It is well known that Nb leads to
softening kinetics retardation, due to solute drag or strain-induced precipitation effect [6]. As a result, below a temperature
known as non-recrystallization temperature (Tnr) [7], softening between deformation passes is suppressed, leading to strain
accumulation and higher stress levels, as observed in stage 1 for the CMnNb steel. However, it is interesting to note that when
the interpass time decreases to 2 s (stage 2), although the temperature continues decreasing, some stress stabilization or even
decrease is observed for both steels. This indicates that some interpass softening takes place during this period. In stage 3 (finish
rolling stage for ‡20 mm simulation), the temperature increases and this makes difficult the interpretation of the stress-strain
behavior. On the other hand, for both steels, during the second pass of stage 4 (last rolling passes of the ‡8 mm simulation,
tip=0.5 s) the stress slightly increases with respect to the previous pass, but then decreases during the subsequent deformation
passes. This indicates that strain accumulation takes place during the first deformation pass while some softening takes place
during subsequent deformation passes. This shape is characteristic of dynamic recrystallization activation [8].
The mean flow stress (MFS), defined as the area under the stress-strain curve divided by the pass strain, was determined and it
has been represented in Figure 4 for all the conditions investigated. During the first deformation passes, similar MFS are
calculated for the CMn and CMnNb steels. However, for both reheating temperatures, after pass #4 larger increase in the MFS
is observed for the CMnNb steel. This difference remains approximately constant or slightly increases during the rest of the
tests. Based on the standard method [7], from the change of MFS increase rate the Tnr values can be estimated. As shown in the

© 2019 by the Association for Iron & Steel Technology. 2003

figure for the CMnNb steel, Tnr values of #1005ºC and of 1055ºC for 1100ºC and 1220ºC reheating temperatures were
determined, respectively. However, when the interpass time decreases to 2 s (stage 2), some stress decrease or stabilization is
observed for all steels and reheating conditions investigated. As mentioned above, since the temperature is increasing in this
stage, the behavior is indicative of some softening during the interpass time. It is also interesting to note that, in stage 4 (tip=0.5
s), although higher stress was determined for the Nb steel, the mean flow stress (MFS) follows the same trends for both steels
and reheating conditions analyzed: it increases during the second pass and then it decreases during the last deformation passes.

250
CMn, 8 mm, 1220ºC
a)
1: tip = 5 s, 'T/'t=-5ºC/s
200 3 2: tip = 2 s, 'T/'t=-2ºC/s
2 3: tip = 2 s, 'T/'t=2ºC/s
1 4: tip= 0.5 s, 'T/'t=10ºC/s
150 4
Stress (MPa)

100

50

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Strain

250
b) CMnNb, 8 mm, 1220ºC
2 3
200
1 4
Stress (MPa)

150

100

50

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Strain
Figure 3. Stress-strain curves obtained in the 8 mm – 1220ºC reheating temperature tests.

The austenite microstructures, characterized from specimens quenched after the deformation sequences, are presented in Figure
5. It can be observed that for both steels, fine and equiaxed microstructures are obtained at all the conditions analyzed. As
above mentioned, in other rolling configurations Nb is added mainly to promote austenite pancaking during hot deformation.
However, the austenite microstructures obtained in these sequences indicate that this role is not applicable for the current rolling
schedules, and that regardless reheating temperature or rolling sequence, recrystallization takes place during the last
deformation passes of the torsion tests. Therefore, this indicates that these routes does not necessarily need the application of
classical TMCP. Nevertheless, some effect of Nb microalloying is observed. For both sequences at 1220ºC reheating
temperature, and at 1100ºC for the ‡20 mm sequence, significant austenite grain refinement is observed for the CMnNb steel
compared to the CMn one. In addition, it can be observed that for the CMn steel, increasing reheating temperature leads to
coarser microstructures. This can be attributed to the higher deformation temperatures applied in this sequence. On the other

2004 © 2019 by the Association for Iron & Steel Technology.

hand, for the Nb steel this is not observed, showing that for this steel the microstructure is less dependent on process parameter
modification.

200 200
a) Reheating 1100ºC 1: tip = 5 s, 'T/'t=-5ºC/s b) Reheating 1220ºC
180 2: tip = 2 s, 'T/'t=-2ºC/s 180 2
3: tip = 2 s, 'T/'t=2ºC/s 3
1 4: tip= 0.5 s, 'T/'t=10ºC/s
160 160 1 4
140 140
MFS (MPa)

MFS (MPa)
Tnr#1000ºC
Tnr#1050ºC
120 2 3 4 120

100 CMn, 8 mm 100 CMn, 8 mm

CMn, 20 mm CMn, 20 mm
80 CMnNb, 8 mm 80 CMnNb, 8 mm
CMnNb, 20 mm CMnNb, 20 mm
60 60
0 5 10 15 20 25 3 0 5 10 15 20 25 3
Pass number Pass number

Figure 4. MFS determined for both steels at: (a) 1100ºC and (b) 1220ºC reheating temperatures.

Reheating at 1100ºC:
8 mm 20 mm

50 μm 50 μm 50 μm 50 μm

CMn, D=15.2±0.5 μm CMnNb, D=15.8±0.5 μm CMn, D=17.4±0.7 μm CMnNb, D=13.4±0.7 μm

Reheating at 1220ºC:
8 mm 20 mm

50 μm 50 μm 50 μm 50 μm
CMn, D=21.2±1 μm CMnNb, D=14.4±0.5 μm CMn, D=22.3±0.6 μm CMnNb, D=12.1±0.3 μm

Figure 5. Austenite microstructures obtained from quenched specimens.

Carbon extraction replicas were prepared from the ‡8 mm CMnNb quenched samples to investigate their precipitation state.
Figure 6 shows some examples of the precipitates observed for both reheating conditions. In the micrograph from the 1220ºC
condition, some coarse particles that could be precipitates that remain undissolved during reheating can be observed. However,
more abundantly, particles smaller than 15 nm can be detected, which indicates that some strain-induced precipitation has taken
place during the sequence. In the 1100ºC sample, the amount of undissolved precipitates was larger and their size coarser. As
shown in Figure 6, some small Nb particles could also be detected in this sample although much less frequently than at 1220ºC.

© 2019 by the Association for Iron & Steel Technology. 2005

It is well known that Nb strain-induced precipitation can stop or greatly retard static recrystallization [6, 9]. This is in good
agreement with the detection of a Tnr temperature during stage 1 of the deformation schedule, where an interpass time of 5 s
was applied. However, when the interpass time decreases to 2 s or 0.5 s, the stress-strain behavior and the microstructures
obtained after the cycles indicate that during these stages precipitates are not able to stop recrystallization and, independently
of the reheating temperature considered, equiaxed microstructures are obtained for both steels. It is also worth noting the small
size of some of these precipitates (<10 nm). This means that if precipitate growth rate is not high during phase transformation,
they could provide some contribution to mechanical strength due to precipitation hardening.

Nb
Nb

Ni
Ni

Ni

Ni
Nb

0 2 4 6 8 10

(a) Reheating 1220ºC

O
Ni Ni
Ni
Cu

Nb
Nb

Cu
Cu
Ni
Nb Cu

0 2 4 6 8 10

(b) Reheating 1100ºC

Figure 6. Precipitates detected in the carbon replicas extracted from 8 mm quenched specimens.

In the frame of this work, a microstructural evolution model, MicroSim-Bars© based on the approaches described in references
[6] and [10], with improved equations to consider the singularities observed in these processes, has been developed. Examples
of the predictions of the model for the ‡8 mm – 1220ºC sequences for both steels are shown in Figure 7. It can be observed
that for the CMn steel, the predicted final austenite grain size agrees well with the measured values (23 μm model vs. 21.2 μm
experimental). For the CMnNb steel, this is slightly overestimated (18.3 μm model vs. 14.4 μm experimental), although the
austenite refining effect that it is observed experimentally with Nb addition is well captured.
It can be observed that during the first passes, for both steels static recrystallization is predicted to occur and 100% recrystallized
fractions are predicted between passes. However, for the CMnNb steel, after 4th pass, where the Tnr is detected, the recrystallized
fraction starts to decrease, first due to solute drag effect, and next due to the occurrence of strain-induced precipitation. This is
in good agreement with the experimental TEM results, which show the presence of some precipitation at the end of the sequence
(Figure 6(a)). However, for both steels, during the last passes, when the interpass time decreases, the model predicts the
activation of dynamic recrystallization during deformation and within the interpass times, metadynamic recrystallization is the
main softening process. This agrees well with the equiaxed microstructures observed for both steels at the end of the sequences.

2006 © 2019 by the Association for Iron & Steel Technology.

 (a) CMnNb steel

(b) CMn Steel

Figure 7. Predictions of MicroSim-Bars© program for the austenite microstructural evolution during the 1220ºC-8 mm hot
rolling sequences.

Torsion tests were also performed simulating final air cooling conditions. The cooling rates were calculated for the different
bar diameters using a temperature model. The model predicted two approximately linear cooling ranges: the first one from exit
temperature to 800ºC (#14ºC/s and 5ºC/s for ‡8 and 20 mm respectively), and a final slower one to 500ºC (#5ºC/s and 2ºC/s
for 8 and 20 mm, respectively). It can be observed from Table 1 that for the CMn steel, the Mn content (0.55%) is significantly
lower than for the CMnNb steel (1.25%). While this difference is not expected to affect significantly the microstructural
evolution during hot deformation, it can have large effect on the microstructure that it is obtained after phase transformation.
Therefore, due to the lack of comparability, simulations were only performed with the CMnNb steel. Figure 8 shows optical
micrographs of the microstructures obtained, together with phase fraction and hardness measurements. For the ‡20 mm

© 2019 by the Association for Iron & Steel Technology. 2007

simulation samples, ferritic-pearlitic microstructures that are slightly coarser than for ‡8 mm can be observed. At this
condition, slightly higher hardness and ferrite fraction is determined for the 1220ºC reheating temperature. As well as the ferrite
fraction, the ferrite grain size is also slightly larger in this sample. Therefore, this suggests some contribution of precipitation
hardening to the mechanical strength for this specimen. For the 8 mm sequences, the microstructures are more acicular and in
addition, in the case of the 1220ºC sample, a small amount of bainite is detected (#4%). The highest hardness value was also
determined for this sample.

Reheating at 1100ºC: Reheating at 1220ºC:

‡ 8mm ‡ 20mm ‡ 8mm ‡ 20mm

20 μm 20 μm 20 μm 20 μm

73.6%α + 26.4%p ; dα = 7.4 μm; 68.3%α + 31.7%p; dα = 8.6 μm; 71.5%α + 24.5%p + 77.9%α + 22.1%p; dα = 9.1 μm;
182 HV 187 HV 4.1%bainite; dα = 8.0 μm; 199 HV
203 HV

Figure 8. Transformed microstructures obtained from the specimens after air cooling simulation.

CCT Curves
To further study the phase transformation behavior, CCT diagrams were constructed for the CMnNb steel. Different conditions
were applied (Figure 2) to produce three initial austenite microstructures before phase transformation: reheating at 1200ºC and
low conditioning temperature to produce a fine initial austenite grain size (6.3 μm) or reheating at 1050ºC and low and high
austenite conditioning temperatures (austenite grain sizes of 6.8 μm and 20.2 μm, respectively). Trials were also performed
using higher conditioning temperatures after 1200ºC reheating to produce a coarser grain size. However, abnormal austenite
grain growth took place, and due to this, it was not possible to produce comparable homogeneous austenite microstructures
with #20 Pm grain size. In Figure 9, the CCT diagrams determined for the three conditions are compared. It can be observed
that increasing reheating temperature and, therefore, the amount of Nb in solid solution, does not lead to large differences in
the transformation start temperatures. On the other hand, except for 0.1ºC/s, increasing the initial grain size results in a decrease
of the temperature of approximately 20º to 70ºC. Some decrease of the transformation finish temperature is also observed,
although mainly for the highest cooling rates.

1000
1200ºC, 860ºC, 6 microns
900 1050ºC, 820ºC, 7 microns
1050ºC, 975ºC, 20 microns
800
F
700
Temperature (⁰C)

P
600 B

500

400 Ms = 375ºC

300

200
⁰C/s 20 10 5 2 1 0.1
100

0
1 10 100 1000 10000
Time (s)

Figure 9. CCT (continuous cooling transformation) curves determined for the CMnNb steel at different reheating and initial
grain size conditions.

2008 © 2019 by the Association for Iron & Steel Technology.

Several examples of the transformed microstructures obtained at constant reheating conditions (1050ºC) and different initial
austenite grain sizes are included in Figure 10. The transition from ferritic-pearlitic microstructures to more acicular phases can
be observed in both cases. However, it is evident from the figure that for the coarse austenite grain size, the appearance of
bainite is significantly enhanced. For this condition, some bainite is observed even at the relatively low 2ºC/s cooling rate and
it clearly increases with increasing cooling rate. On the other hand, for the small initial grain size condition, a significant ferrite
fraction is maintained on the microstructure even for 10ºC/s cooling rate.

Reheating 1050ºC, 20 Pm initial austenite grain size:

1ºC/s 2ºC/s 5ºC/s 10ºC/s

50 P m 50 Pm 50 Pm 50 Pm

Reheating 1050ºC, 7 Pm initial austenite grain size:

1ºC/s 2ºC/s 5ºC/s 10ºC/s

50 Pm 50 Pm 50 Pm 50 Pm

Figure 10. Transformed microstructures observed for the CMnNb steel after 1050ºC reheating and different initial austenite
grain sizes of #7 and 20 Pm.

The phase volume fractions measured by quantitative metallography from the different microstructures are summarized in
Figure 11. It can be observed that for the lowest cooling rates (0.1ºC/s – 1ºC/s) ferritic-pearlitic microstructures are obtained in
all the cases. However, as shown above, at 2ºC/s, for the low reheating temperature and coarse initial austenite grain size
condition, some bainite appeared. In addition, for 5ºC/s this increases drastically to 66%. On the other hand, at these cooling
rates, for the small austenite grain size samples bainite is not detected or the fraction is much more limited. Within the low
initial grain size specimens, large bainite fractions can only be detected at 20ºC/s cooling rate, and more abundantly for the
1200ºC reheating condition. It can be noted that at 20ºC/s, some martensite was also present in the 20 μm initial austenite grain
size sample.
The ferrite grain sizes and hardness values measured from the transformed specimens are summarized in Figure 12. Figure 12
(a) shows that decreasing the initial austenite grain size leads to significant ferrite grain size refinement, as usually reported
[11]. For the fine initial grain size samples, similar ferrite grain sizes were determined for cooling rates from 1 to 10ºC/s (3.0
Pm to 4.5 Pm), although for 2ºC/s and 1ºC/s they tend to be slightly coarser for the high reheating temperature condition. A
slight grain size increase with decreasing cooling rate to 0.1ºC/s is observed (5.2 Pm). On the other hand, for the 20 μm austenite
grain size, the ferrite grain size increase at 0.1ºC/s is much more marked (from 7.2 Pm to 12.8 Pm). Regarding hardness, for
the lowest cooling rates of 0.1ºC/s and 1ºC/s, these tend to be higher for the samples reheated at 1200ºC. At intermediate cooling
rates of 2ºC/s and 5ºC/s the differences are not large. Finally, for the highest cooling rates (10ºC/s-20ºC/s) hardness increases
significantly for the coarse initial austenite grain size condition. This can be attributed to the presence of more acicular phases
(note that some martensite was detected at 20ºC/s).

© 2019 by the Association for Iron & Steel Technology. 2009

 100 0
90 10

Pearlite volume fraction (%)

Acicular ferrite/Bainite

Bainite volume fraction (%)

80 20
70 30
60 40
50 50
40 Pearlite 60
30 70
20 Martensite 80
10 90
0 100
0.1 1 2 5 10 20
Cooling rate (ºC/s)
1200ºC, 6 microns 1050ºC, 7 microns 1050ºC, 20 microns
Figure 11. Phase volume fraction as a function of cooling rate for the different initial conditions investigated.

16 350
a) 1200ºC, 6 microns b) 1200ºC, 6 microns
14 1050ºC 7 microns 1050ºC, 7 microns
300
1050ºC, 20 microns 1050ºC, 20 microns
Ferrite grain size (μm)

12

10 250
HV1
8
200
6

4 150

2
100
0 0.1 1 10
0.1 1 10 Cooling rate (⁰C/s)
Cooling rate (ºC/s)

Figure 12. (a) Ferrite grain size and (b) hardness measured from the different specimens.

SUMMARY AND CONCLUSIONS

-Multipass torsion tests simulating hot rolling of ‡8 mm and ‡20 mm rebars have been performed at two different reheating
conditions (1100ºC and 1220ºC) with two 0.24%C steels, one of them microalloyed with 0.03%Nb. For the Nb microalloyed
steel, during the first deformation passes, the behavior is similar to that observed conventionally for flat products and some
strain accumulation is observed. As a result, Tnr values of #1000ºC and 1050ºC can be determined for reheating temperatures
of 1100ºC and 1220ºC, respectively. However, during the last deformation passes, decreasing interpass time leads to strain
accumulation and to the activation of dynamic recrystallization.

-As a result of this, after the deformation sequences, fine and recrystallized austenite grain sizes were determined for both steels
and all reheating and rolling conditions investigated. However, higher austenite grain size refinement was observed for the
CMnNb steel compared to the CMn one, especially at high reheating temperature and high diameter conditions. Lower
dependence of austenite grain size with processing conditions was also observed for the CMnNb steel.

-In good agreement with the detection of a T nr during the first deformation passes, the occurrence of Nb(C,N) strain-induced
precipitation was detected via TEM analysis. However, these precipitates were not able to stop recrystallization during the last
deformation passes, where the interpass time decreased.

-These trends are well predicted by an austenite microstructural evolution model, MicroSim-Bars©, developed to take into
account the singularities of bar hot rolling.

2010 © 2019 by the Association for Iron & Steel Technology.

-The phase transformation behavior of the CMnNb steel was also investigated as a function of reheating temperature and initial
grain size conditions. Modifying the initial reheating temperature did not affect much the transformation temperatures or the
size or the morphology of the phases developed. However, increasing initial austenite grain size lead to the development of
more bainitic microstructures in the range of cooling rates that can be considered representative of rebar air-cooling conditions.

ACKNOWLEDGEMENTS

The authors would like to gratefully acknowledge Companhia Brasileira de Metalurgia e Mineração (CBMM) for funding this
project.

REFERENCES

1. H.L. Andrade, M.G. Akben, J.J. Jonas, “Effect of Molybdenum, Niobium, and Vanadium on Static Recovery and
Recrystallization and on Solute Strengthening in Microalloyed Steels”, Metall. Trans. A, Vol. 14, 1983, pp. 1967-1977.
2. M.A. Altuna, A. Iza-Mendia, I. Gutierrez, “Precipitation in Nb in Ferrite after Austenite Conditioning. Part II:
Strengthening Contribution in High-Strength Low-Allow (HSLA) Steels, Metall. Mater. Trans. A., Vol. 43, 2012, pp.
4571-4586.
3. G. Glover, C.M. Sellars, “Recovery and Recrystallization during High Temperature Deformation of D-iron”, Met.
Trans., Vol. 4, 1973, pp. 765-775.
4. M. Olasolo, P. Uranga, J.M. Rodriguez-Ibabe, B. Lopez, “Effect of Microstructure and Cooling Rate on Transformation
Characteristics in a Low Carbon Nb-V Microalloyed Steel”, Mater. Sci. Eng. A, Vol. 528, 2011, pp. 2559-2569.
5. S. Bechet, L. Beaujard, “New Reagent for the Micrographical Demonstration of the Austenite Grain of Hardened or
Hardened-Tempered Steels”, Rev. Met., Vol. 52, 1995, pp. 830-836
6. B. Pereda, J.M. Rodriguez-Ibabe, B. Lopez, “Improved Model of Kinetics of Strain Induced Precipitation and
Microstructure Evolution of Nb Microalloyed Steels during Multipass Rolling”, ISIJ Int., Vol. 48, 2008, pp. 1457-1466.
7. D.Q. Bai, S. Yue, W.P. Sun, J.J. Jonas, “Effect of Deformation Parameters on the No-recrystallization Temperature in
Nb Bearing Steels”, Metall. Trans. A, Vol. 24, 1993, pp. 2151-2159.
8. B. Lopez, J.M. Rodriguez-Ibabe, "Recrystallisation and Grain Growth in Hot Working of Steels", in Microstructure
Evolution in Metal Forming Processes, 2012, pp. 67-113.
9. S.S. Hansen, J.B. Vander Sande, M. Cohen, “Niobium Carbonitride Precipitation and Austenite Recrystallisation in
Hot-Rolled Microalloyed Steels”, Metall. Trans. A, 1980, pp. 387-402.
10. P. Uranga, J. Rodriguez-Ibabe, D.G. Stalheim, R. Barbosa, M. Rebellato, “Application of Practical Modeling of
Microalloyed Steels for Improved Metallurgy, Productivity and Cost Reduction in Hot Strip Mill Applications”,
Proceedings of AISTech 2016 Conference, 16-19 May 2016, Pittsburgh, PA, pp. 1769-1778.
11. P.D. Hodgson, R.K. Gibbs, “A Mathematical Model to Predict the Mechanical Properties of Hot Rolled C-Mn and
Microalloyed Steels”, ISIJ Int., Vol. 32, 1992, pp. 1329-1338.

© 2019 by the Association for Iron & Steel Technology. 2011