metals

Article
The Influence of Mg-Based Inclusions on the Grain
Boundary Mobility of Austenite in SS400 Steel
Chih-Ting Lai 1 , Hsuan-Hao Lai 1 , Yen-Hao Frank Su 2 , Fei-Ya Huang 1 , Chi-Kang Lin 1 ,
Jui-Chao Kuo 1, * and Hwa-Teng Lee 3
1 Department of Materials Science and Engineering, National Cheng-Kung University, Tainan 701, Taiwan;
tedchen1118@hotmail.com (C.-T.L.); lai57215308@gmail.com (H.-H.L.); a2780490@gmail.com (F.-Y.H.);
cold19871025@gmail.com (C.-K.L.)
2 China Steel Corporation, Kaohsiung 81233, Taiwan; 150151@mail.csc.com.tw
3 Department of Mechanical Engineering, National Cheng-Kung University, Tainan 701, Taiwan;
htlee@mail.ncku.edu.tw
* Correspondence: jckuo@mail.ncku.edu.tw; Tel.: +886-6-2754194




Received: 23 January 2019; Accepted: 20 March 2019; Published: 22 March 2019


Abstract: In this study, the effects of the addition of Mg to the grain growth of austenite and
the magnesium-based inclusions to mobility were investigated in SS400 steel at high temperatures.
A high-temperature confocal scanning laser microscope (HT-CSLM) was employed to directly observe,
in situ, the grain structure of austenite under 25 torr Ar at high temperatures. The grain size
distribution of austenite showed the log-normal distribution. The results of the grain growth curves
using 3D surface fitting showed that the n and Q values of the growth equation parameters ranged
from 0.2 to 0.26 and from 405 kJ/mole to 752 kJ/mole, respectively, when adding 5.6–22 ppm of Mg.
Increasing the temperature from 1150 to 1250 ◦ C for 20 min and increasing the addition of Mg by 5.6,
11, and 22 ppm resulted in increases in the grain boundary velocity. The effects of solute drag and
Zener pinning on grain boundary mobility were also calculated in this study.

Keywords: confocal scanning laser microscope; grain growth; magnesium addition; low-carbon steel;
mobility; pinning effect

1. Introduction
Both medium and thick steel plates are used in the building of high-rise buildings, bridges, and
in ships. A high-heat-input welding technique with heat input of over 50 kJ/cm has been employed
for high-efficiency, low-cost fabrication of these steel plates. Increasing the heat input energy during
welding enhances the coarsening of the austenite grains in the heat-affected zone (HAZ). However,
coarse grains have detrimental effects on the impact toughness of weldments [1].
TiN inclusions were first applied to improve the toughness of HAZs, which serve as potential
nucleation sites for acicular ferrite (AF) formation [2–4]. The development of TiN inclusions [5–8] can
be expressed as the zero generation of inclusions for AF formation, before the use of oxide inclusions.
However, Nb, V, and Ti nitrides will dissolve into austenite because of the high heat input from
welding temperatures above 1300 ◦ C [9]. Thus, Mizoguchi et al. [10] first proposed oxide inclusions to
improve the impact toughness of HAZs under high-heat-input welding.
Titanium and magnesium oxides can be called the first and second generations of oxide inclusions
for AF formation. In this study, Titanium oxides serve as potent nucleation sites for the formation of
intragranular AF [11–19]. Magnesium oxide serves as fine dispersed inclusions in steel [20] because
Mg has a strong affinity for oxygen and a reasonable amount of Mg is soluble in steel [5]. Kojima
et al. [5] first proposed magnesium oxide metallurgy technology, in which MgO, MgS, and Mg(O,S)

Metals 2019, 9, 370; doi:10.3390/met9030370 www.mdpi.com/journal/metals

Metals 2019, 9, 370 2 of 14

inclusions were used to pin grain boundaries. According to previous research, Zener pinning was
the influence of the dispersion of spherical particles [21], ellipsoidal particles [22] and cylindrical
particles [23] on the movement of grain boundaries. The particles acted to prevent the motion of grain
boundaries by exerting a pinning force which counteracted the driving force of the grain boundaries.
The pinning efficiency against grain growth for grain coarsening was determined by the particle
size [21]. This process refined the HAZ microstructure. Suito et al. [24–27] reported that MgO and
soluble Mg can hinder excessive growth of austenite grains. In addition, Zhu et al. found that
the addition of 0.005 wt% Mg can effectively inhibit austenite grain growth [28,29] and lead to the
formation of AF [30].
The chemical composition of steel, the cooling rate at temperatures ranging from 800 to 500 ◦ C,
the size of austenite grains, and the inclusion parameters affect the formation of intragranular AF in
metal [31]. In this study, we investigated the effect of the addition of Mg on austenite grain growth
in SS400 low-carbon steel using a high-temperature confocal scanning laser microscope (HT-CSLM)
in order to observe the high temperature microstructure in situ. In addition, the pinning effect of the
magnesium-based inclusions on the mobility of austenite was studied.

2. Materials and Experiments

SS400 low-carbon steels produced by China Steel Corporation were selected as the experimental
materials, for which the chemical composition of the steels is listed in Table 1, and Fe-5% Mg alloy wire
was used for the addition of Mg during the ladle treatment. The experimental materials obtained from
the as-cast slab, 8000 mm long, 1800 mm wide, and 270 mm thick, were cut into specimens 50 mm
thick and 200 mm wide. After hot rolling at 1200 ◦ C, the thickness was reduced from 50 mm to 10 mm,
and 6 mm × 5 mm × 2 mm specimens were prepared.

Table 1. Chemical composition of SS400 steel with Mg of 1.5–22 ppm (Unit: wt%).

Mg C Si Mn P S Al N O
1.5 ppm 0.120 0.17 1.18 0.0071 0.0006 0.028 0.0053 0.0041
5.6 ppm 0.118 0.17 0.74 0.0102 0.0035 0.033 0.0054 0.0082
11.0 ppm 0.136 0.28 0.94 0.0075 0.0012 0.020 0.0043 0.0027
22.0 ppm 0.127 0.28 0.87 0.0104 0.0021 0.018 0.0052 0.0016

The specimens were ground using 1500, 2500, and 4000 SiC paper, polished using 3, 1, and 0.1
µm diamond suspensions, respectively, and finally polished with 0.02 µm SiO2 suspensions. After the
final polish, the austenite grain structure was observed in situ at high temperature under 25 torr Ar
using an HT-CSLM (VL2000DX-SVF17SP, Yonekura, Japan). First, the specimen was placed into an
Al2 O3 crucible with a diameter of 8 mm and a height of 3.5 mm. Prior to heating, the chamber was
evacuated for 5 min using a diffusion pump backfilled with ultrahigh purity Ar gas (>99.999%), with
a discharge rate of 300 mL/min passed through a gas cleaning system to reduce the oxygen content.
Then, the specimens were heated to 1000, 1100, 1200, and 1250 ◦ C for 3, 5, 10, and 20 min, respectively.
Afterward, the specimens were cooled at a cooling rate of 100 ◦ C/s. After cooling, images of the
grain structure were examined with a Leica DM6000 M optical microscope (Leica, Wetzlar, Germany).
Microscope Imaging Software (Grain Expert, Leica, Wetzlar, Germany) was used to calculate the
average grain size according to the ASTM E112 international standards. Then, the HT-CSLM was
employed to investigate the movement of the austenite grains. We used an ASPEX EXplorer system
(ASPEX, LLC, Boulder, CO, USA) combining SEM and EDS to automatically classify different types of
inclusions and to determine the inclusion size and number within a large measured area.
Metals 2019, 9, 370 3 of 14

3. Results and Discussion

3.1. Parameters of Grain Growth Equation

Metals 2019, 9, x FOR PEER REVIEW 3 of 14
To determine the number of grains, we adopted the Leica analysis Grain Expert software to detect
Metals 2019, 9, x FOR PEER REVIEW 3 of 14
the grains,
image and Figure
is shown 1b shows
in Figure an image
1a. Then, of the detected
a representative valuegrains in minimal
of the austenite.number
The original image
of grains was
is shown
obtained.
image in Figure
Figures
is shown 1a.
in 2a–d Then,
Figure show a representative
1a. aThen,
histogram value of the
of grain sizevalue
a representative minimal
in terms number
of the
of the number
minimal of grains was
of grains,
number obtained.
while Figure
of grains was
Figure
2e 2a–d
shows show
the a
effecthistogram
of the of
numbergrain size
of in
grains terms
on of
the the number
average of
grain grains,
size andwhile Figure
standard
obtained. Figures 2a–d show a histogram of grain size in terms of the number of grains, while Figure 2e shows
deviation. the
The
effect of the
distributions number
of of
grain grains
size andon the
the average
average grain
grain size
sizes and
were standard
similar. deviation.
However,
2e shows the effect of the number of grains on the average grain size and standard deviation. The The distributions
increasing the number of
grain size and
of grains
distributions ofthe average
decreased
grain size grain
the and sizes
standard
the were similar.
deviation.
average grainThe However,
minimal
sizes increasing
number
were similar. thegrains
of
However,number wasof set
grains
increasing at decreased
the300 for the
number
the standard deviation. The minimal number of grains was set at 300 for the
of grains decreased the standard deviation. The minimal number of grains was set at 300 for the subsequent analysis.

(a) (b)
Figure 1. (a) Initial (a) (b) (b) after the detection
image of austenite obtained using optical microscopy and
process
Figure
Figure 1. using
1. (a) analysis
(a) Initial
Initial imagesoftware
image of after obtained
of austenite
austenite quenching.
obtained using
using optical
opticalmicroscopy
microscopyand
and(b)
(b) after
after the
the detection
detection
process
process using
using analysis
analysis software
software after
after quenching.
quenching.

0.3 1.0
0.3
Relative Frequency

1.0
Relative Frequency

0.3 0.3
Relative Frequency

0.2 0.2
Relative Frequency

0.2 0.5
0.2

0.1 0.5 0.1

0.1 0.1
0.0
0.0 0.0 0 20 40 60 80 100 120 140 160
0 20 40 60 80 100 120 140 160
0.0 0.0
0.0 Grain size(m)
0 20 40
Grain
60
size(m)
80 100 120 140 160
0 20 40 60 80 100 120 140 160

(a) m)
Grain size( (b) m)
Grain size(

(a) (b)
0.3 0.3
Relative Frequency
Relative Frequency

0.3 0.3
Relative Frequency

0.2 0.2
Relative Frequency

0.2 0.2
0.1 0.1

0.1 0.1
0.0 0.0
0 20 40 60 80 100 120 140 160
0 20 40 60 80 100 120 140 160
0.0 Grain size(m) 0.0
0 20 40 60 80 100 120 140 160 Grain size(m)
0 20 40 60 80 100 120 140 160
(c) m)
Grain size( (d) m)
Grain size(
(c) Figure 2. Cont. (d)
 Metals2019,
Metals 2019,9,9,x370
FOR PEER REVIEW 44ofof14
14

Metals 2019, 9, x FOR PEER REVIEW 50 0.8 4 of 14

Average grain size

Average grain size (m)

40
50 0.8

Standard deviation
Average grain size
3040 Standard deviation

Average grain size (m)

0.7

Standard deviation
2030 Standard deviation
0.7
1020

010 0.6
200 300 400
0 Number of grains 0.6
200 300 400
(e)of grains
Number
(e)
Figure
Figure2.2. Histogram
Histogram of of austenite grain size
austenite grain sizeininterms
termsofofthe
themeasured
measured number
number of of grains:
grains: (a) (a)
200,200,
(b) (b)
250,
250, Figure
(c) 300, 2. Histogram
and (d) of
400. austenite
(e) is the grain size
average in terms
grain of
size the
and measured
standardnumber of
deviationgrains:
as a (a) 200,
function
(c) 300, and (d) 400. (e) is the average grain size and standard deviation as a function of the measured (b)
of the
250, (c)
measured 300, and
of(d) 400. (e) is the average grain size and standard deviation as a function of the
number ofnumber
grains. grains.
measured number of grains.
Aftermeasuring
After measuringthe thenumber
numberof ofgrains,
grains,aasinglesinglerepresentative
representativevalue valuefor forthe
thegraingrainsizes
sizesshould
should
After measuring the number of grains, a single representative value for the grain sizes should
beobtained
be obtainedstatistically.
statistically.Statistical
Statisticalvalues,
values,including
includingthe themean,
mean,median,
median,and andmode,
mode,were wereconsidered.
considered.
be obtained statistically. Statistical values, including the mean, median, and mode, were considered.
Figure
Figure 33illustrates
Figure illustrates thethe
the
3 illustrates histogram
histogram
histogram and
and cumulative
andcumulative frequency
frequency
cumulative frequency distributions
distributions
distributions ofthe
of
of the thegrain
graingrain size
size size
of ofofaustenite
austenite
austenite
inSS400
in SS400 with 1.5 ppm Mg. Figure 4 shows the evolution of the austenite
in SS400 with 1.5 ppm Mg. Figure 4 shows the evolution of the austenite grain size using the threethree
with 1.5 ppm Mg. Figure 4 shows the evolution of the austenite grain
grain size
size using
using the
the three
statistical
statistical values. The mean is found by adding all given data and dividing
statistical values. The mean is found by adding all given data and dividing by the number of data data
values. The mean is found by adding all given data and dividing by
by the
the number
number of
of data
entries.
entries. TheThe
The
entries. median
median
medianis the
the middle
middle
is the middlenumber
number
number ofofallall
of values,
all values,
values,while
while
while the mode
the
the mode
mode isisthe number
is the
the numbernumber that occurs
that that
occurs most
occurs
oftenmost
most in the
often often
indatain set.
the the
datadata
set.set.
Thus, theThus,
Thus, mean the
the mean
has
mean thehashas the highest
highest
the value, value,
highest while while
value,the thethe
mode
while mode
is the
modeislowest.
theis lowest.
theThe The Theis
median
lowest.
medianmedian
between isthe ismean
between
between thethe
and mean
mode
mean andmode
(Figure
and mode (Figure 4).
4).(Figure
Therefore, 4). Therefore, ininthis
in this study,
Therefore, westudy,
this adapted
study, wewe adapted
theadapted
mode theasmode
the as
themode
average
as
value
the the average
to avoid
average value value to
calculating avoid
to avoidthe calculating
large grains
calculating the large
the because grains
large grains because
largebecause large
grains lead grains
largetograins lead
abnormal to abnormal
leadgrain grain
growth.grain
to abnormal
growth.
growth.
0.25 1.0
0.25 1.0 1.0 0.25 1.0

Cumulative Frequency
0.25 1.0 1.0 0.20
Relative Frequency
Cumulative Frequency

Cumulative Frequency
0.20 0.2
Cumulative Frequency

0.20
Relative Frequency

Relative Frequency

Relative Frequency

0.15
Cumulative Frequency

0.20 0.2
Cumulative Frequency

0.15
Relative Frequency

Relative Frequency

0.5
0.5 0.5 0.10
0.15
0.15 0.10 0.1 0.5
0.5 0.5 0.05 0.10
0.10 0.05 0.1
0.00 0.05 0.0
0.05 0.00 0.0 0.0 0.0 0 500 1000
0 100 200 300 0 200 400 600 800 Grain size(m)
0.00 0.0
Grain size(m) Grain size(m) 0 500 1000
0.00 0.0 0.0 0.0
0 100 200 300 0 200 400 600 800 Grain size(m)
Grain size((a)
m) (e)
Grain size(m) (i)
0.25 1.0
0.25 (a) 1.0 (e) 1.0
(i)
Cumulative Frequency

0.20
Cumulative Frequency

Relative Frequency
Cumulative Frequency

0.20 0.2 0.25 1.0

Relative Frequency

Relative Frequency

0.25 1.0 1.0 0.15

Cumulative Frequency

0.15 0.20 0.5

Cumulative Frequency

Relative Frequency

0.5 0.5 0.10

Cumulative Frequency

0.20 0.2
0.10 0.1
Relative Frequency

Relative Frequency

0.15
0.05
0.15 0.05 0.5
0.5 0.5 0.00 0.10 0.0
0.10 0.00 0.0 0.1
0.0 0.0 0 500 1000
0 100 200 300 0 200 400 600 800
0.05 Grain size(m)
0.05 Grain size(m) Grain size(m)
0.00 0.0
0.00 (b) 0.0 0.0 (f) 0.0 0 (j) 500 1000
0 100 200 300 0 200 400 600 800
Grain size(m)
Grain size(m) Grain size(m)
Figure 3. Cont.
(b) (f) (j)
 Metals 2019, 9, x FOR PEER REVIEW 5 of 14

0.25 1.0
0.25 1.0 1.0

Cumulative Frequency
0.20

Cumulative Frequency

Relative Frequency
0.2

Cumulative Frequency
0.20 2019, 9, 370
Metals 5 of 14

Relative Frequency
Relative Frequency
Metals 2019, 9, x FOR PEER REVIEW 0.15 5 of 14
0.15 0.5
0.5 0.5 0.10
0.10 0.1
0.25 0.05 1.0
0.050.25 1.0 1.0

Cumulative Frequency
0.20

Cumulative Frequency
0.00 0.0

Relative Frequency
0.2

Cumulative Frequency
0.000.20 0.0 0.0 0.0 0 500 1000

Relative Frequency
Relative Frequency

0 100 200 300 0 200 400 600 800 0.15

0.15 Grain size(m)0.5
Grain size(m) 0.5 Grain size(m) 0.5 0.10
0.10 0.1
0.05
0.05 (c) (g) (k)
0.00 0.0
0.250.00 0 0.0 1.0 0.0 0.0 0 500 1000
100 200 300 0 200 400 600 800 0.25 1.0
1.0 Grain size(m)

Cumulative Frequency
Grain size(m) Grain size(m)

Cumulative Frequency
0.20 0.20
0.3

Cumulative Frequency
Relative Frequency

Relative Frequency
Relative Frequency
0.15 (c) (g) 0.15 (k)
0.5 0.2 0.5
0.100.25 1.0 0.5 0.25 0.10 1.0
1.0
Cumulative Frequency

Cumulative Frequency
0.050.20 0.30.1 0.20 0.05

Cumulative Frequency
Relative Frequency

Relative Frequency
0.15 Relative Frequency 0.15 0.00 0.0
0.00 0.0
0.5 0 500 0.5 1000
0 100 200 300 0.20.0 0.0
0.10 0.5 0.10
0 200 400 600 800 Grain size(m)
Grain size(m)
0.05 0.1 Grain size(m) 0.05

0.00 0.0
0.00 0.0
0 100 (d) 200 300 0.0 (h) 0.0
0 500 (l)1000
0 200 400 600 800 Grain size(m)
Grain size(m)
Grain size(m)
Figure 3. Histogram and cumulative frequency distributions of grain size of austenite in SS400 steel
with 1.5 (d)
ppm Mg for (a, e, i) 3 min, (b, f, j) 5 min, (h) (c, g, k) 10 min, and (d, h, l) 20 min. (l) Broken lines
represent
Figure
the curves
3. Histogramand
fitted by
andcumulative
the log-normal
cumulative frequency
function: (a–d) at 1150 °C, (e–h) at 1200 °C,steel
and (i–l) at
Figure 3. Histogram frequencydistributions
distributions of grain sizesize
of grain of austenite in SS400
of austenite in SS400 steel
1250 °C.
withwith 1.5 ppm
1.5 ppm MgMgforfor (a, e,3 i)
(a,e,i) 3 min,
min, (b, 5f, min,
(b,f,j) j) 5 min, (c, g,
(c,g,k) 10k) 10 min,
min, and and (d, 20
(d,h,l) h, l) 20 min.
min. Broken
Broken lineslines
represent
represent the curves fitted by the log-normal function: (a–d) at 1150 °C, (e–h) at 1200 °C, and (i–l) at
the curves fitted by the log-normal function: (a–d) at 1150 ◦ C, (e–h) at 1200 ◦ C, and (i–l) at 1250 ◦ C.
1250 °C.

300

300
(m)(m)

200
sizesize

200
Grain

100
mean
Grain

100 median
mean
mode
median
0 mode
0 0 500 1000 1500
0 500 1000 1500
Time (sec)
Time (sec)
Figure
Figure 4. 4. Grain
Grain size
size asas a function
a function ofof annealing
annealing time
time forfor SS400
SS400 with
with 1.51.5 ppm
ppm MgMg
atat ◦ C°C
1200
1200 using
using the
the
Figure 4. Grain size as a function of annealing time for SS400 with 1.5 ppm Mg at 1200 °C using the
mean,
mean, median,
median, and
and mode.
mode.
mean, median, and mode.

Next,
Next,Next, the
the grain
thegrain growth
graingrowth equationwas
growth equation was
was derived
derived
derived after
afterafter the
theaveraging
the averagingaveraging
methodmethod
methodwaswasdetermined.
was determined.determined.
The The
numerous
The numerous
numerous equations
equations
equations that havehave
that that
have beenproposed
been proposed
been to predict
proposed
to predicttothe the grain
predict
grain growth
the grain
growth equation
growth
equation be can
can equationbe can
divided divided
be
into
divided
intothree
intotypes.
three three First,
types. types. Beck etetal.
First,
First, Beck al.[32]
Beck [32] reported
etreported a power
al. [32]areported
power law a law
powerfor law
normal
for normal grain
for normal
grain growth,
growth, grain whileetSellars
growth,
while Sellars whileet
al.al.
Sellars [33–36]
[33–36] modified
et al. modified
[33–36] Beck’s equation
Beck’s
modified equation asasfollows:
Beck’s equation follows:
as follows:

ddmmddom mdk−
0k exp
do = kQ0·/Qexp
 exp (−
 t Q/RT
RT/ RT t
)·t m m
(1) (1)
(1)
o 0

where d is the final grain diameter; d0 is the initial grain diameter; t is the annealing time, and m and k0
are constants. Q is the activation energy for grain growth, and T is the temperature in absolute degrees.
where d is the final grain diameter; d0 is the initial grain diameter; t is the annealing time, and m and
The simplified
where is theBeck’s
k0 aredconstants.
final modelactivation
Q grain
was used
is the diameter; d0toispredict
energy the the growth,
forinitial
initialdiameter;
grain grain
austenitic grain
and T istthe
is the size as follows
annealing
temperature
[37]:
intime, and m and
absolute
k0 are constants. Q is the activation energy for grain growth, and T is the temperature in absolute
dm = k0 · exp (− Q/RT ) · t (2)
Metals 2019, 9, x FOR PEER REVIEW 6 of 14

Metals 2019, 9, 370 d 2 − d o2 = k0 ⋅ exp(− Q / RT ) ⋅ t (3) 6 of 14

Moreover, Yoshie et al. [38] and Nishizawa [39] reported a model represented as follows:
In this study, we applied the grain growth formulation in Equations (1) and (2) to predict the
grain growth curves for SS400 with d2 1.5
− dppm
o = k 0Mg
2
at (−
· exp 1200
Q/RT °C as
) · tan example. The parameters of the (3)
fitting growth equations are summarized in Table 2. The fitting curves, using Equation (2), agree with
the experimental
In this study,data with an R-value
we applied the grain ofgrowth
0.96, and the results in
formulation areEquations
remarkably (1)better than
and (2) to those using
predict the
◦
Equation
grain (1) with
growth curves anfor
R-value
SS400of 0.691.5
with [Figure
ppm Mg 5a,b]. The values
at 1200 C as an Q and k0 using
of example. Equation (2)
The parameters are
of the
similar to those of Equation (1). A comparison of the experimental data in Figure 5a,b showed that
fitting growth equations are summarized in Table 2. The fitting curves, using Equation (2), agree with
Equation
the (2) wasdata
experimental the with
best anfitting equation.
R-value of 0.96,Thus,
and theEquation
results (2)
are was employed
remarkably forthan
better the those
subsequent
using
investigations.
Equation (1) with an R-value of 0.69 [Figure 5a,b]. The values of Q and k0 using Equation (2) are similar
to those of Equation (1). A comparison of the experimental data in Figure 5a,b showed that Equation
(2) was Table 2. Parameters
the best of the fitting
fitting equation. Thus,equations
Equationof(2)Equations (1) and (2)
was employed forthe
for SS400 with 1.5 ppm
subsequent Mg.
investigations.

d0 of Equations (1) and

Q(2) for SS400 with 1.5 ppm Mg.
Fitting equation
Table 2. Parameters of the fitting equations m k0 (μm/s)
(μm) (kJ/mole)
Equation
Fitting (1)
equation 4.3
d0 (µm) 2
m 232
Q (kJ/mole) 7.2 × 109
k0 (µm/s)
Equation
Equation(2)
(1) 4.30 2.9
2 622
232 3.2 ×
7.210 9 1025

Equation (2) 0 2.9 622 3.21025

(a) (b)
Figure 5.
Figure 5. Mean
Mean value
value of
of grain
grain size
size as
as aa function
function of
of time
time using
using (a)
(a) Equation
Equation (1)
(1) with
with an
an R-value
R-value of
of 0.96
0.96
and (b)
and (b) Equation
Equation (2)
(2) with
with an
an R-value
R-value ofof 0.69
0.69 for
for SS400
SS400 with
with 1.5
1.5 ppm
ppm Mg.
Mg.

3.2.
3.2. Effects
Effects of
of Temperature
Temperature and
and Mg-Content
Mg-Content on
on Austenite
AusteniteGrain
GrainGrowth
Growth
Figure
Figure 33 illustrates
illustrates the
the histogram
histogram and and cumulative
cumulative frequency
frequency distributions
distributions of
of the
the grain
grain size
size of
of
austenite with an Mg content of 1.5–22 ppm at 1150–1250 ◦ C. The average grain size was plotted
austenite with an Mg content of 1.5–22 ppm at 1150–1250 °C. The average grain size was plotted in
in terms
terms of the
of the annealing
annealing timetime
andand temperature
temperature withwith different
different Mg Mg content,
content, as shown
as shown in Figure
in Figure 6,
6, and
and the corresponding
the corresponding parameters
parameters of the
of the fitting
fitting curves
curves areare listedininTable
listed Table3.3.The
Thegrain
grain size
size distributions
distributions
followed the log-normal distribution, which led to an increase in the average grain size
followed the log-normal distribution, which led to an increase in the average grain size with increases with increases
in
in annealing
annealing temperature.
temperature. TheThe Q
Q and
and nn values
values are
are listed
listed inin Table
Table3. 3.

Table 3. Parameters of the fitting equations for steel with Mg content of 1.5-22.0 ppm using Equation (2).
Table 3. Parameters of the fitting equations for steel with Mg content of 1.5-22.0 ppm using Equation (2).

Mg (ppm)
Mg (ppm) m m QQ(kJ/mole)
(kJ/mole) k0 (µm/s)
k0 (μm/s)
1.5 1.5 2.92.9 622
622 3.2 × 103.2
25
× 1025
5.6 4.9 405 2.5 × 1019 19
5.6 11.0 4.93.8 405
493 2.8 × 102.5
21 × 10

1.7 × 1031 21
11.0 22.0 3.8 6 493
752
2.8 × 10
31
22.0 6 752 1.7 × 10
Metals
Metals2019, 9, 9,
2019, x FOR
370 PEER REVIEW 7 of7 of
1414

(a) (b)

(c) (d)
Figure 6. 6.
Figure Grain size
Grain asas
size a function of of
a function time
timeforfor
(a)(a)
1.5, (b)(b)
1.5, 5.6, (c)(c)
5.6, 11,11,
and (d)(d)
and 2222ppm
ppm Mg,
Mg,where
wheresymbols
symbols
indicate experimental
indicate experimental data and
data the
and thefitting function
fitting functionis isindicated
indicatedbybybroken
brokenlines.
lines.

Except
Except for
forthe
thecase
caseofof1.5ppm
1.5ppmMg,
Mg,thetheQQvalue
valueincreased
increasedfrom
from8181toto150150kJ/mole
kJ/mole and the
and nn
the value
value
was inin
was thethe
range
rangeof 0.2–0.26
of 0.2–0.26as Mg content
as Mg increased
content fromfrom
increased 5.6 to5.6
22ppm. It was It
to 22ppm. noted
was that Q value
thethat
noted the Q
obtained from thefrom
value obtained references in Tablein4Table
the references ranged from 64
4 ranged to 437
from 64 tokJ/mole with the
437 kJ/mole exception
with of 914of
the exception
kJ/mole. Thus, our
914 kJ/mole. Thus,Q our
and Q n values lie in the
and n values lie range
in the obtained from the
range obtained fromreferences.
the references.

Table
Table4.4.Parameters
Parameters n,n,Q,Q,and
ofof andk0kfor the
0 for growth
the equations.
growth equations.

Steels Steels m m Q (kJ/mol)

Q (kJ/mol) kk00(μm/s)
(µm/s) References
References
Low C-MnLow C-Mn2 2 67 67 4.3
4.3 ×× 10
10 12
12 [40] [40]
Low C-MnLow C-Mn 5.6 5.6 126 126 9.4
9.4 ×× 10 38
1038 [37] [37]
× 12
C-Mn C-Mn 2 2 64 64 1.4 × 10 32
1.4 10 12 [41] [41]
C-Mn 1 400 (T>1273) 3.9 × 1032 [36]
C-Mn 1 400 (T>1273) 3.9 × 10 53 [36]
C-Mn 1 914 (T<1273) 5.0 × 10 [36]
C-Mn C-Mn-V 1 7914 (T<1273) 400 5.0
1.5 ×× 10 53
1027 [36] [36]
C-Mn-V C-Mn-Ti 7 10 400 437 1.5
2.6 ×× 10
1027
28 [33] [36]
C-Mn-Ti C-Mn-Nb10 4.5 437 435 2.6
4.1 ×× 10 23
1028 [33] [33]
25
C-Mn-Nb High C-Mn4.5 5.3 435 366 4.1 × 10
9.1 × 10 23 [37] [33]
High C-Mn 0.12 1170 4.1 x 1063 [31]
High C-Mn 5.3 366 9.1 × 10 25 [37]
High C-Mn 0.12 1170 4.1 x 1063 [31]
We differentiated Equation (2) with respect to time (t) to reveal the grain boundary velocity
We differentiated Equation (2) with respect to time (t) to reveal the grain boundary velocity (
(V = dD ) as follows:
dDdt V = n ∗ k o ∗ exp (− Q/RT ) ∗ [t − to ]n−1
V= ) as follows: (4)
dt
The grain boundary velocity was plotted as a function of annealing time in Figure 7. When 1.5 ppm
Mg was added, the grain boundary velocity increased from 0.025 µm/s to 0.1 µm/s after 20 min as
V = n *increased
the annealing temperature ko ∗ exp from (
− Q /1150
RT to ∗ 1250 ) [n −1
]
t − to◦ C. The grain boundary velocity
(4) after 20 min
increased from 0.005 µm/s to 0.01 µm/s, 0.01 µm/s to 0.02 µm/s, and 0.005 µm/s to 0.07 µm/s
when 5.6, 11, and 22 ppm of Mg were added, respectively, as the annealing temperature increased
The
from grain
1150 1250 ◦ C. Considering
to boundary velocity wasthe
plotted as a function
annealing time of of 20 annealing time boundary
min, the grain in Figure 7.velocity
When 1.5was
◦
ppm Mg was
plotted added,
in terms thecontent
of Mg grain boundary velocity
at 1150–1250 C inincreased
Figure 8. from 0.025the
However, μm/s to 0.1
grain μm/s after
boundary 20 min
velocity was
as the annealing temperature increased from 1150 to 1250 °C. The grain boundary velocity after 20
Metals 2019, 9, x FOR PEER REVIEW 8 of 14

min increased from 0.005 μm/s to 0.01 μm/s, 0.01 μm/s to 0.02 μm/s, and 0.005 μm/s to 0.07 μm/s
when 5.6, 11, and 22 ppm of Mg were added, respectively, as the annealing temperature increased
from
Metals1150
2019, to 1250 °C. Considering the annealing time of 20 min, the grain boundary velocity was
9, 370 8 of 14
plotted in terms of Mg content at 1150–1250 °C in Figure 8. However, the grain boundary velocity
was reduced with increases in the Mg content, and steels with the addition of 5.6 ppm and 22 ppm
ofreduced with increases
Mg exhibited similar in the Mg
trends. content,
This and steelssuggests
phenomenon with thethat
addition of 5.6boundary
the grain ppm and velocity
22 ppm of Mg
was
exhibited similar trends. This phenomenon suggests that the grain boundary
significantly decreased by increasing Mg content at 1250 °C. The results in Table 3 show that the velocity was significantly
decreased by increasing Mg 1.5content ◦ C. The results in Table 3 show that the increase in Mg
increase in Mg content from to 22atppm
1250increased the Q exponent. However, the Q exponent
content from
decreased when 1.5the
to 22
Mgppm increased
content was 5.6 Q exponent.
theand 11 ppm. In However, theQactivation
general,the exponentenergy
decreased when
Q for grainthe
Mg content was 5.6 and 11 ppm. In general, the activation energy Q
growth was affected by the amount and type of alloying elements. For materials with inclusionby
for grain growth was affected
the amount
pinning and type of
and elements of alloying elements.
solute drag, grain For
growthmaterials with inclusion
was difficult pinning and
and activation elements
energy of solute
increased. In
drag, grain growth was difficult and activation energy increased. In
some mechanisms, the Q exponent decreased. The m exponent depended on the grain growth some mechanisms, the Q exponent
decreased. The
mechanisms. Whenm exponent depended
the m exponent was on2,the
thegrain growth
materials mechanisms.
had no defects When the m exponent
or inclusions. When the was m2,
exponent was 3, several phenomena, such as inclusions in the grain, were observed, which indicatesas
the materials had no defects or inclusions. When the m exponent was 3, several phenomena, such
the occurrence
inclusions in of
thethe pinning
grain, wereeffect. Whenwhich
observed, the m exponent
indicates was 4 the alloy element
the occurrence exhibited
of the pinning diffusion
effect. When
inthe
them grain boundaries
exponent was 4and
the produced
alloy elementinclusion,
exhibitedwhich indicatesinthe
diffusion theoccurrence of pinning
grain boundaries andand solute
produced
drag effects [42]. Thus, the effects of the pinning force and solute drag on austenite grain mobility
inclusion, which indicates the occurrence of pinning and solute drag effects [42]. Thus, the areof
effects
considered in the subsequent section.
the pinning force and solute drag on austenite grain mobility are considered in the subsequent section.

(a) (b)

(c) (d)
Metals
Figure
Figure The9,grain
2019, grain
7.7.The x FOR PEER REVIEW
boundary
boundaryvelocity
velocity(V) asas
(V) a function ofof
a function time
timefor Mg
for Mgatat(a)(a)1.5,
1.5,(b) 5.6,
(b) 5.6,(c)(c)
11, and
11, 9 of 14
and(d)
(d)
2222ppm.
ppm.

Figure Figure
8. The8. The grain
grain boundary velocity
boundary velocity (V)
(V)in in
terms of annealing
terms temperature
of annealing after 20 min
temperature of annealing
after 20 min of
time.
annealing time.

3.3. Zener Pinning Effect of Inclusions on the Austenite Grain Mobility

According to the model [36], the driving force of grain growth is expressed by:

2σ b
FP = (5)
R
Metals 2019, 9, 370 9 of 14

3.3. Zener Pinning Effect of Inclusions on the Austenite Grain Mobility

According to the model [36], the driving force of grain growth is expressed by:

2σb
FP = (5)
R
where R is the radius of the curvature of grains, and σb is the surface energy of the grain boundary.
Moreover, inclusions exert the retarding force (FR ) on the grain boundaries proposed by Zener [43] as:

3 f v σb
FR = (6)
2r
where f v is the volume fraction of inclusions, σb is the surface energy of the grain boundary, and r is
the radius of the inclusions. Here the surface energy of the grain boundary σb is for austenite, the value
of which was obtained from Reference [44]. The retarding forces (also called the Zener pinning force)
depend on the radius and volume fraction of inclusions, as listed in Table 5. Here the radius (r) is equal
to half of the average diameter, and fA is obtained by dividing the number of inclusions (N) by the
measured area (A). Next, the equation fA = 2NfV was employed. It has been reported that the addition
of 2 ppm Mg led to a change in oxide formation from the Al2 O3 to MgO·Al2 O3 phase and a change in
inclusion formation from Al2 O3 –MnS to MgO·Al2 O3 –MnS [45].

Table 5. The retarding force (FR ) of inclusions.

Mg A D fA fV FR FR
Inclusion
(ppm) (m2 ) (m) (10−6 × %) (10−4 × %) (kJ/m3 ) (kJ/m3 )
1.5 MgO-MnS 87.6 1.9 2.5 15.9 2.9
MnS 67.9 1.4 2.0 14.0 3.5 11.7
MgAl2 O4 693 2.9 19.9 44.6 5.3
5.6 MgO-MnS 26.8 1.9 0.8 8.8 1.6
MnS 26.9 2.1 0.8 8.8 1.5 8.5
MgAl2 O4 834 3.1 23.9 48.9 5.4
11 MgO 10.4 1.7 0.2 4.7 1.0
MgO-MnS 46.6 1.9 1.0 10.0 1.8
MnS 861 3.4 18.5 43.0 4.3 20.3
MgAl2 O4 2030 3.8 43.7 66.1 6.0
Mg-Al-MnS 2790 3.8 60.0 77.5 7.1
22 MgO 108 2.7 3.1 17.6 2.2
MgO-MnS 224 4.4 6.4 25.3 2.0
MnS 310 2.5 8.9 29.8 4.2 17.1
MgAl2 O4 235 2.8 6.7 26.0 3.2
Mg-Al-MnS 759 3.0 21.8 46.6 5.5
* Note: σb is 1.159 J/m2 and the measured areas for 1.5, 5.6, 11, 22 ppm Mg are 34.865, 34.865, 46.487 and
34.865 mm2 , respectively. A—area; D—average diameter; fA —area fraction; fV —volume fraction; FR —retarding
force, and FR —sum of the retarding forces for inclusions.

In the case of 1.5 and 5.6 ppm Mg, the effective inclusion of MgAl2 O4 exerted the greatest retarding
force, and, for 11 and 22 ppm Mg, it was Mg-Al-MnS that exerted the greatest retarding force because
the volume fraction of MgAl2 O4 was the largest phase for 1.5 and 5.6 ppm Mg. The volume fraction of
Mg-Al-MnS was the largest phase for 11 and 22 ppm Mg. The average diameter of the inclusions was
in the range of 1.4–4.4 µm, and the volume fraction of the inclusions was in the range of 4.7–77.5%.
According to Equation (6), the retarding force was mainly determined by the volume fraction because
there was no significant difference in the average diameter of inclusions seen in Table 5. Furthermore,
there was no a linear relationship found between the addition of Mg and the retarding force.
volume fraction because there was no significant difference in the average diameter of inclusions seen
in Table 5. Furthermore, there was no a linear relationship found between the addition of Mg and the
retarding force.
The mobility of grain growth M is proportional to the net driving force, which is equal to the
driving force
Metals 2019, of grain growth (FP) subtracted from the Zener pinning force (FR), which was expressed
9, 370 10 of 14

by Reference [43]:
The mobility of grain growth M is proportional to the net driving force, which is equal to the
= M FP −from
driving force of grain growth (FP )Vsubtracted ( )
FR the Zener pinning force (FR ), which was
(7) expressed
by Reference [43]:
V = M ( FP − FR ) (7)

whereVVisisthe
where thegrain
grainboundary
boundary velocity.
velocity. Figure
Figure 99 shows
shows the
thegrain
grainboundary
boundaryvelocity
velocityininterms
terms ofof
thethe
net
net driving force (F – F ), where the grain boundary velocity was proportional
driving force (FP – FR ), where the grain boundary velocity was proportional to the net driving force.
P R to the net driving
force.
Here,Here,
the ratiothe ratio
of theofgrain
the grain boundary
boundary velocity
velocity to the to
netthe net driving
driving force corresponded
force corresponded to the
to the mobility
mobility
of grainsofM, grains
that M, thatslope,
is, the is, theand
slope,
theand the corresponding
corresponding values ofvalues of mobility
mobility areinlisted
are listed Tablein6.Table
It was
6.observed
It was observed that increasing the temperature resulted in increases in mobility
that increasing the temperature resulted in increases in mobility when 1.5–22 ppm of Mg when 1.5–22 ppm
ofwere
Mg were
added added because
because the mobility
the mobility waswas dependent
dependent on on
thethe growth
growth rate
rate VV alsobeing
also beingproportional
proportionalto
Q𝑄𝑄
to exp(−RT ) in Equation (4). Here, we observed that an increase in Mg content led to a decrease ininthe
exp( − ) in Equation (4). Here, we observed that an increase in Mg content led to a decrease the
𝑅𝑅𝑅𝑅
◦ ◦
mobility
mobilityatat1150 1150°CCasaswell
wellasasatat1200
1200and and1250
1250°CCexcept
exceptfor
for5.6
5.6ppm
ppmMg Mg content,
content, as
as shown
shown in in Table
Table 6.
−−4
4 × m44 /(kJ·s) at 1150 ◦ C,
6. The grain boundary mobility was significantly reduced from 19.8 to 1.3 × 10 × m /(kJ∙s) at 1150 °C,
The grain boundary mobility was significantly reduced from 19.8 to 1.3 × 10
from
from52.6
52.6toto 1.71.7 × −410×−m
× 10
4 4× m4 /(kJ·s) at 1200 ◦ C, and from 105 to 3.7
/(kJ∙s) at 1200 °C, and from 105 to 3.7 × 10−4 × × m410
−4 4 /(kJ·s) at 1250 ◦ C,
/(kJ∙s)×atm1250 °C, when the
Mg content was increased from 1.5 to 22 ppm.
when the Mg content was increased from 1.5 to 22 ppm.

Metals 2019, 9, x FOR PEER REVIEW 11 of 14

(a) (b)

(c) (d)
Figure
Figure9. 9.
TheThegrain boundary
grain boundary velocity (V)
velocity asas
(V) a afunction
functionofofnet
netdriving
drivingforce
force(F(F
PP− F−R)Ffor (a) (a)
R ) for 1.51.5
ppm,ppm,
(b)(b)
5.65.6
ppm,
ppm, (c)(c)
1111
ppmppmand (d)(d)
and 2222
ppm
ppm Mg,
Mg,where
where FPFis
P the
is driving
the drivingforce of
force grain
of graingrowth,
growth, andandF RFthe
R the
retarding
retardingforce
forceonon
boundaries.
boundaries.

Table 6. The values of mobility [10−4 × m4 /(kJ·s)] for 1.5–22 ppm Mg at temperatures of 1150, 1200 and
Table 6.◦ The values of mobility [10−4 × m4/(kJ∙s)] for 1.5–22 ppm Mg at temperatures of 1150, 1200 and
1250 C.
1250 °C.
Temperature ( oC)o
Mg (ppm) Temperature ( C)
Mg (ppm)
1150
1150 1200
1200 1250
1250
1.51.5 19.8
19.8 52.6
52.6 105
105
5.65.6 8.0
8.0 5.9
5.9 7.5
7.5
11.0 7.7 14.2 21.6
11.0 7.7 14.2 21.6
22.0 1.3 1.7 3.7
22.0 1.3 1.7 3.7

In this study, all measurements of grain size were based on a grooved surface. Grooving has a
pinning effect on the migration of grain boundaries, which is therefore slow. The grain boundary
mobility measured here was the lower limit of the grain boundary mobility.
Metals 2019, 9, 370 11 of 14

In this study, all measurements of grain size were based on a grooved surface. Grooving has
a pinning effect on the migration of grain boundaries, which is therefore slow. The grain boundary
mobility measured here was the lower limit of the grain boundary mobility.

3.4. Effect of Solute Drag on the Grain Mobility of Austenite

A lot of studies have reported that solute elements can segregate from the matrix to the grain
boundary, and grain boundary migration may be limited by the diffusivity of the dragged solute atoms.
The pressure induced by the solute drag is dependent on the boundary velocity [46–50]. According to
Cahn et al., the drag effect of mobility (M) is given by [51]:

−1
1
M( XMg , T ) = ( + α· X Mg ) (8)
MPure

and
Nv (kT )2 δ E E
α= Int
(sinh( b ) − ( b )) (9)
Eb D Mg kT kT

where Mpure is the mobility of grain boundary in Fe; XMg is the concentration of Mg; δ is the width of a
grain boundary; Nv is the number of atoms per unit volume; k is the Boltzmann’s constant, and Eb is
the binding energy to the Fe grain boundary. The Mpure is according to Kang et al. [52]:

gb
DFe Vm δ
M pure = β (10)
b2 RT
gb
where b is the burger vector; R is the gas constant; T is the temperature; DFe is the diffusivity of Fe
atoms in ferrite, and Vm is the molar value of ferrite. The values of all the parameters used are listed
◦
in Table 7. The mobility calculation result was 1.8 × 10−3 × m4 /(kJ·s) at 1250 C for 22 ppm Mg.
According to the pinning force results, the mobility of grain boundary was 3.4 × 10−4 × m4 /(kJ·s).
Comparing the results, the effect of solute drag on grain boundary mobility is an order higher than the
effect of Zener pinning. This implies that the effect of Zener pinning is the dominant mechanism for
grain boundary migration.

Table 7. Values of parameters used in the calculation.

Parameter Value Reference

δ 1 × 10−9 m [53]
b 2.48 × 10−10 m [53]
Eb 85 kJ/mol [54]
β 0.7 [49]
Vm 8.26 x10−5 m3 -
1.8·10−4 m2 /s -
g.b −
DFe 1.5·10 4 m2 /s [55]

4. Conclusions
In this study, the effect of Mg addition in the range of 1.5–22 ppm on the grain growth behavior of
austenite in SS400 steel was investigated at 1150, 1200, and 1250 ◦ C. The grain size distribution exhibited
a log-normal distribution. The grain boundary velocity and grain mobility were reduced as Mg was
increased from 1.5 ppm to 22 ppm. As the temperature increased from 1150 to 1250 ◦ C, the grain
boundary velocity also increased. The inclusion with the most retarding force was MgAl2 O4 , and it
was Mg-Al-MnS when 11 and 22 ppm of Mg were added. The mobility of the grain boundary affected
by Zener pinning was one order smaller than that affected by solute drag. It was therefore concluded
that Zener pinning is a dominant factor by which to effectively retard grain boundary migration.
Metals 2019, 9, 370 12 of 14

Author Contributions: Conceptualization, J.-C.K. and Y.-H.F.S.; methodology, C.-T.L.; software, H.-H.L.;
validation, J.-C.K., Y.H. and H.-T.L.; formal analysis, H.-H.L.; investigation, Y.-H.F.S.; resources, H.-T.L.; data
curation, C.-T.L.; writing—original draft preparation, H.-H.L.; writing—review and editing, C.-K.L.; supervision,
J.-C.K.; project administration, F.-Y.H.; funding acquisition, Y.-H.F.S.
Funding: This research was funded by the Ministry of Science and Technology, grant number MOST
103-2622-E-006-037 and MOST 107-2221-E-006-018” and “The APC was funded by the Ministry of Science and
Technology”.
Acknowledgments: The authors would like to thank the Ministry of Science and Technology and China Steel
Corporation for supporting the funding.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Thewlis, G. Transformation Kinetics of Ferrous Weld Metals. Mater. Sci. Technol. 1994, 10, 110–125. [CrossRef]
2. Bramfitt, B.L. The effect of carbide and nitride additions on the heterogeneous nucleation behavior of liquid
iron. Metall. Trans. 1970, 1, 1987–1995. [CrossRef]
3. Zhang, S.H.; Hattori, N.; Enomoto, M.; Tarui, T. Ferrite Nucleation at Ceramic/Austentie Interfaces. ISIJ Int.
1996, 36, 1301–1309. [CrossRef]
4. Jin, H.H.; Shim, J.H.; Chao, Y.W.; Lee, H.C. Formation of Intragranular Acicular Ferrite Grains in a
Ti-containing Low Carbon Steel. ISIJ Int. 2003, 43, 1111–1113. [CrossRef]
5. Kojima, A.; Kiyose, A.; Uemori, R.; Minagawa, M.; Hoshino, M.; Nakashima, T.; Ishida, K.; Yasui, H. Super
High HAZ Toughness Techbology with Fine Microstructure Imoarted by Fine Particles. Nippon Steel Tech.
Rep. 2004, 380, 2–5.
6. Chen, Z.; Loretto, M.H.; Cochrane, R.C. Nature of Large Precipitates in Titanium-containing HSLA Steels.
Mater. Sci. Technol. 1987, 3, 836–844. [CrossRef]
7. Prikryl, M.; Kroupa, A.; Weatherly, G.C.; Subramanian, S.V. Precipitation behavior in a Medium Carbon,
Ti-V-N Microalloyed Steel. Metall. Mater. Trans. A 1996, 27, 1149–1165. [CrossRef]
8. Yan, W.; Shan, Y.Y.; Yang, K. Effect of TiN Inclusions on the Impact Toughness of Low-carbon Microalloyed
Steels. Metall. Mater. Trans. A 2006, 37, 2147–2158. [CrossRef]
9. Baker, T.N. Role of Zicronium in Microalloyed Steels: A Review. Mater. Sci. Technol. 2015, 31, 265–294.
[CrossRef]
10. Mizoguchi, S.; Takamura, J. Roles of Oxides in Steels Performance-Metallurgy of Oxides in Steels.
In Proceedings of the 6th International Iron Steel Congress, Nagoya, Japan, 21–26 October 1990; Volume 1,
pp. 591–597.
11. Grong, O.; Matlock, D.K. Microstructural Development in Mild and Low-alloy Steel Weld Metals. Int. Mater.
Rev. 1986, 31, 27–48. [CrossRef]
12. Shim, J.H.; Oh, Y.J.; Suh, J.Y.; Cho, Y.W.; Shim, J.D.; Byun, J.S.; Lee, D.N. Ferrite Nucleation Potency of
Non-metallic Inclusions in Mediun Carbon Steels. Acta Mater. 2001, 49, 2115–2122. [CrossRef]
13. Byun, J.S.; Shim, J.H.; Cho, Y.W.; Lee, D.N. Non-metallic inclusion and intragranular nucleation of ferrite in
Ti-killed C-Mn Steel. Acta Mater. 2003, 51, 1593–1606. [CrossRef]
14. Barbaro, F.J.; Krauklis, P.; Easterling, K.E. Formation of Acicular Ferrite at Oxide Particles in Steels.
Mater. Sci. Technol. 1989, 5, 1057–1068. [CrossRef]
15. Mills, A.R.; Thewlis, G.; Whitemen, J.A. Nature of Inclusions in Steel Weld Metals and their Influence on
Formation of Acicular Ferrite. Mater. Sci. Technol. 1987, 3, 1051–1061. [CrossRef]
16. Mundra, K.; DebRoy, T.; Babu, S.S.; David, S.A. Towards Predicting Weld Metal Microstructure from Fundamentals
of Transport Phenomena; Department of Energy: Washington, DC, USA, 1995.
17. Goto, H.; Miyazawa, K.; Amada, W.Y.; Tanaka, K. Effect of Cooling Rate on Composition of Oxides
Precipitated during Solidification of Steels. ISIJ Int. 1995, 35, 708–714. [CrossRef]
18. Chao, Y.W.; Byun, J.S.; Shim, J.H. Effect of Ti Addition on Mixed Microstructure of Allotriomorphic and
Bainitic Ferrite in Wrought C-Mn Steels. Mater. Sci. Forum 2003, 426–432, 1511–1516. [CrossRef]
19. Byun, J.S.; Shim, J.H.; Cho, Y.W. Influence of Mn on Microstructural Evolution in Ti-Killed C-Mn Steel.
Scr. Mater. 2003, 48, 449–454. [CrossRef]
20. Kim, H.S.; Chang, C.H.; Lee, H.G. Evolution of Inclusions and Resultant Microstructural change with Mg
Addition in Mn/Si/Ti Deoxidized Steels. Scr. Mater. 2005, 53, 1253–1258. [CrossRef]
Metals 2019, 9, 370 13 of 14

21. Smith, C. Zener pining. Met. Technol. Trans. Metall. Soc. 1948, 175, 15–51.
22. Nes, E.; Ryum, N.; Hunderi, O. On the Zener Drag. Acta Metall. 1985, 1, 11–22. [CrossRef]
23. Schwarze, C.; Kamachali, R.D.; Steinbach, I. Phase-field study of Zener Drag and Pinning Cylindrical
Particles in Polycrystalline Materials. Acta Mater. 2016, 106, 59–65. [CrossRef]
24. Sakata, K.; Suito, H. Grain-growth-inhibiting Effects of Primary Inclusion Particles of ZrO2 and MgO in
Fe-10 mass pct Ni Alloy. Metall. Mater. Trans. A 2000, 31, 1213–1223. [CrossRef]
25. Ohta, H.; Suito, H. Effect of Nonrandomly Dispersed Particles on Austenite Grain Growth in Fe-10 mass%
Ni and Fe-0.20 mass% C-0.02 mass% P Alloys. ISIJ Int. 2006, 46, 832–839. [CrossRef]
26. Karasev, A.V.; Suito, H. Effect of Particle Size Distribution on Austenite Grain Growth in Fe–0.05mass%C
Alloy Deoxidized with Mn–Si, Ti, Mg, Zr and Ce. ISIJ Int. 2006, 46, 718–727. [CrossRef]
27. Karasev, A.V.; Suito, H. Effects of Oxide Particles and Solute Elements on Austenite Grain Growth in
Fe–0.05mass%C and Fe–10mass%Ni Alloys. ISIJ Int. 2008, 48, 658–666. [CrossRef]
28. Zhu, K.; Yang, J.; Wang, R.Z.; Yang, Z.G. Effect of Mg Addition on Inhibiting Austenite Grain Growth in
Heat Affected Zones of Ti-bearing Low Carbon Steels. J. Iron Steel Res. Int. 2011, 18, 60–64. [CrossRef]
29. Zhu, K.; Yang, Z.G. Effect of Magnesium on Austenite Grain Growth of the Heat-affected Zone in Low-carbon
High-strength Steels. Metall. Mater. Trans. A 2011, 42, 2207–2213. [CrossRef]
30. Zhu, K.; Yang, Z.G. Effect of Mg Addition on the Ferrite Grain Boundaries Misorientation in HAZ of Low
Carbon Steels. J. Mater. Sci. Technol. 2011, 27, 252–256. [CrossRef]
31. Sarma, D.S.; Karasev, A.V.; Jonsson, P.G. On the Role of Non-metallic Inclusions in the Nucleation of Acicular
Ferrite in Steels. ISIJ Int. 2009, 49, 1063–1074. [CrossRef]
32. Beck, P.A. Annealing of cold worded metals. Adv. Phys. 1954, 3, 245–324. [CrossRef]
33. Sellars, C.M.; Whiteman, J.A. Recrystallization and Grain Growth in Hot Rolling. Met. Sci. 1979, 13, 187–194.
[CrossRef]
34. Sellars, C.M.; Davies, G.J. Hot Working and Forming Process; The Society: London, UK, 1990; pp. 3–15.
35. Sellars, C.M. Modelling of Structural Evolution during Hot Working Processes; Risol National Laboratory:
Roskilde, Denmark, 1986; pp. 167–187.
36. Hodgson, P.D.; Gibbs, R.K. A Mathematical Model to Predict the Mechanical Properties of Hot Rolled C-Mn
and Microalloyed Steels. ISIJ Int. 1992, 32, 1329–1338. [CrossRef]
37. Anelli, E. Application of Mathematical Modelling to Hot Rolling and Controlled Cooling of Wire Rods and
Bars. ISIJ Int. 1992, 32, 440–449. [CrossRef]
38. Yoshie, A.; Fujioka, M.; Watanabe, Y.; Nishioka, K.; Morikawa, H. Modelling of Microstructural Evolution
and Mechanical Properties of Steel Plates Produced by Thermo-Mechanical Control Process. ISIJ Int. 1992,
32, 395–404. [CrossRef]
39. Nishizawa, T. Grain Growth in Single- and Dual-Phase Steels. Tetsu-to-hagané 1984, 70, 1984–1992. [CrossRef]
40. Namba, S.; Kitamura, M.; Shimada, M.; Katsumata, M.; Inoue, T.; Imamura, H.; Maeda, Y.; Hattori, S.
Prediction of Microstructure Distribution in the Through-thickness Direction during and after Hot Rolling in
Carbon Steels. ISIJ Int. 1992, 32, 377–386. [CrossRef]
41. Senuma, T.; Yada, H. Microstructural Evolution of Plain Carbon Steels in Multiple Hot Working; Risol Natl Lab:
Roskidle, Demark, 1986; pp. 547–552.
42. Atkinson, H.V. Overview no. 65: Theories of Normal Grain Growth in Pure Single Phase Systems. Acta Metall.
1988, 36, 469–491. [CrossRef]
43. Burke, J.; Turnbull, D. Recrystallization and Grain Growth. Process Met. Phys. 1952, 3, 220–292. [CrossRef]
44. Vynokur, B.B. Influence of Alloying on the Free Energy of Austenitic Grain Boundaries in Steel. Mater. Sci.
1996, 32, 448–455. [CrossRef]
45. Luo, S.J.; Su, Y.H.; Lu, M.J.; Kuo, J.C. EBSD Analysis of Magnesium Addition on Inclusion Formation in
SS400 Structural Steel. Mater. Charct. 2013, 82, 103–112. [CrossRef]
46. Masayo, S. An Analysis of the Solute Drag Effect of Nb on Recrystallization of Ultra Low Carbon Steel.
ISIJ Int. 1998, 38, 547–552.
47. Enomoto, M. Influence of Solute Drag on the Growth of Proeutectoid Ferrite in Fe-C-Mn Alloy. Acta Mater.
1999, 47, 3533–3540. [CrossRef]
48. Suehiro, M.; Liu, Z.K.; Agren, J. Effect of Niobium on Massive Transformation in Ultra Low Carbon Steels:
A Solute Drag Treatment. Acta Mater. 1996, 44, 4241–4251. [CrossRef]
Metals 2019, 9, 370 14 of 14

49. Kim, S.G.; Park, Y.B. Grain Boundary Segregation, Solute Drag and Abnormal Grain Growth. Acta Mater.
2008, 56, 3739. [CrossRef]
50. Yu, Q.; Sun, Y. Abnormal Growth of Austenite Grain of Low-carbon Stele. Mater. Sci. Eng. A 2006, 420, 34–38.
[CrossRef]
51. Cahn, J.W. The Impurity-drag Effect in Grain Boundary Motion. Acta Metall. 1962, 10, 789–798. [CrossRef]
52. Kang, K.B.; Kwon, O.; Lee, W.B.; Park, C.G. Effect of Precipitation on the Recrystallization Behavior of a Nb
Containing Steel. Scr. Mater. 1997, 36, 1303–1308. [CrossRef]
53. Sinclair, C.W.; Hutchinson, C.R.; Brechet, Y. The Effect of Nb on the Recrystallization and Grain Growth of
Ultra-High-Purity α-Fe: A Combinatorial Approach. Metal. Mater. Trans. A 2007, 38, 821–830. [CrossRef]
54. Amram, D.; Schuh, C.A. Interplay between Thermodynamic and Kinetic Stabilization Mechanisms in
Nanocrystalline Fe-Mg Alloys. Acta Mater. 2018, 144, 447–458. [CrossRef]
55. Bornstein, L. Numerical Data and Functional Relationships in Science and Technology; Springer: Berlin, Germany, 1987.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).