metals

Article
On the Microstructure Development under Cyclic Temperature
Conditions during WAAM of Microalloyed Steels
Chang Huang 1 , Mohamed Soliman 1,2, * , Kai Treutler 3 , Volker Wesling 3 and Karl-Heinz Spitzer 1, *

1 Institute of Metallurgy, Clausthal University of Technology, 38678 Clausthal-Zellerfeld, Germany

2 Faculty of Engineering, Galala University, Galala City 43511, Egypt
3 Institute of Welding and Machining, Clausthal University of Technology, 38678 Clausthal-Zellerfeld, Germany
* Correspondence: mohamed.soliman@tu-clausthal.de (M.S.); karl.heinz.spitzer@tu-clausthal.de (K.-H.S.)

Abstract: This paper shed light on the kinetics of transformation and the developed microstructure
during wire arc additive manufacturing (WAAM). Three microalloyed alloys, two of them are high
strength low alloyed steel (HSLA) grades and the third is a Ni-Cr-Mo steel, from which the welding
wires are being produced, were investigated. Repeated cycles around varied temperatures from a
reheating temperature of 1350 ◦ C and down to a temperature 35 ◦ C below the Ae1 are applied using
dilatometer on samples from the steels. After applying the cycles, the dilatometric-samples were
investigated metallographically and their macro- and microhardness values were measured. It is
shown that the WAAM using HSLA steels produce softer structure than the steel of the welding
wire. Combined microalloying with Ti and Nb can present a useful strategy for producing finer
structure in the WAAM components due to the effect of Ti in inhibiting the prior austenite grain-
growth and that of Nb in refining the final structure. Additionally, repeated heating near Ae3 refines
the prior austenite grains and produced fine ferrite-pearlite structure in case of HSLA steels and
a microstructure predominated by the granular bainite in case of welding wire alloy. The former
microstructure was the softest one for the case of HSLA steels, whereas the softest structure in case of
the welding wire alloy was the tempered martensite structure developed by reheating below Ae1 .
Citation: Huang, C.; Soliman, M.;
Idealized temperature curves were chosen for the heat treatment, which could be characterized in
Treutler, K.; Wesling, V.; Spitzer, K.-H.
On the Microstructure Development
a well-defined manner. In future work such idealized curves together with temperature histories
under Cyclic Temperature Conditions obtained in WAAM-process will be used to set up a database to train an AI-model for predicting
during WAAM of Microalloyed structure and material properties.
Steels. Metals 2022, 12, 1913. https://
doi.org/10.3390/met12111913 Keywords: wire arc additive manufacturing (WAAM); high strength low alloyed steels (HSLA);
microalloyed steels; microstructure development; dilatometry; thermal cycle
Academic Editor: Andrea Di Schino

Received: 7 October 2022

Accepted: 2 November 2022
Published: 8 November 2022 1. Introduction
Publisher’s Note: MDPI stays neutral New developments in automation and computer-aided design have led to a renais-
with regard to jurisdictional claims in sance in welding process-based additive manufacturing, formerly known as “shape weld-
published maps and institutional affil- ing” [1]. This is accelerated by further developments based on control and sensor technol-
iations. ogy [2–7] and path planning algorithms [8]. Wire and Arc Additive Manufacturing WAAM
has established itself as one of the main methods of welding process-based additive manu-
facturing. This includes all arc welding processes that use a wire-shaped filler metal, which
allow a wide range of potential materials, e.g., steels [9–11], nickel-based [12], complex
Copyright: © 2022 by the authors.
concentrated alloys [13] and many others to be manufactured. A wide-ranging overview of
Licensee MDPI, Basel, Switzerland.
the research status of these processes to date is given in [1,14,15] and others.
This article is an open access article
A key advantage of this additive manufacturing process is the high deposition rate
distributed under the terms and
compared to laser and powder-based processes. However, this leads to a reduced surface
conditions of the Creative Commons
Attribution (CC BY) license (https://
quality of the components, which must be finished using conventional subtractive man-
creativecommons.org/licenses/by/
ufacturing processes such as milling. To simplify this finishing process, new materials
4.0/).

Metals 2022, 12, 1913. https://doi.org/10.3390/met12111913 https://www.mdpi.com/journal/metals

Metals 2022, 12, 1913 2 of 16

with improved cutting properties are currently being developed for additive manufactur-
ing (AM) [12,16–18]. Another advantage of WAAM is the relatively simple realization of
multi-material components, also called functionally graded materials. These components
have different properties based on different heat management [19–21] or the use of differ-
ent materials [22–24]. However, the key point here is the knowledge of the relationships
between the alloy composition of the metal used, the heat management in the additive
manufacturing process and the resulting properties.
The understanding of the microstructure development during additive manufacturing
is the main point for the realization of certain material properties. In case of continuous
cooling or heating, the metastable microstructure formation is described by the associated
CCT- or the CHT-diagram, respectively. In general, however, this data cannot be transferred
directly to conditions in additive manufacturing. Repeating thermal cycles around certain
temperatures lead to a different microstructure.
The influence of a cyclical temperature development on the structure formation dur-
ing welding was analyzed by Samardžić et. al. [25] studied on a welding simulator. This
simulator allows the application of transient, in particular cyclic, temperature profiles.
The resulting dynamic transformation temperatures were determined by dilatometer mea-
surements. Furthermore, microstructure investigations and hardness measurements were
carried out. Celin et al. [26] investigated the microstructure development in different areas
of the heat-affected zone during multi-path welding of a fillet weld for a high-strength
fine-grain structural steel. The multi-layer welding process results in thermal cycles that
correspond to those in the WAAM process. The microstructure during the real welding
process was compared with that of dilatometer simulations, in which a temperature profile
similar to that during the welding process was applied. From the results it was concluded
that dilatometer investigations are well suited for simulating microstructure development
in the heat-affected zone under cycling temperature conditions. The microstructure de-
velopment in general during the WAAM process was investigated by several working
groups [27–31].
Mathematical models for the microstructure and the resulting property are of great
importance for the process design and monitoring of the WAAM process. Data-driven
Artificial Intelligence (AI) based models (e.g., neural networks) are increasingly playing a
major role. AI-techniques have been used to predict CCT diagrams [32–34]. However, it is
repeatedly pointed out [35] that a direct transfer of the results of CCT diagram models to
the microstructure development in additive manufacturing is not possible due to the special
temperature history. Nevertheless, the modeling of CCT diagrams is a useful starting point
for modeling the microstructure development in additive manufacturing.
In order to be able to predict the relation between the temperature history and the ma-
terial properties for a given alloy using an artificial intelligence, AI-model, the temperature
profile must be classified and quantified. For this purpose, the duration and the number of
cycles when the temperature is in certain temperature ranges during heating and cooling
can be considered. The corresponding temperature range selection can be based on the
transformation temperatures Ae1 and Ae3 . As a starting point for the development of such
a classification method, it makes sense to use exemplary temperature profiles with defined
cycle numbers, holding temperatures, holding times and heating/cooling speeds.
In this work, dilatometer trials have been performed. Exemplary temperature curves
were chosen which could be characterized in the described manner. Critical phase transfor-
mation temperatures were determined, the microstructure was analyzed and the hardness
was measured. These thermal cycle simulations were performed on two High Strength Low
Alloyed (HSLA) microalloyed steels and on a steel from which welding wires are made.
Such data shall serve as a starting point to build up a database for an AI microstructure
and material property model for the WAAM process.
Metals 2022, 12, 1913 3 of 16
Metals 2022, 12, x FOR PEER REVIEW  3  of  17 
 

2. Experimental Procedure
2. Experimental Procedure 
2.1. Investigated Materials
2.1. Investigated Materials 
The thermal-cycle simulations were applied on three microalloyed steels with chemical
The  thermal‐cycle 
compositions simulations 
given in Table 1. Steels Awere 
and applied  on strength
B are high three  microalloyed 
low alloyed steels 
(HSLA)with 
steels.
chemical 
Steel compositions with
A is microalloyed given  in  Table 
Nb, while 1.  Steels 
steel B isA microalloyed
and  B  are  high 
withstrength 
Ti andlow 
Nb.alloyed 
Steel C is
(HSLA) steels. Steel A is microalloyed with Nb, while steel B is microalloyed with Ti and 
alloyed with Ni, Cr and Mo and microalloyed with Ti and is used for the production of
Nb. Steel C is alloyed with Ni, Cr and Mo and microalloyed with Ti and is used for the 
welding wires.
production of welding wires. 
Table 1. Chemical composition (wt.%) of the steels used in this work.
Table 1. Chemical composition (wt.%) of the steels used in this work. 
Alloy C Si Mn Al Cr Ni Mo Ti Nb N
Alloy 
A
C 
0.059
Si 
0.023
Mn 
0.766
Al 
0.039
Cr 
0.039
Ni 
0.033
Mo 0.003
Ti 
0.002
Nb 
0.031
N 0.003
B A  0.0640.059  0.023  0.766 
0.028 1.494 0.039 
0.049 0.039 
0.041 0.033 
0.032 0.003 
0.008 0.002 
0.030 0.031 
0.059 0.003 
0.005
C B  0.1330.064  0.028  1.494 
0.714 1.36 0.049 
0.0056 0.041 
0.340 0.032 
2.24 0.008 
0.563 0.030 
0.033 0.059 - 0.005 
0.005
C iron. 0.133 
Balance 0.714  1.36  0.0056  0.340  2.24  0.563  0.033  ‐  0.005 
balance iron. 
2.2. Dilatometry and Heat Treatment
2.2. Dilatometry and Heat Treatment 
For applying the thermal cycles and dilatometric study, a Dil 805A dilatometer was
used. The For applying the thermal cycles and dilatometric study, a Dil 805A dilatometer was 
simulations were performed on cylindrical samples of 4 mm diameter and 10 mm
used. The simulations were performed on cylindrical samples of 4 mm diameter and 10 
length. Sheathed type S “Pt/Pt-10% Rh” thermocouples with a nominal diameter of 0.1 mm
mm length. Sheathed type S “Pt/Pt‐10% Rh” thermocouples with a nominal diameter of 
were individually spot welded to the specimens surface in central position. Each sample
0.1 mm were individually spot welded to the specimens surface in central position. Each 
was held between two Al2 O3 rods, with its 10-mm side along the rods. The thermal cycles
were performed under a vacuum2Oof
sample was held between two Al 3 rods, with its 10‐mm side along the rods. The thermal 
0.005 Pa, and helium was used for cooling. Further
cycles on
details were  performed 
dilatometric under  a  vacuum 
measurements of  0.005 
using Pa,  and 
Dil 805A helium 
is given in was 
[36].used 
Afterfor  cooling. the
applying
Further  details  on  dilatometric  measurements  using  Dil  805A 
thermal cycles in the dilatometer, the dilatometric samples were used for investigating is  given  in  [36].  After 
applying the thermal  cycles  in  the  dilatometer, the  dilatometric 
the effect of the applied heat treatment on microstructure development and mechanical samples  were used  for 
investigating the effect of the applied heat treatment on microstructure development and 
characteristics in terms of hardness values.
mechanical characteristics in terms of hardness values. 
The specimens were subjected to schedules shown in Figure 1. In these schedules, the
The specimens were subjected to schedules shown in Figure 1. In these schedules, the 
specimens were heated up to 1350 ◦ C and held for 3 min. The main objective of this step is
specimens were heated up to 1350 °C and held for 3 min. The main objective of this step 
to allow for the metal carbides to dissolve in austenite and to obtain microstructures that are
is to allow for the metal carbides to dissolve in austenite and to obtain microstructures 
similar to that obtain when cooling from the melt-status. To investigate the effect of thermal
that are similar to that obtain when cooling from the melt‐status. To investigate the effect 
cycles on thecycles 
of  thermal  microstructure development,
on  the  microstructure  three schedules
development,  three were applied.
schedules  were Schedules
applied.  II
and III simulate two and four repeated cycles of adding layers during the WAAM process,
Schedules II and III simulate two and four repeated cycles of adding layers during the 
respectively by heating to temperature TB and holding forB and holding for 30 s. After the 
WAAM process, respectively by heating to temperature T 30 s. After the last thermal cycle,
the samples are cooled to room temperature. The heating rate to TA and TB was 100
last thermal cycle, the samples are cooled to room temperature. The heating rate to T Ks−1 .
A and 
InTall cycles
B  was  100  the
Ks−1. holding temperature,
In  all  cycles  the  holding  TBtemperature, 
is related toTthe equilibrium
B  is  related  to  the transformation
equilibrium 
temperatures A and
transformation temperatures A
e1 A e3 as given in Table 2. These temperatures of the three alloys
e1 and Ae3 as given in Table 2. These temperatures of the 
arethree alloys are calculated by using the software ThermoCalc and applying the database 
calculated by using the software ThermoCalc and applying the database TCFE9. The
natural cooling in all schedules is simulated by applying a t8/5 time8/5 time of 18 s. The 
TCFE9. The natural cooling in all schedules is simulated by applying a t of 18 s. The t8/5 is
the
t8/5required time for cooling a weld bead and its heat-affected zone to pass through the
 is the required time for cooling a weld bead and its heat‐affected zone to pass through 
temperature range range 
the  temperature  from 800 ◦ C800 
from  °C  ◦to 
to 500 C.500 
At this temperature-interval,
°C.  At  the most
this  temperature‐interval,  important
the  most 
important structural changes occur in the steel. 
structural changes occur in the steel.

(a)   (b)   (c)  T = 1350 °C

TA = 1350 °C TA = 1350 °C A
Temperature

Temperature

Temperature

TB TB TB TB TB TB

300 °C  300 °C 

Schedule I Schedule II Schedule III

Time Time Time
 
  Figure 1. Applied thermal schedules.

 
Metals 2022, 12, 1913 4 of 16

Table 2. Predicted critical temperature and the temperature TB (◦ C) applied in different trials.

Temp. A B C
Ae1 700 675 570
Ae
Ae3 855 830 785
(1350 + Ae3 )/2 1100 1090 1070
Ae3 + 35 890 865 820
TB Ae3 − 25 830 805 -
50% austenite 780 - 725
Ae1 − 35 665 640 535

2.3. Microstructure Investigations

Light optical microscopy of the samples, after applying various processing conditions,
were performed on sections cut at central position parallel to the longitudinal axes. After
mounting, the specimens were prepared by mechanical grinding using abrasive grinding
papers ranging from coarse, 180 grit number to fine 1000 grit number followed by polishing
up to 0.05 µm alumina. The microstructure was revealed by etching with 2% Nital. After
etching, the samples were rinsed with ethyl alcohol and dried under a warm air drier.
The different microstructural constituents observed in the current study were classified
according to the features given in [37].

2.4. Hardness Test

In this study both macro- and microhardness measurements were performed on
as-received, and heat-treated samples. Macro-hardness testing was carried out at room
temperature using Vickers hardness according to ASTM E384 applying a load of 30 kg
(HV30). Wolpert hardness tester equipped with a digital measurement system enabling data
transmission to PC was used. The hardness data are the average of at least 4 measurements.
For microhardness measurements, the Mitutoyo microhardness tester with an optical
microscope was used. The hardness of the phases was measured with a Vickers hardness
HV0.025 (25 g load, and 15 s duration time).

3. Results and Discussions

3.1. Microstructure Charactersitics
3.1.1. As Received Microstructure
Figure 2 shows the as-received microstructure of the three alloys. The microstructure
of alloy A is basically consisting of polygonal ferrite (PF) and pearlite (P), with average
ferrite grain size of 5.7 µm. The as-received alloy B has a microstructure consisting of
granular bainite (GB), PF and P. The GB is characterized by a non-smooth outer boundary
and interior area featured with carbide phases dispersed within the grain, whereas the PF
grains have smoother outer boundary and featureless interior area [37]. The term granular
bainite originated from the granular appearance of the carbides within this type of bainite.
The GB is essentially the main microstructure of the C alloy. Indeed, this form of bainite
is reported to form habitually in alloy steel like Ni-Cr-Mo steels (refer to composition of
C in Table 1), that have been cooled at slow-to-intermediate rates and to have dispersed
islands of retained austenite and martensite (M/A islands) [38]. The M/A islands are
correspondingly observable in the microstructure of C and marked with arrows within the
GB structure in Figure 2c. These islands are formed during continuous cooling from the
untransformed austenite with different austenite to martensite ratios. The abundance of
the granular carbide phase in the GB of alloy C compared to alloy B is related to the higher
content of carbide forming elements in the former alloy. The high level of alloying in C
increased its hardenability and therefore suppressed the formation of ferrite and pearlite
during cooling.
 Metals 2022, 12, x FOR PEER REVIEW  5  of  17 
 

Metals 2022, 12, 1913 5 of 16

in  C  increased  its  hardenability  and  therefore  suppressed  the  formation  of  ferrite  and 
pearlite during cooling. 

 
Figure 2. As-received microstructures of alloys A, B and C.
Figure 2. As‐received microstructures of alloys A, B and C. 
3.1.2. Microstructure Development during Reheating and Austenitization
3.1.2. Microstructure Development during Reheating and Austenitization 
During heating, the as-received structure transforms to austenite. The very high
heating During 
rate of heating, 
100 Ksthe −1 as‐received 
simulating structure 
the heating transforms 
during to  theaustenite.  The  very shifts
WAAM process high  the
heating rate of 100 Ks −1 simulating the heating during the WAAM process shifts the critical 
critical temperatures, Ac1 and Ac3 , of the three alloys to higher temperatures compared to
temperatures,  Ac1  and  Ac3,  of  the  three  alloys  to  higher  temperatures  compared  to  the 
the corresponding equilibrium temperatures Ae1 and Ae3 . The Ac1 and Ac3 of the three
corresponding equilibrium temperatures Ae1 and Ae3. The Ac1 and Ac3 of the three alloys 
alloys are illustrated in Figure 3a. The increased alloying contents from alloy A to C that
are illustrated in Figure 3a. The increased alloying contents from alloy A to C that resulted 
resulted in decreasing Ae1 and Ae3 (Table 2) has a similar effect on Ac1 and Ac3 as shown
in decreasing  Ae1  and Ae3 (Table 2) has a  similar effect  on Ac1 and  Ac3  as shown  in  this 
in this figure. In addition to the austenite formation, the continuous heating and holding
figure. In addition to the austenite formation, the continuous heating and holding at 1350 
at 1350 ◦ C results in dissolving of the NbC precipitates in alloys A and B as shown in
°C results in dissolving of the NbC precipitates in alloys A and B as shown in Figure 3b. 
Figure 3b. Nevertheless, the TiN formed in the Ti microalloyed steels B and C, has a high
Nevertheless, the TiN formed in the Ti microalloyed steels B and C, has a high solubility 
solubility temperature;
temperature;  therefore, therefore, an undissolved
an  undissolved  quantity quantity
of  its  of its particles
particles  remains
remains  in the
in  the 
microstructure ◦
during the austenitization stage at 1350 C. This is important for limiting the
microstructure during the austenitization stage at 1350 °C. This is important for limiting 
coarsening of austenite
the  coarsening  grainsgrains 
of  austenite  duringduring 
austenitization via the via 
austenitization  pinning effect [39].
the  pinning  Conversely,
effect  [39]. 
NbC that is completely dissolves during austenitization, precipitates during the subsequent
Conversely, NbC that is completely dissolves during austenitization, precipitates during 
cooling processes and, therefore, plays an important role in refining the final structure.
the subsequent cooling processes and, therefore, plays an important role in refining the 
Accordingly, the mutual microalloying with the Ti and Nb, like in the case of alloy B, is
final structure. Accordingly, the mutual microalloying with the Ti and Nb, like in the case 
common in low carbon steels because of the conjoint reinforcing role that they are playing
of alloy B, is common in low carbon steels because of the conjoint reinforcing role that 
inthey 
refining the microstructure
are  playing  [40].microstructure 
in  refining  the  The Ti forms TiN [40]. that
The limits the coarsening
Ti  forms  of austenite
TiN  that  limits  the 
grains during austenitization and leads to finer prior austenite grains and Nb
coarsening  of  austenite  grains  during  austenitization  and  leads  to  finer  prior  austenite  forms NbC
grains and Nb forms NbC that precipitates during the subsequent cooling processes and 
that precipitates during the subsequent cooling processes and refines the final structure.
In order to reveal the  prior austenite after holding at 1350 ◦ C for 180 s, samples
refines the final structure. 
to room temperature applying a cooling rate of approx. −120 Ks−1 after
In order to reveal the prior austenite after holding at 1350 °C for 180 s, samples were 
were quenched
quenched  to During
austenitizing. room  temperature 
quenching the applying  a  cooling 
prior austenite rate  transforms
grains of  approx.  to
−120  Ks−1  after 
martensite, with
austenitizing. During quenching the prior austenite grains transforms to martensite, with 
grain boundaries corresponding to the prior austenite boundaries. Figure 4 shows the
grain  boundaries 
quenched corresponding 
microstructures. to  the  prior  austenite 
The considerable large prior boundaries. 
austeniteFigure 
grains4 ofshows 
alloythe 
A are
quenched to
attributed microstructures. 
the dissolutionThe  considerable 
of the TiN and NbC large and
prior  austenite 
absence grains  of of
of pinning alloy 
the A  are 
austenite
attributed to the dissolution of the TiN and NbC and absence of pinning of the austenite 
boundaries during heating at 1350 ◦ C (Figure 3b). Alloy B showed the strongest pinning of
boundaries during heating at 1350 °C (Figure 3b). Alloy B showed the strongest pinning 
austenite and therefore the smallest grain sizes. The TiN in alloy C seems to have a partial
of  austenite  and  therefore  the  smallest  grain  sizes.  The  TiN  in  alloy  C  seems  to  have a 
effect in limiting the growth of the austenite grains, although its complete dissolution under
partial  effect  in  limiting  the  growth  of  the  austenite  grains,  although  its  complete 
equilibrium takes place at about 1200 ◦ C. Probably, the equilibrium condition was not
dissolution  under  equilibrium  takes  place  at  about  1200  °C.  Probably,  the  equilibrium 
reached, hence, complete dissolution was delayed. On the other hand, the thermodynamic
condition was not reached, hence, complete dissolution was delayed. On the other hand, 
calculations showed that the Mo and Cr-carbides precipitates are predicted to dissolve
below 800 ◦ C.

 
 Metals 2022, 12, x FOR PEER REVIEW  6  of  17 
 
Metals 2022, 12, x FOR PEER REVIEW  6  of  17 
 

the  thermodynamic  calculations  showed  that  the  Mo  and  Cr‐carbides  precipitates  are 
Metals 2022, 12, 1913 the  6 of 16
thermodynamic  calculations  showed  that  the  Mo  and  Cr‐carbides  precipitates  are 
predicted to dissolve below 800 °C. 
predicted to dissolve below 800 °C. 
(a)  (b) 
(a)  (b) 
1.6
1.6
A
Rel. change in length (%)

1.4
A
Rel. change in length (%)

1.4 B
1.2 B
1.2 C
1 C
1
0.8
0.8
0.6
750 800 850 900 950 1000
0.6
750 800
Temperature (°C)
850 900 950 1000    
Temperature (°C)    
Figure 3. (a) Austenite formation kinetics in terms of relative change in length during heating of 
alloys  A,  B  and  C.  The  temperature  on  the  curves  marks  the  Ac1  and  Ac3  temperatures.  (b) 
Figure 3. (a) Austenite formation kinetics in terms of relative change in length during heating of 
Figure 3. (a) Austenite formation kinetics in terms of relative change in length during heating of alloys
Dependance 
alloys  of  C. 
A,  B  and  the The 
precipitates  volume 
temperature  % curves 
on  the  on  the marks 
annealing 
the  Atemperature  (predicted  using 
c1  and  Ac3  temperatures.  (b) 
A, B and C. The temperature on the curves marks the Ac1 and Ac3 temperatures. (b) Dependance of
ThermoCalc). 
Dependance  of  the  precipitates  volume  %  on  the  annealing  temperature  (predicted  using 
theThermoCalc). 
precipitates volume % on the annealing temperature (predicted using ThermoCalc).

 
◦C  
Figure 4. Comparison of the prior austenite grain sizes after austenitization at 1350 for 180 s. The
Figure 4. Comparison of the prior austenite grain sizes after austenitization at 1350 °C for 180 s. The 
red arrows indicate the grain boundaries in (a–c). In (a,b) the grain boundaries are decorated by
red arrows indicate the grain boundaries in (a–c). In (a,b) the grain boundaries are decorated by 
Figure 4. Comparison of the prior austenite grain sizes after austenitization at 1350 °C for 180 s. The 
allotriomorphic grain boundaries ferrite.
allotriomorphic grain boundaries ferrite. 
red arrows indicate the grain boundaries in (a–c). In (a,b) the grain boundaries are decorated by 
allotriomorphic grain boundaries ferrite. 
3.1.3. Microstructure Development during Applying Schedule I
   
Applying schedule I (see Figure 1) on the three alloys results in the microstructures
   
given in Figure 5. Figure 6 shows a comparison of the transformation kinetics of the three
alloys in terms of length change observed during cooling in the dilatometer investigations.
For alloy A, Widmanstätten ferrite (WF) growing from the grain boundary allotriomor-
phic ferrite (GF) is the major constituent of the microstructure, which also contains GB as a
secondary phase. Referring to the dilatometric curve observed during cooling, it is obvious
that the transformation starts at about 630 ◦ C, which corresponds to the formation of the
GF on the austenite grain boundaries. By continuous cooling the WF grows on the GF
before the formation of GB. The dilatometric curve indicates the start of formation of small
quantity of martensite at about 450 ◦ C. This is a lower martensite-start temperature (Ms) of
alloy A by about 41 ◦ C, compared to that one predicted applying Andrews’ equation [41].
  The enrichment of the untransformed austenite with carbon during the formation of GF,
  WF and GB is the reason for its lower Ms. The GB showed marginally higher microhardness
than WF as shown in Figure 7. On the other hand, B and C alloys exhibited single-phase
microstructure of acicular ferrite (AF) and martensite (M), respectively. The AF poses an
assemblage of interwoven ferrite laths. The disorganized microstructure of the AF gives
 the formation of the GF on the austenite grain boundaries. By continuous cooling the WF 
grows on the GF before the formation of GB. The dilatometric curve indicates the start of 
formation of small quantity of martensite at about 450 °C. This is a lower martensite‐start 
temperature  (Ms)  of  alloy  A  by  about  41  °C,  compared  to  that  one  predicted  applying 
Andrews’  equation  [41].  The  enrichment  of  the  untransformed  austenite  with  carbon 
Metals 2022, 12, 1913 during the formation of GF, WF and GB is the reason for its lower Ms. The GB showed  7 of 16
marginally higher microhardness than WF as shown in Figure 7. On the other hand, B and 
C alloys exhibited single‐phase microstructure of acicular ferrite (AF) and martensite (M), 
itrespectively. The AF poses an assemblage of interwoven ferrite laths. The disorganized 
its larger ability to deflect cracks, thus it is the most favorable structure in many steels
microstructure 
because of  the 
of its high AF  gives 
fracture it  its  larger 
toughness ability 
[37]. Also,to the
deflect 
highcracks,  thus  it density,
dislocation is  the  most 
caused
favorable structure in many steels because of its high fracture toughness [37]. Also, the 
by heavy accommodated strain occurred due to the transformation of austenite into AF,
high  dislocation 
contributes density, incaused 
considerably by  heavy 
achieving accommodated 
high strength strain 
in the final occurred 
product [37]. due 
Onto 
thethe 
other
transformation of austenite into AF, contributes considerably in achieving high strength 
hand, martensite is a hard, brittle phase of steel with a distorted tetragonal crystal lattice. It
isin the final product [37]. On the other hand, martensite is a hard, brittle phase of steel with 
formed by rapid cooling of austenite, a process known as quenching. The formation of
a distorted tetragonal crystal lattice. It is formed by rapid cooling of austenite, a process 
AF in alloy B is observed to take place at about 565 ◦ C and that of M in alloy C at about
known as quenching. The formation of AF in alloy B is observed to take place at about 565 
396 ◦ C. Indeed, the predicted Ms of alloy C according to Andrews’ equation [41] is 393 ◦ C,
°C and that of M in alloy C at about 396 °C. Indeed, the predicted Ms of alloy C according 
which is in consistence with the measured temperature. The diversity of microstructures
to  Andrews’  equation  [41]  is  393  °C,  which  is  in  consistence  with  the  measured 
developed in the three alloys is due to the different chemistries of the alloys and Prior
temperature. The diversity of microstructures developed in the three alloys is due to the 
Austenite Grain Size (PAGS). As revealed in Figure 4 the prior austenite grains (PAG) of
different chemistries of the alloys and Prior Austenite Grain Size (PAGS). As revealed in 
alloys B and C (recorded ~250 and ~450 µm, respectively) are significantly smaller com-
Figure 4 the prior austenite grains (PAG) of alloys B and C (recorded ~250 and ~450 μm, 
pared
respectively) of
to that alloy
are  A (recorded
significantly  ~840compared to 
smaller  µm), thanksthat 
to the pinning effect of TiN
of alloy A (recorded  inμm), 
~840  B and C.
Indeed, increasing the grain size should decelerate the transformation kinetics. Alloy A
thanks to the pinning effect of TiN in B and C. Indeed, increasing the grain size should 
showed,
decelerate however, faster kinetics
the  transformation  of transformation
kinetics.  compared
Alloy  A  showed,  to B and
however,  C despite
faster  its of 
kinetics  larger
austenite grains. This points out that the diluted chemistry of A dominated the kinetics
transformation compared to B and C despite its larger austenite grains. This points out 
ofthat the diluted chemistry of A dominated the kinetics of transformation. 
transformation.

  (a)      (b)    (c) 

AF 
M 
WF 

GF 

GB 

Metals 2022, 12, x FOR PEER REVIEW  8  of  17 

 

Figure 5. Microstructure of alloys A (a), B (b) and C (c) after applying schedule I.
Figure 5. Microstructure of alloys A (a), B (b) and C (c) after applying schedule I 

3.5
A
3
 
Rel. change in length (%)

2.5
B
2
C
1.5

0.5

0
200 300 400 500 600 700 800
Temperature (°C)

Figure 6. Austenite decomposition kinetics in terms of relative change in length during cooling of
Figure 6. Austenite decomposition kinetics in terms of relative change in length during cooling of 
alloysalloys A, B and C according to schedule I. The marked temperatures on the curves of alloys A and 
A, B and C according to schedule I. The marked temperatures on the curves of alloys A and B
designate the Ar1 and r1A and A
B designate the A r3 temperatures and on the curve of alloy C designate the martensite start
r3 temperatures and on the curve of alloy C designate the martensite start 

(MS) and martensite finish (M
(MS ) and martensite finish (Mf ). f). 

(a)  (b) 
 200 300 400 500 600 700 800
Temperature (°C)

Figure 6. Austenite decomposition kinetics in terms of relative change in length during cooling of
alloys A, B and C according to schedule I. The marked temperatures on the curves of alloys A and
Metals 2022, 12, 1913
B designate the Ar1 and Ar3 temperatures and on the curve of alloy C designate the martensite8 of 16
start
(MS) and martensite finish (Mf).
Metals 2022, 12, x FOR PEER REVIEW  9  of  17 
 
(a) (b)

The two cyclic austenitization just above Ae3 (Figure 8e–h) and just below Ae3 (Figure 
8i–l)  “normalizes”  the  grains  of  the  previous  structure.  During  holding  at  these 
temperatures new grains of refined austenite are formed. The normalizing treatment also 
reduces the structure irregularities and produces structure with improved toughness and 
ductility. This is clearly observable in the microstructures of A and B, which revealed a 
homogeneous ferritic pearlitic fine microstructure with grain size of ferrite ranging from 
5 to 10 μm. Figure 8k showed that, in addition to the ferritic‐pearlitic microstructure, a 
second phase of GB is formed in alloy B. In alloy C, a microstructure composed of GB and 
M is formed during  cooling after  holding 35  °C  above Ae3,  as shown  in Figure  8h.  The 
formation of GB indicates the decrease in hardness of alloy C. The decreased hardness in 
this case is basically attributed to the refining of PAG during normalizing at 820 °C. Figure 
HV
9  0.025 = 216.0
compares  the  PAGS  of  alloy  C  after  holding  HV at 0.025 = 204.2
1350  °C  and  that  developed  after 
normalizing at 820 °C. The normalizing process at 820 °C has significantly refined the PAG 
Figure 7. HV 0.025 values together with the micro-hardness indentations (a) in GB and (b) in WF of
from 450 μm to 235 μm, and therefore provided more nucleation sites for transformation 
Figure 7. HV 0.025 values together with the micro-hardness indentations (a) in GB and (b) in WF of
the microstructure
during  developed
cooling,  which  after applying
resulted  schedule I the 
in  accelerating  on alloy A.
transformation  kinetics.  The 
the microstructure developed after applying schedule I on alloy A.
dilatometric curve of alloy C in Figure 8e shows that the GB transformation begins at ~560 
3.1.4. Microstructure Development during Applying Schedule II
°C and the M starts to form at ~360 °C. The PF formation of alloys A and B begin at ~760 
3.1.4. Microstructure Development during Applying Schedule II
Applying two thermal-cycles at varied temperature T
°C  and ~750 °C, respectively. Nevertheless, fast heating and  B (schedule
relatively  II,holding of 
short  Figure 1) has a
Applying
significant two thermal-cycles
influence on the formed at varied temperature
microstructures as shown TB (schedule
in Figure II,The
8. Figure 1) hasana
TBinto 
plays
alloy  C  at  725  °C  was  not  sufficient  for  complete  transformation  of  martensite 
significant
importantThis influence
role in on the formed microstructures as shown in Figure 8. The T B plays an
austenite.  is  defining the resulting
clearly  revealed  from microstructure. The dilatometric
the  small  dilatation  curves observed
observed  during  the 
important role in defining the resulting microstructure. The dilatometric
during the last cooling after two cycles of holding at TB is shown in Figure 8 as well. curves observed
subsequent cooling of C to room temperature (Figure 8i). Cooling from 1350 °C to room 
during the last
Figure 8b cooling
shows after
for twoA
alloy cycles of holding at T is shown sizeinPFFigure 8 as well.
temperature  results  in  formation  of amartensite 
microstructure(Figure ofB5c). 
largeSubsequent  grains
heating  with
and P and
AF. Figure
Figure 8b
8c shows
shows for
for Balloy
a A a microstructure
microstructure composed of largeof size
AF
holding at 725 °C in the two‐phase region results mainly in tempering the martensite and  PF
and grains
GB. with
Contrary P and
to AF.
the
in formation of marginal quantity of newly formed austenite. The main microstructure in  the
Figure 8c shows
characteristics of for
AF, B
as a
a microstructure
desirable composed
microconstituent of
withAF and
high GB. Contrary
toughness, Zhou toet al.
characteristics
reported that GBof AF,
is as a
detrimental desirable
for microconstituent
toughness and bestwith
be high
avoided
this  case  is  composed  of  tempered  martensite  (tM)  with  small  quantity  of  freshly  toughness,
during Zhou
productionet al.
reported[42].
process
transformed  that
new GB
Refining is detrimental
martensite  for toughness
GB is(un‐tempered 
predicted to limit
formed thisand best
effect,
from  the anbeenhancement
avoided
newly  formed duringin theproduction
austenite  cleavage
at 
process [42]. Refining GB is predicted to limit this effect, an enhancement
fracture is reported via decreasing grain size of GB [43]. Alloy C exhibits an un-tempered
725 °C) as shown in Figure 8l. The tempered martensite has an acicular form. This acicular  in the cleavage
fracture is reported
martensite via decreasing
microstructure (Figure grain size of
8d). Two GB [43].
cyclic AlloyatC(1350
heating
shape fades due to tempering, yielding a globular appearance [44]. For alloys A and B,  ◦ C an
exhibits +A un-tempered
e3 )/2 has a
martensiteeffectmicrostructure
holding 25 °C below A
significant (Figure 8d). Two cyclic heating
of allat (1350This
e1 has no significant effect in altering the microstructure observed 
in accelerating the transformation kinetics alloys. °C is+ also
Ae3)/2 has a
revealed
in Figure 5, as shown in Figure 8n,o. Regarding alloy C, holding at 535 °C has resulted in 
significant
in effect in
the dilatometric accelerating
curves of Figurethe 8a astransformation kinetics ofthealltransformation
indicated by increasing alloys. This isstarts,also
producing tM. The tempered martensite in Figure 8p is with longer and less faded acicular 
revealed
this might in the dilatometric
be attributed curves of
to the improved Figure 8aofas
homogeneity indicated ofby
distribution increasing
alloying elements the
form compared to that of Figure 8l due to the lower tempering temperature. 
during the two cyclic
transformation starts,austenitization.
this might be attributed to the improved homogeneity of
distribution of alloying elements during the two cyclic austenitization.
TB  A  B  C 
(a) (1350 + Ae3)/2  (b) 1100 °C  (c) 1090 °C  (d) 1070 °C 
Rel. change in length (%)
0.5

1.5

2.5

3.5
1

3
200
300
400

PF 
M 
500

GB 
600

AF  AF 
700

Temperature (°C)
800

A
B
C
900

Figure 8. Cont.

 
Metals 2022, 12, 1913
Metals 2022, 12, x FOR PEER REVIEW  10  of  17  9 of 16
 

    (e) Ae3 + 35  (f) 890 °C  (g) 865 °C  (h) 820 °C 

Rel. change in length (%)
0.5

1.5

2.5
0

3
200
300

P 
PF 
PF 
400

M 
500

P 
600
700

C
Temperature (°C)

GB 
800

A
B
900

(i) “Ae3‐25”/“50%γ”  (j) 830 °C  (k) 805 °C  (l) 725°C 

Rel. change in length (%)
0.5

1.5

2.5
0

PF 
100
200

PF 
300

P 
GB 
400

tM 
500
600

C
Temperature (°C)
700

B
800

A
900

(m) Ae1 − 35  (n) 665 °C  (o) 640 °C  (p) 535 °C 

Rel. change in length (%)
0.5

1.5

2.5
0

3
200

GB  AF 
300

WF  tM 
GF 
400
500

GB 
Temperature (°C)

B
600

A
700

Dilatometric
Figure8. 8.Dilatometric 
Figure  curves
curves  (the
(the  left left column)
column)  recorded
recorded  during
during  the final
the  final  cooling
cooling  after applying
after  applying 
schedule II (two cycles) with different reheating temperatures (T ) on steels A, B and
schedule II (two cycles) with different reheating temperatures (TB) on steels A, B and C together 
B C together with
with the developed microstructure in the three alloys. 
the developed microstructure in the three alloys.

    The two cyclic austenitization just above Ae3 (Figure 8e–h) and just below Ae3 (Figure 8i–l)
“normalizes” the grains of the previous structure. During holding at these temperatures
new grains of refined austenite are formed. The normalizing treatment also reduces the
structure irregularities and produces structure with improved toughness and ductility. This
is clearly observable in the microstructures of A and B, which revealed a homogeneous
ferritic pearlitic fine microstructure with grain size of ferrite ranging from 5 to 10 µm.
Figure 8k showed that, in addition to the ferritic-pearlitic microstructure, a second phase
 
Metals 2022, 12, 1913 10 of 16

of GB is formed in alloy B. In alloy C, a microstructure composed of GB and M is formed

during cooling after holding 35 ◦ C above Ae3 , as shown in Figure 8h. The formation of
GB indicates the decrease in hardness of alloy C. The decreased hardness in this case is
basically attributed to the refining of PAG during normalizing at 820 ◦ C. Figure 9 compares
the PAGS of alloy C after holding at 1350 ◦ C and that developed after normalizing at 820 ◦ C.
The normalizing process at 820 ◦ C has significantly refined the PAG from 450 µm to 235 µm,
and therefore provided more nucleation sites for transformation during cooling, which
resulted in accelerating the transformation kinetics. The dilatometric curve of alloy C in
Figure 8e shows that the GB transformation begins at ~560 ◦ C and the M starts to form at
~360 ◦ C. The PF formation of alloys A and B begin at ~760 ◦ C and ~750 ◦ C, respectively.
Nevertheless, fast heating and relatively short holding of alloy C at 725 ◦ C was not sufficient
for complete transformation of martensite into austenite. This is clearly revealed from
the small dilatation observed during the subsequent cooling of C to room temperature
(Figure 8i). Cooling from 1350 ◦ C to room temperature results in formation of martensite
(Figure 5c). Subsequent heating and holding at 725 ◦ C in the two-phase region results
mainly in tempering the martensite and in formation of marginal quantity of newly formed
austenite. The main microstructure in this case is composed of tempered martensite (tM)
with small quantity of freshly transformed new martensite (un-tempered formed from
the newly formed austenite at 725 ◦ C) as shown in Figure 8l. The tempered martensite
has an acicular form. This acicular shape fades due to tempering, yielding a globular
appearance [44]. For alloys A and B, holding 25 ◦ C below Ae1 has no significant effect in
altering the microstructure observed in Figure 5, as shown in Figure 8n,o. Regarding alloy
C, holding at 535 ◦ C has resulted in producing tM. The tempered martensite in Figure 8p is
Metals 2022, 12, x FOR PEER REVIEW  11  of  17 
  with longer and less faded acicular form compared to that of Figure 8l due to the lower
tempering temperature.

(a)  (b) 

FigureFigure  9.  Micrographs 

9. Micrographs showing showing  different 
different martensite 
martensite grains  from
grains formed formed  from 
prior prior  austenite 
austenite with 
with different
different PAGS after quenching from (a) 1350 °C and (b) 820 °C for alloy C. 
◦ ◦
PAGS after quenching from (a) 1350 C and (b) 820 C for alloy C.

3.1.5. Microstructure Developed after Applying Schedule III 
3.1.5. Microstructure Developed after Applying Schedule III
FigureFigure  10  shows 
10 shows representative 
representative microstructure 
microstructure forfor  alloy 
alloy AA  comparing four
comparing four cycles
cycles 
(schedule III, Figure 1) to “two cycles (schedule II)”, reheated to the same TB temperature. 
(schedule III, Figure 1) to “two cycles (schedule II)”, reheated to the same TB temperature.
It is shown that the number of cycles plays a notable role in the fineness of the developed 
It is shown that the number of cycles plays a notable role in the fineness of the developed
microstructure, but it does not alter its constituents. 
microstructure, but it does not alter its constituents.
TB  Schedule II  Schedule III 
3.2. Hardness Characteristics
The effect of different thermal cycles on the hardness is shown in Figure 11. The lowest
hardness is observed in the normalized microstructures of alloy A and B. These microstruc-
tures are expected to have improved ductility and toughness due to its regular and fine
grains, of
Ae3course,
 + 35  yield strength is expected to be reduced. The finest of these structures is
that obtained by isothermal holding at Ae3 + 35 ◦ C as disclosed by the measured ferrite
grain sizes (gray curves of Figure 11a,b). The ferrite grain is indeed a key microstructural
component suppressing the propagation of cracks across the grain/grain boundary. Wu
et al. showed that the intracrystalline polygonal ferrite improves the fracture toughness of
Metals 2022, 12, 1913 11 of 16

high strength steels [45]. Additionally, ductile fracture toughness has been found to increase
with decreasing ferrite grain size [46]. Another factor strongly affecting the material tough-
ness is the material impurities. As the impurity level is increased, the inclusion volume
fraction, average inclusion radius, inclusion spacing and number of inclusion particles per
unit area all increase. These increases are detrimental to the toughness and ductility of the
steel [47]. Alloy C (Figure 11c) exhibits higher hardness compared to alloys A and B due to
its structure that is basically composed of martensite/tempered martensite and GB. The
hardness is also lower when isothermal holding at Ae3 + 35 ◦ C is applied. The lower value
in this case is, however, due to refining the prior austenite grains and therefore replacing the
un-tempered M by GB. As discussed above, treating alloy C at 725 ◦ C results in two types
of martensite at room temperature, mostly tM with smaller fraction of quenched martensite,
M. The absence of GB in this case is the reason for the higher hardness observed at this
temperature compared to that recorded at 820 ◦ C. Additionally, a secondary hardening
occurs due to precipitation of Mo and Cr carbides during tempering. Tempering at 535 ◦ C
results in the lowest observed hardness in this alloy. The tempered martensite formed
at this temperature should have higher hardness compared to that formed at 725 ◦ C, as
increasing the tempering temperature reduces the hardness of martensite. Nonetheless,
there is no formed quenched martensite in this microstructure. The microstructure is a
fully tempered one, therefore, it has lower hardness than the microstructures formed by
quenching from 725 ◦ C. The absence of this secondary hardening might also contribute
to the inferior hardness after quenching from 535 ◦ C. It is worth mentioning here that
the hardness of steel B and C treated at (1350 ◦ C + Ae3 )/2 showed similar value to that
obtained by treating at 1350 ◦ C. This can be related to the effective pinning of austenite by
microalloying in these two steels (Figure 3b). The parent austenite of the microstructures
formed by quenching from 1350 ◦ C is similar in size to that of the microstructure formed
by quenching from TB = (1350 ◦ C + Ae3 )/2.

3.3. Transformation Temperatures during the Thermal Cycles

It can be inferred from Figures 6 and 8 that the transformation kinetics during cooling
has a significant correlation with the holding temperatures TA and TB . Table 3 sums up
the measured transformation-start and -end temperatures during cooling from TA and
TB together with that predicted equilibrium temperatures (i.e., Ae1 and Ae3 ) and that
recorded during heating at 100 Ks−1 (i.e., Ac1 and Ac3 ). For alloys A and B, decreasing the
holding temperature preceding the last cooling results in a general trend of shifting the
transformation-start and -end temperatures to higher values. Generally, the microstructure
formed at higher transformation temperatures are softer than that formed at lower tem-
peratures. Therefore, the obtained microstructures are correspondingly becoming softer
by decreasing the holding temperature until reaching the temperature range around Ae3 ,
as shown in Figure 11a,b. Nevertheless, holding at TB below Ae1 does not result in phase
changes and accordingly the hardness is increased compared to higher TB values. Re-
garding alloy C, the transformation is basically dominated by the martensite formation as
previously shown under Section 3.1.4, with exception of the transformation-start tempera-
ture for TB = Ae3 + 35 ◦ C. The transformation-start and -finish temperatures listed in Table 3
for this alloy present Ms and Mf temperatures, respectively. The refined PAGS formed by
holding at Ae3 + 35 ◦ C accelerated the transformation kinetics, therefore, this case records
the highest transformation start of 560 ◦ C, which corresponds to the GB formation. Yet,
the carbon diffused from GB to the untransformed austenite, enriching it and enhancing
its stability. Consequently, the Mf showed lower value compared to that recorded after
holding at TA and TB = (1350 ◦ C + Ae3 )/2.
 Figure  9.  Micrographs  showing  different  martensite  grains  formed  from  prior  austenite  with 
different PAGS after quenching from (a) 1350 °C and (b) 820 °C for alloy C. 

3.1.5. Microstructure Developed after Applying Schedule III 
Figure  10  shows  representative  microstructure  for  alloy  A  comparing  four  cycles 
Metals 2022, 12, 1913 (schedule III, Figure 1) to “two cycles (schedule II)”, reheated to the same TB temperature. 12 of 16
It is shown that the number of cycles plays a notable role in the fineness of the developed 
microstructure, but it does not alter its constituents. 
TB  Schedule II  Schedule III 

Ae3 + 35 

Ae3 − 25 

WF 

Ae1 − 35 
GF 
WF 

GB  GF 

GB 
Metals 2022, 12, x FOR PEER REVIEW  13  of  17 
Figure 10. Effect of applying 2 Cycles (left) and 4 Cycles (right) on the formed microstructure of
Figure 10. Effect of applying 2 Cycles (left) and 4 Cycles (right) on the formed microstructure of 
alloy A. 
alloy A.

(a)
  260 12
Alloy A
GB+GF+WF

240
GB+GF+WF

10
4 Cycles
Ferrite grain size (µm)

220
8
WF+GB+PF

200
AF+PF
HV30

180 6
PF+P
PF+P

160
4
140
HV30
2
120 Ferrite grain size
A1

A3

100 0
600 700 800 900 1000 1100 1200 1300 1400
Temperature (°C)

Figure 11. Cont.

(b)

260 12
AF

Alloy B
240
AF+GB

10
AF

 grain size (µm) 
PF+P

220
PF+P

8
200
HV30

180 6
 160
4

Ferrit
P
140
HV30
2
120 Ferrite grain size

A1

A3
100 0
Metals 2022, 12, 1913 600 700 800 900 1000 1100 1200 1300 1400 13 of 16
Temperature (°C)

(b)

260 12

AF
Alloy B
240

AF+GB
10

AF

Ferrite grain size (µm) 
PF+P
220

PF+P
8
200

HV30
180 6
160
4
140
A1 2

A3
120 HV30
Ferrite grain size
100 0
600 700 800 900 1000 1100 1200 1300 1400
Temperature (°C)

(c)

380
Alloy C

M
370

360

350
HV30

340

330

320
A1

A3

310

300
530 630 730 830 930 1030 1130 1230 1330
Temperature (°C)

Figure 11. Dependence of hardness and ferrite grain size on the treatment temperature of alloys
A, B and C. The developed microstructures are also designated on the diagrams at the respective
treatment temperature.

Table 3. Transformation temperatures under equilibrium together with those recorded during heating
and cooling applying different thermal cycles. St: start and Fi: finish. Under the present conditions
no significant difference was found between 2 and 4 cycles.

Temp. A B C
Ae1 700 675 570
Ae
Ae3 855 830 785
Ac1 816 804 763
Ac
Ac3 954 924 890
TA or TB St Fi St Fi St Fi
schedule I:1350 ◦ C 630 528 565 425 396 298
Ar1 /Ar3 schedule II: (1350 + Ae3 )/2 680 553 625 496 427 308
MS /Mf schedule II: Ae3 + 35 757 651 750 619 560 281
schedule II: Ae3 − 25/50%γ 774 658 733 657 331 180
schedule II: Ae1 − 35 - - - - - -
Metals 2022, 12, 1913 14 of 16

4. Summary and Conclusions

Three different microalloyed steels, two HSLA steels and the third is a welding wire
one, were subjected to different exemplary thermal cycles in dilatometer. The aim is to
enlighten the dependence of the microstructure development on the alloying composition
and thermal-cycling parameters during WAAM. Under the present conditions, the transfor-
mation temperatures given in Table 3 were found. The following conclusions can be drawn
from the current study:
Ti-Nb microalloying strategy can present a method of refining the microstructure of
WAAM components. The undissolved precipitates can hinder the growth of the prior
austenite during the repeated heating cycles at high temperature and therefore, results
finally in a finer WAAM microstructure.
In case of the HSLA-steels, the softest microstructures are those developed by cycling
heating around Ae3 , where fine ferritic-pearlitic microstructure develops from normalized
(fine grained) austenite.
For the wire welding steel, which has higher hardenability compared to the studied
HSLA steels, martensite is formed during cooling from the reheating temperature. In the
simulation of the WAAM process, cyclic thermal processing around Ae3 can also result
in soft microstructure due to the formation of GB from the refine/normalized austenite.
Nonetheless, the softest structure formed in this steel is that one developed by thermal
cycling below Ae1 . This microstructure is a soft tempered martensite one and free from any
quench-martensite.
It is suggested here that the HSLA steels can be more suitable for producing softer
and consequently tougher components than that produced from the conventional welding
wires, where the high additions of alloying elements results in developing of martensite
with lower toughness.
Steel composition, temperature history in terms of reheating temperatures and dwell
time in certain temperature intervals chosen according to equilibrium transformation
temperatures is a reasonable characterization and data structure candidate for prediction of
microstructure by AI methods. Of course, much more work has to be done in this respect.

Author Contributions: Conceptualization, M.S., K.-H.S. and V.W.; methodology, M.S. and K.-H.S.;
software, M.S.; validation, C.H. and M.S.; formal analysis, C.H and M.S.; investigation, C.H. and
M.S.; resources, K.-H.S. and V.W.; data curation, C.H.; writing—original draft preparation, M.S.,
K.T. and K.-H.S.; writing—review and editing, M.S., K.T. and K.-H.S.; visualization, C.H. and M.S.;
supervision, M.S., V.W. and K.-H.S.; project administration, M.S. and K.-H.S. All authors have read
and agreed to the published version of the manuscript.
Funding: This research was funded by Open Access Publishing Fund of Clausthal University
of Technology.
Data Availability Statement: The raw/processed data required to reproduce these findings cannot
be shared at this time as the data also forms part of an ongoing study.
Acknowledgments: We acknowledge financial support by Open Access Publishing Fund of Clausthal
University of Technology.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Treutler, K.; Wesling, V. The Current State of Research of Wire Arc Additive Manufacturing (WAAM): A Review. Appl. Sci. 2021,
11, 8619. [CrossRef]
2. Mbodj, N.G.; Abuabiah, M.; Plapper, P.; El Kandaoui, M.; Yaacoubi, S. Bead Geometry Prediction in Laser-Wire Additive
Manufacturing Process Using Machine Learning: Case of Study. Appl. Sci. 2021, 11, 11949. [CrossRef]
3. Richter, A.; Scheck, M.; Gehling, T.; Bohn, C.; Wesling, V.; Rembe, C. Erfassung geometrischer Daten des Schmelzbades zur
Regelung eines WAAM-Prozesses. tm Tech. Mess. 2021, 88, s95–s100. [CrossRef]
4. Richter, A.; Gehling, T.; Treutler, K.; Wesling, V.; Rembe, C. Real-time measurement of temperature and volume of the weld pool
in wire-arc additive manufacturing. Meas. Sens. 2021, 17, 100060. [CrossRef]
Metals 2022, 12, 1913 15 of 16

5. Richter, A.; Rembe, C.; Gehling, T.; Treutler, K.; Wesling, V. Echtzeittemperaturmessung bei additivem Lichtbogenschweißen/Real-
time temperature measurement at wire arc additive welding. tm Tech. Mess. 2019, 86, 112–116. [CrossRef]
6. Richter, A.; Scheck, M.; Bohn, C.; Rembe, C. Erfassung der Schmelzbadfläche mit Korrektur der Perspektive zur Prozessregelung
eines Wire and Arc Additive Manufacturing. tm Tech. Mess. 2022, 89, 525–533. [CrossRef]
7. Scheck, M.; Franz, J.; Richter, A.; Gehling, T.; Treutler, K.; Beitler, S.; Wesling, V.; Rembe, C.; Bohn, C. Identification and Modeling
of Wire Arc Additive Manufacturing under consideration of Interpass Temperature. In Proceedings of the 2022 UKACC 13th
International Conference on Control (CONTROL), Plymouth, UK, 20–22 April 2022; IEEE: Toulouse, France, 2022; pp. 219–225.
[CrossRef]
8. Ehlers, R.; Treutler, K.; Wesling, V. SAT Solving with Fragmented Hamiltonian Path Constraints for Wire Arc Additive Manu-
facturing. In Theory and Applications of Satisfiability Testing—SAT 2020; Pulina, L., Ed.; Springer International Publishing: Cham,
Switzerland, 2020; pp. 492–500.
9. Wächter, M.; Leicher, M.; Hupka, M.; Leistner, C.; Masendorf, L.; Treutler, K.; Kamper, S.; Esderts, A.; Wesling, V.; Hartmann, S.
Monotonic and Fatigue Properties of Steel Material Manufactured by Wire Arc Additive Manufacturing. Appl. Sci. 2020, 10, 5238.
[CrossRef]
10. Kühne, R.; Feldmann, M.; Citarelli, S.; Reisgen, U.; Sharma, R.; Oster, L. 3D printing in steel construction with the automated
Wire Arc Additive Manufacturing. Ce/Papers 2019, 3, 577–583. [CrossRef]
11. Plangger, J.; Schabhüttl, P.; Vuherer, T.; Enzinger, N. CMT Additive Manufacturing of a High Strength Steel Alloy for Application
in Crane Construction. Metals 2019, 9, 650. [CrossRef]
12. Schroepfer, D.; Treutler, K.; Boerner, A.; Gustus, R.; Kannengiesser, T.; Wesling, V.; Maus-Friedrichs, W. Surface finishing of
hard-to-machine cladding alloys for highly stressed components. Int. J. Adv. Manuf. Technol. 2021, 114, 1427–1442. [CrossRef]
13. Treutler, K.; Lorenz, S.; Hamje, J.; Wesling, V. Wire and Arc Additive Manufacturing of a CoCrFeMoNiV Complex Concentrated
Alloy Using Metal-Cored Wire—Process, Properties, and Wear Resistance. Appl. Sci. 2022, 12, 6308. [CrossRef]
14. Pan, Z.; Ding, D.; Wu, B.; Cuiuri, D.; Li, H.; Norrish, J. Arc Welding Processes for Additive Manufacturing: A Review. In
Transactions on Intelligent Welding Manufacturing; Springer: Singapore, 2018; pp. 3–24.
15. Wu, B.; Pan, Z.; Ding, D.; Cuiuri, D.; Li, H.; Xu, J.; Norrish, J. A review of the wire arc additive manufacturing of metals: Properties,
defects and quality improvement. J. Manuf. Process. 2018, 35, 127–139. [CrossRef]
16. Eissel, A.; Engelking, L.; Treutler, K.; Wesling, V.; Schröpfer, D.; Kannengießer, T. Modification of Co–Cr alloys to optimize
additively welded microstructures and subsequent surface finishing. Weld. World 2022, 66, 2245–2257. [CrossRef]
17. Leicher, M.; Treutler, K.; Wesling, V. Development of an Alternative Alloying Concept for Additive Manufacturing Using PVD
Coating. Appl. Sci. 2022, 12, 6619. [CrossRef]
18. Eissel, A.; Engelking, L.; Treutler, K.; Schroepfer, D.; Wesling, V.; Kannengiesser, T. Nickel-Iron-Alloy Modification to Enhance
Additively Welded Microstructure for Subsequent Milling. In Proceedings of the 2nd International Conference on Advanced Joining
Processes (AJP 2021), Sintra, Portugal, 21–22 October 2021, 1st ed.; Da Silva, L.F.M., Martins, P.A.F., Reisgen, U., Eds.; Springer
International Publishing: Cham, Switzerland, 2022; pp. 85–99.
19. Reisgen, U.; Sharma, R.; Mann, S.; Oster, L. Increasing the manufacturing efficiency of WAAM by advanced cooling strategies.
Weld. World 2020, 64, 1409–1416. [CrossRef]
20. Luo, J.; You, G.; Lu, D.; Zeng, S.; Peng, L.; Liu, Q. Effect of Modified Water-Bath Method on Microstructure and Mechanical
Properties of Wire Arc Additive Manufactured Low-Carbon Low-Alloy Steel. Steel Res. Int. 2021, 92, 523. [CrossRef]
21. Panchenko, O.; Kladov, I.; Kurushkin, D.; Zhabrev, L.; Ryl’Kov, E.; Zamozdra, M. Effect of thermal history on microstructure
evolution and mechanical properties in wire arc additive manufacturing of HSLA steel functionally graded components. Mater.
Sci. Eng. A 2022, 851, 143569. [CrossRef]
22. Treutler, K.; Kamper, S.; Leicher, M.; Bick, T.; Wesling, V. Multi-Material Design in Welding Arc Additive Manufacturing. Metals
2019, 9, 809. [CrossRef]
23. Pütz, R.D.; Pratesa, Y.; Oster, L.; Sharma, R.; Reisgen, U.; Zander, D. Microstructure and Corrosion Behavior of Functionally
Graded Wire Arc Additive Manufactured Steel Combinations. Steel Res. Int. 2021, 92, 2100387. [CrossRef]
24. Leicher, M.; Kamper, S.; Treutler, K.; Wesling, V. Multi-material design in additive manufacturing—Feasibility validation. Weld.
World 2020, 64, 1341–1347. [CrossRef]
25. Samardžić, I.; Stoić, A.; Kozak, D.; Kladaric, I.; Dund̄er, M. Application of Weld Thermal Cycle Simulator in Manufacturing
Engineering. J. Manuf. Ind. Eng. 2013, 12, 2–177. [CrossRef]
26. Celin, R.; Burja, J.; Kosec, G. A comparison of as-welded and simulated heat affected zone (HAZ) microstructures. Mater. Tehnol.
2016, 50, 455–460. [CrossRef]
27. Dai, Y.; Yu, S.; Huang, A.; Shi, Y. Microstructure and mechanical properties of high-strength low alloy steel by wire and arc
additive manufacturing. Int. J. Miner. Metall. Mater. 2020, 27, 933–942. [CrossRef]
28. Rodrigues, T.A.; Duarte, V.; Avila, A.J.; Telmo, G.S.; Miranda, R.M.; Oliveira, J.P. Wire and arc additive manu-facturing of HSLA
steel: Effect of thermal cycles on microstructure and mechanical properties. Addit. Manuf. 2019, 27, 440–450. [CrossRef]
29. Sun, L.; Jiang, F.; Huang, R.; Yuan, D.; Guo, C.; Wang, J. Microstructure and mechanical properties of low-carbon high-strength
steel fabricated by wire and arc additive manufacturing. Metals 2020, 10, 216. [CrossRef]
30. Müller, J.; Hensel, J.; Dilger, K. Mechanical properties of wire and arc additively manufactured high-strength steel structures.
Weld. World 2021, 66, 395–407. [CrossRef]
Metals 2022, 12, 1913 16 of 16

31. Wesling, V.; Leicher, M.; Gräbner, M.; Treutler, K.; Lorenz, S.; Esderts, A.; Hupka, M.; Wächter, M.; Spitzer, K.H. Werkstoffken-
nwerte additiv gefertigter Strukturen. In Tagungsband 4. Niedersächsisches Symposium Materialtechnik: 25. Bis 26. Februar 2021; TU
Clausthal: Clausthal-Zellerfeld, Germany, 2021.
32. Trzaska, J.; Dobrzański, L.A. Modelling of CCT diagrams for engineering and constructional steels. J. Mater. Process. Technol. 2007,
192, 504–510. [CrossRef]
33. Dobrzanski, L.A.; Trzaska, J. Application of neural networks for the prediction of continuous cooling transformation diagrams.
Comput. Mater. Sci. 2004, 30, 251–259. [CrossRef]
34. Wang, J.; van der Wolk, P.J.; van der Zwaag, S. Effects of carbon concentration and cooling rate on continuous cooling transforma-
tions predicted by artificial neural network. ISIJ Int. 1999, 39, 38–46. [CrossRef]
35. Kouraytem, N.; Li, X.; Tan, W.; Kappes, B.; Spear, A.D. Modeling process–structure–property relationships in metal additive
manufacturing: A review on physics-driven versus data-driven approaches. J. Phys. Mater. 2021, 4, 032002. [CrossRef]
36. Soliman, M.; Weidenfeller, B.; Palkowski, H. Metallurgical Phenomena during Processing of TRIP Steel. Steel Res. Int. 2009,
80, 57–65.
37. Soliman, M. Microstructural Control and Properties Optimization of Microalloyed Pipeline Steel. Metals 2020, 10, 1499. [CrossRef]
38. Bramfitt, B.L.; Benscoter, A.O. Metallographer’s Guide: Practices and Procedures for Irons and Steels; ASM International: Materials
Park, OH, USA, 2002.
39. Gomez, M.; Valles, P.; Medina, S.F. Evolution of microstructure and precipitation state during thermomechanical processing of a
X80 microalloyed steel. Mater. Sci. Eng. A 2011, 528, 4761–4773. [CrossRef]
40. Bhadeshia, H.; Honeycombe, R. Steels: Microstructure and Properties, 4th ed.; Butterworth-Heinemann: Oxford, UK, 2017.
41. Andrews, K.W. Empirical formulae for the calculation of some transformation temperatures. J. Iron Steel Inst. 1965, 203, 721.
42. Zhou, T.; Yu, H.; Wang, S. Effect of microstructural types on toughness and microstructural optimization of ultra-heavy steel
plate: EBSD analysis and microscopic fracture mechanism. Mater. Sci. Eng. A 2016, 658, 150–158. [CrossRef]
43. Tian, D.; Karjalainen, L.; Qian, B.; Chen, X. Cleavage Fracture Model for Granular Bainite in Simulated Coarse-Grained Heat-
Affected Zones of High-Strength Low-Alloyed Steels. JSME Int. J. Ser. A Mech. Mater. Eng. 1997, 40, 179–188. [CrossRef]
44. Lescano, D.E.; Silvetti, S.P. Study of Microstructure and Tempered Martensite Embrittlement in AISI 15b41 Steel. Procedia Mater.
Sci. 2012, 1, 134–140. [CrossRef]
45. Wu, Z.; Liu, Z.; Li, X. Microstructural effects on the fracture toughness of advanced high strength steels. Mater. Lett. 2020,
271, 127761. [CrossRef]
46. Srinivas, M.; Malakondaiah, B.; RamaRao, P. Influence of polycrystal grain size on fracture toughness of and fatigue threshold in
Armco iron Author. Eng. Fract. Mech. 1987, 28, 561–576. [CrossRef]
47. Chen, D.; Zhang, X.; Wu, H.; Lang, D.; Xiao, D.; Wang, Z.; Su, B.; Meng, D. The influence of impurities on the ductility and
toughness of a low-temperature-aged U-Nb alloy. Mater. Sci. Eng. A 2018, 739, 1–16. [CrossRef]