Content uploaded by Florian Riedlsperger
Author content
All content in this area was uploaded by Florian Riedlsperger on Feb 17, 2022
Content may be subject to copyright.
Tailoring the alloy composition for wire arc additive manufacturing
utilizing metal-cored wires in the cold metal transfer process
Florian Pixner
a,
⇑
, Ricardo Buzolin
a
, Anto Zelic
´
a
, Florian Riedlsperger
a
, Marta Orlowska
b
,
Fernando Warchomicka
a
, Mathieu Decherf
c
, Michael Lasnik
d
, Norbert Enzinger
a
a
Institute of Materials Science, Joining and Forming, Graz University of Technology, Kopernikusgasse 24/I, A-8010 Graz, Austria
b
Faculty of Mechanical Engineering, Military University of Technology, Kaliskiego 2, 00-908 Warsaw, Poland
c
Voestalpine BÖHLER welding – Belgium S.A., Rue de L’Yser 2, B-7180 Seneffe, Belgium
d
Voestalpine BÖHLER Aerospace GmbH & Co KG, Mariazeller Straße 25, A-8605 Kapfenberg, Austria
highlights
Metal cored wires can be utilized in
the modified short-circuit cold metal
transfer process to customize alloy
composition and microstructure.
By varying the chemical composition
and increasing the nickel content of a
hot-work tool steel alloy, the amount
of detrimental delta ferrite can be
reduced and its morphology changed.
By changing the chemical
composition and the resulting
microstructure of the AM parts, the
mechanical properties can be
significantly improved.
graphical abstract
article info
Article history:
Received 19 November 2021
Revised 19 January 2022
Accepted 5 February 2022
Available online 07 February 2022
Keywords:
Wire-Based Additive Manufacturing
Cold Metal Transfer
Metal Cored Wires
Hot-Work Tool Steels
Alloy Design
abstract
In the present study, the application and tailoring of the alloy composition of chromium martensitic hot-
work steels using metal cored wires (MCW) for wire arc additive manufacturing (WAAM) in a modified
short-circuit metal transfer process is demonstrated. The nickel content was varied and the alloys were
fabricated as tubular-cored wires with various powder fillings. By recording the material transfer at high
speed during processing, evidence was gathered indicating the suitability of the fabricated cored wires
for WAAM. Optimized process parameters were identified by taking a Design of Experiment (DoE)
approach and additive manufacturing (AM) structures were fabricated from the chromium martensitic
hot-work tool steel alloys. The microstructure and mechanical properties of the parts were subsequently
characterized. The phase fraction of the polygonally shaped delta ferrite could be reduced and
microstructural refinement could be achieved by adding nickel to the investigated hot-work tool steel.
In addition to molybdenum-enriched precipitates that covered the grain boundaries, randomly scattered
non-metallic inclusions and oxides were observed. Modifying the microstructure by adding nickel also
affects the mechanical properties of the product: an increase in hardness, impact toughness and yield
strength as the nickel content increased in the AM structures fabricated by WAAM was observed.
Ó2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
https://doi.org/10.1016/j.matdes.2022.110453
0264-1275/Ó2022 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
⇑
Corresponding author.
E-mail address: florian.pixner@tugraz.at (F. Pixner).
Materials & Design 215 (2022) 110453
Contents lists available at ScienceDirect
Materials & Design
journal homepage: www.elsevier.com/locate/matdes
1. Introduction
Additive manufacturing (AM) is an emerging, future-oriented
technology that offers novel production possibilities that can com-
plement well-established and more conventional processes, such
as joining, casting and subtractive manufacturing. The term addi-
tive manufacturing embraces various techniques that can be gen-
erally categorized by applying the ISO 17296–2:2015 standard:
Vat photopolymerization, material jetting, binder jetting, powder
bed fusion (PBF), material extrusion, directed energy deposition
(DED) and sheet lamination. The PBF process is widely used in
metallic AM component processing. However, more economic
DED processes, such as laser metal deposition (LMD) and WAAM,
are also suitable and offer other application possibilities, mainly
due to the increased deposition rates.
AM has a high innovation potential, especially in the field of
tooling, and may be used to restore the geometries of worn/
cracked volumetric structures or manufacturing tool inserts, lend-
ing optimized geometrical, technological and metallurgical proper-
ties. Martensitic chromium hot-work steels with a medium carbon
content are the most commonly used steels in the tooling industry,
but processing them remains a challenge. Hot-work tool steels are
subjected to complex thermomechanical loading conditions in ser-
vice (e.g. dies in forging applications) [1]. The resulting require-
ments include a high wear resistance at the service temperature,
form stability, long service life and the possibility of an economical
repair process, including weldability [2]. Therefore, various hot-
work tool steels have been designed to display high-temperature
strength, wear resistance, toughness and temper resistance [3,4].
Additionally, these steel grades should resist abrasion, mechanical
cracking, thermal cracking and plastic deformation [2]. Since the
listed thermomechanical properties can contradict one another, a
wide range of hot-work tool steels is available. Nevertheless, the
most widespread and frequently commercially used grades are
the X38CrMoV5-1 (1.2343, AISI H11) and X40CrMoV5-1 (1.2344,
AISI H13) grades; these are mainly used for making conventional
forging tools and tools for presses.
Dies can experience different failure modes, either when the
different materials are formed or when being subjected to exten-
sive service conditions. Generally, failures during operation can
be attributed to the material being exposed to thermal shocks,
mechanical stress, cyclic loading and corrosion, which result in
heat checking, wear, plastic deformation and fatigue [5]. The
causes of failure can be described as [6]: 1) operational (e.g. due
to loading conditions, material handling, reconditioning), 2) manu-
facturing (e.g. due to defective material, manufacturing process
failure) and 3) catastrophic (e.g. due to faulty design, improper
storage, mishandling during transport) defects. These defects cause
a degradation in the performance due to accuracy losses or size and
shape changes that occur at critical positions in the tools. Such
defects can even result in tool breakage [5]. Rejects and downtime
increase the costs and profit losses. In industrial economics, the use
of a standby system is a suitable method to overcome this problem
[7]. However, the high tool costs caused by expensive materials,
long lead times and complex manufacturing make repairing/
restoring the geometries with AM a more economical approach
[6,8].
Powder-based processes such as Laser Powder Bed Fusion (L-
PBF) [9–19] or Laser Metal Deposition (LMD) [20–24] are predom-
inantly used to produce tool steels and are widespread in AM. Such
processes are among the leading technologies used to create com-
plex geometries and surface qualities from the scratch. However,
due to fact that the powder grain diameters are in the micrometre
range and that the dimensions of the working chambers are lim-
ited, the achievable deposition rates and component sizes are also
limited [25]. Hybrid structure/manufacturing methods (i.e. the
additive manufacturing of structures on pre-existing convention-
ally manufactured parts) are only partially suitable for PBF. As a
consequence, DED processes and especially WAAM are more suit-
able. WAAM increases the number of available established repair
and AM techniques and gives engineers a novel opportunity to
restore/rebuild a (worn and cracked) near-net-shape volume on a
larger scale. This, in turn, can result in significant cost savings
and shorter lead times. Cold metal transfer (CMT) has proven to
be a promising WAAM method due to the low heat input and dilu-
tion characteristics of the process [26–32]. This lower heat input
and dilution enables continuous deposition and prevents ‘‘burn-
through” on already deposited material [26]. Using the high-end
CMT process, it is possible to control the arc and droplet transfer
accurately to create more sophisticated structures [26,29].
However, few studies [33] on WAAM have been carried out on
the processing hot-work tool steels. Processing medium-carbon
martensitic hot-work tool steel grades in AM, in particular WAAM,
remains a challenge due to susceptibility to cracking due to the
carbon content and complex thermal cycling/residual stresses.
X38CrMoV5-1 represents this particular group of materials, and
is widely used for conventional forging dies or casting mould. It
is generally and conditionally weldable in both the annealed, hard-
ened and tempered states. Sufficient preheating above the marten-
site starting temperature (M
s
) is mandatory. The preheating
temperature should not drop below 325 °C during the actual weld-
ing process and should not exceed 475 °C[34]. Just a few alloys are
available and suitable as solid filler wires for welding/processing
on chromium hot-work tool steels. The guidelines issued by weld-
ing consumable manufacturers [34,35] indicate that the solid wire
grade (SW) X10CrMo6-3 is one of the few suitable alloys, because
it is characterized by a high hot wear resistance and a sufficient
toughness level. Casati et al. [10] stated that such leaner composi-
tion of hot-work tool steels in terms of carbon concentration is of
great interest for 3D printing of complex parts, as they are
expected to have higher toughness as well as higher damage toler-
ance to residual stresses and processing defects. Stockinger et al.
[28] demonstrated that such carbon lean grades could also be used
for wire-based additive manufacturing and the fabrication of
hybrid structures or structural parts. Though, the availability of
suitable solid wires/welding consumables is extremely limited.
Unlike in conventional joint welding, where the filler metal and
base metal are diluted in the fusion zone, the microstructure cre-
ated by AM consists entirely of the filler metal. Conventional filler
metals are actually designed mainly for joint welding and not for
pure additive manufacturing; therefore, dilution is considered in
the chemical composition. Due to processing challenges and the
growing market, the interest in developing new alloys that are
specifically designed for fusion-based AM processes and especially
for wire-based AM is steadily growing [36–38]. However, such
developments are technically challenging and time-consuming,
which is why the portfolio of additive filler materials offered by
welding filler material suppliers remains limited: The available fil-
ler materials mainly remain the same or only vary in terms of their
specification in chemical composition for existing grades [39].
While powder-based processes greatly facilitate the develop-
ment/utilization of new alloying systems – as only the respective
powders need to be mixed in the appropriate composition – a
rather large amount of effort needs to be invested to create solid
wires regarding the alloy development, from the points of melting,
casting and drawing the wire to subjecting the finished wire to
experimental investigations. As alternatives to the complex and
wear-intensive approach normally used to study wire-based pro-
cesses, the following approaches described in the literature can
be used to develop/utilize new alloy systems: 1) tandem/dual solid
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
2
wire feeding [40–43], 2) additions of particle/refiner to solid wire
[44,45], or 3) applying the tubular metal-cored wire approach
[46–48].
In tandem/dual wire feeding, two filler materials are fed simul-
taneously at different feed rates, thus increasing the number of
possible alloys. This provides a certain degree of flexibility, but
does not solve the actual problem, since again only standard filler
materials can be used (i.e. no ‘‘freedom in alloy design”). This
approach is particularly suitable for (intermetallic) binary systems
that are difficult and costly to produce (e.g. Ti-Al [49–51], Ni-Ti
[52], Fe-Al [53]). The addition of smaller amounts of particles or
refiners to the pre-existing chemical composition can be made
directly to the solid wire [54–56] , between the layer deposition
steps by surface coating [57–60], or directly into the melt pool
[61]. Depending on the method of addition, the particles may be
uniformly or non-uniformly dispersed. Particles or refiners can be
added to provide secondary strengthening, prevent wear, increase
the number of nucleation sites for grain refinement, or for other
reasons. Even if these particles/refining additives improve the
existing properties, they are an additional option and do not allow
the free choice of an alloying system. Applying cored wires offers
the most flexible solution in terms of alloy development and prop-
erty customization.
Up until now, only a few standard solid filler wires have been
metallurgically optimized for welding and have been qualified
for joining various tool steel grades. The development of filler
wires optimized for AM processes is of interest, as this develop-
ment would establish a more efficient manufacturing process with
higher deposition rates and quality properties, such as increased
toughness and strength. To improve the properties of AM compo-
nents and increase the applicability of wire-based AM, a focus is
placed on solid wires and especially tubular cored wires [26].
Regarding wire-based additive manufacturing, the application of
new and customized grades by tubular metal-cored wires (MCW)
has great potential. The chemical composition of the deposited
material can be easily adjusted by changing the metal powder fill-
ing composition. Consequently, the material properties can be
easily tailored. Despite the promising applications, the use of
MCW in WAAM is still extremely limited [47]. The present work
was carried out to demonstrate how the chemical composition of
a chromium martensitic hot-work tool steel and its use as tubular
metal-cored wires in wire arc additive manufacturing using a mod-
ified short-circuit metal transfer can be applied and customized.
2. Experimental procedure
2.1. Materials
The chromium-martensitic hot-work tool steel X38CrMoV5-1
(1.2343, AISI H11) was used as substrate, and metal-cored wires
X10CrMoNi6-3-X (carbon lean), with a diameter of 1.2 mm, with
Ni additions of 0.5 wt% and 1 wt% were developed and used as filler
material in present study. The chemical composition of the sub-
strate and the filler metals are listed in Table 1.
High alloyed steel grades and their microstructures can be
described schematically by referring to the Schaeffler diagram
and the chromium Cr
eq
and nickel Ni
eq
equivalents of the materials.
Depending on the Cr
eq
and Ni
eq
of the individual materials or their
combinations, a specific range with a particular microstructure
composition can be estimated. Therefore, a phase prediction can
be made by following the phase field in the constitutional diagram,
as given by the Cr
eq
and Ni
eq
coordinates [62]. The Cr
eq
and Ni
eq
for
the applied materials is calculated as shown below (Eq. (1) and
(2)), and the particular coefficients represent the relative contribu-
tions of each element to the austenitic or ferritic stability [62]:
Cr
eq
¼%Cr þ%Mo þ1:5%Si þ0:5%Nb þ2:0%Ti ð1Þ
Ni
eq
¼%Ni þ30 %Cþ0:5Mn þ30 %Nð2Þ
The Cr
eq
,Ni
eq
and the chemical composition of the cored wires
and the base material are summarized in Table 1; their corre-
sponding positions in the Schaeffler diagram are marked in Fig. 1.
The Cr
eq
and the Ni
eq
are 7.85% and 11.60% for the base material
and 10.40% and 3.30% for the cored wire with the initial chemical
composition (i.e. without addition of nickel), respectively.
While the base material is in the fully martensitic range, the fil-
ler wire is in the transition from martensite + ferrite to full marten-
site (Fig. 1a). Since such filler metals were originally designed for
cladding purposes, diluting these filler metals with the base metal
in this particular case results in the formation of a fully martensitic
microstructure (represented by a solid red line in Fig. 1a). In the
case of AM, however, the AM components are chemical composed
solely of the filler metal (i.e. according to the Schaeffler diagram),
and a martensitic and ferritic microstructure is expected for the
metal filler wire in its initial state.
By adding alloying elements and thus increasing the Ni
eq
,a
transition from martensite + ferrite to fully martensite can be facil-
itated. By considering Equation (2), the respective alloying ele-
ments of carbon, nickel and manganese can be added to increase
Ni
eq
. However, the addition of carbon leads to an increased suscep-
tibility to cracking, limiting the weldability of these tool steel
grades. The addition of manganese also proves suitable only to a
limited extent. When active gas is used as a shielding gas, the
alloying elements of silicon and manganese have an affinity to
react with oxygen, act as getters and form detrimental silicon
and manganese oxides [63–68]. Non-metallic inclusions and oxi-
des form especially on a large scale in AM-manufactured compo-
nents and have a detrimental effect on the mechanical properties
of these components [69]. Thus, nickel was added at 0.5 wt% and
1.0 wt% to increase the Ni
eq
from 3.30% to 3.80% and 4.30%, respec-
tively. According to Fig. 1b, the addition of nickel would be suffi-
cient to form a completely martensitic microstructure.
Once the desired chemical composition had been selected, a
cored / filler wire design was needed. Cored wires can be produced
in two types, seamless and folded, the folded types being of inter-
est for the present study. The outer sheath of the folded cored
wires can be made from a strip of unalloyed or high-alloy base
material, which is rolled into a U-shape and continuously filled
Table 1
Nominal chemical composition and derived chromium and nickel equivalents of the used steel grade substrate X38CrMoV5-1 and modified metal-cored wires X10CrMoNi6-3-X
with different Ni contents.
CCrMoNiMnSiCr
eq
Ni
eq
[wt%] [wt%] [wt%] [wt%] [wt%] [wt%] [wt%] [wt%]
Substrate
X38CrMoV5-1
0.38 5.20 1.30 – 0.40 0.45 7.85 11.60
Metal Cored Wires
X10CrMoNi6-3-X
0.10 6.50 3.3 0.00
0.50
1.00
0.60 0.40 10.40 3.30
3.80
4.30
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
3
with the powder mixture of previously determined chemical com-
position. The U-shape is then closed by rolling equipment to form a
tubular wire, which is finally reduced to the final diameter by
drawing and coiled onto a spool for processing in arc welding pro-
cesses. The design of the cored wires in traditional joint welding
with cored wires aims to reduce the thickness of the outer metal
sheath. This results in a high current density and thus a deep pen-
etration. High deposition rates and high productivity can be
achieved in Metal Active Gas (MAG) welding by using a spray-arc
metal transfer together with fine powder alloying elements [70–
73]. In contrast, CMT is a modified short-circuit metal transfer pro-
cess and the geometric/electrical requirements for the wire
changes. The outer sheath of the cored wires conducts the electric
current. To facilitate the short-circuiting required for the CMT pro-
cess and to achieve similar welding characteristics as compared to
those of solid wires, the thickness of the outer unalloyed steel
sheath was maximized (i.e. no additional iron was added to the
powder filling). The conductive cross-sectional area of the MCW
was 0.81 mm
2
, i.e. 28% smaller than the cross-sectional area of a
comparable solid wire. This corresponds to an increased current
density of 40% as compared to current density of the solid wire.
While the unalloyed steel sheath donates the iron content, the
primary alloying elements of chromium, molybdenum and a cer-
tain amount of nickel were added as powder filling (Fig. 2). The
powdered alloying elements are usually added in different weight
proportions with varying size distributions to achieve the maxi-
mum density of the cored wires.
2.2. Welding equipment
The shielding gas M21-ArC-18, which is 18 vol% CO
2
and 82 vol
% argon according to ISO 14175:2008, was used with a gas flow
rate of 15 l/min. A CMT system (Fronius International, Wels, Aus-
tria) was used for welding, consisting of a Fronius TPS 400i welding
power source, a WF 60i Robacta Drive CMT push–pull torch unit
and a wire buffer. The motion control system consisted of an ABB
IRB 140 six-axis articulated robot, an ABB IRBP A positioner and
the IRC5 control system. The robot arm movement and the
mounted welding torch were programmed using RAPID, a pro-
gramming language that enables the control of ABB industrial
robots. The ABB controller hosted the program, which controls
the robot movements and communicates with the welding power
source via a bus interface.
2.3. Experimental parameters
A Central Composite Design (CCD) was carried out in advance to
determine the welding parameters and the influence of the process
parameters on the properties of the weld bead. The following pro-
cess parameters were varied and investigated: 1) welding current,
ranging from 178 to 262 A and 2) welding speed, ranging from 0.65
to 1.35 cm/sec. In the present study, the ‘‘low power” condition
was selected, and the following combination of process parameters
was used: 190 A welding current and 0.75 cm/sec welding speed.
The process parameters/configurations used are listed in Table 2.
Volumetric AM structures were fabricated from the different
cored wire grades (i.e. nickel content), and the microstructure
and mechanical properties were subsequently characterized. Min-
imum dimensions of 80 20 120 mm are required for sampling
and multi-axis mechanical testing (Fig. 3b). The structures were
manufactured by stacking them layer-by-layer. Each layer con-
sisted of three weld beads with an axial offset
g
of 65% of the weld
bead width to the adjacent weld track (Fig. 3a). The value for the
axis offset/overlap distance and the applied welding sequence
were selected based on values reported in the literature (axial off-
set [28,29,74,75], welding sequence [28,29,76]). The interlayer
temperature between each layer was manually monitored with K
thermocouples (Ø 0.2 mm). The subsequent layer was deposited
on top of the previous one once the interlayer temperature had
reached 350–400 °C (above the M
S
temperature). The total number
of layers and the average build-up rates depended on the welding
parameters used, i.e. the material input. For the specific parameter
set, 50 layers with an average build-up rate of 1.75 mm/layer were
applied to build the predefined structure.
2.4. Characterization of the built structures
2.4.1. High-speed imaging
The metal transfer in CMT for the X10CrMo6-3 cored wire and
reference solid wire was captured with a Photron Fastcam SA1
high-speed camera (Photron, Tokyo, Japan) at 2,000 frames/s with
a resolution of 640 640 pixels. The filler wire grades were depos-
ited on a non-preheated X38CrMoV5-1 substrate at a welding cur-
rent and welding speed of 232 A (equivalent to a wire feed rate of
7.6 m/min) and 0.75 cm/s, respectively. The shielding gas used was
M21-ArC-18 at a gas flow rate of 15 l/min. The results obtained
Fig. 1. a) Schematic illustration of the Schaeffler diagram and b) detailed view of a specific area with the materials used marked in red; F.: ferrite, M.: martensite, A.: austenite,
BM.: base metal, MCW.: metal-cored wire. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
4
were compared based on the material transfer frequency, arc
geometry and visible arc length.
2.4.2. Microstructure characterization
The microstructures of the as-built structures were character-
ized by light optical microscopy (LOM), scanning electron micro-
scopy (SEM) equipped with an electron backscattered diffraction
(EBSD) camera and an energy-dispersive X-ray spectroscopy
(EDXS) detector.
The light optical microscopy analysis was performed using a
Zeiss Observer Z1m microscope (Carl Zeiss Microscopy, Jena, Ger-
many). The cross-sections, perpendicular to the weld direction
(ND-TD plane), were hot mounted, ground with SiC paper up to
Fig. 2. Scanning electron microscopy (SEM) analysis using energy dispersive X-ray spectroscopy (EDXS) and showing the elemental maps of the reference solid wire
X10CrMo6-3 and the metal-cored wires X10CrMoNi6-3-X for iron and the primary alloying elements of chromium, molybdenum and nickel. Scale bar: 500 mm.
Table 2
Selected process parameters/configuration for metal-cored wire based WAAM via
Cold Metal Transfer.
Attribute Unit Values
Welding current I
weld
A 190
Welding speed V
weld
cm/sec 0.75
Electrode stick-out –mm 15
Work angle –°0
Travel angle –°0
Gas flow rate –l/min 15
Arc length correction –– None
Arc dynamic correction –– None
Fig. 3. Schematic illustration of the a) applied welding sequence and b) side view of the volumetric AM structure with the minimum required dimensions; ND.: normal
direction, TD.: transverse direction, WD.: welding direction.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
5
#2000 grit and polished with a silicon oxide polishing suspension
(OPS). The cross-sections were wet etched for 20 s by using a mod-
ified Lichtenegger Bloech etchant composed of 100 ml distilled
water, 0.75 g ammonium hydrogen difluoride and 0.9 g potassium
disulfate.
For electron microscopy, the samples with the different chemi-
cal compositions were prepared similarly without wet etching. The
scanning electron microscopy (SEM) investigation was performed
using a TESCAN Mira3 microscope (TESCAN, Brno, Czech Republic)
equipped with a Super Octane energy dispersive spectroscopy
(EDS) and a Hikari camera and a TEAM software package for the
electron backscatter diffraction (EBSD) analysis. The EDXS analysis
was carried out using an acceleration voltage of 15 kV, working
distance of 15 mm and a spot size of 40 nm. The EBSD measure-
ments were performed on an area of 450 mm450 mm at a step
size of 0.3 mm. An acceleration voltage of 25 kV, a working distance
of 20 mm and a spot size of 15 nm were used during the EBSD mea-
surements. The generally accepted transition angle between the
low-angle grain boundary (LAGB) and high-angle grain boundary
(HAGB) of 15°in the case of an ideal arrangement of dislocations
[77] was used to define a high-angle grain boundary. The EBSD
measurement was analyzed using the OIM Analysis v.8 software.
A minimum grain size of 0.5 mm was chosen to guarantee that each
grain consisted of at least five measured pixels. The confidence
index was then standardized. Finally, the EBSD data were cleaned
considering a minimum confidence index of 0.3 correlated to the
neighbour grains.
X-ray diffraction (XRD) was performed on AM cross-sections in
the ND-TD plane using a Rigaku MiniFlex600 benchtop X-ray
diffractometer (Rigaku, Tokyo, Japan). The testing was carried out
at room temperature on trimmed, polished specimens with a min-
imum 10 10 mm cross-sectional area.
The entire AM cross-sections (ND-TD plane) were scanned auto-
matically using a Keyence VHX 6000 3D microscope. In total, 2,706
(0.0 wt% Ni), 2,652 (0.5 wt% Ni), and 2,800 (1.0 wt% Ni) images
were captured at 500x magnification and further processed. The
captured RGB images were converted to BW images using a devel-
oped routine in Matlab (The Mat Works, Natick, MA, United States)
and an RGB filter mask to capture the etched delta ferrite colour
shades. The ratio of dark to bright pixels in the BW images indi-
cates the phase fraction of delta ferrite. However, the converted
images still contain randomly scattered individual dark and bright
pixels that do not match the local appearance (i.e. salt-and-pepper/
impulsive noise). To effectively reduce/remove this noise, a median
filter was applied with different M M window sizes (3 3to
11 11 px). According to the median filter, the pixel in the centre
of an M M neighbourhood is replaced by the median value of the
corresponding window, resulting in a reduction in the number of
noisy pixels. Therefore, the threshold values for the delta ferrite
content could be determined for cross-sections with different
chemical compositions.
2.4.3. Characterization of mechanical properties
For the mechanical characterization, different sample orienta-
tions (i.e. horizontal and vertical orientations) were considered to
examine the mechanical anisotropy of the materials. Mechanical
properties were determined by 1) taking hardness measurements,
2) conducting Charpy V-notch tests and 3) tensile tests. To statisti-
cally validate the results, three specimens each were subjected to
the tensile and Charpy V-notch tests, and the mean values and
standard deviations were derived.
The hardness measurements were performed with an auto-
mated EMCO M1C (EMCO Test, Kuchl, Austria) hardness machine
according to the standard DIN EN ISO 6507–4: 2006–03. The Vick-
ers hardness (HV) using a load of 1 kp and a dwell time of 15 s was
measured. The distance between adjacent indentations was 1 mm,
and hardness maps for the AM cross-sections (ND-TD plane) with
different chemical composition were derived from the measure-
ments. In addition, microhardness measurements were carried
out with a load of 0.01 kp / 0.1 kp and a dwell time of 15 s in an
Anton Paar MHT4 (Anton Paar, Graz, Austria) hardness machine,
enabling us to determine the Vickers hardness of the individual
phases.
The tensile tests were carried out at room temperature for ver-
tically (fracture plane WD-TD) and horizontally (fracture plane
ND-TD) oriented specimens taken from the AM structures. Tensile
tests were performed according to DIN EN ISO 6892–1 using a
Zwick Roell ZMART PRO machine (Zwick Roell, Ulm, Germany).
Round specimens according to DIN 50125-B 6x30 were selected
as the geometry. The samples were initially preloaded to 200 N
and then loaded under stroke control to fracture at a constant test
speed of 1 mm/min.
Impact toughness was determined for the WD-TD plane (verti-
cal orientation) and ND-TD plane (horizontal orientation) of the
different AM structures. Testing was performed in accordance with
DIN EN ISO 148–1 using Charpy ISO V-notch specimens. The
Charpy tests were performed on a 300 J pendulum impact tester
(Otto Wolpert Werke) at room temperature.
2.4.4. Chemical composition
The chemical composition of the AM structures prepared from
the different cored wires were determined with atomic absorption
spectroscopy. Samples with minimum dimensions of
25 25 15 mm were taken from the centre of the structure
height (about 40 mm) to avoid dilution with the substrate. The
position and orientation (ND-TD plane) of the samples were always
the same to ensure consistency.
2.4.5. Thermodynamic equilibrium calculations
To describe the influence of the nickel addition in more detail,
thermodynamic equilibrium calculations based on the nominal
composition of the filler wires (Table 1) were performed using Mat-
calc 6 software and applying the database mc_fe_v2.060.tdb (Mat-
Calc Engineering, Vienna, Austria).
3. Results and discussion
3.1. High-speed imaging
Fig. 4 shows the high-speed images captured of weld bead
deposition in the front-view for standard solid wire and tailored
cored wires in different stages of the CMT process. First, the arc
is ignited by retracting the filler wire (Fig. 4a,e), followed by a max-
imum visible arc length 2 ms later when the direction of motion is
reversed from retraction to advancement (Fig. 4b,f). The arc geom-
etry and the visible arc length for the solid wire and the cored wire
differ in their visual appearances. This difference can be attributed
to the welding current concentration in the peripheral area of the
cored wire cross-section. While the solid wire shows a more coni-
cally shaped arc and a reduced arc length (Fig. 4b), the cored wire
shows a bowl-shaped arc and an increased arc length (Fig. 4f). The
maximum visible arc length for the solid wire is 1.99 mm, and
for the cored wire, 2.64 mm, respectively. In the arcing phase,
an increased amount of fine spatter for the cored wire (Fig. 4f) as
compared to the solid wire (Fig. 4b) is observed. Later, the filler
wire is moved towards the substrate, and the arc starts to diminish
until it extinguishes upon contact with the substrate and a short
circuit occurs (Fig. 4c,g). Finally, after the short-circuit period ends
and the filler wire is retracted again, the arc is re-ignited (Fig. 4d,g),
and the sequence is repeated until the process stops.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
6
The analysis of the high-speed captured images of weld beads
deposited once with solid wire and once with cored wire in the
CMT process shows only minor discernible deviations and similar
welding characteristics. The arc and short-circuit phases, 5.5 /
4.5 ms for the solid wire and 6.5 / 4.0 ms for the cored wire, differ
slightly. This results in an almost identical total duty cycle until the
point of arc re-ignition, i.e. 10.0 ms for the solid wire and 10.5 ms
for the cored wire, respectively. A duty cycle and material transfer
rate of approx. 100 Hz was derived.
3.2. Chemical composition
After the appropriate material transfer mode of the MCW was
established, the chemical composition of the AM structures was
measured to determine the absence of severe elemental loss during
processing, e.g. by evaporation or spattering.
Table 3 shows the chemical compositions of the AM structures
fabricated from the different grades of cored wire as determined by
atomic absorption spectroscopy in the ND-TD plane. Since the
unalloyed steel sheath supplies iron, and the sheath remains con-
stant for all cored wire grades, the addition of nickel to the pow-
dered filler also slightly changes the ratio of the main alloying
elements. The fractions of the main alloying elements chromium
and molybdenum increase slightly as the Ni content increases.
The actual nickel content of the AM structures, i.e. 0.62 wt% and
1.22 wt%, is marginally higher than the desired nominal values of
0.50 wt% and 1.00 wt%. The manganese and silicon contents for
the cored wires produced are about equal, at approx. 0.7 wt% and
0.4–0.5 wt%, respectively. The derived actual Cr
eq
and Ni
eq
are
10.16% and 4.90%, 11.22% and 5.78%, as well as 11.47% and
6.40%, respectively, when the nickel content is increased. No signif-
icant deviation from the target composition or significant element
loss was measured. Considering the usual manufacturing toler-
ances of chemical composition for filler wires, the fabricated filler
wires and the chemical analysis of the AM components in this
study indicate that the desired chemical composition could be
achieved.
The positions in the Schaeffler diagram, indicating the actual
chemical compositions of the AM structures, are shown in Fig. 5.
Considering the actual composition and the different Cr
eq
/Ni
eq
,it
can be assumed that all AM structures are in the full martensitic
range. These are slightly offset and arranged along the transition
from martensite + ferrite to full martensite.
3.3. Microstructure characterization
The light optical micrographs (ND-TD plane) of the AM struc-
tures fabricated with the different cored wires with different
chemical compositions ((a,d) 0.0 wt% Ni, (b,e) 0.5 wt% Ni, (c,f)
1.0 wt% Ni) are shown in Fig. 6. The microstructure of an AM built
using the cored wires without the addition of nickel consists of
predominantly polygonally shaped delta ferrite islands in a
Fig. 4. Front-view high-speed captured images of weld bead deposition for solid wire X10CrMo6-3 (a-d) and metal-cored wire X10CrMo6-3 (e-h) on X38CrMoV5-1 for
different stages of the CMT process.
Table 3
Actual chemical composition of the fabricated AM structures from the cored wires as measured by optical emission spectroscopy in the ND-TD plane and derived from the actual
chromium and nickel equivalent.
CCrMoNiMnSiCr
eq
Ni
eq
[wt%] [wt%] [wt%] [wt%] [wt%] [wt%] [wt%] [wt%]
X10CrMo6-3
+ 0.0 wt% Ni
0.15 6.16 3.34 0.05 0.71 0.44 10.16 4.90
X10CrMo6-3
+ 0.5 wt% Ni
0.16 6.86 3.63 0.62 0.72 0.49 11.22 5.78
X10CrMo6-3
+ 1.0 wt% Ni
0.16 7.01 3.71 1.22 0.75 0.50 11.47 6.40
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
7
martensitic matrix (Fig. 6a,d). The addition of 0.5 wt% and 1.0 wt%
Ni altered the morphology of the delta ferrite from predominantly
polygonal islands to a mixture of polygonal islands/grain boundary
network (Fig. 6b,e) to a continuous grain boundary network
(Fig. 6c,f). The refinement of the microstructure and a reduction
in the delta ferrite content are the main observed effects of adding
nickel. Regarding the microstructure of AM structures of
X10CrMo6-3 + 1.0 wt% Ni, precipitates were specifically formed
in the region of continuous delta ferrite networks (Fig. 6f).
According to Fig. 1a and Fig. 5, the addition of nickel would be
sufficient to form a fully martensitic microstructure; however, the
experimental results slightly deviate, and a small fraction of fine
grain boundary delta ferrite networks remains (Fig. 6c). The nickel
addition leads to a significant reduction in the delta ferrite, but not
to a complete suppression of its formation.
Due to the similarities in the crystal structure of martensite
(bct) and ferrite (bcc), it is challenging to distinguish and quantify
them using diffraction techniques (e.g. XRD, EBSD). Image analysis
was used to determine the phase fraction of delta ferrite at the
macroscopic level in the ND-TD plane. While the presence of many
dark pixels characterizes the polygon shape in areas of high local
density, the grain boundary network is represented by compara-
tively thin, interconnected dark pixels in areas of lower local den-
sity (Fig. 7b). Especially the latter tends to be underestimated in its
phase fraction, if the median filter is applied with large window
size. The finer-grained delta ferrite networks are not considered
at window sizes of larger than 7x7 pixels. Thus, a trade-off occurs
between removing noisy pixels and removing the actual delta fer-
rite that leads to an inherent error in the method. Fig. 7d shows the
delta ferrite in black when a median filter with a window size of
99 pixels was applied. The areas highlighted in red represent
the grain boundary delta ferrite networks which have vanished
(compared to median filter 5 5 px, Fig. 7c). Since the most signif-
icant phase fraction of delta ferrite consists of polygonally shaped
islands and thick grain boundary networks, the error attributed to
the window sizes consists of a minor percentage (Fig. 8a).
The mean values of the phase fraction of delta ferrite in the
entire cross-sections (ND-TD plane) for the AM structures fabri-
cated with MCW with different nickel contents are shown in
Fig. 8. Depending on the noise reduction/removal method applied
(i.e. median filter px px), the delta ferrite content ranges from
31.8 to 24.1, 22.8–13.1 and 20.6–12.1 for the alloy containing
0 wt%, 0.5 wt% and 1.0 wt% nickel, respectively. The nickel addi-
tions to the hot-work tool steel in the range of 0.5 wt% to 1.0 wt
% reduced the delta ferrite by 10%.
Fig. 5. Detail of a specific area of the Schaeffler diagram with the actual
composition materials used marked in red; F.: ferrite, M.: martensite, A.: austenite,
BM.: base metal, 1,2,3.: AM structures fabricated with the different MCW (legend).
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
Fig. 6. Light optical microscopy of representative microstructures of built-up AM structures made out of metal-cored wires with varying Ni-content 0% (a,d), 0.5% (b,e) and
1.0% (c,f).
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
8
Fig. 9 shows the X-ray diffractograms of the AM cross-sections
(ND-TD plane). For all samples, only bcc-
a
iron is indexed, i.e.
the diffraction peaks (110), (200), (21 1) and (22 0) were detected.
The diffraction peaks can be assigned to the bcc/bct crystal struc-
ture. This assignment indicates that the fcc crystal structure is
either not present or cannot be found in a minor fraction and is
undetectable with the XRD measurements. Thus, the presence of
retained austenite is negligible in this work.
The equilibrium phase fractions as a function of temperature for
the cored wires without and with 1 wt% nickel are shown in
Fig. 10a and Fig. 10b. Equilibrium calculations show that the addi-
tion of nickel in the respective percentage range has no significant
Fig. 7. a) Raw RGB image captured at 500x magnification and (b) related converted BW image obtained by using RGB mask to filter the delta ferrite (black); noise reduction/
removal based on median filtering technique with different window sizes: (c) 5 5 px and (d) 9 9 px (delta ferrite grain boundary network that has vanished highlighted in
red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 8. a) Phase fraction of delta ferrite (in ND-TD plane) in area % for X10CrMo6-3 with varying nickel contents, depending on the window sizes used for the median filter and
b) phase fraction (in ND-TD plane) for varying window sizes for the median filter, depending on the nickel content.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
9
effect on the chemical composition, phase fraction and solubility of
the precipitates M
23
C
6
and MX, as these remained the same in both
calculations (Fig. 10a,b). Maximum equilibrium phase fractions of
2.2% M
23
C
6
and of 0.6% MX were predicted by the simulation.
The M
23
C
6
precipitates should consist of molybdenum and chro-
mium in combination with carbon. The nickel donation predomi-
nantly affects the transformation of
a
?
c
?dand vice versa. Ni
acts as austenite stabilizer in the present study, widening the
austenite area by changing the equilibrium critical temperatures
A
1
,A
3
and A
4
, as shown in Fig. 11. The addition of 1.0 wt% nickel
decreases the temperatures of A
1
,A
3
and increases A
4
, as shown
in Table 4.
The results of the energy dispersive spectroscopy (EDS) analysis
of the AM microstructure (ND-TD plane) made of X10CrMo6-3 + 1.
0 wt% Ni are shown in Fig. 12. The backscattered electron (BSE)
micrograph in Fig. 12a and the corresponding element mapping
in Fig. 12b-g show that the microstructure comprises delta ferrite
grain boundary networks, a martensitic matrix and precipitates
covering the boundaries (Fig. 12a). The continuous delta ferrite
network contains increased amounts of molybdenum and chro-
mium (Fig. 12b,c), while a nickel enrichment is observed in the
matrix (Fig. 12d). The large precipitates are molybdenum-based
and are located at the interface between the martensite matrix
and the delta ferrite (Fig. 12b). The presence of macroscopic imper-
fections (i.e. voids, cracking) could not be detected in all examined
AM cross-sections, but randomly scattered cavities containing non-
metallic inclusions (NMI) with an average diameter of approx.
Fig. 9. X-ray diffraction patterns of the AM structures (in ND-TD plane) made of
cored wires X10CrMo6-3 with 0.0 wt%, 0.5 wt% and 1.0 wt% nickel, respectively. The
peaks of bcc-iron (ferrite) are identified.
Fig. 10. Phase fraction as a function of temperature using thermodynamic simulations for the nominal chemical composition of the metal-cored wires used a) without and b)
with 1.0 wt% nickel.
Fig. 11. Phase fraction of
a
- and d-ferrite and austenite as a function of temperature
using thermodynamic simulations for the nominal chemical composition of the
metal-cored wires used without (black) and with 1.0 wt% (red) nickel. (For
interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
Table 4
Transformation temperatures for the X10CrMoNi6-3 alloys used without and with
1 wt% nickel, calculated on the basis of thermodynamic equilibrium calculations.
A
1
A
3
A
4
[°C] [°C] [°C]
X10CrMo6-3 + 0.0 wt% Ni 839 984 1178
X10CrMo6-3 + 1.0 wt% Ni 748 925 1245
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
10
1mm were seen. EDS analysis results show that the NMIs are sili-
con and manganese oxides (Fig. 12e-g).
Fig. 13 shows the EBSD results for the AM cross-section (ND-TD
plane) of the three investigated alloys. The inverse pole figure (IPF)
maps in Fig. 13(a-c) and the image quality (IQ) maps in Fig. 13(g-i)
show the martensitic structure of the matrix with the islands of
polygonally shaped delta ferrite. The geometrically necessary dis-
location (GND) maps show the high values of GNDs in the matrix,
while negligible values were measured in the delta ferrite, as
shown in Fig. 13(d-f).
The low- and high-angle grain boundary (LAGB and HAGB,
respectively) densities are shown in Fig. 14. The increase in Ni con-
tent diminishes the formation of delta ferrite, increasing the
boundary density of both LAGB and HAGBs. Therefore, a higher
Ni content promotes the refinement of the microstructure of the
X10CrMo6 steel. Additionally, the geometrically necessary disloca-
tion (GND) density is another parameter that can be assessed to
evaluate the effect of the Ni content on the microstructural modi-
fications in the X10CrMo6-3 steel, Fig. 14. Particularly in the lower
regions of the GND (highlighted in gray in Fig. 14b), fraction
decreases when increasing the Ni content; this effect is more nota-
ble with the addition of 1 wt% Ni. The range with low values of
GND can be attributed to the polygonal delta ferrite. Thus, a lower
amount of polygonal delta ferrite leads to a decrease in the number
fraction of GND in this low-value range.
3.4. Mechanical characterization
3.4.1. Hardness measurements
The hardness maps of the cross-sections in the ND-TD plane and
the mean hardness values as a function of the height of the AM
structures for the respective alloys are shown in Fig. 15a and
Fig. 15b, respectively. A homogeneous macroscopic hardness dis-
tribution in terms of height and width is observed for each alloy,
Fig. 15a. The mean hardness values per layer along the height is
nearly constant for the different alloys, and these do not deviate
from the mean hardness values for the entire cross-sections
(Fig. 15b). Despite the frequently reported softening effect of the
subsequent layers on the preceding ones [29] for the AM of related
material grades, this effect was not detected in the investigated
alloys. The hardness values for the last layer, i.e. the final weld
beads, fall within the range of the mean hardness values for the
respective alloy (Fig. 15b dashed lines) and do not differ signifi-
cantly. This may be attributed to the fact that the predefined pre-
Fig. 12. a) Scanning electron micrograph of the AM cross-section (ND-TD plane) made of X10CrMo6-3 + 1.0 wt% and b-g) corresponding to the energy dispersive X-ray
mapping analysis for the main metallic alloying elements of b) molybdenum, c) chromium, d) nickel and the non-metallic elements of e) oxygen, f) silicon and g) manganese.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
11
Fig. 13. Electron-backscattered diffraction (EBSD) analysis of the X10CrMo6-3: a,d,g) without Ni addition; b,e,h) with 0.5 wt% Ni addition; c,f,i) with 1 wt% Ni addition; a-c)
Inverse pole Fig. (IPF) maps; d-f) geometrically necessary dislocation (GND) maps; g-i) image quality (IQ) maps; identical investigated position in the AM structures (center),
i.e. similar experienced thermal cycles, for all samples, for all configurations; similar viewfield for all images a-i.
Fig. 14. a) Boundary density of low-angle grain boundaries (LAGB) and high-angle grain boundaries (HAGB); b) Fraction / distribution of geometrically necessary dislocations
(GND) for AM bulk material for the respective alloys; in grey boxed area represents the range of GND for the polygonal delta ferrite.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
12
heating/interpass temperature was above the M
s
temperature.
Thus, the structural transformation did not occur until the welding
process was complete and all parts had cooled to the ambient tem-
perature homogenously.
The mean hardness of the AM structures increases as the Ni
content increases. The average hardness values are 360 ± 42
HV1, 394 ± 43 HV1, to 429 ± 37 HV1 for the addition of 0.0 wt%,
0.5 wt% and 1.0 wt% nickel, respectively.
Representative BSE micrographs of the microstructures are
shown in Fig. 16. While the delta ferrite morphology at 0 wt%
nickel is predominantly polygonal (Fig. 16a), the delta ferrite in
the alloys with additional nickel occurred predominantly in the
form of a network/platelet embedded in the matrix, with numer-
ous ferrite/strain interfaces additionally covered by precipitates
(Fig. 16b,c). The distribution of the present phases (martensite
and delta ferrite) affect the strengthening mechanisms in the
material. Fig. 17a describes the strengthening mechanisms
involved for each phase. The delta ferrite itself shows only lattice
resistance and solid-solution strengthening mechanisms (
r
0
). The
matrix of martensite laths have two other strengthening mecha-
nisms: the Hall-Petch due to presence of the lath boundaries
(
r
HP
) and the high dislocation density (
r
q
). Since the delta ferrite
transforms from polygonal to thin plates as the Ni content
increases, the morphology of the matrix changes from case B (pre-
dominantly martensite matrix) to case A (mixed matrix) when the
nickel content is increased (Fig. 17b). This results in a decrease in
the hardness when increasing the Ni content of thin and soft,
indented delta ferrite platelets (Fig. 17c). Thus, the overall hard-
ness of the matrix decreases as the Ni content increases from
480 ± 32 HV0.1 to 452 ± 24 HV0.1 and 442 ± 9 HV0.1, respectively.
The single polygonal delta ferrite is notably soft with a hardness
of 180 HV0.01.
Overall, increasing the Ni content leads to:
Fig. 15. a) Hardness distribution of the AM bulk material for different Ni contents in the ND-TD plane, as measured by the Vickers HV1 method and as represented by
hardness maps; b) mean hardness levels in height (markings) and mean hardness of the AM bulk material for the respective alloys (dashed line).
Fig. 16. Micrographs (backscattered electron images) of representative microstructures for the X10CrMo6: a) without Ni addition, b) with 0.5 wt% Ni addition and c) with
1 wt% Ni addition.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
13
1. A reduction in the soft polygonal (delta) delta ferrite content,
2. An increase in the fraction of high-angle grain boundaries
(HAGB) (Fig. 14a), promoting higher Hall-Petch strengthening,
3. And a reduction in the low-value range fraction of geometrically
necessary dislocation (GND) densities (Fig. 14b) due to the
decrease in the fraction of polygonal delta ferrite
3.4.2. Impact toughness
Charpy impact tests were performed on V-notched specimens
oriented vertically (fracture plane WD-TD) and horizontally (frac-
ture plane ND-TD) at room temperature. The absorbed impact
energy is shown in Fig. 18 as a function of the chemical composi-
tion and testing direction. The higher the Ni content, the higher
the impact toughness for both fracture planes/orientations. The
absorbed impact energies are 4.0 ± 1.0 J, 6.0 ± 1.8 J, and 8.0 ± 2.8
J in the WD-TD plane (vertical) and 4.3 ± 0.6 J, 6.3 ± 1.0 J, and 8.7
± 3.1 J, in the ND-TD plane (horizontal) for the alloy containing
0.0 wt%, 0.5 wt% and 1.0 wt% nickel, respectively. The fractured
specimens showed no distinct lateral expansion. Moreover, no sig-
nificant differences were observed in the impact energy absorbed
between the WD-TD and ND-TD planes for the investigated alloys.
Representative horizontally (fracture plane ND-TD) oriented
fractured specimens with different nickel contents and their
macroscopic fracture surfaces are shown in Fig. 19. While the sam-
ples without nickel content show a macroscopically irregular frac-
ture surface, the specimens with nickel additions show a rather
planar and smooth fracture surface.
Fractography images for 0.0 wt% and 1.0 wt% nickel fractured
Charpy samples are shown in Fig. 20a-c and Fig. 20d-f, respectively.
The specimens without added nickel show predominantly cleavage
fractures with some minor areas of dimples in the crack initiation
region (Fig. 20b). Moreover, the jagged fracture surface is charac-
terized by different main fracture planes at different height levels
for the alloy without nickel content, showing a predominantly brit-
tle fracture mode and a subsequently smaller connecting ductile
region for height compensation (Fig. 20c).
On the other hand, the samples with 1.0 wt% nickel exhibit duc-
tile behaviour with deformed/sheared voids and plastic deforma-
tion (Fig. 20e). The voids vary in size within the ductile zones
and show no shear in contrast to the crack initiation region. In
addition, non-metallic inclusions (silicon, manganese oxides) on
the fracture surface were observed in all examined samples.
These findings corroborate with the hardness results, since
increasing the Ni content lowers the hardness values of the matrix
despite the higher GND density and density of boundaries. There-
fore, the increase in toughness observed as the Ni content increases
can be explained as the result of the combined effects of:
1. Lower fractions of delta ferrite (Fig. 8): the large polygonally
shaped or grain boundary network of soft delta ferrite leads
to load partitioning between the martensitic matrix and the
delta ferrite. This promotes strain localization and, thus, less
homogeneous deformation. The decrease in the delta fraction
promotes more homogeneous deformation effects during
impact.
Fig. 17. Schematic illustrations: a) the strengthening mechanisms in delta ferrite and martensite;
r
0
.: solid solution and lattice resistance,
r
HP
.: grain boundary
strengthening,
r
q
.: dislocation strengthening; b) hardness indentation of delta ferrite and different types of matrix; c) relationship between nickel content, content of fine
indented delta ferrite and hardness of matrix.
Fig. 18. Absorbed impact energy of AM bulk material of vertically (fracture plane
WD-TD) and horizontally (fracture plane ND-TD) oriented V-notched specimens
tested at room temperature.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
14
2. Softer martensitic matrix: the softer matrix of the 1.0 wt% Ni is
expected to exhibit more ductile behaviour as compared to the
harder martensitic matrix without Ni addition.
3. A higher fraction of boundaries (Fig. 14a): the higher the frac-
tion of boundaries, the finer the microstructure. Consequently,
this higher fraction means a higher density of obstacles, pro-
moting changes in the propagation of the fracture surface dur-
ing failure.
3.4.3. Tensile properties
The tensile properties of AM structures were determined for
vertically (fracture plane WD-TD) and horizontally (fracture plane
ND-TD) oriented specimens. Fig. 21 shows the dependence of the
yield strength R
P0.2
on the chemical composition and the direction.
The higher the Ni content, the higher the yield strength in both
planes/orientations. The yield strengths are 776 ± 18 MPa,
910 ± 18 MPa and 936 ± 4 MPa in the WD-TD plane (vertical)
and 763 ± 13 MPa, 842 ± 22 MPa and 891 ± 41 MPa, in the ND-
TD plane (horizontal) for the alloy containing 0.0 wt%, 0.5 wt%
and 1.0 wt% nickel, respectively. The influence of the Ni content
on the yield strength is significantly related to its effect on the
Fig. 19. Fractured horizontally oriented (fracture plane ND-TD) Charpy-V notch specimens of AM structures built up with metal-cored wires with varying Ni content of a)
0 wt%, b) 0.5 wt% and c) 1.0 wt%, respectively.
Fig. 20. Fractography on representative, fractured, horizontally oriented (fracture plane ND-TD) Charpy-V notch specimens of AM structures built up with metal-cored wires
with varying Ni contents of 0 wt% (a-c) and 1.0 wt% (d-f), respectively. Secondary electron micrographs of crack initiation area (b,e) and crack propagation area (c,f).
Fig. 21. Yield strength R
p0.2
of AM bulk material of vertically (fracture plane WD-
TD) and horizontally (fracture plane ND-TD) oriented specimens tested at room
temperature.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
15
overall hardness (section 3.4.1) and the microstructure of the alloy
(Fig. 13 and Fig. 16). Thus, the distribution of thin plates in the
martensitic matrix and mainly the decrease in the soft delta ferrite
fraction affect the yield strength.
4. Conclusions
The present work demonstrates the successful application of
cored wires for a modified short-circuit welding process and their
ability to adapt existing alloy systems for wire arc additive manu-
facturing. Starting from a reference filler wire of chromium hot-
work tool steel with a medium carbon content, cored wires with
different chemical compositions (i.e. nickel content) were pro-
duced. Due to the customized wire geometries, the arc characteris-
tics of the cored wires are comparable to those of the reference
solid wire. Only minor alterations in the arc geometry and arc
length were observed. The various cored wires produced were used
to fabricate AM structures. The microstructures of these wires
were characterized and their mechanical properties measured.
The results are summarized and allow us to draw the following
conclusions:
The microstructure of AM structures from fabricated cored
wires consists of delta ferrite and a martensitic matrix. The
addition of nickel decreases the fraction of detrimental delta
ferrite and changes its morphology from predominantly polyg-
onally shaped to a preferable fine continuous grain boundary
network.
These delta ferrite networks are enriched with the main alloy-
ing elements molybdenum and chromium and contrasts with
the matrix containing nickel. Molybdenum-based precipitates
are formed in the interfaces between the matrix and delta
ferrite.
The increase in the Ni content leads to a higher density of low-
and high-angle grain boundaries as well as a higher density of
geometrically necessary dislocations (GND), and improves the
mechanical properties.
The hardness increases as the Ni content increases due to a
lower fraction of soft polygonal delta ferrite. The microhardness
of the matrix decreases as the Ni content increases due to the
decrease in chemical element partitioning caused by the pres-
ence of delta ferrite.
The higher the Ni content, the lower the fractions of delta fer-
rite, the softer the martensitic matrix and the higher the frac-
tion of boundaries, enhancing the impact toughness. The
fracture mode changes from a predominantly brittle cleavage
to a mix of brittle/ductile fracture.
The yield strength increases as the Ni content increases due to
the decrease in the soft delta ferrite fraction, which promotes
a more homogeneous deformation.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgements
The authors carried out this work as part of the COMET pro-
gramme within the COMET K2 Center project ‘‘Integrated Compu-
tational Material, Process and Product Engineering (IC-MPPE)”
(Project No 859480). The COMET programme is supported by the
Austrian Federal Ministries for Transport, Innovation and Technol-
ogy (BMVIT) and for Digital Economic Affairs (BMDW), represented
by the Austrian Research Promotion Agency (FFG), and the Aus-
trian federal states of Styria, Upper Austria and Tyrol. RHB
acknowledges the D-1303000107/CD-Laboratory for Design of
High-Performance Alloys by Thermomechanical Processing and
the support of the Christian Doppler Forschungsgesellschaft. FR
gratefully acknowledges the financial support of the Austrian
Science Fund (FWF) as part of the project ‘‘Software Development
on Dislocation Creep in Alloys” (P-31374). The authors thank
Alexander Lattner for performing the tensile and Charpy tests,
and Thomas Rath and Rafael Paiotti Marcondes Guimaraes for per-
forming the XRD measurements and their analysis.
References
[1] E. Doege, B.-A. Behrens, Handbuch Umformtechnik, Springer Berlin Heidelberg,
Berlin, Heidelberg, 2010.
[2] B. Buchmayr, Damage, Lifetime, and Repair of Forging DiesSchädigung,
Lebensdauer und Reparatur von Schmiedegesenken, BHM Berg- und
Hüttenmännische Monatshefte 162 (3) (2017) 88–93.
[3] R. Rahn and U. Schruff, ‘‘Improved tool performance and cost control in the
forging industry,” in Proceedings of the 8th International Tooling Conference,
RWTH Aachen, 2009, pp. 2–4.
[4] J. Mayerhofer and S. Zinner, ‘‘Innovative Stähle für Schmiedewerkzeuge - vom
Anforderungsprofil zur erfolgreichen Anwendung,” in Proceedings of the
Conference Schmiedewerkzeuge, Salzburg, 2011, pp. 117–135.
[5] S. Jhavar, C.P. Paul, N.K. Jain, Causes of failure and repairing options for dies and
molds: A review, Eng. Fail. Anal. 34 (2013) 519–535.
[6] M. Pleterski, J. Tušek, L. Kosec, M. Muhic
ˇ, T. Muhic
ˇ, Laser Repair welding of
molds with various pulse shapes, Metalurgija 49 (1) (2010) 41–44.
[7] G. Pantazopoulos, S. Zormalia, Analysis of the failure mechanism of a gripping
tool steel component operated in an industrial tube draw bench, Eng. Fail.
Anal. 18 (6) (2011) 1595–1604.
[8] J. Grum, J.M. Slabe, A comparison of tool-repair methods using CO 2 laser
surfacing and arc surfacing, Appl. Surf. Sci. 208–209 (1) (2003) 424–431.
[9] F. Huber et al., ‘‘Laser beam melting and heat-treatment of 1.2343 (AISI H11)
tool steel – microstructure and mechanical properties,” Mater. Sci. Eng. A, vol.
742, no. October 2018, pp. 109–115, 2019.
[10] R. Casati, M. Coduri, N. Lecis, C. Andrianopoli, M. Vedani, Microstructure and
mechanical behavior of hot-work tool steels processed by Selective Laser
Melting, Mater. Charact. 137 (January) (2018) 50–57.
[11] A. Temmler et al., Influence of laser polishing on surface roughness and
microstructural properties of the remelted surface boundary layer of tool steel
H11, Mater. Des. 192 (Jul. 2020) 108689.
[12] R. Dörfert, J. Zhang, B. Clausen, H. Freiße, J. Schumacher, F. Vollertsen,
Comparison of the fatigue strength between additively and conventionally
fabricated tool steel 1.2344, Addit. Manuf. 27 (February) (2019) 217–223.
[13] M. Åsberg, G. Fredriksson, S. Hatami, W. Fredriksson, P. Krakhmalev, Influence
of post treatment on microstructure, porosity and mechanical properties of
additive manufactured H13 tool steel, Materials Science and Engineering: A
742 (2019) 584–589.
[14] J. Šafka, M. Ackermann, L. Volesky
´, Structural properties of H13 tool steel parts
produced with use of selective laser melting technology, J. Phys.: Conf. Ser. 709
(2016) 012004, https://doi.org/10.1088/1742-6596/709/1/012004.
[15] J. Krell, A. Röttger, K. Geenen, W. Theisen, General investigations on processing
tool steel X40CrMoV5-1 with selective laser melting, Journal of Materials
Processing Technology 255 (2018) 679–688.
[16] F. Deirmina, N. Peghini, B. AlMangour, D. Grzesiak, M. Pellizzari, Heat
treatment and properties of a hot work tool steel fabricated by additive
manufacturing, Mater. Sci. Eng. A 753 (January) (2019) 109–121.
[17] R. Mertens, B. Vrancken, N. Holmstock, Y. Kinds, J.-P. Kruth, J. Van Humbeeck,
Influence of powder bed preheating on microstructure and mechanical
properties of H13 tool steel SLM parts, Phys. Procedia 83 (2016) 882–890.
[18] M. Narvan, A. Ghasemi, E. Fereiduni, S. Kendrish, M. Elbestawi, Part deflection
and residual stresses in laser powder bed fusion of H13 tool steel, Mater. Des.
204 (2021) 109659.
[19] M. Narvan, A. Ghasemi, E. Fereiduni, M. Elbestawi, Laser powder bed fusion of
functionally graded bi-materials: Role of VC on functionalizing AISI H13 tool
steel, Mater. Des. 201 (2021) 109503.
[20] O. Hentschel et al., ‘‘Processing of AISI H11 tool steel powder modified with
carbon black nanoparticles for the additive manufacturing of forging tools
with tailored mechanical properties by means of laser metal deposition
(LMD),” Metals (Basel)., vol. 8, no. 9, 2018.
[21] D. Junker, O. Hentschel, M. Schmidt, M. Merklein, Investigation of Heat
Treatment Strategies for Additively-Manufactured Tools of X37CrMoV5-1,
Metals (Basel) 8 (10) (2018) 854.
[22] A. Bohlen, H. Freiße, M. Hunkel, F. Vollertsen, Additive manufacturing of tool
steel by laser metal deposition, Procedia CIRP 74 (2018) 192–195.
[23] J. Choi, Y. Chang, Characteristics of laser aided direct metal/material deposition
process for tool steel, Int. J. Mach. Tools Manuf. 45 (4-5) (2005) 597–607.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
16
[24] N.S. Bailey, C. Katinas, Y.C. Shin, Laser direct deposition of AISI H13 tool steel
powder with numerical modeling of solid phase transformation, hardness, and
residual stresses, Journal of Materials Processing Technology 247 (2017) 223–
233.
[25] D. Ding, Z. Pan, D. Cuiuri, H. Li, Wire-feed additive manufacturing of metal
components: technologies, developments and future interests, Int. J. Adv.
Manuf. Technol. 81 (1-4) (2015) 465–481.
[26] P. Gerhard, K. Ferdinand, and C. Harald, ‘‘Manufacturing of turbine blades by
shape giving CMT Welding,” Met. Addit. Manuf. Conf., no. November, p. 10, 2014.
[27] P. M. Sequeira Almeida and S. Williams, ‘‘Innovative process model of Ti-6Al-
4V additive layer manufacturing using cold metal transfer (CMT),” 21st Annu.
Int. Solid Free. Fabr. Symp. - An Addit. Manuf. Conf. SFF 2010, no. March, pp. 25–
36, 2010.
[28] J. Stockinger, C. Wiednig, N. Enzinger, C. Sommitsch, D. Huber, and M.
Stockinger, ‘‘Additive Manufacturing via Cold Metal Transfer,” Met. Addit.
Manuf. Conf., pp. 117–125, Nov. 2016.
[29] J. Plangger, P. Schabhüttl, T. Vuherer, N. Enzinger, Cmt additive manufacturing
of a high strength steel alloy for application in crane construction, Metals
(Basel) 9 (6) (2019) 1–14.
[30] J. Gu, J. Ding, S.W. Williams, H. Gu, P. Ma, Y. Zhai, The effect of inter-layer cold
working and post-deposition heat treatment on porosity in additively
manufactured aluminum alloys, J. Mater. Process. Technol. 230 (2016) 26–34.
[31] T. Lehmann et al., Concurrent geometry- and material-based process
identification and optimization for robotic CMT-based wire arc additive
manufacturing, Mater. Des. 194 (2020) 108841.
[32] X. Zhang et al., Study on microstructure and tensile properties of high nitrogen
Cr-Mn steel processed by CMT wire and arc additive manufacturing, Mater.
Des. 166 (Mar. 2019) 107611.
[33] Y. Ali, P. Henckell, J. Hildebrand, J. Reimann, J.P. Bergmann, S. Barnikol-Oettler,
Wire arc additive manufacturing of hot work tool steel with CMT process, J.
Mater. Process. Technol. 269 (January) (2019) 109–116.
[34] voestalpine BÖHLER Edelstahl GmbH & Co KG, ‘‘voestalpine BÖHLER Edelstahl
- Schweissen im Werkzeugbau,” 2010. [Online]. Available: https://www.
bohler-edelstahl.com/app/uploads/sites/92/2020/02/productdb/api/
bw140d_schweissen_im_werkzeugbau.pdf. [Accessed: 10-Jan-2021].
[35] DSI-Laser, ‘‘DataSheet.” [Online]. Available: https://www.dsi-laser.de/images/
LaserMold60.TD.pdf.
[36] J.H. Martin, B.D. Yahata, J.M. Hundley, J.A. Mayer, T.A. Schaedler, T.M. Pollock,
3D printing of high-strength aluminium alloys, Nature 549 (7672) (2017) 365–
369.
[37] D. Zhang et al., Metal Alloys for Fusion-Based Additive Manufacturing, Adv.
Eng. Mater. 20 (5) (2018) 1700952.
[38] A. Aversa, G. Marchese, A. Saboori, E. Bassini, D. Manfredi, S. Biamino, D. Ugues,
P. Fino, M. Lombardi, New Aluminum Alloys Specifically Designed for Laser
Powder Bed Fusion: A Review, Materials (Basel) 12 (7) (2019) 1007, https://
doi.org/10.3390/ma12071007.
[39] ‘‘voestalpine Böhler Welding - 3DPRINT SOLID WIRES.” [Online]. Available:
https://www.voestalpine.com/welding/content/download/9744/180767/file/
BW_3DPrint+Solid+Wires_EN_2018_GL_089_WEB.pdf. [Accessed: 10-Oct-
2021].
[40] Y. Wang, S. Konovalov, X. Chen, Y. Ivanov, S. Jayalakshmi, R.A. Singh, Research
on Cu-6. 6 % Al-3. 2 % Si Alloy by Dual Wire Arc Additive Manufacturing, J.
Mater. Eng. Perform. 30 (3) (2021) 1694–1702.
[41] J. Gu, J. Bai, J. Ding, S. Williams, L. Wang, K. Liu, Design and cracking
susceptibility of additively manufactured Al-Cu-Mg alloys with tandem wires
and pulsed arc, J. Mater. Process. Tech. 262 (February) (2018) 210–220.
[42] Z. Yang, Q. Liu, Y. Wang, Z. Ma, Y. Liu, Fabrication of multi-element alloys by
twin wire arc additive manufacturing combined with in-situ alloying, Mater.
Res. Lett. 8 (12) (2020) 477–482.
[43] Y. Feng, B. Zhan, J. He, K. Wang, The double-wire feed and plasma arc additive
manufacturing process for deposition in Cr-Ni stainless steel, J. Mater. Proc.
Technology 259 (2018) 206–215.
[44] M. Bermingham, D. Stjohn, M. Easton, L. Yuan, M. Dargusch, Revealing the
Mechanisms of Grain Nucleation and Formation During Additive
Manufacturing, JOM 72 (3) (2020) 1065–1073.
[45] M.J. Bermingham, D.H. StJohn, J. Krynen, S. Tedman-Jones, M.S. Dargusch, Acta
Materialia Promoting the columnar to equiaxed transition and grain re fi
nement of titanium alloys during additive manufacturing, Acta Mater. 168
(2019) 261–274.
[46] D. Gonzales, S. Liu, M. Domack, and R. Hafley, ‘‘Using Powder Cored Tubular
Wire Technology to Enhance Electron Beam Freeform Fabricated Structures,”
in TMS 2016 145th Annual Meeting & Exhibition, Cham: Springer International
Publishing, 2016, pp. 183–189.
[47] Z. Lin, C. Goulas, W. Ya, and M. J. M. Hermans, ‘‘Microstructure and mechanical
properties of medium carbon steel deposits obtained via wire and arc additive
manufacturing using metal-cored wire,” Metals (Basel)., vol. 9, no. 6, 2019.
[48] V. G. Burov, V. E. Rubtsov, and M. V Rashkovets, ‘‘Flux-cored wire for additive
manufacturing of parts with gradient structure,” in AIP Conference Proceedings
2167, 2019, vol. 020043, no. November, p. 020043.
[49] J. Wang et al., Materials Science & Engineering A Characterization of wire arc
additively manufactured titanium aluminide functionally graded material :
Microstructure, mechanical properties and oxidation behaviour, Mater. Sci.
Eng. A 734 (July) (2018) 110–119.
[50] Y. Ma, D. Cuiuri, N. Hoye, H. Li, Z. Pan, Characterization of In-Situ Alloyed and
Additively Manufactured Titanium Aluminides, Metall. Mater. Trans. B 45 (6)
(2014) 2299–2303.
[51] J. Wang, Z. Pan, D. Cuiuri, H. Li, Phase constituent control and correlated
properties of titanium aluminide intermetallic alloys through dual-wire arc
additive manufacturing, Mater. Lett. 242 (2019) 111–114.
[52] J. Wang et al., ‘‘In-situ dual wire arc additive manufacturing of NiTi-coating
on Ti6Al4V alloys: Microstructure characterization and mechanical
properties,” Surf. Coatings Technol., vol. 386, no. November 2019, p.
125439, Mar. 2020.
[53] C. Shen, Z. Pan, Y. Ma, D. Cuiuri, H. Li, Fabrication of iron-rich Fe – Al
intermetallics using the wire-arc additive manufacturing process, Addit.
Manuf. 7 (2015) 20–26.
[54] G. Langelandsvik, M. Grandcolas, K.G. Skorpen, T. Furu, O.M. Akselsen, H.J.
Roven, Development of Al-TiC Wire Feedstock for Additive Manufacturing by
Metal Screw Extrusion, Metals (Basel) 10 (11) (2020) 1485.
[55] D. Oropeza et al., Welding and additive manufacturing with nanoparticle-
enhanced aluminum 7075 wire, J. Alloys Compd. 834 (2020) 154987.
[56] G. Langelandsvik, O. Ragnvaldsen, J.E. Flåm, O.M. Akselsen, H.J. Roven, Wire
and Arc Additive Manufacturing with TiC-Nanoparticle Reinforced AA5183
Alloy, MATEC Web Conf. 326 (2020) 07002.
[57] J. R. Kennedy, A. E. Davis, A. E. Caballero, S. Williams, E. J. Pickering, and P. B.
Prangnell, ‘‘The potential for grain refinement of Wire-Arc Additive
Manufactured (WAAM) Ti-6Al-4V by ZrN and TiN inoculation,” Addit. Manuf.,
vol. 40, no. November 2020, p. 101928, 2021.
[58] S. Mereddy, M.J. Bermingham, D.H. Stjohn, M.S. Dargusch, Grain re fi nement of
wire arc additively manufactured titanium by the addition of silicon, J. Alloys
Compd. 695 (2017) 2097–2103.
[59] M.J. Bermingham, D. Kent, H. Zhan, D.H. StJohn, M.S. Dargusch, ScienceDirect
Controlling the microstructure and properties of wire arc additive
manufactured Ti – 6Al – 4V with trace boron additions, Acta Mater. 91
(2015) 289–303.
[60] M.J. Bermingham, S.D. Mcdonald, M.S. Dargusch, Materials Science &
Engineering A E ff ect of trace lanthanum hexaboride and boron
additions on microstructure, tensile properties and anisotropy of Ti-6Al-
4V produced by additive manufacturing, Mater. Sci. Eng. A 719
(February) (2018) 1–11.
[61] T.A. Rodrigues et al., In-situ strengthening of a high strength low alloy steel
during Wire and Arc Additive Manufacturing (WAAM), Addit. Manuf. 34
(January) (2020) 101200.
[62] S. Zhang, Q. Wang, R. Yang, C. Dong, Composition equivalents of stainless steels
understood via gamma stabilizing efficiency, Sci. Rep. 11 (1) (2021) 5423.
[63] M.V. Upadhyay, M.B.H. Slama, S. Gaudez, N. Mohanan, L. Yedra, S. Hallais, E.
Héripré, A. Tanguy, Non-oxide precipitates in additively manufactured
austenitic stainless steel, Sci. Rep. 11 (1) (2021), https://doi.org/10.1038/
s41598-021-89873-2.
[64] T. Chen, S. Xue, B. Wang, P. Zhai, and W. Long, ‘‘Study on short-circuiting
GMAW pool behavior and microstructure of the weld with different waveform
control methods,” Metals (Basel)., vol. 9, no. 12, 2019.
[65] X. Lou, P. L. Andresen, and R. B. Rebak, ‘‘Oxide inclusions in laser additive
manufactured stainless steel and their effects on impact toughness and stress
corrosion cracking behavior,” J. Nucl. Mater., vol. 499, no. November, pp. 182–
190, 2018.
[66] P. Deng, M. Karadge, R.B. Rebak, V.K. Gupta, B.C. Prorok, X. Lou, The origin and
formation of oxygen inclusions in austenitic stainless steels manufactured by
laser powder bed fusion, Addit. Manuf. 35 (May) (2020) 101334.
[67] M. Song, X. Lin, F. Liu, H. Yang, W. Huang, Effect of environmental oxygen
content on the oxide inclusion in laser solid formed AISI 420 stainless steel,
Mater. Des. 90 (2016) 459–467.
[68] X. Yang, F. Tang, X. Hao, Z. Li, Oxide Evolution During the Solidification of 316L
Stainless Steel from Additive Manufacturing Powders with Different Oxygen
Contents, Metall. Mater. Trans. B Process Metall. Mater. Process. Sci. 52 (4)
(2021) 2253–2262.
[69] A.L.V. Da Costa, E. Silva, The effects of non-metallic inclusions on properties
relevant to the performance of steel in structural and mechanical applications,
J. Mater. Res. Technol. 8 (2) (2019) 2408–2422.
[70] K. Weman, Welding processes handbook, Elsevier, 2011.
[71] ‘‘Voestalpine Böhler Welding - Cored Wires.” [Online]. Available: https://
www.voestalpine.com/welding/Brands/Boehler-Welding/Cored-wires.
[Accessed: 24-Apr-2020].
[72] ‘‘Lincoln Electric - METAL CORED OUTERSHIELD WIRES.” [Online]. Available:
https://www.lincolnelectric.com/en-gb/support/process-and-theory/Pages/
metal-cored-wires.aspx. [Accessed: 24-Apr-2020].
[73] ‘‘ESAB - Advantages and Disadvantages of Metal Cored Wires.” [Online].
Available: https://www.esabna.com/us/en/education/blog/advantages-and-
disadvantages-of-metal-cored-wires.cfm. [Accessed: 24-Apr-2020].
[74] Y. Cao, S. Zhu, X. Liang, W. Wang, Overlapping model of beads and curve fitting
of bead section for rapid manufacturing by robotic MAG welding process,
Robot. Comput. Integr. Manuf. 27 (3) (2011) 641–645.
[75] F. Pixner, F. Warchomicka, P. Peter, A. Steuwer, M.H. Colliander, R. Pederson, N.
Enzinger, Wire-Based Additive Manufacturing of Ti-6Al-4V Using Electron
Beam Technique, Materials (Basel) 13 (15) (2020) 3310, https://doi.org/
10.3390/ma13153310.
[76] J. Xiong, G. Zhang, H. Gao, L. Wu, Modeling of bead section profile and
overlapping beads with experimental validation for robotic GMAW-based
rapid manufacturing, Robot. Comput. Integr. Manuf. 29 (2) (2013) 417–
423.
[77] D.G. Brandon, The structure of high-angle grain boundaries, Acta Metall. 14
(11) (1966) 1479–1484.
F. Pixner, R. Buzolin, A. Zelic
´et al. Materials & Design 215 (2022) 110453
17
