Article

Additive Manufacturing of Tungsten Carbide (WC)-Based

Cemented Carbides and Niobium Carbide (NbC)-Based Cermets
with High Binder Content via Laser Powder Bed Fusion
Fabio Miranda 1,2, * , Marcelo Otavio dos Santos 3,4 , Rodrigo Condotta 3 , Nathalia Marina Gonçalves Pereira 3 ,
Daniel Rodrigues 2 , Suzilene Real Janasi 2 , Fernando dos Santos Ortega 5 , Marcello Vertamatti Mergulhão 6 ,
Rodrigo Santiago Coelho 6 , René Ramos de Oliveira 7 , Luis Gallego Martinez 7 and Gilmar Ferreira Batalha 1

1 Polytechnic School, University of São Paulo, EPUSP–PMR, São Paulo 05508-220, São Paulo, Brazil;
gfbatalh@usp.br
2 BRATS Sintered Filters and Special Powder Composites, Cajamar 07750-000, São Paulo, Brazil;
daniel@brats.com.br (D.R.); srjanasi@yahoo.com.br (S.R.J.)
3 Chemical Engineering Department, University Center FEI, São Bernardo do Campo 09850-901, São Paulo,
Brazil; marcelo.santos@maua.br (M.O.d.S.); rcondotta@fei.edu.br (R.C.); unienpereira@fei.edu.br (N.M.G.P.)
4 Mauá Institute of Technology, University Centre, São Caetano do Sul 09580-900, São Paulo, Brazil
5 Research and Development Institute, University of Vale do Paraíba, São José dos Campos 12244-000,
São Paulo, Brazil; ortega@univap.br
6 SENAI CIMATEC—Institute of Innovation for Forming & Joining of Materials, Salvador 41650-010, Bahia,
Brazil; marcello.mergulhao@fieb.org.br (M.V.M.); rodrigo.coelho@fieb.org.br (R.S.C.)
7 IPEN, Institute of Energy and Nuclear Research, São Paulo 05508-000, São Paulo, Brazil;
reneoliveira.ipen@gmail.com (R.R.d.O.); lgallego@ipen.br (L.G.M.)
* Correspondence: fabio.miranda@usp.br

Abstract: The additive manufacturing technique performed via laser powder bed fusion has matured
Citation: Miranda, F.; dos Santos,
as a technology for manufacturing cemented carbide parts. The parts are built by additive consolida-
M.O.; Condotta, R.; Pereira, N.M.G.;
tion of thin layers of a WC and Co mixture using a laser, depending on the power and scanning speed,
Rodrigues, D.; Janasi, S.R.; Ortega,
making it possible to create small, complex parts with different geometries. NbC-based cermets,
F.d.S.; Mergulhão, M.V.; Coelho, R.S.;
de Oliveira, R.R.; et al. Additive
as the main phase, can replace WC-based cemented carbides for some applications. Issues related
Manufacturing of Tungsten Carbide to the high costs and dependence on imports have made WC and Co powders emerge as critical
(WC)-Based Cemented Carbides and raw materials. Furthermore, avoiding manufacturing workers’ health problems and occupational
Niobium Carbide (NbC)-Based diseases is a positive advantage of replacing WC with NbC and alternative binder phases. This work
Cermets with High Binder Content used WC and NbC as the main carbides and three binders: 100% Ni, 100% Co, and 50Ni/50Co wt.%.
via Laser Powder Bed Fusion. Metals For the flowability and spreadability of the powders of WC- and NbC-based alloy mixtures in the
2024, 14, 1333. https://doi.org/ powder bed with high cohesiveness, it was necessary to build a vibrating container with a pneumatic
10.3390/met14121333 turbine ranging from 460 to 520 Hz. Concurrently, compaction was promoted by a compacting
Academic Editors: Costanzo Bellini system. The thin deposition layers of the mixtures were applied uniformly and were well distributed
and Golden Kumar in the powder bed to minimize the defects and cracks during the direct sintering of the samples.
The parameters of the L-PBF process varied, with laser scanning speeds from 25 to 125 mm.s–1 and
Received: 10 October 2024
laser power from 50 to 125 W. Microstructural aspects and the properties obtained are presented and
Revised: 15 November 2024
discussed, seeking to establish the relationships between the L-PBF process variables and compare
Accepted: 18 November 2024
Published: 25 November 2024
them with the liquid phase sintering technique.

Keywords: L-PBF; WC-based cemented carbides; NbC-based cermets; alternative binder phases;
microstructural characterization; properties
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
1. Introduction
conditions of the Creative Commons
Attribution (CC BY) license (https://
Cemented carbides are considered metal–ceramic materials and are processed by the
creativecommons.org/licenses/by/
powder metallurgy route, containing traditional elements such as tungsten carbides (WCs)
4.0/). and cobalt (Co). Alternative binder phases such as nickel (Ni) or combined (Co, Ni) are

Metals 2024, 14, 1333. https://doi.org/10.3390/met14121333 https://www.mdpi.com/journal/metals

Metals 2024, 14, 1333 2 of 33

used [1]. The most commonly used cemented carbide alloy in the metalworking sector is
based on WC-Co [2], mainly for machining, mining, cold-forming, and mechanical-forming
processes, among other applications [3]. There are considerable efforts to develop new
classes of cemented carbides, or those with alternative compositions, to improve these
products’ microstructure, mechanical properties, and performance. This quest to replace
the WC-Co alloy is related to patent restrictions and the scarcity of Co ore, which have
led Co and W ores to reach increasingly higher prices since the early 21st century [4].
Due to its worldwide consumption for battery manufacturing, Co is in short supply and,
consequently, the commercial price is higher. Another important point is that Co in dust
form, when released in the workplace, results in occupational diseases. Inhalable Co is
classified as A3 (confirmed animal carcinogen with unknown relevance to humans) by the
American Conference on Governmental Industrial Hygiene (ACGIH) and Ni is classified as
A5 (not suspected as a human carcinogen). In 2016, the ACGIH made available specific data
for cemented carbides (WC-Co), classified as A2 (suspected human carcinogen). Replacing
Co with Ni, entirely or in part, with alternative binder phases (Co, Ni) is a great advantage,
minimizing the health risks to workers during production [5,6]. Co has been classified
as a critical raw material and global demand is expected to quadruple in the next four
decades due to the high demand forecast [7]. Replacing Co with Ni may be beneficial from
an occupational hygiene and cost perspective.
NbC-based cermets with alternative binder phases (NbC-Ni) have emerged to mini-
mize the consumption of WC and Co and are manufactured by the conventional powder
metallurgy route [8,9]. Their applications are focused on cutting tools for machining subject
to wear resistance at high temperatures. In addition, they have greater corrosion resistance
than Co [10]. They require high resistance to hot fracture and thermomechanical plastic
deformation to withstand severe conditions, such as cold and hot rolling, extrusion and
stamping [11]. In addition, they are resistant to wear and corrosion, combine at high
temperatures and compete with combined WC-based cemented carbides (WC-Co) with
small amounts of Cr3 C2 and (Ti, Mo) C [12].
There are numerous scientific studies on adding small amounts of NbC to conventional
WC-Co alloys to reinforce the binder phase [2] and as an inhibitor of abnormal WC grain
growth [13]. NbC-based cermets have not been explored as the main phase [8,13,14].
The total replacement of WC by NbC for hard metal alloys was a difficult task. The raw
material was very expensive and only available in small quantities due to there being
few manufacturers or import restrictions [15]. Currently, rolling rolls, guide rolls, and
rollers, among other products applied in the steel industry, are manufactured based on
NbC-Ni cermets. The application of this class is well defined. The manufacture of cermets
based on NbC-20Ni was a great challenge, requiring coarse NbC grains, hardness and
microstructures equivalent to WC-Co alloys [16]. Brazil has the world’s largest niobium
reserves, making the study of new NbC applications a strategic research area [2,4].
Laser powder bed fusion (L-PBF) is a promising alternative for producing refractory
alloys using computer-controlled layer-by-layer deposition [17], having as an advantage the
ability to manufacture W-based alloy components with complex and complicated shapes
and small dimensions, which cannot be easily manufactured by traditional methods, such
as casting, forging machining, and conventional powder metallurgy [18,19]. Several indus-
trial sectors have been using this technique to reduce production costs (eliminating inputs
and manufacturing steps), mainly to manufacture high-value-added and low-volume me-
chanical components. Many researchers have highlighted the difficulties in manufacturing
parts with tungsten-based alloys, as it involves high temperatures, rapid melting and
localized solidification, mainly resulting in cracks [20].
Compared to the centuries-old conventional powder metallurgy technique, L-PBF
has been identified as the best technology for manufacturing complex-shaped cemented
carbide products [21]. The great challenge is to obtain products with high density, with an
almost total absence of porosity and cracks, with a homogeneous, refined and as uniform
Metals 2024, 14, 1333 3 of 33

as possible microstructure, and with good mechanical properties; and, such as the hardness
and transverse rupture stress (TRS), equivalent to products from the conventional route [15].
This work aimed to compare conventional powder metallurgy (PM), via liquid phase
sintering (LPS), with additive manufacturing (AM), via L-PBF (laser powder bed fusion),
considering WC- and NbC-based cemented carbides with binders such as Co, Ni and
Co and Ni mixtures. The great challenge was to improve the flowability to obtain a bed
as homogeneous as possible, so, additionally, a vibrating container was used to spread
the powder, trying to obtain a uniform layer. Conventional samples were produced by
pressing and high-temperature vacuum sintering. For AM, the laser power versus speed
was investigated. Analyses of the porosities, microstructures, hardness, phase analysis, and
residual thermal stresses were evaluated to discuss the potential of L-PBF.

2. Materials and Methods

The conventional route cemented carbide samples were prepared (Table 1) with the
following compositions (Table 2) in wt.%: (I) WC-30Co, (II) WC-30Ni, (III) WC-30(Co, Ni),
(IV) NbC-30Co, (V) NbC-30Ni and (VI) NbC-30(Co, Ni). High-purity commercial Co, Ni,
WC, and NbC powders were used, with the following average particle sizes: 2.0, 4.8, 1.5,
and 1 µm. The refractory and metallic powders, as shown in Figure 1, were mixed and
homogenized using a conventional high shear rate mixer for 2 h, using isopropyl alcohol to
promote a homogeneous suspension, avoiding deformation of the metallic particles, Co
and Ni. After the mixing step, isopropyl alcohol was removed by conventional drying in
an oven at 200 ◦ C for 2 h. The raw materials NbC, WC, Ni, and Co, as shown in Table 1,
were evaluated for the particle size distribution using a particle size analyzer for non-
spherical particles, Microtrac S3550 (Microtrac Retsch GmbH, Haan, Germany), with laser
diffraction with 25 nm precision, which measures particle sizes from 0.02 to 2800 µm. To
analyze the actual densities of the powders, an automatic helium pycnometer, AccuPyc II
1350 (Micromeritics Instrument Corporation, Norcross, GA, USA), was used to measure
the volume and density of the solids, and the Gemini II N2 physisorption equipment
(Micromeritics Instrument Corporation, Norcross, GA, USA) was used to obtain the surface
area and average pore diameter. In Table 1, the theoretical densities of the mixtures are
considered, along with the content of 6.14 wt.% carbon, and compared with the actual
densities obtained in the pycnometer test, the real densities.

Table 1. Properties of refractory ceramic and metallic powders.

Theoretical Average Purity

Density Grain Size Carbon Bulk Density Origin
Powders
(g.cm−3 ) FSSS (µm) (wt.%) (wt.%) (g.cm−3 )
WC 15.63 1.5 99.53 5.8 4.0 Buffalo Tungsten Inc. (Depew, NY, USA)
7.65 <1.0 99.65 7.8 3.0 F&X Electro-Materials Limited
NbC (Jiangmen, China)
Nanjing Hanrui Cobalt Co.
Co 8.91 2.2 99.87 <0.03 1.2
(Nanjing, China)
CVMR Corporation (Toronto,
Ni 8.91 4.8 99.85 - 2.5
ON, Canada)
Information obtained from manufacturers’ quality certificates.

Table 2. Weight balance for the prepared alloys.

Theoretical Pycnometric
Alloy Weight Balance (%) Density Density
Powders
WC NbC Co Ni (g.cm−3 ) (g.cm−3 )
(I) WC-30Co 70.0 30.0 0.0 12.74 12.38
(II) WC-30Ni 70.0 0.0 30.0 12.75 12.71
(III) WC-30(Co, Ni) 70.0 15.0 15.0 12.75 12.54
(IV) NbC-30Co 3.0 67.0 30.0 0.0 8.03 7.86
(V) NbC-30Ni 3.0 67.0 0.0 30.0 8.05 7.99
(VI) NbC-30(Co, Ni) 3.0 67.0 15.0 15.0 8.04 7.93
 Electrolytic Co presents irregular shapes and aggregates with sizes smaller than 1 µm
(submicron). For the refractory powders, WC presents particles of higher apparent density
in an irregular shape (non-spherical), with an average grain size smaller than 10 µm. For
NbC, as shown in Table 1, it presents a lower apparent density when compared to the
Metals 2024, 14, 1333 density of WC. In Figure 1D–F, the particles are observed to be irregular and polygonal,
4 of 33
with an average size of less than 1 µm.

Co Ni (Co, Ni)

WC

NbC

Figure 1. BSE-SEM images of the mixtures: (A) WC-30 Co, (B) WC-30 Ni, (C) WC-30 (Co, Ni), (D)
Figure 1. BSE-SEM images of the mixtures: (A) WC-30 Co, (B) WC-30 Ni, (C) WC-30 (Co, Ni),
NbC-30 Co, (E) NbC-30 Ni and (F) NbC-30 (Co, Ni).
(D) NbC-30 Co, (E) NbC-30 Ni and (F) NbC-30 (Co, Ni).
Table 2. Weight balance for the prepared alloys.
Figure 1A–F show the BSE-SEM images of the WC- and NbC-based mixtures. In
Figure 1C,F, showing the identification of the Ni and Co powders, Theoretical the particle shapes
Pycnometric
are different.Powders Weight Balance (%)
Ni has a dendritic and porous shape with dimensions smallerDensity
than 10 µm.
Alloy Density
Electrolytic Co presents irregular WC shapes
NbC andCo aggregates
Ni with sizes
(g.cm ) −3 smaller than−3)1 µm
(g.cm
(submicron). For the refractory powders, WC presents particles of higher apparent density
(I) WC-30Co 70.0 30.0 0.0 12.74 12.38
in an irregular shape (non-spherical), with an average grain size smaller than 10 µm. For
(II) WC-30Ni 70.0 0.0 30.0 12.75 12.71
NbC, as shown in Table 1, it presents a lower apparent density when compared to the
(III) WC-30(Co, Ni) 70.0 15.0 15.0 12.75
density of WC. In Figure 1D–F, the particles are observed to be irregular and polygonal, 12.54
(IV)an average
with NbC-30Co 3.0 1 µm.
size of less than 67.0 30.0 0.0 8.03 7.86
(V)The flowability
NbC-30Ni of these mixed 3.0 67.0 0.0 30.0 8.05
powders in the vibrating pneumatic device (Figure 2) 7.99
is(VI) NbC-30(Co,
not an inherent Ni) which
property, 3.0 depends
67.0 not 15.0
only on 15.0
the physical8.04properties but 7.93also on
the stress state, the equipment used, and the handling method, since these are conventional
The flowability
metallurgy powders, of with
thesedifficulty
mixed powders
flowinginor the vibratingTo
cohesive. pneumatic
analyze the device (Figure 2)
granulometric
isdistribution
not an inherentof theproperty,
mixtures which depends
according not2,only
to Table on the
a sieve physical
shaker properties Scientific,
was used—LGI but also
onmodelthe LGI-VW-SSO
stress state, the equipment
(Laborglas, São used,
Paulo,and thewith
Brazil) handling method,sieves,
standardized since capacity
these areof
conventional
6 sieves at most: metallurgy
#100 (150powders,
µm), #200with difficulty
(75 µm), flowing
#325 (44 or cohesive.
µm), #400 (37 µm), #500 To analyze
(25 µm) the
and
#600 (15 µm) mesh
granulometric Tyler, inofdecreasing
distribution the mixtures order of openings.
according to Table 2, a sieve shaker was used—
A pneumatic
LGI Scientific, device
model with a vibrating
LGI-VW-SSO containerSão
(Laborglas, wasPaulo,
developed, coupled
Brazil) with with a metallic
standardized
roller, capacity
sieves, as shownofin6Figure
sieves 2A–D,
at most: whose
#100 rotation
(150 µm),speed
#200 was controlled
(75 µm), #325 (44 independently
µm), #400 (37 of
its displacement
µm), #500 (25 µm)speed. Compaction
and #600 (15 µm) meshwas performed with a polished
Tyler, in decreasing order ofstainless
openings.steel metal
roller inside the sintering chamber in the powder bed, as in Figure 2C [22]. Compacting
thin layers with a compactor roller is interesting for extra-fine powder composites for
applications focused on only AM in metals [23]. In this compaction operation with the
metal roller, the powder mixtures needed to flow through the lower part of the pneumatic
container and be concurrently spread and compacted. For this, it was necessary to select
and determine a sieve, or stainless-steel mesh, to control the deposition flow of powders
from the prepared mixtures. For this device, as shown in Figure 2C, it was impossible to
evaluate the rheological behavior of the powder composites to obtain the flow energy of
 [22]. Compacting thin layers with a compactor roller is interesting for extra-fine powder
composites for applications focused on only AM in metals [23]. In this compaction
operation with the metal roller, the powder mixtures needed to flow through the lower
part of the pneumatic container and be concurrently spread and compacted. For this, it
Metals 2024, 14, 1333 was necessary to select and determine a sieve, or stainless-steel mesh, to control the
5 of 33
deposition flow of powders from the prepared mixtures. For this device, as shown in
Figure 2C, it was impossible to evaluate the rheological behavior of the powder
composites
the samples to
to obtain thethe
quantify flow energy to
resistance of flow
the samples to quantify
for a better the resistance
understanding to flow for
of the behavior of
a better understanding of the behavior of the
the mixtures in the powder bed, as in Figure 2D. mixtures in the powder bed, as in Figure 2D.

(A) (B)

(C) (D)

Figure 2.
Figure 2. Pneumatic
Pneumatic vibrating
vibrating device
device coupled
coupled to
to aa metal
metal compactor
compactor roller:
roller: (A)
(A) front
front view;
view; (B)
(B) rear
rear
view; (C) sintering chamber and pneumatic vibrating device installed inside the sintering chamber;
view; (C) sintering chamber and pneumatic vibrating device installed inside the sintering chamber;
and (D) flowability and spreadability test of powder mixtures in the powder bed.
and (D) flowability and spreadability test of powder mixtures in the powder bed.

The layer
The layer onon the
the powder
powder bed bed was
was created
created by by aa vibrating
vibrating device
device (capacity
(capacity 1500
1500 g)g) with
with
a rubber scraper ruler and a compactor roller at the bottom of the device, as in Figure 2B.
a rubber scraper ruler and a compactor roller at the bottom of the device, as in Figure 2B.
Thespeed
The speedofofthe thedevice
deviceand and thethe rotation
rotation of the
of the roller
roller were were controlled.
controlled. The powder
The powder bed
bed area
areavast
was was(about
vast (about
630 cm630
2 ), cm
and2),the
and the compaction
compaction forceapplied
force was was applied
by a 2 by
kgfa(19.613
2 kgf (19.613 N)
N) roller.
roller.
The The powder
powder drained
drained from from the
the device device
through through
a #35 a #35 mesh/Tyler
mesh/Tyler sieve [22]. Thesieve [22]. The
translational
translational
movement ofmovement
the device of the device simultaneously
simultaneously spread andthe
spread and compacted compacted
powder,the powder,
generating
generating
layers in thelayers
powder inbed
the of powder
30 to 100bedµm.of As
30 depicted
to 100 µm. As depicted
in Figure in Figurechamber
2C, the sintering 2C, the
sintering chamber presented a controlled atmosphere of high-purity
presented a controlled atmosphere of high-purity argon (>99.99%), with a continuous flow argon (>99.99%),
withofa 0.3
rate L.min−1 flow
continuous rate operational
. Several of 0.3 L.min−1 . Several operational
parameters associatedparameters
with this new associated
compactionwith
this newcan
strategy compaction
be adjusted strategy can be adjusted
and controlled to modifyand thecontrolled
compactedtolayer modify the
in the compacted
dust bed; the
layer of
effect in the
theroller
dust compaction
bed; the effect forceofonthe
theroller compaction
properties force on the
of the powdered properties
composites of the
is one of
the main variables
powdered composites[23].is one of the main variables [23].
To
To better
better understand
understand the the rheological
rheological behavior
behavior of of these
these composites,
composites, an an FT-4
FT-4 Powder
Powder
Rheometer ® was used. Results were obtained using a patented
Rheometer ® patented measuring principle that
measuring principle that
evaluated
evaluatedthe theresistance
resistance to movement
to movement of a of
specially shaped
a specially twistedtwisted
shaped blade that goesthat
blade through
goes
a precise volume of sample along a prescribed path. The required torque and force were
recorded and converted into flow energy [24]. Repeated testing of a single sample (with
intermediate reconditioning) can be used to evaluate the physical stability of a powder
(described by the Stability Index—SI). Changes in the blade speed can be used to determine
how the powder responds to being made to flow at different rates (described by the
Flow Rate Index—FRI) [25]. The granules’ rheological properties can be obtained in the
Powder Rheometer® and are influenced by the processing parameters: compressibility,
bulk/tapped density, aeration, the primary energy of flow, and shear [24,26]. However, it
is difficult to observe this L-PBF process to obtain critical parameters that demonstrate a
direct influence on the performance of the compacted layer in the powder bed [22]. For
the L-PBF process parameters, as in Figure 3A, the tracks were created using laser powers
Metals 2024, 14, 1333 6 of 33

(PL ) as a function of the scanning speed (Vs ). According to the scanning system of the
OmniSinter-160 equipment (Omnitek Tecnologia, São Paulo, Brazil), PL = 457 W (100%),
the scanning speed was varied from 25–125 mm.s−1 , changing 50–125 W while maintaining
the Hs = 80 µm of the Iz = 30 µm, laser focus diameter 140 µm. The volumetric energy
density (VED) applied to the WC- and NbC-based cemented carbides can be calculated
by Equation (1) [27,28], where the VED (J.mm−3 ) is equivalent to the VED that relates the
power (PL ) with the scanning speed (vS ), track distance (hS ) and layer thickness (IZ ).
Metals 2024, 14, x FOR PEER REVIEW 7 of 34
PL
VED = (1)
vs · h s · I z

(A) (B)

Figure3.3.(A)
Figure (A)Direct
Directsintering
sinteringstrategy,
strategy,via
viaL-PBF,
L-PBF,for
forthe
thesamples
samplesgenerated
generatedby
bydifferent
differentlevels
levelsofof
factors(P(P
factors L
L×× V
VS
S);
); and
and (B)
(B) compaction
compaction system
system for
for mixtures
mixtures in
in the
the powder
powder bed.
bed.

InThe
theprocessing parameters
sintering chamber, affect the 0.3
a continuous L.min−1of
properties flowtheofgranulates:
high-puritypermeability,
argon was
maintained,
compressibility,and some
apparentresidual oxygen, estimated
and compacted densities, at aeration,
300 ppm,elemental
remained flow in theenergy
chamber. and
The
shearsintering parameters
[24]. However, theseviaparameters
L-PBF, as in Figure
must 3A, for the
be observed inVED values (Pchamber
the sintering L × vS ), were
of the
based
L-PBFonprocess
the research by Uhlmann
to obtain et parameters
the critical al. [28], whothat investigated
demonstrate the WC-17Co alloy (17%
a direct influence onwt.
the
Co), with laser of
performance power (PL ) up to 150
the compacted layerW,inscanning
the powder speed (vSand
bed ) uptheto 100 mm.s−1 , line
relationship spacing
between the
(hbulk/tapped
S ) below 100density.
µm and The IZ ofsintered
30 and 50samples
µm (thicknesses).
were hardTherefore, it was decided
Bakelite, ground for the
with diamond
WC- and NbC-based
resinoid alloys grain
grinding wheels, with 30%
size wt.
100 ofµm binder phase,
and grit to useW5
number the(size
following maximum
3.5–5 µm), with a
parameters: PL up of to 0.03
125 mm,
W, vSinup to 125and mm.s −1 and h = 80 µm and I = 30 µm. A
thickness removal 2 passes polished inSrotary polishers,Zwith diamond
pneumatic vibrating
lapping pastes, device
micron created
grades: 25, a15,powder
9, 6, 3 bed
and with
1 µm. a maximum
The apparent capacity of 1500
porosities wereg
(proof-of-concept model) with a rubber scraper ruler and a metal
evaluated with 100× magnification and microstructural analysis, 1500× magnification, by roller at the bottom,
asoptical
in Figure 3B. The speed
microscopy Olympus, of theGX device
seriesand the rotation
(Leco Corporation,of theSt. roller, as well
Joseph, MI,asUSA)the
rotation directions, were controlled. The area of the powder bed was enormous (620 cm 2 ),
Additionally, micrographs of the polished sections were obtained by scanning electron
and the weight
microscopy applied
using both SE to the
anddeposited
BSE detectorslayerunder
was ata least 2 kgf (19.6
high vacuum, N). The
Tescan modelpowder
Vega
flowed through the bottom of the device, through a stainless-steel
LMU (TESCAN Group, Kohoutovice, Czech Republic) and Quanta FEI 650 FEG SEM sieve, and different mesh
sizes could be chosen,
(Thermofisher Scientific,depending
Waltham,onMA, the USA).
agglomerates’ size distribution
For the hardness in the hardness
test, a Vickers powder
mixture [22].
tester, Microindentation System, HV-110 series (Mitutoyo, Jundiaí, Brazil), LED lighting
The and
system, translational
motorized movement of thesystem
load selection devicefor simultaneously
a 1 kgf (9.81 N) spread
load, and compacted
HV test, and systemthe
powder, creating layers in
A: color LCD touch panel were used.the powder bed. With the scraper ruler, the layer resulted in
100 µm, and then with the compacting roller, it resulted in a thin and
The analyses of the residual thermal stresses were performed using a Rigaku X-ray uniform layer of 30 µm.
Adiffractometer,
zigzag scanning strategy
model was used
Automate II,for a singleCorporation,
(Rigaku layer, with a Osaka,
rotationJapan)
between layers
with of
X-rays
67◦ . The scanning speed varied from 25 to 125 mm.s−1 , maintaining a hatch distance (Hs)
generated by a 2.291Å chromium tube at a voltage of 40 kV and current of 20 mA. The
equal to 80 µm. The laser focal diameter was 140 µm and the compacted mixture layer was
method used was sen2ψ. In the WC samples, the applied modulus was 523.7 GPa, and a
30 µm. For the VED shown in Figure 3A, samples measuring 1.2 × 5 × 5 mm3 (30 mm3 )
Poisson’s ratio of 0.22 at an incidence angle of 2θ of 136.63°. For the NbC samples, the
were produced, with an average of 40 layers deposited, establishing at least one critical
elastic modulus was 495 GPa, and a Poisson’s ratio of 0.21 at an incidence angle of 2θ of
function analytically or experimentally to validate the experiment and proof of concept.
125.32°. For the WC-based alloys, a hexagonal structure, space group P-6m2, a = 2.9005 Å
Several operational parameters can be associated with this new compaction strategy, such
and c = 2.8290 Å, the peak hkl:(012) was used; and for the NbC-based alloys, an FFC
as the clockwise or counterclockwise rotation direction of the compaction cylinder, or in
structure, space group Fm-3m, a = 4.47204 Å peak hkl:(222) was used.
XRD measurements were performed on a Rigaku ULTIMA-IV diffractometer
(Rigaku Corporation, Osaka, Japan), using Cu-Kα radiation under the following
conditions: θ-theta normal scan from 25° to 125°, 0.025° step, 2 s/step, 40 kV, and 30 mA.
All the analyses were performed by comparison with the PDF powder pattern databases
(Powder Diffraction File from the International Centre for Diffraction Data (ICDD)) or by
Metals 2024, 14, 1333 7 of 33

the latter case, with the cylinder locked and controlled to modify the compacted layer in
the powder bed (Iz); the effect of the compaction force of the roller on the properties of the
powder composites remains one of the main variables.
The processing parameters affect the properties of the granulates: permeability,
compressibility, apparent and compacted densities, aeration, elemental flow energy and
shear [24]. However, these parameters must be observed in the sintering chamber of the
L-PBF process to obtain the critical parameters that demonstrate a direct influence on the
performance of the compacted layer in the powder bed and the relationship between the
bulk/tapped density. The sintered samples were hard Bakelite, ground with diamond
resinoid grinding wheels, grain size 100 µm and grit number W5 (size 3.5–5 µm), with a
thickness removal of 0.03 mm, in 2 passes and polished in rotary polishers, with diamond
lapping pastes, micron grades: 25, 15, 9, 6, 3 and 1 µm. The apparent porosities were
evaluated with 100× magnification and microstructural analysis, 1500× magnification, by
optical microscopy Olympus, GX series (Leco Corporation, St. Joseph, MI, USA) Addition-
ally, micrographs of the polished sections were obtained by scanning electron microscopy
using both SE and BSE detectors under a high vacuum, Tescan model Vega LMU (TESCAN
Group, Kohoutovice, Czech Republic) and Quanta FEI 650 FEG SEM (Thermofisher Scien-
tific, Waltham, MA, USA). For the hardness test, a Vickers hardness tester, Microindentation
System, HV-110 series (Mitutoyo, Jundiaí, Brazil), LED lighting system, and motorized load
selection system for a 1 kgf (9.81 N) load, HV test, and system A: color LCD touch panel
were used.
The analyses of the residual thermal stresses were performed using a Rigaku X-
ray diffractometer, model Automate II, (Rigaku Corporation, Osaka, Japan) with X-rays
generated by a 2.291Å chromium tube at a voltage of 40 kV and current of 20 mA. The
method used was sen2ψ. In the WC samples, the applied modulus was 523.7 GPa, and
a Poisson’s ratio of 0.22 at an incidence angle of 2θ of 136.63◦ . For the NbC samples, the
elastic modulus was 495 GPa, and a Poisson’s ratio of 0.21 at an incidence angle of 2θ of
125.32◦ . For the WC-based alloys, a hexagonal structure, space group P-6m2, a = 2.9005 Å
and c = 2.8290 Å, the peak hkl:(012) was used; and for the NbC-based alloys, an FFC
structure, space group Fm-3m, a = 4.47204 Å peak hkl:(222) was used.
XRD measurements were performed on a Rigaku ULTIMA-IV diffractometer (Rigaku
Corporation, Osaka, Japan), using Cu-Kα radiation under the following conditions: θ-theta
normal scan from 25◦ to 125◦ , 0.025◦ step, 2 s/step, 40 kV, and 30 mA. All the analyses were
performed by comparison with the PDF powder pattern databases (Powder Diffraction
File from the International Centre for Diffraction Data (ICDD)) or by diffraction patterns
calculated from the crystal structure data from the Inorganic Crystal Structure Database
(ICSD), American Mineralogist Crystal Structure Database (AMCS) and Crystallographic
Open Database (COD) databases, using the Crystal Impact Match and Crystallographica
Search-Match software to identify the crystalline phases of the WC, NbC, Co, Ni and other
crystalline phases formed.

3. Results and Discussion

3.1. Characterization of Metallic Powders
In Figure 4, the Co and NbC powders present narrow particle size distributions,
smaller than 0.1 µm and with a higher percentage of finer particles. Cobalt has a bimodal
distribution, where most particles are smaller than 0.1 µm, and a small concentration
around 10 µm. This is unlike NbC, which has a normal distribution close to 0.1 µm. This is
due to the raw material manufacturing process. In contrast, the other metallic powders, WC
and Ni, have larger particles with a more open particle size distribution. The actual density
values (pycnometric) of the raw materials are presented in Table 3. They are compatible
with the values described in their respective technical data sheets of origin. Co powder has
the highest specific surface area compared to other powders.
 around 10 µm. This is unlike NbC, which has a normal distribution close to 0.1 µm. This
is due to the raw material manufacturing process. In contrast, the other metallic powders,
WC and Ni, have larger particles with a more open particle size distribution. The actual
density values (pycnometric) of the raw materials are presented in Table 3. They are
Metals 2024, 14, 1333 compatible with the values described in their respective technical data sheets of origin. Co
8 of 33
powder has the highest specific surface area compared to other powders.

(A) (B)

Figure4.4.(A)
Figure Frequency
(A) Frequency distribution andand
distribution (B) (B)
Cumulative distribution
Cumulative of primary
distribution particles
of primary (NbC,(NbC,
particles WC,
Co and Ni) present in the raw materials, measured by laser diffraction.
WC, Co and Ni) present in the raw materials, measured by laser diffraction.

Table 3. Characterization of the powders.

Table 3. Characterization of the powders.
Actual Average Total Pores per
Raw Surface Area Total Pores per Average Pore
Raw Material Actual Average
Density Surface Area Volume Average Pore
−3—3 (m2 .g−1 ) 3 .g−31 ) —1 Diameter (nm)
Material Density(g.cm
(g.cm )) (m2.g—1) Volume
(cm(cm .g ) Diameter (nm)
WCWC 15.494 ± 0.040
15.494 ± 0.040 0.624
0.624 0.006
0.006 35.770
35.770
NbC
NbC 7.6387.638 ± 0.045
± 0.045 2.586
2.586 0.017
0.017 26.495
26.495
CoCo 8.422
8.422 ± 0.068
± 0.068 4.667
4.667 0.015
0.015 13.126
13.126
Ni 8.952 ± 0.031 1.252 0.012 38.974
Ni 8.952 ± 0.031 1.252 0.012 38.974

To Toobtain
obtainthe
theideal
idealgrid
gridmesh
mesh(Tyler
(Tylerequivalent)
equivalent)for forthethepneumatic
pneumaticvibrating
vibratingdevice,
device,
the
theparticle
particlesize
sizedistribution
distributionofofthethepowder
powdermixture
mixture(WC- (WC-ororNbC-based
NbC-basedcomposites)
composites)isis
generally
generallyconsidered.
considered. However,
However, different
different characterization
characterization methodsmethods (sieving
(sievingand andlaser
laser
diffraction) may provide different results, mainly to obtain the average
diffraction) may provide different results, mainly to obtain the average Sauter diameter Sauter diameter
(D
(D ) )ofof
psPS thethe
mixture, according
mixture, accordingto Equation (2) [22],
to Equation (2)which
[22], iswhich
considered a valuablea property
is considered valuable
for
property for comparing and controlling the different batches; in this case, forthe
comparing and controlling the different batches; in this case, for selecting stainless
selecting the
steel sievesteel
stainless mesh. Here,
sieve mesh.“DHere,
ps ” means
“Dps”the meanthe
means Sauter
meandiameter (µm), “x(µm),
Sauter diameter i ” is the
“xi”mass
is the
fraction retainedretained
mass fraction in each in
sieve
each (dimensionless); and “Di ”and
sieve (dimensionless); is the diameter
“Di” is the of the particles
diameter of the
retained
particlesinretained
each sieve (µm).sieve (µm).
in each
1
D ps = n (2)
1xi
𝐷 ∑ =i=1 Di 𝑥 (2)
∑
The sieving method is the most widely used in industries 𝐷 that manufacture cemented
carbide products. The particle size distribution of WC- and NbC-based alloy agglomerates
The sieving method is the most widely used in industries that manufacture cemented
is an important process variable that, associated with the selected sieve mesh, directly
carbide products. The particle size distribution of WC- and NbC-based alloy agglomerates
affects the flowability and powder deposition rate. It depends on the particle size of the
is an important process variable that, associated with the selected sieve mesh, directly
raw materials but also on the cohesion of the particles that form the granules, which in
turn results from several factors, such as the density, specific surface area, roughness and
moisture. In Figure 5A,B, the Sauter diameter Di is calculated from Equation (2), resulting
from the knowledge of the particle size distributions of the WC- and NbC-based samples,
where the average particle size of the mixtures was obtained. In Figure 5A, the test for 200 g,
the indicated mesh was the Tyler equivalent #140 (100 µm); in practice, with the pneumatic
device, there was a flowability of the powders in the container. However, for Figure 5B, the
test with 1000 g, the Tyler equivalent meshes that flowed the mixtures in the studies were
the #35, #50, and #60 meshes. When comparing Figure 5A,B, it is seen that the frequency
and cumulative distributions of the agglomerated particles of the dry powder mixtures
behaved differently, interfering in the aggregation of the solid particles; consequently,
the Sauter Dps values were other results in the mass transfer. The Sauter Dps is used in
studies related to interfacial phenomena. The evaluation of the particle size distribution
 Figure 5B, the test with 1000 g, the Tyler equivalent meshes that flowed the mixtures in
the studies were the #35, #50, and #60 meshes. When comparing Figure 5A,B, it is seen
that the frequency and cumulative distributions of the agglomerated particles of the dry
powder mixtures behaved differently, interfering in the aggregation of the solid particles;
Metals 2024, 14, 1333 consequently, the Sauter Dps values were other results in the mass transfer. The Sauter 9 of 33
Dps is used in studies related to interfacial phenomena. The evaluation of the particle size
distribution of powdered composites is essential in all the processes proposed for
producing
of powderedsintered
compositesparts, as it inisall the
is essential intrinsically linked tofor the
processes proposed density
producing and
sintered
linearity/volumetric shrinkage.
parts, as it is intrinsically linked to the density and linearity/volumetric shrinkage.

(A) 200 g (B) 1000 g

Figure
Figure 5.
5. Frequency
Frequency distribution
distribution and
and cumulative
cumulative distribution
distributionof
of the
the agglomerated
agglomerated particles
particles of
of the
the
dry powder mixtures: (A) 200 g and (B) 1000 g, which correspond to the material to be deposited by
dry powder mixtures: (A) 200 g and (B) 1000 g, which correspond to the material to be deposited by
the device.
the device.

Using
Using the
the FT-4
FT-4 Powder
Powder Rheometer
Rheometer®equipment,
®
equipment,the
thebasic
basicflow
flowenergy
energy(BFE),
(BFE),stability,
stability,
and variable flow (VRF) tests were carried out, measuring the resistance
and variable flow (VRF) tests were carried out, measuring the resistance to to thethe
passage of
passage
aofblade through
a blade throughthethe
particle
particlebed.
bed.The
Thesample
samplestability
stabilitywas
wasevaluated
evaluated under
under constant
constant
rotation of 100 mm.s —1
− 1 (blade tip speed of 48 mm diameter) as a function of
rotation of 100 mm.s (blade tip speed of 48 mm diameter) as a function of time (number time (number
of tests from 1 to 7); the torque and force were recorded and converted into flow energy. The
movement of the blade through the bed of particles can modify the packing, the interactions
between the particles and alloy mixtures, or even cause surface modifications: abrasion,
breaks, segregation, etc. The basic flow energy (BFE) is the value of the total energy that
the blade expends, measured in the seventh test after the samples have stabilized. It can
be seen in Figure 6A that in the seventh test, a level of stability is observed, ensuring the
reliability of the measurement of this parameter.
In general terms, in Figure 6, the energy required to move the bed of refractory WC
particles is more significant in the samples containing predominantly NbC; this occurs
because the density of WC (15.494 g.cm−3 ) is higher than that of NbC (7.638 g.cm−3 ), and
more energy is needed to move denser particles. Additionally, for the samples with the
same majority component (WC and NbC), Table 4 shows that the samples containing only
Ni are those that require the most significant amount of energy to be moved by the blade;
that is, the mechanical transport of both materials containing only Ni in their composition
is more complicated. The incorporation of Co in the samples containing only Ni reduces
 more energy is needed to move denser particles. Additionally, for the samples with the
same majority component (WC and NbC), Table 4 shows that the samples containing only
Ni are those that require the most significant amount of energy to be moved by the blade;
that is, the mechanical transport of both materials containing only Ni in their composition
Metals 2024, 14, 1333 10 of 33
is more complicated. The incorporation of Co in the samples containing only Ni reduces
the basic flow energy (BFE). Using only Co proves to be more effective in reducing the
flow energy of the samples. Therefore, Co appears to act as an attenuator of interparticle
the basic flowallowing
interactions, energy more
(BFE).effortless
Using only Co proves
movement to be more
of powder effective in
bed particles. Toreducing
eliminatethe the
flow energy of the samples. Therefore, Co appears to act as an attenuator
effect of the density of each type of material on the energy consumed by the blade when of interparticle
interactions,
crossing the allowing
powder bed,morewe effortless movement
can divide the BFEofbypowder bedofparticles.
the mass materialTo eliminate
that the
constitutes
effect of the density of each type of material on the energy consumed
the powder bed, resulting in the flow energy per gram of material moved: the BFE/mass by the blade when
crossing the powder
of the material. bed,
Thus, thewe can divide
effort per unittheofBFE by required
mass the mass toof material
move thethatWCconstitutes
and NbC
the
samples with only Ni in their composition appears practically identical.BFE/mass
powder bed, resulting in the flow energy per gram of material moved: the Adding Co of
the material.
together withThus,
Ni isthe
noteffort per unitfor
significant of the
massWC required
samples. to move the WC
However, theand
energyNbCreduction
samples
with only Ni in their composition appears practically identical. Adding
observed when adding only Co to the samples is evident, especially for the NbC samples,Co together with
Ni is not
in Table 4.significant for the WC samples. However, the energy reduction observed when
adding only Co to the samples is evident, especially for the NbC samples, in Table 4.

(A) (B)

Figure6.6. (A)
Figure (A) Test
Testof
ofthe
the basic
basic flow
flow energy
energy(test
(testnumber
number11to
to 7)
7) and
and variable
variable flow
flow energy
energy (test
(test number
number
88to
to11)
11)and
and(B)(B)aeration
aerationtest
testof
ofthe
theWC-
WC-and
andNbC-based
NbC-basedmixtures.
mixtures.

4. BFE
Table An and VFR test
interesting test results: relationship
is the aeration test,between thein
as shown bulk and compressed
Figure 6B, which isdensity (BD) and
essentially the
the applied pressure.
basic flow energy (BFE) test with air injection at the base of the sample bed. It measures
the reduction in interactions between particles−(reduction in the BFE) as a function of the
BFE SE Split BFE.g 1 BD CPS BDcomp .
Composition amount SIof air injected into the system. As part of the
(mJ) −
(mJ.g )1 Mass (g) −
(mJ.g ) 1 (g.cm 3 ) initially
volume− occupied(g.cm
15 kPa by a−solid
3)
is now occupied by air, the blade exerts less effort to go through the bed of particles. Thus,
WC-30Co 4007.11 1.11 6.87 634.31 6.317 3.33 9.13 3.66
the air separates the particles to allow them to pass. When the energy stabilizes, the lowest
WC-30Ni 3606.39 1.08 7.96 564.88 6.384 4.11 6.29 4.39
WC-30(Co, Ni) 2623.04 state of interaction6.61
1.18 between the particles is
507.74 reached, and
5.166 3.60the bed reaches
9,74 its fluidization
3.99
NbC-30Co 2424.98 condition.
1.04 Furthermore,
7.80 if the377.95
particles are 6.416
not cohesive,2.45they separate easily, and 2.82
13.08 the layer
NbC-30Ni 2001.91 1.03 8.31 332.56 6.020 2.47 14.32 2.88
NbC-30(Co, Ni) 1486.77 1.19 5.70 375.78 3.956 2.21 14.42 2.59

An interesting test is the aeration test, as shown in Figure 6B, which is essentially the
basic flow energy (BFE) test with air injection at the base of the sample bed. It measures
the reduction in interactions between particles (reduction in the BFE) as a function of the
amount of air injected into the system. As part of the volume initially occupied by a solid is
now occupied by air, the blade exerts less effort to go through the bed of particles. Thus,
the air separates the particles to allow them to pass. When the energy stabilizes, the lowest
state of interaction between the particles is reached, and the bed reaches its fluidization
condition. Furthermore, if the particles are not cohesive, they separate easily, and the layer
can permeate the bed without significant difficulties, reaching levels at shallow values. For
particles that still present specific interactions (cohesiveness), the same level is reached at
higher air flow rates or higher levels for the same flow rates. Table 4 presents the results
of the basic flow energy (BFE) for all the powder samples, as well as the results of the
compressibility tests at 15 kPa. The complete evaluation of the compressibility behavior of
all the samples, when subjected to normal stresses ranging from 0.5 to 15 kPa, is exhibited
 based alloys, respectively, indicating once again that the NbC-30Ni sample exhibits the
most cohesive and difficult-to-flow behavior. Adding Co to the sample enhances its flow
properties, with both samples containing a mixture of Co and Ni exhibiting similar
behavior. The WC-30Ni sample demonstrates the most favorable flow characteristics
Metals 2024, 14, 1333 among the samples containing only Co. Achieving the reliable flow of cohesive powders
11 of 33
presents a significant challenge in various particle-processing operations, including silo
and hopper unloading, feeding, dosing, etc., whereby the shear strength is evaluated
under
in Figurespecific consolidation
7B. Table stress
5 presents the or ofpacking
results conditions.
the shear Typically,
cell experiments this evaluation
conducted at 3, 6, 9,
involves moderate to high stresses
and 15 kPa of consolidation stress. and meager shear strain rates [30].

(A) (B)

Figure7.7.(A)
Figure (A)Bulk
Bulkdensity
density(BD
(BDcomp
comp);
); and
and (B)
(B) compressibility tests of
compressibility tests of the
the WC-
WC- and
andNbC-based
NbC-basedalloys.
alloys.
Table 5. Shear test results: cohesion and angle of internal friction (AIF) under different stress conditions.
In additive manufacturing via L-PBF, this dry powder deposition technique is
promising,
3 kPa with the advantage
6 kPa of being applicable
9 kPa to small quantities 15 kPaof powder
Composition composites based on WC and NbC in the vibrating device. However, further investigation
Cohesion AIF Cohesion AIF Cohesion AIF Cohesion AIF
(kPa) into its(behavior
◦) in(kPa)
the dynamic(◦regime
) is necessary,
(kPa) with(◦ )more significant
(kPa) measurement
(◦ )
repetition, to better differentiate between powders with similar rheological properties in
WC-30Co 0.73 32.4 1.16 32.9 1.45 33.1 1.66 34.4
WC-30Ni 0.62 all the31.8
packing states. 1.06 This type31.9
of dry granulation
1.35 is not suitable or1.44
32.9 the most commonly
35.8
WC-30(Co, Ni) 0.96 used for32.5processing1.39 techniques 33.7
via L-PBF; it1.82 is more often 33.7used for conventional
2.08 sintering
36.3
NbC-30Co 0.69 techniques
29.4 via LPS. 0.93 No information
31.7 in 1.19
the literature 31.0indicates1.23- which formulation
32.4
NbC-30Ni 1.31 properties
35.7 are suitable1.90 or unsuitable
40.4 for this processing
- -
method. The2.96 42.2parts
qualities of the
NbC-30(Co, Ni) 1.09 manufactured
32.7 1.56 33.4 2.07 33.7
by additive manufacturing (AM) are influenced by the characteristics 2.04 37.1 of
the powders, which include the shape, size distribution, surface morphology,
composition,
Figure 7 and flowabilitythat
demonstrates or flowability of the particulates
the NbC samples are more [31].compressible (percentage
The typical particle sizes for sintering via
change in volume after compression, %) than the WC samples, L-PBF are in the range ofof10the
regardless to addition
60 µm. The of
behavior
other of a composite/ceramic
compounds. Thus, it is possible powder
to stateisthat
fundamental
the WC samples to its have
performance in the
lower porosity,
manufacturing
producing a bedprocess, such as
of particles in compaction,
with fewer voids,inwith which thethe filling being
sample operations of molds
WC-30Ni. or
This
die powder spreading depend on the flowability. Various methods
means that the granulation of the NbC samples must be further worked to reach the same or tests can help with
the evaluation
state of compactionof dynamic
as the flow and flowability.
WC samples. The smallStill,oscillations
specifying the best flow
between condition
points 1 and 3is
not always clear, given that different sizes and shapes correlate
for samples of alloys WC-30Ni, WC-30(Co, Ni), WC-30Co, and NbC-30Co, as in Figure 6,[32]. In the AM technique,
indicate an initial accommodation of the samples in the bed, which stabilize from the third
test. Observing the cohesion values of the samples, Figure 8 shows that the WC-30Ni
sample is the most cohesive, corroborating the aeration test results and the primary flow
energy per unit mass (BFE.g−1 ). The WC-30Ni and NbC-30Co samples had the lowest
cohesion values, corroborating the aeration and basic flow energy per unit mass (BFE/g)
tests. Shear tests also allow for the so-called flow function of materials, which classifies
their flow profile. The results indicate complete fluidization of all the samples at an air
velocity of 15 mm.s−1 (for a diameter of 50 mm) but a significant reduction in the BFE at low
flow rates (2.5 to 5.0 mm.s−1 ). Significantly reduced cohesion is observed for the samples
containing only Co, whereas the most excellent cohesion is observed for the NbC-30Ni
samples, as in Figure 8. The samples containing Ni and Co appear to present two levels of
energy stabilization. This may indicate two moments of fluidization, the first for smaller
particles (fines) and the second for the remaining particle size (analyze the particle size
distribution and check if it is bimodal). An important observation is the behavior of the
samples containing Ni, which shows that the mixture is more cohesive than the mixture
containing NbC; however, it presents the lowest cohesion when mixed with WC. This
Metals 2024, 14, 1333 12 of 33

indicates that interactions between different elements are essential for the flowability of
the studied samples. The flowability of powders is not an inherent property—it depends
not only on the physical properties (shape, particle size, moisture content, etc.) but also on
Metals 2024, 14, x FOR PEER REVIEW 13 of 34
the stress state, on the equipment used, and on the handling method. The powder flow
in individual additive technology methods is a complex area of study [29]. Figure 8A,B
show the shear stress test and flow profile classification of the WC- and NbC-based alloys,
the experimental
respectively, characterization
indicating once againisthat
minimal when measuring
the NbC-30Ni composite
sample exhibits theproperties, such
most cohesive
as the
and average diameter,
difficult-to-flow particle
behavior. size distribution,
Adding and packing
Co to the sample enhances density.
its flowItproperties,
is inadequate
withto
solvesamples
both a specific configuration
containing for of
a mixture powder
Co andbed depositionsimilar
Ni exhibiting [31,32].behavior.
Powdered Thecomposites
WC-30Ni
used for
sample AM are spread
demonstrates ontofavorable
the most a compactedflow layer for directamong
characteristics sintering, and thiscontaining
the samples process is
repeated
only from layerthe
Co. Achieving to deposited layer
reliable flow to form the
of cohesive product.
powders The measurement
presents a significanttechniques
challenge
in various
used particle-processing
to evaluate operations,can
the powder flowability including silo into
be divided and separate
hopper unloading,
categories feeding,
[3], such
dosing,
as the flow rate, density rates, bulk and beat density, angle of repose, and shear stress
etc., whereby the shear strength is evaluated under specific consolidation stress,
or packing
among conditions.
other Typically, this evaluation involves moderate to high stresses and
methods [1–3].
meager shear strain rates [30].

(A) (B)

Figure8.8.Shear
Figure stress
Shear testtest
stress for the
for quantitative measurement
the quantitative of cohesion;
measurement (A) 6 kPa
of cohesion; and
(A) 6 (B)
kPaclassifica-
and (B)
tion of the flow profile of WC- and NbC-based alloys.
classification of the flow profile of WC- and NbC-based alloys.

In additive
The manufacturing
difference between thesevia L-PBF,
methods thiscandrybe
powder deposition
significant, as thetechnique is promis-
test method affects
ing, with the advantage of being applicable to small quantities of powder
the stress state and velocity of the particulates, the bulk density, the tapped density, composites based
and
on WC and NbC in the vibrating device. However, further investigation into its behavior
other factors that influence the feed flow behavior [31,32]. These flowability methods for
in the dynamic regime is necessary, with more significant measurement repetition, to better
powders have been known for a long time and are recommended for AM. In general, the
differentiate between powders with similar rheological properties in all the packing states.
physical behavior of powders directly influences the manufacturing process parameters
This type of dry granulation is not suitable or the most commonly used for processing
via L-PBF; the main properties are the shapes of the particles, the distribution and size of
techniques via L-PBF; it is more often used for conventional sintering techniques via LPS.
the grains, the apparent density, the thickness of the deposited layers, and the properties
No information in the literature indicates which formulation properties are suitable or
of the powder composites [33].
unsuitable for this processing method. The qualities of the parts manufactured by additive
The packing of powders can be quantified based on their bulk/tapped density.
manufacturing (AM) are influenced by the characteristics of the powders, which include the
Flowability tests allow for obtaining the mass flow rate and the bulk density [22]. Some
shape, size distribution, surface morphology, composition, and flowability or flowability of
tests promote a denser random packing, allowing the accumulated air (in the particulates)
the particulates [31].
to escape through agitation and the settling of the powders, which is performed with a
The typical particle sizes for sintering via L-PBF are in the range of 10 to 60 µm.
constant
The volume
behavior [32]. Dry granulation
of a composite/ceramic in compaction
powder techniques
is fundamental to its is unsuitable in
performance forthe
all
powder materials. Binders are needed to make this conventional processing
manufacturing process, such as in compaction, in which the filling operations of molds or method
suitable.
die powder However,
spreading thedepend
qualityon of the
the flowability.
dry granulate is known
Various methodsto significantly
or tests can impact
help withthe
compaction processes, which is shown to be a function of the rheological
the evaluation of dynamic flow and flowability. Still, specifying the best flow condition isproperties of the
granules.
not alwaysIn turn,given
clear, otherthat
studies havesizes
different found and that powders
shapes with[32].
correlate the same
In theparticle size can
AM technique,
have
the very different
experimental flow behavior is
characterization due to the effects
minimal of other properties,
when measuring composite such as the texture,
properties, such
surface shape, and accumulation [24].

3.2. Volumetric Flow Rate Test of NbC- and WC-Based Alloy Mixtures
According to the ASTM normative definitions, the volumetric flow rate allows for
controlling the mass (process) and height of the delicate layers in the powder bed
Metals 2024, 14, 1333 13 of 33

as the average diameter, particle size distribution, and packing density. It is inadequate
to solve a specific configuration for powder bed deposition [31,32]. Powdered composites
used for AM are spread onto a compacted layer for direct sintering, and this process is
repeated from layer to deposited layer to form the product. The measurement techniques
used to evaluate the powder flowability can be divided into separate categories [3], such as
the flow rate, density rates, bulk and beat density, angle of repose, and shear stress, among
other methods [1–3].
The difference between these methods can be significant, as the test method affects
the stress state and velocity of the particulates, the bulk density, the tapped density, and
other factors that influence the feed flow behavior [31,32]. These flowability methods for
powders have been known for a long time and are recommended for AM. In general, the
physical behavior of powders directly influences the manufacturing process parameters
via L-PBF; the main properties are the shapes of the particles, the distribution and size of
the grains, the apparent density, the thickness of the deposited layers, and the properties of
the powder composites [33].
The packing of powders can be quantified based on their bulk/tapped density. Flowa-
bility tests allow for obtaining the mass flow rate and the bulk density [22]. Some tests
promote a denser random packing, allowing the accumulated air (in the particulates) to es-
cape through agitation and the settling of the powders, which is performed with a constant
volume [32]. Dry granulation in compaction techniques is unsuitable for all powder materi-
als. Binders are needed to make this conventional processing method suitable. However,
the quality of the dry granulate is known to significantly impact the compaction processes,
which is shown to be a function of the rheological properties of the granules. In turn, other
studies have found that powders with the same particle size can have very different flow
behavior due to the effects of other properties, such as the texture, surface shape, and
accumulation [24].

3.2. Volumetric Flow Rate Test of NbC- and WC-Based Alloy Mixtures
According to the ASTM normative definitions, the volumetric flow rate allows for
controlling the mass (process) and height of the delicate layers in the powder bed (quality).
The flowability is a function of some parameters: the particle size distribution, particulate
or agglomerate size, cohesive strength, friction between particles, and particle morphology.
For this reason, the tests reported in Section 3.1 are indispensable. Among the standards
available for determining the flowability of powders for additive manufacturing, the Hall
flowmeter is the most widely used, being relevant for systems that use pipes, dozers, or
powder injectors in the sintering chambers, assuming that they are spherical particles and
free-flowing. However, the Carney flowmeter funnel is recommended for cemented carbide
powders or mixtures that do not flow freely. Regarding the pneumatic vibrating device, the
importance of the flow rate, or the volumetric flow rate, is related to the time required to
fill the volume in the powder bed, avoiding waste and shortages in the construction of the
powder bed [22].
The deposition and displacement speed of the device needed to achieve a uniform
flow of the metal powder mass are factors with significant influence in terms of the L-PBF
technique. For the volumetric flow test of the WC- and NbC-based mixtures, the device
was used in a stationary mode outside the sintering chamber. The mass flow results are
presented graphically as a function of the device vibration (Figure 9). For this new mass
flow test or measurement methodology, an average quantity of 1300 g of powder was placed
in the container to flow by gravity and by vibration of the device, varying the pressure in
the system from 0.25 MPa (460 Hz), 0.30 MPa (490 Hz), and 0.35 MPa (520 Hz), with the
selection of the #35 mesh sieve, the empirically ideal one as predicted in the particle size
distribution test (Figure 5). The WC-based samples presented higher bulk density values
than the NbC-based alloys and presented a longer flow time under a pressure of 0.35 MPa
(520 Hz). The pressure of 0.25 MPa (460 Hz) resulted in uneven stacking layers forming in
the powder bed, which is not recommended. The best ability of the mixtures to flow within
Metals 2024, 14, 1333 14 of 33

Metals 2024, 14, x FOR PEER REVIEW 15 of 34

the vibrating device is a function of the particle shapes, which increase the friction between
the particles; the higher the friction, the lower the flow rate of the powder mixtures in the
powder bed [22].

WC NbC
Co - 30 wt. %
Ni - 30 wt. %
(Co, Ni) - 30 wt. %

Figure 9.
Figure 9. The
The mass
mass flow
flow rate
rate of
of the
the WC-
WC- and
and NbC-based
NbC-based powder
powder mixtures
mixtures for
for aa #35
#35 mesh
mesh sieve
sieve as
as aa
function of the system pressure or container vibration frequency [22].
function of the system pressure or container vibration frequency [22].

Another important feature of the dry granulation of powder mixtures with metal
rollers is that it is widely applied in the pharmaceutical industry; it can continuously
Metals 2024, 14, 1333 15 of 33

The mixtures containing Ni presented the most challenging flow behavior. Within
the small range of the standard deviation of each mass flow curve, the data tend to be
more concentrated around the mean; that is, the flow is uniform and contributes to a
homogeneous layer in the powder bed. The pressures of 0.35 MPa (520 Hz) presented the
Metals 2024, 14, x FOR PEER REVIEW 16 of 34
best results, except for the NbC-30(Co, Ni) alloy; that is, the values of the dataset are equal
to the means, without much variation.
The NbC-30(Co, Ni) alloy presented a high variance value for all the pressures in the
produceindicating
system, large quantities
that theofobserved
compacted granular
values tend toproducts
be far fromat the
a comparatively
mean, which can lowcause
cost.
Two of the
failures main
in the advantages
deposition of this
layers process
in the bed.are thatstage
In the it is dry and continuous.
of spreading Despiteon
the mixtures being
the
powder
a simplebed usingaonly
process, the scraper
quantitative ruler, the deposition
understanding layers has
of the process in the first
been preliminary
challenging tests
due to
presented
the complex irregularities
behavior of and a lack of adhesion
the particulate materials,between
which canthe result
mixedinparticles, as shown
unsatisfactory in
linear
Figure 10A. [26].
compactions

(A) (B) (C)

Figure 10. Flow rate and spreadability behavior of mixtures in the powder bed, as a function of the
Figure 10. Flow rate and spreadability behavior of mixtures in the powder bed, as a function of the
system pressure or device vibration frequency: (A) lack of adhesion between particulates; (B) craters
system pressure or device vibration frequency: (A) lack of adhesion between particulates; (B) craters
and “alligator shell” and (C) uniform, homogeneous, and continuous flow from the central part of
and “alligator
the powder shell” and (C) uniform, homogeneous, and continuous flow from the central part of
bed.
the powder bed.
3.3. Porosities and Microstructures of WC- and NbC-Based Alloys via L-PBF
With the compacting roller rotating clockwise and counterclockwise, craters appeared
on the The samples
surface from
of the the L-PBF
powder technique
of the “alligator had dimensions
shell” type, as ofin 1.0 × 5.010B.
Figure × 5.0However,
mm (25 mm 3),
with
equivalent
the to 40 deposition
roller locked, layerscreated
it filled the gaps for allby thetheWC- and ruler
scraper NbC-based alloys. the
well, making Thedeposition
apparent
porosity analyses were performed using a light microscope (100×
layer uniform. For this reason, controlling the mass flow rate versus the flow time to control magnification), as
determined
the speed of by theASTM
vibrationB276 [34]. transfer
device Figure 11 is shows the porosity
inevitable. characteristics
In addition, there was aof the WC-
significant
and NbC-based
improvement insamples based on the
the densification, goingpowerfrom parameters
a condition (W)ofas a function
bulk densityoftothe scanning
a beaten or
speeds (mm.s —1). In this case, the apparent porosities are classified as type A, which are
pressed density, as in Figure 10C.
microporosities
An important below 10 µm, in
observation and type
this testB,is which
that forare microporosities
values below 0.20 MPa, between 10 and for
specifically 25
µm; above 25 µm, they are classified as macro porosities. There is
a #35 mesh sieve, the powder did not flow and the particles settled or packed together. For also the classification of
atype C, which
pressure represents
of 0.40 MPa, largefreequantities
or uncombined carbon. Type
of the powder were B and type
expelled C porosities
through can
the upper
impair
part sintered cemented
of the container; therefore,carbide
mass flow alloys’ mechanical
curves were not properties
performed. withFor binder phases
the powder
ranging fromthe
spreadability, 3 to 30% by weight.
displacement speed of Electrolytic
the pneumatic etching was
device necessary
varied from 40 totoanalyze
60 mm.sthe −1 .
microstructures
The moisture in the of the WC- and
mixtures NbC-based
greatly influenced alloys, as shown inbehavior,
the flowability Figure 12. Due to the high
reducing flow
corrosion resistance of cemented carbide alloys, 5 ◦
volts were applied
rate. To avoid this setback, heating the powder at 100 C for 2 h in an oven was standardized for 5 s in the presence
of the
for all Murakami
the mixtures solution:
to perform10 gthis
of potassium
test, and the ferrocyanide
same conditionsK3 [Fe(CN) 6] and 10 g of
were performed in sodium
another
period
hydroxidefor the direct dissolved
(NaOH) sintering process
in 100 cm [22].
3 of distilled water. An optical microscope (OM)

Another
was used important
to observe feature of the dry with
the microconstituents, granulation of powder
magnifications of upmixtures
to 1500×,with metal
according
rollers
to ASTM is B657
that it[35].
is widely applied
The standard inmethod
test the pharmaceutical industry; determination
for the metallographic it can continuously of the
produce large quantities
microstructure of cemented of compacted
carbides was granular
appliedproducts
to identifyat athecomparatively
eta-η phase (W low cost.
3Co 3C)
Two of the main advantages of this process are that it is dry
and α-phase (WC), with fine, medium, and coarse structures, respectively [36]. and continuous. Despite being
a simple process,to
In addition a quantitative
revealing theunderstanding
grain boundaries of theofprocess
WC and has beenthe
NbC, challenging
abnormaldue to
grain
the complex
growth behavior
of the WC- and of the particulatealloys
NbC-based materials, which can
is evidenced result in
[36,37]. The unsatisfactory
presence of linear
fine-,
compactions
medium- and[26]. coarse-grain W2C carbides (heterogeneous structure), together with the eta-
η phase, for the WC-30Co/WC-30(Co, Ni)-based alloys embrittles the material. The same
occurs for other WC- and NbC-based alloys combined with Ni. Increasing the content of
the binder phase in the initial mixture can reduce the incidence of cracking in cemented
carbides produced by the L-PBF technique. However, increasing the Co content in
cemented carbide alloys would significantly increase the manufacturing cost and decrease
the hardness of the sintered product [38].
Metals 2024, 14, 1333 16 of 33

3.3. Porosities and Microstructures of WC- and NbC-Based Alloys via L-PBF
The samples from the L-PBF technique had dimensions of 1.0 × 5.0 × 5.0 mm (25 mm3 ),
equivalent to 40 deposition layers for all the WC- and NbC-based alloys. The apparent
porosity analyses were performed using a light microscope (100× magnification), as deter-
mined by ASTM B276 [34]. Figure 11 shows the porosity characteristics of the WC- and
NbC-based samples based on the power parameters (W) as a function of the scanning
speeds (mm.s−1 ). In this case, the apparent porosities are classified as type A, which are
microporosities below 10 µm, and type B, which are microporosities between 10 and 25 µm;
above 25 µm, they are classified as macro porosities. There is also the classification of
type C, which represents free or uncombined carbon. Type B and type C porosities can
impair the sintered cemented carbide alloys’ mechanical properties with binder phases
ranging from 3 to 30% by weight. Electrolytic etching was necessary to analyze the mi-
crostructures of the WC- and NbC-based alloys, as shown in Figure 12. Due to the high
corrosion resistance of cemented carbide alloys, 5 volts were applied for 5 s in the presence
of the Murakami solution: 10 g of potassium ferrocyanide K3 [Fe(CN)6 ] and 10 g of sodium
hydroxide (NaOH) dissolved in 100 cm3 of distilled water. An optical microscope (OM)
was used to observe the microconstituents, with magnifications of up to 1500×, according
to ASTM B657 [35]. The standard test method for the metallographic determination of the
microstructure of cemented carbides was applied to identify the eta-η phase (W3 Co3 C) and
α-phase (WC), with fine, medium, and coarse structures, respectively [36].
In addition to revealing the grain boundaries of WC and NbC, the abnormal grain
growth of the WC- and NbC-based alloys is evidenced [36,37]. The presence of fine-,
medium- and coarse-grain W2 C carbides (heterogeneous structure), together with the eta-η
phase, for the WC-30Co/WC-30(Co, Ni)-based alloys embrittles the material. The same
occurs for other WC- and NbC-based alloys combined with Ni. Increasing the content of the
binder phase in the initial mixture can reduce the incidence of cracking in cemented carbides
produced by the L-PBF technique. However, increasing the Co content in cemented carbide
alloys would significantly increase the manufacturing cost and decrease the hardness of
the sintered product [38].
No scientific or technological study has been able to produce WC-based cemented
carbides without defects by additive manufacturing in a single direct sintering step. In
addition, direct sintering with a high-power laser, above 200 W, causes Co evaporation, WC
decomposition, and carbon loss, among other undesirable effects, such as the formation of
the embrittlement phases, W2 C and Co3 W3 C, which are inevitably formed in the rapidly
solidified microstructure. Other studies have already indicated that pores can form due
to gases in the molten pool as the volumetric energy density (VED = J.mm−3 ) increases,
resulting in undesirable effects such as keyholes [39]. Anomalies occurred in the structures
of the WC- and NbC-based alloys during direct sintering with rapid cooling, as shown in
Figure 13 A to H. Figure 13A shows the lack of fusion and microcracks in the WC-30Ni
alloy, with parameters of PL = 50 W; vS = 100 mm.s−1 and VED = 167 J.mm−3 . In Figure 13B,
gas was trapped during solidification, resulting in spherical pores. In Figure 13C, there are
transgranular and intergranular cracks in the WC-30Co alloy, with parameters of PL = 75 W;
vs = 25 mm.s−1 and VED = 625 J.mm−3 . Figure 13D highlights the presence of the η-phase
(W3 Co3 C), which contributes to the generation of macro- and microcracks. In Figure 13E,
the top view of sample V of the WC-Co alloy (VED = 625 J.mm−3 ) shows sintered tracks
(laser beam direction), thermal cracks, and the heat-affected zone (HAZ), as well as the
heterogeneous structure. Figure 13F has the same image of the same region as Figure 13E,
without attack, evidencing only the cracks due to thermal or internal residual stresses.
Figure 13G, an enlarged region of Figure 13E, evidences the mixed interface and HAZ, with
cobalt lakes and abnormal WC grain growth, and Figure 13H shows evaporation of the
binder phase.
Metals2024,
Metals 2024,14,
14,1333
x FOR PEER REVIEW 1717ofof34
33

WC NbC
Co - 30 wt. %
Ni- 30 wt. %
(Co, Ni) - 30 wt. %

Figure 11.
Figure 11. Characteristics
Characteristics of
of the
the apparent
apparent porosities
porosities (MO—100×)
(MO—100×)ofofthe
the WC-
WC- and
and NbC-based
NbC-based
cemented carbide samples generated by different levels of energy factors, VED (PL × vS), adapted
cemented carbide samples generated by different levels of energy factors, VED (PL × vS ), adapted
from [36,37].
from [36,37].
Metals
Metals2024,
2024,14,
14,x1333
FOR PEER REVIEW 1818ofof34
33

WC NbC
Co - 30 wt. %
Ni- 30 wt. %
(Co, Ni) - 30 wt. %

Figure
Figure12.
12.Characteristics
Characteristics of
ofthe
themicrostructures
microstructures (MO—1500×)
(MO—1500×)of ofthe
theWC-
WC-and andNbC-based
NbC-basedcemented
cemented
carbides samples generated by different levels of energy factors, VED (P L × vS), adapted from [36,37].
carbides samples generated by different levels of energy factors, VED (PL × vS ), adapted from [36,37].
 = 625 J.mm— 3) shows sintered tracks (laser beam direction), thermal cracks, and the heat-
affected zone (HAZ), as well as the heterogeneous structure. Figure 13F has the same
image of the same region as Figure 13E, without attack, evidencing only the cracks due to
thermal or internal residual stresses. Figure 13G, an enlarged region of Figure 13E,
Metals 2024, 14, 1333
evidences the mixed interface and HAZ, with cobalt lakes and abnormal WC 19 of 33
grain
growth, and Figure 13H shows evaporation of the binder phase.

(A) (B) (C) (D)

(E) (F) (G) (H)

Figure 13. Micrographs (OM): (A) Lack of fusion between the sintered layers; (B) spherical pores
Figure 13. Micrographs (OM): (A) Lack of fusion between the sintered layers; (B) spherical pores
originating from trapped gases; (C) transgranular cracks (binder phase); (D) eta-η phase (W3Co3C);
originating
(E) unmixedfrom trappedtracks
deposited gases;with
(C) transgranular cracks (binder
HAZ (heterogeneous phase);
structure); (D) eta-ηcracks
(F) thermal phaseand
(W3macro
Co3 C);
(E)
pores; (G) Co lakes and abnormal WC grain growth; and (H) evaporation of the binder phase.macro
unmixed deposited tracks with HAZ (heterogeneous structure); (F) thermal cracks and
pores; (G) Co lakes and abnormal WC grain growth; and (H) evaporation of the binder phase.
In the case of WC-Ni-based alloys, there was WC impoverishment to W2C, with a
In the
rounded case
and of WC-Ni-based
non-hexagonal alloys,
prismatic thereThe
shape. wasloss
WCofimpoverishment to W
carbon changes the 2 C, withofa
densities
rounded and non-hexagonal prismatic shape. The loss of carbon changes the densities
the alloys, and the hardness is reduced significantly. WC particles, which are directly of the
alloys, and the hardness is reduced significantly. WC particles, which are directly
exposed to a laser beam, are disintegrated into tungsten and carbon due to the excess exposed
to a laser beam, are disintegrated into tungsten and carbon due to the excess energy [37].
energy [37]. The leading cause of the η-phase appearing in the test specimens is the
The leading cause of the η-phase appearing in the test specimens is the presence of oxygen
presence of oxygen inside the sintering chamber. Even with the dynamic continuous flow
inside the sintering chamber. Even with the dynamic continuous flow of argon (0.3 L.min−1 )
of argon (0.3 L.min−1) in the system, the total elimination of oxygen is impossible. As the
in the system, the total elimination of oxygen is impossible. As the carbon content depends
carbon content depends on the sintering atmosphere, the carbon weight balance cannot
on the sintering atmosphere, the carbon weight balance cannot be fully controlled and
decarburization occurs. However, uncertainties arise in the control of microstructures,
considering the appearance of WC0.5 (called β-W2 C, hexagonal compact HC) and the stable
form WC (6.14 wt.% C, face-centered cubic phase—(FFC) [39,40]. Abnormal WC growth
can compromise the product’s performance. For example, it can contribute to premature
wear of cutting tools or machining inserts, which should be avoided. Refined or sub-micron
WC grains are the most desirable microstructural feature in these cases. However, coarse
grains can be interesting for mechanical forming tools subjected to impact, whereby higher
toughness is required. This mechanism is not fully understood, which justifies the lack of
research on this abnormal grain growth for the L-PBF [36]. In Figure 14, the microstructures
(BSE-SEM images) of WC-based alloys with 30 wt.% of binder phase, sintered by the
conventional LPS route (1370 ◦ C) and L-PBF, are compared for the sample parameters:
XI (PL = 100 W × vs = 100 mm.s−1 ) and XVI (PL = 125 W and vs = 125 mm.s−1 ). Almost
complete dissolution of WC is observed in the liquid phase and during supercooling, with
the precipitation of secondary carbides and abnormal growth of particulates. The greater
the supercooling in the binder phase, the easier the nucleation of columnar grains and
equiaxed dendrites in the region rich in the binder phase.
The formation of the metastable η-phase for the WC-Co alloys is possibly due to
its lower surface energy than WC, meaning that its formation would be energetically
favored at relatively low cooling rates. Still, at relatively fast cooling rates, the amount of
transformation is severely limited, resulting in a high fraction of retained eta-η phase and
relatively few abnormal WC grains. The lack of complete carburization of the eta-η phase
can be explained by kinetic limitations [41]. The decrease in the carbon mass fraction leads
 [39,40]. Abnormal WC growth can compromise the product’s performance. For example,
Metals 2024, 14, x FOR PEER REVIEW 21 of 34
it can contribute to premature wear of cutting tools or machining inserts, which should be
avoided. Refined or sub-micron WC grains are the most desirable microstructural feature
in these cases. However, coarse grains can be interesting for mechanical forming tools
Metals 2024, 14, 1333 transformation
subjected to impact, is severely limited,
whereby resulting
higher in a high
toughness fraction This
is required. of retained
mechanismeta-η phase
is not 20and
fully
of 33
relatively
understood, fewwhich
abnormal WC grains.
justifies the lackThe lack of complete
of research carburization
on this abnormal grainofgrowth
the eta-ηforphase
the L-
can
PBFbe[36].
explained
In Figure by 14,
kinetic limitations [41]. The
the microstructures decrease
(BSE-SEM in the of
images) carbon mass fraction
WC-based leads
alloys with 30
towt.%
the formation
to theofformation
binder phase,of the ɳ phase
of the sintered due
phase due to carbon
to carbon
by the deficiency.
deficiency.
conventional There
LPSThere
routeare are two
two°C)
(1370 typestypes
and eta-ηeta-η
of
of L-PBF, phase.
are
phase. The is first ◦ C.
The first
compared 6is
forMtheC,M 6C, which
which
sample formsforms at XI
equilibrium
at equilibrium
parameters: (PL = 100 temperatures
temperatures
W × vs = 100aboveabove
mm.s ) 1150
1150
—1 and °C.
The
XVI The sec-
(Psecond,
L= 125
ond, ◦
W 12M
Mand C,
vs =forms
C,12forms 125 at temperatures
at mm.s
temperatures below
below
—1). Almost 1150
complete1150C. °C.
In Inthethe
dissolution samples
samples
of WC isobtained
obtained viavia
observed inL-PBF,
L-PBF,
thethe the
two
liquid
two
phasetypes
types of
and of eta-η
eta-η phase
during phase resulted
resulted forwith
for the
supercooling, the
(W, (W,the6Co)
Co) 6C and
Cprecipitation
and (W,12Co)
(W, Co) C samples
C samples
of secondary
12 [42]. The ab-
[42].carbides
The abnormaland
normal
growth
abnormal growth
occurs
growth occurs
as the as thephase
eta-η eta-ηisThe
of particulates. phase is carburized
carburized
greater during
the during solidification
solidification
supercooling to form
in the binderto form
WC and WC
phase, expel
the
andCo, expel
as inCo, as
Figure in Figure
15, 15,
following following
the the
reaction: reaction:
easier the nucleation of columnar grains and equiaxed dendrites in the region rich in the
binder phase. 𝐶𝑜4W 𝑊2𝐶 + 𝐶 → 2𝑊𝐶 + 4𝐶𝑜 (3)(3)
Co 4 2 C + C → 2WC + 4Co

(A) WC-30Co (B)

WC-30Ni WC-30(Co,Ni)
1370 °C - LPS Sample

(A) HV1 = 866 ± 50 (B) HV1 = 694 ± 33 (C) HV1 = 756 ± 28

(D) HV1 = 1140 ± 85 (E) HV1 = 787 ± 30 (F) HV1 = 859 ± 33

XI Sample -L-PBF

Figure 15. BSE-SEM images corresponding to (A) WC-30Co, sample XI, Figure 14D primary den-
drites of Co (hypoeutectic structure) with the presence of eta-η phase dendrites; and (B) WC-30(Co,
Ni), sample XVI, Figure 14F WC carbides in the form of secondary dendrites at the interface between
the layers with the precipitation of small WC grains in the alternative binder phase (Co, Ni).

For the WC-Co- and WC-Ni-based alloys, the presence of carbon not combined with
W in =the chemical
± 52 (H)composition can result HV1in the± presence
330 (I) of type C porosityHV1 (free= 867
carbon,
(G) HV1 1026 = 784 ± 102
XVI Sample -L-PBF

θ-phase) in the binder phase of the sintered products, according to ASTM B276 [36], which
makes the product brittle and reduces the mechanical properties, especially for commer-
cial grades G4, G5 and G6, alloys with 20, 25 and 30 wt.% binder phase, respectively.
However, a lack of carbon can lead to the formation of carbon-poor carbide (W2C) in the
WC-Ni alloy, which in turn partially reacts with Ni to form Ni2W4C (ternary carbide).
Finally, controlling the carbon content of the WC-Ni alloy mixtures prevents the formation
of these compounds, and their harmful effect is on the densification during direct sinter-
ing [43]. Few studies have been reported on manufacturing cemented carbides and cer-
mets
Figure using
14. (A)L-PBF techniques,
WC-30Co, mainly
(B) WC-30Ni andalloys with highNi)binder
(C) WC-30(Co, produced phase contents.
by the The few
LPS conventional
Figure 14. (A) WC-30Co, (B) WC-30Ni and (C) WC-30(Co, Ni) produced by the LPS conventional
reported
route [36], or1370
published worksimages
°C; BSE-SEM have evidenced
corresponding the to:
formation
(D) WC-30Coof cracks,alloy,abnormal
(E) WC-30Ni growth
alloy of (F)
route [36], 1370 ◦ C; BSE-SEM images corresponding to: (D)—1WC-30Co alloy, (E) WC-30Ni alloy
WC-30(Co,
WC grains, Ni) alloys samples
heterogeneous XI (PL = 100
structures, W × vS = 100 mm.s
decarburization, ). BSE-SEM images
and composition
− 1
(G) WC-30Co
alteration, result-
(F) WC-30(Co,
alloy, Ni) alloys samples XI (PLNi)
= 100 W× vS = 100 mm.s ). BSE-SEM
× vS =images (G)—1WC-30Co
ing in (H)
the WC-30Ni
degradationalloy of
(I) WC-30(Co,
the mechanical alloy samples
properties XVI
[44]. L = 1256W
(PTable compares 125 mm.s ), via L-
the chemical
alloy,
PBF. (H) WC-30Ni alloy (I) WC-30(Co, Ni) alloy samples XVI (P L = 125 W × v S = 125 mm.s−1 ),
composition of the 70WC-30Co and 70WC-30(Co, Ni) alloys of samples XI and XVI with
via L-PBF.
their respective direct sintering parameters. Table 6 allows observation of the loss of car-
The formation of the metastable η-phase for the WC-Co alloys is possibly due to its
bon and Forthetheevaporation
WC-Co- and ofWC-Ni-based
the metallic phase.
alloys,Forthethe WC provided
presence of carbon innot
Table 1, the per-
combined with
lower surface energy than WC, meaning that its formation would be energetically favored
centage
W in the of combined carbon is 5.8can
chemical composition wt.% C, and
result for presence
in the NbC, it isof7.8type wt.% C.
C porosity (free carbon,
When in
θ-phase) the VED
the value
binder is very
phase high,
of the aboveproducts,
sintered VED = 300 J.mm—3to
according , the
ASTM microstructures
B276 [36], which of
the WC-Ni-based samples revealed WC grains of different sizes and
makes the product brittle and reduces the mechanical properties, especially for commercial shapes, with a heter-
ogeneous
grades G4, distribution,
G5 and G6,asalloys
shown in Figure
with 16. Figure
20, 25 and 30 wt.% 16A shows
binder that respectively.
phase, it is difficult to iden-
However,
tify what
a lack of occurs
carbon in canthe binder
lead to thephase solidofsolution
formation via OM,
carbon-poor carbidewhen (Wcompared with SEM
2 C) in the WC-Ni alloy,
images,
which as in Figure
in turn 16B,C:
partially the with
reacts WC precipitates,
Ni to form Nior WWC
2 4 C reinforcement
(ternary carbide).in the metal
Finally, matrix,
controlling
orthe
binder
carbonphase; the of
content qualitative
the WC-Ni nature
alloy of segregation
mixtures is shown,
prevents respectively,
the formation of these with a max-
compounds,
imum and aharmful
and their minimum forisWC
effect anddensification
on the Ni. Both resources,
during OMdirectand SEM images,
sintering [43]. Fewarestudies
necessary have
been reported on manufacturing cemented carbides and cermets using L-PBF techniques,
mainly alloys with high binder phase contents. The few reported or published works
Metals 2024, 14, x FOR PEER REVIEW 21 of 34

at relatively low cooling rates. Still, at relatively fast cooling rates, the amount of
Metals 2024, 14, 1333 transformation is severely limited, resulting in a high fraction of retained eta-η phase 21 and
of 33
relatively few abnormal WC grains. The lack of complete carburization of the eta-η phase
can be explained by kinetic limitations [41]. The decrease in the carbon mass fraction leads
to theevidenced
have formationthe of the ɳ phaseofdue
formation to carbon
cracks, abnormal deficiency.
growth There
of WCare two heterogeneous
grains, types of eta-η
structures,
phase. Thedecarburization,
first is M6C, whichand composition alteration,temperatures
forms at equilibrium resulting in the degradation
above 1150 °C.ofThe
the
mechanical
second, properties
M12C, forms at[44]. Table 6 compares
temperatures the chemical
below 1150 composition
°C. In the of the 70WC-30Co
samples obtained via L-PBF,
andtwo
the 70WC-30(Co, Ni) phase
types of eta-η alloys resulted
of samplesforXI
theand
(W,XVI
Co)with their
6C and (W,respective direct sintering
Co)12C samples [42]. The
parameters. Table 6 allows observation of the loss of carbon and the evaporation
abnormal growth occurs as the eta-η phase is carburized during solidification to form ofWCthe
metallic phase. For the WC provided in Table
and expel Co, as in Figure 15, following the reaction: 1, the percentage of combined carbon is
5.8 wt.% C, and for NbC, it is 7.8 wt.% C.
𝐶𝑜4𝑊2𝐶 + 𝐶 → 2𝑊𝐶 + 4𝐶𝑜 (3)

(A) (B)

Figure 15. BSE-SEM images corresponding to (A) WC-30Co, sample XI, Figure 14D primary
Figure 15. BSE-SEM images corresponding to (A) WC-30Co, sample XI, Figure 14D primary dendrites
dendrites of Co (hypoeutectic structure) with the presence of eta-η phase dendrites; and (B) WC-
of Co (hypoeutectic structure) with the presence of eta-η phase dendrites; and (B) WC-30(Co, Ni),
30(Co, Ni), sample XVI, Figure 14F WC carbides in the form of secondary dendrites at the interface
sample XVI,
between the Figure 14F WC
layers with the carbides in theofform
precipitation of secondary
small WC grainsdendrites at the interface
in the alternative binderbetween the
phase (Co,
layers with the precipitation of small WC grains in the alternative binder phase (Co, Ni).
Ni).

6. Comparison
TableFor the WC-Co-ofand
the chemical composition
WC-Ni-based alloys,ofthe
samples sintered
presence via L-PBF.
of carbon not combined with
W in the chemical composition−can result in the presence of type C porosity (free carbon,
Material Sample Parameters VED (J.mm 3 ) C (wt.%) Co (wt.%) Ni (wt.%) W (wt.%)
θ-phase) in the binder phase of the sintered products, according to ASTM B276 [36], which
70WC-30Co makes PL =the
125 Wproduct brittle 5.054445 23.11550 - 71.83005
XVI 333 and reduces the mechanical properties, especially for
70WC-30(Co, Ni) vs = 125 mm.s−1 4.655052 14.71521 12.20967 68.42006
commercial grades G4, G5 and G6, alloys with 20, 25 and 30 wt.% binder phase,
70WC-30Co PL = 100 W
respectively. However, a 333 lack of carbon 4.319959
can lead to30.20653
the formation of- carbon-poor 65.47351
carbide
XI −1
70WC-30(Co, Ni) vs = 100 mm.s 5.784030 16.73984
(W2C) in the WC-Ni alloy, which in turn partially reacts with Ni to form Ni2W4C 13.10717 64.36896
(ternary
carbide). Finally, controlling the carbon content of the WC-Ni alloy mixtures prevents the
formation
Whenofthe these
VEDcompounds,
value is very and theirabove
high, harmful
VEDeffect
= 300isJ.mm
on the
−3 ,densification during
the microstructures
direct sintering [43]. Few studies have been reported on manufacturing
of the WC-Ni-based samples revealed WC grains of different sizes and shapes, with a cemented carbides
and cermets using
heterogeneous L-PBF techniques,
distribution, as shownmainly alloys
in Figure 16.with high
Figure binder
16A shows phase
thatcontents. The
it is difficult
few reported
to identify whator occurs
published
in the works
binder have
phase evidenced the formation
solid solution via OM, whenof cracks,
comparedabnormal
with
growth
SEM images, of WC grains,
as in Figureheterogeneous
16B,C: the WC structures,
precipitates,decarburization,
or WC reinforcement and composition
in the metal
alteration, resulting in the degradation of the mechanical properties
matrix, or binder phase; the qualitative nature of segregation is shown, respectively, [44]. Table 6 compares
with
the chemical composition of the 70WC-30Co and 70WC-30(Co,
a maximum and a minimum for WC and Ni. Both resources, OM and SEM images, Ni) alloys of samples areXI
and XVI with
necessary their respective
for better structural direct sinteringOut-of-equilibrium
identification. parameters. Table 6conditions
allows observation
tend toward of
the
high loss of carbon
micro and theWC
segregation, evaporation
diffusion,ofand theWC metallic phase. For(by
reinforcement theprecipitation)
WC providedin inthe
Table
Ni
1, the percentage
metal matrix due of to combined
the high VED carbon
values is 5.8 wt.% brittle
leaving C, andstructures
for NbC, itwith
is 7.8 wt.% WC
refined C. grains
and lowWhen binder phasevalue
the VED content for Ni-based
is very high, above alloys.
VEDNb-C hasJ.mm
= 300 a wide range
—3, the of stoichiometric
microstructures of
carbon
the and hexagonal
WC-Ni-based Nb2 C
samples in the binary
revealed WC grainsdiagram, from approximately
of different 6.8 wt.%
sizes and shapes, C to
with a
11.8 wt.% C, resulting in NbC
heterogeneous distribution, as1−shown x + C [21].
in Figure 16. Figure 16A shows that it is difficult to
free
identifyIn the
whatNb-C equilibrium
occurs diagram
in the binder phasepresented by Labonne
solid solution via OM, et al. [8],
when unlike W-C
compared with(with
SEM
6.0 wt.%as
images, C),inpresented by Garcia
Figure 16B,C: the WC et al. [21], compounds
precipitates, or WCsuch as NbC1−xin
reinforcement , Nb 2 C,
the Nb4C
metal 3−x or
matrix,
Nb6 C5 can be formed. Therefore, the amount of carbon will strongly influence the structure
of NbC-based cemented carbides [8]. NbC and Nb2 C carbides are stable but have very
different crystal structures, respectively, cubic and hexagonal. The chemical bond of NbC is
 XI 333
70WC-30(Co, Ni) vs = 100 mm.s—1 5.784030 16.73984 13.10717 64.36896

In the Nb-C equilibrium diagram presented by Labonne et al. [8], unlike W-C (with
6.0 wt.% C), presented by Garcia et al. [21], compounds such as NbC1−x, Nb2C, Nb4C3−x or
Metals 2024, 14, 1333 22 of 33
Nb6C5 can be formed. Therefore, the amount of carbon will strongly influence the struc-
ture of NbC-based cemented carbides [8]. NbC and Nb2C carbides are stable but have very
different crystal structures, respectively, cubic and hexagonal. The chemical bond of NbC
very complex,
is very with
complex, a predominant
with a predominantcovalent characteristic,
covalent but but
characteristic, it also presents
it also metallic
presents and
metallic
ionic characteristics [15].
and ionic characteristics [15].

(A) HV1 = 857 ± 49 (B) (C)

Figure 16. (A) OM (1500×) of the WC-30Ni alloy, parameters: PL = 125 W, vs = 100 mm.s—1 and VED
Figure 16. (A) OM (1500×) of the WC-30Ni alloy, parameters: PL = 125 W, vs = 100 mm.s−1 and
= 400 J.mm—3; (B)− 3 BSE-SEI (13,000×) with coarse and abnormal hexagonal grains of the WC-30Ni
VED = 400 J.mm ; (B) BSE-SEI (13,000×) with coarse and abnormal hexagonal grains of the WC-30Ni
Metals 2024, 14, x FOR PEER REVIEWalloy, highlighting the light region WC, SEI image (right); and (C) BSE-SEM (20,000×) dark 23 region,
of 34
alloy, highlighting
reinforcement thebinder
of the light region WC,
phase by theSEI image (right);
precipitation and
of Ni (C) BSE-SEM
interdendritic (20,000×) dark region,
eutectics.
reinforcement of the binder phase by the precipitation of Ni interdendritic eutectics.
For producing NbC-based cermets with Ni, Co, or combined binder phases obtained
These For producing NbC-based cermets with
0.25Ni,
toCo,
3.0 or combined binder
Mo2phases obtained by3
by thesmall additions
conventional should
route vary
(indirect from
sintering), wt.%;
Figures 17WC,
andVC,18 point C,out
TaC,thatand Cr2C
there is
the
are conventional
commonly route (indirect sintering), Figures 17 and 18 point out that there is abnormal
abnormal grainused in cermets
growth for the to inhibit NbC
NbC-30Ni and grain growthNi)
NbC-30(Co, [12]. Both which
alloys, Co andreached
Ni binders
av-
grain good
have growth for the NbC-30Ni and NbC-30(Co, Ni)a alloys, which reachedofaverage sizes of
erage sizes wettability
of 9.0 ± 5.0 µmfor and
NbC7.0 and can
± 2.0 µm,dissolve significant
respectively, amount
after LPS sintering◦
carbide
at 1370 °C in for
the
9.0 ± 5.0
liquid µm4.8 7.0 ±
andwt.% 2.0Coµm, respectively, after
Ni LPS sinteringimportant
at 1370 C for 30 min of
30 minstate,
of stabilization, for
starting and 3.9 NbC
from wt.%particles
for [8]. Another
smaller than 1 µm. Thisobservation
shape of the is
stabilization,
related starting
to theanddifferentfrom
stableNbC particles smaller than 1 µm. This shapeNb of 4the WC Nb grains,
WC grains, for NbC, can and unstableinphases,
be obtained pseudocubic,
different ways; useful NbC,
strategies C3,include
and 6C5,
ad-
and can
that for NbC,
modify canthe
bestoichiometry
obtained in different
of NbC ways; useful
X during
strategies
sintering, include adjusting the
justing the production parameters (time and temperature) for allowing
conventional significant
carbidesvaria-
[45]
production
tions in WC parameters
the mechanical (time
andmade and temperature)
microstructural for conventional carbides [45] so that WC
so that grains can be spherical orproperties, because
faceted prisms ofTo
[21]. a wide
inhibitrange
this of 6.8 to
uncon-
grains can be made spherical or faceted prisms [21]. To inhibit this uncontrolled growth,
11.8 wt.% C, as shown in the Nb-C binary phase diagram by Labonne
trolled growth, considering sintering via LPS, secondary carbides are added [8,13,14]. In et al. [8], for forming
considering sintering via LPS, secondary carbides are added [8,13,14]. In this case, 3 wt.%
NbCx ranging
this case, 3 wt.%from
WC 0.75
was ≤ xadded
≤ 1.0 attothetheratio x = C/Nballoys,
NbC-based [46]. Figures
aiming17 toand 18 allow
obtain a finer obser-
and
WC was added to the NbC-based alloys, aiming to obtain a finer and more homogeneous
vation that even withmicrostructure,
more homogeneous the addition ofwith 3 wt.% WC, there
excellent was considerable
hardness, and then comparedcoalescencewith and
the
microstructure, with excellent hardness, and then compared with the samples sintered via
growth
samplesofsintered
the NbCvia grains.
L-PBF, Theas growth
in Figures behavior
19–21. of the NbC grains is therefore dependent
L-PBF, as in Figures 19–21.
on theAdding
addition of secondary
these secondarycarbides
carbidesand the combined
significantly carbon
improves thecontent [47].of the liquid
wettability
phase among the ceramic phases, improving the densification of the sintered material.
(A) HV1 = 898 ± 71 (B) (C)

Figure 17. Microstructures of the NbC-based carbide, NbC-30Ni alloy, sintered conventional LPS
Figure 17. Microstructures of the NbC-based carbide, NbC-30Ni alloy, sintered conventional LPS
route (1370 ◦°C, 30 min, and 50 MPa); (A) OM image—1000× magnification; (B) BSE-SEM—1500×
route (1370 C, 30 min, and 50 MPa); (A) OM image—1000× magnification; (B) BSE-SEM—1500×
magnification; and (C) BSE-SEM—5000× magnification.
magnification; and (C) BSE-SEM—5000× magnification.

(A) HV1 =Adding

944 ± 54 (B) secondary carbides significantly (C)
these improves the wettability of the liquid
phase among the ceramic phases, improving the densification of the sintered material.
These small additions should vary from 0.25 to 3.0 wt.%; WC, VC, Mo2 C, TaC, and Cr2 C3
are commonly used in cermets to inhibit NbC grain growth [12]. Both Co and Ni binders
have good wettability for NbC and can dissolve a significant amount of carbide in the liquid
state, 4.8 wt.% for Co and 3.9 wt.% for Ni [8]. Another important observation is related
to the different stable and unstable phases, pseudocubic, NbC, Nb4 C3, and Nb6 C5, that

Figure 18. Microstructures of the NbC-based carbide, NbC-30(Co, Ni) alloy, sintered via conven-
tional LPS route (1370 °C, 30 min, and 50 MPa); (A) OM image—1000× magnification; (B) BSE-
 (A) HV1 = 898 ± 71 (B) (C)

Metals 2024, 14, 1333 23 of 33

can modify the stoichiometry of NbCX during sintering, allowing significant variations in
the mechanical and microstructural properties, because of a wide range of 6.8 to 11.8 wt.%
C, as shown in the Nb-C binary phase diagram by Labonne et al. [8], for forming NbCx
ranging from 0.75 ≤ x ≤ 1.0 at the ratio x = C/Nb [46]. Figures 17 and 18 allow observation
that even with the addition of 3 wt.% WC, there was considerable coalescence and growth
Figure 17. Microstructures of the NbC-based carbide, NbC-30Ni alloy, sintered conventional LPS
of the NbC grains. The growth behavior of the NbC grains is therefore dependent on the
route (1370 °C, 30 min, and 50 MPa); (A) OM image—1000× magnification; (B) BSE-SEM—1500×
addition of secondary carbides and the combined carbon content [47].
magnification; and (C) BSE-SEM—5000× magnification.

Metals 2024, 14, x FOR PEER REVIEW 24 of 34

(A) 2024, 14, x FOR PEER REVIEWHV1 = 944 ± 54
Metals (B) (C) 24 of 34

model for the structure of the material, which, for technological applications, allows for
determining
model for the whether theofdeveloped
structure the material, alloys satisfy
which, forthe requirements
technological for a givenallows
applications, applica-for
determining
tion. The direct whether
sintering thetechnique
developed alloys satisfy by
is characterized therapid
requirements
heating and for cooling,
a given andapplica-
the
tion. The direct
solubility sinteringof
or dissolution technique
the WC isand characterized
NbC particles by rapid
in theheating
moltenand zonecooling,
during and the
laser
scanning
solubilityisorinadequate
dissolutionorofinterrupted.
the WC and Consequently,
NbC particles theinW,
theNbC,
molten andzone
C elements dis-
during laser
scanning
solved is inadequate
in the Co and Niorbinder interrupted.
phases,Consequently, the W, NbC,
or their combination (Co, Ni),andundergo
C elements dis-
in situ
synthesis,
solved in forming
the Co and ceramic reinforcement
Ni binder phases, phases
or theirorcombination
the secondary (Co,carbides WC andinWsitu
Ni), undergo 2C

synthesis,
in the binder forming
phase. ceramic
The reinforcement
refractory elements phases
WC or
andthe secondary
NbC diffuse carbides
into
Figure 18. Microstructures of the NbC-based carbide, NbC-30(Co, Ni) alloy, sintered via conven- the WC and
molten W
zone 2C
Figure
in the 18. Microstructures
binder phase.
the Co The of the NbC-based
refractory carbide, NbC-30(Co, Ni) alloy, sintered via conventional
and react
tional LPS with
route
◦ C,(1370
and
°C, 30 Ni min, and elements
elements 50toMPa); WCOM
generate
(A) andimage—1000×
NbC
metallic diffuse
carbide into the
phases. molten
Therefore,
magnification; zone
it is
(B) BSE-
LPS
and route
essential (1370
reactto with the30Co
determine min, andNi
that
and 50elements
the MPa);
X-ray (A) OM image—1000
diffraction
to generate(XRD) × magnification;
patterns
metallic of the
carbide (B) BSE-SEM—1500
samples
phases. align with
Therefore, it×is
SEM—1500× magnification; and (C) BSE-SEM—5000× magnification.
magnification;
essential
the and (C)
to determine
thermodynamic BSE-SEM—5000
that the X-ray
calculations × magnification.
[48]. diffraction (XRD) patterns of the samples align with
the thermodynamic
The distributioncalculations
of the NbC [48]. grains was heterogeneous in the microstructures of the
NbC-30Ni, NbC-30Co, and NbC-30(Co, Ni) alloys, sintered via L-PBF for samples XIII,
(A) HV1 = 1344 ± 160 (B) (C)
with the highest energy VED = 1000 J.mm—3. Observing Figures 19A, 20A, and 21A, it is
(A) HV1 = 1344 ± 160 (B) (C)
clear that there are limitations in evaluating the contiguity of the NbC phase regarding
the use of the OM for 1500× magnification. Better visualization of the NbC/binder phase
interaction is achieved by using SEM; notably, there is the formation of a skeleton (coales-
cence and growth of adjacent grains), the contiguity is much more significant, and abnor-
mal NbC grains are consequently predominant in the microstructure; few areas of the
binder phase are observed, which does not occur with WC-based alloys. Trapping of the
inclusions of the binder phase inside the NbCX is evidenced in the NbC-30Ni alloy, as in
Figure 19C. For a better understanding of the formation of the resulting phases for the
NbC-based alloys, the characterization by X-ray diffraction (XRD) results in an average
Figure
Figure 19.
19. Microstructures
Microstructuresof
ofthe
theNbC-30Ni
NbC-30Nialloy,
alloy, sintered
sintered via
viaL-PBF,
L-PBF, sample
sample XIII
XIII with
withparameters:
parameters:
PFigure
L = 125 19.
W, Microstructures
vs = 25 mm.s—1 and
of VED
the = 1000
NbC-30Ni J.mm —3. (A) OM image, 1500×; (B) FEG–SEM image—
alloy, sintered via L-PBF, sample XIII with parameters:
PL = 125 W, vs = 25 mm.s−1 and VED = 1000 J.mm−3 . (A) OM image, 1500×; (B) FEG–SEM image—
5000×; and
PL = 125 W,(C)
vs FEG–SEM image,
= 25 mm.s—1 20,000×.
and VED = 1000 J.mm—3. (A) OM image, 1500×; (B) FEG–SEM image—
5000×; and (C) FEG–SEM image, 20,000×.
5000×; and (C) FEG–SEM image, 20,000×.

(A) HV1 = 1519 ± 226 (B) (C)

(A) HV1 = 1519 ± 226 (B) (C)

Figure 20. Microstructures of the NbC-30Co alloy, sintered via L-PBF, sample XIII with parameters:
PFigure
L = 12520.
W,Microstructures
vs = 25 mm.s—1 and VED
of the
the = 1000 J.mm
NbC-30Co . (A) OMvia
alloy,—3sintered
sintered image, 1500×;
L-PBF, (B)XIII
sample FEG–SEM image—
with parameters:
parameters:
Figure 20. Microstructures of NbC-30Co alloy, via L-PBF, sample XIII with
5000×; andW,
PL = 125 (C)
vsFEG–SEM
= 25 mm.s−image, 20,000×.
—1 and VED = 1000 J.mm —3. (A) OM image, 1500×; (B) FEG–SEM image—
PL = 125 W, vs = 25 mm.s 1 and VED = 1000 J.mm−3 . (A) OM image, 1500×; (B) FEG–SEM image—
5000×; and (C) FEG–SEM image, 20,000×.
5000×; and (C) FEG–SEM image, 20,000×.
(A) HV1 = 1391 ± 122 (B) (C)
(A) HV1 = 1391 ± 122 (B) (C)
Metals 2024, 14, 1333 Figure 20. Microstructures of the NbC-30Co alloy, sintered via L-PBF, sample XIII with parameters:
24 of 33
PL = 125 W, vs = 25 mm.s—1 and VED = 1000 J.mm—3. (A) OM image, 1500×; (B) FEG–SEM image—
5000×; and (C) FEG–SEM image, 20,000×.

(A) HV1 = 1391 ± 122 (B) (C)

Figure 21. Microstructures of the NbC-30(Co, Ni) alloy, sintered via L-PBF, sample XIII with param-
Figure 21. Microstructures of the NbC-30(Co, Ni) alloy, sintered via L-PBF, sample XIII with param-
eters: PL = 125W, vs = 25 mm.s—1 and VED = 1000 J.mm—3. (A) OM image, 1500×; (B) FEG–SEM im-
eters: PL = 125W, vs = 25 mm.s−1 and VED = 1000 J.mm−3 . (A) OM image, 1500×; (B) FEG–SEM
age—5000×; and (C) FEG–SEM image, 10,000×.
image—5000×; and (C) FEG–SEM image, 10,000×.

The distribution of the NbC grains was heterogeneous in the microstructures of the
NbC-30Ni, NbC-30Co, and NbC-30(Co, Ni) alloys, sintered via L-PBF for samples XIII, with
the highest energy VED = 1000 J.mm−3 . Observing Figures 19A, 20A and 21A, it is clear
that there are limitations in evaluating the contiguity of the NbC phase regarding the use of
the OM for 1500× magnification. Better visualization of the NbC/binder phase interaction
is achieved by using SEM; notably, there is the formation of a skeleton (coalescence and
growth of adjacent grains), the contiguity is much more significant, and abnormal NbC
grains are consequently predominant in the microstructure; few areas of the binder phase
are observed, which does not occur with WC-based alloys. Trapping of the inclusions of
the binder phase inside the NbCX is evidenced in the NbC-30Ni alloy, as in Figure 19C.
For a better understanding of the formation of the resulting phases for the NbC-based
alloys, the characterization by X-ray diffraction (XRD) results in an average model for the
structure of the material, which, for technological applications, allows for determining
whether the developed alloys satisfy the requirements for a given application. The direct
sintering technique is characterized by rapid heating and cooling, and the solubility or
dissolution of the WC and NbC particles in the molten zone during laser scanning is
inadequate or interrupted. Consequently, the W, NbC, and C elements dissolved in the Co
and Ni binder phases, or their combination (Co, Ni), undergo in situ synthesis, forming
ceramic reinforcement phases or the secondary carbides WC and W2 C in the binder phase.
The refractory elements WC and NbC diffuse into the molten zone and react with the Co
and Ni elements to generate metallic carbide phases. Therefore, it is essential to determine
that the X-ray diffraction (XRD) patterns of the samples align with the thermodynamic
calculations [48].

3.4. Microhardness of WC- and NbC-Based Alloys via L-PBF

Table 7 provides the microindentation hardness data of the WC- and NbC-based alloys
as a function of the L-PBF sintering strategy for the different VED levels. The Vickers
hardness test determined the hardness with a load of 9807 N (1 kg). Ten indentations
on each sample were made to determine the average hardness of the samples with their
respective standard deviations. A Vickers hardness using 1 kgf of load is noted as HV1. The
hardness values available in the literature and the commercial catalogs of cemented carbide
product manufacturers indicate a range of HV 700 to 1000 for the traditional WC-30Co
alloy, designated as grade G6/G7. This is a classic grade of cemented carbides with great
prominence and application in the steel industry as cylinders, roll rings, torsion rolls, rollers,
and guide rollers [16,21,49]. This hardness range depends on the average grain size of the
WC. For the Vickers hardness tests, the HV is 0.2 to HV 100, according to the ISO 3878
standard [50].
Metals 2024, 14, 1333 25 of 33

Table 7. Vickers hardness by the microindentation of WC- and NbC-based alloys as a function of the
sintering strategy via L-PBF for different factor levels (PL × vS ).

WC NbC
Sample PL Vs VED
(W) (mm.s−1 ) (J.mm−3 ) 30Co 30Ni 30(Co, Ni) 30Co 30Ni 30(Co, Ni)
(wt. %) (wt. %) (wt. %) (wt. %) (wt. %) (wt. %)
I 25 417 983 ± 33 585 ± 147 778 ± 66 1152 ± 111 518 ± 127 991 ± 289
II 50 208 943 ± 98 762 ± 49 832 ± 93 989 ± 115 595 ± 228 735 ± 164
III 50 100 167 862 ± 148 701 ± 172 818 ± 93 884 ± 135 589 ± 74 694 ± 215
IV 125 139 798 ± 168 565 ± 97 783 ± 62 875 ± 104 577 ± 62 543 ± 115
V 25 625 1186 ± 73 625 ± 50 873 ± 55 1162 ± 232 895 ± 164 1110 ± 193
VI 50 313 1348 ± 83 788 ± 72 734 ± 96 1227 ± 220 808± 179 937 ± 155
VII 75 100 250 940 ± 116 739 ± 90 907 ± 57 1124 ± 206 644 ± 137 905 ± 132
VIII 125 208 1113 ± 69 779 ± 52 835 ± 120 1138 ± 153 815± 164 1032 ± 131
IX 25 833 1242 ± 80 756 ± 84 893 ± 69 1482 ± 238 1013± 214 1409 ± 161
X 50 417 1270 ± 40 793 ± 29 867 ± 84 1318 ± 131 1089 ± 158 1235 ± 163
XI 100 100 333 1140 ± 85 787 ± 41 859 ± 33 1130 ± 136 777 ± 224 1135 ± 215
XII 125 278 1059 ± 36 819 ± 67 838 ± 70 1134± 121 668 ± 134 992 ± 158
XIII 25 1000 1216 ± 64 672 ± 34 865 ± 43 1519 ± 226 1344 ± 160 1391 ± 122
XIV 50 500 988 ± 69 758 ± 71 842 ± 49 1241 ± 202 1087 ± 222 1322 ± 213
XV 125 100 400 1093 ± 103 803 ± 32 857 ± 49 1179 ± 140 897 ± 168 1063 ± 274
XVI 125 333 1026 ± 52 784 ± 30 867 ± 102 1157± 182 605 ± 102 851 ± 237

For grades with 30 wt.% Co binder, with extra-coarse grains, the average hardness is
equivalent to 690 HV10 (81 HRA), with a density of 12.75 g.cm−3 , and for a grade with
30 wt.% Co-Ni-Cr, the hardness is equivalent to 610 HV10 (80 HRA), with a density of
12.70 g.cm−3 . These are cemented carbide grades with high binder content, exhibiting
toughness and good thermal shock resistance, mainly of medium- to extra-coarse WC
grain size used under demanding mechanical, thermal, and often chemical operating
conditions. The mechanical and thermal loads in these applications, such as hard metal
tools and hot rolling, are highly dynamic. There is a growing trend toward extra-coarse WC
grains in mining and construction applications to profit from the higher fracture toughness
and thermal conductivity [49]. For other average Vickers hardness values, the classical
WC-30Co alloys presented ranges of 900–1000 HV30 (refined grains) and 750–850 HV30
(medium grains), respectively [51].
A marked increase in the average microhardness of the samples based on WC and
NbC with the Co binder phase is noticeable, with values above HV1 1000. Combined
with the results of the analysis of the microstructures WC-30Co in Figure 15A and NbC-
30Co in Figure 20, this evidences the diffusion of the refractory elements WC and NbC
to the binder phase, contributing to the precipitation of carbides’ refined grains from the
solid solution. Additionally, there is some evaporation of the binder phase (change in
chemical composition) and an increase in the microhardness of the binder phase due to the
segregation and emergence of dendrites and the formation of embrittlement compounds,
such as the η-phase. There is a gradual decline in microhardness, as observed in alloys
based on WC and NbC containing Ni in the binder phase, as in Table 7, taking the hardness
properties relatively below the minimum value of 700 HV1. Decarburization, abnormal
carbide growth, and the formation of the W3 Ni4 compound are the probable causes of this
undesirable effect. The only grade that maintained the hardness results in the range of
700 to 1000 HV1 was the WC-30(Co, Ni) alloy.

3.5. Thermal Residual Stresses of WC- and NbC-Based Alloys via L-PBF
Thermal residual stresses play an essential role when cracks are evidenced in the
samples, mainly in sintering via the L-PBF technique followed by rapid cooling. They arise
due to the difference in thermal expansion between the binder phase and the carbides [49].
Due to the large number of samples, the following samples were selected for residual stress
analysis: IV (VED = 139 J.mm−3 ); VII (VED = 250 J.mm−3 ); X (VED = 417 J.mm−3 ) and XIII
(VED = 1000 J.mm−3 ), according to the PL × vS strategy, as in Figure 3A, for the WC- and
NbC-based cemented carbide alloys, as shown in Figure 22A,B, respectively.
 Thermal residual stresses play an essential role when cracks are evidenced in the
samples, mainly in sintering via the L-PBF technique followed by rapid cooling. They arise
due to the difference in thermal expansion between the binder phase and the carbides [49].
Due to the large number of samples, the following samples were selected for residual
stress analysis: IV (VED = 139 J.mm—3); VII (VED = 250 J.mm—3); X (VED = 417 J.mm—3) and
Metals 2024, 14, 1333 26 of 33
XIII (VED = 1000 J.mm—3), according to the PL × vS strategy, as in Figure 3A, for the WC-
and NbC-based cemented carbide alloys, as shown in Figures 22A and 22B, respectively.

(A) (B)

Figure
Figure 22.Comparison
22. Comparison of
ofthe
theresidual
residualthermal stresses
thermal (compressive),
stresses after polishing,
(compressive), for the L-PBF
after polishing, for the L-
samples: IV (VED = 139 J.mm—3); VII− 3(VED = 250 J.mm—3); X (VED − 3 = 417 J.mm—3) and XIII−(VED
3 =
PBF samples: IV (VED = 139 J.mm ); VII (VED = 250 J.mm ); X (VED = 417 J.mm ) and XIII
1000 J.mm—3); (A)−WC-based alloys; and (B) NbC-based alloys.
(VED = 1000 J.mm 3 ); (A) WC-based alloys; and (B) NbC-based alloys.
The results show a tendency toward compressive residual stress on the surface of the
The results show a tendency toward compressive residual stress on the surface of the
samples, in general, ranging from −10 MPa to −220 MPa for the WC-based alloys and from
samples,
−10 MPaintogeneral,
−200 MParanging from −10 alloys,
for the NbC-based MPa to −220are
which MPa for the WC-based
compressive stresses, as alloys
shown and
from − 10 MPa to − 200 MPa for the NbC-based alloys, which are compressive
in Figure 22A,B. The thermal residual stresses in the polished samples of the WC-based stresses,
asand
shown in Figure 22A,B. The thermal residual stresses in the polished
NbC-based alloys containing Ni, for energies below VED = 417 J.mm , are observed —3 samples of the
WC-based and NbC-based alloys containing Ni, for energies below VED = 417 J.mm −3 ,
to be significantly lower when compared to the WC-based and NbC-based alloys contain-
areing
observed
Co and to beNi),
(Co, significantly
except for lower
VED =when compared
1000 J.mm —3. Also to verified
the WC-based
is that theandtendency
NbC-based
alloys containing Co and (Co, Ni), except for VED = 1000 J.mm −3 . Also verified is that
toward higher residual thermal stresses (compressive) is related in the WC- and NbC-
thebased
tendency toward
samples higher
containing Coresidual
[37]. Thethermal stresses
uncertainties (compressive)
of the measurements is related in the WC-
of the polished
and NbC-based
samples, samples
for energies VED containing
= 1000 J.mmCo [37].
—3, are The uncertainties
higher, as seen in all ofthe
theWC-measurements
and NbC- of
thebased samples.
polished Due tofor
samples, theenergies
temperature
VEDgradient between
= 1000 J.mm −3 , the
are sintered
higher, as layers
seenN,inN-1,
all theandWC-
andN-2 during direct
NbC-based sintering
samples. Duefor to
various VEDs, variedgradient
the temperature thermal expansions
between the and contractions
sintered layers N,
(linear/volumetric)
N-1, and N-2 duringoccur directin sintering
the rapid cooling, which
for various causevaried
VEDs, distortions
thermal and expansions
compressive and
and tensile stresses [37,48]. The residual compressive stresses on the
contractions (linear/volumetric) occur in the rapid cooling, which cause distortions surface of the sam- and
Metals 2024, 14, x FOR PEER REVIEWples resulted in tensile stresses somewhere below the surface, as
compressive and tensile stresses [37,48]. The residual compressive stresses on in Figure 23A. The 27inter-
the of 34
surface
nal stresses (tensile versus compressive) can separate the material sintered
of the samples resulted in tensile stresses somewhere below the surface, as in Figure 23A. via L-PBF from
The internal stresses (tensile versus compressive) can separate the material sintered via
the substrate,
L-PBF resulting inresulting
from the substrate, horizontal
in and verticaland
horizontal cracks, or only
vertical vertical,
cracks, which
or only resultswhich
vertical, in
slower crack growth, as shown in Figure 23B,C [ 38 ] .
results in slower crack growth, as shown in Figure 23B,C [38].

(A) (B)

Figure
Figure (A)
23.23. FEG–SEM
(A) FEG–SEMimage image(100 ×) of
(100×) ofsample
sampleXXofofthe
theNbC-30Co
NbC-30Co alloy,
alloy, parameters
parameters PL =P100
L = W,
100 W,
vs50
vs = = 50 mm.s
mm.s —— 1 1and
andVED
VED== 417
417 J.mm
J.mm—3−3; ;and
and(B)
(B)SEM
SEMimage of of
image sample
sampleXIIIXIII
of the NbC-30(Co,
of the NbC-30(Co,Ni) Ni)
alloy,
alloy, parameters:
parameters: PLPL= =125
125W,
W,vs
vs== 25
25 mm.s
mm.s—1 −1and
andVED
VED= = 1000 J.mm
1000 J.mm.−3 .
—3

Another
Another hypothesisfor
hypothesis forthe
theoccurrence
occurrenceofofmicrocracks
microcracksduring
duringdirect
directsintering
sintering via
via L-PBF
L-
PBF of Co-containing hard metal alloys is that Co (100 wt.%, pure) is allotropic;
of Co-containing hard metal alloys is that Co (100 wt.%, pure) is allotropic; above 427 ◦ C, above
427 °C, it apresents
it presents a face-centered
face-centered cubic crystalline
cubic (FCC) (FCC) crystalline structure
structure and,and,
upon upon supercool- its
supercooling,
ing, its structure changes to ε-Co hexagonal compact (HC), whereas nickel (FCC), remains
structure changes to ε-Co hexagonal compact (HC), whereas nickel (FCC), remains a FCC
a FCC upon supercooling. Therefore, Co can present two allotropic forms, ε(HC) and
upon supercooling. Therefore, Co can present two allotropic forms, ε(HC) and α(FFC). For
α(FFC). For combining Co and Ni in the proportion 1:1, a solid solution FCC (α-Co, Ni) is
combining Co and Ni in the proportion 1:1, a solid solution FCC (α-Co, Ni) is obtained
obtained after the allotropic transformation of ε-Co FCC. The Co alloy can present a stable
cubic crystalline structure from certain levels at high temperatures, 760 to 980 °C, or even
at room temperature [42]. The powder bed substrate was manufactured in AISI 1020 steel
(0.2%C) due to the manufacturing cost and the fact that it has a different thermal expan-
sion coefficient from cemented carbides, and this may have contributed to the increase in
thermal residual stresses and generation of cracks in the WC- and NbC-based samples.
Metals 2024, 14, 1333 27 of 33

after the allotropic transformation of ε-Co FCC. The Co alloy can present a stable cubic
crystalline structure from certain levels at high temperatures, 760 to 980 ◦ C, or even at
room temperature [42]. The powder bed substrate was manufactured in AISI 1020 steel
(0.2%C) due to the manufacturing cost and the fact that it has a different thermal expansion
coefficient from cemented carbides, and this may have contributed to the increase in thermal
residual stresses and generation of cracks in the WC- and NbC-based samples.

3.6. Phase Composition Analysis via XRD

Since there are several phase modifications in the microstructures, as seen in Figures 14–21,
it is essential to identify the phases, via XRD, for different factor levels (PL × vS ), which can be
used in the metallurgical industry and in the area of materials sciences, both for the conventional
LPS route and for AM [52]. For better phase identification, a sample sintered by the conventional
LPS route at 1370 ◦ C for 30 min in a vacuum furnace at 2 × 10−5 mbar was compared, which
served as a reference standard. Due to the large number of samples available, samples XIII
sintered via L-PBF, with parameters of PL = 125 W, vS = 25 mm.s−1 and VED = 1000 J.mm−3 ,
which presents the highest VED level, were selected for the XRD studies for comparison
purposes. Figure 24 compares the results of samples XIII (L-PBF) with samples sintered via
LPS, which presented significant differences, as the sintering via L-PBF leads to the formation
of new phases and the partial decomposition of chemical compounds. These phase changes in
the WC- and NbC-based alloys occurred due to the decreased carbon and evaporation of the
metallic binder phase (Co and Ni).
This topic details the critical compositional changes that occur during the direct
sintering process for the highest VED (1000 J.mm−3 ) for WC- and NbC- based alloys
with Co and alternative Ni and (Co, Ni) binder phases. As observed in Figure 24, all the
diffractometry peak points in the XRD patterns were identified. In the WC- and NbC-based
samples containing (Co, Ni), particularly for the WC-30(Co, Ni) and NbC-30(Co, Ni) alloys,
the metallic binder phase is noted to be practically a single phase in solid solution, the FCC
binder phase marked as α-(Co, Ni). For the samples containing only the Co binder phase,
the WC-30Co and NbC-30Co alloys, sintered via L-PBF and conventional LPS route, in the
XRD, a compact hexagonal HC form is evidenced, ε-Co (COD ID 9011617); and another
part transformed into the FCC form, α-Co (COD ID 9011619), allotropic transformation.
An important observation is that the mixing and homogenization process was carried out
with a simple mixer (homogenizer), not by ball milling or attrition milling, which can
mechanically induce this transformation from α-Co to ε-Co [53].
The ε-Co is transformed into α-Co above 427 ◦ C, as seen for both sintering techniques
(LPS and L-PBF), as in Figure 24. However, during cooling, the reverse transformation can
occur, with α-Co transforming into a hexagonal form; two modifications occur: α-Co and
ε-Co [42,53]. When comparing the diffractograms of the WC- and NbC-based alloys, the
peaks of the significant carbides, WC (COD ID 2102265) and NbC (COD ID 9008682), are
notable for the samples sintered via LPS and L-PBF. However, for the WC-30Co-based alloys
sintered via L-PBF, the presence of W2 C (COD ID 1539792), W9 Co3 C4 (COD ID 1528857),
and Co7 W6 (COD ID 1528857) is noticeable due to the carbon deficiency in the carburization
process to obtain WC (carbothermal reaction during the mixing of carbon black), the carbon
combined with W is well below 6.14%wt (theoretical), and the decarburization during
sintering, LPS (<50 ppm O2 ) and L-PBF (~300 ppm O2 ).
Peaks of new phases were observed, as shown in Figure 24, possibly indicating small
intermetallic phase formations of Co7 W6 (COD ID 1524837), Co3 W (COD ID 1524836),
Co2 Nb (COD ID 1525052), Co7 Nb6 (COD ID 1525267), and new carbides: CNi3 (COD ID
1563177) e Co2 C (COD ID 1528415), and high amounts of eta-η phase. Intermetallic phases
and ternary phases are key factors in the mechanical properties of cemented carbides [54],
especially for high-temperature or high-energy sintering (J.mm−3 ). The resulting phases
of the W-C-Co ternary phase diagram system are very complex, and the possible phases
include eta-η phases (Co6 W6 C and Co3 W3 C), di-tungsten carbide (W2 C) and the interme-
diate phase Co7 W6 [55]. Chicinas et al. [56] state that traditional WC-Co alloys with low
Metals 2024, 14, 1333 28 of 33

carbon content, in magnetization tests, suggest the formation of carbon-deficient phases

(M6 C and M12 C) and that the eta-η phases are not magnetic and that their formation is
influencing the total magnetic saturation, as given by the fraction of free Co. WC-Co alloys
containing sub-carbide phases, M6 C and M12 C, are of interest for cutting tools, as they
potentially increase wear resistance. To achieve this, the growth of the eta-η phase needs
to be controlled by finely dispersing them and preventing them from agglomerating. For
WC-based alloys with a high binder phase content, with applications in the steel industry,
Metals 2024, 14, x FOR PEER REVIEW 28 of 34
the formation of these large and brittle sub-carbides must be avoided [56,57]. The letter “M”
represents the metallic phase and can be Ni, Co or W in varying amounts, for example, M6 C
in WC-Co cemented carbides range from Co2.2 W3.8 C to Co2 W4 C and for M12 C, Co6 W6 C to
Cocarried out with a simple mixer (homogenizer), not
◦ by ball milling or attrition milling,
5.9 W6.1 C. The M6 C phase appears above 1150 C. After cooling, M6 C transforms into the
which can mechanically induce this transformation from α-Co to ε-Co [53].
M12 C phase and both have FCC crystal structures [55,58].

(A) WC (B) NbC

Co - 30 wt. %
Ni- 30 wt. %
(Co, Ni) - 30 wt. %

Figure 24. XRD patterns of samples XIII with parameters: PL = 125 W, vS = 25 mm.s—1 and VED = −1 and
Figure 24. —3XRD
1000 J.mm (~300patterns
ppm of Oof samples XIII with parameters: PL = 1370
2) and sample of the conventional LPS route:
125 W, vS =ppm
°C (<50 25 of
mm.s
O2),
VED = 1000 −3 (~300 ppm of O ) and sample of the conventional LPS route: 1370 ◦ C (<50 ppm
30 min and J.mm
50 MPa for comparison purposes;
2 (A) WC-based alloys; and (B) NbC-based alloys.
of O2 ), 30 min and 50 MPa for comparison purposes; (A) WC-based alloys; and (B) NbC-based alloys.
The ε-Co is transformed into α-Co above 427 °C, as seen for both sintering techniques
(LPS and L-PBF), as in Figure 24. However, during cooling, the reverse transformation
can occur, with α-Co transforming into a hexagonal form; two modifications occur: α-Co
and ε-Co [42,53]. When comparing the diffractograms of the WC- and NbC-based alloys,
the peaks of the significant carbides, WC (COD ID 2102265) and NbC (COD ID 9008682),
are notable for the samples sintered via LPS and L-PBF. However, for the WC-30Co-based
Metals 2024, 14, 1333 29 of 33

For the WC-30Ni alloy, with a considerable carbon loss, the dissolution of W is most
likely, and three phases can form during solidification, with the first phase being the NiW
compound, with a composition of 75.8 wt.% of W. Second, in inhomogeneous regions, with
44 wt.% of W, the Ni4 W compound and the NiW2 compound (86.3 wt.% W) can form,
whose formation is susceptible to the presence of oxygen in the system [59], forming the
oxides WO3 (COD ID 1521534), WO2.72 (COD ID 1538315), W2 O7 (COD ID 9014038), and
W5 O14 (COD ID 1527783). For the WC-30Co and WC-30(Co, Ni) alloys, the oxides formed
were Co3 O4 (COD ID 5910031), Co2.72 O4 (COD ID 1528446), CoO (COD ID 1541662), and
WO3 (COD ID 1521534). The WC-30(Co, Ni) alloy behaves similarly to the WC-30Ni
alloy, creating the same compounds and oxides. Alternative compositions for W- and
Ni-based refractory alloys can be efficiently designed for specific applications, provided the
embrittlement of the intermetallic precipitations for alloys with Ni-rich binders is avoided,
as they form the Ni4 W (COD ID 1523560), Ni0.85 W0.15 (COD ID 1523162), and Ni0.92 W0.08
(COD ID 1523346) compounds for slow and fast cooling. These phases can be eliminated
by a post-intermediate heat treatment [60].
For NbC-based cermet alloys, the addition of 3 wt.% of WC was observed to dissolve
in the metallic binder during L-PBF sintering, since no significant peaks were observed
in the XRD for WC. Nonetheless, some W combined with Nb was confirmed, forming
the (Nb, W) C phase, showing that WC can be dissolved in both phases: solid–solid and
solid–liquid [13]. The purpose of adding WC to NbC-based alloys was to prevent abnormal
grain growth and contribute to the hardening of the binder phase, which consequently
increases the hardness of NbC-based alloys, matching the hardness of conventional WC-
based alloys obtained via LPS sintering [11,16]. XRD diffractometry also showed that NbC
(COD ID 9008682) transformed into Nb2 C (COD ID 1540810), presenting very different
crystalline structures, FCC and HC, respectively [42]. Nb6 C5 (COD ID 1536531), another
intermetallic compound that appeared during melting and precipitation, was also detected,
which is undesirable, as it increases the brittleness and decreases the mechanical strength
of cemented carbide alloys. In addition, oxides such as NbO2 (COD ID 9009093), NbO3
(COD ID 9015165), and Nb4 O5 (COD ID 1534619) were present in the phase analysis.
Finally, the resulting structures of the WC- and NbC-based alloys via L-PBF were
heterogeneous for the different VED levels compared to the sintering technique via LPS.
Further scientific research is necessary regarding direct sintering techniques via L-PBF
for cemented carbide alloys with alternative binder phases, evaluating different factors,
such as materials (chemical composition, particle size, and shape), sintering conditions and
parameters (PL × vS ), which are the main obstacles. The undesirable effects include porosity,
carbon loss, heterogeneous structure, abnormal grain growth, formation of intermetallic
phases or compounds, microcracks, and good bonding of particles and deposition layers.
These problems are still to be better understood with more advanced studies [43,61].

4. Conclusions
The best powder flow conditions are observed in samples containing only Co, close to
the WC-30Co samples, which have good flowability, whereas the NbC-30Ni samples have
poor flowability. The porosity of the NbC-30Co mixture is more significant, requiring a
better compaction technique in the powder bed to produce a flawless part. The packing
of the mixtures for the NbC-based alloys indicated higher levels of cohesion than the
WC-based alloys due to the increased particle size and density. The WC-30Ni alloy powder
mixture with the presence of larger Ni particulates (~10 µm) had more efficient particle
packing when compared to the NbC-based alloys (<1 µm) with Co, since coarser and denser
particles tend to flow more freely in the powder bed and can therefore slide over each
other to form a more compacted powder bed. The NbC-based alloy samples presented
critical porosities after sintering via L-PBF when compared to the WC-based alloys. The
WC samples containing Ni presented lower porosity, producing a particle bed with fewer
voids during compaction. The strategies for different VED levels (PL × VS ) created in
this work provide the laser operating conditions for validating the critical functions of a
Metals 2024, 14, 1333 30 of 33

prototype (reduced scale of large parts) in a relevant environment. The WC-based alloys
performed better than the NbC-based alloys, which need improvements. The recommended
conditions to avoid undesirable effects (cracks, carbon and binder phase losses) for WC-
based and NbC-based alloys with high binder phase content (30 wt.%) are VEDs from
250 to 330 J.mm−3 . VEDs above 300 J.mm−3 easily aggravated the formation of cracks and
fissures. For VEDs below 417 J.mm−3 , they contributed to the formation of macroporosities,
leading to a poor relative density for the specimens.
Direct sintering via L-PBF generated distinct regions in the microstructures: one region
is similar to the alloys processed via LPS with primary WC carbides, the second region
is composed of W2 C (decomposition) phases and eta-η (ternary) phases, and the third
region with dendrites. For the NbC-based alloys, Nb2 C, Nb5 C6 and ternary phases were
formed. Increasing the laser power contributed to the carbon loss, evaporation of the
binder phase, formation of metal oxides and increased residual thermal stresses, resulting
in microcracks. Increasing the scanning speed results in increased apparent porosity.
Increasing the laser power increases the temperature over a wide range, leading to the
formation and enlargement of the WC decomposition region. The WC-30Co alloy samples
presented the highest number of cracks for VEDs above 333 J.mm−3 due to the thermal
residual stresses; the allotropic transformation of Co contributed to this effect.
The WC- and NbC-based alloys with Ni presented the lowest values of thermal
residual stress, below—50 MPa (compressive stress). Regarding the microstructures of
the WC-based alloys, they presented abnormal grain growth and decomposition, loss
of carbon and formation of eta-n phases, ternary phases and intermetallic compounds.
In addition, there was the presence of small amounts of oxides. The interdendritic and
dendritic regions were evidenced for WC-based alloys with Co, Ni, and their combination
(Co, Ni) binder phases for alloys sintered above VED = 300 J.mm−3 . For WC-30Co, the
binder phase comprises primary dendrites (hypo eutectic structure), and for the WC-30Ni
alloy, interdendritic eutectics.
The microhardness of the WC-30Ni and WC-30(Co, Ni) alloys, via L-PBF, for all
the sintering parameters (PL × vS ), presented values close to the reference values of the
traditional WC-30Co alloy HV1 = 800 to HV1 = 1000. For the WC- and NbC-based alloys
combined with Co, some samples resulted in values higher than HV1 > 1000 due to the
hardening of the binder phase, evaporation of the binder phase, presence of intermetallic
phases, eta-n phase, metal oxides, etc. The WC- and NbC-based alloys with Ni presented
lower thermal residual stress values for energies below VED = 417 J.mm−3 , when compared
to the Co-based alloys, which were the samples with the highest number of cracks due to
allotropic transformation.

5. Future Directions
Demonstrating the critical functions of a prototype in a relevant environment, we
are considering the following suggestions to proceed with this work: corrections of WC-
and NbC-based mixtures with the addition of free carbon (carbon black) or at the request
of carbide manufacturers (raw material), with combined carbon close to the levels of
10 wt.% C for NbC and 6.1 wt.% C for WC, to avoid the emergence of eta-η phases
and other compounds. Furthermore, grain growth inhibitors will be added to WC and
NbC. Improvements and new studies on the rheology of NbC cermet-based mixtures.
Studying the variation in the compaction pressure of WC- and NbC-based mixtures in the
powder bed, with the metal roller rotating clockwise and counterclockwise. The L-PBF
parameters for WC- and NbC-based carbide alloys with high binder phase content, as
defined in this study, can potentially be applied (with the pneumatic vibratory device) to
the manufacturing of pilot-scale parts for complex shapes with high relative density and
without microcracks for energies below VED = 300 J.mm−3 . Comparing the direct sintering
techniques (L-PBF) with a dynamic flow atmosphere (argon or nitrogen) with the vacuum
chamber (2 × 10−5 mbar in terms of carbon loss, wt.% C) and performing heat treatment
via vacuum sintering to eliminate possible cracks and dendrites, if necessary.
Metals 2024, 14, 1333 31 of 33

Author Contributions: F.M.: writing—original draft, resources, conceptualization, data curation,

formal analysis, investigation, methodology, visualization; dos Santos, M.O.d.S.: writing—review and
editing, validation and methodology; R.C.: writing—review and editing, validation and methodology;
Pereira, N.M.G.P.: validation and methodology; D.R.: conceptualization, funding acquisition, project
administration, resources, validation and supervision; S.R.J.: funding acquisition, project adminis-
tration, resources, validation, supervision, conceptualization, writing—review and editing; F.d.S.O.:
conceptualization, validation, methodology and writing—review and editing; M.V.M.: validation and
writing—review and editing; R.S.C.: validation and writing—review and editing; R.R.d.O.: validation
and methodology; L.G.M.: validation and methodology; and G.F.B.: conceptualization, validation
and supervision. All authors have read and agreed to the published version of the manuscript.
Funding: This study was financed, in part, by the São Paulo Research Foundation (FAPESP), Brasil.
Research Grant—Innovative Research in Small Businesses—PIPE, Process Number #2022/06201-7.
Data Availability Statement: Data will be made available on request.
Acknowledgments: The authors thank the University Center FEI, Mauá Institute of Technology, the
Institute of Energy and Nuclear Research (IPEN), and the company Brats Filters and Metal Powders
for carrying out the experimental tests. Furthermore, the authors acknowledge the financial support
from the São Paulo Research Foundation (FAPESP), Brazil.
Conflicts of Interest: The authors declare that they have no known competing financial interests or
personal relationships that could have appeared to influence the work reported in this paper. The
funders had no role in the study design, in the collection, analyses, or interpretation of data, in the
writing of the manuscript, or in the decision to publish the results.

References
1. Roulon, Z.; Missiaen, J.-M.; Lay, S. Carbide grain growth in cemented carbides sintered with alternative binders. Int. J. Refract.
Met. Hard Mater. 2019, 86, 105088. [CrossRef]
2. Acchar, W.; Zollfrankb, C.; Greilb, P. Microstructure and Mechanical Properties of WC-Co Reinforced with NbC. Mater. Res. 2004,
7, 445–450. [CrossRef]
3. Jucan, O.-D.; Gadalean, R.-V.; Chicina¸S, H.-F.; Bâlc, N.; Popa, C.-O. The Assessment of the Transversal Rupture Strength (TRS)
and Hardness of WC-Co Specimens Made via Additive Manufacturing and Sinter-HIP. Metals 2023, 13, 1051. [CrossRef]
4. Miranda, F.; Fernandes, L.J.; Batalha, M.H.; Rodrigues, D.; Stoeterau, R.L.; Batalha, G.F. Critical raw materials in cutting tools for
machining applications—A review. Braz. J. Dev. 2021, 7, 49513–49537. [CrossRef]
5. Miranda, F.; Stoeterau, R.L.; Batalha, G.F. Occupational Risks in the Manufacture of Products Heavy Alloy and Hardmetals
Cutting Tools in the Mechanical Industry. Glob. J. Med. Res. K Interdiscip. 2021, 21, 29–51. [CrossRef]
6. Akira, M.; Suganuma, N. Hard metal lung disease. Health Sci. Rev. 2024, 11, 10016. [CrossRef]
7. McCullough, E.; Nassar, N.T. Assessment of critical minerals: Updated application of an early-warning screening methodology.
Miner. Econ. 2017, 30, 257–272. [CrossRef]
8. Labonne, M.; Missiaen, J.-M.; Lay, S.; García, N.; Lavigne, O. Sintering behavior and microstructural evolution of NbC-Ni
cemented carbides with Mo2C additions. Int. J. Refract. Met. Hard Mater. 2020, 92, 105295. [CrossRef]
9. Alam, M.E.; Wang, J.; Henager, C.H., Jr.; Setyawan, W.; Odette, G.R. The effect of hot rolling on the strength and fracture toughness
of 90W–7Ni3Fe tungsten heavy metal alloys. Mater. Sci. Eng. A. 2021, 824, 141738. [CrossRef]
10. Correa, E.O.; Santos, J.N.; Klein, A.N. Microstructure and mechanical properties of WC Ni–Si-based cemented carbides developed
by powder metallurgy. Int. J. Refract. Met. Hard Mater. 2010, 28, 572–575. [CrossRef]
11. Fernandes, L.J.; Stoeterau, R.L.; Batalha, G.F.; Rodrigues, D.; Borrile, A.V. Wear analysis of NbC-Ni cemented carbides for cutting
tools. Adv. Mater. Process. Technol. 2022, 8, 305–321. [CrossRef]
12. Franco, E.; Costa, C.; Tsipas, S.; Gordo, E. Cermets based on FeAl–NbC from composite powders: Design of composition and
processing. Int. J. Refract. Met. Hard Mater. 2015, 48, 324–332. [CrossRef]
13. Huang, S.G.; Liu, R.L.; Li, L.; Van Der Biest, O.; Vleugels, J. NbC as grain growth inhibitor and carbide in WC–Co hardmetals. Int.
J. Refract. Met. Hard Mater. 2008, 26, 389–395. [CrossRef]
14. Huang, S.G.; Li, L.; Van Der Biest, O.; Vleugels, J. Influence of WC addition on the microstructure and mechanical properties of
NbC-Co cermets. J. Alloys Compd. 2007, 430, 158–164. [CrossRef]
15. Cuppari, M.; Santos, S. Physical Properties of the NbC Carbide. Metals 2016, 6, 250. [CrossRef]
16. Rodrigues, D.; Cannizza, E. The use of NbC20Ni Hard materials for hot rolling applications. In Proceedings of the 19th Plansee
Seminar 2017, Reutte, Austria, 29 May–2 June 2017; Volume 3.
17. Li, J.; Wei, Z.; Zhou, B.; Wu, Y.; Chen, S.-G.; Sun, Z. Densification, Microstructure and Properties of 90W-7Ni-3Fe Fabricated by
Selective Laser Melting. Metals 2019, 9, 884. [CrossRef]
Metals 2024, 14, 1333 32 of 33

18. Sidambe, A.; Tian, Y.; Prangnell, P.; Fox, P. Effect of processing parameters on the densification, microstructure, and crystal-
lographic texture during the laser powder bed fusion of pure tungsten. Int. J. Refract. Met. Hard Mater. 2018, 78, 254–263.
[CrossRef]
19. Wang, D.; Yu, C.; Zhou, X.; Ma, J.; Liu, W.; Shen, Z. Dense Pure Tungsten Fabricated by Selective Laser Melting. Appl. Sci. 2017,
7, 430. [CrossRef]
20. Campanelli, S.; Contuzzi, N.; Posa, P.; Angelastro, A. Printability and Microstructure of Selective Laser Melting of WC/Co/Cr
Powder. Materials 2019, 12, 2397. [CrossRef]
21. García, J.; Collado Ciprés, V.; Blomqvist, A.; Kaplan, B. Cemented carbide microstructures: A review. Int. J. Refract. Met. Hard
Mater. 2018, 80, 40–68. [CrossRef]
22. Miranda, F.; Maiolini, A.; Dos Santos, M.O.; Rodrigues, D.; Janasi, S.R.; Batalha, G.F. Towards the flowability and spreadability
of cemented carbides and cermet powders for Additive Manufacturing: Experimental and numerical approach—Part 1. In
Proceedings of the 27th International Congress of Mechanical Engineering, Rio de Janeiro, Brazil, 4–8 December 2023. [CrossRef]
23. Budding, A.; Vaneker, T. New Strategies for Powder Compaction in Powder-based Rapid Prototyping Techniques. Procedia CIRP
2013, 6, 528–533. [CrossRef]
24. Freeman, T.; Bey, H.; Hanish, M.; Brockbank, K.; Armstrong, B. The influence of roller compaction processing variables on the
rheological properties of granules. Asian J. Pharm. Sci. 2016, 11, 516–527. [CrossRef]
25. Santos, L.C.; Condotta, R.; Ferreira, M.C. Flow properties of coarse and fine sugar powders. J. Food Process Eng. 2017, 41, e12648.
[CrossRef]
26. Gururajan, B.; Seville, J.P.K.; Adams, M.; Greenwood, R.; Fitzpatrick, S. Roll compaction of a pharmaceutical excipient: Experi-
mental validation of rolling theory for granular solids. Chem. Eng. Sci. 2005, 60, 3891–3897. [CrossRef]
27. Aramian, A.; Razavi, M.; Sadeghian, Z.; Berto, F. A Review of Additive Manufacturing of Cermets. Addit. Manuf. 2020, 33, 101130.
[CrossRef]
28. Uhlmann, E.; Bergmann, A.; Gridin, W. Investigation on Additive Manufacturing of Tungsten Carbide-Cobalt by Selective Laser
Melting. Procedia CIRP 2015, 35, 8–15. [CrossRef]
29. Zegzulka, J.; Gelnar, D.; Jezerská, L.; Prokes, R.; Rozbroj, J. Characterization and flowability methods for metal powders. Sci. Rep.
2020, 10, 21004. [CrossRef]
30. Freeman, R. Measuring the flow properties of consolidated, conditioned and aerated powders—A comparative study using a
powder rheometer and a rotational shear cell. Powder Technol. 2007, 174, 25–33. [CrossRef]
31. Debroy, T.; Wei, H.; Zuback, J.; Mukherjee, T.; Elmer, J.; Milewski, J.O.; Beese, A.; Wilson-Heid, A.; De, A.; Zhang, W. Additive
manufacturing of metallic components—Process, structure, and properties. Prog. Mater. Sci. 2018, 92, 112–224. [CrossRef]
32. Marchetti, L.; Hulme-Smith, C. Flowability of steel and tool steel powders: A comparison between testing methods. Powder
Technol. 2021, 384, 402–413. [CrossRef]
33. Pal, S.; Drstvensek, I.; Brajlih, T. Physical behaviors of materials in selective laser melting process. In DAAAM International
Scientific Book 2018; DAAAM International Vienna: Vienna, Austria, 2018; pp. 239–256. [CrossRef]
34. ASTM Standard B276-21; Evaluation of Apparent Porosity in Cemented Carbides. ASTM International: West Conshohocken, PA,
USA, 2021. Available online: https://www.astm.org/b0276-21.html (accessed on 17 March 2023).
35. ASTM Standard B657-92; Standard Test Method for Metallographic Determination of Microstructure in Cemented Tungsten
Carbides. ASTM International: West Conshohocken, PA, USA, 2000. Available online: https://www.astm.org/b0657-92r00.html
(accessed on 16 August 2017).
36. Miranda, F.; Dos Santos, M.O.; Rodrigues, D.; Batalha, G.F.; Janasi, S.R.; Ortega, F.S. WC Cemented Carbides: Microstructural
aspects comparing L-PBF Additive Manufacture and convencional LPS. In Proceedings of the Euro Powder Metallurgy 2023
Congress & Exhibition, Lisbon, Portugal, 1–4 October 2023. [CrossRef]
37. Miranda, F.; Santos, M.O.; Coelho, R.S.; Rodrigues, D.; Batalha, G.F. NbC-based Cermet production comparison: L-PBF Additive
Manufacturing versus conventional LPS powder metallurgy. Mater. Tehnol. 2023, 57, 129–137. [CrossRef]
38. Xing, M.; Wang, H.; Zhao, Z.; Lu, H.; Liu, C.; Lin, L.; Wang, M.; Song, X. Additive manufacturing of cemented carbide inserts
with high mechanical performance. Mater. Sci. Eng. A 2022, 861, 144350. [CrossRef]
39. Schwanekamp, T.; Marginean, G.; Reuber, M.; Ostendorf, A. Impact of cobalt content and grain growth inhibitors in laser-based
powder bed fusion of WC-Co. Int. J. Refrac. Met. Hard Mat. 2022, 105, 105814. [CrossRef]
40. Lee, S.W.; Kim, Y.W.; Jang, K.M.; Lee, J.W.; Park, M.-S.; Koo, H.Y.; Ha, G.-H.; Kang, Y.C. Phase control of WC–Co hardmetal using
additive manufacturing technologies. Powder Metall. 2022, 65, 13–21. [CrossRef]
41. Vreeswijk, M.; Kot, A.; Giuliani, F.; Humphry-Baker, S. Abnormal WC crystal growth from liquid Co flux occurs via eta phase
decomposition. Int. J. Refract. Met. Hard Mater. 2021, 99, 105589. [CrossRef]
42. Fernandes, C.M.; Senos, A.M.R. Cemented carbide phase diagrams: A review. Int. J. Refract. Met. Hard Mater. 2011, 29, 405–418.
[CrossRef]
43. Serkhane-Ouabadi, N.; Benalia-Baguenane, G.; Nechiche, M.; Azemu, S. Influence of the liquid phase and the carbon on the
reactive sintering in mixtures W-Ni-C. U.P.B. Sci. Bull. 2022, 84, 109–122.
44. Jose, S.A.; John, M.; Menezes, P.L. Cermet Systems: Synthesis, Properties, and Applications. Ceramics 2022, 5, 210–236. [CrossRef]
45. Straumal, B.B.; Konyashin, I. Faceting/Roughening of WC/Binder Interfaces in Cemented Carbides: A Review. Materials 2023,
16, 3696. [CrossRef]
Metals 2024, 14, 1333 33 of 33

46. Woydt, M.; Huang, S.; Vleugels, J.; Mohrbacher, H.; Cannizza, E. Potentials of niobium carbide (NbC) as cutting tools and for
wear protection. Int. J. Refract. Met. Hard Mater. 2018, 72, 380–387. [CrossRef]
47. Huang, S.G.; De Baets, P.; Sukumaran, J.; Mohrbacher, H.; Woydt, M.; Vleugels, J. Effect of Carbon Content on the Microstructure
and Mechanical Properties of NbC-Ni Based Cermets. Metals 2018, 8, 178. [CrossRef]
48. Lv, N.; Yue, H.; Guo, C.; Dai, W.; Zhang, J.; Li, Q.; Zhao, G.; Hao, G. A comparative investigation on the effects of reinforcement
phase addition methods on laser melting deposited WC/Co coatings. J. Manuf. Process. 2024, 129, 134–146. [CrossRef]
49. Sarin, V.; Mari, D.; Llanes, L.; Nebel, C.E. Comprehensive Hard Materials; Elsevier: Amsterdam, The Netherlands, 2014; Volume I,
p. 700.
50. ISO Standard 3878-1983(E); Hardmetals—Vickers Hardness Test. ISO International Standard: Geneva, Switzerland, 1983. Available
online: https://www.iso.org/standard/9484.html (accessed on 8 October 2024).
51. Chen, C.; Huang, B.; Liu, Z.; Li, Y.; Zou, D.; Liu, T.; Chang, Y.; Chen, L. Additive manufacturing of WC-Co cemented carbides:
Process, microstructure, and mechanical properties. Addit. Manuf. 2023, 63, 103410. [CrossRef]
52. Khmyrov, R.S.; Shevchukov, A.P.; Gusarov, A.V.; Tarasova, T.V. Phase composition and microstructure of WC–Co alloys obtained
by selective laser melting. Mech. Ind. 2017, 18, 714. [CrossRef]
53. Zhang, F.L.; Wang, C.Y.; Zhu, M. Nanostructured WC/Co composite powder prepared by high energy ball milling. Scripta Mat.
2003, 49, 1123–1128. [CrossRef]
54. Balbino, N.A.N.; Corrêa, E.O.; Huanca, D.R.; de Freitas Matos, F.A.; de Carvalho Valeriano, L. Comparative Study of Corrosion
Behaviors of WC-NiMo and WC-Co Cemented Carbides. Materials 2023, 16, 4480. [CrossRef]
55. Li, X.; Zhang, X.; Zhang, J.; Zhang, Q.; Ji, V.; Liu, J. Effect of Mo and C Additions on Eta Phase Evolution of WC-13Co Cemented
Carbides. Coatings 2022, 12, 1993. [CrossRef]
56. Chicinas, H.F.; Marinca, T.F.; Götze, P.; Eckert, A.; Popa, C.O. Influence of aqueous milling duration on the sintered WC–10Co
hard metal powders. J. Mater. Sci. 2018, 53, 2901–2910. [CrossRef]
57. Tran, S. Microstructure Investigations of WC-Co Cemented Carbide Containing η-Phase and Cr. Master’s Thesis, Uppsala
University, Uppsala, Sweden, 8 August 2018; p. 51.
58. Upadhyaya, G.S. Cemented Tungsten Carbides: Production, Properties, and Testing; Noyes Publications: Norwich, NY, USA, 1998;
p. 27, ISBN 0-8155-1417-4. Available online: https://books.google.com.br/books?id=9UVcIa8M4V8C&printsec=frontcover&hl=
pt-BR&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false (accessed on 15 October 2024).
59. Enneti, R.; Morgan, R.; Wolfe, T.; Harooni, A.; Volk, S. Direct Metal Laser Sintering (DMLS) of Tungsten Powders. Int. J. Powder
Metall. 2017, 53, 23–31.
60. Funch, C.V.; Proust, G. Laser-based additive manufacturing of refractory metals and their alloys: A review. Addit. Manuf. 2024,
94, 104464. [CrossRef]
61. Chhabra, D. Comparison and analysis of different 3d printing techniques. Int. J. Latest Trends Eng. Technol. 2017, 8, 264–272.
[CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.