materials

Article
Selective Laser Sintering of Porous Silica Enabled by
Carbon Additive
Shuai Chang 1,2 , Liqun Li 1 , Li Lu 2 and Jerry Ying Hsi Fuh 2, * ID

1 State Key Laboratory of Advanced Welding and Joining, Harbin Institute of Technology,
Harbin 150001, China; msecs@nus.edu.sg (S.C.); liliqun@hit.edu.cn (L.L.)
2 Department of Mechanical Engineering and Centre for Additive Manufacturing,
National University of Singapore (NUS), Singapore 117576, Singapore; mpeluli@nus.edu.sg
* Correspondence: jerry.fuh@nus.edu.sg; Tel.: +65-6516-6690

Received: 30 September 2017; Accepted: 13 November 2017; Published: 16 November 2017

Abstract: The aim of this study is to investigate the possibility of a freeform fabrication of porous
ceramic parts through selective laser sintering (SLS). SLS was proposed to manufacture ceramic
green parts because this additive manufacturing technique can be used to fabricate three-dimensional
objects directly without a mold, and the technique has the capability of generating porous ceramics
with controlled porosity. However, ceramic printing has not yet fully achieved its 3D fabrication
capabilities without using polymer binder. Except for the limitations of high melting point, brittleness,
and low thermal shock resistance from ceramic material properties, the key obstacle lies in the
very poor absorptivity of oxide ceramics to fiber laser, which is widely installed in commercial
SLS equipment. An alternative solution to overcome the poor laser absorptivity via improving
material compositions is presented in this study. The positive effect of carbon additive on the
absorptivity of silica powder to fiber laser is discussed. To investigate the capabilities of the
SLS process, 3D porous silica structures were successfully prepared and characterized.

Keywords: selective laser sintering (SLS); porous ceramic; carbon additive; laser absorptivity

1. Introduction
Advanced ceramic materials offer great potential for high-end applications due to their unique
properties such as high melting point, exceptional mechanical strength, superior wear and thermal
resistance, and excellent chemical stability [1,2]. Tailored porous ceramic exhibits more special features
including low thermal mass, low thermal conductivity, high surface area, low density, and high specific
strength. Therefore, ceramics containing controlled porosity nowadays find many applications as
end products, especially for operation environments where high temperature, extensive wear, and
corrosive media are involved. The increasing number of applications for such materials include, for
example, the filtration of hot corrosive gases and molten metals, high-temperature thermal insulation,
catalytic carriers in various chemical processes, and membranes for separation and purification [3–5].
Despite their excellent properties, the wider use of porous ceramics is still limited because
traditional manufacturing processes (e.g., dry pressing, slip casting, tape casting, and injection molding)
are time-consuming and shaping through molding is limited, particularly for complex parts [6–8].
However, recent developments in additive manufacturing (AM) technologies allows for a rapid
freeform fabrication of parts with complex geometries that could be very difficult or even impossible to
fabricate by the conventional techniques. AM in general and selective laser sintering (SLS) in particular
have recently gained popularity because of their ability to produce complex porous ceramic parts
directly without a mold [9]. Manufacturing of complex polymeric or metallic parts directly has been
well studied. However, there are very few works on the direct selective laser sintering (SLS) of ceramics
without the use of a polymer binder. This is still a very challenging task due to the brittleness, and

Materials 2017, 10, 1313; doi:10.3390/ma10111313 www.mdpi.com/journal/materials

Materials 2017, 10, 1313 2 of 11

low thermal shock resistance properties of the material [10,11]. Besides, the key obstacle for SLS of
such materials is that oxide ceramics usually show very poor absorptivity to near-infrared laser [12].
Several researchers attempted to obtain the direct SLS of ceramics through the addition of a powder
coating to low melting point ceramics or by using ceramic composites with lower melting points [13].
However, the laser absorptivity of the ceramic powder coated with some transparent material such
as silica was found to be even weaker [14]. Additionally, silica parts built by direct laser melting or
sintering have been examined to fabricate complex tooling via sandcasting. The low purity of the silica
sand used in the study resulted in a reduced melting temperature, and very poor mechanical strength.
The large pores and cracks also limited its practical use [15].
Thus, this work aims to investigate the capabilities of a rapid fabrication of porous ceramic parts
via the SLS technique. In general, the parts that had higher green density after SLS also exhibited
higher strength. The higher the green strength, the easier complex shaped parts could be fabricated
without the risk of breaking. Therefore, the direct SLS of porous ceramics remains challenging [16].
CO2 laser and Nd:YAG fiber laser are generally used in most SLS/M machines. In comparison to
Nd:YAG fiber laser emitting at 1070 nm, CO2 laser with a wavelength of 10.6 µm is much more easily
absorbed by most ceramic materials, although the diameter of CO2 laser spot is much larger than that
of the fiber laser [17,18]. Therefore, the fiber laser is more suited for processing with higher accuracy.
However, silica is almost non-absorbent to Nd:YAG fiber [19,20]. This study on powder absorptivity to
laser aims to improve the process of selective laser sintering (SLS), and it improves our understanding
of the mechanism of the interaction between laser and materials, which is crucial to finding a more
uniform and suitable processing window for SLS [21]. Therefore, a solution of improving ceramic
materials’ absorptivity to laser is proposed. To investigate the influence of a carbon additive on ceramic
material, 3D test specimens are successfully fabricated and characterized via the SLS process.

2. Materials and Methods

2.1. Principle of SLS of Materials

Selective laser sintering was invented by Carl R. Deckard and Joseph J. Beaman in the 1980s at
the University of Texas [18]. The selective laser sintering (SLS) technique, as depicted in Figure 1, is a
powder bed fusion technique in which a laser beam is used to fuse powder materials selectively
according to the digital design of the built part. The CAD/CAM (computer-aided design and
computer-aided manufacturing) model is sliced into thin layers with thicknesses typically less
than 100 µm [22]. This technique allows for the generation of three-dimensional complex and
near-net-shaped parts.
The intrinsic feature of laser processing of ceramics is that, at a certain time of interaction between
the laser beam and ceramic powder bed, only a little laser energy is absorbed; some is transmitted and
the rest is reflected away as depicted in Figure 1. To evaluate the efficiency of laser processing, it is
necessary to know how much of the incident laser intensity is coupled to the sample. This coupling
efficiency is described by “absorptance” (this is also referred to as absorption coefficient or just
absorption). Other quantities often used to evaluate the laser processing are called reflectance and
transmittance. The delineations of these quantities are given as follows. The absorptance, A, is given by:

A = 1 − T− R (1)

A = (Absorbed laser energy in the sample)/(Total incident laser power) (2)

while reflectance, R, is:

R = (Reflected laser energy from the sample)/(Total incident laser power) (3)
Materials 2017, 10, 1313 3 of 11

Materials 2017, 10, 1313 3 of 11

Materials
and 2017, 10, 1313
transmittance, T, is defined by: 3 of 11
and transmittance, T, is defined by:
and transmittance, T, is defined
T = (Transmitted laserby:
energy through the sample)/(Total incident laser power) (4)
T = (Transmitted laser energy through the sample)/(Total incident laser power) (4)
T = (Transmitted laser energy through the sample)/(Total incident laser power) (4)
In this study,
study, reflectance
reflectance (R)
(R) and
and transmittance
transmittance (T)
(T) are
are measured.
measured.
In this study, reflectance (R) and transmittance (T) are measured.

Figure 1. Schematic
Figure 1. Schematic drawing
drawing of
of laser
laser impinging
impinging on
on the
the powder
powder bed
bed in the selective
in the selective laser
laser sintering
sintering
Figureprocess.
(SLS) 1. Schematic drawing of laser impinging on the powder bed in the selective laser sintering
(SLS) process.
(SLS) process.
2.2. Materials and Samples Preparation
2.2.
2.2. Materials
Materials and
and Samples
Samples Preparation
Preparation
Silica powder (SS1206, Industrial Powder, Buffalo Grove, IL, USA) was used for this study.
Silica powder (SS1206, Industrial Powder,
Powder, Buffalo Grove, Grove, IL,
IL, USA)
USA) was
was used
used forfor this study.
study.
SEM Silica powder
micrograph and(SS1206, Industrial
laser diffraction technique Buffalo
are used together to determine particle this
shape and
SEM
SEM micrograph
micrograph and
and laser diffraction
laserMost
diffraction technique
technique are
are used
used together
together toto determine
determine particle
particle shape
shapethanand
and
particle size distributions. of the particles shown are spherical with a diameter no greater
particle
particle size
size distributions.
distributions. Most
Most of
of the
the particles
particles shown
shown are
are spherical
spherical with
with a
a diameter
diameter no
no greater
greater than
than
60 μm (Figure 2). Spherical shape is beneficial to the SLS process, since the conformation towards
60
60 µm
μmcan(Figure
(Figure 2). Spherical
2). Spherical shapeshape is beneficial
beneficial to to the
the SLS
SLS process, since
since the
the conformation towards
unity gradually enhance both is its powder packing process,
density and flowability conformation towards
[23,24]. As shown in
unity can
unity can gradually
gradually enhance both its powder packing density and flowability [23,24]. As shown in
Figure 3, the particleenhance both its powder
size is distributed from 1packing
μm to density and flowability
60 μm (measured [23,24].
by laser As shown
diffraction on LSin
Figure
Figure 3,3,the
theparticle
particle size
sizeis is
distributed
distributed from 1 µm
from 1 to 60toµm
μm 60 (measured
μm by laser
(measured by diffraction
laser on LS 100
diffraction on Q,
LS
100 Q, Coulter International Corporation, West Lafayett, IN, USA). The SLS technique is more
Coulter International
100 Q, Coulter Corporation, West Lafayett,
International IN, USA).IN, The SLS technique SLS is more commonly
commonly utilized with typicalCorporation,
particle size West
rangesLafayett,
of 20–150 μmUSA). Thepowder.
for metal technique is more
We selected the
utilized
commonly with typical
utilized particle
with size
typical ranges
particle of 20–150
size ranges µm offor metal
20–150 μmpowder.
for We
metal selected
powder. the
We particle
selected size
the
particle size range from 1 μm to 60 μm since the melting point of ceramic is much higher than that
range from
particle size1range
µm tofrom 60 µm 1 sincetothe
μm 60 melting
μm sincepoint
the of ceramic
melting is much
point of higheristhan
ceramic much that of metals
higher than [23].
that
of metals [23]. Fine active carbon powder (50–150 nm) with the mass fraction of 0.1–0.3% was doped
Fine active carbon
of metals powder (50–150 nm) with thenm) mass fraction of 0.1–0.3% was dopedwas intodoped
silica
into silica [23].
powderFinetoactive carbon
increase the powder (50–150
laser absorptivity. with the mass fraction of 0.1–0.3%
powder to increase the laser absorptivity.
into silica powder to increase the laser absorptivity.

Figure 2. Morphology of original silica (SiO2) powder.

Figure 2. Morphology of original silica (SiO2) powder.
Figure 2. Morphology of original silica (SiO2 ) powder.
Materials 2017, 10, 1313 4 of 11
Materials 2017, 10, 1313 4 of 11
Materials 2017, 10, 1313 4 of 11

Figure 3. Particle size distribution of the spherical silica powder.

Moreover, particles Figure

Figure with3. Particle
3. a smaller
Particle size
size distribution
size are more
distribution of the
of theeasily
spherical
spherical silica powder.
powder. [23]. The composite
melted/reacted
silica
particles were mechanically mixed with Ball Mill (from Planetary Mono Mill, FRITSCH GmbH,
Moreover,
Moreover,particles
Idar-Oberstein, particleswith
Germany) at a150
with smaller
a rpm size10
smaller
for areh.more
size easily easily
are powder
The more melted/reacted
dried in a[23].
was melted/reacted The
drying composite
[23].
oven The
at 80 particles
composite
°C for 2
were mechanically
hparticles
before the were mixed
SLSmechanically
process. with Ball Mill (from Planetary Mono Mill, FRITSCH
mixed with Ball Mill (from Planetary Mono Mill, FRITSCH GmbH, GmbH, Idar-Oberstein,
Germany) at 150 rpm forused
10 h.atin
The powder forwas dried ◦in
Idar-Oberstein,
The SLS Germany)
machine 150
the rpm
present 10work
h. Theisinapowder
aself-developed
dryingwasovendried
at 80
SLS Csystem.
afor 2 h before
drying Itoven the
atSLS
is equipped process.
80 °C for 2
with
h The
before SLS
the machine
SLS process. used in the present work is a self-developed
a continuous fiber laser (Pmax = 200 W.) with a wavelength of 1070 nm and a laser beam focusing SLS system. It is equipped with
a continuous
The of
diameter SLS~100fiber
machine
μm. laser is(Pmax
It used in the
comprised = 200 ofW.)
present with
a 100 work a is
× 100 wavelength
×a 150 mm3 of
self-developed 1070envelope,
build nmsystem.
SLS and aandlaser is beam
It the equipped
powder focusing
with
layer
diameter of ~100 It is comprised of a 100 × 100 × 150 mm 3 build envelope, and the powder layer
a continuous fiber
µm. laser (Pmax = 200 W.) with a
was deposited by blade. The whole process was conducted under air atmosphere. wavelength of 1070 nm and a laser beam focusing
was deposited
diameter
The of ~100
main byμm.blade.It isThe
parameters whole
comprised
for theprocess
of
SLSa 100 was
× 100conducted
processing × 150aremm under
3 build
illustratedairenvelope,
atmosphere.
in Figure and4.the powder layer
Optimized SLS
was The main
deposited parameters
by blade. for
The the
whole SLS processing
process was are illustrated
conducted in
under
parameters shown in Table 1 were applied to fabricate green specimens with a setup dimension of Figure
air 4. Optimized
atmosphere. SLS parameters
8shown The
× 8 mm main
in2 Table
for 1parameters
were applied
investigations inforto the
this SLS green
fabricate
study. processing
specimens are illustrated
with a setup indimension 8 × 8 mm2SLS
Figure 4. ofOptimized for
parameters
investigations
In order shown
in thisin Table
to increase study. the1 were
bonding applied
of the to fabricate
SLS parts, green
furnacespecimens
sintering with a setup
was applied dimension
in a highof
8 × 8 mm
In 2 forto
order investigations
increase the in
bondingthis study.
of the SLS parts, furnace sintering
temperature chamber furnace (Carbolite HTF 17/5/3216P1, Chelmsford, UK). The parts were heated was applied in a high temperature
In order to(Carbolite
increase the ◦ C/min
chamber
at 2 °C/minfurnace
in air to 1200 °C HTF forbonding
5 h. The of
17/5/3216P1, wholethe SLS parts,
Chelmsford,
furnace furnace
UK).
sintering sintering
Theprocess
parts were wasmore
heated
is much applied in a in
high
at 2 cost-effectiveair
◦
temperature for 5chamber furnace (Carbolite HTF
to
and 1200
lessCtime-consuming
h. The whole furnace
owing tosintering
omission of17/5/3216P1,
process
theisde-binding
much moreChelmsford, UK).and
cost-effective
process, becauseThelessparts
the rawwere heated
time-consuming
material is
at 2
owing°C/min
to in
omissionair
free of polymer binder. to
of 1200
the °C for
de-binding 5 h. The whole
process, furnace
because the sintering
raw materialprocess
is freeis much
of polymermore cost-effective
binder.
and less time-consuming owing to omission of the de-binding process, because the raw material is
free of polymer binder.

Figure 4. Key parameters in the SLS process.

Figure 4. Key parameters in the SLS process.
Figure 4. Key parameters in the SLS process.
Materials 2017, 10, 1313 5 of 11
Materials 2017, 10, 1313 5 of 11

2.3.Measurements
2.3. Measurements
The morphology of the specimens was observed by an optical microscope (OLYMPUS SZX10,
The morphology of the specimens was observed by an optical microscope (OLYMPUS SZX10,
Olympus Corporation, Tokyo, Japan) and a scanning electron microscope (SEM, JEOL JSM5510LV,
Olympus Corporation, Tokyo, Japan) and a scanning electron microscope (SEM, JEOL JSM5510LV,
Tokyo, Japan) equipped with energy dispersive X-ray spectroscopy (EDX, Oxford 7582, Oxfordshire,
Tokyo, Japan) equipped with energy dispersive X-ray spectroscopy (EDX, Oxford 7582, Oxfordshire,
UK). The crystal structure was characterized by X-ray diffraction (XRD-6000 Cu-Ka radiation,
UK). The crystal structure was characterized by X-ray diffraction (XRD-6000 Cu-Ka radiation, Shimadzu,
Shimadzu, Tokyo, Japan). Thermogravimetric analysis and differential thermal analysis (DTA)
Tokyo, Japan). Thermogravimetric analysis and differential thermal analysis (DTA) (DTA-TG, Shimadzu
(DTA-TG, Shimadzu DTG-60H, Shima, Tokyo, Japan) were used in the range from room temperature
DTG-60H, Shima, Tokyo, Japan) were used in the range from room temperature to 1400 ◦ C in a flowing
to 1400 °C in a flowing air atmosphere with a heating rate of 5 °C/min. The UV-vis-NIR spectra
air atmosphere with a heating rate of 5 ◦ C/min. The UV-vis-NIR spectra were obtained with a Cary
were obtained with a Cary 5000 UV-vis-NIR spectrophotometer. X-ray computed tomography (CT)
5000 UV-vis-NIR spectrophotometer. X-ray computed tomography (CT) scan of the printed part®was
scan of the printed part was carried out by using a high-resolution XCT system Phoenix Nanotom m
carried out by using a high-resolution XCT system Phoenix Nanotom® m (GE Phoenix, Lewistown,
(GE Phoenix, Lewistown, PA, USA). The density was measured by the Archimedes’ method in
PA, USA). The density was measured by the Archimedes’ method in water. To calculate the amount of
water. To calculate the amount of shrinkage of the SLS parts, the dimension of cubic parts was
shrinkage of the SLS parts, the dimension of cubic parts was measured by digital Vernier caliper.
measured by digital Vernier caliper.
Table 1. Optimized laser sintering parameters.
Table 1. Optimized laser sintering parameters.
Laser Power (W) Hatch Space (µm) Scan Speed (mm/s) Spot Size (µm) Layer Thickness (µm)
Laser Power (W) Hatch Space (µm) Scan Speed (mm/s) Spot Size (µm) Layer Thickness (µm)
60
60 150
150 50
50 120
120 90
90

3.3.Results
Resultsand
and Discussion
Discussion
SEM
SEM micrograph
micrograph and EDX EDX analyses
analysesofofsilica
silicapowder
powder with
with carbon
carbon addition
addition were
were carried
carried out out
as
as a loose
a loose powder
powder bedbed (Figure
(Figure 5). ball
5). The Themill
ballatmill
low at low rotating
rotating speed achieved
speed achieved uniformwithout
uniform mixing mixing
without
crushingcrushing spherical
spherical silica granules
silica granules (Figure
(Figure 5a). The
5a). The overall
overall EDXEDX mapping
mapping of of
thethepowder
powderbedbed
demonstrates that
demonstrates that the
the powder
powder contains
contains primarily
primarily carbon
carbon (C),
(C), silicon
silicon(Si),
(Si),and
andoxygen
oxygen(O).
(O).No
Notraces
traces
ofofother
otherelements
elementswere
were found.
found. This
This confirms that the powder
powder waswas not
not contaminated
contaminatedduring
duringthe
theball
ball
millingprocess.
milling process. The
The distribution
distribution of carbon was
was consistent
consistentwith
withthat
thatof
ofsilicon
siliconand
andoxygen
oxygenelements,
elements,
whichindicates
which indicatesthat
thatCCwas
wasdistributed
distributed uniformly
uniformly on the micron-size particles.
particles.

Figure5.5.Micrographs
Figure Micrographs showing
showing characteristic
characteristic distribution
distributionofofsilica
silicapowder
powderwith
with0.2
0.2wt
wt%%ofofcarbon
carbon
additive. (a) SEM micrograph of mixed powders; The corresponding EDX mapping images of (b) of
additive. (a) SEM micrograph of mixed powders; The corresponding EDX mapping images (b)
silicon;
silicon; (c) carbon; and (d) oxygen
(c) carbon; and (d) oxygen elements. elements.
Materials 2017, 10,
Materials 2017, 10, 1313
1313 66 of
of 11
11

As introduced previously, fine carbon was doped into the silica granules to enhance the
As introduced
absorptivity previously,
to laser beam. The fine carbon wasof
absorptance doped
silicainto the silica
powder granules
measured fortodifferent
enhance the absorptivity
carbon doping
to laser beam. The absorptance of silica powder measured for different
quantities is presented in Figure 6. The UV-vis-NIR absorption spectra was measured on a greencarbon doping quantities is
presented
pellet with inaFigure 6. The
thickness ofUV-vis-NIR
4 mm, which absorption
shows the spectra was
clear measured onof
improvement a green pellet with
absorptance by aincreasing
thickness
of 4 mm, which
amounts shows
of carbon the clear
additive improvement
(Figure of absorptance
6a). Comparison of the by increasing to
absorptance amounts of carbon
fiber laser additive
(λ = 1070 nm)
(Figure 6a). Comparison of the absorptance to fiber laser (λ = 1070 nm) in
in pellets containing different amount of the additive reveals significant changes (Figure 6b).pellets containing different
amount
The of the additive
absorptance at 1070reveals significant changes
nm wavelength of SiO2 (Figure
+ 0.1 wt6b). The
% C, absorptance
SiO 2 + 0.2 wt % at C,
1070
andnmSiO
wavelength
2 + 0.3 wt of
%
SiO
C +
reached
2 0.1 wt %
~4%, C, SiO
~16%, 2 +
and 0.2 wt
~19%, % C, and SiO
respectively. 2 + 0.3
The wt % C reached
absorbance to ~4%,
fiber ~16%,
laser of and
SiO 2~19%,
+ 0.2 respectively.
wt % C was
The
up toabsorbance
eight times tohigher
fiber laser
thanofpure
SiO2silica.
+ 0.2 wt % C was up to eight times higher than pure silica.

Figure 6. Comparison
Figure 6. Comparison of
of absorptance
absorptance to to near-infrared
near-infrared light
light with
with varied
varied amount
amount of
of carbon
carbon additive.
additive.
UV-vis-NIR absorption
(a) UV-vis-NIR absorption spectra;
spectra; (b)
(b) absorptance
absorptance toto UV
UV light
light at
at1070
1070nm
nmwavelength.
wavelength.

The carbon nanoparticles exhibited a high tendency to absorb laser, owing to the fact that they
have aa much
muchhigher
higherlaser
laserabsorption, and
absorption, that
and they
that have
they an extremely
have largelarge
an extremely surface area/volume
surface ratio
area/volume
(i.e.,
ratiobecause of theof
(i.e., because melting point depression
the melting phenomenon)
point depression [25]. The
phenomenon) significant
[25]. benefitbenefit
The significant from doped
from
carbon carbon
doped was revealed in the comparison
was revealed of the SLS
in the comparison ofof silica
the SLSwith different
of silica withcarbon additions
different carbon (Figure 7).
additions
The
(Figureone-layer
7). Thesamples shown
one-layer in Figure
samples 7 werein
shown all Figure
prepared 7 under same
were all process parameters,
prepared under samewhich are
process
parameters, which are listed in Table 1. It is evident that the conditions of powder melting with
Materials 2017, 10, 1313 7 of 11
Materials 2017, 10, 1313 7 of 11

listed
variedincarbon
Table 1. It is evident
addition that the The
are different. conditions
ceramicofpowders
powder withmelting withwere
carbon varied
at carbon addition
least joined are
to form
different. The ceramic powders with carbon were at least joined to form a solid piece
a solid piece with the absorption of laser energy. However, there was almost no trace of melting in with the absorption
of
thelaser
pureenergy.
silica However,
powders there was of
because almost
poornoabsorptivity.
trace of melting
It isinworth
the pure silica powders
mentioning thatbecause of poor
an unsuitable
absorptivity.
amount of the It iscarbon
worth mentioning
additive wouldthat an unsuitable
introduce amount of the
macro-cracks carbon
(Figure additive
7d) due towould
excessintroduce
doping,
macro-cracks (Figure
while insufficient 7d) due
doping to excess
may resultdoping, while insufficient
in uncontentious joiningdoping
(Figuremay result
7b). in uncontentious
In the range of the
joining
studied(Figure
amounts 7b).ofIncarbon
the range of thegood
additive, studied amounts
qualities of carbonand
of melting additive,
bonding good qualities
without of melting
visible cracks
and
werebonding
achievedwithout
by doping visible cracks
with were
0.2 wt % ofachieved
carbon by doping
(Figure 7c).with 0.2 wt % of carbon (Figure 7c).

Figure 7. Optical graphs showing the appearance and cross-section of the SLS sample surface with
Figure 7. Optical graphs showing the appearance and cross-section of the SLS sample surface with
varying the amount of carbon additive. (a) Pure silica; (b) silica + 0.1 wt % carbon; (c) silica + 0.2 wt %
varying the amount of carbon additive. (a) Pure silica; (b) silica + 0.1 wt % carbon; (c) silica + 0.2 wt %
carbon; (d) silica + 0.3 wt % carbon.
carbon; (d) silica + 0.3 wt % carbon.
Materials 2017, 10, 1313 8 of 11
Materials 2017,
Materials 2017, 10,
10, 1313
1313 88 of
of 11
11

Along
Alongwith
Along withdoping
with dopingaaasmall
doping smallamount
small amount of
amount of carbon,
of carbon, the
carbon, the laser
laser absorptivity
laser absorptivityat
absorptivity at1070
at 1070nm
1070 nmwas
nm wasobviously
was obviously
obviously
enhanced. On
enhanced. On
enhanced. the
On the other
the other hand,
other hand, it
hand, it was
it was proven
was proven
proven thatthat no
that no reaction
no reaction with
reaction with active
with active carbon
active carbon occurred,
carbon occurred,
occurred, as as
asshown
shownin
shown
the
in X-ray
in the diffraction
the X-ray
X-ray (XRD)(XRD)
diffraction
diffraction resultsresults
(XRD) (Figure(Figure
results 8). The8).
(Figure X-ray
8). Thediffraction
The pattern pattern
X-ray diffraction
X-ray diffraction revealed that the structures
pattern revealed
revealed that the
that the
structures
are are well
well crystallized
structures are well crystallized
crystallized
into the quartz into phase
into the quartz
the quartz phase
after phase after laser
laser sintering.
after laserThe
sintering. The transformation
transformation
sintering. The transformation
of quartz into of
of
quartz into
cristobalite cristobalite
and tridymide and tridymide
after furnace after furnace
sintering was sintering
also was
analyzed. also analyzed.
There
quartz into cristobalite and tridymide after furnace sintering was also analyzed. There is no carbon is no There
carbon is no
phase carbon
or any
phase or
phase
composite or any
any
phase composite
composite
detectedphasephase
in both detected
detected
green and in both
in both green
finalgreen
ceramic andparts
and finalasceramic
final ceramic parts as
parts
a consequence as of
aa consequence
consequence
carbon addition. of
of
carbonthere
carbon
Hence, addition.
addition. Hence,
is noHence, there
risk ofthere
inducingis no
is noundesirable
risk of
risk of inducing
inducing undesirable
undesirable
contamination contamination
orcontamination
a new phase, or or aa new
which new mayphase,
phase,
have
which may
which
negative may have
have
effects negative
negative
on the effects on
effects
final properties. on The
the carbon
the final properties.
final properties. The carbon
The
may be consumed carbon may
may
by the be consumed
be
reactionconsumed
with theby by the
the
oxygen
reaction
inreaction with
the air, with
which the oxygen
thewill
oxygen in the
in the air,
contribute air,
some which
which will
heatwill energycontribute
contribute some heat
someprocess.
to the SLS energy
heat energy to
Thisto the SLS process.
the SLSbehavior
thermal process. of This
Thisthe
thermal
thermal
carbon behavior
behavior
additive of the
of the carbon
is shown carbon additive
by theadditive
DTA/TGA is shown
is shown
curve by by the DTA/TGA
the
(Figure DTA/TGA
9). curve (Figure
curve (Figure 9).
9).

Figure 8.
Figure
Figure 8.8.XRD
XRD pattern
XRD pattern of
pattern of the
of the as-SLS
the as-SLS green
as-SLS green part
green part and final
and final
partand sintered part.
sinteredpart.
finalsintered part.

Figure9.9.
Figure
Figure 9.Thermogravimetric
Thermogravimetricanalysis
Thermogravimetric analysis
analysis and
and
and differential
differential
differential thermal
thermal
thermal analysis
analysis
analysis (DTA/TGA)
(DTA/TGA)
(DTA/TGA) curve
curve
curve of silica
of
of the the
the
silica powder
silica
powder powder with
with
with 0.2 0.2of
wt 0.2
% wt
wt %% of
of carbon
carbon carbon additive.
additive.
additive.

Goodmelting
Good
Good meltingand
melting andbonding
and bondingisis
bonding isthe
theprerequisite
the prerequisitefor
prerequisite forcreating
for creatingporous
creating porousstructures.
porous structures.One
structures. Oneefficient
One efficientway
efficient wayof
way
of generating
of generating
generating the porous
the
the porousporous part
partpart isenlarge
is tois to enlarge
to enlarge the hatch
the
the hatchhatch distance
distance
distance (H).
(H). (H). In this
In
In this this case,
case,case, the particles
the particles
the particles would would
would
bond
bond
togethertogether
bond together but a void
butcould
but a void a void could
could
be left be left
be left
inside. inside. Therefore,
inside. Therefore,
Therefore, the
the porositythe porosity
canporosity can
canby
be tailored be tailored
bevarying by
tailoredH.by varying
varying
The H.
H.
fabrication
 Materials
Materials 2017,
2017, 10,10, 1313
1313 9 of 11 11
9 of
Materials 2017, 10, 1313 9 of 11

The fabrication of a 3D porous silica part was conducted with optimized parameters and lager
The fabrication of a 3D porous silica part was conducted with optimized parameters and lager
ofhatch
a 3D porous
distancesilica
(300part
μm)was conducted
according with1.optimized
to Table parameters
A porous silica and lager
cubic sample withhatch distance
20 layers (300 10)
(Figure µm)
hatch distance (300 μm) according to Table 1. A porous silica cubic sample with 20 layers (Figure 10)
according to Table 1.
was successfully A porousvia
fabricated silica
SLScubic sample
followed with 20 sintering,
by furnace layers (Figure 10) was
and the successfully
relative density was fabricated
only
was successfully fabricated via SLS followed by furnace sintering, and the relative density was only
via46.8%.
SLS followed by furnace
Firstly, the sintering,
silica sample wasand the relative
evaluated density
by visual was onlysince
inspection 46.8%. it Firstly,
is quick,thesimple,
silica sample
and
46.8%. Firstly, the silica sample was evaluated by visual inspection since it is quick, simple, and
wascheap to perform
evaluated by visualto most specimens
inspection since it[26], and simple,
is quick, the sample showed
and cheap a porous-like
to perform to most and crack-free
specimens [26],
cheap to perform to most specimens [26], and the sample showed a porous-like and crack-free
andsurface.
the Defects
sample such aasporous-like
showed cracks were andnot observed
crack-free in the Defects
surface. cross-section
such ofcracks
as the sample
were (Figure
not 10c). in
observed
surface. Defects such as cracks were not observed in the cross-section of the sample (Figure 10c).
theContinuous
cross-section pores withsample
of the random distribution
(Figure were formed
10c). Continuous between
pores the
withthe particles,
random which bonded
distribution via a
were formed
Continuous pores with random distribution were formed between particles, which bonded via a
neckingthe
between mechanism
particles, (Figure
which 10d). The viasuccessful
a neckingfabrication of silica with tailored porosity fabrication
was also
necking mechanism (Figurebonded
10d). The successful mechanism
fabrication of (Figure 10d).
silica with The successful
tailored porosity was also
demonstrated
of silica in
with tailored the 3D reconstruction from the micro-CT scan (Figure 11). The linear and volume
demonstrated in theporosity was also demonstrated
3D reconstruction from the micro-CTin the 3Dscanreconstruction
(Figure 11). The from the micro-CT
linear and volume scan
shrinkage:
(Figure 11). TheΔL/L, ΔW/W,
linear ΔH/H, and
and volume ΔV/V, were
shrinkage: ∆L/L,measured
∆W/W, ∆H/H, by digital Vernier
∆V/V, were
andVernier caliper, and the
measured
shrinkage: ΔL/L, ΔW/W, ΔH/H, and ΔV/V, were measured by digital caliper, and the by
results
digital were caliper,
Vernier 1.45%, 1.57%, the2.7%,
resultsand 6.75% in 1.57%,
length, width, height, and volume, respectively.
results were 1.45%, and 1.57%, 2.7%, were
and 1.45%,
6.75% in length,2.7%,
width, andheight,
6.75% and
in length,
volume, width, height, and
respectively.
Excellent
volume, bonding
respectively.
Excellent bonding and and low
Excellent shrinkage
bonding and
low shrinkage of the porous
lowporous
of the silica
shrinkage structure
of the
silica were
porouswere
structure obtained
silicaobtained owing
structureowing to the
were obtained
to the
direct
direct SLS with carbon additive and without the use of a polymer binder in laser sintering.
owing toSLSthe with
directcarbon additive
SLS with carbon and without
additive thewithout
and use of athe
polymer
use ofbinder in laser
a polymer bindersintering.
in laser sintering.

Figure
Figure10.10.Silica
Silica part
part produced
produced by
by the SLS + sintering. (a) Top view of the silica
silica cubic;
cubic; (b)
(b) side
side view
view of
of
Figure 10. Silica part produced by the SLS + sintering. (a) Top view of the silica cubic; (b) side view of
the
the silica
silica cubic.
cubic. SEM
SEM micrographs
micrographs of fracture surface of final sintered sample;
sample; (c)(c) lowmagnification;
lowmagnification;
the silica cubic.
(d) SEM micrographs of fracture surface of final sintered sample; (c) lowmagnification;
(d)high
high magnification.
magnification.
(d) high magnification.

Figure 11. Cont.

Materials 2017, 10, 1313 10 of 11
Materials 2017, 10, 1313 10 of 11

Figure
Figure11.11.Image
Imagerenderings
renderingsfrom
frommicro-
micro- computed
computed tomography
tomography (CT) (CT) scans
scansofoffinal
finalporous
poroussilica
silica
specimen
specimenproduced
producedby bySLS.
SLS. (a)
(a) 3D
3D image;
image; (b)
(b) cross-section parallel toto the
cross-section parallel the building
buildingdirection;
direction;
(c)(c)
cross-section
cross-section vertical tothe
vertical to the laser
laser scanning
scanning direction;
direction; (d) cross-section
(d) cross-section parallel toparallel toscanning
the laser the laser
scanning direction.
direction.

4. 4. Conclusions
Conclusions
InIn
thisthis paper,
paper, the capabilities
the capabilities of manufacturing
of manufacturing porous porous
silica silica by means
by means of selective
of selective laser
laser sintering
sintering (SLS) without using a polymer binder have been demonstrated.
(SLS) without using a polymer binder have been demonstrated. The significant improvement of laser The significant
improvement
absorptivity of laser carbon
by doping absorptivity
on thebysilica
doping carbon
material onbeen
has the silica materialFurthermore,
highlighted. has been highlighted.
there was
no trace of undesirable contamination or new phase, which may result in negative effects onresult
Furthermore, there was no trace of undesirable contamination or new phase, which may in
the final
negative effects on the final properties. The well-bonded 3D porous silica structure
properties. The well-bonded 3D porous silica structure with low shrinkage was successfully fabricated with low
shrinkage
without visiblewasdefects.
successfully fabricated
Moreover, without isvisible
this approach defects. Moreover,
also applicable this approach
to the fabrication of otherisporous
also
applicable
ceramic partstothrough
the fabrication of otherThe
laser sintering. porous ceramic
material parts through
processing laser
technique, sintering.
although The material
time-consuming,
processing technique, although time-consuming, makes the direct SLS of porous silica structures
makes the direct SLS of porous silica structures possible.
possible.
Acknowledgments: The work described in this paper was jointly supported by the Overseas Exchange of
Acknowledgments:
NUS-HIT, NUS StrategicThe work described
Research in this
Programme, andpaper
MorganwasAdvanced
jointly supported
Materialsby
Pt the
Ltd.,Overseas Exchange of
Singapore.
NUS-HIT, NUS Strategic Research Programme, and Morgan Advanced Materials Pt Ltd., Singapore
Author Contributions: Shuai Chang and Jerry Ying Hsi Fuh conceived and designed the experiments; Shuai
Chang,
AuthorLiqun Li, and Li Lu
Contributions: analyzed
Shuai Changthe material
and data. Hsi
Jerry Ying Shuai
FuhChang performed
conceived the experiments
and designed and wrote
the experiments;
theShuai
paper.Chang, Liqun Li, and Li Lu analyzed the material data. Shuai Chang performed the experiments and
wrote the
Conflicts of paper.
Interest: The authors declare no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.
References
References
1. Liang, Y.H.; Sourin, P.D. Application trend in advanced ceramic technologies. Technovation 2001, 21, 61–65.
1. [CrossRef]
Liang, Y.H.; Sourin, P.D. Application trend in advanced ceramic technologies. Technovation 2001, 21, 61–65.
2. 2. Ohmura,
Ohmura, A.;A.;
Ouchi,
Ouchi,N.;N.;
Morisaki,
Morisaki, S.;S.;
Watanabe,
Watanabe, C. C.
Functionality
Functionalitydevelopment
development as as
a survival strategy
a survival strategyforfor
fine
ceramics. Technovation
fine ceramics. 2003, 23,
Technovation 2003,833–842. [CrossRef]
23, 833–842.
3. 3. Scheffler,
Scheffler,
M.;M.; Colombo,P.P.
Colombo, CellularCeramics:
Cellular Ceramics:Structure,
Structure, Manufacturing,
Manufacturing, Properties
Propertiesand
andApplications;
Applications;John
JohnWiley
Wiley
& Sons:
& Sons: Hoboken,NJ,
Hoboken, NJ,USA,
USA,2006.
2006.
4. 4. Gauckler,
Gauckler, L.J.;
L.J.; Waeber,
Waeber, M.M.;Conti,
M.M.; Conti,C.; C.;Jacob-Duliere,
Jacob-Duliere,M. M. Ceramic
Ceramic foam forfor molten
moltenmetal
metalfiltration.
filtration.JOM
JOMJ. J.
Miner.
Miner. Met.
Met. Mater.Soc.
Mater. Soc.1985,
1985,37,
37,47–50.
47–50.[CrossRef]
5. 5. Tang,
Tang, H.H.;
H.H.; Chiu,Chiu, M.L.;
M.L.; Yen,Yen,
H.C. H.C. Slurry-based
Slurry-based selective
selective laser sintering
laser sintering of polymer-coated
of polymer-coated ceramicceramic
powders
powders to fabricate high strength alumina parts. J. Eur. Ceram. Soc. 2011,
to fabricate high strength alumina parts. J. Eur. Ceram. Soc. 2011, 31, 1383–1388. [CrossRef] 31, 1383–1388.
6. 6. Wu,Wu,H.;H.;
Li,Li,
D.;D.;Tang,
Tang,Y.;Y.; Sun,B.;B.;Xu,
Sun, Xu,D.D.Rapid
Rapidfabrication
fabrication of
of alumina-based
alumina-based ceramic
ceramiccores
coresfor
forgas
gasturbine
turbine
blades by stereolithography and gelcasting. J. Mater. Process. Technol. 2009, 209, 5886–5891.
blades by stereolithography and gelcasting. J. Mater. Process. Technol. 2009, 209, 5886–5891. [CrossRef]
7. 7. Travitzky,
Travitzky,N.N. Processing
Processing ofof ceramic-metalcomposites.
ceramic-metal composites.Adv.Adv. Appl.
Appl. Ceram.
Ceram. 2012,
2012, 111,
111,286–300.
286–300. [CrossRef]
8. 8. Kruth, J.P.; Leu, M.C.; Nakagawa, T. Progress in additive manufacturing
Kruth, J.P.; Leu, M.C.; Nakagawa, T. Progress in additive manufacturing and rapid prototyping. and rapid prototyping.CIRPCIRP
Ann.
Ann. Manuf. Technol. 1998, 47, 525–540.
Manuf. Technol. 1998, 47, 525–540. [CrossRef]
Materials 2017, 10, 1313 11 of 11

9. Bertrand, P.; Bayle, F.; Combe, C.; Gœuriot, P.; Smurov, I. Ceramic components manufacturing by selective
laser sintering. Appl. Surf Sci. 2007, 254, 989–992. [CrossRef]
10. Shishkovsky, I.; Yadroitsev, I.; Bertrand, P.; Smurov, I. Alumina-zirconium ceramics synthesis by selective
laser sintering/melting. Appl. Surf. Sci. 2007, 254, 966–970. [CrossRef]
11. Tolochko, N.K.; Khlopkov, Y.V.; Mozzharov, S.E.; Ignatiev, M.B.; Laoui, T.; Titov, V.I. Absorptance of powder
materials suitable for laser sintering. Rapid Prototyp. J. 2000, 6, 155–161. [CrossRef]
12. Kruth, J.P.; Levy, G.; Klocke, F.; Childs, T.H.C. Consolidation phenomena in laser and powder-bed based
layered manufacturing. CIRP Ann. Manuf. Technol. 2007, 56, 730–759. [CrossRef]
13. Gan, M.X.; Wong, C.H. Properties of selective laser melted spodumene glass-ceramic. J. Eur. Ceram. Soc.
2017, 30, 4147–4154. [CrossRef]
14. Gahler, A.; Heinrich, J.G.; Guenster, J. Direct Laser Sintering of Al2 O3 -SiO2 Dental Ceramic Components by
Layer-Wise Slurry Deposition. J. Am. Ceram. Soc. 2006, 89, 3076–3080. [CrossRef]
15. Sing, S.L.; Yeong, W.Y.; Yeong, W.Y.; Wiria, F.E.; Zhao, L. Direct selective laser sintering and melting of
ceramics: A review. Rapid Prototyp. J. 2017, 23, 611–623. [CrossRef]
16. Deckers, J.P.; Deckers, J.P.; Shahzad, K.; Shahzad, K.; Cardon, L.; Kruth, J.P. Shaping ceramics through
indirect selective laser sintering. Rapid Prototyp. J. 2016, 22, 544–558. [CrossRef]
17. Qian, B.; Shen, Z.J. Laser sintering of ceramics. Asian Ceram. Soc. 2003, 1, 315–321. [CrossRef]
18. Juste, E.; Petit, F.; Lardot, V.; Cambier, F. Shaping of ceramic parts by selective laser melting of powder bed.
J. Mater. Res. 2014, 29, 2086–2094. [CrossRef]
19. Duley, W. Laser Processing and Analysis of Materials; Springer Science & Business Media: Berlin, Germany,
2012.
20. Beaman, J.J.; Deckard, C.R. Selective Laser Sintering with Assisted Powder Handling. U.S. Patent US4938816 A,
3 July 1990.
21. Ko, S.H.; Pan, H.; Grigoropoulos, C.P.; Luscombe, C.K.; Fréchet, J.M.; Poulikakos, D. All-inkjet-printed
flexible electronics fabrication on a polymer substrate by low-temperature high-resolution selective laser
sintering of metal nanoparticles. Nanotechnology 2007, 18, 345. [CrossRef]
22. Kruth, J.P.; Vandenbroucke, B.; Vaerenbergh, V.J.; Naert, I. Rapid manufacturing of dental prostheses by
means of selective laser sintering/melting. In Proceedings of the Les 11emes Assises Europeennes du
Prototypage Rapide (AFPR), Paris, France, 4–5 October 2005.
23. Sutton, A.T.; Kriewall, C.S.; Leu, M.C.; Newkirk, J.W. Powder characterisation techniques and effects of
powder characteristics on part properties in powder-bed fusion processes. Virtual Phys. Prototyp. 2017, 12,
3–29. [CrossRef]
24. Tan, J.H.; Wong, W.L.E.; Dalgarno, K.W. An Overview of Powder Granulometry on Feedstock and Part
Performance in the Selective Laser Melting Process. Addit. Manuf. 2017, 18, 228–255. [CrossRef]
25. AlMangour, B.; Grzesiak, D.; Yang, J.M. Rapid fabrication of bulk-form TiB2 /316L stainless steel
nanocomposites with novel reinforcement architecture and improved performance by selective laser melting.
J. Alloys Compd. 2016, 680, 480–493. [CrossRef]
26. Lu, Q.Y.; Wong, C.H. Applications of non-destructive testing techniques for post-process control of additively
manufactured parts. Virtual Phys. Prototyp. 2017, 12, 301–321. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).