polymers

Article
Design and Synthesis of Free-Radical/Cationic
Photosensitive Resin Applied for 3D Printer with
Liquid Crystal Display (LCD) Irradiation
Junyang Shan 1 , Zijun Yang 1 , Guoguang Chen 1 , Yang Hu 1 , Ying Luo 1 , Xianming Dong 1,2 ,
Wenxu Zheng 1,2, * and Wuyi Zhou 1,2, *
1 Research Center of Biomass 3D Printing Materials, College of Materials and Energy,
South China Agricultural University, Guangzhou 510642, China; [email protected] (J.S.);
[email protected] (Z.Y.); [email protected] (G.C.); [email protected] (Y.H.);
[email protected] (Y.L.); [email protected] (X.D.)
2 Key Laboratory for Biobased Materials and Energy of Ministry of Education,
College of Materials and Energy, South China Agricultural University, Guangzhou 510642, China
* Correspondence: [email protected] (W.Z.); [email protected] (W.Z.)




Received: 5 May 2020; Accepted: 10 June 2020; Published: 15 June 2020


Abstract: In this work, aiming at a UV-curing 3D printing process with liquid crystal display
(LCD) irradiation, a novel free-radical/cationic hybrid photosensitive resin was designed and
prepared. After testing, the results showed that the acrylate monomers could be polymerized through
a free-radical mechanism, while the epoxides were polymerized by a cationic curing mechanism.
During the process of UV-curing, the acrylate and epoxide polymers were crosslinked and further
locked together by non-covalent bonds. Therefore, an interpenetrating polymer network (IPN)
structure could be formed through light-curing 3D-printing processes. Fourier transform infrared
spectroscopy (FT-IR) revealed that the 3,4-epoxy cyclohexyl methyl-3,4-epoxy cyclohexyl formate and
acrylic resin were both successfully involved in the UV-curing process. Furthermore, in order to make
the 3D-printed objects cured completely, post-processing was of great importance. The results from the
systematic study of the dynamic mechanical properties of the printed objects showed that the heating
treatment process after UV irradiation was very necessary and favorable for the complete cationic
polymerization of UV-6110 induced by Irgacure 261. The optimum heating treatment conditions were
achieved at a temperature of 70 ◦ C for 3 h.

Keywords: 3D printing; photosensitive resin; liquid crystal display irradiation; UV-curing

1. Introduction
To date, as an additive manufacturing technology, 3D printing has been developed to build many
3D objects by a layer-by-layer printing process using some special materials, such as micron-sized
powdery metals, thermoplastic or liquid photosensitive resins, and other adhesive materials [1–5].
Compared with the conventional subtractive manufacturing process, 3D printing has many advantages
including a short manufacturing cycle, the capability of fabricating complex models [6], and reduced
material waste. Thus, 3D printing has been applied to a wide range of fields, involving tissue engineering,
the digitized manufacturing of metallic components, energy materials, and food processing. To date,
3D printing technologies have been developed rapidly and become a research hotspot in many countries
all over the world [7–11]. Among all kinds of 3D printing technology, the stereolithography apparatus
(SLA) rapid prototyping method is a typical 3D printing technology with a special control system
using an ultraviolet (UV) laser beam to scan photosensitive polymers according to the profile data of
each cross-section [10–13]. In a local region, the photosensitive resin absorbs the energy from UV light

Polymers 2020, 12, 1346; doi:10.3390/polym12061346 www.mdpi.com/journal/polymers

Polymers 2020, 12, 1346 2 of 14

with special wavelengths and is then cured rapidly by a photo-polymerization process, and finally
forms a thin layer [14]. The above process is repeated until the designed objects are completely formed.
Photosensitive resins are those materials that can be cured quickly under the excitation of UV
light [15,16]. In general, the resins include reactive oligomers, reactive diluents such as reactive
monomers, photoinitiators, filler pigments, and other auxiliaries. However, most of those compositions
have strong odors and certain toxicities. In practical applications, the products need to be non-toxic
and have no stimulative effect, with low odor. Therefore, there are few photosensitive resins that can
be used for 3D printing. Furthermore, most of them are initiated by free radicals [17], which will
lead to the hybrid resins being hard and brittle after curing, resulting in poor bending and stretching
properties, severely limiting the range of application [18,19]. Used as the diluent of acrylic resin,
N-acryloyl-morpholine (ACMO) is very stable with typically low irritative, corrosive, and toxic
properties. As a key monomer, it can greatly enhance the mechanical performance of the cured
resin. Furthermore, it has a rapid reaction rate and good flexibility. Polyurethane acrylate (PUA) is
an oligomer with high wearability and good weather resistance properties [20–22]. During the UV
light-curing process, the long-chain PUA can react with the ACMO to form a flexible spline owing
to the low degree of cross-linking. After the 3D printing process, the conversion of reactive groups
is usually incomplete, so the internal part of the sample is not cured completely. The curing rate of
photosensitive resin is about 65–90 percent along with the incorporation of epoxy [23]. In order to
complete the reaction, heat, ultrasonic energy, and UV light can be used in post processing [24,25].
The mechanical properties such as tensile strength and hardness can be effectively improved to a certain
extent after post-curing treatment.
At present, the thermal conductivity for the most of photosensitive resins is not uniform. In the
post processing, the internal part of the products remains cool when the temperature is heated above
its glass transition point, resulting in differential thermal strain and further cracking and warping [23].
In order to solve this issue, we explored the influence of the post-treatment (hot) conditions
on the mechanical properties of 3D printing products formed by free-radical/cationic hybrid
photosensitive resins.

2. Materials and Methods

2.1. Materials
Acryl-morpholine (ACMO) was purchased from the Tianjin Jiuri Chemical Company (Tianjin, China).
3,4-epoxy cyclohexyl methyl-3,4-epoxy cyclohexyl formate (UV-6110) was purchased from the
GuangZhou Kute Chemical Company (Guangzhou, China). Long-chain PUA and Antioxidant
was donated by the Taiwan Changxing Chemical company (Taiwan, China). Phenyl bis(2,4,6-
trimethylbenzoyl)-phosphine oxide (819) and (cumene) cyclopentadienyliron (ii) hexa-fluorophos (261)
was obtained from the Changzhou Qiangli Chemical Company (Changzhou, China). Defoamer was
donated by Byk (Wesel, Germany).

2.2. Preparation of Photosensitive Resins and 3D Printing Tests

After studying the tensile strength and elongation properties of the solidified resin samples by
using varying ratios of PUA and ACMO, the optimal ratio of PUA to ACMO was confirmed as 3:7.
The PUA/ACMO/UV-6110 hybrid resin was prepared by mixing different concentrations of PUA,
ACMO, UV-6110, the free-radical photoinitiator Irgacure 819, and cationic photoinitiator Irgacure 261.
The obtained photosensitive resins were stirred in a 3-neck flask at 50 ◦ C for 30 min. A desk 3D printer
with liquid crystal display (LCD) irradiation was used to print some designed objects for the evaluation
of the mechanical properties of the resin. Three samples were designed and printed simultaneously
with an inclination of 0◦ with respect to the printing platform. After printing, the formed samples were
rinsed with ethanol in two successive baths for 15 min and then placed to dry. After that, the 3D-printed
objects were post-cured in a dark vacuum oven at different temperatures. The heating treatment time
Polymers 2020, 12, 1346 3 of 14

and temperature for the post-curing process were set as 150 ◦ C and 1 h, 100 ◦ C and 1 h, 100 ◦ C and 2 h,
80 ◦ C and 1 h, 70 ◦ C and 1 h, 70 ◦ C and 2 h, and 70 ◦ C and 3 h, respectively.

2.3. Soxhlet Extraction Process

The gel content of the newly different 3D-printed samples, such as 100 ◦ C 1 h, 70 ◦ C 2 h, and 70 ◦ C
3 h, was determined gravimetrically by five Soxhlet extraction cycles with the use of boiling THF
(Tetrahydrofuran) as solvent (~5 h). Approximately, 0.1 g of 3D-printed sample was cut into small
pieces and placed in a cellulose thimble. After that, the samples were dried in a vacuum oven at 50 ◦ C
till a constant weight was reached. The gel fraction was calculated using Equation (1), where W0 is
the initial weight of the sample and W1 is the weight of the sample after extraction. For each sample,
the gel fraction determination was repeated twice.

Gel Fraction (wt%) = (W1/W0) × 100 (1)

2.4. Characterization of the Samples

The thermal properties of the liquid photosensitive resins and new 3D-printed samples treated as
100 ◦C for 1 h, 70 ◦ C for 2 h, and 70 ◦ C for 3 h, respectively, were measured with a MKIII Analyser
of TA (New Castle, DE, USA). The loss and storage modulus and the loss tangent (tanδ, ratio of loss
to storage modulus) were recorded as a function of the temperature. For these samples, the glass
transition temperature was evaluated as the inflection point in the DSC curve observed. Of the samples,
5~15 mg of samples before or after post-curing process were put into different aluminum pans under
a constant nitrogen flow of 50 mL min−1 . All of the samples were heated from 0 ◦ C to 130 ◦ C with
a heating rate of 5 k/min (frequency of 1 Hz) by using an extension mode. The thermal stability of those
samples was investigated by thermogravimetric analysis (TGA). The decomposition temperature of the
UV-cured test sample was measured with a TA-5000 Ir thermal analyzer at a heating rate of 10 ◦ C/min.
The heating temperature was set in a range from 25 to 600 ◦ C with a heating rate of 10 ◦ C·min−1 .
All the samples were heated and treated in nitrogen with a flow of 50 mL min−1 for non-oxidative
degradation. The mechanical properties were characterized with a SHIMADZU universal testing
machine. The dumbbell-shaped samples with a standard size of 150 mm × 10 mm × 4 mm were printed
by a 3D printer with LCD light irradiation (LD-001, Shenzhen Chuangxiang 3D technology co. LTD).
After that, the tensile strength of those samples was tested at a 5 mm min−1 stretching rate according to
the national standard of GB 1040-2006. The rectangular samples with a size of 80 mm × 10 mm × 4 mm
were printed for flexural tests with a deformation speed of 20 mm/min according to the national
standard of GB 9341-2008. The fracture surface morphology of the printed objects was characterized
by scanning electron microscopy (Hitachi S 4800). The acrylates part of the cured hybrid resins was
selectively removed by using an alkaline solution of 10% KOH in methanol. The surface static contact
angles of the samples were measured with an OCA 20 (Dataphysics, Stutt-gart, Germany) with the use
of a 5 µL distilled water droplet placed automatically on the cured samples. The volume shrinkage
of the samples was measured by reflective laser beam scanning. The Fourier transform infrared
spectrophotometer (FT-IR) spectra of the UV-cured resins were recorded by using a Nicolet 5700
spectrometer with a range from 4000 to 650 cm−1 and a total of 64 scans at a resolution of 1 cm−1 in the
transmission mode.

3. Results and Discussions

3.1. Mechanism of UV-Curing and LCD 3D Printing

In general, free-radical polymerization is made up of four continuous processes involving initiation,
propagation, chain transfer, and termination. In the initiation process, the photo-initiator can transform
from the ground state to the excited state after absorbing UV radiation, thus producing a chemical
change to form some reactive intermediates of radicals, which can initiate the next polymerization
 3.1. Mechanism of UV-Curing and LCD 3D Printing
In general, free-radical polymerization is made up of four continuous processes involving
initiation, propagation, chain transfer, and termination. In the initiation process, the photo-initiator
can2020,
Polymers transform
12, 1346 from the ground state to the excited state after absorbing UV radiation, 4 of 14thus
producing a chemical change to form some reactive intermediates of radicals, which can initiate the
next polymerization between radicals and molecules of monomers or oligomers (e.g., with
between radicals and molecules of monomers or oligomers (e.g., with acrylates), forming reactive
acrylates), forming reactive alkylmonomers. Those free-radical monomers with high reactivity can
alkylmonomers. Those free-radical monomers with high reactivity can further react with other
further react with other monomers or oligomers within the process of propagation. Chain transfer is
monomers or oligomers within the process of propagation. Chain transfer is a reaction process in
a reaction process in which the reactive center interacts with a monomer or oligomer to produce
which the reactive center interacts with a monomer or oligomer to produce another new free radical,
another new free radical, which implies that the initial growth of the polymer radical has been
which implies that the initial growth of the polymer radical has been terminated. The termination
terminated. The termination reaction is mainly completed by the interaction of two radicals, leading
reaction is mainly completed by the interaction of two radicals, leading to the disappearance of the
to the disappearance of the reactive center and the formation of stable polymers. One way of
reactive center and the formation of stable polymers. One way of termination is the binding of two
termination is the binding of two radicals in the active chain, and the other way happens through the
radicals in the active chain, and the other way happens through the disproportionation for bimolecular
disproportionation for bimolecular termination. Sometimes, the two ways happen simultaneously.
termination. Sometimes, the two ways happen simultaneously. The polymerization process ends when
The polymerization process ends when all the chemicals have reacted each other completely. The
all the chemicals have reacted each other completely. The obvious fact is shown that the increase in the
obvious fact is shown that the increase in the viscosity of the resins will hinder the movement of the
viscosity of the resins will hinder the movement of the molecular chains in spite of the presence of
molecular chains in spite of the presence of unreacted compounds [26]. In general, the cationic
unreacted compounds [26]. In general, the cationic polymerization process also contains four steps.
polymerization process also contains four steps. Differently from in the free-radical polymerization
Differently from in the free-radical polymerization process, the cations act as the reactive center in the
process, the cations act as the reactive center in the cationic polymerization. During UV light
cationic polymerization. During UV light initiation, the cationic initiator can transform into an excited
initiation, the cationic initiator can transform into an excited state, forming a new reactive center. For
state, forming a new reactive center. For example, when using aryl-ferrocenium salts such as (cumene)
example, when using aryl-ferrocenium salts such as (cumene) cyclopentadienyliron(ii)
cyclopentadienyliron(ii) hexa-fluorophos (Irgacure 261), the salt can be decomposed into cumene and
hexa-fluorophos (Irgacure 261), the salt can be decomposed into cumene and ferrocene lewis acid
ferrocene lewis acid under UV light irradiation. In terms of the hydrogen donor produced from the
under UV light irradiation. In terms of the hydrogen donor produced from the polyethers
polyethers commonly used in the hybrid photosensitive resin, the aryl-ferrocenium cations produce
commonly used in the hybrid photosensitive resin, the aryl-ferrocenium cations produce protons
protons combined with the anion and offer Brönsted acids, which can initiate the ring-opening reaction
combined with the anion and offer Brönsted acids, which can initiate the ring-opening reaction of
of epoxides. After reaction with Bronsted acids, the epoxide ring is opened and carbonium cations are
epoxides. After reaction with Bronsted acids, the epoxide ring is opened and carbonium cations are
produced, which can initiate the opening of other epoxide groups and lead to chain propagation [26].
produced, which can initiate the opening of other epoxide groups and lead to chain propagation
The depicted mechanism of the cationic polymerization process is illustrated in Scheme 1.
[26]. The depicted mechanism of the cationic polymerization process is illustrated in Scheme 1.

PF6
PF6
Fe hv
+
Fe

Lewis acid

PF6
PF6
PF6 hv O R
Fe Fe
O
Fe O O
R
R 3 R 2

Scheme 1. Photo
Scheme decomposition
1. Photo of iodonium
decomposition salt. salt.
of iodonium

The super acids produced by the photolysis of cationic photoinitiators can still initiate cationic
The super acids produced by the photolysis of cationic photoinitiators can still initiate cationic
polymerization after UV light irradiation or heat treatment. Therefore, the synthetic photosensitive
polymerization after UV light irradiation or heat treatment. Therefore, the synthetic photosensitive
resin with the use of Irgacure 261 can form an interpenetrating polymer network (IPN) structure
through a post-curing process [27].
According to the imaging principle of a photosensitive 3D printer with LCD light irradiation,
the image signals are provided by a computer program and displayed on a screen driving circuit [28].
Some transparent areas are selected and appear on the screen before irradiation. Under the irradiation
 According to the imaging principle of a photosensitive 3D printer with LCD light irradiation,
the image signals are provided by a computer program and displayed on a screen driving circuit
[28]. Some transparent areas are selected and appear on the screen before irradiation. Under the
irradiation
Polymers 2020, 12,with
1346the LCD lights, the transparent region of the images displayed on the liquid crystal 5 of 14
screen reduces the barriers to the LCD lights. However, in those regions without image display, the
UV lights are blocked. The UV lights transmit through the LCD screen to form an area with images
with thelights.
of UV LCD lights,
Therethe is atransparent
liquid tankregion
locatedof the images
on the displayed
surface of the on the liquid
liquid crystalcrystal
screen.screen reduces
A transparent
the barriers to the LCD lights. However, in those regions without image
film is set at the bottom of the liquid tank. The LCD lights radiate directly to these liquid display, the UV lights are
blocked. The UVresins
photosensitive lightsthrough
transmitathrough the LCD
transparent filmscreen
so thattothose
form an areacan
resins withbeimages of UV lights.
polymerized by the
There is a liquid tank located on the surface of the liquid crystal screen.
curing reaction and then solidified [29]. The opaque part of the LCD screen blocks the UV A transparent filmlights,
is setand
at
the bottom of the liquid tank. The LCD lights radiate directly to these liquid
those resins without UV light irradiation still remain in the liquid state in the printing process. The photosensitive resins
through
display aaccuracy
transparent of thefilm
LCDso screen
that those resins can
is generally be The
high. polymerized
resolutionby ofthe
thecuring reaction
3D printer that weandused
thenis
solidified
1280 × 768 pixels, and the dot matrix accuracy of the inch splay is 0.16 × 0.16 mm, which meansUV
[29]. The opaque part of the LCD screen blocks the UV lights, and those resins without that
light irradiation still remain in the liquid state in the printing process. The
the size accuracy of the product formed through the LCD can reach 0.16 mm. The liquid crystal display accuracy of the LCD
screen
screenis images
generally high.direct
from The resolution
molding have of thea3D lowprinter that we used
deformation rate isfor
1280 × 768 pixels,
graphics and the
and high dot
product
matrix accuracy of the inch splay is 0.16 × 0.16 mm, which means that the
accuracy. Typical cationic photopolymers, such as epoxides or oxetanes, are insensitive to oxygensize accuracy of the product
formed
but have through the LCD canofreach
the disadvantage a low0.16 mm.rate,
curing Theresulting
liquid crystal
in somescreen imagesepoxies
unreacted from direct moldingin
still existing
have a low deformation rate for graphics and high product accuracy. Typical cationic
the printed objects. It was difficult to form a perfect IPN structure except by a post-curing process, photopolymers,
such
suchasasepoxides or oxetanes,
heat treatment are insensitive
as shown in Schemeto2.oxygen but have the disadvantage of a low curing rate,
resulting in some unreacted epoxies still existing in the printed objects. It was difficult to form a perfect
IPN structure except by a post-curing process, such as heat treatment as shown in Scheme 2.

Scheme 2. Formation of cross-linked polymeric networks through the UV-curing and heat treatment
Scheme
process for2.free-radical/cational
Formation of cross-linked
hybrid polymeric networks
photosensitive through the UV-curing and heat treatment
resins.
process for free-radical/cational hybrid photosensitive resins.
3.2. FT-IR Analysis
Figure 1 shows the FTIR spectra of the cationic/radical initiated resin systems with different contents
of epoxy under UV irradiation by a desk 3D printer with an LCD light source. For the uncured hybrid
3.2. FT-IR Analysis
resins, the absorption features at 1634 cm−1 and 960 cm−1 correspond to the acrylates. The absorption
featureFigure
at 9141 cm −1 corresponds to the epoxy groups of 3,4-epoxy cyclohexyl methyl-3,4-epoxy
shows the FTIR spectra of the cationic/radical initiated resin systems with different
cyclohexyl
contents offormate.
epoxyItunder
could UVbe observed
irradiation frombythe FTIR spectra
a desk 3D printer that with
the absorption features
an LCD light of acrylates
source. For the
disappeared afterresins,
3D printing. The absorption peak −1
uncured hybrid the absorption features at 1634atcm
1634and
−1 cm960disappeared
cm correspond
−1 in spectra b and e,
to the acrylates.
which indicated that
The absorption the double
feature at 914bonds
cm−1 reacted completely
corresponds to theduring
epoxythe UV-curing
groups of the 3Dcyclohexyl
of 3,4-epoxy printing
process. At the beginning
methyl-3,4-epoxy of the
cyclohexyl UV irradiation,
formate. It could bethe absorption
observed fromfeatures
the FTIRofspectra
epoxy that
groupsthe were also
absorption
decreased toacrylates
features of some extent. It could be
disappeared concluded
after fromThe
3D printing. theabsorption
gradually decreasing
peak at 1634intensity of the peakin
cm disappeared
−1

located
spectraatb914
andcm e,−1 in spectra
which b to ethat
indicated thatthe
thedouble
UV-6110 had completed
bonds the ring-opening
reacted completely during thepolymerization
UV-curing of
successfully. The process.
strongestAtpeaks located at of 1720 −1 were attributed to the stretching vibration
the 3D printing the beginning thecm
UV irradiation, the absorption features of epoxy
ofgroups
the carbonyl
were also groups in ester
decreased to and
somecarboxylic acid. In
extent. It could bethe present from
concluded system,
the acrylate
gradually resins were
decreasing
polymerized
intensity ofthrough
the peak free-radical
located atreactions
914 cm−1while epoxides
in spectra b were polymerized
to e that by cationic
the UV-6110 mechanisms.
had completed the
Additionally, the comparison of the FTIR spectra between the cured hybrid photopolymers and the
uncured acrylate or epoxy polymers showed that there was no new absorption peak appearing in the
cured hybrid photopolymers, which meant that no reaction happened between the acrylate resin and
epoxy resin. In terms of those UV light-cured hybrid resins, acrylate and epoxy polymers were locked
together by cross-linking in a non-covalent manner. In the end, an IPN structure could be formed
in the 3D-printed objects. As we know, the acrylate monomers were polymerized faster and more
strongly than the epoxy monomers [30], which limited the diffusion ability of the epoxy molecules in
photopolymers and the uncured acrylate or epoxy polymers showed that there was no new
absorption peak appearing in the cured hybrid photopolymers, which meant that no reaction
happened between the acrylate resin and epoxy resin. In terms of those UV light-cured hybrid
resins, acrylate and epoxy polymers were locked together by cross-linking in a non-covalent
Polymers
manner. In 12,
2020, the1346
end, an IPN structure could be formed in the 3D-printed objects. As we know, 6 ofthe
14

acrylate monomers were polymerized faster and more strongly than the epoxy monomers [30],
which
the limited
highly the diffusion
cross-linked ability
polymeric of theAepoxy
network. molecules
small part in the highly
of the uncured cross-linked
epoxy resin polymeric
did not participate
network. A small part of the uncured epoxy resin did not participate in the UV-curing
in the UV-curing process. Therefore, in order to make the samples cured completely, a post-curing process.
Therefore,
process wasinnecessary
order to make the 3D-printed
for those samples cured completely, a post-curing process was necessary for
objects.
those 3D-printed objects.

Transmittance（%） d

914cm-1
1634cm-1 960cm-1

1800 1600 1400 1200 1000 800 600

-1
Wavenumber（cm ）

Figure 1. FT-IR spectra of the liquid resin and UV-cured hybrid resin with different content of UV-6110
Figure 1. FT-IR spectra of the liquid resin and UV-cured hybrid resin with different content of
(a, uncured hybrid resin; b,10% epoxy resin; c, 20% epoxy resin; d, 30% epoxy resin; e, 40% epoxy resin).
UV-6110 (a, uncured hybrid resin; b,10% epoxy resin; c, 20% epoxy resin; d, 30% epoxy resin; e, 40%
epoxy resin).
3.3. Shrinkage of Photopolymer Resins
The contact
3.3. Shrinkage angle of the 3D-printed
of Photopolymer Resins samples to water was around 90◦ on the surface. As shown in
Figure 2a, with an increasing content of epoxy resin, the wettability of the sample surface changed
Thewhich
slightly, contactmeant
angle that
of thethe3D-printed samples
contact angle to water
did not was
clearly aroundon
depend 90° onchange
the the surface.
in theAs shown
UV-6110
in Figure 2a, with an increasing content of epoxy resin, the wettability of the sample surface
concentration. It was clear that a medium-polarity surface was formed owing to the coexistence of changed
slightly, which
Polymers 2020,
hydrophobic
12, x meant
FOR PEERthat the contact
REVIEW
materials (UV-6110) angle
and the did
polar not clearly
groups depend
of acrylic on thesuch
oligomers, change in the UV-6110
7 of 15
as morpholinyl.
concentration. It was clear that a medium-polarity surface was formed owing to the coexistence of
hydrophobic materials (UV-6110) and the polar groups of acrylic oligomers, such as morpholinyl.

Under the UV light irradiation, liquid acrylate resins quickly transformed to a solid
homogeneous film since the Van der Waals forces were converted into covalent bonds, significantly
narrowing the average distance between the molecules. The epoxy resins could be incorporated into
the acrylate resin system as fillers to decrease their shrinkage properties. It was found from Figure
2b and Table 1 that the shrinkage of the hybrid resins was obviously decreased along with the
adding of epoxy.

Figure 2. Influence of the modification ratio on the water contact angles and polymerization shrinkage of
Figure 2. Influence of the modification ratio on the water contact angles and polymerization
the UV-cured polymer. ((A), contact angles of samples with different contents of epoxy; (B), Shrinkage of
shrinkage of the UV-cured polymer. (A, contact angles of samples with different contents of epoxy;
samples with different contents of epoxy).
B, Shrinkage of samples with different contents of epoxy).
Under the UV light irradiation, liquid acrylate resins quickly transformed to a solid homogeneous
Table
film since the Van der 1.Waals
The shrinkage of the
forces were samples with
converted intodifferent
covalentcontent
bonds,ofsignificantly
epoxy resin. narrowing the
average distance between the molecules.
Content (%) The epoxy resins could be incorporated
Shrinkage (%) into the acrylate
resin system as fillers to decrease their shrinkage properties. It was found from Figure 2b and Table 1
0
that the shrinkage of the hybrid resins was obviously decreased along10.44with the adding of epoxy.

10 9.15

20 8.57

30 7.83
Polymers 2020, 12, 1346 7 of 14

Table 1. The shrinkage of the samples with different content of epoxy resin.

Content (%) Shrinkage (%)

0 10.44
10 9.15
20 8.57
30 7.83
40 7.41

3.4. Tensile Strength and Elongation Properties

The mechanical properties of the printed objects before and after post-curing were studied
and analyzed, and the results are shown in Figure 3 and Table 2. The tensile strength and bending
strength of the optimal samples before post-curing were 29 ± 1.4 MPa and 58± 2.1 MPa, respectively.
The tensile strength and bending strength of the samples at 20% of epoxy monomers after post-cuing
were 28.46 ± 0.5 MPa and 106 MPa, respectively. Therefore, the tensile strength of the samples after
post-curing increased by up to 64% compared with the samples without post-curing. The bending
strength increased up to 150%, which indicated that the mechanical properties of the 3D-printed
Polymers 2020, 12, x FOR PEER REVIEW 8 of 15
samples were apparently improved after post-curing treatment.

50
46
unpost-cured 120 unpost-cured
41 106
100oC 2h 100oC 2h
40 96
90
Bending strength(Mpa)
Tensile strength(Mpa)

29 29
30 27 73
25 24
22 59
60
20 48 49
14 42
34
10 30 27
10 19

0 0
0 10 20 30 40 0 10 20 30 40
Content (%) Content (%)

Figure 3. Influence of the modification ratio on the mechanical properties of hybrid resins.
Figure 3. Influence of the modification ratio on the mechanical properties of hybrid resins.
Table 2. Mechanical properties of photosensitive resin.
Table 2. Mechanical properties of photosensitive resin.
Content /% Tensile Strength/MPa Breaking Elongation/% Bending Strength/MPa
Content /%0 Tensile Strength/MPa
25 Breaking17Elongation/% Bending
47 Strength/MPa
10 29 12 58
020 25 28 14 17 49 47
30 22 21 44
1040 29 13 30 12 27 58

20 28 14 49
When the epoxy resin content reached 20%, there were more open-ring bonds that reacted with
the acrylate 30resin to form the 22IPN structure after the post-curing
21 treatment, further44increasing the
tensile strength
40 of the hybrid13resins compared with that of30the resins containing 10% 27 of epoxy resins.
However, when the epoxy resin content reached 30%, the efficiency of the cationic photo-initiator was
insufficient under the post-curing at 100 ◦ C for 2 h. The situation had reversed due to the facts that the
IPN structure did not form completely in the samples and that those uncured epoxy resins acted as
3.5. Effect of decreasing
impurities, Heat Treatment at Different
the tensile Temperatures on Spline Morphology
strength.
The heat
3.5. Effect treatment
of Heat of 3D-printed
Treatment objects was performed
at Different Temperatures for 1 h at 150 °C, 100 °C, and 80 °C in a
on Spline Morphology
vacuum oven. The results shown in Figure 4 indicated that the printed samples were warped after
The exhibiting
heat treatment of 3D-printed ◦ C, 100 ◦ C, and 80 ◦ C in
heating, higher shrinkage objects
along was
the performed for 1 This
y-axis [31–32]. h at 150
observation revealed that
a vacuumtemperature
excessive oven. The results
duringshown in Figure 4 process
the post-curing indicated that the
would printed
induce samples
parts were warped
of distortion. after
The reason
was probably that while the external part of the sample heated up during the heat treatment, the
internal part remained cool, leading to an uneven thermal change, which resulted in the external
resins bearing bigger tensile stress. Ultimately, the differential thermal strains were generated, and
cracking and warping were produced. As shown in Figure 5, many cracks were formed around the
sample treated at 150 °C for 1 h.
 The heat treatment of 3D-printed objects was performed for 1 h at 150 °C, 100 °C, and 80 °C in a
vacuum oven. The results shown in Figure 4 indicated that the printed samples were warped after
heating, exhibiting higher shrinkage along the y-axis [31–32]. This observation revealed that
excessive temperature during the post-curing process would induce parts of distortion. The reason
was probably that while the external part of the sample heated up during the heat treatment,
Polymers 2020, 12, 1346
the
8 of 14
internal part remained cool, leading to an uneven thermal change, which resulted in the external
resins bearing bigger tensile stress. Ultimately, the differential thermal strains were generated, and
heating, exhibiting
cracking higherwere
and warping shrinkage alongAs
produced. theshown
y-axis in
[31,32]. This
Figure observation
5, many cracks revealed
were formedthat excessive
around the
temperature during the post-curing
sample treated at 150 °C for 1 h. process would induce parts of distortion. The reason was probably
that while the external part of the sample heated up during the heat treatment, the internal part
remained In order to make to
cool, leading 3D-printed
an uneven products
thermalwith highwhich
change, accuracy, the post-heating
resulted treatment
in the external should be
resins bearing
bigger tensile stress. Ultimately, the differential thermal strains were generated, and crackingatand
carried out at an appropriate temperature. It was found that when the sample was treated 70 °C
for 3 h,were
warping no warp occurred,
produced. and the
As shown shrinkage
in Figure ratioscracks
5, many of thewere
length and width
formed aroundwere 0.97% and
the sample 1.5%,
treated
respectively,
◦
at 150 C for 1 h. which indicated that a highly accurate sample was formed.

80 °C 1 h

Polymers 2020, 12, x FOR PEER REVIEW 9 of 15

Polymers 2020, 12, x FOR PEER REVIEW 100 °C 1 h 9 of 15

150 °C 1h
150 °C 1h

150 °C 1h
150 °C 1h

Figure 4. Morphological changes of the resin with different heat treatment conditions.
Figure 4. Morphological changes of the resin with different heat treatment conditions.
Figure 4. Morphological changes of the resin with different heat treatment conditions.

70 °C 1 h 70 °C 3h
70 °C 1 h 70 °C 3h

70 °C 3 h
70 °C 1 h
70 °C 3 h
70 °C 1

Figure
Figure 5. Morphological
5. Morphological changes
changes in the
in the resin
resin after
after different
different times
times at 70 70 ◦ C.
at °C.
Figure 5. Morphological changes in the resin after different times at 70 °C.
In order
3.6. Soxhlet to make 3D-printed products with high accuracy, the post-heating treatment should be
Extraction
3.6. Soxhlet
carried out atExtraction
an appropriate temperature. It was found that when the sample was treated at 70 ◦ C
As a means
for 3 h,As noawarp of stereolithography
occurred, and the3D printing,ratios
shrinkage UV light irradiation
of the length is very
and crucial
width wereand can initiate
0.97%
means of stereolithography 3D printing, UV light irradiation is very crucial andand
can1.5%,
initiate
polymerization
respectively, reactions
which or
indicated crosslinking
that a highly between
accurate resins.
sample Towasassess the
formed. degree of crosslinking,
polymerization reactions or crosslinking between resins. To assess the degree of crosslinking,
soxhlet extraction was employed to measure the gel content of the samples. The gel fraction (wt%)
soxhlet extraction was employed to measure the gel content of the samples. The gel fraction (wt%)
reflected the Extraction
3.6. Soxhlet reaction degree of the resins after UV-curing and post-heating treatment. The
reflected the reaction degree of the resins after UV-curing and post-heating treatment. The
experimental data are shown in Table 3 in detail. It could be observed that the gel content increased
As a meansdata
experimental of are
stereolithography
shown in Table3D 3 inprinting,
detail. It UV
could light irradiation
be observed thatisthe
very
gelcrucial
contentand can
increased
dramatically from 0 to 83.1 wt% after the 3Dprinting process, which meant that there were some
initiate polymerization reactions or crosslinking between resins. To assess
dramatically from 0 to 83.1 wt% after the 3Dprinting process, which meant that there were some the degree of crosslinking,
uncured resins in the sample. After post-curing at 100 °C for 1 h, the gel content values of the
soxhlet
uncured extraction
resins inwas employed
the sample. to measure
After the gel at
post-curing content
100 °Coffor the1samples.
h, the gelThecontent
gel fraction
values(wt%)
of the
samples increased up to 100 wt%, as they did after that at 70 °C for 3 h. The results revealed that the
reflected
samples theincreased
reaction degree
up to 100of the resins
wt%, afterdid
as they UV-curing and
after that atpost-heating
70 °C for 3 h.treatment.
The resultsThe experimental
revealed that the
polymer chains of the resins had been reacted completely. Figure 6 shows that upon increasing the
data are shown
polymer chainsin of
Table
the 3resins
in detail.
had It could
been be observed
reacted that the
completely. gel 6content
Figure showsincreased
that upondramatically
increasing the
heating time from 1 h to 3 h at 70 °C, the tensile strength and bending strength were increased to 45 ±
from 0 to 83.1
heating timewt%
fromafter
1 h tothe3 h3Dprinting
at 70 °C, theprocess,
tensile which
strength meant that there
and bending were some
strength wereuncured
increased resins
to 45 ±
1 MPa and 99 ± 2 MPa, respectively. Therefore, the hybrid resins did not react completely during the
◦ C for
in 1the
MPasample.
and 99After post-curing
± 2 MPa, at 100
respectively. Therefore, 1 h,
thethe gel content
hybrid values
resins did of thecompletely
not react samples increased
during the
3D printing process. It is necessary to make a◦ perfect and stable product with the use of a heating
up3D to printing
100 wt%,process.
as they Itdidis after that atto70
necessary makeC for 3 h. The
a perfect andresults
stablerevealed
productthat withthe
thepolymer
use of achains
heating
treatment after 3D printing. In the current work, the hybrid resins made by us could prepare stable
treatment after 3D printing. In the current work, the hybrid resins made by us could prepare stable
products by LCD light irradiation-based 3D printing with a post-curing treatment at 70 °C for 3 h.
products by LCD light irradiation-based 3D printing with a post-curing treatment at 70 °C for 3 h.
This was due to an integrated IPN structure being formed completely in the hybrid resins, resulting
This was due to an integrated IPN structure being formed completely in the hybrid resins, resulting
in excellent mechanical properties.
in excellent mechanical properties.
Table 3. Gel fractions of the samples with various heat treatment conditions.
 Polymers 2020, 12, 1346 9 of 14

of the resins had been reacted completely. Figure 6 shows that upon increasing the heating time
from 1 h to 3 h at 70 ◦ C, the tensile strength and bending strength were increased to 45 ± 1 MPa and
99 ± 2 MPa, respectively. Therefore, the hybrid resins did not react completely during the 3D printing
process. It is necessary to make a perfect and stable product with the use of a heating treatment after
3D printing. In the current work, the hybrid resins made by us could prepare stable products by
LCD light irradiation-based 3D printing with a post-curing treatment at 70 ◦ C for 3 h. This was due
to an integrated IPN structure being formed completely in the hybrid resins, resulting in excellent
mechanical properties.

Table 3. Gel fractions of the samples with various heat treatment conditions.

Sample Gel Fraction (wt%)

Liquid resin 0
Freshly printed resin 83.1
Polymers 2020, 12, x FOR PEER REVIEW 100 ◦ C 1 h 100 10 of 15
70 ◦ C 2 h 98.4
70 ◦ C 3 h 100

99
100 tenile strength 95 100
bending srength

Bending strength （MPa ）

80
80
80 80
Tenile strength (MPa)

60 60

43 45
44
42 42
40 40
34
34
27
27
20 20

0 0
0 1 2 3
Time (h)

Figure 6. Mechanical property changes of the cured resin after different times at 70 ◦ C.
Figure 6. Mechanical property changes of the cured resin after different times at 70 °C.
3.7. Thermal Stability of the 3D-Printed Objects

3.7. ThermalTGA analysis

Stability allows
of the the studying
3D-Printed of the thermal stability of the resins. The results of the TGA
Objects
analysis are shown in Figure 7, and the degradation temperatures are listed in Table 4. Obviously, all of
TGA analysis
the samples allows
after the studying
the post-heating of thehad
treatment thermal
a good stability of the
heat resistance resins.ItThe
property. was results
observedofthat
the TGA
analysistheare shown
weight loss in
wasFigure 7, 10.0
less than andwt%the asdegradation temperatures
the temperature was increasedare
uplisted ◦
to 200 inC Table
and then4. less
Obviously,
than all
20.0 wt% as the temperature was increased up to had ◦
300 a C.good
The weight
of the samples after the post-heating treatment heat loss in the first
resistance stage wasItattributed
property. was observed
to those small molecules, such as photoinitiator 819 and 261, and moisture
that the weight loss was less than 10.0 wt% as the temperature was increased up to 200 °C and then inside of the samples.
The second stage happened in the range of 200–350 ◦ C and could be regarded as the decomposition
less than 20.0 wt% as the temperature was increased up to 300 °C. The weight loss in the first stage
of the hard section where the post-cured sample reached a nearly constant weight. Therefore, in this
was attributed to those small molecules, such as photoinitiator 819 and 261, and moisture inside of
stage, the unreacted acrylic resins and epoxy monomers were evaporated constantly while the internal
the samples. Thesample
part of the second stagecompletely.
reacted happenedIninthethe range
range of 200–350
of 350–450 °Ctoand
◦ C, due could be regarded
the decomposition of the as the
decomposition
cross-linked bonds and other chemical bonds, those uncured epoxy resins accelerated the thermalweight.
of the hard section where the post-cured sample reached a nearly constant
Therefore, in this stage,
decomposition the unreacted
of the UV-cured resins asacrylic
shown in resins and
Table 4. epoxythose
Although monomers
resins had were
formed evaporated
an
IPN structure
constantly while the between
internalthe part
acrylate andsample
of the epoxy resins in the
reacted process of UV-curing,
completely. In the range the uncured epoxy°C, due
of 350–450
monomers decreased their thermal stability. However, after post-curing,
to the decomposition of the cross-linked bonds and other chemical bonds, those uncured epoxy the hybrid resins initially
increased the thermal stability due to the complete reaction of the hybrid resins and further formation
resins accelerated the thermal decomposition of the UV-cured resins as shown in Table 4. Although
of IPN structures between the acrylate resins and epoxy resins.
those resins had formed an IPN structure between the acrylate and epoxy resins in the process of
UV-curing, the uncured epoxy monomers decreased their thermal stability. However, after
post-curing, the hybrid resins initially increased the thermal stability due to the complete reaction of
the hybrid resins and further formation of IPN structures between the acrylate resins and epoxy
resins.
 resins accelerated the thermal decomposition of the UV-cured resins as shown in Table 4. Although
those resins had formed an IPN structure between the acrylate and epoxy resins in the process of
UV-curing, the uncured epoxy monomers decreased their thermal stability. However, after
post-curing, the hybrid resins initially increased the thermal stability due to the complete reaction of
the hybrid
Polymers 2020, 12,resins
1346 and further formation of IPN structures between the acrylate resins and epoxy 10 of 14
resins.

100
unheated
heated
80

Weight(%)
Polymers 2020, 12, x FOR PEER REVIEW
60
11 of 15

40
Table 4. TGA data of the hybrid-cured films.

Sample Decomposition Temperature/°C Residual Mass/%

20

70 °C, 3 h 395.22 15.74

0
0 100 200 300 400 500 600
unheated 380.22
Tempeture(%)
14.31

Figure 7. TGA curves of the before after 70 ◦ C 3h heat treatment films.

Figure 7. TGA curves of the before after 70 °C 3h heat treatment films.
Table 4. TGA data of the hybrid-cured films.
3.8. Dynamic Mechanical Properties
Sample analysis
Decomposition ◦
Dynamic mechanical (DMA) Temperature/
conducted in C flexion
Residual Mass/%
mode was used to study the
◦
variation in the 70 C, 3 h
mechanical properties 395.22
of the 3D-printed samples 15.74
with changes in time and
unheated
temperature. Moreover, DMA was also used 380.22 14.31
to study the difference between the samples before and
after post-curing. The thermal mechanical properties of the 3D-printed samples before and after
3.8. Dynamic Mechanical
post-curing Properties
were studied through a DMA test, and the results are shown in Figure 8 and Table 5,
which contain
Dynamic the flexion
mechanical storage
analysis modulus
(DMA) (E′) and
conducted tan δ mode
in flexion measured for those
was used fresh
to study the3D-printed
variation
samples at 6 Hz frequencies.
in the mechanical properties of the 3D-printed samples with changes in time and temperature.
Moreover, DMA was also used to study the difference between the samples before and after post-curing.
The thermal mechanical properties of the 3D-printed samples before and after post-curing were studied
through a DMA test, and the results are shown in Figure 8 and Table 5, which contain the flexion
storage modulus (E0 ) and tan δ measured for those fresh 3D-printed samples at 6 Hz frequencies.

0.65 2000

0.60 A
B
0.55 1500 C
0.50
A B
E'（MPa）

0.45 1000
tan σ

0.40

0.35
A 500
0.30 B
C
0.25
0
0.20

0 20 40 60 80 100 120 0 20 40 60 80 100 120

o
Tmep ( C ) Temp(oC)

Figure 8. Dynamic mechanical analysis (DMA) curves of hybrid polymer storage modulus loss factor
tan δ (A) and
Figure E0 (B). (A:
8. Dynamic 70 ◦ C 3 h 20%
mechanical epoxy(DMA)
analysis monomer content
curves of hybrid 70 ◦ C 3 hstorage
resin; B:polymer 0% epoxy monomer
modulus loss
content
factor resin; C: 20%
tan δ (A) andepoxy
E′ (B).monomer
(A: 70 °Cun
3 hpost-cured
20% epoxycontent resin).
monomer content resin; B: 70 °C 3 h 0% epoxy
monomer content0 resin; C: 20% epoxy monomer un post-cured content resin).
Table 5. E and Tg of the UV-cured polymer and hybrid films obtained from DMA.

As observed from the DMA Sample curves ofTthe ◦ hybridE0polymer

(MPa) storage modulus loss factor tan δ,
g ( C)
there was only one glass transitionA temperature 82.6 (Tg) detected, suggesting only a single phase
1826.9
structure in the sample and no phase B separation
79.8 [33,34], which meant that the acrylic resin and
1347.1
C 60.5 953.2
epoxy resin had good compatibility under UV light irradiation. Epoxy molecular chains and the
acrylic molecular chains interweaved with each other and prevented phase separation. Figure 8
showed that, compared
As observed from the with
DMA post-cured
curves of thepure acrylic
hybrid resins,storage
polymer the unreacted
modulus epoxy monomers
loss factor tan δ,
decreased
there was onlytheone
thermal stability oftemperature
glass transition the 3D-printed objects. After
(Tg) detected, the post-curing
suggesting process,
only a single phase the hybrid
structure
sample showed a higher glass transition temperature (Tg) than the pure acrylic resins,
in the sample and no phase separation [33,34], which meant that the acrylic resin and epoxy resin had indicating
that the thermal stability of the resins after post-heat treatment had been improved, probably due to
the formation of IPN structures.

Table 5. E′ and Tg of the UV-cured polymer and hybrid films obtained from DMA.

Sample Tg (°C) E′ (MPa)

Polymers 2020, 12, 1346 11 of 14

good compatibility under UV light irradiation. Epoxy molecular chains and the acrylic molecular chains
interweaved with each other and prevented phase separation. Figure 8 showed that, compared with
post-cured pure acrylic resins, the unreacted epoxy monomers decreased the thermal stability of the
3D-printed objects. After the post-curing process, the hybrid sample showed a higher glass transition
temperature (Tg) than the pure acrylic resins, indicating that the thermal stability of the resins after
post-heat treatment had been improved, probably due to the formation of IPN structures.

3.9. Finite Element Analysis of the Effect of the Temperature on the Spline Warp
In order to analyze the effect of the temperature on the spline warp, a finite element analysis (FEA)
was carried out [35–37]. The FEA process was as follows: (1) Calculate the heat flux of the spline within
300 s under contact with the thermostatic heating baseplate; (2) Set the heat flux data as the load of the
spline, and calculate the temperature field change within 300 s of the spline; (3) According to the data
of the maximum temperature and the minimum temperature obtained in Step 2, select the load step
with the maximum temperature difference to calculate the warping caused by the thermal expansion
of the spline. The simulation model was a cuboid spline with a length of 80 mm, a width of 10 mm,
and a height of 4 mm. A thin plate was added at the bottom of the model as the heating bottom plate of
the dark oven. The average density of the material was set as 1.2 g·cm−3 , which remained unchanged
during the heat treatment. The Young’s modulus, Poisson’s ratio, and specific heat capacity of the
material were 213.29 MPa, 0.35, and 1600 J·kg−1 ·K−1 . The isotropic thermal conductivity of the material
was 0.18 w·m−1 ·K−1 , and the isotropic average thermal expansion coefficient was 4.8875·10−5 ◦ C−1 .
The heating base plate defaults to structural steel. All the figures obtained from FEA calculations are
illustrated in the supporting information.
The heat treatment of the spline was to place the spline in a dark oven, and the spline was heated
by the heating bottom plate without air convection and thermal radiation. Assuming that the spline
was heated by direct contact with the bottom plate, the warping phenomenon of the spline could be
better analyzed by calculating the heat flux of the spline.
Figures S1–S3, included in supporting information, show the heat flux maps of the spline when
it just came into contact with the heating baseplate at 80 ◦ C, 100 ◦ C, and 150 ◦ C. The greater the
temperature difference between the heating baseplate and the spline, the greater the heat flux would be.
At 80 ◦ C, the maximum heat flux of the spline was about 7775 w·m−2 , while at 150 ◦ C, the maximum
heat flux of the spline reached 17,160 w·m−2 . The heating power of oven was generally large. When
the initial temperature difference between the heating source and the spline was large, the heat flux
of the spline would be relatively large. The spline belonged to the polymer material, the thermal
conductivity was low, and under the high heat flux, the internal temperature difference easily occurred
and caused warping.
When the heat flux was loaded onto the spline, the maximum internal temperature difference
could be calculated, thus the maximum deformation could be calculated. By comparing Figures S4–S6,
it can be seen that under high heat flux, the higher the heating temperature was, the greater the
temperature difference inside the spline would be. During heat treatment at 150 ◦ C, the temperature
difference inside the spline could be as high as 88 ◦ C. Figure S7 shows that the thermal deformation
of the spline in the Y-axis direction could reach 1.5 mm at 80 ◦ C, Figure S8 shows that the thermal
deformation of the spline in the Y-axis direction could reach 2.0 mm at 100 ◦ C, and Figure S9 shows
that the thermal deformation of the bending spline in the Y-axis direction could reach 3.4 mm at 150 ◦ C.
By comparing Figures S7–S9, it can be seen that the higher the heating bottom plate temperature
was, the more serious the warping phenomenon would be. The deformation of the thermosetting
materials was irreversible, and when the temperature difference of the spline became small gradually,
the warping was reduced a little but did not disappear.

3.10. Microstructure of the Objects Printed by 3D Printer with LCD Light Source
Some typically complicated 3D objects were selected and fabricated by a 3D printer with an LCD
light source. These samples were printed directly without any supporting structures. The printing
Polymers 2020, 12, x FOR PEER REVIEW 13 of 15

Some typically complicated 3D objects were selected and fabricated by a 3D printer with an
LCD light source. These samples were printed directly without any supporting structures. The
Polymers 2020, 12, 1346 12 of 14
printing accuracy of the objects was evaluated through calculating the sizes of the 3D models and
actual printed objects. After that, those printed 3D objects were transferred to an oven for further
thermal post-curing.
accuracy Thewas
of the objects photographs in Figure
evaluated through 10a,b were
calculating the samples
the sizes with a hollow
of the 3D models structure
and actual printed and
a honeycomb
objects. Afterstructure,
that, those respectively.
printed 3D objectsDue totransferred
were similar refractive index
to an oven for of thermal
further 3,4-epoxy cyclohexyl
post-curing.
The photographs
methyl-3,4-epoxy in Figure 10a,b
cyclohexyl wereand
formate the samples withalla hollow
acrylates, of thosestructure
printed and a honeycomb
objects lookedstructure,
transparent.
respectively. Due to similar refractive index of 3,4-epoxy cyclohexyl methyl-3,4-epoxy
Figure 9c was a printed black elk from blending with 0.05 wt% of carbon black powders in hybrid cyclohexyl
formate
resins. Afterand acrylates, all
comparison of of
thethose printed objects
dimensions betweenlooked
thetransparent.
printed objectsFigure 9c was
and a printed
the 3D black
models, it was
elk from blending with 0.05 wt% of carbon black powders in hybrid resins. After comparison of the
found that the printed objects owned excellent printing accuracy. The SEM images of the fracture
dimensions between the printed objects and the 3D models, it was found that the printed objects owned
surfaces of the printed samples are shown in Figure 10, where no particles were found on the
excellent printing accuracy. The SEM images of the fracture surfaces of the printed samples are shown
fracture surface
in Figure of the no
10, where synthesized photosensitive
particles were found on the hybrid
fracture resins,
surface indicating that the
of the synthesized compatibility of
photosensitive
the acrylic resin with the epoxy resin was fine. As shown in Figure 10, the
hybrid resins, indicating that the compatibility of the acrylic resin with the epoxy resin was sample after post-curing
fine.
treatment as shown in Figure 10b exhibited a smooth and glassy
As shown in Figure 10, the sample after post-curing treatment as shown in Figure 10b exhibited surface representing a
homogeneous
a smooth and structure. However,
glassy surface the fracture
representing surfacestructure.
a homogeneous of the sample
However, without post-curing
the fracture surface of heat
treatment showed rougher features and more micro-cracks than the post-cured samples, which
the sample without post-curing heat treatment showed rougher features and more micro-cracks than led
the post-cured
to a low mechanical samples, which led to a low mechanical strength.
strength.

Figure
Figure 9. 3D
9. 3D printing
printing of of hybridresin.
hybrid resin.(a)
(a)hollow
hollow structure,
structure,(b)
(b)honeycomb
honeycombstructure, (c) elk.
structure, (c) elk.

Figure 10. SEM micrographs of the printed the before (A) and after post-cured (B) samples.
Figure 10. SEM micrographs of the printed the before (A) and after post-cured (B) samples.
4. Conclusions
4. Conclusions
In the present work, a kind of novel sensitive hybrid resin aimed at 3D printing with LCD light
irradiation
In the presentthrough a free
work, a kindradical/cationic dual curing
of novel sensitive hybridprocess was prepared.
resin aimed The curing
at 3D printing withprocess
LCD light
involved an opening of the double bond of the acrylate resin and a ring-opening
irradiation through a free radical/cationic dual curing process was prepared. The curing process polymerization of
the epoxy resin. Followed by further UV-curing, an interpenetrating polymer
involved an opening of the double bond of the acrylate resin and a ring-opening polymerization ofnetwork (IPN) and
a complex 3D shape were obtained. Due to the incomplete curing of the resins in the 3D printing process,
the epoxy resin. Followed by further UV-curing, an interpenetrating polymer network (IPN) and a
the post-curing treatment was necessary and has been proved by this study in detail. The mechanical
complex 3D shape were obtained. Due to the incomplete curing of the resins in the 3D printing
properties of the hybrid composite depended on the UV-curing degree. The optimal ratio of epoxy in
process, the post-curing
the hybrid treatment
resin was 1:5, where thewas necessary
Tg value and has
and storage been proved
modulus by this
of the hybrid study
films in detail.
increased after The
mechanical properties of the hybrid composite depended on the UV-curing degree.
the heat treatment. The optimal post-curing conditions for the cationic polymerization were confirmed The optimal
ratioasofa epoxy
heatingin the hybrid
temperature of resin wasa1:5,
70 ◦ C and where
heating timethe
of 3Tg
h. value and
Both the storage
thermal modulus
stability of the hybrid
and mechanical
filmsstrength
increased
of theafter the heat
synthesized treatment. The
free-radical/cational optimal
hybrid post-curing
photosensitive resinsconditions
were higher for
thanthe
thosecationic
of
polymerization were confirmed as a heating temperature of 70 °C and a heating time of 3 h. Both the
the pure oligomer resins.
thermal stability and mechanical strength of the synthesized free-radical/cational hybrid
photosensitive resins were higher than those of the pure oligomer resins.
Polymers 2020, 12, 1346 13 of 14

Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4360/12/6/1346/s1,

Figure S1 Heat flux maps of the spline at 80 ◦ C; Figure S2 Heat flux maps of the spline at 100 ◦ C; Figure S3 Heat
flux maps of the spline at 150 ◦ C; Figure S4 Temperature map of the spline with heat treatment at 80 ◦ C for 22 s;
Figure S5 Temperature map of the spline with heat treatment at 100 ◦ C for 22 s; Figure S6 Temperature map of the
spline with heat treatment at 150 ◦ C for 22 s; Figure S7 Maximum thermal deformation map of the spline on Y-axis
at 80 ◦ C; Figure S8 Maximum thermal deformation map of the spline on Y-axis at 100 ◦ C; Figure S9 Maximum
thermal deformation map of the spline on Y-axis at 150 ◦ C.
Author Contributions: Conceptualization, J.S. and W.Z. (Wuyi Zhou); methodology, J.S.; validation, Z.Y.; formal
analysis, J.S.; FEA calculation, G.C. and W.Z. (Wenxu Zheng); resources, W.Z. (Wuyi Zhou), X.D., and W.Z.
(Wenxu Zheng); data curation, J.S.; writing—original draft preparation, J.S.; writing—review and editing, W.Z.
(Wuyi Zhou), Y.H. and Y.L.; supervision, W.Z. (Wuyi Zhou). All authors have read and agreed to the published
version of the manuscript.
Funding: This research was funded by Natural Science Foundation of Guangdong Province (2020A151501823),
special scientific research project of Guangdong universities (2020KZDZX1035), International Science and
Technology Cooperation Project of Guangzhou Economic Technological Development Zone (2017GH09),
2019 Innovation Team Project of Shaoguang City.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Chattopadhyay, D.; Panda, S.S.; Raju, K. Thermal and mechanical properties of epoxy acrylate/methacrylates
UV cured coatings. Prog. Org. Coat. 2005, 54, 10–19. [CrossRef]
2. Kardar, P.; Ebrahimi, M.; Bastani, S.; Jalili, M. Using mixture experimental design to study the effect of
multifunctional acrylate monomers on UV cured epoxy acrylate resins. Prog. Org. Coat. 2009, 64, 74–80.
[CrossRef]
3. Buchanan, C.; Gardner, L. Metal 3D printing in construction: A review of methods, research, applications,
opportunities and challenges. Eng. Struct. 2019, 180, 332–348. [CrossRef]
4. He, Y.; Zhou, M.; Wu, B.; Jiang, Z.; Nie, J. Synthesis and properties of novel polyurethane acrylate containing
3-(2-hydroxyethyl) isocyanurate segment. Prog. Org. Coat 2010, 67, 264–268. [CrossRef]
5. Bayramoglu, G.; Kahraman, M.V.; Kayaman-Apohan, A. The coating performance of adipic acid modified
and methacrylated bisphenol-A based epoxy oligomers. Polym. Adv. Technol. 2007, 18, 173–179. [CrossRef]
6. Hawker, J.; Hedrick, J.L.; Malmstrom, E.E. Dual living free radical and ring opening polymerizations from
a double-headed initiator. Macromolecules. Macromolecules 1998, 31, 213–219. [CrossRef]
7. Ni, Y.; Zheng, S.; Nie, K. Morphology and thermal properties of inorganic–organic hybrids involving epoxy
resin and polyhedral oligomeric silsesquioxanes. Polymer 2004, 45, 5557–5568. [CrossRef]
8. Chiang, C.L.; Ma, C.C.M.; Wang, F.Y.; Kuan, H.C. Thermooxidative degradation of novel epoxy containing
silicon and phosphorous nanocomposites. Eur. Polym. J. 2003, 39, 825–830. [CrossRef]
9. Xu, F.; Yang, J.L.; Gong, Y.S.; Ma, G.P.; Nie, J. A fluorinated photoinitiator for surface oxygen inhibition
resistance. Macromolecules 2012, 45, 1158–1164. [CrossRef]
10. Hakeim, O.A.; Arafa, A.A. Effect of process conditions on the properties of surface-modified organic pigments
encapsulated by UV-curable resins. Coloration Technol. 2018, 134, 44–58. [CrossRef]
11. Ceccia, S.; Turcato, E.A.; Maffettone, P.L.; Bongiovanni, R. Nanocomposite UV-curedcoatings: Organoclay
intercalation by an epoxy resin. Prog. Org. Coat. 2008, 63, 110–115. [CrossRef]
12. Bongiovanni, R.; Turcato, E.A.; Gianni, A.D.; Ronchetti, S. Epoxy coatings containing clays and organoclays:
Effect of the filler and its water content on the UV-curing process. Prog. Org. Coat. 2008, 62, 336–343. [CrossRef]
13. Huang, B.; Du, Z. Preparation of a novel hybrid type photosensitive resin for stereolithography in 3D printing
and testing on the accuracy of the fabricated parts. J. Wuhan Univ. Technol.-Mate 2017, 32, 726–732. [CrossRef]
14. Sasaki, H. Curing properties of cycloaliphatic epoxy derivatives. Prog. Org. Coat. 2007, 58, 227–230. [CrossRef]
15. Crivello, J.V. UV and electron beam-induced cationic polymerization. Nucl. Instrum. Methods Phys. Res. 1999,
151, 8–21. [CrossRef]
16. Bulut, U.; Crivello, J.V. Investigation of the reactivity of epoxide monomers in photo-initiated cationic
polymerization. Macromolecules 2005, 38, 3584–3595. [CrossRef]
17. Teramoto, N.; Kogure, H.; Kimura, Y.; Shibata, M. Thermal properties and biodegradability of the copolymers
of l-lactide, ε-caprolactone, and ethylene glycol oligomer with maleate units and their crosslinked products.
Polymer 2004, 45, 7927–7933. [CrossRef]
Polymers 2020, 12, 1346 14 of 14

18. Tai, Q.; Song, L.; Lv, X.; Lu, H.; Hu, Y.; Yuen, R.K.K. Flame-retarded polystyrene with phosphorus- and
nitrogen-containing oligomer: Preparation and thermal properties. J. Appl. Polym. 2011, 123, 770–778. [CrossRef]
19. Kwak, R.; Park, H.; Ko, H. Partially Cured Photopolymer with Gradient Bingham Plastic Behaviors as
a Versatile Deformable Material. ACS Macro Lett. 2017, 6, 561–565. [CrossRef]
20. Ahn, B.U.; Lee, S.K.; Lee, S.K.; Park, J.H.; Kim, B.K. UV curable polyurethane dispersions from polyisocyanate
and organosilane. Prog. Org. Coat. 2008, 62, 258–264. [CrossRef]
21. Unal, S.; Oguz, C.; Yilgor, E.; Gallivan, M.; Long, T.E. Understanding the structure development in
hyperbranched polymers prepared by oligomeric A2+B3 approach: Comparison of experimental results and
simulations. Polymer 2005, 46, 4533–4543. [CrossRef]
22. Zhao, T.; Yu, R. A comparative study on 3D printed silicone-epoxy/acrylate hybrid polymers via pure
photopolymerization and dual-curing mechanisms. J. Mater. Sci. 2019, 54, 5101–5111. [CrossRef]
23. Colton, J.; Blair, B. Experimental study of post-build cure of stereolithography polymers for injection molds.
Rapid Prototyp. J. 1999, 5, 72–81. [CrossRef]
24. Hague, R.; Mansour, S.; Saleh, N. Materials analysis of stereolithography resins for use in Rapid Manufacturing.
J. Mater. Sci. 2004, 39, 2457–2464. [CrossRef]
25. Yang, Z.; Wu, G.; Wang, S. Dynamic postpolymerization of 3D-printed photopolymer nanocomposites: Effect
of cellulose nanocrystal and postcure temperature. J. Polym. Sci. Part B Polym. Phys. 2018, 56, 935–946.
[CrossRef]
26. Green, W.A. Industrial Photoinitiators a Technical Guide; CRC Press: Boca Raton, FL, USA, 2010.
27. Huang, J.; Yuan, T. UV/thermal dual curing of tung oil-based polymers induced by cationic photoinitiator.
Prog. Org. Coat. 2019, 126, 8–17. [CrossRef]
28. Wong, K.V.; Hernandez, A. A Review of Additive Manufacturing. ISRN Mech. Eng 2012. [CrossRef]
29. Melchels, F.P.W.; Feijen, J.; Grijpma, D.W. A review on stereolithography and its applications in biomedical
engineering. Biomaterials 2010, 31, 6121–6130. [CrossRef]
30. Yang, C.H.; Liu, F.J.; Liu, Y.P.; Liao, W.T. Hybrids of colloidal silica and waterborne polyurethane. J. Colloid
Interface Sci. 2006, 302, 123–132. [CrossRef] [PubMed]
31. Van, G.A.; Fy, F. Theory of shrinkage stresses in oriented glass-reinforced plastics. Mechanika Polymerov. 1965,
6, 61–68.
32. Karalekas, D.; Aggelopoulos, A. Study of shrinkage strains in a stereolithography cured acrylic photopolymer
resin. J. Mater. Process. Technol. 2003, 136, 146–150. [CrossRef]
33. Jin, H.; Yoon, S.; Kim, S.C. Synthesis and characterization of interpenetrating polymer networks from
polyurethane and poly(ethylene glycol) diacrylate. J. Appl. Polym. 2008, 109, 805–812. [CrossRef]
34. Okhawilai, M.; Pudhom, K.; Rimdusit, S. Synthesis and characterization of sequential interpenetrating
polymer networks of polyurethane acrylate and polybenzoxazine. Polym. Eng. 2014, 54, 1151–1161. [CrossRef]
35. Hossain, M.; Possart, G.; Steinmann, P. A small-strain model to simulate the curing of thermosets.
Comput. Mech. 2009, 43, 769–779. [CrossRef]
36. Wua, J.; Zhaoa, Z.; Hamel, C.M. Evolution of material properties during free radical photopolymerization.
J. Mech. Phys. Solids 2018, 112, 25–49. [CrossRef]
37. Hossain, M.; Possart, G.; Steinmann, P. A finite strain framework for the simulation of polymer curing. Part I:
Elasticity. Comput. Mech. 2009, 44, 621–630. [CrossRef]

© 2020 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/).