Self-Extinguishing Resin Transfer Molding Composites Using Non-Fire-Retardant Epoxy Resin


Geng, Z.; Yang, S.; Zhang, L.; Huang, Z.; Pan, Q.; Li, J.; Weng, J.; Bao, J.; You, Z.; He, Y.; Zhu, B.

Materials 11(12)

2018


Introducing fire-retardant additives or building blocks into resins is a widely adopted method used for improving the fire retardancy of epoxy composites. However, the increase in viscosity and the presence of insoluble additives accompanied by resin modification remain challenges for resin transfer molding (RTM) processing. We developed a robust approach for fabricating self-extinguishing RTM composites using unmodified and flammable resins. To avoid the effects on resin fluidity and processing, we loaded the flame retardant into tackifiers instead of resins. We found that the halogen-free flame retardant, a microencapsulated red phosphorus (MRP) additive, was enriched on fabric surfaces, which endowed the composites with excellent fire retardancy. The composites showed a 79.2% increase in the limiting oxygen index, a 29.2% reduction in heat release during combustion, and could self-extinguish within two seconds after ignition. Almost no effect on the mechanical properties was observed. This approach is simple, inexpensive, and basically applicable to all resins for fabricating RTM composites. This approach adapts insoluble flame retardants to RTM processing. We envision that this approach could be extended to load other functions (radar absorbing, conductivity, etc.) into RTM composites, broadening the application of RTM processing in the field of advanced functional materials.

materials
Article
Self-Extinguishing
Resin
Transfer
Molding
Composites
Using
Non-Fire-Retardant
Epoxy
Resin
Zhi
Geng
1
'
21E
,
Shuaishuai
Yang
3,t,
Lianwang
Zhang
4
,
Zhenzhen
Huang
2,*
,
Qichao
Pan
1,2,
Jidi
Li
2
,
Jianan
Weng
2
,
Jianwen
Bao
4,*
,
Zhengwei
You
1
,
Yong
He
5
'
4
and
Bo
Zhu
2
'
*
1
State
Key
Laboratory
for
Modification
of
Chemical
Fibers
and
Polymer
Materials,
College
of
Materials
Science
and
Engineering,
Donghua
University,
Shanghai
201620,
China;
(Z.G.);
(Q.P.);
(Z.Y.)
2
School
of
Materials
Science
and
Engineering,
Shanghai
University,
333
Nanchen
Road,
Baoshan,
Shanghai
200444,
China;
(f.L.);
(f.W.)
3
SAMAC
Shanghai
Aircraft
Manufacturing
Co.,
Ltd.,
Shangfei
Road,
Pudong
New
District,
Shanghai
201324,
China;
4
Avic
Advanced
Composites
Center,
Shijun
South
Street,
Aviation
Industrial
Park,
Shunyi,
Beijing
101300,
China;
5
Collaborative
Innovation
Center
for
Civil
Aviation
Composites,
Donghua
University,
Shanghai
201620,
China
Correspondence:
(Z.H.);
cf.B.);
(Y.H.);
(B.Z.)
t
These
authors
contributed
equally
to
this
work.
Received:
6
October
2018;
Accepted:
11
December
2018;
Published:
15
December
2018
Abstract:
Introducing
fire-retardant
additives
or
building
blocks
into
resins
is
a
widely
adopted
method
used
for
improving
the
fire
retardancy
of
epoxy
composites.
However,
the
increase
in
viscosity
and
the
presence
of
insoluble
additives
accompanied
by
resin
modification
remain
challenges
for
resin
transfer
molding
(RTM)
processing.
We
developed
a
robust
approach
for
fabricating
self-extinguishing
RTM
composites
using
unmodified
and
flammable
resins.
To
avoid
the
effects
on
resin
fluidity
and
processing,
we
loaded
the
flame
retardant
into
tackifiers
instead
of
resins.
We
found
that
the
halogen-free
flame
retardant,
a
microencapsulated
red
phosphorus
(MRP)
additive,
was
enriched
on
fabric
surfaces,
which
endowed
the
composites
with
excellent
fire
retardancy.
The
composites
showed
a
79.2%
increase
in
the
limiting
oxygen
index,
a
29.2%
reduction
in
heat
release
during
combustion,
and
could
self-extinguish
within
two
seconds
after ignition.
Almost
no
effect
on
the
mechanical
properties
was
observed.
This
approach
is
simple,
inexpensive,
and
basically
applicable
to
all
resins
for
fabricating
RTM
composites.
This
approach
adapts
insoluble
flame
retardants
to
RTM
processing.
We
envision
that
this
approach
could
be
extended
to
load
other
functions
(radar
absorbing,
conductivity,
etc.)
into
RTM
composites,
broadening
the
application
of
RTM
processing
in
the
field
of
advanced
functional
materials.
Keywords:
fire
retardant;
tackifier;
resin
transfer
molding
1.
Introduction
Fiber-reinforced
epoxy
composites
are
widely
used
as
structural
materials
to
replace
metal
materials
in
airplanes,
trains,
and
automobiles,
as
they
possess
high
mechanical
strength
with
light
weight
[1-8].
Resin
transfer
molding
(RTM)
is
one
of
the
most
important
techniques
used
to
fabricate
epoxy
composites
due
to
its
cost-effectiveness,
being
solvent-free,
the
ability
to
be
mass
produced,
the
high
automatization
potential,
and
its
excellent
performance
in
manufacturing
large
parts
and
molding
complex
shapes
[9-15].
Materials
2018,
11,
2554;
doi:10.3390/ma11122554
www.mdpi.com/journal/materials
Materials
2018,
11,
2554
2
of
18
The
epoxy
resin
is
highly
flammable
and
needs
further
modification
to
reduce
the
hazard
of
accidental
fires.
Numerous
efforts
have
focused
on
reducing
the
flammability
of
epoxy
resins
and
their
composites
with
considerable
progress
[16-19].
These
fire-retardant
epoxy
composites
passed
the
UL94
VO
flammability
rating
and
achieved
a
limiting
oxygen
index
value
higher
than
30%
[_
].
The
epoxy
resins
used
in
these
composites
are
normally
prepared
through
introduction
into
resin
nonflammable
hardeners
[20,21],
building
blocks
for
epoxy
polymers
[22,23],
additives
of
small
molecules
or
polymers
[24-27],
and
nano/micromaterials
[28-30].
The
chemical
structure,
molecular
weight,
and
composition
of
RTM
epoxy resins
were
optimized
to
have
a
very
low
viscosity
to
satisfy
the
fluidity
requirement
for
RTM
processing,
whereas
most
of
these
modification
procedures
increase
the
resin
viscosity
[22,29,31-35].
The
nano/micromaterials
would
either
be
filtered
by
the
mixer
of
the
RTM
instrument
or
enriched
on
the
outer
layers
of
reinforcing
three-dimensional
(3D)
preforms,
leading
to
a
non-uniform
distribution
of
the
nano/micromaterials,
[14,36],
therefore
potentially
introducing
a
negative
effect
on
fire
resistance
and
the
mechanical
properties
of
the
composites.
Due
to
these
issues,
we
were
interested
in
developing
an
approach
to
manufacture
self-extinguishing
RTM
composites
without
modifying
the
epoxy
resin.
Our
idea
was
to
use
fire
retardant-loaded
tackifiers
or
binders
to
fabricate
3D
glass-fiber
preforms,
and
then
the
RTM
epoxy
composites.
Tackifiers
are
commonly
used
in
the
RTM
process
to
offer
preforming
and
improved
handling
for
the
unimpregnated
woven
fabrics.
As
the
tackifier
resin
does
not
soften
and
flow
at
the
injection
temperature,
the
fire-retardant
tackifier
layer
anchored
on
the
fabrics
would
not
be
diluted
by
the
injected
resin
in
the
molding.
This
means
that
the
composite
could
efficiently
retard
fire
propagation
due
to
the
localized
high
content
of
the
fire
retardant.
Red
phosphorus
is
widely
employed
to
enhance
the
fire
retardancy
of
a
variety
of
polymers,
including
epoxy
resins,
polyesters,
and
polyurethanes,
due
to
its
low
cost
and
ecofriendly
properties
[37-42].
Red
phosphorus
is
an
efficient
flame
retardant
for
epoxy
resins.
It
works
in
both
oxygen
and
nitrogen-containing
polymers
via
the
formation
of
thermally
stable
char
in
the
condensed
phase
and
phosphorus
radicals
in
the
gas
phase
[43,44].
In
this
work,
we
instead
used
microencapsulated
red
phosphorus
(MRP)
to
prepare
the
flame-retardant
tackifier,
as
MRP
additives
are
more stable
and
easier
to
handle.
Using
the
reinforcing
textile
that
was
preformed
by
the
modified
tackifier,
we
prepared,
by
a
non-fire-retardant
epoxy
resin
and
standard
RTM
approach,
an
excellent
fire-retardant
epoxy
composite.
The
thermal
stability
and
flame
retardancy
of
the
composite
were
both
investigated
by
thermogravimetric
analysis
(TGA),
the
limiting
oxygen
index
(LOI)
test,
the
cone
calorimeter
test
(CCT),
and
flammability
examination.
Char
residues
remained
after
the
CCT,
and
flammability
tests
were
observed
using
optical
and
scanning
electron
microscopy.
To
evaluate
the
effects
of
MRP
loading
on
the
mechanical
properties
of
composites,
their
tensile
and
compression
strength,
modulus
and
interlaminar
shear
strength
were
further
measured.
2.
Materials
and
Methods
2.1.
Materials
The
epoxy
resin
EP3228
and
tackifier
powder
ET3228,
averaging
at
10
lim,
used
in
this
study
was
supplied
by
the
Avic
Advanced
Composites
Center
(Beijing,
China).
The
MRP
(HP
1250),
averaging
13
µm,
was
purchased
from
the
Shanghai
Champ
Chemical
Co.,
Ltd.
(Shanghai,
China).
The
2/2
twill
glass-fiber
fabric
EW-220B-100a
was
purchased
from
the
Nanjing
Fiberglass
Research
and
Design
Institute
Co.
Ltd.
(Jiangsu,
China).
The
modified
tackifier
powder
was
prepared
by
manually
mixing
tackifier
ET3228
with
MRP.
MRP
weight
percentages
in
the
modified
tackifiers
ranged
from
0
to
48
wt.%.
The
tackifiers
containing
0,
23,
38,
and
48
wt.%
MRP
are
denoted
by
T(0%P),
T(23%P), T(38%P),
and
T(48%P),
respectively.
Materials
2018,
11,
2554
3
of
18
2.2.
Preforming
and
RTM
Processing
The
flame-retardant
tackifier
powders
of
4.0
to
7.75
wt.%
that
were
used
for
tackifier-fabric
loading
were
applied
uniformly
on
the
fabrics
using
a
sifter.
The
tackifier
powders
were
melted
onto
the
fabrics
by
placing
fabrics
under
an
infrared
(IR)
lamp
at
90
°C
for
10
s.
Ten
plies
of
fabrics
with
a
layup
(340
mm
x
70
mm)
were
pressed
at
85
°C
for
10
min
by
vacuum
bagging
to
fabricate
the
preform.
U-shape
bending
was
employed
to
evaluate
the
control
of
the
tackifier
on
the
preform
dimensions.
Ten
plies
of
fabric
with
a
layup
(340
mm
x
70
mm)
were
fixed
on
a
U-shape
bending
tool
under
the
pressure
produced
by
the
vacuum
bagging
at
85
°C
for
10
min.
After
cooling
to
room
temperature
(25
°C),
the
preform
was
removed
from
the
tool
and
the
springback
angle
was
measured
every
1
h
(Scheme
).
The
t-test
was
used
to
evaluate
whether
the
differences
among
the
springback
angles
of
the
fabrics
that
were
preformed
by
tackifiers
with
different
MRP
contents
were
statistically
significant.
e
170
mm
111
Scheme
1.
The
U-shape
bending
test.
The
springback
of
the
preform
angle
was
determined
by
averaging
0
1
and
0
2
.
The
standard
RTM
approach
was
carried
out
to
fabricate
glass-fiber-reinforced
epoxy
resin
composites.
The
preform
containing
10
plies
of
fabrics
was
placed
into
a
rectangular
RTM
mold
with
injection
and
outlet
ports.
The
assembled
mold
was
heated
to
50
°C
while
a
vacuum
was
applied
to
the
mold
cavity.
Epoxy
resin
was
injected
into
the
mold
at
a
pressure
in
the
range
of
0.05
to
0.35
MPa
while
the
thermostat
was
set
at
50
°C.
After
resin
injection,
the
mold
temperature
was
maintained
at
100
°C
for
1
h
to
cure
the
composite.
A
post-cure
process
was
performed
by
maintaining
the
temperature
at
140
°C
for
2
h.
After
the
assembled
mold
was
cooled
to
room
temperature,
the
composite
was
demolded
(Scheme
2).
Layer
up
Preform
...
Transfer
into
mold
111111
111
Cure
and
demokl
hd 1=#
Inject
epox
resin
Scheme
2.
Schematic
of
the
resin
transfer
molding
(RTM)
processing
used
to
fabricate
the
fire-retardant
composites
using
non-fire-retardant
epoxy
resin.
Materials
2018,
11,
2554
4
of
18
The
composites
that
were
preformed
by
the
tackifiers
containing
0,
23,
38,
and
48
wt.%
MRP
are
denoted
as
COMP-T(0%P),
COMP-T(23°AT),
COMP-T(38%P),
and
COMP-T(48%P),
respectively.
The
cured
resin
samples
containing
0,
2.5,
5,
and
7.5
wt.%
MRP
are
denoted
as
Resin(0%P),
Resin(2.5%P),
Resin(5%P),
and
Resin(7.5%P),
respectively.
2.3.
Rheological
Measurement
Dynamic
rheological
measurements
of
resins
were
performed
with
a
dynamic
oscillatory
rheometer
(AR2000,
TA
Instrument,
New
Castle,
DE,
USA).
For
each
measurement,
the
sample
was
heated
from
80
to
110
°C
at
a
heating
rate
of
2
°C/min.
2.4.
Thermal
Analysis
Thermogravimetric
(TG)
experiments
were
performed
using
an
analyzer
(STA
409PC,
Netzsch
Group,
Bavaria,
Germany).
Eight
milligram
samples
were
heated
in
alumina
pans
from
room
temperature
to
800
°C
at
a
heating
rate
of
10
°C/min
in
air.
To
evaluate
the
effect
of
MRP
loading
on
the
thermal
stability
of
ET3228,
a
simple
sum
of
char
residuals
of
MRP
and
ET3228
(m
sum
)
was
calculated
on
the
basis
of
the
Equation
(1).
The
char
residuals
of
ET3228
and
MRP
at
600
°C
are
m
t
and
m
p
,
respectively,
and
c
t
and
c
p
are
the
weight
percentages
of
ET3228
and
MRP
in
the
flame-retardant
tackifier,
respectively.
Maim
=
mt
X
Ct
+
mp
X
Cp
(1)
2.5.
Morphological
Characterization
Optical
microscopy
(BX51
M,
Olympus,
Tokyo,
Japan)
was
performed
to
observe
the
distribution
of
the
ET3228/MRP
tackifier
in
the
molded
composites.
The
cross-sections
of
the
samples
were
wet-polished
to
prepare
smooth
surfaces
for
microscopy
observation.
Scanning
electron
microscopy
(SEM;
S-4800,
Hitachi, Hitachi,
Japan)
was
carried
out
to
examine
the
sample
morphology
before
and
after
combustion.
2.6.
Fire-Retardancy
Evaluation
A
butane
torch
(BS-271,
Wenzhou
Wans
Torch
Co.
Ltd.,
Wenzhou,
China)
was
used
to
ignite
the
composite
samples
for
10
s;
then,
we
measured
the
burning
time
ti.
If
the
flame
died,
we
reignited
it
for
10
s
more
and
again
measured
the
secondary
burning
time
t
2
.
The
results
of
the
flammability
examinations
are
shown
in
Video
S1
in
the
Supplementary
Material.
The
limiting
oxygen
index
(LOI)
was
measured
on
an
oxygen
index
meter
(PX-01-005,
Suzhou
Phinix
Instrument
Co.,
Ltd.,
Suzhou,
China)
according
to
ISO
4589-2:1996.
The
size
of
the
specimens
used
in
the
test
was
150
mm
x
10
mm
x
4
mm.
The
t-test
was
used
to
evaluate
whether
the
differences
among
the
LOI
values
of
the
resins
or
composites
with
different
MRP
contents
was
statistically
significant.
The
cone
calorimeter
test
was
performed
on
a
Fire
Testing
Technology
cone
calorimeter
(Stanton
Redcroft,
Fire
Testing
Technology
Ltd.,
East
Grinstead,
UK)
according
to
ISO
5660-1:2015
[
].
Each
sample
(with
a
size
of
100
mm
x
100
mm
x
3
mm)
was
placed
on
an
aluminum
foil
and
exposed
horizontally
to
an
external
heat
flux
of
50
kW/m
2
.
2.7.
Mechanical
Properties
The
mechanical
properties
were
determined
on
a
fatigue-testing
system
(8803,
Instron,
Norwood,
MA,
USA).
The
tensile,
compression,
and
interlaminar
shear
strengths
and
modulus
were
each
tested
according
to
ASTMD3093
[46],
ASTMD6641
[
],
and
ASTMD2344
[48],
respectively.
The
t-test
was
used
to
evaluate
whether
the
differences
among
the
strengths
and
moduli
of
the
composites
with
different
MRP
contents
were
statistically
significant.
Sp
r
ing
bac
k
Ang
le
(
°)
25
20
15
10
5
Materials
2018,
11,
2554
5
of
18
3.
Results
and
Discussion
3.1.
Preforms
by
Flame-Retardant-Loaded
Tackifier
We
prepared
the
fire-retardant
ET3228
tackifiers
with
MRP
weight
contents
of
23,
38,
and
48
wt.%
and
included
pristine
ET3228
as
a
control.
The
main
reason
for
using
the
ET3228
tackifier
to
stabilize
the
preform
dimension
was
that
its
viscosity
is
much
higher
than
that
of
the
EP3228
resin
at
the
injection
temperature
of
50
°C.
To
clarify
the
effect
of
MRP
addition
on
tackifier
viscosities,
we
registered
the
temperature
dependences
of
complex
viscosity
at
1
Hz
for
these
flame-retardant-loaded
tackifiers.
We
found
that
the
addition
of
MRP
enhanced
the
viscosity
of
tackifier
resins.
As
noted,
with
the
MRP
content
increasing
from
0
to
48
wt.%,
the
complex
viscosity
of
the
tackifier
at
85
°C
increased
from
381
to
940
Pa•s
(Figure
la).
The
increase
in
viscosity
normally
restrains
tackifier
resins
from
wetting
fabrics
and
reduces
their
control
over
the
preform
dimension
[49].
We
employed
the
U-shape
bending
test
to
evaluate
the
effect
of
MRP
addition
on
its
control
over
the
preform
dimension
[
].
When
we
prepared
the
preforms,
the
total
amount
of
the
tackifier
consisting
of
the
ET3228
resin
and
MRP
varied
from
4.0
to
7.75
wt.%
of
the
preform,
whereas
the
amount
of
the
ET3228
resin
was
fixed
at
4.0
wt.%.
Ten
plies
of
fabric
were
molded
on
an
85
°C
U-shape
bending
tool,
where
the
pressure
was
produced
through
vacuum
bagging
for
10
min
(Scheme
).
After
cooling
to
room
temperature
(25
°C),
the
preform
was
removed
from
the
tool
and
the
springback
angle
was
monitored
over
time.
The
springback
angles
for
the
tackifiers
with
0,
23,
38,
and
48
wt.%
MRP
additives
were
found
to
be
1.0,
2.3,
3.5,
and
5.2°,
respectively
(Figure
1
b).
Compared
to
the
springback
angle
for
the
preform
without
using
a
tackifier
(21.1°),
the
springback
angles
for
the
modified
tackifiers
were
much
smaller.
Based
on
previous
studies
and
our
experiences,
a
preform
with
a
springback
angle
smaller
than
is
acceptable,
especially
when
the
fiber
volume
faction
is
controlled
to
be
lower
than
55
vol.%
[50].
The
preforming
control
was
improved
by
extending
the
duration
of
preforming
or
processing
at
a
temperature
higher
than
85
°C.
As
such,
MRP-loaded
tackifiers
more
thoroughly
impregnated
the
fabrics
[9,50].
However,
to
fairly
evaluate
the
effects
of
MRP
addition
on
the
composite
properties,
we
did
not
change
any
condition
or
parameter
of
RTM
processing
in
this
study.
a
T(0%P)
T(23%P)
T(38%P)
V
T(48%P)
A
A
77A
10min
nn
2h
12h
77
2411
2000
1500
e
g
woo
.t>
5.
1
500
0
80
90
100 110
T(0%P)
T(23%P) T(38%P)
T(48%P
no
Temperature
(°C)
tackifier
Figure
1.
(a)
Temperature
dependence
of
complex
viscosities
for
the
ET3228
tackifiers
with
microencapsulated
red
phosphorus
(MRP)
contents
ranging
from
0
to
48
wt.%.
(b)
Springback
angles
monitored
for
the
fabrics
preformed
by
0-48
wt.%
MRP-loaded
ET3228
tackifiers
and
the
fabric
without
preforming.
All
the
t-tests
carried
out
on
the
springback
angles
of
the
fabrics
preformed
by
different
tackifiers
resulted
in
a
p
value
less
than
0.01,
indicating
that
the
dependence
of
the
springback
angle
on
MRP
content
is
statistically
significant.
We
used
TG
experiments
to
evaluate
the
effects
of
MRP
addition
on
the
thermal
stability
of
ET3228
tackifiers
in
air.
The
results
of
thermogravimetric
analysis
(TGA)
and
the
char
residual
at
600
°C
are
shown
in
Figure
2a,b,
respectively,
and
the
first
derivative
TGA
results
(DTG)
are
shown
in
Figure
Si.
The
ET3228
tackifier
without
MRP
presents
a
two-step
decomposition
with
two
maximum
mass-loss
Res
i
due
we
ig
ht
(
/o
)
40
30
20
10
0
Materials
2018,
11,
2554
6
of
18
rates
at
395
and
525
°C.
The
first
step
is
the
fast
thermal
decomposition
of
the
ET3228
tackifier
and
the
second
step
is
further
oxidative
degradation
of
char
residues
[51].
The
tackifier
without
MRP
did
not
leave
any
residue
at
600
°C.
MRP
decomposed
in
a
single
step
with
a
maximum
loss
rate
at
445
°C
and
left
a
residue
of
34.6
wt.%.
By
mixing
the
ET3228
tackifier
with
MRP,
its
decomposition
behaviors
occurred
in
three
steps,
with
maximum
mass-loss
rates
at
around
330,
475
and
600
°C.
This
indicates
that
the
MRP
loading
promotes
the
thermal
decomposition
of
the
tackifier
but
distinctively
depresses
the
oxidative
degradation
of
char
residues.
In
addition,
an
increase
in
weight
was
observed
around
450
°C
for
most
MRP-loaded
tackifiers,
which
could
be
attributed
to
the
oxidation
of
phosphorus
in
the
condensed
phase
[
].
100
t
80
60
40
-
T(0%P)
-
T(23%P)
20
-
-
T(38%P)
-
T(48%P)
_
-
MRP
200
400
600
Temperature
(°C)
b
0
23
38
48
MRP
contents
in
tackifier
(wt%)
Figure
2.
(a)
Thermogravimetric
registrations
and
(b)
residues
when
heated
to
600
°C
under
an
air
atmosphere
for
ET3228
tackifiers
with
MRP
contents
in
the
range
of
0
to
48
wt.%.
By
comparing
the
decomposition
behaviors
and
residues
at
600
°C,
we
found
that
the
addition
of
MRP
dramatically
enhanced
thermal
stability
under
an
air
atmosphere.
With
increasing
MRP
content
from
0
to
48
wt.%,
the
residues
of
the
flame-retardant
tackifiers
increased
sharply
at
600
°C.
Notably,
their
residues
were
much
more
than
the
simple
sum
of
the
residues
of
the
two
individual
materials.
For
example,
the
residue
for
the
tackifier
with
48
wt.%
MRP
was
40.0%,
whereas
the
simple
sum
of
the
residues
of
the
two
individual
materials
was
calculated
to
be
17.3
wt.%.
The
details
of
this
calculation
were
described
in
the
Materials
and
Methods
section.
This
phenomenon
is
attributed
to
the
enhanced
formation
of
carbonaceous
layers
containing
a
large
amount
of
stable
P-0
and
P—C
complexes
[
].
The
char
residue
inhibited
oxygen
penetration
and
heat
conduction
and
retarded
the
thermal
decomposition
process.
3.2.
Self-Extinguishing
RTM
Composites
with
Flammable
Resins
The
MRP
weight
percentages
in
the
resins
were
0,
2.5,
5,
and
7.5
wt.%
for
the
composites
preformed
by
the
tackifiers,
with
MRP
contents
of
0,
23,
38,
and
48
wt.%,
respectively.
The
MRP
and
tackifier
contents
in
the
composites
and
preforms
are
shown
in
Tables
S1
and
S2,
respectively.
The
design
of
the
flame-retardant
tackifier
possessed
significantly
beneficial
properties
for
fabricating
flame-retardant
composites
with
typical
RTM
processing
conditions.
Due
to
the
MRP
being
loaded
in
the
tackifiers,
the
viscosity
of
the
EP3228
resin
could
not
be
influenced
by
the
flame-retardant
MRP,
which
ensures
the
high
resin
fluidity
necessary
for
wetting
fabrics
and
driving
air
bubbles
at
50
°C.
To
evaluate
the
effects
of
MRP
addition
on
resin
viscosity,
we
directly
mixed
the
MRP
into
the
EP3228
resin
in
similar
contents
and
tested
the
resin
viscosity.
As
shown
in
Figure
S2,
the
viscosity
of
the
EP3228
resin
increased
from
0.35
(0
wt.%
MRP)
to
0.63
(7.5
wt.%
MRP),
corresponding
to
an
80%
viscosity
increase.
The
80%
increased
viscosity
for
the
MRP-loaded
EP3228
resin
weakened
its
fluidity,
which
made
it
difficult
to
mold
the
composites
following
the
standard
RTM
approach
I
43.
1
0
5
Otim
-
ow
C
a
500ftm
.
.
1
4
Age
,
500µm
'4•2
r
4
t
4
Materials
2018,
11,2554
7
of
18
for
EP3228
resin.
Adopting
a
higher
injection
pressure
and
temperature
may
help
resins
to
wet
fabrics
and
improve
the
composite
quality.
However,
we
found
the
13
1..tm
MRP
particles
would
be
trapped
in
the
outer
fabric
layer
of
the
preforms
as
the
gap
between
fibers
of
fabrics
is
around
2
Even
for
nanoparticles,
previous
works
have
revealed
their
non-uniform
distribution
and
uneven
performance
[14,36].
These
issues
normally
restrain
the
use
of
non-soluble
additive
modified
resins
in
RTM
processing.
Employing
flame-retardant
tackifiers
leads
to
MRP
concentration
on
the
fabric
surface.
We
used
an
optical
microscope
to
observe
the
composite
section
(Figure
3).
MRP
additives
were
enriched
on
the
surface
without
penetrating
the
fabric.
This
was
a
reasonable
finding
as
the
tackifier
resin
did
not
soften
at
the
injection
temperature
and
the
MRP
additives
were
anchored
on
the
fabric
surface
during
resin
injection.
The
unique
MRP
distribution,
i.e.,
the
enrichment
of
MRP
particles
on
the
fabric
surface,
should
be
responsible
for
the
composites'
excellent
fire
retardancy,
which
is
demonstrated
below.
Figure
3.
Spatial
distribution
of
MRP
in
resin
transfer
molded
(RTM)
composites.
Optical
microscope
images
for
the
sections
of
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
containing
(a)
0,
(b)
23,
(c)
38
and
(d)
48
wt.%
MRP.
The
sections
of
the
composites
are
vertical
to
the
reinforced
fabrics.
We
used
thermogravimetric
measurements
to
investigate
the
effect
of
MRP
addition
on
the
thermal
decomposition
of
composites.
As
shown
in
Figure
a,
the
weight
of
the
composite
without
MRP
loading
in
the
tackifier
reached
a
constant
when
the
temperature
increased
above
600
°C.
This
indicates
that
all
the
organic
materials
were
gasified
and
that
the
remaining
63.2
wt.%
residue
originated
from
the
glass
fabric
[54].
Thus,
the
weight
percentage
of
the
organic
materials
was
36.8
wt.%
for
the
composite.
Materials
2018,
11,
2554
8
of
18
Notably,
the
organic
materials
consisted
not
only
of
the
resin
and
tackifier,
but
also
the
fiber
sizing
coating,
the
presence
of
which
was
proven
by
its
3.8
wt.%
weight
loss
at
600
°C.
To
better
understand
the
effect
of
MRP
addition
on
the
decomposition
behavior,
we
calculated
the
residues
relative
to
the
organic
materials
at
600
°C
for
all
composites
by
dividing
the
difference
between
the
composite
residue
with
MRP
and
that
of
the
composite
without
MRP
with
the
weight
percentage
of
the
organic
materials
(36.8
wt%),
which
is
shown
in
Figure
4b.
In
this
case,
the
density
difference
between
the
MRP
additive
and
the
tackifier
were
not
considered.
With
the
addition
of
MRP,
their
residues
relative
to
the
organic
materials
increased
sharply.
As
noted,
the
residue
reached
30.9
wt.%
when
the
tackifier
was
loaded
with
48
wt.%
MRP
(corresponding
to
7.5
wt.%
of
the
whole
resin).
The
30.9
wt.%
residue
was
much
more
than
the
residue
that
the
MRP
additive
produced
by
itself,
as
it
only
accounted
for
7.5
wt.%
of
the
resin.
Thus,
this
result
provides
evidence
for
the
enhanced
thermal
stability
of
the
composites.
100
b
30
90
80
on
70
bA
.
20
10
COMP-T(O%P)
-
COMP-T(23%P)
60
r
-
COMP-T(38%P)
rz
a
COMP-T(48%P)
50
Glass
Fiber
0
200
400
600
0
23
38
48
Temperature
(°C)
MRP
contents
in
tackifier
(wt%)
Figure
4.
(a)
Thermogravimetric
registrations
with
a
heating
rate
of
10
°C
min
-1
under
an
air
atmosphere
and
(b)
calculated
residues
(600
°C)
of
the
organic
material,
including
the
resin,
tackifier,
and
fiber
sizing
coating,
for
the
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
with
MRP
content
in
the
range
of
0
to
48
wt.%.
To
evaluate
the
fire
retardancy
of
the
composites
in
the
lab,
we
directly
examined
the
flammability
of
the
composite
samples
[55,56].
The
results
are
summarized
in
Figure
a—d
and
Table
S3.
The
composite
sample
without
MRP
was
easily
ignited
and
kept
burning
until
all
the
organic
materials
were
consumed.
However,
when
23
wt.%
MRP
(corresponding
to
2.5
wt.%
of
the
whole
resin)
was
added
into
the
tackifier,
the
composite
self-extinguished
within
25
s
(t
1
).
When
the
MRP
content
in
the
tackifier
was
higher
than
38
wt.%
(corresponding
to
5
wt.%
of
the
whole
resin),
the
first
burning
time
was
dramatically
reduced
to
less
than
4
s,
indicating
excellent
fire
retardancy.
By
increasing
the
MRP
weight
content
to
48
wt.%,
the
first
burning
time
was
further
reduced
to
2
s.
We
also
noted
that
the
secondary
burning
behaviors
of
the
composites
were
very
similar
to
the
first
behaviors
presented
above.
To
accurately
define
the
composites'
fire
retardancy,
we
used
the
limiting
oxygen
index
(LOI)
measurement
and
cone
calorimeter
test
to
clarify
the
flammability
of
composites.
The
LOI
is
defined
as
the
minimum
fraction
of
oxygen
in
an
oxygen
nitrogen
mixture
to
sustain
combustion
after
ignition.
The
LOI
value
of
the
composite
without
MRP
was
about
25.9%
(Figure
e).
By
introducing
23
wt.%
MRP
(corresponding
to
2.5
wt.%
of
the
whole
resin)
into
the
tackifier,
the
LOI
value
of
composite
increased
to
46.0%.
By
increasing
the
MRP
content
to
38
wt.%
(corresponding
to
5.0
wt.%
of
the
whole
resin)
in
the
tackifier,
the
LOI
reached
a
maximum
value
at
48.8%,
which
meant
an
88.42%
increase
in
fire
retardancy.
We
noted
that
any
further
increase
in
MRP
content
did
not
increase
the
LOI
value
any
more.
The
LOI
value
for
the
composite
preformed
by
the
tackifier
containing
48
wt.%
MRP
became
smaller
(46.4%).
For
the
cured
EP3228
resins
mixed
with
MRP,
their
LOI
values
increased
from
20.4
(0
wt.%)
to
35.1
(7.5
wt.%),
which
is
also
shown
in
Figure
5e.
However,
it
was
difficult
to
evaluate
Lim
it
ing
Oxyg
en
In
dex
(
%)
60
50
40
30
20
10
Materials
2018,
11,
2554
9
of
18
the
effect
of
MRP
distribution
on
fire
performance
by
comparing
the
two
samples,
as
the
fabrics
also
contribute
to
the
fire
retardance
of
composites.
b
120s
25s
111111&._
4s
d
Os
2
s
S
ao
o
co
(7'
r
-
'
‘c:
A
O
O
0
Figure
5.
Digital
pictures
taken
during
the
first
burning
of
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
containing
(a)
0,
(b)
23,
(c)
38,
and
(d)
48
wt.%
MRP,
where
the
time
was
counted
after
the
sample
had
been
ignited
for
10
s.
(e)
Limiting
oxygen
index
(LOI)
values
for
the
RTM
composite
samples
preformed
by
the
ET3228
tackifiers
containing
0,
23,
38,
and
48
wt.%
MRP
(white
columns),
and
the
cured
EP3228
resin
samples
containing
0,
2.5,
5,
and
7.5
wt.%
MRP
(gray
columns).
All
the
t-tests
carried
out
on
the
LOI
values
of
the
composites
and
cured
resins
resulted
in
a
p
value
less
than
0.001,
indicating
that
the
dependences
of
LOI
value
on
MRP
content
for
the
composites
and
cured
resins
are
statistically
significant.
The
cone
calorimetry
tests
are
widely
considered
to
be
one
of
the
most
critical
bench-scale
measurements
in
the
field
of
fire
testing.
At
a
constant
heat
flux
of
50
kW/m
2
,
we
recorded
the
_a
-°-
COMP-T(0%P)
-0-
COMP-T(23%P)
-6-
COMP-T(38%P)
-
v
-
COMP-T(48%P)
-0-
COMP-T(0%P)
COMP-T(23%P)
COMP-T(38%P)
-
v
-
COMP-T(48%P)
c
i
's
250
200
50
gi's
50
F
.,
40
`
c
f
s
30
g
20
10
"rs
E-
1
©
0
Materials
2018,
11,
2554
10
of
18
heat-release
rate
curves,
weight
loss
curves
and
total
smoke
release
(TSR)
to
evaluate
the
composite
flammability
(Figure
6).
The
main
cone
calorimeter
testing
data
are
shown
in
Table
S4.
As
shown
in
Figure
6a,
the
heat
release
rate
for
the
composite
was
distinctively
reduced
by
adding
only
23
wt.%
MRP
(corresponding
to
2.5
wt.%
of
the
whole
resin)
into
the
tackifier.
By
further
increasing
the
MRP
content
in
the
tackifier,
the
heat
release
rate
tended
to
be
slightly
faster,
but
still
slower
than
the
composite
without
MRP.
To
understand
the
total
heat
release
of
the
composites,
we
integrated
the
heat-release
rate
curve
and
plotted
the
results
in
Figure
b.
We
noted
that
the
addition
of
23
wt.%
MRP
(2.5
wt.%
of
the
whole
resin)
into
the
tackifier
markedly
decreased
the
total
heat
release
from
46.12
to
33.02
MJ/kg.
This
corresponded
to
a
28.4%
reduction
in
the
total
heat
release.
Any
further
increase
in
MRP
contents
did
not
noticeably
reduce
the
total
heat
release.
The
total
heat
release
for
the
composite
preformed
by
the
tackifier
containing
48
wt.%
MRP
was
32.64
MJ/kg,
which
indicated
a
29.2%
reduction
in
the
total
heat
release.
Figure
c
presents
the
mass-loss
curves
of
these
composites
measured
by
the
CCT.
The
composites
preformed
by
the
ET3228
tackifiers
containing
23
wt.%
MRP
started
to
decompose
at
180
s,
while
the
neat
composite
started
at
135
s.
This
indicates
the
enhanced
thermal
stability
of
the
former.
By
increasing
the
amount
of
MRP
in
the
tackifiers,
the
decomposition
of
composites
started
earlier,
but
was
still
much
slower
than in
the
neat
composite.
All
the
MRP-loaded
composites
produced
more
residue
than
the
neat
sample.
Figure
6d
illustrates
the
TSR
of
the
composites
during
the
CCT.
The
TSR
showed
a
33%
increase
after
adding
23
wt.%
MRP
in
the
tackifier
when
compared
to
the
neat
epoxy
composite.
However,
the
TSR
did
not
noticeably
increase
for
composites
with
higher
MRP
contents
in
the
tackifier.
0
100
200
300
400
0
100
200
300
400
Time
(s)
Time
(s)
100
e
90
2
80
70
—•
3000
4
"
at
4
2000
e
1000
"
C
r
S
-
COMP-T(0%
P)
-
COMP-T(23%
P)
-
COMP-T(38%
P)
COMP-T(48%
P)
-
COMP-T(0%
P)
COMP-T(23%
P)
-
COMP-T(38%
P)
COMP-T(48%
P)
0
100
200
300
400
0
100
200
300
400
Time
(s)
Time
(s)
Figure
6.
(a)
Heat
release
rate,
(b)
total
heat
release,
(c)
mass
loss,
and
(d)
total
smoke
release
for
the
RTM
epoxy
composites
preformed
by
the
ET3228
tackifiers
with
a
MRP
content
in
the
range
of
0
to
48
wt.%.
The
reduction
in
heat
release
could
be
attributed
to
the
enhanced
char
formation
with
the
addition
of
MRP,
which
was
obvious
in
the
digital
images
of
the
composites
after
the
cone
calorimetry
tests
e
-
kJ
,
%-;
1-
Materials
2018,
11,
2554
11
of
18
(Figure
7).
The
char
residue
microstructures
for
the
composites
after
the
burning
test
were
also
observed
by
SEM.
As
shown
in
Figure
8a—d,
more
compressed
and
continuous
char
residues
were
formed
on
the
external
composite
surface
with
high
MRP
contents.
Figure
8e—h
illustrates
the
morphologies
of
the
composite
inner
layers
after
the
burning
test.
The
tackifier
and
resin
around
the
glass-fiber
fabrics
were
completely
burned
in
the
composite
without
MRP.
In
the
composites
containing
MRP,
residue
formation
was
clearly
observed
inside
the
composites,
and
the
residue
amount
correlated
well
with
the
MRP
content
in
the
tackifiers.
The
SEM
images
for
the
external
and
fracture
surfaces
of
the
composites
before
flaming
are
shown
in
Figure
S3
for
comparison.
1
C
.
••
,
.'
11
4.'
%
I
.,
4
Figure
7.
Digital
pictures
for
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
containing
(a)
0,
(b)
23,
(c)
38
and
(d)
48
wt.%
MRP
after
the
cone
calorimeter
test.
-
.4,71
,
M
N•yet..
:
To:,
Rit
a
`
t
1
1 ;
14
.
_.•
.
4w•
V.
...
7
4-
1..,34:
..,.
•,4
..
...
VT.
;
dct
-
to•
,
'4•
••
4%
'*
4
1;
1%,
-
.
1,
••
*Z
,
4
Materials
2018,11,
2554
12
of
18
Ill.
...
4
t
.4"
s••
'
•••
..••
A*
`
,It•
4
11P
••
*A 3.
O.
"
fr
we
,
.••
.•
.
20µm
.,
)
*
%
la
.
1
s
"
e
.
.0
.
:
,
t
:
S
e
- '
1.
•.
.
4
.
6.1•
11
11
11
4...
*
...
kr
, .
e
l
:
,
il•
sr-1
616
:
4
;
0,
c
%
..
ie
,
.
...
:
1
ll
II
..,
I .
41
.
l
iv
)
'
.
,
.
a.
1
'
*
$
1
.
1
_
.
•.
*
IV'
q
lil
V
e
a
v.
t_
a
2kn
'
sr
7
g
,
.
"
i
lb
P
to
,,
w
il
un
••
p
.1.
'
1
-
**'
-
20prti
St
.4
4
to.
41r•
11
(
1
d
tiX
Pk.
t*
tiR
Ite
de
1.
4
Figure
8.
Cont.
_
irda
4
Z -
"‘"
4'
-
•••
-
•••••••
-20pm
Materials
2018,
11,
2554
13
of
18
J4
fr...04
4
'
6411104•11414.
----------
_
Figure
8.
Scanning
electron
microscopy
(SEM)
images
of
the
external
surfaces
of
burned
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
containing
(a)
0,
(b)
23,
(c)
38
and
(d)
48
wt.%
MRP,
and
inner
surfaces
of
burned
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
containing
(e)
0,
(f)
23,
(g)
38
and
(h)
48
wt.%
MRP.
We
considered
attributing
this
dramatically
enhanced
fire
retardancy
to
the
unique
heterogenous
distribution
of
MRP
additives
in
the
composites.
The
MRP
additives
were
enriched
on
the
fabric
surface
without
being
diluted
by
the
injected
resin
because
the
tackifier
did
not
soften
during
resin
injection.
MRP
enrichment
promoted
the
formation
of
concentrated
char
residues
on
the
fabric
surface
[57].
The
thick
char
layer
on
the
fabrics
could
act
as
a
barrier
to
effectively
retard
heat-flux
and
flammable-volatile
permeation,
suppressing
burning
propagation.
We
noted
that
the
MRP
additives
could
not
penetrate
and
distribute
into
the
resins
inside
fabrics
(Figure
),
which
prevents
promoting
the
charring
of
resins
inside
fabrics.
Due
to
the
size
of
the
MRP
particles
being
much
larger
than
the
gap
between
the
fibers,
the
MRP
particles
could
not
penetrate
inside
the
fabrics
and
the
optimal
conditions
could
not
be
reached
for
further
improving
the
flame
retardancy
of
the
composites.
However,
further
work
combining
the
fire-retardant
epoxy
resin
with
our
MRP
modified
preforms
is
expected
to
produce
a
much
better
fire-retardant
composite.
3.3.
Composite
Mechanical
Properties
We
successfully
demonstrated
that
the
application
of
a
flame-retardant
tackifier
significantly
enhanced
the
flame
resistance
of
RTM
composites.
The
unique
beneficial
aspect
to
this
approach
is
that
when
molding
fire-retardant
composites
by
RTM,
we
did
not
need
to
modify
the
resins.
However,
this
technique
can
potentially reduce
the
mechanical
strength
of
the
composites,
as
it
introduces
MRP
additives
to
the
fabric-resin
interface.
With
respect
to
this
issue,
we
carried
out
a
systematic
study
on
the
mechanical
properties
of
the
composites.
The
interlamellar
shear
strength
slightly
decreased
when
23
wt.%
MRP
was
added
in
the
tackifier
but
recovered
with
further
loading
of
MRP
additives
(Figure
9a).
This
indicates
that
the
interlamellar
shear
strength
is
not
sensitive
to
MRP
loading
in
tackifier
resin.
We
observed
a
slight
fluctuation
(either
an
increase
or
decrease)
in
the
tensile
strength
and
a
very
slight
increase
in
the
compression
strength
for
the
composite
preformed
by
tackifiers
with
MRP
(Figure
a).
Thus,
these
data
do
not
support
the
notion
that
the
effect
of
the
MRP
addition
on
the
mechanical
strengths
of
composites
was
distinctive.
In
comparison,
the
effect
on
the
tensile
and
compression
moduli
was
clear
and
regular,
although
the
moduli
increase
with
MRP
loading
was
also
slight
(Figure
9b).
We
noted
that
without
loading
MRP,
the
interlamellar
shear
strength
(ILSS,
66.8
MPa)
-
a
—M—
Interlaminar
Shear
Strength
Tensile
Strength
—7—
Compression
Strength
_
4
A
4_______
A
ir
10
,______----*
+
I
0
0
0
23
38
48
MRP
contents
in
tackifier
(wt%)
Streng
t
h
(
MPa
)
600
500
400
300
200
100
30
Ts'
a,
(„)
20
o
10
0
2
10
0
Materials
2018,
11,
2554
14
of
18
was
much
higher
than
those
(30-40
MPa)
reported
previously
[58,59],
and
the
compression
and
tensile
strengths
were
comparable
to,
or
even
higher
than
those
reported
in
previous
works
[60-62].
Several
reasons
explain
the
indistinctive
effects
of
MRP
loading
on
the
mechanical
properties.
First
of
all,
the
small
size
(13
um)
of
the
MRP
additives
used
here
would
reduce
the
effect
that
adding
MRP
would
have
on
mechanical
properties
[
].
Previous
studies
found
that
the
mechanical
properties
are
sensitive
to
the
size
of
MRP
additives.
When
the
size
of
MRP
additives
is
smaller
than
16
um,
the
damage
caused
by
the
addition
of
MRP
to
the
mechanical
properties
can
be
greatly
reduced
[38].
In
addition,
the
MRP
additives
used
in
this
work
could
have
formed
a
strong
interfacial
interaction
between
the
encapsulating
layer
and
the
resins,
which
benefited
MRP
dispersion
and
the
retention
of
any
mechanical
properties
[38].
Finally,
increasing
viscosity
in
the
tackifiers
with
the
addition
of
MRP
could
also
help
improve
the
composites'
mechanical
properties.
As
we
discussed
above,
the
viscosity
increase
in
the
tackifiers
would
restrain
the
tackifier
resins
from
wetting
the
fibers
inside
the
fabric
tows.
This
would
also
reduce
tackifier
control
over
the
preform
dimension
and
leave
more
tackifiers
outside
the
fiber
tows.
However,
in
this
case,
the
void
content
would
decrease
and
the
composite
mechanical
properties
would
instead
improve
[49]
b
—M—
Compression
Modulus
-
—7—
Tensile
Modulus
0
23
38
48
MRP
contents
in
tackifier
(wt%)
Figure
9.
Mechanical
properties
of
RTM
epoxy
composites.
(a)
Interlaminar
shear,
tensile,
and
compression
strength
values
of
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
with
an
MRP
content
in
the
range
of
0
to
48
wt.%.
(b)
Tensile
and
compression
modulus
values
of
RTM
epoxy
composites
preformed
by
ET3228
tackifiers
with
an
MRP
content
in
the
range
of
0
to
48
wt.%.
All
the
t-tests
on
the
interlaminar
shear,
tensile,
and
compression
strengths
of
composites
resulted
in
a
p
value
less
than
0.05,
indicating
that
the
dependences
of
interlaminar
shear,
tensile
and
compression
strength
on
MRP
content
are
statistically
significant.
The
t-test
on
the
moduli
of
composites
resulted
in
a
p
value
larger
than
0.05,
indicating
that
the
dependences
of
tensile
and
compression
moduli
on
MRP
content
are
not
statistically
significant.
4.
Conclusions
The
introduction
of
fire-retardant
additives
or
building
blocks
into
resins
is
generally
adopted
to
improve
the
fire
retardancy
of
resins.
However,
the
concomitant
increase
in
resin
viscosity
and
the
enrichment
of
insoluble
additives
on
the
preform
outer
layers
remain
challenges
for
RTM
processing.
Targeting
this
issue,
we
developed
a
robust
approach
to
fabricate
self-extinguishing
RTM
composites
using
non-fire-retardant
epoxy
resins.
We
loaded
the
halogen-free
flame
retardant,
microencapsulated
red
phosphorus
(MRP)
into
tackifiers
instead
of
resins.
The
flame-retardant-loaded
tackifiers
retained
the
preform
shapes
and
integrity
of
the
preforms
during
resin
injection.
The
flame-retardant
enrichment
on
fabric
surfaces,
due
to
the
high
viscosity
of
tackifiers,
endowed
the
composites
with
excellent
fire
retardancy
while
causing
almost
no
effect
on
the
mechanical
properties.
After
loading
48
wt.%
MRP
in
the
tackifier
(accounting
for
5
wt.%
of
the
whole
resin),
the
flammable
composite
became
Materials
2018,
11,
2554
15
of
18
self-extinguishing
within
two
seconds
while
retaining
a
springback
angle
of
5.2°.
Its
LOI
index
was
enhanced
by
79.2%,
and
its
total
heat
release
was
reduced
by
29.2%.
As
this
approach
does
not
modify
the
resins,
the
RTM
processing
does
not
need
to
be
reoptimized.
One
of
most
attractive
properties
of
this
approach
is
that
it
makes
the
insoluble
flame
retardant
adaptable
to
RTM
processing.
This
approach
is
basically
applicable
to
all
RTM
resins
for
fabricating
composites
if
an
appropriate
tackifier
is
available.
This
approach
is
not
limited
to
fire-retardant
loading,
but
could
be
extended
to
load
other
functions,
such
as
radar
absorbing,
conductivity,
etc.,
into
RTM
composites.
Supplementary
Materials:
The
following
are
available
online
at
http:/
/www.mdpi.com/1996-1944/11/12/25541
s1:
Figure
S1:
Derivative
thermogravimetric
analysis
(DTG)
for
MRP
and
ET3228
tackifiers
with
microencapsulated
red
phosphorus
(MRP)
contents
in
the
range
of
0-48
wt.%,
Figure
S2:
Temperature
dependence
of
complex
viscosities
for
the
EP3228
resin
mixing
with
MRP
content
in
the
range
of
0-7.5 wt.%,
Figure
S3:
Scanning
electron
microscopy
(SEM)
images
of
the
external
surfaces
of
resin
transfer
molded
(RTM)
epoxy
composites
preformed
by
the
ET3228
tackifiers
containing
(a)
0,
(b)
23,
(c)
38,
and
(d)
48
wt.%
MRP,
and
fracture
surfaces
of
RTM
epoxy
composites
preformed
by
the
ET3228
tackifiers
containing
(e)
0,
(f)
23,
(g)
38,
and
(h)
48
wt.%
MRP
before
flaming,
Table
S1:
MRP
and
ET3228
contents
in
composites,
Table
S2:
MRP
and
ET3228
contents
in
preforms,
Table
S3:
First
burning
time
(t
1
),
secondary
burning
time
(t
2
),
and
dripping
observed
for
RTM
composite
samples
preformed
by
the
ET3228
tackifiers
containing
0,
23,
38,
and
48
wt.%
MRP,
and
cured
resin
samples
containing
0,
2.5,
5,
and
7.5
wt.%
MRP,
Table
S4:
Combustion
calorimetric
test
(CCT)
results
for
the
composites.
Video
S1:
Flammability
examination.
Author
Contributions:
B.Z.,
J.B.,
Z.H.,
and
Y.H.
initiated
the
project
and
designed
the
experiments.
S.Y.
and
L.Z.
prepared
the
fire-retardant-loaded
tackifiers,
preforms,
and
composites.
S.Y.,
L.Z.,
and
J.W.
evaluated
the
viscosity,
preforming
control,
and
thermal
stability
of
the
tackifiers.
Z.G.,
S.Y.,
Z.H.,
and
J.L.
evaluated
the
thermal
and
mechanical
properties
and
the
fire
retardance
of
the
composites.
Z.G.,
J.L.,
and
J.W.
conducted
the
SEM
experiments.
Z.G.
and
Z.H.
wrote
the
original
draft.
B.Z.,
J.B.,
Z.Y.,
and
Y.H.
reviewed
and
revised
the
manuscript.
Funding:
This
research
was
funded
by
the
National
Natural
Science
Foundation
of
China
(NSFC)
(21474014,
21704013),
the
Natural
Science
Foundation
of
Shanghai
(14ZR1400200),
and
the
China
Postdoctoral
Science
Foundation
(2017M611416).
Conflicts
of
Interest
The
authors
declare
no
conflict
of
interest.
References
1.
Yu,
A.;
Ramesh,
P.;
Sun,
X.;
Bekyarova,
E.;
Itkis,
M.E.;
Haddon,
R.C.
Enhanced
Thermal
Conductivity
in
a
Hybrid
Graphite
Nanoplatelet—Carbon
Nanotube
Filler
for
Epoxy
Composites.
Adv.
Mater.
2008,
20,
4740-4744.
[CrossRef]
2.
Wang,
L.;
Zhang,
C.;
Gong,
W.;
Ji,
Y.;
Qin,
S.;
He,
L.
Preparation
of
Microcellular
Epoxy
Foams
through
a
Limited-Foaming
Process:
A
Contradiction
with
the
Time-Temperature-Transformation
Cure
Diagram.
Adv.
Mater.
2018,
30,
1703992.
[CrossRef]
[PubMed]
3.
Watanabe,
H.;
Kunitake,
T.
A
Large,
Freestanding,
20
nm
Thick
Nanomembrane
Based
on
an
Epoxy
Resin.
Adv.
Mater.
2007,
19,
909-912.
[CrossRef]
4.
Morell,
M.;
Ramis,
X.;
Ferrando,
F.;
Yu,
Y.;
Serra,
A.
New
improved
thermosets
obtained
from
DGEBA
and
a
hyperbranched
poly(ester-amide).
Polymer
2009,
50,
5374-5383.
[CrossRef]
5.
Yuan,
L.;
Liang,
G.Z.;
Xie,
J.Q.;
Li,
L.;
Guo,
J.
The
permeability
and
stability
of
microencapsulated
epoxy
resins.
I.
Mater.
Sci.
2007,
42,
4390-4397.
[CrossRef]
6.
Demir,
B.;
Henderson,
L.C.;
Walsh,
T.R.
Design
Rules
for
Enhanced
Interfacial Shear
Response
in
Functionalized
Carbon
Fiber
Epoxy
Composites.
ACS
Appl.
Mater.
Interfaces
2017,
9,11846-11857.
[CrossRef]
[PubMed]
7.
Song,
K.;
Chen,
D.;
Polak,
R.;
Rubner,
M.F.;
Cohen,
RE.;
Askar,
K.A.
Enhanced
Wear
Resistance
of
Transparent
Epoxy
Composite
Coatings
with
Vertically
Aligned
Halloysite
Nanotubes.
ACS
Appl.
Mater.
Interfaces
2016,
8,
35552-35564.
[CrossRef]
8.
Wang,
Y.;
Raman
Pillai,
S.K.;
Che,
J.;
Chan-Park,
M.B.
High
Interlaminar
Shear
Strength
Enhancement
of
Carbon
Fiber/Epoxy
Composite
through
Fiber-
and
Matrix-Anchored
Carbon
Nanotube
Networks.
ACS
Appl.
Mater.
Interfaces
2017,
9,
8960-8966.
[CrossRef]
9.
Wu,
Q.;
Zhang, C.;
Liang,
R.;
Wang,
B.
Fire
retardancy
of
a
buckypaper
membrane.
Carbon
2008,
46,
1164-1165.
[CrossRef]
Materials
2018,
11,
2554
16
of
18
10.
Hohne,
C.C.;
Wendel,
R.;
Kabisch,
B.;
Anders,
T.;
Henning,
F.;
Kroke,
E.
Hexaphenoxycyclotriphosphazene
as
FR
for
CFR
anionic
PA6
via
T-RTM:
A
study
of
mechanical
and
thermal
properties.
Fire
Mater.
2017,
41,
291-306.
[CrossRef]
11.
Li,
C.;
Kang,
N.J.;
Labrandero,
S.D.;
Wan,
J.;
Gonzalez,
C.;
Wang,
D.Y.
Synergistic
Effect
of
Carbon
Nanotube
and
Polyethersulfone
on
Flame
Retardancy
of
Carbon
Fiber
Reinforced
Epoxy
Composites.
Ind.
Eng.
Chem.
Res.
2013,
53,
1040-1047.
[CrossRef]
12.
Steffen,
S.;
Bauer,
M.;
Decker,
D.;
Richter,
F.
Fire-retardant
hybrid
thermosetting
resins
from
unsaturated
polyesters
and
polysilazanes.
J.
Appl.
Polym.
Sci.
2014,
131,
40375.
[CrossRef]
13.
Bulgakov,
B.A.;
Sulimov,
A.V.;
Babkin,
A.V.;
Afanasiev,
D.V.;
Solopchenko,
A.V.;
Afanaseva,
E.S.;
Kepmana,
A.V.;
Avdeeva,
V.V.
Flame-retardant
carbon
fiber
reinforced
phthalonitrile
composite
for
high-temperature
applications
obtained
by
resin
transfer
molding.
Mendeleev
Commun.
2017,
27,
257-259.
[CrossRef]
14.
Pomazi,
A.;
Toldy,
A.
Particle
Distribution
of
Solid
Flame
Retardants
in
Infusion
Moulded
Composites.
Polymers
2017,
9,250.
[CrossRef]
15.
Tang,
Y.;
Zhuge,
J.;
Gou,
J.;
Chen,
R-H.;
lbeh,
C.;
Hu,
Y.
Morphology,
thermal
stability,
and
flammability
of
polymer
matrix
composites
coated
with
hybrid
nanopapers.
Polym.
Adv.
Technol.
2011,
22,1403-1413.
[CrossRef]
16.
Dong,
C.;
Wirasaputra,
A.;
Luo,
Q.;
Liu,
S.;
Yuan,
Y.;
Zhao,
J.;
Fu,
Y.
Intrinsic
Flame-Retardant
and
Thermally
Stable
Epoxy
Endowed
by
a
Highly
Efficient,
Multifunctional
Curing
Agent.
Materials
2016,
9,
1008.
[CrossRef]
[PubMed]
17.
Zhang,
Y.;
Yu,
B.;
Wang,
B.;
Liew,
K.;
Song,
L.;
Wang,
C.;
Hu,
Y.
Highly
Effective
P-P
Synergy
of
a
Novel
DOPO-Based
Flame
Retardant
for
Epoxy
Resin.
Ind.
Eng. Chem.
Res.
2017,56,1245-1255.
[CrossRef]
18.
Zhou,
X.;
Qiu,
S.;
Xing,
W.;
Gangireddy,
C.S.R.;
Gui,
Z.;
Hu,
Y.;
Qiu,
S.
Hierarchical
Disulfide
Hybrid
Structure
for
Enhancing
the
Flame
Retardancy
and
Mechanical
Property
of
Epoxy
Resins.
ACS
Appl.
Mater.
Interfaces
2017,
9,29147-29156.
[CrossRef]
19.
Rwei,
S.;
Chen,
Y.;
Chiang,
W.;
Ting,
Y.
A
Study
of
the
Curing
and
Flammability
Properties
of
Bisphenol
A
Epoxy
Diacrylate
Resin
Utilizing
a
Novel
Flame
Retardant
Monomer,
bis[di-acryloyloxyethyl]-
p-tert-butyl-phenyl
Phosphate.
Materials
2017,
10,
202.
[CrossRef]
20.
Liu,
Y.L.
Flame-retardant
epoxy
resins
from
novel
phosphorus-containing
novolac.
Polymer
2001,
42,
3445-3454.
[CrossRef]
21.
Shieh,
J.Y.;
Wang,
C.S.
Synthesis
of
novel
flame
retardant
epoxy
hardeners
and
properties
of
cured
products.
Polymer
2001,
42,7617-7625.
[CrossRef]
22.
Liu,
W;
Wang,
Z.;
Xiong,
L.;
Zhao,
L.
Phosphorus-containing
liquid
cycloaliphatic
epoxy
resins
for
reworkable
environment-friendly
electronic
packaging
materials.
Polymer
2010,
51,4776-4783.
[CrossRef]
23.
Sun,
J.;
Wang,
X.;
Wu,
D.
Novel
spirocyclic
phosphazene-based
epoxy
resin
for
halogen-free
fire
resistance:
Synthesis,
curing
behaviors,
and
flammability
characteristics.
ACS
Appl.
Mater.
Interfaces
2012,
4,4047-4061.
[CrossRef]
[PubMed]
24.
Oliwa,
R.;
Heneczkowski,
M.;
Oleksy,
M.;
Galina,
H.
Epoxy
composites
of
reduced
flammability.
Compos.
Part.
B-Eng.
2016,
95,
1-8.
[CrossRef]
25.
Guzel,
G.;
Sivrikaya,
0.;
Deveci,
H.
The
use
of
colemanite
and
ulexite
as
novel
fillers
in
epoxy
composites:
Influences
on
thermal
and
physico-mechanical
properties.
Compos.
Part.
B-Eng.
2016,
100,
1-9.
[CrossRef]
26.
Khalili,
P.;
Tshai,
K.Y.;
Hui,
D.;
Kong,
I.
Synergistic
of
ammonium
polyphosphate
and
alumina
trihydrate
as
fire
retardants
for
natural
fiber
reinforced
epoxy
composite.
Compos.
Part.
B-Eng.
2017,
114,
101-110.
[CrossRef]
27.
Wang,
X.;
Hu,
Y.;
Song,
L.;
Xing,
W.;
Lu,
H.;
Lv,
P.;
Jie,
G.
Flame
retardancy
and
thermal
degradation
mechanism
of
epoxy
resin
composites
based
on
a
DOPO
substituted
organophosphorus
oligomer.
Polymer
2010,
51,2435-2445.
[CrossRef]
28.
Feng,
Y.;
Li,
X.;
Zhao,
X.;
Ye,
Y.;
Zhou,
X.;
Liu,
H.;
Liu,
C.;
Xie,
X.
Synergetic
Improvement
in
Thermal
Conductivity
and
Flame
Retardancy
of
Epoxy/Silver
Nanowires
Composites
by
Incorporating
"Branch-Like"
Flame-Retardant
Functionalized
Graphene.
ACS
Appl.
Mater.
Interfaces
2018,
10,
21628-21641.
[CrossRef]
29.
Zhang,
X.;
He,
Q.;
Gu,
H.;
Colorado,
H.A.;
Wei,
S.;
Guo,
Z.
Flame-Retardant
Electrical
Conductive
Nanopolymers
Based
on
Bisphenol
F
Epoxy
Resin
Reinforced
with
Nano
Polyanilines.
ACS
Appl.
Mater.
Interfaces
2013,
5,
898-910.
[CrossRef]
Materials
2018,
11,
2554
17
of
18
30.
Yu,
B.;
Xing,
W.;
Guo,
W.;
Qiu,
S.;
Wang,
X.;
Lo,
S.;
Hu,
Y.
Thermal
exfoliation
of
hexagonal
boron
nitride
for
effective
enhancements
on
thermal
stability,
flame
retardancy
and
smoke
suppression
of
epoxy
resin
nanocomposites
via
sol-gel
process.
J.
Mater.
Chem.
A
2016,
4,
7330-7340.
[CrossRef]
31.
Gerard,
C.;
Fontaine,
G.;
Bellayer,
S.;
Bourbigot,
S.
Reaction
to
fire
of
an
intumescent
epoxy
resin:
Protection
mechanisms
and
synergy.
Polym.
Degrad.
Stab.
2012,
97,1366-1386.
[CrossRef]
32.
Gu,
H.;
Guo,
J.;
He,
Q.;
Tadakamalla,
S.;
Zhang,
X.;
Yan,
X.;
Huang,
Y.;
Colorado,
H.A.;
Wei,
S.;
Guo,
Z.
Flame-Retardant
Epoxy
Resin
Nanocomposites
Reinforced
with
Polyaniline-Stabilized
Silica
Nanopartides.
Ind.
Eng.
Chem.
Res.
2013,
52,
7718-7728.
[CrossRef]
33.
Kashiwagi,
T.;
Du,
F.;
Douglas,
J.F.;
Winey,
K.I.;
Harris,
R.H.,
Jr.;
Shields,
J.R.
Nanoparticle
networks
reduce
the
flammability
of
polymer
nanocomposites.
Nat.
Mater.
2005,
4,
928-933.
[CrossRef]
[PubMed]
34.
Konnicke,
D.;
Kuhn,
A.;
Mahrholz,
T.;
Sinapius,
M.
Polymer
nanocomposites
based
on
epoxy
resin
and
ATH
as
a
new
flame
retardant
for
CFRP:
Preparation
and
thermal
characterisation.
J.
Mater.
Sci.
2011,
46,
7046-7055.
[CrossRef]
35.
Xu,
M.;
Ma,
K.;
Jiang,
D.;
Zhang,
J.;
Zhao,
M.;
Guo,
X.;
Shao,
Q.;
Wujcik,
E.;
Li,
B.;
Guo,
Z.
Hexa-[4-(glycidyloxycarbonyl)
phenoxylcyclotriphosphazene
chain
extender
for
preparing
high-performance
flame
retardant
polyamide
6
composites.
Polymer
2018,
146,
63-72.
[CrossRef]
36.
Van
Velthem,
P.;
Bailout,
W;
Dumont,
D.;
Daoust,
D.;
Sclavons,
M.;
Cordenier,
F.;
Pardoen,
T.;
Devaux,
J.;
Bailly,
C.
Phenoxy
nanocomposite
carriers
for
delivery
of
nanofillers
in
epoxy
matrix
for
resin
transfer
molding
(RTM)-manufactured
composites.
Compos.
Part.
A
2015,
76,
82-91.
[CrossRef]
37.
Savas,
L.A.;
Deniz,
T.K.;
Tayfun,
U.;
Dogan,
M.
Effect
of
microcapsulated
red
phosphorus
on
flame
retardant,
thermal
and
mechanical
properties
of
thermoplastic
polyurethane
composites
filled
with
huntite&hydromagnesite
mineral.
Polym.
Degrad.
Stab.
2017,
135,
121-129.
[CrossRef]
38.
Kim,
J.;
Yoo,
S.;
Bae,
J.Y.;
Yun,
H.C.;
Hwang,
J.;
Kong,
B.-S.
Thermal
stabilities
and
mechanical
properties
of
epoxy
molding
compounds
(EMC)
containing
encapsulated
red
phosphorous.
Polym.
Degrad.
Stab.
2003,
81,
207-213.
[CrossRef]
39.
Liu,
J.;
Guo,
Y.;
Zhang,
Y.;
Liu,
H.;
Peng,
S.;
Pan,
B.;
Ma,
J.;
Niu,
Q.
Thermal
conduction
and
fire
property
of
glass
fiber-reinforced
high
impact
polystyrene/magnesium
hydroxide/microencapsulated
red
phosphorus
composite.
Polym.
Degrad.
Stab.
2016,
129,
180-191.
[CrossRef]
40.
Cao,
Z.J.;
Dong,
X.;
Fu,
T.;
Deng,
S.B.;
Liao,
W.;
Wang,
Y.Z.
Coated
vs.
naked
red
phosphorus:
A
comparative
study
on
their
fire
retardancy
and
smoke
suppression
for
rigid
polyurethane
foams.
Polym.
Degrad.
Stab.
2017,
136,
103-111.
[CrossRef]
41.
Liu,
J.;
Peng,
S.;
Zhang,
Y.;
Chang,
H.;
Yu,
Z.;
Pan,
B.;
Lu,
C.;
Ma,
J.;
Niu,
Q.
Influence
of
microencapsulated
red
phosphorus
on
the
flame
retardancy
of
high
impact
polystyrene/magnesium
hydroxide
composite
and
its
mode
of
action.
Polym.
Degrad.
Stab.
2015,
121,
208-221.
[CrossRef]
42.
Xu,
L.;
Xiao,
L.;
Jia,
P.;
Goossens,
K.;
Liu,
P.;
U,
H.;
Cheng,
C.;
Huang,
Y.;
Bielawski,
C.W.;
Geng,
J.
Lightweight
and
Ultrastrong
Polymer
Foams
with
Unusually
Superior
Flame
Retardancy.
ACS
Appl.
Mater.
Interfaces
2017,
9,
26392-26399.
[CrossRef]
43.
Wang,
Z.;
Wu,
G.;
Hu,
Y.;
Ding,
Y.;
Hu,
K.;
Fan,
W.
Thermal
degradation
of
magnesium
hydroxide
and
red
phosphorus
flame
retarded
polyethylene
composites.
Polym.
Degrad.
STable
2002,
77,
427-434.
[CrossRef]
44.
Peters,
E.N.
Flame-retardant
ther49moplastics.
I.
Polyethylene-red
phosphorus.
J.
Appl.
Polym.
Sci.
1979,
24,
1457-1464.
[CrossRef]
45.
International
Organization
for
Standardization
Publication.
Reaction-to-Fire
Tests—Heat
Release,
Smoke
Production
and
Mass
Loss
Rate—Part
1:
Heat
Release
Rate
(Cone
Calorimeter
Method)
and
Smoke
Production
Rate
(Dynamic
Measurement);
ISO
5660-1:2015;
International
Organization
for
Standardization
Press:
Geneva,
Switzerland,
2015.
46.
American
Society
for
Testing
and
Materials
International.
Tensile
Testing
of
Ppolymer
Matrix
Composites;
ASTM
D3039;
American
Society
for
Testing
and
Materials
International
Press:
West
Conshohocken,
PA,
USA,
2000.
47.
American
Society
for
Testing
and
Materials
International.
Standard
Test
Method
for
Compressive
Properties
of
Polymer
Matrix
Composite
Materials
Using
a
Combined
Loading
Compression
(CLC)
Test
Fixture;
ASTM
D6641;
American
Society
for
Testing
and
Materials
International
Press:
West
Conshohocken,
PA,
USA,
2016.
48.
American
Society
for
Testing
and
Materials
International.
ASTM
D2344;
Standard
Test
Method
for
Short-Beam
Strength
of
Polymer
Matrix
Composite
Materials
and
Their
Laminates;
American
Society
for
Testing
and
Materials
International
Press:
West
Conshohocken,
PA,
USA,
2000.
Materials
2018,
11,
2554
18
of
18
49.
Shih,
C.H.;
Liu,
Q.;
Lee,
L.J.
Vacuum-assisted
resin
transfer
molding
using
tackified
fiber
preforms.
Polym.
Composite.
2001,
22,
721-729.
[CrossRef]
50.
Shih,
C.H.;
Lee,
L.J.
Tackification
of
Textile
Fiber
Preforms
in
Resin
Transfer
Molding.
J.
Compos.
Mater.
2001,
35,1954-1981.
[CrossRef]
51.
Wang,
X.;
Hu,
Y.;
Song,
L.;
Xing,
W.;
Lu,
H.
Thermal
degradation
mechanism
of
flame
retarded
epoxy
resins
with
a
DOPO-substitued
organophosphorus
oligomer
by
TG-FTIR
and
DP-MS.
J.
Anal. Appl.
Pyrol.
2011,
92,
164-170.
[CrossRef]
52.
Braun,
U.;
Schartel,
B.
Flame
Retardant
Mechanisms
of
Red
Phosphorus
and
Magnesium
Hydroxide
in
High
Impact
Polystyrene.
Macromol.
Chem.
Phys.
2004,
205,
2185-2196.
[CrossRef]
53.
Wu,
Q.;
Lii,
J.;
Qu,
B.
Preparation
and
characterization
of
microcapsulated
red
phosphorus
and
its
flame-retardant
mechanism
in
halogen-free
flame
retardant
polyolefins.
Polym.
Int.
2003,
52,1326-1331.
[CrossRef]
54.
Moon,
C.R.;
Bang,
B.R.;
Choi,
W.J.;
Kang,
G.H.;
Park,
S.-Y.
A
technique
for
determining
fiber
content
in
FRP
by
thermogravimetric
analyzer.
Polym.
Test.
2005,24,376-380.
[CrossRef]
55.
Manfredi,
A.;
Carosio,
F.;
Ferruti,
P.;
Ranucci,
E.;
Alongi,
J.
Linear
polyamidoamines
as
novel
biocompatible
phosphorus-free
surface-confined
intumescent
flame
retardants
for
cotton
fabrics.
Polym.
Degrad.
Stab.
2018,
151,52-64.
[CrossRef]
56.
Guo,
W.;
Wang,
X.;
Zhang,
P.;
Liu,
J.;
Song,
L.;
Hu,
Y.
Nano-fibrillated
cellulose-hydroxyapatite
based
composite
foams
with
excellent
fire
resistance.
Carbohyd.
Polym.
2018,195,71-78.
[CrossRef]
[PubMed]
57.
Ballistreri,
A.;
Montaudo,
G.;
Puglisi,
C.;
Scamporrino,
E.;
Vitalini,
D.;
Calgari,
S.
Mechanism
of
flame
retardant
action
of
red
phosphorus
in
polyacrylonitrile.
J.
Polym.
Sci.
Polym.
Chem.
1983,
21,
679-689.
[CrossRef]
58.
Liu,
Y.;
Yang,
J.;
Xiao,
H.;
Qua,
C.;
Feng,
Q.;
Fu,
S.;
Shindo,
Y.
Role
of
matrix
modification
on
interlaminar
shear
strength
of
glass
fibre/epoxy
composites.
Compos.
Part.
B-Eng.
2012,
43,95-98.
[CrossRef]
59.
Kwon,
D.;
Shin,
P.;
Kim,
J.;
Baek,
Y.;
Park,
H.;
DeVries,
K.;
Park,
J.
Interfacial
properties
and
thermal
aging
of
glass
fiber/epoxy
composites
reinforced
with
SiC
and
SiO2
nanoparticles.
Compos.
Part.
B-Eng.
2017,
130,
46-53.
[CrossRef]
60.
Manjunatha,
C.;
Taylor,
A.;
Kinloch,
A.;
Sprenger,
S.
The
tensile
fatigue
behaviour
of
a
silica
nanoparticle-modified
glass
fibre
reinforced
epoxy
composite.
Compo.
Sci.
Technol.
2010,
70,
193-199.
[CrossRef]
61.
Kim,
J.;
Kwon,
D.;
Shin,
P.;
Beak,
Y.;
Park,
H.;
DeVries,
K.;
Park,
J.
Interfacial
properties
and
permeability
of
three
patterned
glass
fiber/epoxy
composites
by
VARTM.
Compos.
Part.
B-Eng.
2018,
148,
61-67.
[CrossRef]
62.
Shin,
P.;
Wang,
Z.;
Kwon,
D.;
Choi,
J.;
Sung,
I.;
Jin,
D.;
Kang,
S.;
Kim,
J.;
DeVries,
K.;
Park,
J.
Optimum
mixing
ratio
of
epoxy
for
glass
fiber
reinforced
composites
with
high
thermal
stability.
Compos.
Part.
B-Eng.
2015,
79,
132-137.
[CrossRef]
©
2018
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/).