polymers
Article
Tannic Acid as a Bio-Based Modifier of
Epoxy/Anhydride Thermosets
Xiaoma Fei, Fangqiao Zhao, Wei Wei, Jing Luo, Mingqing Chen * and Xiaoya Liu
The Key Laboratory of Food Colloids and Biotechnology, Ministry of Education,
School of Chemical and Material Engineering, Jiangnan University, Wuxi 214122, China;
[email protected] (X.F.); [email protected] (F.Z.); [email protected] (W.W);
[email protected] (J.L.); [email protected] (X.L.)
* Correspondence: [email protected]; Tel.: +86-510-8532-6066
Academic Editor: Antonio Pizzi
Received: 22 June 2016; Accepted: 11 August 2016; Published: 26 August 2016
Abstract: Toughening an epoxy resin by bio-based modifiers without trade-offs in its modulus,
mechanical strength, and other properties is still a big challenge. This paper presents an approach
to modify epoxy resin with tannic acid (TA) as a bio-based feedstock. Carboxylic acid-modified
tannic acid (TA–COOH) was first prepared through a simple esterification between TA and
methylhexahydrophthalic anhydride, and then used as a modifier for the epoxy/anhydride curing
system. Owing to the chemical modification, TA–COOH could easily disperse in epoxy resin
and showed adequate interface interaction between TA–COOH and epoxy matrix, in avoid of
phase separation. The use of TA–COOH in different proportions as modifier of epoxy/anhydride
thermosets was studied. The results showed that TA–COOH could significantly improve the
toughness with a great increase in impact strength under a low loading amount. Moreover, the
addition of TA–COOH also simultaneously improved the tensile strength, elongation at break and
glass transition temperature. The toughening and reinforcing mechanism was studied by scanning
electron microscopy (SEM), dynamic mechanical analysis (DMA) and thermal mechanical analysis
(TMA), which should be owned to the synergistic effect of good interface interaction, aromatic
structure, decreasing of cross linking density and increasing of free volume. This approach allows us
to utilize the renewable tannic acid as an effective modifier for epoxy resin with good mechanical
and thermal properties.
Keywords: tannic acid; epoxy resin; bio-based; toughening
1. Introduction
Because of their excellent performance properties, good processability and low cost, epoxy resins
are used as one of the most versatile thermosetting polymers with a wide range of applications
including coatings, adhesives, structural composites and electronic materials [1–3]. However, their
inherent brittle nature because of high degree of chemical crosslinking severely limits their uses
in many applications. Up to now, various modifiers, such as rubber [4], thermoplastic [5], clay [6],
hyperbranched polymers (HBPs) [7,8], nanomaterial [9,10] and layered double hydroxides (LDHs) [11]
have been incorporated into epoxy resin, which can significantly improve the toughness of the epoxy
resin. However, the key point for all these approaches is to toughen epoxy resin without sacrificing its
strength, modulus and other thermal properties.
In recent years, the fast depletion of petroleum reserve and increasing environmental problems
have led to a growing interest in the use of bio-based sustainable feedstock in the synthesis of bio-based
chemicals and products. In this regard, some researches had focused on the synthesis and utilization
of renewable material as efficient epoxy modifier [12–16]. Liu reported an approach to toughen
Polymers 2016, 8, 314; doi:10.3390/polym8090314
www.mdpi.com/journal/polymers
Polymers 2016, 8, 314
2 of 12
epoxy resin with lignin, which could significantly toughen and simultaneously reinforce the epoxy
resin. However, the glass transition temperature (Tg ) of the resin was reduced to a certain extent [13].
Satheesh reported the utilization of chitosan to modify epoxy resin and investigated the influence of
chitosan loading on the thermal, mechanical, and morphological properties. The results showed that
when the chitosan loading increased above 5 wt %, chitosan tended to agglomerate in epoxy resin,
with the formation of clear phase separation. What is more, the tensile strength decreased after adding
chitosan [14]. It is important to note that toughening an epoxy resin by bio-based modifiers without
trade-offs in its modulus, mechanical strength, and thermal properties is still a big challenge.
Tannic acid (TA) is water-soluble high molecular weight polyphenolic compounds, mostly
extracted from plants and microorganisms. It has a macromolecular structure composed of gallic units
and abundant terminal phenolic hydroxyl groups. Owing to such a structure, TA shows remarkable
properties and is widely used in many application, such as coatings, adsorption and antibacterial
materials, mucoadhesive compounds, separator for lithium-ion batteries, and nanomaterials [17–21].
It has been proved that a highly branched structure can introduce more internal cavities or free
volumes in cured thermosets, which is favor of improving toughness [22–24]. In addition, an aromatic
structure is beneficial to the Tg and modulus of cured epoxy thermoset. In this context, the use of tannic
acid as an epoxy modifier seems to be an interesting proposition because its architectural structure is
similar to the hyperbranched aromatic polyester with abundant terminal phenolic hydroxyl groups.
Furthermore, the utilization of bio-based tannic acid will contribute to the long-term sustainability
including environmental and health safety issues [25].
Therefore, the motivation for this work is to utilize TA as a bio-based modifier for epoxy resins
with simultaneous improvement in toughness and other properties. We had tried to add TA into
epoxy formula without any modification. However, because of intermolecular hydrogen bonds,
Van der Waals interactions and π–π stacking of aromatic groups, TA is immiscible with epoxy resin
and tends to precipitation during curing. Therefore, a certain extent of chemical modification of TA
is necessary to improve the miscibility of TA in epoxy matrix and enhance the interface interaction
between TA and epoxy matrix.
In the study reported herein, a carboxylic acid-modified tannic acid (TA–COOH) was
prepared through the simple esterification between TA and a commercial epoxy hardener
methylhexahydrophthalic anhydride (MeHHPA). In addition, TA–COOH was added into the
epoxy formula and used as an all-purpose epoxy modifier. It was found that TA–COOH could
significantly improve the toughness of cured thermosets with an increase in impact strength under low
loading amount, and simultaneously improve the elongation at break, Tg and strength. In addition,
other thermal properties and fracture surfaces were also studied.
2. Experimental Section
2.1. Materials
TA (the main component is a kind of polygallic acid as given in Scheme 1, whose purity is 99%)
and MeHHPA (99%) were purchased from Aladdin (Shanghai, China). Epoxy resin (diglycidyl ether
of bisphenol A, trade name E-51) with an epoxy equivalent weight of 171–175 g per equivalent was
obtained from Kukdo Chemical (Kunshan, China). All other reagents were of analytical grade and
used as received from Sinopharm Chemical Reagent (Shanghai, China).
Polymers
2016,2016,
8, 314
Polymers
8, 314
3 3ofof1212
Scheme
1. 1.Synthetic
Scheme
Syntheticroute
routeof
ofTA–COOH.
TA–COOH.
Synthesis
of Carboxylic
Acid-FunctionalizedTannic
TannicAcid
Acid(TA–COOH)
(TA–COOH)
2.2. 2.2.
Synthesis
of Carboxylic
Acid-Functionalized
2 mM)
was
dissolved
mLofofpyridine
pyridinein
inaaround-bottomed
round-bottomed flask
TA TA
(3.4(3.4
g, 2g,mM)
was
dissolved
inin
3030mL
flaskwith
withaamagnetic
magnetic
stirrer
and
a
gas
inlet
to
fill
the
flask
with
N
2
.
Then
16.8
g
of
MeHHPA
(100
mM)
was
added
stirrer and a gas inlet to fill the flask with N2 . Then 16.8 g of MeHHPA (100 mM) was addedtotothe
the
◦C
solution.
mixture
was
stirred
°Cfor
for3636h.h.After
Afterreaction,
reaction, the
the mixture
mixture was
solution.
TheThe
mixture
was
stirred
atat
8080
was precipitated
precipitatedinto
into
diethyl
ether
to remove
pyridine
and
theunreacted
unreactedMeHHPA.
MeHHPA.The
The precipitate
precipitate was
diethyl
ether
to remove
pyridine
and
the
was dissolved
dissolvedininTHF
THF
and
precipitated
into
diethyl
ether
twice,
and
then
dried
under
vacuum
at
30
°C
to
give
a
brown
solid
◦
and precipitated into diethyl ether twice, and then dried under vacuum at 30 C to give a brown
product.
solid
product.
Fabrication
of TA–COOH/Epoxy
Thermosets
2.3. 2.3.
Fabrication
of TA–COOH/Epoxy
Thermosets
TA–COOH/epoxy thermosets were prepared by the following procedure. A required
TheThe
TA–COOH/epoxy
thermosets were prepared by the following procedure. A required amount
amount of TA–COOH was dispersed in epoxy resin (E-51) under mechanical stirring
at 60 °C. Then,
of TA–COOH was dispersed in epoxy resin (E-51) under mechanical stirring at 60 ◦ C. Then, the curing
the curing agent (MeHHPA) and catalyst (ethyl triphenyl phosphonium bromide) were added to the
agent (MeHHPA) and catalyst (ethyl triphenyl phosphonium bromide) were added to the above
above mixture (the epoxy and curing agent were in a 1:1 equivalent ratio, and the catalyst loading
mixture (the epoxy and curing agent were in a 1:1 equivalent ratio, and the catalyst loading was
was 1 wt % of the total weight). Finally, the mixture was degassed and poured into mold. The samples
1 wt % of the total weight). Finally, the mixture was degassed and poured into mold. The samples for
for thermal and mechanical characterization were cured using the following profile:
80 °C for 1.5 h,
◦ for 1.5 h, 100 ◦ C
thermal
and
were
usingthermosets
the following
100 °C
formechanical
1 h, 120 °Ccharacterization
for 1 h, and 140 °C
forcured
2 h. These
withprofile:
0.5, 1.080
andC2.0
wt % of TA–
◦ C for 1 h, and 140 ◦ C for 2 h. These thermosets with 0.5, 1.0 and 2.0 wt % of TA–COOH
for 1COOH
h, 120were
coded as TA–COOH0.5, TA–COOH1.0 and TA–COOH2.0, respectively. Neat epoxy was
wereprepared
coded asfollowing
TA–COOH0.5,
TA–COOH1.0
TA–COOH2.0,
the same
procedure asand
mentioned
above. respectively. Neat epoxy was prepared
following the same procedure as mentioned above.
2.4. Characterization
2.4. Characterization
The 13C NMR spectra of TA–COOH were recorded on a Bruker AV400M nuclear magnetic
13 C-NMR spectra of TA–COOH were recorded on a Bruker AV400M nuclear magnetic
The
resonance
spectrometer (400 MHz, Bruker, Karlsruhe, Germany). In order to calculate the degree of
resonance
spectrometer
(400
MHz,
Bruker, Karlsruhe,
Germany).
calculate
thetransform
degree of
modification,
an inverse
gated
decoupling
technique with
the time In
of order
4 h wastoused.
Fourier
modification,
an inverse
gated
technique
with the
timeScientific
of 4 h was
used.iS50
Fourier
transform
infrared (FTIR)
spectra
weredecoupling
obtained using
a Thermo
Fisher
Nicolet
spectrometer
infrared
(FTIR)
spectra
wereWaltham,
obtained MA,
using
a Thermo
Scientific
Nicolet
iS50 spectrometer
(Thermo
Fisher
Scientific,
USA)
at roomFisher
temperature
in the
wavenumber
range of
600–4000
cm−1Scientific,
. TGA experiment
was
carried
out
a METTLER
TOLEDO
(Mettler range
Toledo,of
(Thermo
Fisher
Waltham,
MA,
USA)
at on
room
temperature
in theTGA/1
wavenumber
Greifensee,
in N2 atmosphere
rate of 10
°C min−1TGA/1
. DSC was
recorded
on
600–4000
cm−1Switzerland)
. TGA experiment
was carriedwith
out aonheating
a METTLER
TOLEDO
(Mettler
Toledo,
◦
−
1
a
NETZSCH
204
F1
thermal
analyzer
(Mettler
Toledo,
Greifensee,
Switzerland).
The
samples
of
~5 a
Greifensee, Switzerland) in N2 atmosphere with a heating rate of 10 C min . DSC was recorded on
mg in weight
placed
in aluminum
pans under
nitrogenSwitzerland).
atmosphere. The
Dynamic
mechanical
NETZSCH
204 F1 were
thermal
analyzer
(Mettler Toledo,
Greifensee,
samples
of ~5 mg
analysis
(DMA)
wasinconducted
onpans
a TAunder
Instruments
DMA
Q800 (TADynamic
Instruments,
Newcastle,
UK)
in weight
were
placed
aluminum
nitrogen
atmosphere.
mechanical
analysis
(DMA) was conducted on a TA Instruments DMA Q800 (TA Instruments, Newcastle, UK) at a heating
Polymers 2016, 8, 314
Polymers 2016, 8, 314
4 of 12
4 of 12
rate of 3 ◦ C min−1 and a frequency of 1 Hz under an air atmosphere. The linear coefficients of thermal
−1 and
expansion
were
measured
using
a METTLER
TMA/SDTA841e
thermal
analyzer
at a heating
rate
of 3 °C min
a frequencyTOLEDO
of 1 Hz under
an air atmosphere.
Themechanical
linear coefficients
(Mettler
Toledo,
Greifensee,
Switzerland).
tensile
strengths ofTOLEDO
the cured TMA/SDTA841e
hybrids were characterized
of thermal
expansion
were
measured The
using
a METTLER
thermal
(Mettler
Toledo,
Greifensee,
tensile
strengths
the527:1993.
cured
bymechanical
an Instronanalyzer
1185 test
machine
(Instron
Corp., Switzerland).
Canton, MA,The
USA)
according
to of
ISO
hybrids were
characterized
by anwere
Instron
1185 teston
machine
Corp., tester
Canton,
MA, USA)
Un-notched
impact
strength tests
performed
a Ceast(Instron
Resil impact
(CEAST,
Turin,
according
to ISO
527:1993.
Un-notched
strengthattests
performed
a Ceast Resil
Italy)
according
to ISO
179:1982.
For eachimpact
composition,
leastwere
5 samples
wereon
measured.
Theimpact
fracture
tester (CEAST,
Turin,toughness
Italy) according
to ISO
179:1982. For
composition,
least 5 samples
were
surfaces
from fracture
tests were
investigated
by aeach
HITACHI
S-4800atfield-emission
scanning
measured.
The
fracture
surfaces
from
fracture
toughness
tests
were
investigated
by
a
HITACHI
Selectron microscope (FESEM, HITACHI, Tokyo, Japan). Rheological measurements were performed
4800
field-emission
scanning
electron
microscope
(FESEM,
HITACHI,
Tokyo,
Japan).
Rheological
on a discovery DHR-2 hybrid rheometer (TA Instruments, Newcastle, UK) equipped with cone and
measurements
were
performed
on a discovery
DHR-2
hybrid
rheometer
(TA All
Instruments,
Newcastle,
plate
geometry (25
mm
cone diameter,
1.986◦ cone
angle,
50 mm
gap size).
the experiments
were
UK)
equipped
with
cone
and
plate
geometry
(25
mm
cone
diameter,
1.986°
cone
angle,
50
mm
gap
◦
performed at 25 C.
size). All the experiments were performed at 25 °C.
3. Results and Discussion
3. Results and Discussion
3.1. Synthesis and Characterization
3.1. Synthesis and Characterization
The commercial TA is given as C76 H52 O46 . Its chemical structure was given in Scheme 1. It has
The commercial
TA is linked
given astoCa76glucose
H52O46. Its
chemical
structure it
was
given in
1. Ithydroxyl
has 10
10 esterified
galloyl groups
core.
Theoretically,
contains
25Scheme
phenolic
esterified
galloyl
groups
linked
to
a
glucose
core.
Theoretically,
it
contains
25
phenolic
hydroxyl
groups, which could serve as reactive sites and be exploited for the functionalization of TA through
groups,
which could
serve
reactive
and modified
be exploited
the functionalization
of TA through
various
reactions
[26–28].
Inas
this
work,sites
TA was
by for
a pyridine-catalyzed
esterification
with
various
reactions
[26–28].
In
this
work,
TA
was
modified
by
a
pyridine-catalyzed
esterification
with
a common epoxy hardener MeHHPA, generating terminal carboxyl groups. The terminal carboxyl
a common epoxy hardener MeHHPA, generating terminal carboxyl groups. The terminal carboxyl
groups could react with epoxide groups in curing process and then provide a good interface interaction
groups could react with epoxide groups in curing process and then provide a good interface
between TA–COOH and epoxy matrix. The proposed structure of TA–COOH is shown in Scheme 1.
interaction between TA–COOH and epoxy matrix. The proposed structure of TA–COOH is shown in
First, the characterization of the TA–COOH was carried out by FTIR spectroscopy, as shown in
Scheme 1. First, the characterization of the TA–COOH was carried out by FTIR spectroscopy, as
Figure 1a. TA shows a broader OH band, which is attributed to the great hydrogen bonding interaction
shown in Figure 1a. TA shows a broader OH band, which is attributed to the great hydrogen bonding
of phenol groups. After modification, the OH region weakens and a new carboxylic group band in the
interaction of phenol groups. After modification, the OH region weakens and a new carboxylic group
region of 2500−3500 cm−1 appears (as
arrow has pointed out). The carbonylic region was also given
band in the region of 2500−3500 cm−1 appears (as arrow has pointed out). The carbonylic region was
−1
in also
the inset.
As
can
be
seen,
compared
to
the IR spectrum
TA, a new
absorption
peak at 1730
cm
given in the inset. As can be seen, compared
to the IRof
spectrum
of TA,
a new absorption
peak
at
occurrs,
which
couldwhich
be assigned
the stretching
C=O inofthe
ester
−1 occurrs,
1730 cm
could betoassigned
to the vibration
stretching of
vibration
C=O
in groups
the estergenerated
groups
−1 and 2959 cm−1 are due to the
bygenerated
the ring-opening
of
anhydride.
Moreover,
the
bands
at
2804
cm
−1
by the ring-opening of anhydride. Moreover, the bands at 2804 cm and 2959 cm−1 are due
aliphatic
C–H vibrations
of MeHHPA
moietymoiety
in the TA–COOH.
to the aliphatic
C–H vibrations
of MeHHPA
in the TA–COOH.
13 C-NMR spectroscopy. Figure 1b shows
The
structure
The
structureofofTA–COOH
TA–COOHwas
wasfurther
furtherconfirmed
confirmed by
by 13C
NMR spectroscopy. Figure 1b shows
13
13C NMRspectrum
thethe C-NMR
of
TA–COOH
with
the
corresponding
assignments.ItItcan
canbebeseen,
seen,after
after
spectrum of TA–COOH with the corresponding assignments.
modification,
TA–COOH
shows
a resonance
at 176
ppm,
which
can be
to theto–COOH
groups
modification,
TA–COOH
shows
a resonance
at 176
ppm,
which
canassigned
be assigned
the –COOH
generated
by anhydride.
The resonances
at 23, 29,at
33,23,
41,29,119
133 and
ppm133
areppm
assigned
to the aliphatic
groups generated
by anhydride.
The resonances
33,and
41, 119
are assigned
to the
C from MeHHPA
moiety.
The resonances
at 109,
are assigned
toin
C aliphatic
from MeHHPA
moiety. The
resonances
at 109, 140,
146, 140,
151 146,
and 151
155 and
ppm155
areppm
assigned
to the C
the C inring
benzene
benzene
from ring
TA. from TA.
Figure1.1.(a)
(a)FTIR
FTIRspectra
spectraof
ofTA
TAand
andTA–COOH
TA–COOH and
and (b)
(b) 13
13C-NMR
Figure
C-NMRspectrum
spectrumofofTA–COOH.
TA–COOH.
Polymers 2016, 8, 314
Polymers 2016, 8, 314
5 of 12
5 of 12
Because of steric hindrance affect, not all the 25 hydroxyl groups of TA are equally active. Thus,
Because of steric hindrance affect, not all the 25 hydroxyl groups of TA are equally active.
the degree of modification of MeHHPA was calculated from the integration of the aliphatic and
Thus, the degree of modification of MeHHPA was calculated from the integration of the aliphatic
aromatic signals. After calculation, it is found that about 15 hydroxyl groups of TA reacted with
and aromatic signals. After calculation, it is found that about 15 hydroxyl groups of TA reacted with
MeHHPA. Although the modification degree is not as high as expected, TA–COOH still shows good
MeHHPA. Although the modification degree is not as high as expected, TA–COOH still shows good
solubility in epoxy resin by stirring at 60 ◦°C for 2 h. In addition, the terminal carboxyl groups offer a
solubility in epoxy resin by stirring at 60 C for 2 h. In addition, the terminal carboxyl groups offer a
good interface interaction between TA–COOH and epoxy matrix.
good interface interaction between TA–COOH and epoxy matrix.
3.2.
3.2. Rheological
Rheological Properties
Properties
Generally,
when epoxy
epoxy resin
resin is
is toughened
toughened by
a linear
linear liquid
liquid rubber,
its viscosity
viscosity increases
increases to
Generally, when
by a
rubber, its
to aa
great
extent.
However,
for
hyperbranched
toughener,
it
usually
has
a
low
melt
viscosity
because
of
great extent. However, for hyperbranched toughener, it usually has a low melt viscosity because of its
its
globular
structure.
Such
a
low
melt
viscosity
is
good
for
processing,
which
shows
a
very
little
globular structure. Such a low melt viscosity is good for processing, which shows a very little increase
increase
in the viscosity
thanrubber.
liquid rubber.
The rheological
of theepoxy
neat epoxy
and epoxy
in the viscosity
than liquid
The rheological
studiesstudies
of the neat
and epoxy
resin resin
with
with
different
TA–COOH
loading
were
performed.
All
of
the
samples
have
Newtonian
different TA–COOH loading were performed. All of the samples have Newtonian behaviors behaviors
and show
and
show
similar rheological
range of studied
frequencies
studied
(Figure
2). As
similar
rheological
behavior inbehavior
the rangeinofthe
frequencies
(Figure
2). As
expected,
theexpected,
viscosity
the
viscosity
values
of
the
samples
with
different
TA–COOH
loadings
show
only
a
little
increase
as
values of the samples with different TA–COOH loadings show only a little increase as that of neat
that
of
neat
epoxy.
Such
a
low
viscosity
is
still
benefit
for
processability.
epoxy. Such a low viscosity is still benefit for processability.
Figure
resins.
Figure 2.
2. Viscosity
Viscosity versus
versus shear
shear rate of neat epoxy and TA–COOH modified epoxy resins.
3.3.
3.3. Curing
Curing Study
Study
As
an aim
aimtotomaintain
maintain
a certain
interface
interaction
between
the TA–COOH
and
the matrix
epoxy
As an
a certain
interface
interaction
between
the TA–COOH
and the
epoxy
matrix
after
curing,
we
selected
anhydride
as
curing
agent.
The
use
of
anhydride
as
curing
agents
after curing, we selected anhydride as curing agent. The use of anhydride as curing agents has been
has
been extensively
reportedHere
[29,30].
the influence
of introducing
TA–COOH
on the
curing
extensively
reported [29,30].
the Here
influence
of introducing
TA–COOH
on the curing
behavior
behavior
of
epoxy/anhydride
system
was
studied.
Figure
3
compares
the
DSC
and
conversion
curves
of epoxy/anhydride system was studied. Figure 3 compares the DSC and conversion curves of the
of
theepoxy
neat epoxy
the samples
with different
TA–COOH
loadings.
As is mentioned
neat
and theand
samples
with different
TA–COOH
loadings.
As is mentioned
above, above,
the usethe
of
use of TA–COOH
produces
an increase
in the viscosity
of the mixture.
Thus, itdecrease
would decrease
the
TA–COOH
produces
an increase
in the viscosity
of the mixture.
Thus, it would
the mobility
mobility
of the propagating
species
andthe
lower
the reaction
However,
compared
to the
neat
of the propagating
species and
lower
reaction
activity.activity.
However,
compared
to the neat
epoxy
epoxy
the TA–COOH
modified
epoxysystems
resin systems
showcuring
lower curing
temperature
the
system,system,
the TA–COOH
modified
epoxy resin
show lower
temperature
and theand
curing
curing
temperature
withthe
increase
loading of This
TA–COOH.
Thistheimplied
that the
temperature
decreasesdecreases
with increase
loadingthe
of TA–COOH.
implied that
incorporation
of
incorporation
of
TA–COOH
could
accelerate
the
epoxy/anhydride
curing
reaction.
TA–COOH could accelerate the epoxy/anhydride curing reaction.
The
byby
thethe
non-isothermal
integral
isoconversional
procedure.
The
The curing
curingkinetics
kineticswere
werestudied
studied
non-isothermal
integral
isoconversional
procedure.
activation
energy
(E
a
)
and
pre-exponential
factor
(A)
was
calculated
and
the
results
were
shown
The activation energy (Ea ) and pre-exponential factor (A) was calculated and the results were shown in
in
Table 1.
1. As
Aswe
wecan
cansee,
see,the
thepre-exponential
pre-exponential
factor
decrease
slightly,
however,
the activation
energy
Table
factor
decrease
slightly,
however,
the activation
energy
also
also
decreases
to a certain
In consideration
of the compensation
effect between
activation
decreases
to a certain
extent.extent.
In consideration
of the compensation
effect between
activation
energy
energy
and
pre-exponential
factor,
we
consider
the
curing
process
was
accelerated
by
TA–COOH.
As
and pre-exponential factor, we consider the curing process was accelerated by TA–COOH. As we said
we
saidthe
above,
the viscosity
did not increase
a lot,
such an acceleration
affect
be explained
as
above,
viscosity
did not increase
a lot, such
an acceleration
affect could
becould
explained
as follows:
follows:
the modification
of MeHHPA,
TA with MeHHPA,
TA–COOH
containscarboxyl
terminalgroups
carboxyl
groups
After
theAfter
modification
of TA with
TA–COOH
contains terminal
and
some
and
some unreacted
hydroxyl
groups.
Thegroups
carboxyl
can
serveresin
as epoxy
resinIn
hardener.
In the
unreacted
hydroxyl groups.
The
carboxyl
cangroups
serve as
epoxy
hardener.
the meantime,
Table 1. DMA and DSC results of the neat epoxy and TA–COOH modified epoxy resins.
Sample
NEAT
TA–COOH0.5
Polymers 2016, 8, 314
TA–COOH1.0
TA–COOH2.0
Ea a (kJ/mol)
68.3
63.5
-
ln A (min−1)
9.15
7.78
-
Tg (°C)
136.8
142.6
146.7
144.6
Eg b (MPa)
2,515
2,412
2,421
2,500
Er c (MPa)
13.9
11.7
13.2
12.7
ρ (10−3 mol/cm3)
1.27
1.05
1.18
1.14
6 of 12
the carboxyl
groups
and residual
groups
can initiate
the
mechanism
of
a Apparent
b Storage
activation
energy athydroxyl
50% conversion;
modulus
at polycondensation
50 °C; c Storage modulus
at Tg
reaction+ between
epoxide
and
anhydride,
accelerating
the
curing
[31,32].
30 °C.
Figure
3. (a)
DSC
thermograms
of neat
epoxy
the TA–COOH
modified
epoxy
resins
anddegree
(b)
Figure
3. (a)
DSC
thermograms
of neat
epoxy
andand
the TA–COOH
modified
epoxy
resins
and (b)
degree of conversion against temperature of the curing of neat epoxy and the TA–COOH modified
of conversion against temperature of the curing of neat epoxy and the TA–COOH modified epoxy resins.
epoxy resins.
TableMechanical
1. DMA and
DSC results of the neat epoxy and TA–COOH modified epoxy resins.
3.4. Dynamic
Properties
The dynamica mechanical behaviors−1of neat epoxy
andb the TA–COOH
modified epoxy
resins
Sample
Ea (kJ/mol)
T g (◦ C)
Er c (MPa)
ln A (min )
ρ (10−3 mol/cm3 )
Eg (MPa)
were measured. The storage moduli (E’) and loss tangent (tan δ) as a function of temperature for the
NEAT
68.3
2,515
13.9 are shown1.27
9.15
cured
neat epoxy and
the thermosets
containing136.8
0.5, 1.0 and 2.0
wt % TA–COOH
in Figure
TA–COOH0.5
142.6
2,412
11.7
1.05
4, and the data are summarized in Table 1. Compared with neat epoxy, the TA–COOH modified
TA–COOH1.0
63.5
146.7
2,421
13.2
1.18
7.78
epoxy shows a slightly
decrease in -the rubbery
plateau modulus
(Er). Following
classical
TA–COOH2.0
144.6
2,500
12.7
1.14rubber
elasticity,
Er isactivation
proportional
to
the
average
crosslinking
density.
The
crosslinking
density
(ρ) of a
a Apparent
b
◦
c
energy at 50% conversion; Storage modulus at 50 C; Storage modulus at Tg + 30 ◦ C.
cured epoxy network can be calculated using the equation:
3.4. Dynamic Mechanical Properties
(1)
ρ
3
The dynamic mechanical behaviors of neat epoxy and the TA–COOH modified epoxy resins were
Where ρ represents the crosslinking density per unit volume (mol·cm−3), Er is rubbery modulus
measured. The storage moduli (E’) and loss tangent (tan δ) as a function of temperature for the cured
(MPa), R is the gas constant, and T is the absolute temperature. Theoretically, adding of TA–COOH
neat epoxy and the thermosets containing 0.5, 1.0 and 2.0 wt % TA–COOH are shown in Figure 4,
may enhance the crosslinking density due to its higher functionality. However, the thermosets with
and the data are summarized in Table 1. Compared with neat epoxy, the TA–COOH modified epoxy
different TA–COOH loadings show lower crosslinking density than neat epoxy. This presumably can
shows a slightly decrease in the rubbery plateau modulus (Er ). Following classical rubber elasticity,
be explained as following. Although the carboxylic content of TA–COOH is relative low, which has
Er been
is proportional
crosslinking
density.phenolic
The crosslinking
(ρ) of
a cured
epoxy
13C average
confirmed to
by the
NMR (Figure
1), the external
hydroxyl density
groups can
also
react with
network
using
equation:
epoxy can
resinbeincalculated
some ways.
This the
makes
the TA–COOH an extra hardener. As the ratio of epoxy resin
and hardener kept the same in all the formulas, the adding of TA–COOH would lead to an
Er
ρ =the crosslinking density. On the other hand, the(1)
unbalanced formulation, which may decrease
3RT
incorporation of bulk TA–COOH unit into the cured network enlarged the average molecular weight
−3 ), E is rubbery modulus
where
ρ represents
crosslinking
density
unit volume
(mol
·cmsuch
r
between
crosslinks,the
thus
lead to a decrease
in per
crosslinking
density.
With
a complex
mechanism,
(MPa),
R
is
the
gas
constant,
and
T
is
the
absolute
temperature.
Theoretically,
adding
of TA–COOH
the thermoset with 1.0 wt % TA–COOH loading shows the maximum of crosslinking
density.
Similar
may
enhance
thewt
crosslinking
to its
higher
functionality.
However,
the expansion
thermosets(see
with
maxima
at 1.0
% loadingsdensity
are alsodue
found
in T
g and linear
coefficients
of thermal
different
TA–COOH loadings show lower crosslinking density than neat epoxy. This presumably can
later sections).
be explained as following. Although the carboxylic content of TA–COOH is relative low, which has
been confirmed by 13 C-NMR (Figure 1), the external phenolic hydroxyl groups can also react with
epoxy resin in some ways. This makes the TA–COOH an extra hardener. As the ratio of epoxy resin
and hardener kept the same in all the formulas, the adding of TA–COOH would lead to an unbalanced
formulation, which may decrease the crosslinking density. On the other hand, the incorporation of bulk
TA–COOH unit into the cured network enlarged the average molecular weight between crosslinks,
thus lead to a decrease in crosslinking density. With such a complex mechanism, the thermoset with
1.0 wt % TA–COOH loading shows the maximum of crosslinking density. Similar maxima at 1.0 wt %
loadings are also found in Tg and linear coefficients of thermal expansion (see later sections).
Polymers 2016, 8, 314
7 of 12
Polymers 2016, 8, 314
7 of 12
Figure 4. Storage modulus (a) and tan delta (b) versus temperature for the neat epoxy and TA–COOH
Figure 4. Storage modulus (a) and tan delta (b) versus temperature for the neat epoxy and TA–COOH
modified epoxy resins.
modified epoxy resins.
Tan δ is defined as the ratio of the loss modulus to the storage modulus, and the peak of the tan
Tan
δ is defined
as thecurve
ratioisof
the as
loss
to thermosets,
the storagethe
modulus,
and
the peak
the tan δ
δ versus
temperature
taken
Tg. modulus
For all of the
curves are
unimodal
andof
only
one
T
g
is
observed.
That
is,
in
all
cases
the
network
structures
are
homogeneous
and
no
obvious
phase
versus temperature curve is taken as Tg . For all of the thermosets, the curves are unimodal and only
occurs.
loading
wt %, Tgstructures
increases with
loading
one Tg separation
is observed.
ThatBelow
is, in aall
cases of
the1.0network
are TA–COOH
homogeneous
andand
no reaches
obviousa phase
highest value of 146.7 °C, which is 10 °C higher than that of neat epoxy. For the thermoset with 2.0
separation occurs. Below a loading of 1.0 wt %, Tg increases with TA–COOH loading and reaches
wt % TA–COOH loading,
Tg slightly decreases. Tg was also measured by DSC and the results were
a highest
value of 146.7 ◦ C, which is 10 ◦ C higher than that of neat epoxy. For the thermoset with
consistent with the results of DMA. Generally, decreasing the crosslink density alone would decrease
2.0 wt %
loading,
Tg slightly
Tg was
measured
and thedensity
results were
theTA–COOH
material rigidity,
thereby
reducingdecreases.
Tg. However,
Tg is also
affected
by both by
theDSC
crosslinking
consistent
theflexibility.
results ofAs
DMA.
Generally,
decreasingdensity
the crosslink
density
aloneTA–COOH.
would decrease
and with
the chain
is mentioned,
the crosslinking
decreases
after adding
Therefore,
the increase
of Treducing
g is probably
to the high
of aromatic
structure
in TA–COOH,
the material
rigidity,
thereby
Tg .due
However,
Tgcontent
is affected
by both
the crosslinking
density
which
partly
enhances
the
chain
rigidity.
and the chain flexibility. As is mentioned, the crosslinking density decreases after adding TA–COOH.
Therefore, the increase of Tg is probably due to the high content of aromatic structure in TA–COOH,
3.5. Thermal Expansion
which partly enhances the chain rigidity.
The values of the coefficient of thermal expansion (CTE) measured in the glassy region (αg) and
3.5. Thermal
Expansion
the rubber
region (αr) as well as their difference (Δα = αr − αg) are listed in Table 2. Compared with
neat
epoxy,
TA–COOH
epoxy resins(CTE)
show measured
lower CTE in
values
in the region
glassy region.
The
values
ofall
thethe
coefficient
of modified
thermal expansion
the glassy
(αg ) and the
Such a small αg in cured hybrids could lower the internal stress when processing and is very
rubber region (αr ) as well as their difference (∆α = αr − αg ) are listed in Table 2. Compared with neat
beneficial for the composite matrix, because it helps to maintain better interfacial strength during
epoxy, temperature
all the TA–COOH
modified
epoxy
resins
show
lowerinCTE
the
glassy region.
cycles and
shocks. On
the other
hand,
as shown
Tablevalues
2, wheninthe
TA–COOH
loadingSuch a
small αincreases,
hybrids
could
lower
the
internal
stress
when
processing
and
is
very
beneficial
g in cured
αg decreases, and αr increases. Thus, the Δα increases. Based on the free volume theory [33],for the
composite
matrix,
it helps
to maintain
interfacial
during temperature cycles and
the fractionalbecause
free volume
at temperature
T (fbetter
T) can be
expressedstrength
as:
shocks. On the other hand, as shown in Table 2, when
Δ the TA–COOH loading increases, αg decreases,
(3)
and αr increases. Thus, the ∆α increases. Based on the free volume theory [33], the fractional free
Where Δα = αr − αg is the difference between the CTE in rubbery and glassy states, and fg is the
volume at temperature T (f T ) can be expressed as:
fractional free volume at Tg. Therefore, the increased Δα for the TA–COOH modified epoxy resins
is incorporated into the epoxy
indicates the existence of more free volume when TA–COOH
f T = f g + ∆α T − Tg
(3)
networks. Such a result is also observed
in other reports that introducing a hyperbranched polymer
increase the
free volume.
It should
be noted
thatglassy
these states,
increaseand
in free
where into
∆α =epoxy
αr −networks
αg is thecan
difference
between
the CTE
in rubbery
and
f g is the
volume could significantly improve the toughness in cured thermosets [22–24,34–36].
fractional free volume at Tg . Therefore, the increased ∆α for the TA–COOH modified epoxy resins
indicates the existence
more
free volume
when
TA–COOH
is incorporated
into the epoxy networks.
Table 2. of
Linear
coefficients
of thermal
expansion
determined
from TMA measurements.
Such a result is also observed in other reports that introducing a hyperbranched polymer into epoxy
αr
Δα = αr − αg
αg
Sample
networks can increase the free
volume. It should
be noted that these increase in free volume could
（× 10−6 K−1） （× 10−6 K−1） （× 10−6 K−1）
significantly improve the toughness
in cured
thermosets180.1
[22–24,34–36].
NEAT
81.3
98.8
TA–COOH0.5
76.4
190.6
114.2
TA–COOH2.0
75.5
200.1
124.6
Table 2. Linear coefficients
of thermal
expansion determined
from
TMA measurements.
TA–COOH1.0
67.9
198.7
130.9
Sample
αg
(× 10−6 K−1 )
αr
(× 10−6 K−1 )
∆α = αr − αg
(× 10−6 K−1 )
NEAT
TA–COOH0.5
TA–COOH1.0
TA–COOH2.0
81.3
76.4
67.9
75.5
180.1
190.6
198.7
200.1
98.8
114.2
130.9
124.6
Polymers 2016, 8, 314
Polymers 2016, 8, 314
8 of 12
8 of 12
3.6.
3.6.Mechanical
MechanicalProperties
Properties
To
To investigate
investigate the
the reinforcing
reinforcing and
and toughening
toughening effects
effects of
of the
the incorporated
incorporated TA–COOH,
TA–COOH, the
the
thermosets
with
different
TA–COOH
loadings
were
prepared
and
their
tensile
and
impact
properties
thermosets with different TA–COOH loadings were prepared and their tensile and impact properties
were
weretested.
tested.From
FromFigure
Figure5a,
5a,ititcan
canbe
befound
foundthat
thatthe
theimpact
impactstrength
strengthwas
wasdistinctly
distinctlyimproved
improvedwith
with
2
the
theincorporation
incorporationof
ofTA–COOH.
TA–COOH.ItItreaches
reachesthe
themaximum
maximumofof33.9
33.9kJ/m
kJ/m2 at
at 1.0
1.0 wt
wt %
% loading,
loading,which
whichisis
159%
159%higher
higherthan
thanthat
thatof
ofneat
neatepoxy.
epoxy.The
Thefurther
furtherincrease
increaseof
ofTA–COOH
TA–COOHloading
loadingover
over1.0
1.0wt
wt%
%leads
leads
to
to aa slightly
slightly decrease
decrease in
in the
the impact
impact strength,
strength, which
which can
can be
be ascribed
ascribed to
to the
the “crosslinking
“crosslinking density
density
reduction”
reduction”effect.
effect.
The
Thetensile
tensiletest
testresults
resultsare
areshown
shownin
inFigure
Figure5b.
5b.The
Theelongation
elongationat
atbreak
breakincreases
increasescontinuously
continuously
from
fromabout
about3.0%
3.0%to
to5.9%
5.9%with
withincreasing
increasingTA–COOH
TA–COOHloading
loadingfrom
from00toto2.0
2.0wt
wt%.
%.Such
Suchan
anincrease
increase
should
be
related
to
the
good
interface
interaction
via
a
chemical
reaction
between
the
terminal
should be related to the good interface interaction via a chemical reaction between the terminal
carboxyl
carboxylgroup
groupof
ofTA–COOH
TA–COOHand
andepoxy
epoxymatrix.
matrix.For
Fortypical
typicalepoxy
epoxymaterial,
material,enhancing
enhancingthe
theelongation
elongation
at
atbreak
breakusually
usuallyaccompanies
accompanieswith
withdecreasing
decreasingtensile
tensilestrength.
strength. However,
However, aacontinuous
continuous increase
increase in
in
tensile
strength
was
also
observed
as
the
loading
of
TA–COOH
increases.
The
thermoset
with
2.0
wt
%
tensile strength was also observed as the loading of TA–COOH increases. The thermoset with 2.0 wt
TA–COOH
loading
has has
a maximum
tensile
strength
of 67ofMPa,
which
is about
42.5%42.5%
higher
than that
% TA–COOH
loading
a maximum
tensile
strength
67 MPa,
which
is about
higher
than
of
neat
The increase
in tensile
strengthstrength
may be may
due to
aromatic
structure structure
and goodand
interface
that
ofepoxy.
neat epoxy.
The increase
in tensile
bethe
due
to the aromatic
good
interaction
between TA–COOH
and epoxyand
matrix.
interface interaction
between TA–COOH
epoxy matrix.
Figure5.5.Impact
Impact(a)
(a) and
and tensile
tensile (b)
(b) properties
properties of
of the
the neat
neat epoxy
epoxy and
and TA–COOH
TA–COOHmodified
modifiedepoxy
epoxyresins
resins
Figure
withdifferent
differentTA–COOH
TA–COOHloadings.
loadings.
with
3.7. Morphology of Fractured Surfaces
3.7. Morphology of Fractured Surfaces
The morphology of the fracture surfaces of neat epoxy and the TA–COOH modified epoxy resins
The morphology of the fracture surfaces of neat epoxy and the TA–COOH modified epoxy resins
were investigated by SEM. Figure 6 presents the SEM micrographs of impact fractured surfaces of
were investigated by SEM. Figure 6 presents the SEM micrographs of impact fractured surfaces of
the thermosets prepared. It can be observed that the fracture surface of neat epoxy (Figure 6a) is very
the thermosets prepared. It can be observed that the fracture surface of neat epoxy (Figure 6a) is
smooth except for some river-like lines, indicating a brittle failure mode without any ductility. In
very smooth except for some river-like lines, indicating a brittle failure mode without any ductility.
contrast, the fracture surfaces of the thermosets with TA–COOH are much rougher than that of the
In contrast, the fracture surfaces of the thermosets with TA–COOH are much rougher than that of
neat epoxy and without traces of phase separation. In addition, a lot of oriented “protonema” or
the neat epoxy and without traces of phase separation. In addition, a lot of oriented “protonema” or
“fibrils” are clearly observed, which indicates that the thermosets undergo more plastic deformation.
“fibrils” are clearly observed, which indicates that the thermosets undergo more plastic deformation.
As we already know, large plastic deformation and crazing processes could significantly absorb the
As we already know, large plastic deformation and crazing processes could significantly absorb the
energy, thus result in an increase in the amount of energy needed for crack propagation and for the
energy, thus result in an increase in the amount of energy needed for crack propagation and for the
formation of new surfaces. Such a result was also observed by other researchers and this is in good
formation of new surfaces. Such a result was also observed by other researchers and this is in good
agreement with in situ toughening mechanism [36–39].
agreement with in situ toughening mechanism [36–39].
Polymers 2016, 8, 314
9 of 12
Polymers 2016, 8, 314
Polymers 2016, 8, 314
9 of 12
9 of 12
Figure 6. SEM images of the impact fracture surfaces of (a) neat epoxy; (b) TA–COOH0.5; (c) TA–
Figure
6. SEM images of the impact fracture surfaces of (a) neat epoxy; (b) TA–COOH0.5;
Figure
6. SEM
of the impact fracture surfaces of (a) neat epoxy; (b) TA–COOH0.5; (c) TA–
COOH1.5;
andimages
(d) TA–COOH2.0.
(c) TA–COOH1.5; and (d) TA–COOH2.0.
COOH1.5; and (d) TA–COOH2.0.
3.8. Thermogravimetric Analysis
3.8.3.8.
Thermogravimetric
ThermogravimetricAnalysis
Analysis
TGA curves of cured neat epoxy and thermosets with different TA–COOH loadings are shown
curves
neat
epoxy
thermosets
withdifferent
different
TA–COOH
loadings
shown
TGA
curves
ofcured
cured
neat
epoxy
and
thermosets
with
TA–COOH
loadings
areare
shown
inTGA
Figure
7. Asof
we
can see,
both
neatand
epoxy
and TA–COOH
modified
epoxy show
only one
similar
in in
Figure
7. 7.AsAsstage,
we
see,
neat
epoxy
TA–COOH
modified
epoxy
show
only
one
similar
degradation
indicating
that
a homogenous
structure ofmodified
the matrix
is formed
and
theone
breakage
Figure
wecan
can
see,both
both
neat
epoxy and
and TA–COOH
epoxy
show
only
similar
degradation
stage,
indicating
that
structure
ofthe
thematrix
matrix
formed
and
breakage
of bonds in
the
network
structure
simultaneously.
After
careful
observation,
itand
can
bethe
seen
that
degradation
stage,
indicating
thataoccurs
ahomogenous
homogenous
structure
of
isisformed
the
breakage
the
initial
degradation
temperatures
(T5%
)simultaneously.
of the thermosets
with
different
TA–COOH
are
of of
bonds
ininthe
structureoccurs
occurs
After
careful
observation,
it
seen
bonds
thenetwork
network structure
simultaneously.
After
careful
observation,
it canloadings
becan
seenbe
that
slightly
lower
than
that
of
neat
epoxy
(as
shown
in
the
inset
of
Figure
7).
In
addition,
it
is
worth
thethe
initial
degradation
temperatures
(T5%(T
) of
with with
different
TA–COOH
loadings
are
that
initial
degradation
temperatures
) ofthermosets
the thermosets
different
TA–COOH
loadings
5%the
mentioning
that
thethat
Tthat
5% of
increases
with
an
increase
inin
thethe
amount
of
TA–COOH.
The
combination
of
than
neat
epoxy
(as(as
shown
in
inset
of of
Figure
7).7).
In In
addition,
it is
areslightly
slightlylower
lower
than
of
neat
epoxy
shown
inset
Figure
addition,
it worth
is worth
some effects
could
this experimental
behavior.
On
the
oneof
hand,
the thermal
of TA–
mentioning
that
theTexplain
T5%
5% increases
with
inin
the
amount
TA–COOH.
Thestability
combination
of
mentioning
that
the
increases
withan
anincrease
increase
the
amount
of TA–COOH.
The
combination
COOH
is
relatively
low,
its
T
5%experimental
was
about
200
°C.
On
the
other
hand,
the
terminal
carboxyl
groups
of
some
effects
could
explain
this
behavior.
On
the
one
hand,
the
thermal
stability
of
TA–
of some effects could explain this experimental behavior. On the one hand, the thermal stability of
the TA–COOH
provide
miscibility
during200
curing
and allow
reaction
with
other network
functional
COOH
is is
relatively
low,
its
T5%Twaswas
about
°C. ◦On
hand,
the
terminal
carboxyl
groups
of
TA–COOH
relatively
low,
its
about 200
C. the
On other
the other
hand,
the terminal
carboxyl
groups
5%
groups,
which
in
turn
produces
greater
adhesion
to
the
matrix
and
may
prevent
the
elimination
of
with
other
network
functional
of the
the TA–COOH
TA–COOH provide
providemiscibility
miscibilityduring
duringcuring
curingand
andallow
allowreaction
reaction
with
other
network
functional
volatilewhich
fragments.
groups,
in turn produces greater adhesion to the matrix and may prevent the elimination of
groups, which in turn produces greater adhesion to the matrix and may prevent the elimination of
volatile fragments.
volatile fragments.
Figure 7. TGA curves of the neat epoxy and TA–COOH modified epoxy resins.
Figure
epoxy and
andTA–COOH
TA–COOHmodified
modifiedepoxy
epoxyresins.
resins.
Figure7.7.TGA
TGAcurves
curves of
of the
the neat epoxy
Polymers 2016, 8, 314
10 of 12
4. Conclusions
A bio-based carboxyl-terminated tannic acid (TA–COOH) had been synthesized through a simple
esterification between TA and MeHHPA. Then TA–COOH was used as a modifier for anhydride cured
epoxy system. Owing to the chemical modification, TA–COOH could easily disperse in epoxy resin
and showed a good interface interaction between TA–COOH and epoxy matrix. When TA–COOH
was used as modifier, it can simultaneously improve toughness, elongation at break, Tg , and strength.
Especially for the thermoset with 1.0 wt % TA–COOH loading, it showed the impact strength and
tensile strength of 33.9 kJ/m2 and 62.0 MPa, respectively, which are 159% and 32% higher than that of
neat epoxy, respectively. In the meantime, the Tg increased to 146.7 ◦ C. According to the results of DMA,
TMA and SEM, no phase separation occurred. The simultaneous enhancements in Tg , tensile strength,
and impact strength are mainly because of the synergistic effect of aromatic structure, decreasing of
cross linking density, increasing of free volume and good interface interaction.
Acknowledgments: We are grateful for the financial support from the Enterprise-University-Research Prospective
Program, Jiangsu Province (BY2013015-08 and BY2015019-08) and MOE & SAFEA for the 111 Project (B13025).
Author Contributions: All authors contributed to the technical review of the manuscript. Xiaoma Fei and
Mingqing Chen conceived and designed the experiments. Xiaoma Fei, Fangqiao Zhao and Wei Wei performed
the experiments. Xiaoma Fei and Wei Wei analyzed data. Xiaoma Fei, Jing Luo and Xiaoya Liu contributed
reagents/materials/analysis tools. Xiaoma Fei wrote the manuscript.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Comstock, M.J. Epoxy Resin Chemistry; Bouer, R.S., Ed.; ACS publication: Washington, DC, USA, 1983.
Auvergne, R.; Caillol, S.; David, G.; Boutevin, B.; Pascault, J.P. Biobased thermosetting epoxy: Present and
future. Chem. Rev. 2014, 114, 1082–1115. [CrossRef] [PubMed]
Jin, F.; Li, X.; Park, S. Synthesis and application of epoxy resins: A review. J. Ind. Eng. Chem. 2015, 29, 1–11.
[CrossRef]
Chonkaew, W.; Sombatsompop, N.; Brostow, W. High impact strength and low wear of epoxy modified
by a combination of liquid carboxyl terminated poly(butadiene-co-acrylonitrile) rubber and organoclay.
Eur. Polym. J. 2013, 49, 1461–1470. [CrossRef]
Gan, W.J.; Zhan, G.Z.; Wang, M.H.; Yu, Y.F.; Xu, Y.Z.; Li, S.J. Rheological behaviors and structural transitions
in a polyethersulfone-modified epoxy system during phase separation. Colloid Polym. Sci. 2007, 285,
1727–1731. [CrossRef]
Wang, K.; Chen, L.; Wu, J.S.; Toh, M.L.; He, C.B.; Yee, A.F. Epoxy nanocomposites with highly exfoliated clay:
Mechanical properties and fracture mechanisms. Macromolecules 2005, 38, 788–800. [CrossRef]
Bibekananda, D.; Karak, N. Ultralow dielectric, high performing hyperbranched epoxy thermosets:
Synthesis, characterization and property evaluation. RSC Adv. 2015, 5, 35080–35088.
Miao, X.P.; Meng, Y.; Lia, X.Y. A novel all-purpose epoxy-terminated hyperbranched polyether sulphone
toughener for an epoxy/amine system. Polymer 2015, 60, 88–95. [CrossRef]
Wang, R.; Zhuo, D.X.; Weng, Z.X.; Wu, L.X.; Cheng, X.Y.; Zhou, Y.; Wang, J.L.; Xuan, B.W. A novel
nanosilica/graphene oxide hybrid and its flame retarding epoxy resin with simultaneously improved
mechanical, thermal conductivity, and dielectric properties. J. Mater. Chem. A 2015, 3, 9826–9836. [CrossRef]
Ma, Q.; Luo, J.; Chen, Y.X.; Wei, W.; Liu, R.; Liu, X.Y. Reactive copolymer functionalized graphene sheet for
enhanced mechanical and thermal properties of epoxy composites. J. Polym. Sci. Part A 2015, 53, 2776–2785.
[CrossRef]
Tseng, C.H.; Hsueh, H.B.; Chen, C.Y. Effect of reactive layered double hydroxides on the thermal and
mechanical properties of LDHs/epoxy nanocomposites. Compos. Sci. Technol. 2007, 67, 2350–2362. [CrossRef]
Alamri, H.; Low, I.M. Mechanical properties and water absorption behaviour of recycled cellulose fibre
reinforced epoxy composites. Polym. Test. 2012, 31, 620–628. [CrossRef]
Liu, W.S.; Zhou, R.; Goh, H.L.S.; Huang, S.; Lu, X.H. From waste to functional additive: Toughening epoxy
resin with lignin. ACS Appl. Mater. Interfaces 2014, 6, 5810–5817. [CrossRef] [PubMed]
Polymers 2016, 8, 314
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
11 of 12
Satheesh, B.; Tshai, K.Y.; Warrior, N.A. Effect of chitosan loading on the morphological, thermal,
and mechanical properties of diglycidyl ether of bisphenol A/hexamethylenediamine epoxy system.
J. Compos. 2014. [CrossRef]
Enjoji, M.; Yamamoto, A.; Shibata, M. Wood-derived phenol novolaks and their wood/epoxy biocomposites.
J. Appl. Polym. Sci. 2015. [CrossRef]
Masoodi, R.; El-Hajjar, R.F.; Pillai, K.M.; Sabo, R. Mechanical characterization of cellulose nanofiber and
bio-based epoxy composite. Mater. Des. 2012, 36, 570–576. [CrossRef]
Liu, R.; Zhu, J.J.; Luo, J.; Liu, X.Y. Synthesis and application of novel UV-curable hyperbranched
methacrylates from renewable natural tannic acid. Prog. Org. Coat. 2014, 77, 30–37. [CrossRef]
Luo, J.; Lai, J.P.; Zhang, N.; Liu, Y.B.; Liu, R.; Liu, X.Y. Tannic acid induced self-assembly of three-dimensional
graphene with good adsorption and antibacterial properties. ACS Sustain. Chem. Eng. 2016, 4, 1404–1413.
[CrossRef]
Shin, M.; Kim, K.; Shim, W.; Yang, J.W.; Lee, H. Tannic acid as a degradable mucoadhesive compound.
ACS Biomater. Sci. Eng. 2016, 2, 687–696. [CrossRef]
Rivero, S.; García, M.A.; Pinotti, A. Crosslinking capacity of tannic acid in plasticized chitosan films.
Carbohydr. Polym. 2010, 82, 270–276. [CrossRef]
Pan, L.; Wang, H.B.; Wu, C.L.M.; Liao, C.B.; Li, L. Tannic-acid-coated polypropylene membrane as a separator
for lithium-ion batteries. ACS Appl. Mater. Interfaces 2015, 7, 16003–16010. [CrossRef] [PubMed]
Liu, D.; Wang, H.M.; Jiang, H.S.; Zhou, D.P. Improving the heat-resistance and toughness performance of
phenolic resins by adding a rigid aromatic hyperbranched polyester. J. Appl. Polym. Sci. 2016. [CrossRef]
Miao, X.P.; Meng, Y.; Li, X.Y. Epoxide-terminated hyperbranched polyether sulphone as triple enhancement
modifier for DGEBA. J. Appl. Polym. Sci. 2015. [CrossRef]
Li, T.; Qin, H.J.; Liu, Y.; Zhong, X.H.; Yu, Y.F.; Serra, A. Hyperbranched polyester as additives in filled and
unfilled epoxy-novolac systems. Polymer 2012, 53, 5864–5872. [CrossRef]
Tian, H.; Wang, Y.; Zhang, L.; Quan, C.; Zhang, X. Improved flexibility and water resistance of soy protein
thermoplastics containing waterborne polyurethane. Ind. Crops Prod. 2010, 32, 13–20. [CrossRef]
Liu, R.; Zheng, J.C.; Guo, R.X.; Luo, J.; Yuan, Y.; Liu, X.Y. Synthesis of new biobased antibacterial
methacrylates derived from tannic acid and their application in UV-cured coatings. Ind. Eng. Chem. Res.
2014, 53, 10835–10840. [CrossRef]
Gogoi, S.; Karak, N. Biobased biodegradable waterborne hyperbranched polyurethane as an ecofriendly
sustainable material. ACS Sustain. Chem. Eng. 2014, 2, 2730–2738. [CrossRef]
Saha, P.; Chowdhury, S.; Manna, S.; Roy, D.; Adhikari, B.; Thomas, S. New biobased surface treatment to
improve strength and durability of bombax ceiba. ACS Sustain. Chem. Eng. 2016, 4, 76–84. [CrossRef]
Belmonte, A.; Däbritz, F.; Ramis, X.; Serra, A.; Voit, B.; Fernández-Francos, X. Cure kinetics modeling
and thermomechanical properties of cycloaliphatic epoxy-anhydride thermosets modified with hyperstar
polymers. J. Polym. Sci. Part B 2014, 52, 1227–1242. [CrossRef]
Foix, D.; Yu, Y.F.; Serra, A.; Ramis, X.; Salla, J.M. Study on the chemical modification of epoxy/anhydride
thermosets using a hydroxyl terminated hyperbranched polymer. Eur. Polym. J. 2009, 45, 1454–1466.
[CrossRef]
Tomuta, A.; Ferrando, F.; Serra, À.; Ramis, X. New aromatic–aliphatic hyperbranched polyesters with vinylic
end groups of different length as modifiers of epoxy/anhydride thermosets. React. Funct. Polym. 2012, 72,
556–563. [CrossRef]
Fernández-Francos, X.; Rybak, A.; Sekula, R.; Ramis, X.; Serra, A. Modification of epoxy-anhydride
thermosets using a hyperbranched poly(ester-amide): I. Kinetic study. Polym. Int. 2012, 61, 1710–1725.
[CrossRef]
Kaji, M.; Nakahara, K.; Endo, T. Synthesis of a bifunctional epoxy monomer containing biphenyl moiety and
properties of its cured polymer with phenol novolac. J. Appl. Polym. Sci. 1999, 74, 690–698. [CrossRef]
Ratna, D.; Simon, G.P. Epoxy and hyperbranched polymer blends: Morphology and free volume. J. Appl.
Polym. Sci. 2010, 117, 557–564. [CrossRef]
Luo, L.J.; Qiu, T.; Meng, Y.; Guo, L.H.; Yang, J.; Li, Z.X.; Cao, X.Z.; Li, X.Y. A novel fluoro-terminated
hyperbranched poly(phenylene oxide) (FHPPO): Synthesis, characterization, and application in low-k epoxy
materials. RSC Adv. 2013, 3, 14509–14520. [CrossRef]
Polymers 2016, 8, 314
36.
37.
38.
39.
12 of 12
Liu, T.; Nie, Y.X.; Chen, R.S.; Zhang, L.D.; Meng, Y.; Li, X.Y. Hyperbranched polyether as an all-purpose
epoxy modifier: Controlled synthesis and toughening mechanisms. J. Mater. Chem. A 2015, 3, 1188–1198.
[CrossRef]
Mezzenga, R.; Månson, J.A.E. Thermo-mechanical properties of hyperbranched polymer modified epoxies.
J. Mater. Sci. 2001, 36, 4883–4891. [CrossRef]
Ratna, D.; Varley, R.; Raman, R.K.S.; Simon, G.P. Studies on blends of epoxy-functionalized hyperbranched
polymer and epoxy resin. J. Mater. Sci. 2003, 38, 147–154. [CrossRef]
Tomuta, A.M.; Fernández-Francos, X.; Ferrando, F.; Ramis, X.; Serra, À. Enhanced chemical reworkability
of DGEBA thermosets cured with rare earth triflates using aromatic hyperbranched polyesters (HBP) and
multiarm star HBP-b-poly(ε-caprolactone) as modifiers. Polym. Adv. Technol. 2013, 24, 962–970. [CrossRef]
© 2016 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/).