Embed Size (px)
Citation preview
ORI GIN AL PA PER
Influence of rubber on the curing kinetics of DGEBAepoxy and the effect on the morphology and hardnessof the composites
Angel Romo-Uribe • Jose Antonio Arcos-Casarrubias •
Araceli Flores • Cintya Valerio-Cardenas •
Agustin E. Gonzalez
Received: 2 September 2013 / Revised: 31 December 2013 / Accepted: 2 March 2014 /
Published online: 13 March 2014
� Springer-Verlag Berlin Heidelberg 2014
Abstract The influence of the end groups of two liquid rubbers on curing kinetics,
morphology, and hardness behavior of diglycidyl ether of bisphenol-A based epoxy
resin (DGEBA) has been studied. The rubbers are silyl-dihydroxy terminated
(PDMS-co-DPS-OH) and silyl-diglycidyl ether terminated (PDMS-DGE). Cross-
linking reactions, investigated by shear rheometry, ranged 90–110 �C, using a
constant concentration (5 phr) of liquid rubbers and 1,2-Diamino cyclohexane (1,2-
DCH) as hardener agent. The gel time, tgel, of the neat epoxy significantly decreased
when adding the elastomers, more so for the silyl-dihydroxy terminated elastomer;
at 110 �C the reaction was nearly complete before rheological test started. The
results suggest that the elastomers induced a catalytic effect on the curing reaction.
Scanning electron microscopy revealed phase separation of the elastomer during the
curing reaction with rubber domains about 5 lm size. However, the DGEBA/
dihydroxy terminated elastomer composite cured at 110 �C exhibited a homogenous
morphology, that is, the rapid reaction time would not allow for phase separation.
Water contact angle tests evidenced either more hydrophilic (silyl-diglycidyl ether
terminated rubber) or more hydrophobic (silyl-dihydroxy terminated rubber)
behavior than the neat epoxy. The latter effect is attributed to the presence of
aromatic rings in the backbone structure of PDMS-co-DPS-OH. Microindentation
measurements show that the elastomers significantly reduced the hardness of the
A. Romo-Uribe (&) � C. Valerio-Cardenas � A. E. Gonzalez
Laboratorio de Nanopolimeros y Coloides, Instituto de Ciencias Fisicas, Universidad Nacional
Autonoma de Mexico, 62210 Cuernavaca, Morelos, Mexico
e-mail: [email protected]
J. A. Arcos-Casarrubias
Division de Ingenieria Quimica y Bioquimica, Tecnologico de Estudios Superiores de Ecatepec,
Av. Tecnologico s/n esq. Av. Hank Gonzalez, 55210 Ecatepec, Edo. de Mex., Mexico
A. Flores
Instituto de Estructura de la Materia, IEM-CSIC, Serrano 119, Madrid, Spain
123
Polym. Bull. (2014) 71:1241–1262
DOI 10.1007/s00289-014-1121-6
epoxy resin, the DGEBA/ether terminated composite exhibiting the lowest hardness
values. Moreover, hardness increased as reaction temperature did, correlating with a
reduction of microdomains size thus enabling the tuning of mechanical properties
with reaction temperature.
Keywords Epoxy � DGEBA � Rubber � Shear rheometry � Curing
reaction
Introduction
Cross-linking reactions are able to connect macromolecules into a tridimensional
polymeric network. During the initial stages of the cross-linking process, branched
molecules of broadly distributed sizes and of various architectures are formed. Their
average molecular weight increases as the cross-linking reaction progresses. The gel
point determines the transition from liquid to solid and is defined by the instant at
which the system reaches a critical extent of reaction for which either the weight-
average molecular weight diverges to infinity (infinite sample size) [1, 2] or a first
macromolecular cluster extends across the entire sample (finite sample size).
Consequently, the system loses its solubility, the steady shear viscosity g diverges to
infinity, and the equilibrium modulus starts to rise to a finite value. The newly
formed macroscopic network structure coexists with the remaining branched
molecules which are not yet attached. Beyond the gel point, the network stiffness
continues to increase steadily with increasing cross-linking density until the system
reaches completion of the chemical reaction.
There are two main methods for the rheological study of cross-linking processes
in polymers [3]. In one approach, the polymer in its liquid state is subjected to shear
flow. The measured viscosity increases with increasing extent of reaction until the
stress reaches the limit of the instrument or until the material breaks. For
characterization beyond the gel point the material is subjected to strain and the
steady-state modulus is measured during its growth with increasing extent of
reaction. Measurements in either the liquid state or the solid state give reliable data
away from the gel point. However, the transition itself is only accessible by
extrapolation and this introduces some uncertainty in the determination of the gel
point time.
The second method uses small-amplitude oscillatory experiments to obtain the
components of the complex modulus (G0, G00) during the cross-linking process. The
viscous behavior of the oligomeric material dominates the initial part of the
experiment, i.e., the viscoelastic behavior of the material is dominated by the loss
modulus (G00[ G0). As the cross-linking reaction proceeds, the molecular weight
increases, the loss modulus increases and, as more energy is stored than dissipated,
the storage modulus rises sharply until it intersects and then exceeds the loss
modulus. The point of intersection occurs at the time tgel, at which G0 = G00, which
is in the vicinity of the gel point of the reaction. Afterwards, the storage modulus
keeps increasing with increasing crosslink density while the loss modulus changes
1242 Polym. Bull. (2014) 71:1241–1262
123
very little during the reaction. Both moduli level off as the reaction comes to
completion [1, 4–6].
Epoxy polymers are mainly used to form rigid structural materials. The epoxy
prepolymer reacts further with a wide variety of primary or secondary amines to
form a rigid polymer network [7, 8]. The conversion of an epoxy polymer to an
interconnected network structure is formally similar to the vulcanization of rubber,
but the process is termed curing in the epoxy system [9–11].
The addition of rubbers to epoxy resins has been shown to not only alter the final
morphology and properties but also the curing kinetics [12–24]. Poly(dimethylsi-
loxane) (PDMS) has been effectively used to improve the thermomechanical
properties and toughness of epoxy resins [12–18]. Because phase separation of the
elastomer currently takes place during curing [19–23], the resultant heterogeneous
morphology obtained after the reaction controls to an important extent the properties
of the final product. It has been shown that the rubber particle size in rubber-
modified epoxy resins using a rubber butadiene acrylonitrile copolymer with
carboxyl or hydroxyl end-groups influences the toughness of the composite [20].
Significantly higher impact strength was found in the composite using the hydroxyl-
terminated rubber with 10–50 lm particle size with respect to that including the
carboxyl-terminated rubber displaying a particle size of 2–4 lm [20]. The latter
material has been found to exhibit an average particle size ranging from 5 to 10 lm
for rubber concentrations below 20 phr [21]. It has also been shown that,
irrespective of the rubber end-group, the curing kinetics slows down due to dilution
and viscosity effects, as determined from gelation time measurements [22, 23].
However, in blends of diglycidyl ether of bisphenol-A and (DGEBA/DGEBF), the
butadiene-acrylonitrile rubber induced a catalytic effect on the polymerization of
the epoxy resin, thus inducing a higher rate of gel formation and higher cure rate
[24].
This research focused on the curing behavior of an epoxy amine system modified
with two different liquid elastomers through shear rheometry. The elastomers were
chosen based on either chemical or physical compatibility with the epoxy amine
system to hinder phase separation or alternatively reduce the size of the
microdomains. One of the modified epoxy resins includes a PDMS diglycidyl
ether terminated elastomer while the other one incorporates a poly(dimethylsilox-
ane-co-diphenylsiloxane) dihydroxy terminated elastomer. The aim of the work was
to enhance the processing behavior and the mechanical properties of the matrix
through the incorporation of tailored elastomers to obtain tougher composites for
more demanding applications.
Experimental
Materials
The epoxy resin used was diglycidyl ether of bisphenol-A (DGEBA), with
equivalent weight of epoxy group equal to 172–176 Eq. g-1. 1,2-Diaminocyclo-
hexane (1,2-DCH) was used as hardener. The two liquid rubbers employed were
Polym. Bull. (2014) 71:1241–1262 1243
123
(i) poly(dimethylsiloxane) diglycidyl ether terminated (PDMS-DGE), a low-
molecular-mass elastomer (typical Mn = 980 g mol-1) that contains reactive
functional groups towards diamine with equivalent weight 490 Eq. g-1 and (ii)
poly(dimethylsiloxane-co-diphenylsiloxane) dihydroxy terminated (PDMS-co-DPS-
OH) with 95:5 mol ratio of dimethylsiloxane:diphenylsiloxane. All the chemicals
were supplied by Aldrich Chemical Co (St Louis, MO, USA) and were used as
received without purification. Figure 1 shows the structures of the materials used in
the study, and Table 1 summarizes their properties.
Sample preparation
The neat epoxy material was prepared dissolving diglycidyl ether of bisphenol-A
based epoxy resin DGEBA with 1,2-DCH, which was added drop by drop to reach a
stoichiometric ratio of 2:1. The mixture was vigorously agitated using a mechanical
stirrer until obtaining a homogenous mixture. The blends with PDMS-DGE or
PDMS-co-DPS-OH were prepared as follows: DGEBA was heated up to 40 �C and
1,2-DCH was added in a stoichiometric ratio of 2:1. Thereafter, different weighted
amounts of PDMS-DGE or PDMS-co-DPS-OH were added as needed to attain 5 phr
of elastomer in the respective mixture. The mixture was stirred again at 40 �C until
a homogeneous mixture was obtained; immediately, the prepared blends were
quickly introduced in the rheometer.
Shear rheometry
The curing kinetics and the gel point were studied by shear rheometry utilizing a
CVO rheometer (Malvern Ltd), equipped with parallel plates fixtures of 25 mm
diameter. Rheological measurements were carried out at different isothermal
temperatures, ranging from 90 to 120 �C, maintaining the elastomers concentration
at 5 phr. Prior to measurements, the rheometer was pre-heated and the gap was
zeroed.
FT-IR characterization
FTIR spectroscopy was used to characterize the reagents and cured samples.
Infrared spectra were obtained with the FTIR Nicolet iS10 of Thermo Scientific
(USA). For each spectrum, sixteen scans between 4,000 and 400 cm-1 were
averaged with a resolution of 4 cm-1.
Phase-morphology studies
Scanning electron microscopy (SEM) was performed using a CARL ZEISS
DSM940A model SEM with electron voltage of 15 kV. The samples were fractured
under liquid nitrogen and the dispersed elastomer phase was extracted in the soxleth
using refluxed acetone for 10 h at atmospheric pressure. A thin section of the dried
samples were sputter-coated with a thin layer of gold prior to SEM examination.
1244 Polym. Bull. (2014) 71:1241–1262
123
Fig. 1 Chemical structures of a 1,2-diaminocyclohexane (DCH), b diglycidyl ether of bisphenol-Aepoxy resin (DGEBA), c poly(dimethylsiloxane) diglycidyl ether terminated (PDMS-DGE), andd poly(dimethylsiloxane-co-diphenylsiloxane) dihydroxy terminated (PDMS-co-DPS-OH)
Table 1 Physical properties of DGEBA epoxy and siloxane elastomers
Component Mn
(g/mol)
FW Density
(g/mL)
Viscosity
(cSt)
Tm
(�C)
Functionality Equiv
(g-1)
DGEBA 340.4 1.160 – 40 2 172–176
1,2 DCH 114.19 0.931 – – 4 28.5
PDMS-DGE 980 0.99 15 – 2 490
PDMS-co-DPS-OH 1.05 60 – 0 0
Polym. Bull. (2014) 71:1241–1262 1245
123
Image analyses of the micrographs were carried out using the commercial software
ImageJTM [25].
Contact angle measurements
Contact angle measurements of the epoxies were conducted with an in-house
instrument which consists of an optical microscope Stereomaster II, Fisher
Scientific Model SPT-ITH equipped with a Motic1000 digital camera [26]. The
volume of the drop was maintained at 1 lL in all cases using a microsyringe. For
accuracy, measurements were repeated five times on different regions of the same
sample.
Microhardness measurements
Micro-indentation measurements were carried out at room temperature (19 �C)
using a Vickers diamond attached to a Leica tester. A load, P, of 0.25 N was applied
for 6 s. The diagonal of the residual impression left behind upon load release, d, was
measured using an optical microscope. Microhardness, H, was calculated according
to [27]:
H ¼ 1:85P
d2: ð1Þ
Results and discussion
It is noteworthy that both the epoxy resin DGEBA and the PDMS-DGE elastomer
contain oxirane rings as functional end groups (see Fig. 1). Hence, it is expected
that during the curing process, the curing agent DCH reacts with both the DGEBA
epoxy system and the PDMS-DGE elastomer. Polycondensation involves the
opening of the oxirane rings in the structures. The unreacted oxirane rings would
react with other diamine molecules to form a molecular network. It is suggested that
the elastomer is chemically bonded to the resin network, presumably forcing the
elastomer to remain inside the polymeric matrix. The mechanism of reaction
suggested is shown schematically in Fig. 2.
On the other hand, the second siloxane reactant utilized, PDMS-co-DPS-OH,
presents aromatic groups on the chemical structure (see Fig. 1), as DGEBA does,
and hence some compatibility between the elastomer and the epoxy resin is
presumed. The enthalpic interaction would promote the dissolution of the elastomer
in the epoxy matrix, at least partially; the elastomer being occluded within the
molecular network. In this system, the elastomer is not expected to be chemically
bonded to the crosslinked network; however, the presence of the hydroxyl end
groups in the elastomer may catalyze the curing reaction. The aim was to keep the
elastomer well dispersed within the epoxy matrix avoiding migration towards the
surface. The expectation was to obtain a composite with microdomains of limited
size.
1246 Polym. Bull. (2014) 71:1241–1262
123
The chemical structure of the epoxy-rubber composites was investigated using
FTIR spectroscopy. Figure 3 shows the spectrum for every system studied.
Figure 3a shows that at 2,961 cm-1 the intensity of the band in the epoxy-rubber
composites is increased (relative to the neat epoxy resin), which can be attributed to
the CH2 groups of the elastomers. On the other hand, Fig. 3b shows that the
DGEBA-(PDMS-co-DPS-OH) and the DGEBA-(PDMS-DGE) composites exhibit
peaks at about 1,258, 1,019, 799 cm-1 that are not present in the neat epoxy resin.
These vibrations are assigned to the Si–O bond in the elastomers.
Gel point determination
The gel point was determined for all the materials investigated during their curing
processes at temperatures in the range 90–110 �C using the method of the crossing
point of G0 and G00 in small-amplitude oscillatory shear experiments. Figure 4
shows the plots of dynamic shear moduli as a function of time during the curing
reaction at 95 �C for (a) DGEBA-(PDMS-DGE) and (b) DGEBA-(PDMS-co-DPS-
Fig. 2 Mechanism of reaction of DGEBA, DCH and (PDMS-DGE): a initiation, and b propagation
Polym. Bull. (2014) 71:1241–1262 1247
123
OH). Figure 4a shows that during the first 200 s the rheological behavior of
DGEBA-(PDMS-DGE) is liquid-like. After this dormant period the elastic modulus
G0 starts to increase rapidly, at a higher rate than the viscous modulus G00, increasing
over four orders of magnitude in less than 200 s. The G0 and G00 values coincide
Fig. 3 FTIR spectra of epoxy systems a 4,000–2,000 cm-1, b 1,800–600 cm-1. (i) DGEBA-DCH, (ii)DGEBA-(PDMS-co-DPS-OH), and (iii) DGEBA-(PDMS-DGE)
1248 Polym. Bull. (2014) 71:1241–1262
123
(G0 = G00 = 4 9 103 Pa) at tgel = 540 s defining the gel point. Thereafter, G0 still
increases two orders of magnitude before leveling off at about 1,050 s, thus defining
the transition into the glassy state and G0 reaching a value of 3 9 106 Pa. On the
other hand, the shear viscous modulus G00 still increases after the gel point although
at a lower rate than G0 and finally decreases with increasing reaction time as a
consequence of solidification because the elastic behavior of the glassy state
dominates the viscoelastic behavior of the epoxy [28].
Fig. 4 Evolution of elastic shear modulus, G0, and viscous shear modulus, G0 0, as a function of time, at95 �C. a DGEBA-(PDMS-DGE), and b DGEBA-(PDMS-co-DPS-OH)
Polym. Bull. (2014) 71:1241–1262 1249
123
The rheological response of DGEBA-(PDMS-co-DPS-OH) shown in Fig. 4b
exhibits some differences with respect to that described for DGEBA-(PDMS-DGE).
In the first place, the data in Fig. 4b show that the reaction occurred immediately
after starting the experiment, i.e., the elastic modulus G0 increasing rapidly over
three orders of magnitude. In addition, the gel point occurs at tgel = 252 s, i.e.,
about half the time necessary for gelation when using the ether-terminated
elastomer. Moreover, in case of the PDMS-co-DPS-OH material both shear moduli
increase at about the same rate above the gel point until solidification starts at
around 1,200 s, where finally G0 tend to level off and G0’ starts to decrease.
The results shown in Fig. 4 stresses out the distinct rheological behavior of the
two epoxy modified systems under curing reaction. The slower rate of shear moduli
growth observed for the diol terminated reactive, DGEBA-(PDMS-co-DPS-OH)
after the gel point suggests that the curing reaction is diffusion controlled. Although
the gel point was reached in about half the time than the ether-terminated elastomer
DGEBA-(PDMS-DGE) the moduli of DGEBA-(PDMS-co-DPS-OH) at the gel
point was 10 times smaller (3 9 104 vs. 4 9 103 Pa, respectively). It is presumed
that due to this lower modulus, there is still enough mobility within the three-
dimensional molecular network to enable the chemical reaction to continue.
Figure 5a, b shows the shear dynamic moduli as a function of reaction time at
100 �C for the two systems investigated. It is noteworthy that the rate of growth of
the shear moduli for the system DGEBA-(PDMS-DGE) (Fig. 5a) is now faster with
respect to measurements at 95 �C, and consequently the gel point is significantly
reduced, occurring at tgel = 430 s, about 100 s less than when curing at 95 �C (see
Fig. 4a). The modulus at the gel point is also higher, obtaining ca. 105 Pa.
Interestingly, the elastic modulus starts to level off (and the viscous modulus starts
to decrease) at ca. 550 s marking the onset towards the glassy state which is about
500 s less than that at 95 �C.
On the other hand, curing the DGEBA-(PDMS-co-DPS-OH) system at 100 �C
(Fig. 5b) the gel point occurred at tgel = 251 s, i.e., the same gel time obtained
when at 95 �C. However, the moduli values at the gel point are now substantially
increased, G0 = G00 = 6 9 104 Pa, an order of magnitude greater than those at
95 �C (see Fig. 3b). Moreover, the elastic shear modulus starts to level off (and the
viscous shear modulus starts to decrease) at 300 s, approaching a value at this
reaction time of G0300 s = 3 9 105 Pa. This rapid increase of shear moduli is
followed by a slowing down of the curing reaction reflected in the continuous and
slow growth of the elastic modulus, only reaching the onset for the glassy state
(G0 = 106 Pa) after holding for 1,200 s.
A further increase in reaction temperature to 105 �C produced the shear moduli
results illustrated in Fig. 6a, b. The plots in Fig. 6a show that tgel for the DGEBA-
(PDMS-DGE) system has now decreased to 350 s, and the modulus at the gel point
is G0 = G00 = 5 9 104 Pa. Thereafter, the elastic shear modulus increased rather
slowly, reaching the onset for the glassy state (G0 = 106 Pa) at about 1,000 s. On the
other hand, Fig. 6b shows that for DGEBA-(PDMS-co-DPS-OH), the gel point
occurred at tgel = 210 s, and the modulus at the gel point was G0 = G00 = 5 9 103
Pa. After the gel point the curing reaction continued rather slowly and reached the
glassy state at about 1,100 s.
1250 Polym. Bull. (2014) 71:1241–1262
123
Finally, Fig. 7a, b shows the plots of the shear moduli as a function of time of
isothermal holding at 110 �C. Figure 7a shows that for the DGEBA-(PDMS-DGE)
system the time for the gel point is further reduced, tgel = 100 s. On the other hand,
the onset for the glassy state is also significantly reduced, now occurring at about
600 s. Figure 7b shows that for the DGEBA-(PDMS-co-DPS-OH) system the
curing reaction is extremely fast, because at this temperature the curing reaction
occurred as soon as the material was placed onto the pre-heated rheometer, i.e.,
there was no observation of the gel point.
The results of Figs. 4, 5, 6, 7 demonstrate the influence of the elastomers on the
DGEBA system. Figure 8 shows a plot of the gel time, tgel, as a function of reaction
Fig. 5 Evolution of elastic shear modulus, G0, and viscous shear modulus, G0 0, as a function of time, at100 �C. a DGEBA-(PDMS-DGE), and b DGEBA-(PDMS-co-DPS-OH)
Polym. Bull. (2014) 71:1241–1262 1251
123
temperature. Figure 8a includes the gel time for the neat epoxy DGEBA resin at one
reaction temperature. On the other hand, Fig. 8b and c shows the gel time data for
DGEBA-(PDMS-DGE) and DGEBA-(PDMS-co-DPS-OH), respectively. The
results clearly show that the elastomers accelerate the curing reaction of DGEBA,
and the hydroxyl-terminated elastomer, PDMS-co-DPS-OH, is more efficient to
accelerate the curing reaction.
Assuming that the gel point occurred at a constant conversion, the dependence of
the gel time on temperature may be interpreted using an Arrhenius expression,
Fig. 6 Evolution of elastic shear modulus, G0, and viscous shear modulus, G0 0, as a function of time, at105 �C. a DGEBA-(PDMS-DGE), and b DGEBA-(PDMS-co-DPS-OH)
1252 Polym. Bull. (2014) 71:1241–1262
123
1
tgel
¼ Ae�Ea=RT ; ð2Þ
where Ea is the activation energy, T the absolute temperature, and R the ideal gas
constant.
Figure 9a, b shows the plot of natural logarithm of the reciprocal gel time as a
function of reciprocal absolute temperature for (a) DGEBA-(PDMS-DGE) and
(b) system. The results show that the data obey the Arrhenius expression. Thus, the
corresponding activation energies were extracted utilizing Eq. (2), and the results
showed that Ea = 122.6 kJ mol–1 for DGEBA-(PDMS-DGE), and Ea = 54.9 kJ
mol–1 for DGEBA-(PDMS-co-DPS-OH). These results indicate that the hydroxyl
Fig. 7 Evolution of elastic shear modulus, G0, and viscous shear modulus, G0 0, as a function of time, at110 �C. a DGEBA-(PDMS-DGE), and b DGEBA-(PDMS-co-DPS-OH)
Polym. Bull. (2014) 71:1241–1262 1253
123
groups indeed caused the catalysis that accelerated the curing reaction. These
activation energy values are in reasonable agreement with literature data for other
epoxy systems [22, 29–33].
Morphology
Cured epoxy samples were fractured under liquid nitrogen and examined using
SEM to evaluate whether there was phase separation during the curing reaction
between the DGEBA and the elastomers. Moreover, the fractured surfaces were
washed several times in hot acetone to extract any elastomer present on the surface.
As a cure reaction occurs, the molecular weight increases and an average molecular
weight is reached where a homogeneous mixture is no longer favored and,
consequently, the reacting rubbers may begin to separate into two phases [17–23].
Figure 10 shows a SEM micrograph for the neat epoxy resin DGEBA cured at
110 �C. The micrograph shows that the fractured surface exhibits sharp edges
typical of brittle mechanical behavior.
On the other hand, Fig. 11 shows micrographs of the surfaces of DGEBA-
(PDMS-DGE) composite cured at (a) 90 �C, (b) 100 �C, and (c) 110 �C,
respectively. The surfaces exhibit a microvoid morphology indicative of phase
separation between the epoxy resin and the ether-terminated elastomer, that is, the
microvoids correspond to space previously occupied by the elastomer. It is
interesting to find that the droplet size is relatively uniform at each temperature and
tends to decrease at higher curing temperatures. At 90 �C (Fig. 10a) the droplet size
is about 2 lm (±0.3 lm), whereas at 110 �C (Fig. 10c) the droplet size is within
Fig. 8 Gel time, tgel, as a function of temperature. a DGEBA, b DGEBA-(PDMS-DGE) composite,c DGEBA-(PDMS-co-DPS-OH) composite
1254 Polym. Bull. (2014) 71:1241–1262
123
1 lm. A slightly larger particle size has been reported for carboxyl-terminated
butadiene-acrylonitrile rubbers [20, 21].
Finally, Fig. 12 shows the fractured surfaces of DGEBA-(PDMS-co-DPS-OH)
composite cured at (a) 90 �C, (b) 100 �C, and (c) 110 �C, respectively. The
micrographs show a phase separated morphology at 90 and 100 �C (Fig. 12a, b,
respectively), with droplet sizes ranging from 1 to 3 lm. However, the sample cured
at 110 �C (Fig. 12c) does not show microvoid morphology, i.e., there is no evidence
of microphase separation. It is suggested that there was not microphase separation in
ln (
1/t g
el)
1/T (°K-1)
0.00260 0.00265 0.00270 0.00275 0.00280-6.4
-6.0
-5.6
-5.2
-4.8
-4.4(a)
(b)
Fig. 9 Arrhenius plot of reciprocal gel time as a function of reciprocal curing temperature for a DGEBA-(PDMS-DGE) and b DGEBA-(PDMS-co-DPS-OH) composites
Fig. 10 Scanning electronmicrograph of the epoxy resinDGEBA cured at 110 �C
Polym. Bull. (2014) 71:1241–1262 1255
123
Fig. 11 Scanning electronmicrographs of the DGEBA-(PDMS-DGE) composite curedat a 90 �C, b 100 �C, andc 110 �C
1256 Polym. Bull. (2014) 71:1241–1262
123
Fig. 12 Scanning electronmicrographs of the DGEBA-(PDMS-co-DPS-OH) compositecured at a 90 �C, b 100 �C, andc 110 �C
Polym. Bull. (2014) 71:1241–1262 1257
123
this case due to the faster cure reaction afforded by PDMS-co-DPS-OH. The
rheological results showed that at 110 �C the hydroxyl-terminated elastomer did
react at the moment the sample was loaded in the rheometer and a gel point could
not be detected (see Figs. 7b, 8b). Moreover, it is also suggested that the hydroxyl
terminated PDMS was solubilized within the DGEBA as initially hypothesized, and
this also contributed to avoid microphase separation between DGEBA and the
elastomer.
Surface properties
The contact angle is a thermodynamic parameter for insoluble solids that can be
used to determine the interfacial tension and show the change in the hydrophilic
nature of the materials. The determination of surface tension of a solid from contact
angle measurements has been extensively studied [34–37]. Water contact angle
measurements were carried out on all the cured samples, including samples cured at
140 �C in the oven, and the results and representative photographs of water drops on
the samples surfaces are shown in Fig. 13. The results show that the DGEBA epoxy
cured without rubbers displays a contact angle of 78� (Fig. 13a). When PDMS-DGE
is added to the DGEBA epoxy, the contact angle diminishes, the difference being
more conspicuous if the composite is cured at low temperatures (Fig. 13b). In
contrast, when PDMS-co-DPS-OH is added to the DGEBA epoxy the water contact
angle at low curing temperatures (Fig. 13c) is greater than that of the neat DGEBA.
It is believed that the greater contact angle (more hydrophobic behavior) can be
attributed to the hydrophobic nature of the PDMS-co-DPS-OH structure, which
contains aromatic rings that possess this property. Despite having OH groups, its
Curing temperature (°C)
65
70
75
80
85
90
80 90 100 110 120 130 140 150
Co
nta
ct a
ng
le (
°)
(a)
(b)
(c)
Fig. 13 Water contact angle (h) as a function of curing temperature. a DGEBA, b DGEBA-(PDMS-DGE), and c DGEBA-(PDMS-co-DPS-OH)
1258 Polym. Bull. (2014) 71:1241–1262
123
behavior is of low polarity because the electronic resonance of the aromatic rings
reduces the dipole moment of the terminal OH groups. Figure 13c shows that
increasing the curing temperature produces a decrease of the contact angle, i.e., the
surface shows a tendency to become more hydrophilic. Furthermore, at high curing
temperatures the water contact angle was the same as that for neat DGEBA. This
result suggests that the PDMS-co-DPS-OH elastomer was indeed solubilized within
the epoxy matrix, in agreement with the SEM results shown in Fig. 12.
Microhardness
Finally, the mechanical behavior of the epoxy-rubber composites was investigated
via micro-indentation measurements. Figure 14 shows a plot of hardness as a
function of curing temperature for (a) the neat epoxy resin, and the composites
(b) DGEBA-(PDMS-co-DPS-OH), and (c) DGEBA-(PDMS-DGE). First, the results
show that there is indeed a reduction in hardness due to the inclusion of elastomers
in the epoxy resin. Samples cured at 140 �C exhibited about 12 and 18 % reduction
in hardness for the hydroxyl-terminated and ether-terminated rubber-epoxy
composites, respectively, relative to the neat epoxy resin. Second, the hardness of
the DGEBA-(PDMS-co-DPS-OH) composite is a growing function of the curing
temperature. These results suggest that the hardness of the hydroxyl-terminated
elastomer-epoxy composite increases due to the gradual decrease of droplet size at
higher temperatures, until eventually there appears no microphase separation (see
Fig. 12). Indeed, the dispersion of elastomer droplets in the epoxy matrix introduces
interfaces (or interphases) that hamper the transfer of load across the sample. Hence,
morphologies with important matrix-particle interfaces are expected to be
detrimental to the mechanical properties with respect to those with no significant
200
220
240
260
90 100 110 120 130 140 150
Temperature (°C)
Har
dn
ess
H (
MP
a)
(a)
(b)
(c)
20 µm
Fig. 14 Hardness as a function of curing temperature for a DGEBA, b DGEBA-(PDMS-co-DPS-OH),and c DGEBA-(PDMS-DGE). The inset shows, as an example, one indentation produced on the surfaceof the DGEBA sample cured at 140 �C
Polym. Bull. (2014) 71:1241–1262 1259
123
phase separation. In the same direction, the lower hardness value for the DGEBA-
(PDMS-DGE) composite may be due to the phase-separated morphology that the
composite exhibits at all curing temperatures (see Fig. 11).
Conclusions
The curing reaction of the DGEBA resin and the 1,2-DCH hardener with 5 phr
concentration of two different liquid elastomers (one elastomer is ether terminated,
PDMS-DGE; the other one is dihydroxy terminated with diphenyl rings in the main
backbone, PDMS-co-DPS-OH) is investigated at different temperatures by means of
shear rheometry. The morphology of the resultant composites is explored using
SEM and correlated to the hydrophilic properties and the mechanical behavior by
indentation testing.
Results show that the gel point decreases as the curing temperature is increased.
It is clearly shown that the elastomers accelerate the curing reaction of DGEBA
leading to significantly shorter gel times. This is especially marked in the case of the
hydroxyl-terminated elastomer where the gel point at the highest curing temperature
(110 �C) was not detected presumably due to the rapid curing kinetics.
Electron microscopy studies on the fractured surfaces of the composite samples
revealed microvoid morphology for all the composites investigated except for that
incorporating the hydroxyl-terminated elastomer and further cured at the highest
temperature (110 �C). The microvoid morphology is indicative of a two-phase
composite where the elastomer droplets have been pulled out from the epoxy
matrix. It is found that the droplet size is relatively uniform in all cases, in the range
1–3 lm, and exhibits a tendency to decrease with increasing curing temperature.
Most interesting is the morphology of the DGEBA-(PDMS-co-DPS-OH) material
cured at 110 �C where there is no evidence of phase separation possibly due to the
rapid curing kinetics. It is suggested that the diphenyl and hydroxy groups promoted
elastomer miscibility with the thermoset resin and microphase separation is only
observed at long reaction times.
Water contact angle measurements revealed that hydrophobicity of the epoxy
matrix diminishes with the incorporation of PDMS-DGE. In contrast, the PDMS-co-
DPS-OH elastomer enhances hydrophobicity as a consequence of the presence of
aromatic rings in the backbone structure. The surface of the DGEBA-(PDMS-co-
DPS-OH) composites induces higher water contact angles than the neat epoxy only
for low curing temperatures; at high temperatures, the contact angle values approach
those of the neat epoxy suggesting that PDMS-co-DPS-OH elastomer was indeed
solubilized within the epoxy matrix, in agreement with the morphological results.
Indentation measurements revealed that inclusion of either elastomer reduces the
microhardness values of the epoxy resin. In addition, it is found that the
microhardness of the DGEBA-(PDMS-co-DPS-OH) composite increases as the
curing temperature is raised. This result is correlated with the morphological studies
obtained by SEM and shows that phase separation drives the microhardness
behavior.
1260 Polym. Bull. (2014) 71:1241–1262
123
Acknowledgments C. Valerio-Cardenas was supported by a postdoctoral fellowship from DGAPA–
UNAM. This research was financially supported by PAPIIT (IG101313) and CONACyT, program CB-
2011 (Grant 168095). A. E. G. is grateful to DGAPA-UNAM (PAPIIT Project IN-106008) for partial
support. One of the authors (AF) wishes to thank the MICINN (Ministerio de Ciencia e Innovacion),
Spain, for financial support of the research (Grant FIS2010-18069).
References
1. Winter HH (1987) Can the gel point of a cross-linking polymer be detected by the G0–G0 0 crossover?
Polym Eng Sci 27:1698–1702
2. Winter HH (1987) Evolution of rheology during chemical gelation. Progr Col Poly Sci 75:104–110
3. Larson RG (1999) The structure and rheology of complex fluids, 2nd edn. Oxford University Press,
Oxford
4. Winter HH, Chambon F (1986) Analysis of linear viscoelasticity of a crosslinking polymer at the gel
point. J Rheol 30:367–382
5. Tung C-YM, Dynes PJ (1982) Relationship between viscoelastic properties and gelation in ther-
mosetting systems. J Appl Polym Sci 27:569–574
6. Nunez L, Gomez-Barreiro S, Gracia-Fernandez CA (2005) Study of the influence of isomerism on the
curing properties of the epoxy system DGEBA (n = 0)/1,2-DCH by rheology. Rheol Acta
45:184–191
7. Rudin A (1999) The elements of polymer science and engineering, 2nd edn. Academic Press, New
York, p 11
8. Liu P, He L, Song J, Liang X, Ding H (2008) Microstructure and thermal properties of silyl-
terminated polycaprolactone-polysiloxane modified epoxy resin composites. J Appl Polym Sci
109:1105–1113
9. Laza JM, Julian CA, Larrauri E, Rodrıguez M, Leon LM (1999) Thermal scanning rheometer
analysis of curing kinetic of an epoxy resin: 2. An amine as curing agent. Polymer 40:35–45
10. Carrozzino S, Levita G, Rolla P, Tombari E (1990) Calorimetric and microwave dielectric moni-
toring of epoxy resin cure. Polymer Eng Sci 30:366–373
11. Vyazovkin S, Sbirrazzuoli N (1996) Mechanism and kinetics of epoxy-amine cure studied by dif-
ferential scanning calorimetry. Macromolecules 29:1867–1873
12. Zhao F, Sun Q, Fang DP, Yao K (2000) Preparation and properties of polydimethylsiloxane-modified
epoxy resins. J Appl Polym Sci 76:1683–1690
13. Hou S-S, Chung Y-P, Chan C-K, Kuo P-L (2000) Function and performance of silicone copolymer.
Part IV. Curing behavior and characterization of epoxy-siloxane copolymers blended with diglycidyl
ether of bisphenol-A. Polymer 41:3263–3272
14. Rey L, Poisson N, Maazouz A, Sautereau H (1999) Enhancement of crack propagation resistance in
epoxy resins by introducing poly(dimethylsiloxane) particles. J Mater Sci 34:1775–1781
15. Konczol L, Doll W, Buchholz U, Mulhaupt R (1994) Ultimate properties of epoxy resins modified
with a polysiloxane-polycaprolactone block copolymer. J Appl Polym Sci 54:815–826
16. Anand Prabu A, Alagar M (2005) Thermal and morphological properties of silicone-polyurethane-
epoxy intercrosslinked matrix materials. J Macromol Sci Part A Pure Appl Chem 42:175–188
17. Lin S-T, Huang S-K (1996) Preparation and structural determination of siloxane-modified sulfone-
containing epoxy resins. J Polym Sci Part A Polym Chem 34:869–884
18. Lin S-T, Huang S-K (1996) Synthesis and impact properties of siloxanes-DGEBA epoxy copolymers.
J Polym Sci Part A Polym Chem 34:1907–1922
19. Tripathi G, Srivastava D (2007) Effect of carboxyl-terminated poly(butadiene-co-acrylonitrile)
(CTBN) concentration on thermal and mechanical properties of binary blends of diglycidyl ether of
bisphenol-A (DGEBA) epoxy resin. Mat Sci Eng A 443:262–269
20. Ramos VD, da Costa HM, Soares VLP, Nascimento RSV (2005) Modification of epoxy resin: a
comparison of different types of elastomers. Polym Test 24:387–394
21. Tripathi G, Srivastava D (2009) Toughened cycloaliphatic epoxy resin for demanding thermal
applications and surface coatings. J Appl Polym Sci 114:2769–2776
22. Thomas R, Durix S, Sinturel C, Omonov T, Goossens S, Groeninckx G, Moldenaers P, Thomas S
(2007) Cure kinetics, morphology, and miscibility of modified DGEBA-based epoxy resin: effects of
a liquid rubber inclusion. Polymer 48:1695–1710
Polym. Bull. (2014) 71:1241–1262 1261
123
23. Thomas R, Yumei D, Yuelong H, Le Y, Moldenaers P, Weimin Y, Czigany T, Thomas S (2008)
Miscibility, morphology, thermal and mechanical properties of a DGEBA based epoxy resin
toughened with a liquid rubber. Polymer 49:278–294
24. Calabrese L, Valenza A (2003) Effect of CTBN rubber inclusions on the curing kinetic of DGEBA-
DGEBF epoxy resin. Eur Polym J 39:1355–1363
25. ImageJ�. Software developed by the National Institutes of Health, USA
26. Castillo-Perez R, Romo-Uribe A (2012) Diseno y construccion de un instrumento para medir angulo
de contacto. Memorias del XVIII Congreso Internacional Anual de la SOMIM, ISBN: 978-607-
95309-6-9, pp 772–779
27. Flores A, Ania F, Balta-Calleja FJ (2009) From the glassy state to ordered polymer structures. A
microhardness study. Polymer 50:729–746
28. Ferry JD (1980) Viscoelastic properties of polymers. Wiley, New York
29. Canamero-Martınez P, Fernandez-Garcıa M, de la Fuente JL (2010) Rheological cure character-
ization of a polyfunctional epoxy acrylic resin. React Funct Polym 70:761–766
30. Wisanrakkit G, Gillham JK (1990) The glass transition temperature (Tg) as an index of chemical
conversion for a high-Tg amine/epoxy system: chemical and diffusion-controlled reaction kinetics.
J Appl Polym Sci 41:2885–2929
31. Mortimer S, Ryan AJ, Stanford JL (2001) Rheological behavior and gel-point determination for a
model lewis acid-initiated chain growth epoxy resin. Macromolecules 34:2973–2980
32. Takiguchi O, Ishikawa D, Sugimoto M, Taniguchi T, Koyama K (2008) Effect of rheological
behavior of epoxy during procuring on foaming. J Appl Polym Sci 110:657–662
33. Choe Y, Kim M, Kim W (2003) In situ detection of the onset of phase separation and gelation in
epoxy/anhydride/thermoplastic blends. Macromol Res 11:267–272
34. Neumann AW, Good RJ, Hope CJ, Sejpal M (1974) An equation-of-state approach to determine
surface tensions of low-energy solids from contact angles. J Colloid Interface Sci 4:291–304
35. Spelt JK, Li D, Neumann AW (1992) Modern approaches to wettability. In: Schrader ME, Loeb GI
(eds) Plenum Press, New York, pp 101–142
36. Grundke K, Bogumil T, Werner C, Janke A, Poschel K, Jacobasch H (1996) Liquid-fluid contact
angle measurements on hydrophilic cellulosic materials. J Colloid Surf A Physicochem Eng Asp
116:79–91
37. Gulec HA, Sarioglu K, Mutlu M (2006) Modification of food contacting surfaces by plasma poly-
merization technique. Part I: determination of hydrophilicity, hydrophobicity and surface free energy
by contact angle method. J Food Eng 75:187–195
1262 Polym. Bull. (2014) 71:1241–1262
123
