close

Вход

Забыли?

вход по аккаунту

?

818

код для вставкиСкачать
Polymer International
Polym Int 48:787±793 (1999)
Bulk properties of epoxy resin modified by
epoxy–aminosilane copolymers
MG Lu,1 MJ Shim2 and SW Kim1*
1
Department of Chemical Engineering, University of Seoul, Seoul 130–743, Korea
Department of Life Science, University of Seoul, Seoul 130–743, Korea
2
Abstract: A linear epoxy±aminopropyltriethoxysilane addition polymer was used as a new epoxy
modi®er. In this paper, the thermal and mechanical properties have been determined by differential
scanning calorimetry (DSC) and mechanical testing. The experimental results show that this copolymer modi®er can effectively improve the toughness of resins without sacri®cing their thermal
resistance, stiffness and strength. As a comparison, the properties of epoxy resin blended with aminopropyltriethoxysilane (g-APS) have been carried out simultaneously.
# 1999 Society of Chemical Industry
Keywords: epoxy; thermal analysis; fracture toughness; mechanical property
INTRODUCTION
Epoxides constitute a very important class of thermosetting polymers and are used in adhesives, as matrices
for ®bre-reinforced composites, and as coatings.
Epoxides have excellent mechanical and thermal
properties, such as high strength, elastic modulus
and glass transition temperature (Tg). These properties result from the chemical nature of the monomers
and the high crosslink density of the ®nal materials.
However, this structure also leads to low resistance to
crack initiation and propagation, so that methods to
increase toughness are needed.1,2
Two strategies are being applied: matrix ¯exibilization by means of ¯exible segments, and dispersion of
stress-concentrating microphases. Because matrix
¯exibility and reduced crosslink density are associated
with a drastic softening of the matrix, the ®rst strategy
leads to improved toughness at the expense of stiffness
and glass transition temperature. As a consequence,
the second strategy, involving dispersion of discrete
nano- and microphases in the continuous rigid epoxy
matrix, represents the method of choice to promote
toughness and to preserve high stiffness, strength and
heat distortion temperature.3,4
For decades, the addition polymerization of amines
and diepoxides has been investigated intensively,5±7
and using special polymerization conditions, thermoplastic linear addition polymers have been obtained.8
The principle of linear epoxide±amine addition polymerization makes it possible to incorporate other
functional groups and amines with special properties
into polymer chains.
It is well known that silicone compounds have
unique chemical and physical properties, such as low
glass transition temperature, extremely low wateruptake, low dielectric constant, and excellent thermooxidative stability. Reactive liquid silicone rubber has
been used as an epoxy toughening agent, and silane
coupling agents can change the interface between an
organic polymer and an inorganic substrate, enhance
bond strength and prevent debonding at the interface
of composites. Unfortunately, most silicones and also
silicone rubbers are immiscible with common epoxy
resins and hardeners. They must be compatibilized
either via advancement reaction or via incorporation of
more polar substituents.9,10
Copolymers synthesized by addition polymerization
of epoxy and g-APS not only combine the properties of
silicones but, more importantly, conveniently enhance
the compatibility between polysiloxane and resin. It
then becomes possible to use this copolymer as a new
toughening modi®er for epoxy resins.
In this paper, we shall discuss the effects of an
addition copolymer on the thermal and mechanical
behaviour of epoxy resins and compare the results with
the blend system of epoxy and g-APS.
EXPERIMENTAL
Materials
The epoxy resin systems used in this study consist of a
commercial diglycidylether of bisphenol A (DGEBA)
(Shell Chemicals, USA, Epon 828), 4,4'-methylene
dianiline (MDA) (Fluka Chemie AG, Buchs) and gaminopropyltriethoxysilane (g-APS) (Aldrich Chemical Co Inc, USA). These materials were used as
supplied, without further puri®cation. The epoxide
equivalent weight of the resin was taken as 184. The
* Correspondence to: Sangwook Kim, Department of Chemical Engineering, University of Seoul, 90 Jeonnong Dong, Dongdaemun Gu, Seoul
130–743, Korea
(Received 28 August 1998; revised version received 8 March 1999; accepted 22 April 1999)
# 1999 Society of Chemical Industry. Polym Int 0959±8103/99/$17.50
787
MG Lu, MJ Shim, SW Kim
copolymer of epoxy and g-APS was synthesized
previously in our laboratory with a number average
molecular weight of 1280.11,12
Experimental procedure
The resin was cured employing stoichiometric quantities of diamine. The epoxy resin was mixed with
various concentrations of copolymer of g-APS (the
percentage composition of copolymer is recorded with
reference to g-APS) and melted MDA with ef®cient
stirring at about 80 °C then cured at 80 °C for 1.5 h and
150 °C for 30 min. The products were then slowly
cooled to room temperature inside the oven.
Samples consisting of 3±6 mg of cured resins were
placed in open aluminium DSC pans and run on a
Seiko I-5000 series (Seiko, Japan) instrument. The
glass transition temperature and thermal degradation
character of the polymers were determined by dynamic
differential scanning calorimetry (DSC) and thermogravimetric (TG) analysis, respectively, at a heating
rate of 10 K minÿ1. The curing reaction rates and
degrees of conversion were obtained from isothermal
curing experiments. All the thermal analysis experiments were carried out under an atmosphere of
nitrogen at a ¯ow rate of 50 ml minÿ1
The cast resin sheets (4 mm thick) were cut into
rectangular plates (6 45 4 mm3), and a notch was
made in the middle of the specimen (single edge
notch). The depth of the notch was 2.5 mm.
The ¯exure test was performed in three-point
bending mode using non-notched specimens by an
Instron mechanical testing machine (Shimadzu,
Japan) at a crosshead speed of 1 mm minÿ1. The
fracture intensive factor was determined using the
same method and operation, but notched specimens
were used.13 The morphology of the fracture surface
was examined using a scanning electron microscope.
Figure 1. A typical DSC curve for Tg measurement (sample with 2%
copolymer).
RESULTS AND DISCUSSION
Figure 1 is a typical DSC scan for the measurement of
glass transition temperature. Figure 2 shows the glass
transition temperature as a function of siloxane
content. As can be seen, for the resin modi®ed by
copolymer, the introduction of ¯exible siloxane
linkages slightly decreases the Tg of the modi®ed
resins. Partial miscibility between siloxane and epoxy
is suggested by the relatively lower Tg values for some
of the modi®ed resins. Although dimethyl siloxane
exhibits little compatibility with epoxy resin, the
solubility can be improved by the epoxy units of the
copolymer. Therefore, It is possible that the siloxane
moiety was aggregated to form domains or resulted in
microscopic phase separation. The values of Tg are
listed in Table 1 for both systems with various concentrations.
By comparing the two curves in Fig 2, we can
determine the difference between the two systems. For
the blend system, the Tg of the resins decreases when
low concentrations of g-APS are added, and then
788
Figure 2. Tg as a function of siloxane content.
slightly increases with increasing concentrations of
g-APS. It is clear that the solubility of polysiloxane in
epoxy resin is limited by the process of the cure
reaction between epoxy and siloxane. At lower
concentration of g-APS, polysiloxane is miscible with
neat resin, but the siloxane component in the resin
migrates to the surface during curing, or macroscopic
phase separation may take place with an increase in
siloxane composition. This phenomenon was observed
by other authors.14 The increase of Tg at higher
concentration of siloxane is attributed to the higher
degree of crosslinking, because the hydrogens in
g-APS have the same reactivity as those in an aliphatic
amine, and can participate in the cure reaction of the
epoxy resin. Figures 3 and 4 show the relationships
between curing rate, degree of conversion and reaction
time. It is clear that both the curing rates and degrees
Polym Int 48:787±793 (1999)
Bulk properties of modi®ed epoxy resin
Table 1. Thermal properties of polymers
DGEBA-MDA/Copolymer
DGEBA-MDA/g-APS
Content of siloxane (%)
T0.5 PDT (max) Char yield (%) up to 600 °C
0
1
2
5
8
397
392
393
403
513
377
381
380
380
383
18
23
23
29
46
Figure 3. Curing rate as a function of curing time for both copolymer and
blend systems with 2% siloxane at 100°C.
Figure 4. Degree of cure as a function of time for copolymer and blend
systems with 2% of siloxane at 100°C.
of conversion for blends are higher than those for
copolymer systems at given curing temperatures and
times.
The experimental results of polymer thermal dePolym Int 48:787±793 (1999)
Tg
162
147
153
151
143
T0.5 PDT (max) Char yield (%) up to 600 °C
397
401
414
419
534
377
376
374
376
380
18
24
26
31
48
Tg
162
148
149
156
155
Figure 5. TG curves for epoxy resins modified by copolymer with different
contents of siloxane.
composition determined by TG analysis are summarized in Table 1 for epoxy resin with different contents
of copolymer and g-APS. As is already known for
general rubber modi®cation of epoxy resin, the
thermal stability of materials is more or less sacri®ced
to improve their inherent brittleness. Especially for
polymeric diene modi®cation, there will be a lower
thermo-oxidation stability because these materials
contain unsaturated double bonds in their molecular
skeleton.15 However, an epoxy resin modi®ed by its
copolymer with siloxane exhibits a higher thermal
resistance because of the higher chemical bond-energy
of Si-O-Si chain segments (see Fig 5). It is worth
noting that higher char residues were observed with an
increase of the extent of siloxane. For example, nearly
50% of char yield is obtained up to 650 °C after adding
8% of siloxane. This may imply good ¯ame retardation
for this modi®ed epoxy resin system.16
Having a different glass transition temperature, the
system blended with g-APS exhibits similar thermal
degradation behaviour to that of the copolymer
system. As can be seen from Table 1, the polymer
decomposition temperature is gradually increased with
increasing siloxane concentration, and becomes even
higher than that of the copolymerized system. This can
789
MG Lu, MJ Shim, SW Kim
Figure 6. Variation of flexural strength (sc) with content of siloxane.
be attributed to the polysiloxane phase being more
temperature resistant than the epoxy resin, and to a
higher degree of crosslinking in blend systems, as
described above. The temperature at which half of the
polymer is decomposed (T0.5), the temperature at
which the polymer maximum rate of weight loss
occurred (PDTmax), and the char yields are given in
Table 1.
The variation of some ¯exural properties of the
epoxy resins with different contents of copolymer of
g-APS was determined. Figure 6 shows the relationship between the ¯exural strength of resins and the
contents of siloxane in copolymer or blends. As can be
seen, the ¯exural strength increases with increasing
copolymer contents up to 5%, then remains practically
constant. For the blend system, the ¯exural strength
increases ®rst until about 2% of g-APS has been
added, and then decreases abruptly because of the
compatibility limitation between neat resin and polymerized siloxane. This is in agreement with the
analytical results by glass transition temperature, ie
microphase separation only occurred in the system
modi®ed by copolymer. The decrease of ¯exural
strength in blends might be a re¯ection of macroscopic
phase separation and the result of over-crosslinking.
The behaviour of elongation at break shown in Fig 7
is similar to that of ¯exural strength. Meanwhile, the
experimental results show that the Young's modulus
(E) is almost independent of the siloxane contents for
both systems in the range investigated (see Fig 8). It is
not surprising, in view of the above comments, that the
dispersion of discrete nano- and microphases in the
continuous rigid epoxy matrix makes it possible to
promote toughness and to preserve high stiffness. For
the blend systems, a small quantity of siloxane
component will also not decrease the modulus of the
cured resins because of its incompatibility with epoxy.
However, a high g-APS concentration may result in a
bad microstructure that induces an unexpected
790
Figure 7. Variation of elongation at break (εb) with content of siloxane.
Figure 8. Relations of Young’s modulus (E) and content of siloxane.
physical property in the blend system. The ¯exural
properties of modi®ed epoxy resins are given in Table
2.
The fracture of materials is measured by a test on a
specimen of ®nite size, containing a machined notch
which has been extended by controlled fatigue loading, so that the end of the notch should be as sharp as
possible. For the three-point-bend specimen mode
KIC ˆ PL 1
a2
BW 2
where KIC is the critical stress intensity factor (fracture
toughness); P is the load at fracture; L is the span
between the supports; B is specimen thickness; W is
specimen width; a is depth of the precracked notch; a*
is a geometrical factor which is a function of the (a/w)
Polym Int 48:787±793 (1999)
Bulk properties of modi®ed epoxy resin
Table 2. Mechanical properties of polymers
DGEBA-MDA/Copolymer
Content of siloxane (%)
0
1
2
5
8
DGEBA-MDA/g-APS
E (GPa) sc (MPa) Elongation (%) KIC (MPa m1/2) E (GPa) sc (MPa) Elongation (%) KIC (MPa m1/2)
2.3
2.0
2.1
2.1
1.9
65.7
69.8
74.0
80.0
73.3
4.9
5.6
6.2
8.6
9.0
0.57
0.89
0.94
1.05
1.14
2.3
2.2
2.0
2.1
2.1
65.7
67.2
71.5
51.5
50.9
4.9
5.2
5.5
4.2
4.0
0.57
0.64
0.71
0.51
±
Figure 10. Scanning electron micrograph of an unmodified epoxy sample.
Figure 9. Fracture toughness (KIC) as a function of content of siloxane.
ratio. In this case
ˆ 2:03ÿ
a
a 2
a 3
a 4
ÿ 33:51
‡ 35:41
‡ 12:69
3:27
w
w
w
w
The values of KIC shown in Table 2 and Fig 9 indicate
that the fracture toughness of the materials obviously
increases with increasing contents of siloxane, even for
the system in which the epoxy±siloxane copolymer was
at relatively low concentration; as expected, such an
improvement of mechanical property was achieved,
but not at the expense of the stiffness expressed by
Young's modulus (E) and ¯exural strength (sc).
However, for the blend system, the values of KIC
decreased when an excess of g-APS was added, which
is consistent with the results of ¯exural strength.
Similar results were observed by other investigators.
In their work a poly(ether ester), linear or branched
segmented silicone rubber was used to modify epoxy
resins.17 Sue et al 18 evaluated particulate modi®ed
grafted rubber concentrates (GRC) for performance in
toughening a brittle epoxy resin cure system DGEBA/
DDS (diaminodiphenyl sulphone). The GRC rubber
Polym Int 48:787±793 (1999)
provided the most effective toughening without the
usual loss of Tg.
To toughen epoxides effectively, some possible
toughening models have been proposed, including
crazing, shear yielding and cavitation. The toughening
principles for relatively ductile polymers and brittle
epoxies were reviewed and discussed by Yee and
Pearson,19,20 Kinloch et al,21 and Garg and Mai.22
Based on the experimental work conducted in the
present study, an approach for toughening epoxies is
discussed below.
Figures 10±12 show scanning electron micrographs
of the fracture surfaces for both copolymerized and
blended systems. The materials used were chosen on
the basis of their mechanical behaviour. SEM of the
samples shows that an epoxy modi®ed by a copolymer
has a signi®cantly different morphology. Compared
with pure resin (Fig 10), the matrices containing
copolymers have undergone plastic deformation that
increases with the siloxane content (Fig 12). This also
occurred in low concentration blends (Fig 11B, C).
However, the specimen with a high siloxane content in
a blend system (Fig 11A) and pure resin show typical
brittle fracture, which is consistent with the results
obtained from the mechanical measurements. The
specimen with higher elongation and toughness
exhibited ductile fracture with cavitation, and the
damage zone can also be observed (Fig 12C). The
plastic deformation observed may originate from
deformation of the siloxane zone that is too small to
791
MG Lu, MJ Shim, SW Kim
Figure 11. Scanning electron micrographs of blended samples (siloxane
content: A, 1%; B, 2%; C, 5%).
Figure 12. Scanning electron micrographs of samples with copolymers
(siloxane content: A, 1%; B, 2%; C, 5%).
be seen in low magni®cation micrographs, or the
interface between the siloxane domain and the matrix
may be obscured because of the low siloxane content
and improved compatibility by copolymers. In order to
understand the toughening mechanisms and the
sequence of failure events in modi®ed epoxies, it is
imperative that the damage zone of the modi®ed
systems, the micromechanical deformation process
and the interface adhesion between the matrix and the
toughener phase be studied.
CONCLUSIONS
792
Our results show that the composition of siloxane in
copolymer improves the fracture toughness of modi®ed resins without sacri®cing the stiffness and
strength, even at levels of 5% siloxane moiety in
modi®ed resins. However, a miscibility limitation
exists between neat resin and siloxane with the
progress of cure in the blend system, and only at low
g-APS contents (2%) can slight increases in the
mechanical properties of resins be achieved. However,
Polym Int 48:787±793 (1999)
Bulk properties of modi®ed epoxy resin
similar thermal degradation characteristics are observed for those two systems. The modi®ed resin
exhibits a higher char yield. Therefore, it is possible to
use an epoxy±silane copolymer as a toughening agent
for epoxy resin.
ACKNOWLEDGEMENTS
This work was supported by SK group, Jeong Moon
Information Co. Ltd., and Keuk Dong Design &
Communication Co. Ltd. in Korea.
REFERENCES
1 Kinloch AJ, in Advances in Polymer Science, Vol 72, Ed by DuÃsek
K, Springer-Verlag, Berlin. p 45 (1985).
2 May CA, Epoxy Resins ± Chemistry and Technology, Marcel
Dekker, New York. p 551 (1987).
3 Siebert AR, in Rubber Modi®ed Thermoset Resins, Ed by Riew CK
and Gillham JK, Advances in Chemistry Series, No. 208,
American Chemical Society, Washington DC. p 179 (1984).
4 Arends CB, Polymer Toughening, Marcel Dekker, NY. p 175
(1996).
5 Gramain P and Frere Y, Polymer 21:921 (1980).
6 Zentel R, Makromol Chem 194:859 (1993).
Polym Int 48:787±793 (1999)
7 Klee JE, Horhold HH and Schutz H, Acta Polym 38:293 (1987).
8 Chien LC, Lin C and Fredley DS, Macromolecules 25:133 (1992).
9 Yorkgitis EM, Eiss NS, Tran C Jr, Wilkes GL and McGrath JE,
in Advances in Polymer Science, Vol 72, Ed by DuÃsek K,
Springer-Verlag, Berlin. p 79 (1985).
10 Hu TY, Eiss NS, Yorkgitis EM, Wilkes GL, Tran C, Yilgor I and
McGrath JE, Polym Mater Sci Eng, 49: (1983).
11 Lu MG and Kim SW, Proceedings of IUMRS-ICA, September
1997, Chiba, Japan, p 125.
12 Ramis X and Salla JM, J Appl Polym Sci 45:227 (1992).
13 Williams JG, Fracture Mechanics of Polymers, Ellis Horwood,
Chichester. p 272 (1984).
14 Hedrick JI, Yilgor I, Jurek M, Hedrick JC, Wilkes GL and
McGrath JE, Polymer 32:2020 (1991).
15 Yamini S and Young R, J Mater Sci 15:1814 (1980).
16 Turi EA, Thermal Characterization of Polymer Materials, Academic Press, New York, p 435 (1983).
17 Shi YB, Groleau MR and Lee AF, Polym Mater Sci Eng 70:110
(1994).
18 Sue HJ, Garcia-Meitin EI and Orchard NA, J Polym Sci Polym
Phys Ed 31:595 (1993).
19 Yee AF and Pearson RA, J Mater Sci 21:2462 (1986).
20 Pearson RA and Yee AF, J Mater Sci 26:3828 (1991).
21 Kinloch AJ Shaw SJ, Tod DA and Hunston DJ, Polymer 24:1341
(1983).
22 Garg AC and Mai YW, Compos Sci Technol 31:179 (1988).
793
Документ
Категория
Без категории
Просмотров
2
Размер файла
577 Кб
Теги
818
1/--страниц
Пожаловаться на содержимое документа