Polymer 45 (2004) 89578966

www.elsevier.com/locate/polymer

Effects of polyamide 6 incorporation to the short glass fiber reinforced

ABS composites: an interfacial approach
Guralp Ozkoca, Goknur Bayramb,*, Erdal Bayramlic
a

Department of Polymer Science and Technology, Middle East Technical University, 06531 Ankara, Turkey
b
Department of Chemical Engineering, Middle East Technical University, 06531 Ankara, Turkey
c
Department of Chemistry, Middle East Technical University, 06531 Ankara, Turkey
Received 14 May 2004; received in revised form 4 September 2004; accepted 13 October 2004
Available online 28 October 2004

Abstract
The properties of 30 wt% short glass fiber (SGF) reinforced acrylonitrile-butadiene-styrene (ABS) terpolymer and polyamide 6 (PA6)
blends prepared with extrusion were studied using the interfacial adhesion approach. Work of adhesion and interlaminar shear strength values
were calculated respectively from experimentally determined interfacial tensions and short beam flexural tests. The adhesion capacities of
glass fibers with different surface treatments of organosilanes were evaluated. Among the different silanes tested, g-aminopropyltrimethoxysilane (APS) was found to be the best coupling agent for the glass fibers, possibly, because of its chemical compatibility with PA6. Tensile
test results indicated that increasing amount of PA6 in the polymer matrix improved the strength and stiffness of the composites due to a
strong acidbase interaction at the interface. Incorporation of PA6 to the SGF reinforced ABS reduced the melt viscosity, broadened the fiber
length distributions and increased the toughness of the composites. Fractographic analysis showed that the incorporation of PA6 enhanced
the interactions between glass fibers and the polymeric matrix.
q 2004 Elsevier Ltd. All rights reserved.
Keywords: Interface; ABS; Glass fiber reinforcement

1. Introduction
The incorporation of fibrous glass is known to improve
the properties of the thermoplastic materials by applying
traditional processes such as extrusion and injection
molding [14]. The mechanical performance of glass fiber
reinforced composites depends on not only the properties of
individual components but also the interfacial interactions
established between the reinforcing agent and the matrix
material [59].
An effective interface can be obtained if there exists
sufficient bonding between constituents. Silane coupling
agents, which are generally applied to the surface of
inorganic fillers, are well known and most widely used as
interfacial agents to bind glass fibers to polymeric matrices
* Corresponding author. Tel.: C90 312 210 2632; fax: C90 312 210
1264.
E-mail address: gbayram@metu.edu.tr (G. Bayram).
0032-3861/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2004.10.026

[10]; however, in some cases because of the lack of

functional groups of polymers that allow reactions at the
interface, silane coupling agents are incapable of binding
the fibers to polymers. One of the well-known solutions to
this problem is the introduction of new functional groups
externally to the polymer. It has been reported by many
researchers that to introduce polarity or acidity to a nonpolar matrix, great variety of acidic additives including
acrylic acid, methacrylic acid, maleic acid and their
anhydrides have been used to enhance the adhesion between
glass fibers and polymers [11]. The most commonly applied
process is the melt blending of reactive species, especially
through reactive extrusion [12].
Acrylonitrile-butadiene-styrene (ABS) terpolymer gains
industrial and scientific importance due to its toughness,
dimensional stability and good surface appearance. The
previous studies showed that incorporation of short glass
fibers (SGF) to neat ABS to balance the toughness and
stiffness results in an improvement in tensile strength and

8958

G. Ozkoc et al. / Polymer 45 (2004) 89578966

modulus, on the other hand, a decrease in toughness is

observed. In addition to that, microscopic studies showed
lack of adhesion between ABS and glass fibers [1,13,14].
The current study focuses on the effects of incorporation
of acidic groups by blending of ABS with polyamide 6
(PA6) for functionalization of the polymeric matrix to be
reinforced with SGFs; so that an improved interfacial
adhesion between fiber and polymer can be established. The
study also includes a comparison of neat-ABS and PA6
incorporated-ABS, which are reinforced with SGFs. At the
beginning, the adhesion capacities of the different coupling
agents at the fiber/matrix interface were experimentally
examined. The coupling agent, which exhibited the best
adhesion performance, was used for compounding SGFs
with the polymeric matrices in the study that follows.

2. Background
The bonding of glass fibers to polymer matrices has been
recognized since glass fibers were first used in composites
[10]. Organosilicone compounds (or silane coupling agents)
are generally used for this purpose since the silicone ends
are chemically similar to the structure of glass, and the
organic groups on the silicone are potentially capable of
interacting with polymer [15].
The reactions of silane coupling agent promoting the
interfacial adhesion are represented in Fig. 1 in which
trialkoxysilane hydrolyzes in aqueous media to form a
silanol compound, then, silanol reacts with the hydroxyl
group of the glass surface. The coupling reaction occurs
during processing between glass fibers functionalized by
organo-functional silane and polymer, if it is chemically
favorable. R is the organic functional group to interact
with the functionality of polymer *R.

For a better understanding of fiber/matrix interphase

mechanics, it is necessary to determine the interfacial
strength and possible interactions between polymer and
fiber. This is usually carried out by means of contact angle
measurements [9,16,17] to determine the thermodynamical
energy of adhesion and specific mechanical tests on real or
model composites to determine the strength of interphase
formed between fiber and polymer [18].
The method of estimating the thermodynamical adhesion
at fiber matrix interphase, which is generally called work of
adhesion, from the surface properties of both materials, is a
theoretical approach. This thermodynamical model of
adhesion is certainly the most widely used approach in
adhesion science. It considers that the adhesive will adhere
to the substrate because of interatomic and intermolecular
forces established at the interphase, provided that an
intimate contact between both materials is achieved [19].
The total surface energy, gTOT
, of a given non-metallic
i
material (i) can be considered as being composed of two
parts: the Liftshitz-van der Waals (LW), gLW
i , and the acid
base (AB) component, gAB
i [9,21]. This is written as the sum
of the two components,
Z gLW
C gAB
gTOT
i
i
i

(1)

where the acidbase term is a property based on the mutual

interaction of two unlike species, an acid and a base. gAB
is
i
composed of two surface parameters: gC
,
the
Lewis
acid
i
component and gK
i is the Lewis base component of the
surface free energy. These, together, yield the acidbase
component of surface free energy, gAB
i .
C K 1=2
gAB
i Z 2gi gi

(2)

The most characteristic feature of these Lewis acid and

base components is their non-additivity. That is to say, if

Fig. 1. Reactions of organosilanes (R and *R are the functionalities of silane coupling agent and polymer, respectively).

G. Ozkoc et al. / Polymer 45 (2004) 89578966

8959

F
bd

(7)

K
phase (i) possesses only gC
i or gi , this component does not
participate in the total surface free energy of the phase (i).
However, it will interact with the complementary component (j) of the contacting phase.
C
K
The values of gAB
i , gi and gi can be determined by
using the contact angle, q and Complete Young Equation
[9,20],

tILSS Z 0:75

where F is the rupture force under bending force, b is the

width and d is the thickness of the specimen.

3. Experimental

1 C cos qgTOT
i
LW 1=2
K 1=2
C 1=2
Z 2gLW
C gC
C gK
i gj
i gj
i gj

3.1. Materials
(3)

The LW component of a solid surface (i) can also be

found from the contact angle of a non-polar liquid (j), where
gTOT
Z gLW
j , on the solid surface. In this case, Eq. (3)
j
reduces to,
LW 1=2
1 C cos qgTOT
Z 2gLW
j
j gi

(4)

As a result, the LW component of a solid surface can be

calculated by determining the contact angle of a non-polar
liquid on the solid surface in Eq. (4).
For a bipolar liquid (L) contacting with the solid (S), with
surface tension gL, acidic and basic surface parameters gC
L
and gK
L , respectively, and non-polar surface component,
gLW
L , the corresponding complete equation is as follows,
1 C cos qL gTOT
L
LW 1=2
K 1=2
C 1=2
Z 2gLW
C gC
C gK
L gS
L gS
L gS

(5)

This is also written for a second bipolar liquid. A set of two

simultaneous equations are formed in terms of the
C
parameters of the solid, gK
S , gS and two advancing contact
angles q1 and q2, which are measured on the solid surface.
These two equations are then simultaneously solved for gK
S
LW
C
K
and gC
S provided that the gi , gi and gi for the probe
liquids are known [21].
Providing the known surface components of the contacting phases (i.e. polymer and fiber), the work of adhesion
(Wa) between phase 1 and phase 2 can be calculated from
the summation of dispersive and acid/base components by
using [20],
WaTOT Z WaLW C WaAB
LW 1=2
K 1=2
C 1=2
Z 2gLW
C gC
C gK
1 g2
1 g2
1 g2

(6)

The strength of interphase formed can be determined by

applying some specific mechanical tests on model or real
composites systems. Short Beam Flexural Test is a
standard testing method to evaluate the interlaminar shear
strength (ILSS) of fiber reinforced composites [22]. The
measurement is performed applying a three point bending
test on a short beam with the span-to-depth ratio chosen to
produce interlaminar shear failure. For a rectangular crosssection, the apparent interlaminar shear strength, tILSS, is
given by

Information on materials used in this study are

summarized in Table 1. ABS, PA6 and glass fibers were
obtained from Emas Plastik, Tekno Polimer, and Cam Elyaf
Glass Fiber companies, respectively. Silane coupling agents
used during surface modification of glass fibers were gaminopropyltrimethoxysilane (APS hereafter, Aldrich
Chemical Co.), g-(trimethoxysilyl)propylmethacrylate
(MPS hereafter, Aldrich Chemical Co.), and a blend of
styrylsilane and methacrylosilane, which had already been
applied to the glass fibers by Cam Elyaf company, the
manufacturer, (As received hereafter).
3.2. Experimental procedure
Before new coupling agents were applied onto glass fiber
surfaces, initial film formers and couplings had to be
removed. The removal was carried out in a furnace by
burning off them at approximately 600 8C for 30 min, then
by using a soxhlet apparatus glass fibers were washed with
acetone for 1 h to remove organic degradation deposits.
At first, solutions including 1 wt% coupling agent were
prepared with distilled water for hydrolysis. The pH of the
solution was adjusted to 4.5 with acetic acid to catalyze the
MPS reaction. For APS, to avoid the possible reaction
between amino group and acid, no catalyst was used. Since
the MPS is not water soluble, it was dissolved in a solution
of 10 wt% water in ethanol. After hydrolysis, continuous
glass fiber bundles were dipped into the solution and kept
for 1 h to allow formation of silanol bonds with continuous
stirring, then, the solution was removed and glass fibers
were washed with water and then with ethanol to remove the
physisorbed coupling agents. The fibers were then dried in
an oven for 10 min at 110 8C. To check the amount of
coupling agent bonded to the fiber surface, a small sample of
fiber was analyzed by using TGA (thermo gravimetric
analyzer). This measurement indicated that the weight
fraction of coupling agent on the surface was around 0.5%.
Contact angles were measured tensiometrically using an
electronic microbalance (Sartorious Microbalance, M25D)
equipped with a motor-driven stage that has vertical
displacement capability of 10 mm. The digital signals
from the microbalance were recorded. In the case of glass
fibers, a single fiber specimen was prepared first, by tapping
1 cm of a 2 cm length of fiber between two pieces of
adhesive tape, with about 1 cm of the fiber exposed. The

8960

G. Ozkoc et al. / Polymer 45 (2004) 89578966

Table 1
Information on materials used in this study
Material

Trade name
w

Manufacturer

Description
Extrusion grade Sp. gravity: 1.04
Textile yarn extrusion grade amino-end groups: 44G
5 meq/kg, Sp. gravity: 1.14
Diameter: 16.0 mm, coupling agent: blend of styrylsilane
and methacrylosilane
Diameter: 13.0 mm, coupling agent: aminosilane

ABS
PA6

Tairilac
Domamid 27w

Formosa, Taiwan
Domo, Germany

Continuous glass fibers

WR 3-1200w

Cam Elyaf, Turkey

Short glass fibers

PA-1w

Cam Elyaf, Turkey

specimen was then placed to the hook of the microbalance.

In all experiments a stage velocity of 1 mm/s was used to
bring the fiber into contact with the liquid. The force (F)
measured by the microbalance was then recorded to
calculate the advancing contact angle (q) by Eq. (8),
F Z Pgcos q

(8)

where P is the perimeter of the sample and g is the surface

tension of the probe liquid. Each contact angle measurement
was repeated at least three times. The contact angle was
calculated from the averages of the measurements. After
each measurement to specify the perimeter of sample,
0.5 cm of the fiber tip was cut; the remainder was then
brought into contact with n-decane, a completely wetting
liquid. It is assumed that n-decane makes zero angle with the
sample. Properties of these probe liquids are given in Table
2 [21]. Diiodomethane (DIM) was used as the probe liquid
for Liftshitz-van der Waals interactions. Ethyleneglycol
(EG) and formamide (FA) were used to identify the acid
base interactions between sample and the probe liquid.
For the polymeric samples, the same procedure was
repeated with a tiny, very thin strip-shaped sample with
dimensions of 2!20!0.2 mm. The samples were prepared
with a hot-press at 230 8C and 100 bars.
ILSS was determined according to ASTM D 2344 [22]
using the short beam flexural test. Bundles of glass fibers
treated with different silane coupling agents were embedded
into the polymer matrix using a hot-press, which is
illustrated in Fig. 2. The short beams including unidirectional parallel glass fibers at the central position were
then cut from the sheets obtained and tested by using a
Lloyd 30 K Universal Testing Machine. The support span to
depth ratio used was 5:1 with a corresponding specimen
length to depth of 7:1, and crosshead speed was 2.0 mm/
min. The force versus displacement curve for each sample
was recorded.

Before blending, ABS and PA6 pellets were dried in a

vacuum oven at 80 8C for 4 and 12 h, respectively. The 10,
20 and 30 wt% PA6 containing ABS/PA6 batches were
processed in a co-rotating twin-screw extruder (Thermoprism TSE 16 TC, L/DZ24) at a screw speed of 200 rpm
and barrel temperature profile of 195230230235
240 8C. The extrudate was water cooled and chopped into
small pellets. The produced ABS/PA6 pellets were again
vacuum dried again at 80 8C for 12 h. The 30 wt% glass
fiber reinforced ABS/PA6 composites were prepared by
extrusion at the same operating conditions applied previously in the ABS/PA6 blending. The extrudate was again
water cooled and chopped into small pellets. The specimens
for mechanical characterization experiments were molded
by using a laboratory scale plunger type injection-molding
machine (Microinjector, Daca Instruments) at a barrel
temperature of 230 8C and mold temperature of 80 8C.
All mechanical tests were carried out at room temperature. At least five samples were tested and average results
with standard deviations were reported for each type of
composite. Tensile tests were performed on dog-bone
specimens according to ASTM D638 by using a Lloyd
30 K Universal Testing Machine. Charpy impact tests were
performed by using a Pendulum Impact Tester of Coesfeld
Material Test, according to the ASTM D256 on rectangular
specimens (6!60!2 mm).
Melt flow index values were measured by using an
extrusion plastometer (Omega Lab. Ins. Lmt.) according to
ASTM D 1238 with 5 kg load and at 230 8C.
A low voltage SEM (JEOL JSM-6400) was used to
analyze the tensile fractured surfaces of the composites. The
specimens were coated with gold to eliminate arcing of the
beam. To obtain the fiber length distributions after
processing, approximately 5 g of random samples from
extruded pellets were placed in an oven at 600 8C for 30 min
to remove the polymeric matrix. The remaining glass fibers

Table 2
Surface tension components of probe liquids, (mN/m)
Surface tension (mN/m)

DIM

EG

FA

n-Decane

gTOT
L
gLW
L
gAB
L
gK
L
gC
L

50.8
50.8

58
39
19
29
2.3

48
29
19
39.6
2.3

23.83
23.83

G. Ozkoc et al. / Polymer 45 (2004) 89578966

8961

Fig. 2. The symbolic diagram of preparation of polymer/glass fiber

sheet for ILSS measurements (REC: as received).

were dispersed in distilled water and transferred to glass

slides. The slides were then placed on an optical microscope
(Prior Laboratory Microscope Model B 3000) and photographs of glass fibers were taken. For each sample, a
population of about 400 fibers were scaled and counted to
obtain fiber length distributions.

4. Results and discussion

4.1. Adhesion measurements
The WaTOT is the sum of the surface chemical affinity of
the components in the composite, which can be through
H-bonding or other secondary attractions; and in addition to
these interactions, the ILSS values contains the term related
to mechanical interlocking. In this study, the WaTOT and
ILSS values were calculated from experimentally determined interfacial tensions and short beam flexural tests,
respectively.
The surface free energy components of the glass fibers
with different surfaces and polymers calculated from
measured contact angles using the probe liquids are shown
in Table 3. Work of adhesion between contacting phases can
be calculated from Eq. (6) by utilizing the contact angles
and surface free energy components. The thermodynamical
work of adhesion, WaTOT , values including dispersive, WaLW ,
and acid/base, WaAB , components between matrix materials
and various glass fiber surface coatings are given in
Figs. 3 and 4.

Fig. 3. Work of adhesion between ABS and glass fibers with

different surface treatments.

It is observed from Fig. 3 that dispersive component

WaLW of WaTOT remained almost constant for all of the
species when ABS was concerned. As expected, APS, MPS
and as received silane treated fibers have basic surface
characteristics due to electron rich groups in their structures.
Thus, the acid/base interaction is not possible between
electron donor groups, phenyl, nitrile and vinylidene, of
ABS. As is seen from Fig. 3, the WaAB between unsized fiber
and ABS is higher than the others due to the presence of
acidic OH groups on the surface of the glass, which can
interact with electron donor groups of ABS.
Fig. 4 represents the thermodynamical adhesion between
PA6 and glass fibers with different surface coatings. Similar
dispersive components as in ABS are observed in PA6,
therefore, it can be said that the variation of thermodynamical adhesion values arises from the difference in acid/
base terms. Especially in APS, amines may interact with
both the amide groups on the backbone and carboxylic acid
groups at the end of the PA6 chains [5], hence, it is
reasonable to achieve higher values of WaTOT for APS.

Table 3
Surface free energy components, (g, mJ/m2)
Species
As received fibers
APS treated fibers
MPS treated fibers
Unsized fibers
ABS
PA6

Surface free energy components (mJ/m2)

gLW

gC

gK

32.55
34.58
33.95
32.90
28.96
37.14

0.02
0.03
0.09
4.38
0.42
4.11

5.19
7.31
0.38
0.82
4.96
1.07

Fig. 4. Work of adhesion between PA6 and glass fibers with

different surface treatments.

8962

G. Ozkoc et al. / Polymer 45 (2004) 89578966

Fig. 5. ILSS values between ABS and glass fibers with different
surface treatments.

The ILSS values calculated from short beam flexural

tests are given in Figs. 5 and 6. In the case of ABS (Fig. 5),
unlike the highest WaTOT observed for unsized fiber, its ILSS
value is the lowest with respect to the others. A possible
reason for this may be lack of mechanical interlocking
between the fiber surface and molten-ABS during specimen
preparation. When the coupling agents are present on the
surface, the variation of ILSS values may be attributed to
pure mechanical interlocking in the absence of acid/base
interaction. The presence of the coupling agents on the
surface may allow the chains to penetrate inside through the
sizing layer, hence, the highest ILSS was obtained with APS
which is less bulky compared to MPS and silane treated
glass fibers used as received; and may allow inter-chain
diffusion to a larger extent.
The ILSS values for the PA6 and fibers are harmonious
with thermodynamical adhesion values because of the
synergetic effect of both acid/base interactions and mechanical interlocking (see Fig. 6). The measured adhesion
strength value for APS coated fibers is about two times
larger than the others.
The dependency between ILSS and WaTOT shown in Fig.
7 was also investigated to check the consistency of these

Fig. 6. ILSS values between PA6 and glass fibers with different
surface treatments.

Fig. 7. The interlaminar shear strength versus work of adhesion.

measurements. For the system involving poorly polar

matrices or non-polar matrices such as PE, PVC, PU and
surface treated glass fibers; it has been shown that a linear
relationship can be established between ILSS and WaTOT
[18]. Thus, it is possible to write the following relation:
ILSS Z kWaTOT

(9)

where k is the coefficient of proportionality. This equation

indicates that physicochemical interactions between the
fiber and the matrix determine, to a large extent, the
mechanical behavior of the interphase from the relationship
given by Cox [23]. The constant k depends on the modulus
of matrix and fiber, and the equilibrium center-to-center
distance involved in physical interactions such as van der
Waals and acid/base interactions [24]. Fig. 7 shows the
dependence of WaTOT on ILSS measured between glass
fibers with different surface coatings and polymer matrices
(ABS and PA6). While trend lines were being added to the
plots of ILSS versus WaTOT in Fig. 7, the unsized fibers were

Fig. 8. Representative stressstrain curves of materials produced.

G. Ozkoc et al. / Polymer 45 (2004) 89578966

8963

Table 4
Tensile test results of materials produced
Material

Tensile strength (MPa)

Youngs modulus (MPa)

Strain at break (%)

Neat ABS
30% SGF reinforced ABS
30% SGF reinforced (90% ABSC10% PA6) blend
30% SGF reinforced (80% ABSC20% PA6) blend
30% SGF reinforced (70% ABSC30% PA6) blend

44.0G6.0
78.0G2.5
85.0G3.0
91.0G1.1
93.0G1.8

2240G183
6500G352
6910G205
8045G512
8415G333

7.5G0.8
2.5G1.3
3.2G0.6
3.4G0.9
3.2G0.9

not included because of the lack of mechanical interactions.

It is observed that there is an approximate linear dependence
between WaTOT and ILSS for ABS/GF system when the fiber
surfaces are treated with silane coupling agents. Similar
results were obtained experimentally by Park et al. [16] for
systems including continuous glass fibers treated with
various silane coupling agents and unsaturated polyester;
and by Schultz and Lavielle [25] for various carbon fiber
reinforced epoxy resins [25]. These studies indicate that
good wettability of matrix, which is identified by high WaTOT
between fiber and matrix, results in high interfacial strength
identified by ILSS.
The significant result obtained from the plot (Fig. 7) is
the divergence from the linear dependence between ILSS
and WaTOT values (R2Z0.948 and 0.703) when the polarity
of the matrix increases from ABS to PA6, which gives rise
to strong interactions between contacting phases, between
PA6 and glass fibers. Nardin and Schultz [18] stated that
linearity between WaTOT and ILSS is no longer valid, when
strong specific interactions are established at the interface.
The adhesion measurements indicate that acid/base
interaction at the interface is the key factor to obtain the
desired interfacial adhesion. As a result of adhesion
measurements, it was decided to use the APS treated SGF
to reinforce the ABS and ABS/PA6 blends due to its highest

acidbase interaction capacity with PA6 and its highest

ILSS compared to other coupling agents studied.
4.2. Stressstrain behavior
The representative tensile stressstrain curves for
30 wt% SGF reinforced ABS blended with various amounts
of PA6 are shown in Fig. 8. The variation of the response of
the composites under tensile load gives useful information
about the changes of the microstructure at the fiber/matrix
interface. It was reported that a strong interfacial interaction
enhances efficient load transfer from matrix to fibers in fiber
reinforced systems [2628]. The ultimate strength of
composites is a direct indicator of the strength of the
interfacial bonds since the applied stress is more efficiently
transferred through the interface [29]. The results of tensile
tests tabulated in Table 4 exhibit an increasing trend in
ultimate strength of composites as the PA6 content in the
matrix increases. This result can be attributed to the
increasing extent of acidbase reaction between acidic end
group of PA6 chains and aminopropyl functional group of
coupling agent on the glass fiber surface as the amount of
PA6 in the matrix increases. Moreover, these interactions
might stiffen the polymer chains close to the fibers to resist
against the applied tension leading to an improved Youngs
modulus with increasing PA6 content (Table 4).
The fiber length distributions of composites after
extrusion are illustrated in Fig. 9. Because of the nature of
the injection molding machine used for preparing specimens, it can be assumed that all of the fiber length reduction
occurred during extrusion and the distribution curves are
representative of the specimens tested. It is observed that the
number of long fibers in the composites increases as the PA6
content increases in the matrix from 0 to 30%, although the
corresponding average fiber length remains nearly the same
(427, 447, 471, 482 mm). The melt viscosity of PA6 is lower
Table 5
Melt flow index values of materials produced with 5 kg load and 230 8C
Material

Fig. 9. Fiber length distributions as a function of PA6 concentration

in the matrix for 30 wt% SGF reinforced ABS/PA6 composites.

Neat ABS
30% SGF reinforced ABS
30% SGF reinforced (90% ABSC10% PA6) blend
30% SGF reinforced (80% ABSC20% PA6) blend
30% SGF reinforced (70% ABSC30% PA6) blend
Neat PA6

Melt flow index

(g/10 min)
15.6
3.2
11.5
12.2
13.7
36.4

8964

G. Ozkoc et al. / Polymer 45 (2004) 89578966

Fig. 10. Unnotched impact strength of materials produced.

than the viscosity of ABS, thus the shear forces are lower in
the case of ABS/PA6 blends with high PA6 content which
reduces fiber length degradation [30]. This can be followed
from the MFI values given in Table 5 which shows that MFI
values increase as the PA6 content increases. It is known
that the load bearing capacity of fibers embedded into a
matrix increases as their length gets longer; therefore
improved tensile strength and Youngs modulus must have
resulted from increasing number of longer fibers together
with the improved adhesion [3,31,32].
4.3. Impact strength
Fig. 10 illustrates the variation of unnotched charpy
impact strength of materials produced. The neat ABS resin
has the highest impact strength as expected. The incorporation of SGF to ABS results in a sharp reduction in impact
strength because of the restricted deformation ability of
matrix in the presence of fibers. When 10 wt% of PA6 is
incorporated to the matrix, a drastic improvement in the
impact strength of composites is obtained. Further addition
of PA6 decreases the impact strength value but it is still
higher than the SGF/ABS system. This behavior may be
attributed to the high chemical affinity between the amino
functional coupling agent on the fiber surface and PA6 to
decrease the total free energy of the system. The PA6 chains
fed can concentrate near the surface of the glass fibers
during melt processing and there might be a PA6 sheath
formed on the fibers; therefore the ABS matrix will be free
to deform similarly to the neat-ABS. This theory is
supported by the SEM analysis that follows.
4.4. Fractographic analysis
In order to investigate the microstructure and extent of

the adhesion between fibers and matrix, SEM study of

tensile fracture surfaces of 30 wt% SGF containing specimens is performed (see Fig. 11). In the case of 30 wt% SGF
reinforced ABS, deformation of matrix is restricted and a
brittle fracture is observed. Also it can be seen that the fibers
are debonded, leaving a dark ring occurring at the interface,
which also observed in the study of Fu and Lauke [31]. In
that study, the authors interpreted the dark rings resulting
from local deformation of the matrix around the fibers after
fiber debonding. Most of the fibers were pulled-out from the
matrix during deformation; however, the matrix is still
attached to the fibers. Failure occurs at the fiber matrix
interface. In close-up view (see Fig. 11(a)), the fiber surface
is clean and smooth, which supports the work of adhesion
results indicating that there is lack of interaction between
APS treated SGFs and ABS.
Fig. 11(b)(d) show the SEM micrographs of fracture
surfaces of PA6 containing composites reinforced with
30 wt% SGF. In the case of strong interphase, the physical
properties (especially density) of matrix near the fiber
surface are different than that of the bulk polymer; therefore
the failure under tensile loading occurs at the matrix near the
fibers. It is clear from these figures that the failure occurred
at the matrix. The sheath of polymer observed on the surface
of the fibers can be interpreted as an evidence of strong
interactions.

5. Conclusions
The effects of incorporation of acidic groups by blending
ABS with PA6 to obtain improved interfacial adhesion
between glass fibers and polymer were studied by observing
the tensile and impact behavior of 30 wt% SGF reinforced
composites with ABS/PA6 matrix. The adhesion capacities

G. Ozkoc et al. / Polymer 45 (2004) 89578966

8965

Fig. 11. SEM micrographs (!1000) of tensile fractured surfaces of (a) 30 wt% SGF reinforced ABS, (b) 30 wt% SGF reinforced (90%
ABSC10% PA6), (c) 30 wt% SGF reinforced (80% ABSC20% PA6) and (d) 30 wt% SGF reinforced (70% ABSC30% PA6) composites.

of the different coupling agents at the fiber/matrix interface

were experimentally examined. Comparison of thermodynamical adhesion values with the experimental ILSS values
indicates the importance of inter-chain diffusion and
mechanical interlocking of polymer chains at the interfacial
region between the fiber and the matrix. Among the silane
coupling agents, APS functions best as manifested by high
ILSS and thermodynamical adhesion values with PA6; so
that, it was selected as the coupling agent for glass fibers to
reinforce blends of ABS and PA6.
The tensile test results indicate that incorporation of PA6
to the ABS improves the stiffness and strength of 30 wt%
SGF containing composites because of enhanced interfacial
adhesion. It is concluded from the impact tests that
incorporation of SGF to the ABS restricts the deformability
of polymeric matrix resulting in brittle fracture. The PA6
containing ABS/SGF composites tend to have higher impact
strength than the one which does not contain PA6. It is seen
from the SEM micrographs of fracture surfaces of ABS/SGF
composites that adhesion between ABS and APS sized glass
fiber is poor and fibers are pulled-out of the matrix during
tensile loading. On the other hand, the SEM micrographs of
PA6 containing ABS reinforced with 30 wt% SGF show

that the fibers are coated with a sheath of polymer, which

may be due to the strong interaction between the fibers and
the matrix.

Acknowledgements
The authors would like to thank Emas Plastik, Tekno
Polimer and Cam Elyaf Glass Fiber companies for
supplying the materials used in this study.

References
[1] Ylmazer U. Compos Sci Technol 1992;44:11925.
[2] Fu YS, Lauke B, Mader E, Yue CY, Hu X. Composites Part A 2000;
31:111725.
[3] Thomason JL, Vlug MA. Composites Part A 1996;27:44784.
[4] Tjong SC, Xu SA, Li RKY, Mai YW. Compos Sci Technol 2002;62:
201727.
[5] Laura DM, Keskkula H, Barlow JW, Paul DR. Polymer 2002;43:
467387.
[6] Frenzel H, Bunzel U, Hassler R, Pompe G. J Adhes Sci Technol 2000;
14(5):65160.

8966

G. Ozkoc et al. / Polymer 45 (2004) 89578966

[7] Gonzalez VA, Uc JMC, Olayo R, Franco PJH. Composites Part B

1999;30:30920.
[8] Bikiaris D, Matzinos P, Prinos J, Flaris V, Larena A, Panayiotou C.
J Appl Polym Sci 2001;80:287788.
[9] Akbay AR, Bayraml E. J Adhes 1995;50:15564.
[10] Plueddeman EP. Silane coupling agents, 2nd ed. New York: Plennum
Press; 1991.
[11] Pak HS, Caze C. J Appl Polym Sci 1997;65:14353.
[12] Xanthos M, editor. Reactive extrusion principles and practice. 1st ed.
New York: Carl Hanser Verlag; 1992.
[13] Fu YS, Lauke B. Composites Part A 1998;29:63141.
[14] Fu YS, Lauke B. Composites Part A 1998;29:57583.
[15] Park SJ, Jin JS, Lee RJ. J Adhes Sci Technol 2000;14(13):167789.
[16] Park SJ, Jin JS. J Polym Sci Part B 2003;41:5562.
[17] Pisanova E, Mader E. J Adhes Sci Technol 2000;14(3):41536.
[18] Nardin M, Schultz J. In: Akoval G, editor. Interfacial interactions in
polymeric composites. Nato ASI series. Netherlands: Kluwer
Academic Publisher; 1992.
[19] Sharpe LH. In: Akoval G, editor. Interfacial interactions in polymeric
composites. Nato ASI series. Netherlands: Kluwer Academic
Publisher; 1992.

[20] van Oss JC, Good JR, Chaudhury MK. Langmuir 1988;4:88494.
[21] Akoval G, Torun TT, Bayraml E, Erinc NK. Polymer 1998;39:
13638.
[22] Annual book of ASTM standards, vol. 08.01; 1993. Plastics (1), D
2344 (standard test method for inter laminar shear strength by short
beam flexural test).
[23] Cox HL. Br J Appl Phys 1952;3:7284.
[24] Israelachvili JN. Intermolecular and surface forces. London: Academic Press; 1985.
[25] Schultz J, Lavielle L. Science and new applications of carbon fibers.
Proceedings of the international symposium in University of
Toyohashi, Japan; 1987.
[26] Hamada H, Ikuta N, Nishida N, Maekawa Z. Composites 1994;25:512.
[27] Dilsiz N, Erinc NK, Bayraml E, Akoval G. Carbon 1995;33(6):
8538.
[28] Zafeiropoulos NE, Baillie CA, Hodgkjnson JM. Composites Part A
2002;33:118590.
[29] Yue CY, Cheung WL. J Mater Sci 1992;27(12):318191.
[30] Turkovich R, Erwin L. Polym Eng Sci 1983;23(13):7439.
[31] Fu YS, Lauke B. Compos Sci Technol 1996;56:117990.
[32] Thomason JL. Compos Sci Technol 1999;59:231528.