Article

Journal of Reinforced Plastics

and Composites

Fiber length, thermal, mechanical, and 30(6) 488–498

! The Author(s) 2011
Reprints and permissions:
dynamic mechanical properties of sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0731684410397898
injection-molded glass-fiber/polyamide jrp.sagepub.com

6,6: plasticization effect

Aziz Hassan, Noordini M. Salleh, Rosiyah Yahya

and M.R.K. Sheikh

Abstract
Unreinforced- and glass fiber-reinforced-polyamide 6,6 specimens were subjected to dry, 50% RH, and wet conditions.
FLD study showed that more fiber degradation occurred during processing of the composites with higher fiber loading.
TGA results revealed that about 2% and 5% levels of moisture have been absorbed by the specimens subjected to 50%
RH and wet conditions, respectively. From DSC results, with moisture uptake, there were no significance changes in the
melting temperature, crystallization temperature, and enthalpy of crystalline; however, degree of crystallinity and
enthalpy of melting decreased. Tensile strength, tensile modulus, flexural strength, and flexural modulus decreased
with moisture uptake, while tensile strain and flexural displacement increased. DMA study showed that tan delta at
room temperature and tan delta maxima increased with moisture absorption, while temperatures at maximum tan delta
in a- and b-transition regions decreased.

Keywords
glass fiber composites, plasticization effect, thermal properties, mechanical properties, dynamic mechanical properties

Introduction
preferentially penetrate the amorphous regions, readily
Short fiber-reinforced thermoplastic composites con- breaking the hydrogen bonds to the amide group and
taining polyamide 6,6 (PA, polyamide) and E-glass increasing molecular chain mobility. As a result, the
fibers have been extensively used in many engineering absorbed water lowers the tensile modulus, the work
applications owing to their outstanding mechanical to elongate and to recovery, and the glass-transition
properties. PA 6,6 is relatively easy to process while temperature, Tg. The crystallinity of PA also increases
E-glass fibers are commonly used as reinforcement as it absorbs water due to the break up of intermole-
due to low cost, high tensile strength, high chemical cular hydrogen bonding and thus allows the hydrocar-
resistance, high dimensional stability, and excellent bon chain segments to pack better.
insulation properties.1,2 However, PA 6,6 is a hygro- In this article, the effects of moisture absorption
scopic material. The water absorption of PA 6,6 can (conditioning) on thermogravimetric analysis (TGA),
reach a maximum up to 2.5% w/w. This is related to
the polar amide groups around which water molecules
can become coordinated.3 Department of Chemistry, Polymer and Composite Materials Research
Moisture affinity significantly influences the physical Laboratory, University of Malaya, Malaysia.
and mechanical properties of both the PA and fiber.
The absorption of water causes the fiber to swell, mod- Corresponding author:
Aziz Hassan, Department of Chemistry, Polymer and Composite
ifying its dimensions and in turn changes the size, Materials Research Laboratory, University of Malaya, 50603 Kuala
shape, stiffness, and permeability of final products. Lumpur, Malaysia
Water is also a powerful plasticizer for PA and will Email: [email protected]

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

Hassan et al. 489

differential scanning calorimetry (DSC), tensile, flex- fibers were counted and divided according to the length
ural, and dynamic mechanical properties of glass range, at every 0.10 mm interval. A mid-point of fiber
fiber-reinforced PA 6,6 with different fiber volume frac- length range was taken as the average fiber length.
tions, Vf, were investigated.
Thermogravimetric analysis
Experimental TGA was used to investigate the thermal decomposi-
tion behavior of materials using a TGA 6
Materials and sample preparation Thermogravimetric Analyser (Perkin Elmer). Tests
For the characterization, commercially compounded were done under nitrogen at a scan rate of 20 C/min
materials were used, i.e., short glass fiber-reinforced within a temperature range of 50–900 C. A sample of
PA 6,6 composite, TechnylÕ A216 V30 with fiber 5–10 mg was used.
weight fraction, Wf of 30% (volume fraction, Vf of
16%). To prepare specimens with lower fiber loading,
composite TechnylÕ A216 V30 was diluted with unrein-
Differential scanning calorimetry
forced PA 6,6 matrix, TechnylÕ A216. DSC experiments were performed with a Diamond
Injection molding machine, BoyÕ 55M was DSC (Perkin Elmer). A test sample of 5–10 mg was
employed for the specimen preparation. For diluted used for the characterization. Each sample was heated
composites, the mixtures were physically mixed and (first heating) from 0 C to 290 C, then cooled to 0 C,
shaken before feeding into the feed hopper of the injec- and heated again (second heating) to 290 C for each
tion molding machine. A single-gated double cavity run. Results shown are from the second heating.
(one impact and one tensile standard test bar) mold
was used in the molding. The dimensions of the dumb- Tensile testing and fracture surface morphology
bell-shaped tensile test pieces were in accordance with
the ASTM Standard D638-80, type 1.4
characterization
The tensile tests were performed using an Instron 5569
Universal Testing Machine with a constant crosshead
Specimen conditioning
speed of 1.0 mm/min at room temperature of about
Specimens under dry condition refer to the specimens 23 C. ASTM standard D638-80 was used as a stan-
as they were molded. The specimens were kept in vac- dard4 in calculating the tensile properties.
uumed desiccators with silica gel immediately after the After testing, the fractured surface was cut from the
molding. For the preparation of specimens conditioned selected test specimen. A layer of gold was sputter-
under 50% relative humidity (RH), injection-molded coated to improve the sample conductivity and avoid
composites were exposed to saturated sodium hydrogen electrical charging during scanning electron microscopy
sulfate (NaHSO4) solution environment for at least 1 (SEM) examination. Fractured surfaces were examined
month before testing.5 Hot water bath was used to pre- using JEOL Field Emission Auger Microprobe, JAMP-
pare the wet condition of the specimens. The specimens 9500F under SEM mode.
were boiled for 24 h to ensure that the specimens were
completely wet and there was no further increase in
weight.
Flexural testing
Three-point bending flexural tests were performed
using an Instron 5569 Universal Testing Machine at
Determination of fiber volume fraction (Vf) room temperature of about 23 C at a constant cross-
Fiber volume fraction of the composite specimens was head speed of 1.28 mm/min. Specimen support span, L,
determined by ashing. Samples from the central portion was fixed at 50 mm. ASTM D790 was used as a stan-
of injection-molded test pieces were cut and the poly- dard6 to calculate the flexural properties.
mer matrix was removed by heating the specimen in a
muffle furnace at a temperature of up to 700 C for
6–7 h.
Dynamic mechanical analysis
This analysis, shortly DMA, was carried out to inves-
tigate the effect of temperature on viscoelastic proper-
Determination of fiber length distribution
ties.7 Dynamic mechanical analyser, Q800 (Thermal
For the determination of fiber length distribution Analysis Instruments), was used for the characteriza-
(FLD), lengths of the extracted fibers were measured tion. The specimens were taken from the middle section
using image analyser, model Leco IA-32. At least 500 of the molded dumbbell bar, with the dimension of

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

490 Journal of Reinforced Plastics and Composites 30(6)

60 mm length  13 mm width  3.3 mm thickness

40
(height) and subjected to three-point bending mode
with a span length of 35 mm. Measurements were con-
: Vf = 8%
ducted over the temperature range of 100–200 C with 30
: Vf = 16%
a heating rate of 3 C/min, under a constant frequency

Fiber count (%)

of 1.0 Hz.
20

Results and discussion 10

Fiber volume fraction (Vf)

Vf of composites is calculated using the following 0
0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 0.95 1.05
equation: Fibre length (mm)

Mf Figure 1. FLD of injection-molded (tensile specimen)

f composites.
Vf ¼ ð1Þ
Mf Mm
þ
f m
volume fraction. Jones et al.10 also reported that fiber
where M and  are weight and density, respectively, and degradation is due to an increase in fiber volume frac-
subscripts f and m are referred to the fiber and matrix, tion of composites.
respectively. FLD data were also calculated for the number aver-
In order to reduce Vf, physical blending was used in age fiber length, Ln , that gives a measure of fiber ends
mixing the compounded materials with PA 6,6 matrix. density and weight average fiber length, Lw , that gives a
As reported by Hassan et al.,8 a precaution should be greater emphasis to the proportion of long fibers in the
taken in preparing the materials before feeding them distribution. These terms are given by Equations (2)
into injection molding machine. At the intended fiber and (3), respectively:
loading, the blends were prepared in batches with a P
total weight of 500 g each. If a large amount of blend ðn L Þ
Ln ¼ Pi i ð2Þ
is used, the compounded materials tend to settle down ni
in the feed hopper of the injection molding machines,
P  2
leaving the polymer matrix at the top. Therefore, the Vf ni L
of molded specimen is not consistent. In this study, the Lw ¼ P i ð3Þ
ni Li
experimental values of fiber volume fractions are the
same as the intended values. where ni is the number of fiber count (frequency) of
fibers of species i with length Li. A mid-point of fiber
length range was taken as an average value of fiber
Fiber length distribution
length, Li, in the calculation. Ln values of the compos-
Histograms of FLD of composites taken from the ten- ites are found to be 0.31 and 0.28 for composites with
sile test specimen are shown in Figure 1. From this Vf of 8% and 16%, respectively, whereas Lw values are
figure, it can be seen that at both fiber loadings of 0.39 and 0.35, respectively. These results further rein-
8% and 16%, fiber distribution is quite normal with a force the earlier statement made on the more fiber deg-
long tail toward the longer fiber length. The histograms radation occurrence with an increase in fiber
also show that more fiber population with shorter loading.11,12
length has occurred in composites with higher Vf
(16%). This means that composites with lower Vf
(8%) have more fiber population with longer length,
Thermogravimetric analysis
indicating that more fiber degradation has taken place TGA was used to study the thermal stability of the
during compounding and molding, probably due to materials. Figure 2 shows the TGA thermograms of
fiber–fiber, fiber–matrix, and fiber–machinery surface 8% Vf composites under dry, 50% RH, and wet con-
friction. The same trend of behavior was also reported ditions. Normally, for hygroscopic materials, gradual
by Denault et al.9 In explaining this, the authors quoted weight loss within the temperature range of 50–200 C
that high melt viscosity and increased tendency for high corresponds to the evaporation of water/moisture. In
fiber contact are probably responsible for the higher the dry samples, there is no indication of existence of
fiber breakage in composites with a higher fiber moisture within this temperature range. For specimen

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

Hassan et al. 491

conditioned under 50% RH, the samples show some crystalline (
Hc ). The degree of crystallinity (Xc ) of
low-molecular weight species burning off below 200 C the composites can also be calculated after the melt-
with the amount of about 2%, which may likely be due crystallization process using a reference heat of fusion
to the evaporation of moisture. For wet samples, 4–5% for the fully crystalline PA 6,6 that equals to 197 J/g
weight loss is recorded, indicating the maximum mois- (
Hm ).13,14 As there is no indication of the existence of
ture absorption of samples subjected to 24 h in boiling any major transition below 200 C, at least for the inter-
water. Weight loss due to the decomposition of PA 6,6 est of this research, thermograms shown were only for
matrix starts at about 510 C and is the same for all the temperature range of between 200 C and 290 C.
specimens under different conditions. These are given in Figure 3. Thermal properties
extracted from these analyses are tabulated in
Table 1. PAs are able to crystallize because of strong
Differential scanning calorimetry
intermolecular hydrogen bonds through the amide
DSC analyses were carried out to determine the groups,15 and because of van der Waals forces between
melting temperature (Tm ), enthalpy of melting (
Hm ), the methylene chains. The degree of crystallinity of PA
crystallization temperature (Tc ), and enthalpy of 6,6 decreases from dry to wet conditions, whereby the

Table 1. DSC thermal properties of glass fiber composite at

various conditions

Sample Tm
Hm Tc 
Hc Xc

(Vf, condition) ( C) (J/g) ( C) (J/g) (%)

0%, Dry 260.1 72.03 233.7 44.52 36.6

0%, 50% RH 262.5 71.40 235.7 45.68 36.2
0%, Wet 262.9 66.14 236.0 43.11 33.6
8%, Dry 260.6 59.96 234.1 34.94 30.4
8%, 50% RH 260.8 58.07 233.8 37.54 29.4
8%, Wet 260.9 55.88 234.3 34.83 28.4
16%, Dry 262.4 50.30 236.1 30.84 25.5
16%, 50% RH 262.3 48.88 236.2 30.39 24.8
Figure 2. TGA thermographs of glass fiber composite (8% Vf) 16%, Wet 262.5 47.05 236.2 30.49 23.9
subjected under different conditions.

Figure 3. DSC thermogram of glass fiber composite (8% Vf) subjected under different conditions.

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

492 Journal of Reinforced Plastics and Composites 30(6)

moisture weaken the hydrogen bonds and affects the absorption of some heat energy by the glass fiber
polymer internal molecular arrangement. during melting the composites. When samples were sub-
However, when glass fibers were added to the PA 6,6 jected to 50% RH and wet conditions, the decrements in
for 8% and 16% Vf, the existence of moisture does not
Hm were also observed. This phenomenon may be
affect the degree of crystallinity as much as for the pure attributed to the changes of crystal structure of polymer
matrix. Comparing with the dry specimens, Xc of wet matrix or variation of crystal thickness and size.
specimens decreased by 8.2%, 6.6%, and 6.3% for spe- The crystalline peak temperatures, Tc, of dry unrein-
cimens with Vf values of 0%, 8%, and 16%, respec- forced PA 6,6 is 233.7 C and the addition of glass fiber
tively. The values of melting temperature, Tm of 8% (Vf of 16%) increases the Tc to 236.1 C. Glass fibers
and 16% Vf, are essentially the same as the unrein- may act as nucleating agents that can increase the crys-
forced PA 6,6 under all conditions. This suggests that tallization temperature of the matrix. When the samples
there is no significant change in the microstructure of were exposed to the 50% RH and wet environment, no
the matrix with the addition of glass fiber and when significant changes are observed in Tc.
samples were subjected under 50% RH and wet condi- The crystalline enthalpy,
Hc of dry unreinforced PA
tions. A similar observation has been reported for short 6,6 is 44.52 J/g. As glass fiber is added, the magnitude
glass fiber-reinforced poly(butylene terephthalate) of
Hc decreases and it is believed that this is related to
composites.16 the changes in crystal structure, from triclinic to hexag-
The
Hm of unreinforced PA 6,6 is 72.03 J/g, while onal form since the addition of glass fiber will interfere
the addition of glass fiber decreases the
Hm to with crystallization. The crystallization occurs because
59.96 J/g and 50.30 J/g for composites with 8% and the glass fiber plays a role as a nucleating agent.17,18 As
16% Vf, respectively. This could be due to the for the samples subjected to 50% RH and wet condi-
tions, no significant changes are observed in
Hc .

Tensile properties
180 Tensile strength 4
Histogram of tensile strengths and tensile moduli of
Tensile modulus (GPa)
Tensile strength (MPa)

Tensile modulus
150 composites is shown in Figure 4. For comparison, the
3 values of property index (PI) of these properties are
120 calculated using the following general equation:

90 2  
Pc
Vf
60 Property index ¼  r  ð4Þ
1 Pc
30 Vrf
0 0 where Prc and Vrf are the respective property and fiber
0%, 50%RH

8%, 50%RH

16%, 50%RH
0%, Dry

8%, Dry

16%, Dry
0%, Wet

8%, Wet

16%, Wet

volume fraction of a reference composite, and Pc and Vf

are the corresponding property and fiber volume frac-
tion of the composite from which a comparison is to be
made.
Vf , Condition The values of PI are given in Table 2. This result
suggests that an increase in moisture absorption
Figure 4. The tensile strength and tensile modulus of com- reduces the composites strength and modulus of all
posites subjected under different conditions. specimens. The tensile strengths for unreinforced PA

Table 2. PI of tensile properties of composites under different conditions

Vf 0% Vf 8% Vf 16%

Property Dry 50% RH Wet Dry 50% RH Wet Dry 50% RH Wet

Strength Reference 0.94 0.66 Ref 0.61 0.48 Reference 0.66 0.53
Modulus Reference 0.86 0.53 Ref 0.52 0.30 Reference 0.74 0.61
Strain Reference 1.51 1.25 Ref 1.99 1.82 Reference 1.21 1.18

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

Hassan et al. 493

6,6 under dry, 50% RH and wet conditions are 53, 50,
and 35 MPa, respectively; we get 6% and 34% decre-
ments for 50% RH and wet specimens, respectively,
compared to the dry specimen. Tensile strength of com-
posites with 8% Vf under dry, 50% RH and wet con-
ditions are 121, 75, and 59 MPa, respectively; 39% and
52% decrements are recorded for 50% RH and wet
samples, respectively, compared to the dry specimens.
For composites with 16% Vf under dry, 50% RH, and
wet conditions, tensile strength are 160, 104 and
85 MPa, respectively; 34% and 47% decrements by ten-
sile strength are recorded for 50% RH and wet samples,
respectively, compared to dry specimen.
It is believed that the amount of water in the com-
posites causes the fiber to swell, which alters its dimen-
sions and in turn changes the size, shape, stiffness, and
permeability. The moisture acts as a plasticizer that
reduces the entanglement and intermolecular forces
between molecules, thereby increasing their volume
and mobility.19 Huang and Sun20 also reported that
the tensile strength decreased gradually when glass/ Figure 5. SEM micrograph of fracture surface of tensile speci-
polyester composites were immersed in water. The reac- men under wet condition.
tion between water molecules and the matrix deterio-
rated the interphase and hence weakened the
composites fiber–matrix interfacial shear strength. The
authors also believed that water can cause matrix swell-
ing and the softening of the matrix material resulted in
a decrease of the composite strength.
The tensile moduli for unreinforced PA 6,6 under
dry, 50% RH, and wet conditions are 0.71, 0.62, and
0.38 GPa, respectively (Figure 4). The tensile moduli of
50% RH and wet condition specimen decrease by 14%
and 47%, respectively, compared to the dry specimen.
The same trend is observed for composite specimens
with 8% and 16% Vf under both conditions. For com-
posite specimens with Vf of 8% and 16%, conditioned
at 50% RH, the decrements in tensile moduli are 48%
and 26%, respectively, compared to the dry specimens.
Meanwhile, for wet specimens, the equivalent tensile
moduli decrements are 70% and 39%, respectively,
compared to the dry specimens (see PI values in
Table 2). From the micrograph of the fracture surface,
it can also be seen that the matrix phase for the wet
specimen failed in ductile mode with clear yielding
behavior (Figure 5), compared to a brittle mode seen
in the dry specimen (Figure 6). Figure 6. SEM micrograph of fracture surface of tensile speci-
It has been shown that at higher fiber contents, the men under dry condition.
stress transfer capability between fiber and matrix inter-
face reduced as the moisture content increased. Mohd
Ishak et al.16 has suggested that the absorbed moisture lowered the stress-transfer capability of the fiber–
significantly changed the fracture mode from brittle to matrix interface. A similar phenomenon has been
ductile resulting in the reduction of the tensile strength reported by Huang and Sun20 for the glass/polyester
and tensile modulus. In other studies, Thomason21 composite when the sample was immersed in water.
reported that the reduction in tensile strength and ten- The reduction in modulus when the samples were
sile modulus was due to the existence of water which conditioned in wet environment is due to the region

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

494 Journal of Reinforced Plastics and Composites 30(6)

created by the presence of water where larger surfaces Further increase in moisture uptake from 50% RH
of fiber and matrix participate in the adhesion process. to wet environment, leads to an increase in tensile strain
Abdel-Magid et al.22 speculated that the larger contact compared to the dry specimens; however, a decrement
area may allow the load to transfer to the fiber more is recorded by wet specimens compared to the 50% RH
effectively thus maintaining the strength. However, the specimens. The unexpected lower tensile strain of wet
effect of plasticization is much greater; it hence reduces specimens compared to the 50% RH specimens can be
the strength and modulus. explained along these lines: as the matrix absorbs more
The histogram of tensile strain for composites under water, the crystallinity increases due to disruption of
different conditions is shown in Figure 7. The tensile the intermolecular hydrogen bonding, thus allowing
strain for all conditioned specimens are increased com- the hydrocarbon chain segments to pack better.24
pared to the dry specimens (see PI values in Table 2). Polymer matrix may have also swollen. The same
The increments are 51%, 99%, and 21% for specimens trend of behavior occurred in all specimens with Vf
with Vf of 0%, 8%, and 16%, respectively. The values of 0%, 8%, and 16%.
absorbed moisture in polymer matrix also increases
the ductility of the composites. The increase in tensile
strain upon exposure of the samples to wet environ-
Flexural properties
ment can be attributed to the plasticization of PA 6,6 Histograms of flexural strengths and plot of flexural
caused by moisture absorption. The presence of mois- moduli are shown in Figure 8. For comparison, the
ture gives rise to the lubrication effect which allows the PI values of these properties are calculated using
polymer chains to slip past each other. This is expected Equation (4) and tabulated in Table 3. The flexural
due to the formation of hydrogen bonding between strengths of unreinforced PA 6,6 under dry, 50% RH,
water molecule and PA in the composites and thus and wet conditions are 140, 110, and 82 MPa, respec-
the dipole–dipole interactions between matrix and tively. The flexural strength of 50% RH and wet speci-
fibers become less effective.23 When a fiber breaks, the mens are decreased by 21% and 41%, respectively,
load it supports is transferred to surrounding fibers by compared to the dry specimen. The same trend is
matrix resin. As the relative moisture absorbed observed for the composite specimens with 16% Vf,
increases, adhesion between matrix resin and fiber where the flexural strength of 50% RH and wet speci-
becomes poorer; hence, the matrix can no longer effec- mens decreased by 8% and 31%, respectively, com-
tively distribute the applied load over an appreciable pared to the dry specimen. However, 8% Vf
length of the adjacent fibers. Therefore, there is less composites does not follow the same trend as unrein-
occurrence of fiber breakage and thus fracture strain forced PA 6,6 and composite with 16% Vf. Compared
of composites is increased. to the dry specimen, flexural strength of 50% RH and
wet specimens show an increment and decrement,
respectively. This awkward trend is probably due to

400 0.20
Flexural displacement (mm/mm)

Tensile strain
Flex displacement
Tensile strain (%)

Flexural modulus (GPa)

Flexural strength (MPa)

300 0.15 300 7

Flexural strength
250 Flexural modulus 6
5
200 0.10 200
4
150
3
100 0.05 100
2
50 1
0 0.00 0 0
0%, 50%RH

8%, 50%RH

16%, 50%RH
0%, 50%RH

8%, 50%RH

16%, 50%RH

0%, Dry

8%, Dry
0%, Dry

8%, Dry

16%, Dry
16%, Dry
0%, Wet

8%, Wet

16%, Wet

0%, Wet

8%, Wet

16%, Wet

Vf , Condition Vf , Condition

Figure 7. The tensile strain and flexural displacement of com- Figure 8. The flexural strength and flexural modulus of com-
posites subjected under different conditions. posites subjected under different conditions.

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

Hassan et al. 495

Table 3. PI of flexural properties of composites under different conditions

Vf 0% Vf 8% Vf 16%

Property Dry 50% RH Wet Dry 50% RH Wet Dry 50% RH Wet

Strength Reference 0.79 0.59 Reference 1.14 0.43 Reference 0.92 0.69
Modulus Reference 0.51 0.23 Reference 0.70 0.39 Reference 0.73 0.50
Displacement Reference 1.43 2.13 Reference 1.48 1.24 Reference 1.45 1.47

the optimization of the interfaces between fiber and plasticizer to favor the bending strength. Other authors,
matrix which leads to a higher value of flexural strength Dhakal et al.25 and Stamboulis et al.26 have been
in 50% RH compared to dry condition. Generally, the reported that water molecules act as a plasticizer
flexural strength drops abruptly as the moisture content agent in composite material which can lead to the
increases. This may be caused by the formation of weak increase of the maximum strain for the composites
hydrogen bonding between the water molecules and PA after water absorption. However, the flexural displace-
6,6 matrix. The water molecules diffuse through the ments show the highest value at 50% RH compared to
material, seeking out any charged areas and forcing wet condition in reinforced composite samples. This
polymer chains apart along the way. This is the might be due to the existence of microvoids which
reason why some parts of PA swell after being exposed allow the moisture to fill the space and enhance the
to moisture. The separation of the polymer chains bending behavior as reported by Huang and Sun.20
reduces the polar attraction between chains and
increases chain mobility, resulting in decrease mechan-
ical properties.
Dynamic mechanical analysis
Flexural moduli of 50% RH and wet samples are DMA results are shown in Figures 9–11, depicting the
found to decrease significantly compared to the dry tan  against temperature curves. From these figures,
samples. The flexural moduli of unreinforced PA 6,6 over a temperature range of 100–200 C, two transi-
under dry, 50% RH, and wet conditions are 2.2, 1.1, tion regions as indicated by two damping maxima, are
and 0.5 GPa, respectively, (refer Figure 8). The flexural recorded.
moduli of 50% RH and wet composite specimens with From the tan  curves, the thermomechanical data of
Vf of 8% are reduced by 30% and 61%, respectively, the composites are summarized in Table 4. tan 25 and
compared to the dry specimens. At higher Vf of 16%, tan max are the tan  values at a temperature of 25 C
the equivalent reductions in flexural moduli are 27% and at the maximum value of tan , respectively. Ta and
and 50%. See Table 3 for their PI values. Tß are the temperatures at the maximum value of tan 
Except for composite specimen with Vf of 8% under in a-transition (at higher temperature) and ß-transition
50% RH, generally, flexural displacement of 50% RH (at lower temperature), respectively. The a-transition
and wet specimens have increased dramatically com- temperature is commonly known as the glass transition
pared to the dry specimens (Figure 7). The flexural dis- temperature, Tg.27 This method of analysis has also
placements of unreinforced PA 6,6 under dry, 50% been reported by previous researchers.28–30 It can
RH, and wet conditions are 0.09, 0.13, and 0.20 mm/ clearly be seen that major transition peaks are shifted
mm, respectively. Compared to the dry specimen, the to a lower temperatures and higher tan  values when
displacements of 50% RH and wet specimens are exposed to humid environment.
increased by 43% and 113%, respectively. For the com- For the unreinforced PA 6,6 under dry condition
posite specimens, as presented in Table 3 in the form of (Figure 9), Tß and Tg are recorded at 70 C and
PI, flexural displacements of 50% RH and wet speci- 55 C, respectively. Under 50% RH and wet conditions,
mens at both fiber loadings were also increased com- Tg values are reduced to 17 C and 13 C, respectively.
pared to the dry specimens, however, with lower Similar trends are observed for composites with Vf of
magnitude compared to the pure matrix. 8% and 16% (Figures 10 and 11, respectively). Under
The composites tend to be more ductile when all conditions, Tß is reduced to 76 C to 76 C for
exposed to humid environment and as the fiber 50% RH specimen and completely eliminated for wet
volume fraction decreases. Huang and Sun20 believed specimen. With an increase in water absorption for all
that when the wet samples are being bended, the water the specimens, Tß and Tg were reduced to lower tem-
molecules would enhance the integrity of the samples as peratures with the corresponding higher values of tan .
a result of the greater bending resistance. In addition, This behavior is believed to be due to the weakening of
all the voids will be filled with water and acts as a the polymer matrix arising from the water absorption.

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

496 Journal of Reinforced Plastics and Composites 30(6)

Table 4. DMA thermomechanical data of glass fiber composite

at various conditions

a-transition b-transition

Sample tan 25 tan max Ta tan max Tb

(Vf, Condition) (102) (102) ( C) (102) ( C)

0%, Dry 5.12 8.21 55 3.83 70

0%, 50% RH 9.43 10.10 17 3.98 74
0%, Wet 5.71 11.43 13 – –
8%, Dry 0.87 5.87 68 2.67 64
8%, 50% RH 5.24 5.50 19 2.68 75
8%, Wet 3.08 7.38 18 – –
16%, Dry 0.35 4.32 52 2.41 69
16%, 50% RH 4.03 4.50 11 2.12 76
Figure 9. The DMA tan –temperature curves of unreinforced 16%, Wet 3.27 5.93 16 – –
PA 6,6 matrix under different conditions.

80
Dry
50% RH
Wet
60

40
Tg (°C)

20

0
0 8 16

–20
Fibre volume fraction (%)
Figure 10. The DMA tan –temperature curves of composites
with 8% Vf under different conditions. Figure 12. Glass transition temperature, Tg, of composites
under different conditions.

The variation of Tg and Tß with Vf are shown in

Figures 12 and 13, respectively.
As tabulated in Table 4, tan 25 is highest when
exposed to 50% RH environment and lowest in dry
condition. This is believed to be due to the increase in
crystallinity of the matrix which breaks up intermole-
cular hydrogen bonding and allows the hydrocarbon
chain segments to pack better. This behavior is in the
same order as the values of tensile strains of unrein-
forced and reinforced polymers, i.e., highest shown by
the specimens under 50% RH and lowest for the dry
specimens, also tested at 25 C.
In the a-transition region, tan max is increased and
Tg is decreased when moisture absorption increased.
Figure 11. The DMA tan –temperature curves of composites This indicates that humidity acts as a plasticizer
with 16% Vf under different conditions. which induces an increase in the amorphous chain

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

Hassan et al. 497

plasticization produced significant effects on the prop-

Fibre volume fraction (%)
–60 erties of unreinforced and glass fiber-reinforced PA 6,6.
0 8 16
Acknowledgements
–65 We thank the Government of Malaysia and the University of
Malaya as the study reported in this article was supported by
IRPA 33-02-03-3054 and RG092/10AFR.
Tb (°C)

–70
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
–75
Dry
50% RH References
Wet – Eliminated 1. Shiao ML and Nair SV. Effect of glass fibre-fibre rein-
–80 forcement and annealing on microstructure and mechan-
ical behaviour on polyamide 6,6. J Mater Sci 1994; 29:
Figure 13. Tb of composites under different conditions. 1973–1981.
2. Velmurugan R and Manikandan V. Mechanical proper-
ties of palmyra/glass fibre hybrid composites. Composites
Part A 2007; 38: 2216–2226.
mobility in the material and hence reduces the Tg sig- 3. Garbassi F and Po R. Engineering Thermoplastics –
nificantly.31 In ß-transition region, tan max is increased Overview. In: Mark HF (ed.) Encyclopedia of Polymer
and Tß is slightly decreased when exposed to 50% RH. Science and Technology. 3rd edn. Volume 2, New York:
John Wiley and Sons, 2003, pp.307–328.
However, Tß was eliminated (no value recorded) for all
4. ASTM Standard, Standard Test Method for Tensile
wet specimens. All these parameters suggest that the Properties of Plastics, 2003, D638-03, pp. 228–244.
higher content of moisture introduced the plasticizing 5. ASTM Standard, Standard Practice for Conditioning
effect in the composites which can lower the transition Plastics for Testing, 2000, D618-00, pp. 39–42.
region. Aitken et al.32 have also reported the reduction 6. ASTM Standard, Standard Test Method for Flexural prop-
in Tg from 90 C to 57 C in their study on the effect of erties of Unreinforced and Reinforced Plastics and
moisture on acrylic fibers. This further supports our Electrical Insulating Materials, 2003, D790-03, pp. 149–159.
results on the plasticizing action of water. 7. Senthilvelan S and Gnanamoorthy R. Damping charac-
teristics of unreinforced, glass and carbon fibre reinforced
polyamide 6/6 spur gears. Polym Test 2006; 25: 56–62.
Conclusion 8. Hassan A, Yahaya AH, Ibrahim ZA and Hornsby PR.
From the FLD characterization, it is found that Ln and Short- and long glass fibre reinforced polyamide 6,6 com-
posites: tensile properties of injection moulded specimens.
Lw of composites decreased with increase in fiber load-
Malaysian J Sci 2001; 20: 51–55.
ing, indicating that more fiber degradation has 9. Denault J, Vu-Khanh T and Foster B. Tensile properties
occurred during processing. TGA results showed that of injection molded long fiber thermoplastic composites.
PA 6,6 absorbed about 2% and 5% w/w of water when Polym Compos 1989; 10: 313–321.
subjected to 50% RH and wet conditions, respectively. 10. Jones DP, Leach CD and Moore DR. The application of
DSC study revealed that, with the absorption of water, instrumented falling weight impact techniques to the
there were negligible changes in the melting tempera- study of toughness in thermoplastics. Plast Rubber
ture, crystallization temperature, and enthalpy of crys- Process Appl 1986; 6(1): 67–79.
talline of unreinforced and reinforced specimens while 11. Hassan A, Yahya R, Yahaya AH, Tahir ARM and
degree of crystallinity and enthalpy of melting Hornsby PR. Tensile, impact and fiber length properties
decreased. Moisture uptake produced the same effect of injection-molded short and long glass fiber-reinforced
polyamide 6,6 composites. J Reinf Plast Compos 2004;
on the tensile and flexural properties where tensile
23(9): 969–986.
strength, tensile modulus, flexural strength, and flexural
12. Hassan A, Hornsby PR and Folkes MJ. Structure-
modulus decreased, while tensile strain and flexural dis- property relationship of injection-molded carbon fiber-
placement increased. DMA results showed that tan  at reinforced polyamide 6,6 composites: the effect of
room temperature and tan  maxima in both a- and compounding routes. Polym Test 2003; 22(2): 185–189.
b-transition regions increased with moisture absorp- 13. Tamara E, Maurice B and Jacques S. Crystallinity mea-
tion, while temperatures at maximum tan  in both surements of polyamides adsorbed as thin films. Polymer
transitions decreased. It can be concluded that water 2002; 43: 4811–4822.

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015

498 Journal of Reinforced Plastics and Composites 30(6)

14. Vasanthan N and Salem DR. Infrared spectroscopic chloride)/wood sawdust composites. Polym Int 2004; 53:
characterization of oriented polyamide 66: band assign- 1210–1218.
ment and crystallinity measurement. J Polym Sci 1999; 24. Bao LR, Yee AF and Lee CYC. Moisture absorption and
38: 516. hygrothermal aging in a bismaleimide resin. Polymer
15. Dasgupta S, Hammond WB and Goddard III WA. 2001; 42: 7327–7333.
Crystal structures and properties of nylon polymers 25. Dhakal HN, Zhang ZY and Richardson MOW. Effect of
from theory. J Am Chem Soc 1996; 118: 12291–12301. water absorption on the mechanical properties of hemp
16. Mohd Ishak ZA, Ariffin A and Senawi R. Effects of Ebre reinforced unsaturated polyester composites.
hygrothermal aging and a silane coupling agent on the Compos Sci Technol 2007; 67: 1674–1683.
tensile properties of injection moulded short glass fibre 26. Stamboulis A, Baillie CA and Peijs T. Effects of environ-
reinforced poly(butylenes terephthalate) composites. mental conditions on mechanical and physical properties
Eur Polym J 2001; 37: 1635–1647. of flax fibres. Composites Part A 2001; 32: 1105–1115.
17. Klata E, Van de Velde K and Krucińska I. DSC investi- 27. Nielsen LE. Mechanical properties of polymers and com-
gations of polyamide 6 in hybrid GF/PA 6 yarns and posites. Vol. 2, New York: Marcel Dekker, Inc., 1974.
composites. Polym Test 2003; 22: 929–937. 28. Sakellariou P, Hassan A and Rowe RC. Interactions and
18. Lee SY, Kang IA, Doh GH, Kim WJ, Kim JS, Yoon HG partitioning of diluents/plasticizers in hydroxypropyl
and Wu Q. Thermal, mechanical and morphological methylcellulose and poly(vinyl alcohol) homopolymers
properties of polypropylene/clay/wood flour nanocom- and blends. Part II: Glycerol. Colloid Polym Sci 1994;
posites. Express Polym Lett 2008; 2: 78–87. 272(1): 48–56.
19. Kohan M. Polyamide Plastics Handbook. New York: 29. Sakellariou P, Hassan A and Rowe RC. Phase separation
Hanser-Gardner Publications, Inc., 1995, p.631. and polymer interactions in aqueous poly(vinyl alcohol)/
20. Huang G and Sun H. Effect of water absorption on the hydroxypropyl methylcellulose blends. Polymer 1993;
mechanical properties of glass/polyester composites. 34(6): 1240–1248.
Mater Des 2007; 28: 1647–1650. 30. Sakellariou P, Hassan A and Rowe RC. Plasticization
21. Thomason JL. The influence of fibre length, diameter and and interactions of polyethylene glycol 6000 with hydro-
concentration on the impact performance of long glass- xypropyl methyl cellulose/polyvinyl alcohol blends.
fibre reinforced polyamide 6,6. Composites Part A 2009;
Int J Pharm 1994; 102(1–3): 207–211.
40: 114–124.
31. Adams RD and Singh MM. The dynamic properties of
22. Abdel-Magid B, Ziaee S, Gass K and Schneider M. The
fibre reinforced polymers exposed to hot, wet conditions.
combined effects of load, moisture and temperature on
Compos Sci Technol 1996; 56: 977–997.
the properties of E-glass/epoxy composites. Compos
32. Aitken D, Burkinshaw SM, Cox R, Catherall J, Litchfield
Struct 2005; 71: 320–326.
RE, Price DM and Todd NG. Determination of the Tg of
23. Sombatsompop N and Chaochanchaikul K. Effect of
wet acrylic fibres using DMA. J Appl Polym Sci: Appl
moisture content on mechanical properties, thermal and
Polym Symp 1991; 47: 263–269.
structural stability and extrudate texture of poly(vinyl

Downloaded from jrp.sagepub.com at UNIV OF PITTSBURGH on March 9, 2015