See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/297898134

Tensile Behavior of Thermoplastic Composites

Including Temperature, Moisture, and
Hygrothermal Effects
Article in Polymer Testing March 2016
DOI: 10.1016/j.polymertesting.2016.03.011

CITATIONS

READS

96

2 authors:
Mohammadreza Eftekhari

Ali Fatemi

Ford Motor Company

University of Toledo

9 PUBLICATIONS 21 CITATIONS

151 PUBLICATIONS 3,720 CITATIONS

SEE PROFILE

All content following this page was uploaded by Mohammadreza Eftekhari on 25 March 2016.

The user has requested enhancement of the downloaded file.

SEE PROFILE

Polymer Testing 51 (2016) 151e164

Contents lists available at ScienceDirect

Polymer Testing
journal homepage: www.elsevier.com/locate/polytest

Material behaviour

Tensile behavior of thermoplastic composites including temperature,

moisture, and hygrothermal effects
Mohammadreza Eftekhari, Ali Fatemi*
Mechanical, Industrial and Manufacturing Engineering Department, The University of Toledo, 2801 West Bancroft Street, Toledo, OH 43606, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 29 January 2016
Accepted 8 March 2016
Available online 11 March 2016

An experimental study was conducted to investigate the effect of temperature, moisture, and hygrothermal aging on the tensile behavior of thermoplastic composites. Four different composites including
talc lled and short glass ber reinforced polypropylene, short glass ber reinforced polyamide-6.6, and a
blend of polyphenylene ether and polystyrene with short glass bers were used for the study. Kinetics of
water absorption and desorption were investigated for polyamide-6.6 composites and Fickian behavior
was observed. The reductions in tensile strength and elastic modulus due to water absorption are represented by mathematical relations as a function of moisture content. In addition to moisture content,
aging time was also found to inuence the tensile behavior. A parameter is introduced for correlation of
normalized stiffness and strength with different aging times and temperatures. Higher strength and
stiffness obtained for re-dried samples after aging was explained by an increase in crystallinity.
2016 Elsevier Ltd. All rights reserved.

Keywords:
Tensile behavior
Elevated temperature
Temperature effect
Moisture effect
Hygrothermal effect

1. Introduction
The use of thermoplastic composites in load-bearing applications, including in the automotive industry, is constantly increasing
due to a large number of advantages they present. These include
ease of processing for complex geometries at high production rate,
outstanding cost to performance ratio, ability to reprocess and
corrosion resistance. Fillers such as talc and reinforcements such as
glass are widely used to improve strength, elastic modulus and
stability of neat thermoplastics.
Components made of thermoplastic composites have applications in harsh environmental conditions such as elevated temperature and moisture, as well as their combination which is known as
hygrothermal effect. For example, most under the hood (bonnet)
parts in an automobile are subjected to elevated temperature
which can be as high as 130
C because of engine temperature and
weather conditions, as well as moisture because of weather which
can be 100% RH. Mechanical performance of all thermoplastic
composites is highly temperature dependent, while the effect of
moisture on their mechanical properties is dependent on matrix
composition, ber type and ber/matrix interface.

* Corresponding author.
E-mail addresses: mohammadreza.eftekhari@rockets.utoledo.edu (M. Eftekhari),
afatemi@eng.utoledo.edu (A. Fatemi).
http://dx.doi.org/10.1016/j.polymertesting.2016.03.011
0142-9418/ 2016 Elsevier Ltd. All rights reserved.

Tensile properties as the basic material properties are still

widely used in design. A simple tension test provides valuable information about stiffness, strength, toughness and ductility of a
material. Fiber content, length, diameter and direction can affect
tensile properties of polymer composites. These aspects have been
studied in previous studies, such as in Refs. [1,2]. In a recent literature survey by Eftekhari and Fatemi [3] a broad range of effects on
tensile properties of short ber reinforced polymer composites
including temperature, moisture and hygrothermal effects was
presented.
Tensile strength and elastic modulus of thermoplastic composites have been observed to decrease with increasing temperature,
with a sudden drop near the glass transition temperature, Tg [4,5].
Strain at fracture increases with increasing temperature, particularly at temperatures above Tg, as molecular chain motion of polymers intensies. This effect is much more accentuated in
unreinforced materials than in reinforced materials [6]. The temperature dependency of tensile properties is generally more matrix
dominant when loaded in the normal to the molding direction than
when loaded in the molding direction. Tensile properties are more
ber dependent in the molding direction and temperature sensitivity is less [5]. Under tensile loading, ber pull-out and matrix
brittle fracture are the dominant failure mechanisms at low temperatures or high strain rates, while failure occurs via matrix
crazing and crack propagation near the ber ends at elevated
temperatures or low strain rates [7]. Time-temperature

152

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Nomenclature
D
D0
E
Ea
h
K
Mm
Mt
n
R

diffusion coefcient
pre-exponential coefcient
elastic modulus
activation energy
thickness
strength coefcient
% wt of maximum water absorption
% wt of water absorption
strain hardening exponent
gas constant

dependency of tensile properties has been modeled using empirical

models [8].
Kinetics of water absorption of polymers and their composites
has also been studied considering the effects of temperature,
thickness, mold ow direction and reinforcement content. In most
cases, water absorption has been reported to follow Fick's law
[9e11], however, exceptions at some conditions have also been
reported [10,12,13]. For example, at elevated temperatures where
thermal aging, oxidation and leaching are important, more deviation from Fick's law has been observed [12].
Nearly all polymers absorb moisture and the amount of water
uptake depends on the polymer composition. For example, polyamide groups are highly sensitive to moisture because of their polar
structure. The rate of water absorption increases with increasing
temperature, with a sudden change near Tg, while the maximum
water uptake is constant [9]. Increasing temperature enhances
segmental mobility which results in greater zone of activation,
therefore, the rate of water absorption increases. In polymer composites, only the resin typically absorbs moisture, but the ber/
matrix interface can change the rate of water absorption [14].
A signicant reduction in Tg and a slight increase in crystallinity
of the materials has been observed with increasing amount of
water absorption [12]. Tensile strength and elastic modulus reduce
signicantly with increasing water absorption [9,14,15]. The
reduction is lower for composites with enhanced ber/matrix bond
[14]. This effect has been related to plasticization of the matrix and
weakness of ber and matrix interface due to water absorption.
There are inconsistent reports in the literature about the recovery of water absorption effect on materials properties. Some
studies suggest that the water absorption process is a physical
phenomenon and can be recovered with desorption of water [9,12],
while in some other studies full recovery has not been obtained
with complete drying of aged samples [16,17]. The nature of
moisture attack plays an important role in recovery of the mechanical properties. For example, in polyamide composites, the
plasticization effect of water on the resin is a reversible physical
process, however, for PBT, hydrolysis of ester groups is an irreversible chemical process [14]. In addition, exposure of glass bers
to water can result in a chemical reaction. Therefore, composites
without a coupling agent where bers are not protected from
diffusion of water do not show recoverability of mechanical properties with complete drying, while for composites with treated bers full recovery has been reported [18].
In this study, effects of temperature, moisture and hygrothermal
aging on tensile properties of talc lled polypropylene, and short
glass ber reinforced polypropylene, polyamide-6.6 and a blend of
polyphenylene ether and polystyrene were investigated. For the
polyamide-6.6 composite, a complete moisture study at room as well
as elevated temperatures, including the kinetics of water absorption

Su
Sy
t
T
Tg
Tm
Wd
Ww

M
P

ultimate tensile strength

0.2% yield strength
time
temperature
glass transition temperature
melting temperature
dry as-molded specimen weight
wet specimen weight
true stress
true strain
strain at ultimate strength
true plastic strain

and its effect on the tensile properties, was conducted. Scanning

electron microscopy was also performed on fracture surfaces to
study the moisture effect at the microstructural level. For the other
materials, the moisture study was limited to only elevated temperature, since they were less prone to absorbing moisture. The experimental program conducted, the results obtained and the models
applied are presented and discussed in the following sections.
2. Experimental program
2.1. Materials and specimen
One talc lled and three short glass ber reinforced thermoplastic composites with different glass transition temperature, Tg,
were used for the experimental study. The 40% by weight (wt%) talc
lled polypropylene (designated as PP-T) and 30 wt% short glass
ber reinforced polypropylene (designated as PP-G) have Tg of
about 11
C and 23
C, respectively, and a melting point of 165
C.
Thermylene is the trade name for PP-T and PP-G.
A 30 wt% short glass ber reinforced polyamide-6.6 (here
designated PA66) with the trade name of Ultramid is another material used with Tg in the range of studied temperature (55
C) and
melting pint of 260
C. Another material is a 20 wt% short glass ber
reinforced modied polyphenylene ether and polystyrene resin
(here designated PPE/PS), which is known as Noryl. This is a re
retardant material with high values of Tg (135
C) and melting
temperature (325
C), as compared to the other thermoplastic
composites used in the experimental study.
A summary of the materials used in the experimental study is
given in Table 1, including their Tg and melting temperatures. Tg was
measured in dried as-molded condition from loss tangent peak in
the dynamic mechanical analysis (DMA) tests for PA66 and PPE/PS,
and from differential scanning calorimetry (DSC) tests for PP-T and
PP-G [19].
Rectangular plates with dimensions of 100  200 mm x 2.8 mm
were injection molded. Rectangular strips were cut in 90
(transverse) direction with respect to the mold ow direction, according
to Fig. 1(a). A portion of plaques from injection molding side with
the length of 60 mm was discarded to prevent end effects. The
specimens were machined from the strips with the geometry
shown in Fig. 1(b), using a CNC milling machine.
All the specimens were dried and kept in a desiccator prior to
the temperature effect study tests and prior to immersion in water
for the moisture and hygrothermal effects study tests. PA66 specimens were dried for 24 h at 80
C in a vacuum chamber, and PPE/
PS specimens were dried for 4 h at 100
C in a vacuum chamber. As
PP-T and PP-G are less prone to absorbing moisture compared with
PA66 and PPE/PS, they were dried for 4 h at 80
C in a regular
chamber.

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

153

Table 1
Composites used for the experimental study with their melting and glass transition temperatures.
Material

Matrix

Reinforcement

Tm (
C)

Tg (
C)

PP-T
PP-G
PA66
PPE/PS

Polypropylene
Polypropylene
Polyamid-6.6
Polyphenylene ether Polystyrene

40%
30%
30%
20%

165
165
260
325

11
23
55
135

talc
short glass ber
short glass ber
short glass ber

Fig. 1. (a) Location of specimens cut from injection molded plaques in the transverse to mold ow direction (dashed areas are discarded materials). (b) Specimen geometry used for
tension tests (All dimensions are in mm).

Table 2
Tensile properties of PP-T, PP-G, PA66 and PPE/PS at different testing temperatures under displacement rate of 1 mm/min in transverse to mold ow direction and dry-asmolded condition.
Material Temp
(
C)

Strain rate (1/

sec)

PP-T

5
5
2
2
2
2
4
4
6
6
7
7
4
4
4
4
4
4
4
4
4
4
3
3
4
4

PP-G

PA66

PPE/PS

23
23
85
85
120
120
23
23
85
85
120
120
23
23
85
85
120
120
150
150
23
23
85
85
120
120

(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)
(104)

0.2% offset yield strength

(MPa)

Ultimate strength
(MPa)

Strain at ultimate strength (mm/

mm)

Elastic modulus (ASTM)

(GPa)

K
(Mpa)

21.1
21.3
7.4
7.7
3.2
3.4
27.7
28.5
12.1
12.4
6.2
5.9
67.7
71.2
25.2
26.1
19.5
20.1
16.8
18.1
52.4
52.7
38.2
40.8
22.4
22.8

29.6
29.2
13.8
13.7
7.5
7.8
47.1
46.6
21.6
21.9
12.0
11.8
105.4
102.4
58.1
59.5
47.2
46.3
40.2
40.0
71.2
70.5
46.3
46.8
29.9
30.7

0.023
0.021
0.039
0.041
0.056
0.049
0.041
0.041
0.067
0.065
0.140
0.150
0.036
0.032
0.085
0.085
0.089
0.090
0.081
0.088
0.031
0.032
0.024
0.020
0.023
0.017

4.40
4.10
1.60
1.50
0.65
0.70
3.46
3.43
1.40
1.63
0.93
1.00
5.30
5.20
2.27
2.22
1.60
1.53
1.47
1.53
4.40
4.50
4.00
4.00
3.50
3.40

59.5
58.4
34.4
32.6
20.2
21.0
111
101
39.5
42.3
20.7
20.5
258
246
136
143
121
106
106
92.5
139
133
73.7
69.3
55.2
57.5

0.157
0.152
0.233
0.221
0.278
0.272
0.209
0.191
0.172
0.180
0.174
0.181
0.198
0.185
0.245
0.250
0.269
0.238
0.277
0.242
0.147
0.139
0.098
0.075
0.131
0.124

154

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Fig. 2. Effect of temperature on tensile stress-strain curve of (a) PP-T and PP-G, (b) PA66, and (c) PPE/PS in the transverse direction under displacement rate of 1 mm/min.

2.2. Experimental method

Tension tests were conducted based on ASTM D638 [20] and
ISO-527 [21] at a displacement rate of 1 mm/min using an Instron
servo-hydraulic testing machine. A pair of pneumatic grips with
5 kN capacity and adjustable pressure system which were suitable
for testing polymers at elevated temperatures was used. Strain was

measured using an Instron video extensometer. Two dots 8 mm

apart in the gage section were marked and the video extensometer
measured the distance between these two points, and hence strain,
during the test. A load cell with a capacity of 10 kN was used to
measure the applied load. An environmental chamber employing
an electronic heating element with accuracy better than 1
C was
used for performing elevated temperature tests.

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

155

Fig. 3. Variation of ultimate (Su) and yield (Sy) strength and elastic modulus (E) with temperature for (a) PP-T and PP-G, (b) PA66, and (c) PPE/PS. Hashed areas are the range for Tg.

A Hitachi scanning electron microscope was used to observe

fractured surface of specimens. The fractured surface was sputtercoated with gold for 55 s, prior to microscopy.
To investigate the effect of temperature, tension tests were
conducted at 23
C, 85
C, and 120
C for PP-T, PP-G and PPE/PS, and
at these temperatures in addition to at 150
C for PA66. To consider
the hygrothermal aging effects on tensile properties, PP-T, PP-G,
PPE/PS and PA66 samples were immersed in 85
C water for 4 days
(saturation time for PA66 in 85
C water), then tested at 23
C. In
addition to the aforementioned conditions, to investigate the kinematics of water absorption and assess the effect of water absorption on tensile properties of PA66, samples were immersed in
23
C water for a duration between one day to 140 days, and in
85
C water for a duration of a few hours up to 6 days.
The tensile properties for duplicate tests (considered sufcient)
for each material at different temperatures for dry as-molded

conditions are listed in Table 2. These include strain rate, the 0.2%
offset yield strength (Sy), ultimate strength (Su), strain at ultimate
strength (M) and elastic modulus (E).
The Ramberg-Osgood relation has been shown [5] to represent
stress-strain curves of short ber reinforced polymers very well.
This model was also used in this study to represent stress-strain
curves of the materials, given in the following form:

s  s 1n

E
K

(1)

where K is the strength coefcient and n is the strain hardening

exponent. These properties were obtained based on ASTM E646 for
metallic sheet materials [22] from a t of true stress (s) versus true
plastic strain (P) on a log-log scale in the form:

156

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Fig. 4. Variation of (a) normalized ultimate tensile strength and, (b) elastic modulus (with respect to their values at 23
C) with temperature for PP-T, PP-G, PA66, and PPE/PS.

s KP n

(2)

These values are also listed in Table 2.

3. Temperature effect
Stress-strain curves are considerably inuenced by temperature
for all the materials, as shown in Fig. 2. Large deformations were
observed for PP-T and PP-G at 85
C and 120
C, and PP-T specimens
became gummy at these temperatures. The plastic deformation of
PP-G was less than PP-T at all temperatures, related to the
strengthening effect of glass bers. As can be seen in Fig. 2, PA66 at
23
C behaves in a brittle way, as its response shows linear elastic
behavior for a large portion of the stress-strain curve. At elevated
temperatures, this material behaves in a ductile manner with signicant plastic deformation. Strain at rupture increases with
increasing temperature from 23
C to 85
C and remains constant
up to 150
C. For PPE/PS, the behavior is different to the other
studied materials, where a reduction in strain at rupture is
observed with increasing temperature. Since the tested temperatures are lower than Tg, which is 135
C, more stability in deformation is observed with increasing temperature.
Variations of Su, Sy, and E with temperature are shown in Fig. 3
for all the materials. A linear reduction in tensile properties with
temperature is observed for PP-T, PP-G and PPE/PS. For these materials, Tg is not in the range of studied temperatures. However, for
PA66, two different slopes can be identied for variations of
strength and stiffness at temperatures near and above Tg. The
reduction is more signicant with increasing temperature from 23
to 85
C as compared to the reduction from 85 to 150
C.
The strength (Su and Sy values) of PP-G are higher than PP-T at all
temperatures due to the strengthening effect of the glass bers. The
stiffness (E value) of PP-T is higher than PP-G at 23
C, equal to PP-G
at 85
C, and less than PP-G at 120
C. The main reason for incorporating talc in polymers is to increase the stiffness. The degree of
rigidity depends on the lling level, aspect ratio (diameter to
thickness) and neness of the talc [23]. Talc has relatively high
stiffness and increases the crystallinity index of the composite [23]
and, therefore, increases the stiffness of polypropylene. For PP-T at
elevated temperatures (85
C and 120
C), the polypropylene matrix becomes gummy, resulting in reduction of bonds between talc
and matrix and, therefore, talc loses its stiffening effect.
Addition of talc decreases the tensile strength as compared to
neat polypropylene [24,25] due to its stress concentration effect. It

also has a weak interfacial bond with polypropylene and creates

voids and areas for damage initiation, therefore, reducing the
strength increasingly with increasing talc content.
The sensitivity of strength and stiffness to temperature can be
compared for all the materials using Fig. 4, which shows the variation of normalized Su and E with respect to their values at 23
C.
Effect of temperature on strength of both PP-T and PP-G is the same.
This indicates the dominancy of the matrix in temperature dependency on strength. Temperature sensitivity of stiffness is higher
for PP-T as compared to PP-G. Temperature sensitivity of elastic
modulus for PPE/PS is much less compared to other materials. This
material has an amorphous matrix which is glassy below Tg where
secondary bonds determine elastic properties and are less affected
by a change in temperature [26].
The Ramberg-Osgood relation presented by Eq. (1) was used to
represent the stress-strain curves up to ultimate tensile strength of
the studied materials at all temperatures. As can be seen in Fig. 5,
this relation can represent the true stress-true strain curves of all
the materials very well. Variation of strength coefcient, K, with
temperature was observed to be very similar to the variations of
tensile strength and elastic modulus with temperature. A uniform
trend for variation of strain hardening exponent, n, with temperature was not observed. For PP-T and PA66, strain hardening
exponent increased with increasing temperature, while for PP-G
and PPE/PS the opposite trend was observed.

4. Moisture and hygrothermal effects

Hygrothermal aging effect on tensile properties of PP-T, PP-G
and PPE/PS was observed to be much less than for PA66, as shown
in Fig. 6. Polypropylene, because of its non-polar structure, is much
less sensitive to moisture as compared to polyamide [23]. Hygrothermal aging affected the tensile strength, elastic modulus (mainly
for PP-T) and the strain at failure for PP-T and PP-G, which can be
seen in Fig. 6a and b. About 7% and 4% reduction in tensile strength
with water absorption was observed for PP-T and PP-G, respectively, with respect to DAM condition.
PPE/PS sample absorbed 0.21 wt% with immersion in 85
C
water for four days. A small reduction of elastic modulus (about 4%)
and larger reduction of tensile strength (about 17%) was observed
for PPE/PS with hygrothermal aging, as can be seen in Fig. 6c. Since
PPE/PS matrix is amorphous, a small amount of water absorption
causes a more signicant reduction in tensile strength, as compared
to PP-T and PP-G, and even PA66 with the same amount of water

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

157

Fig. 5. Stress-strain curves (solid lines) and corresponding Ramberg-Osgood equation representation (dash lines) at different temperatures for (a) PP-T, (b) PP-G, (c) PA66, and (d)
PPE/PS in the transverse direction at displacement rate of 1 mm/min in DAM condition.

absorption.
PA66 sample absorbed 5.4% moisture with immersion in 85
C
water for four days which is much higher as compared to the other
studied materials. Reduction in elastic modulus and tensile

strength were obtained to be 49% and 62%, respectively. Polyamide

is a semi-crystalline polymer for which polar amide groups form
most of its crystalline regions [27]. The electrons shared between
some atoms are not shared equally, resulting in regions of slight

158

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Fig. 6. Effect of water absorption on tensile stress-strain curve of (a) PP-T, (b) PP-G, (c) PPE/PS, and (d) PA66 at testing temperature of 23
C. [%wt of moisture absorption, immersion
time, immersion temperature]. Duplicate tests were conducted for DAM condition of all the materials.

positive and slight negative charges in the polymer chain. When

water molecules come into contact with polyamide, weak bonds
form and water molecules diffuse through the material forcing
polymer chains apart, causing swelling. Water is absorbed only in
the amorphous phase, while crystalline regions resist being pulled
apart. This is because the bonds between the amide groups are
stronger than the attraction to water, therefore, water acts as a

plasticizer rather than a solvent. The separation of polymer chains

reduces the polar attraction between chains and allows for
increased chain mobility, resulting in diminished mechanical
properties such as strength and stiffness.
The polarity of polyamides also produces a strong bond with
reinforcements such as glass, which is also polar. In addition, a
silane coupling agent is usually used to increase the interface

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

159

23
C with 2
C variation. The weight percentage (%wt) of moisture absorption was periodically measured by weighing the specimens and calculated as:

Mt

Ww  Wd
 100
Wd

(3)

where Wd and Ww are DAM specimen and wet specimen weight,

respectively. Mt values were obtained to be 0.65 wt% after one year,
0.75 wt% after 14 months and 1.45 wt% after two years. As observed
previously for hygrothermally aged samples, this amount of moisture absorption can dramatically affect the mechanical properties.
To achieve DAM condition, moisture should be desorbed
completely. Full recovery of tensile properties were obtained by
drying the 14 months aged sample at 80
C for 24 h in a vacuum
chamber, as shown in Fig. 7. Full water desorption was not obtained
by drying in a regular chamber.
PA66 samples were immersed in 23
C and 85
C water for
different durations and %wt of water absorption was measured
based on Eq. (3). Fick's law is the most commonly used approach to
model single free-phase diffusion [29] and predicts that the %wt of
absorbed uid increases linearly with square root of time, and then
gently slows until it reaches a plateau (saturation). The simplest
form of the model can be expressed in two regions as [30]:

 
Mt
4 Dt 1=2
1=2 2
Mm p
h

for

Dt
< 0:05
h2

   
Mt
8
Dt
1  2 exp  2 p2
Mm
p
h
Fig. 7. Stress-strain curves of PA66 samples aged for 14 months in the laboratory
environment with different drying conditions.

Fig. 8. Water absorption at 23
C and 85
C water, and water desorption at 85
C in

vacuumed oven for PA66. The curves correspond to Fick's law.

adhesion between glass ber and matrix such as polyamide and to

protect the interphase region from water damage [18,28]. Since
tensile properties of PA66 were much more affected by moisture
absorption as compared to the other studied materials, a more
comprehensive moisture study was conducted on it.
4.1. Moisture absorption of PA66
As received or dry-as-molded (DAM) PA66 samples after
molding process were kept in the laboratory environment which
has average of 20% RH with 10% variation and temperature of

for

(4)
Dt
> 0:05
h2

(5)

where t is time, h is the thickness of the sample, D is the diffusion

coefcient and Mm is the maximum water absorption. D values can
be obtained at each temperature from the t of Eq. (4) to the initial
linear portion of the moisture uptake curve.
Fig. 8 shows the %wt of water uptake as a function of time at
immersion temperatures of 23 and 85
C and the corresponding
Fick's model predictions. Each data point is the average weight of
three different samples. At both temperatures, good agreement is
observed between experimental results and Eqs. (4) and (5), which
conrms the Fickian behavior of PA66. Mm value is 5.4% at 23
C and
85
C, indicating Mm is independent of immersion temperature. D
values were obtained to be 3  1013 and 1.3  1011 (m2/s) at 23
C
and 85
C, respectively, indicating that unlike Mm, diffusion coefcient is highly inuenced by immersion temperature. Due to
higher mobility of polymer chains at elevated temperatures, the
rate of water absorption increases dramatically. Mm and D values
Obtained are in agreement with values reported by Ishak and Bery
[15] for 30 wt% short carbon ber reinforced polyamide-6.6, suggesting that glass or carbon do not affect the water uptake process.
For polymers and polymer composites, diffusion coefcient D
has been related to the immersion temperature by an Arrhenius
type relationship [14], given by:



Ea
D D0 exp
RT

(6)

where D0 is the pre-exponential coefcient, R is the gas constant

(8.31 J/mol K), Ea is the activation energy and T is the absolute
temperature in Kelvin. Values of D0 and Ea were reported to be
8.5  103 m2/s and 58.65 kJ/mol, respectively, for short carbon
reinforced polyamide-6.6 composite. Using Eq. (6) with these
constants results in reasonable values for D at 23
C and 85
C
(3.6  1013 m2/s and 2.3  1011 m2/s, respectively). Broudin et al.
[31] mention that the Arrhenius equation is accurate for the glassy

160

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Fig. 9. Tensile stress-strain curves at different water absorption conditions showing the effect of water absorption for PA66 at (a) 23
C and (b) 85
C. [% wt of moisture absorption,
immersion time, immersion temperature]. Duplicate tests were conducted for DAM condition in both gures.

the polymer resin. One sample after 130 days immersion in 23
C

water with saturated water content of 5.4 wt% was dried in vacuum
chamber at 85
C. The water desorption during time, similar to
water absorption, follows Fick's model, as shown in Fig. 8. After four
days, full drying was obtained.
4.2. Effect of water absorption on tensile properties of PA66

Fig. 10. DSC thermographs of unaged and aged PA66.

state of polymer (T < Tg), but for rubbery state (T > Tg) a more
complicated behavior should be considered. Therefore, more variation at 85
C is expected. Estimated D values can be used with Eqs
(4) and (5), and the fact that maximum water uptake is constant at
different temperatures (5.4% for PA66), to predict the amount of
absorbed moisture at different temperatures and time.
As mentioned earlier, water molecules diffuse through the PA66
samples, forcing polymer chains apart and causing swelling of the
polyamide. Comparison of the dimensions of the samples before
and after immersion in water showed 4% increase in thickness and
0.1% increase in width at both 23
C and 85
C for the saturated
condition. These are in agreement with data reported by Thomason
and Porteus [32]. As the bers are oriented in the transverse to the
mold ow direction, dimension change in width should be less,
since the bers which are along the width prevent movement of

Tensile test were conducted on samples aged at 23
C and

hygrothermally aged at 85
C at the same temperature as that of
aging. Stress-strain curves for different moisture contents tested at
23
C and 85
C are shown in Fig. 9. Tensile strength, elastic
modulus and strain at failure are considerably inuenced by the
amount of absorbed water. Tg of polyamide-6.6 decreases considerably with increasing moisture content and can reach 0
C from
about 55
C [31]. At 23
C, for aged samples with Tg higher than
23
C, strain at failure is similar to that of DAM condition, but for
aged samples with Tg lower than 23
C strain at failure increased
(see Fig. 9a). At 85
C, which is above Tg, because of postcrystallization [10], a reduction in strain at failure is observed
compared to DAM condition (see Fig. 9b).
Comparison of stress-strain curves of samples conditioned in
laboratory environment, shown in Fig. 7 (aging time of 14 months),
with stress-strain curve of samples conditioned at 23
C in water,
shown in Fig. 9a (aging time of 4 days) shows that, in addition to %
wt of absorbed moisture, aging time has also affected the mechanical behavior.
The stress-strain curve of re-dried sample after immersion for
130 days in 23
C water, shown in Fig. 9a, indicates that water
desorption somewhat increases both stiffness and strength of the
PA66 as compared to the DAM condition. The percentage of crystallinity of DAM and re-dried PA66, obtained from DSC thermograms shown in Fig. 10, was found to be 25.9% and 27.1%,
respectively. Narrower melting peak for the aged sample shows

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

161

Fig. 11. Variation of ultimate tensile strength (Su) and elastic modulus (E) with water absorption at (a) 23
C and (b) 85
C. Variation of normalized ultimate tensile strength and
elastic modulus as a function of (c) moisture content and, (d) Mt t0.5 at 23
C and 85
C.

formation of more organized crystalline structure, as compared to

the DAM sample. This increase in crystallinity resulted in the increase of both stiffness and strength, which was also reported in
Refs. [12,33].
Exponentially decaying t with %wt of absorbed water was used
for tensile strength and elastic modulus at both 23 and 85
C, as
shown in Fig. 11a and b. Additional reduction of strength and
stiffness after 4%wt water absorption, which is the point of transition from linear to exponential form in Fick's model, is nearly
negligible at both temperatures. Normalized tensile strength and
elastic modulus with respect to their values in DAM condition is
used to compare the water aging effects at 23
C and 85
C, as
shown in Fig. 11c. It can be seen that the effect of water aging on Su
and E is almost the same at both temperatures.
The water aging effect on strength and stiffness were signicantly more at 23
C as compared to 85
C, as can be seen in Fig. 11c.
The aging time is much less at 85
C compared to 23
C (22 times
less for saturated condition). When the factor of time is combined
with %wt of water absorption in the form Mt t0.5, a good correlation
of both strength and stiffness at both 23
C and 85
C is obtained, as
can be seen in Fig. 11d. An exponentially decaying curve is obtained
up to the transition point mentioned earlier (i.e. from linear to
exponential form in Fick's model). After this point, the tensile
properties remain constant. The equation of the t is expressed as:
0:5
X
0:904e0:004Mt t
X0

(7)

where X can be Su or E and X0 is the Su or E in DAM condition.

Fig. 12 shows SEM micrographs of fracture surfaces of the core

layer of DAM samples tested at 23
C and 85
C, aged samples tested

at 23
C and 85
C, and a re-dried sample after saturation in 23
C
water. Samples were cut transverse to the mold ow direction,
therefore, the orientation of the majority of bers in the core layer
are perpendicular to the fracture plane due to the core-shell
morphology [2]. Matrix cracking with no evidence of matrix
deformation along with ber pullouts, indicating brittle failure of
the sample, are observed for DAM sample tested at 23
C (Fig. 12a).
Signicant plastic deformation of the matrix, mainly at ber tips
due to stress concentration, is observed at 85
C for DAM condition
(Fig. 12b). For the aged samples, matrix yielding and plastic deformation, illustrating the physical damage of the matrix, and
debonding between ber and matrix (mainly at 85
C) which results to an increase in deformation, are evident (Fig. 12c and d).
Fracture surface of the re-dried sample shows brittle failure, similar
to that of the DMA sample at 23
C, and a strong ber/matrix
interface is observed (Fig. 12e).
In order to illustrate the effect of water aging and re-drying on
ber/matrix interface, SEM micrographs of the ber in the fracture
surface in different conditions are shown in Fig. 13. For DAM
samples tested at 23
C and 85
C, an excellent bond between ber
and matrix is observed (Fig. 13a and b). Continuous cover of ber by
matrix at 23
C, and signicant plastic deformation and brils of
matrix at the tip of ber at 85
C, are indications of brittle and
ductile failures, respectively [34]. Diffused water at 85
C affected
the interface between ber and matrix (Fig. 13d). A nearly clean
ber with small retention of the matrix reveals drastic damage at
the ber/matrix interface with hygrothermal effect (Fig. 13d).
Finally, for the re-dried water aged sample at 23
C, a very good
bond can be seen (Fig. 13e), which conrms recoverability of water

162

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

Fig. 12. SEM micrographs of the fracture surface in the core layer after tensile tests of (a) DAM at 23
C, (b) DAM at 85
C, (c) with 5.4 wt% moisture (saturated) tested at 23
C, (d)
with 5.35 wt% moisture (saturated) tested at 85
C, and (e) re-dried sample after desorption of 5.4 wt% moisture tested at 23
C.

aging at 23
C with complete drying.

5. Conclusions
Temperature, moisture and hygrothermal effects on tensile
behavior of talc lled and short glass ber reinforced polypropylene
(PP-T and PP-G), and short glass ber reinforced polyamide (PA66)
and a blend of polyphenylene ether and polystyrene (PPE/PS) were
investigated. Based on the observed experimental behavior, the
analysis performed and SEM of fracture surfaces, the following
conclusions can be made:
1) Stress-strain curve was considerably inuenced by temperature
for all materials. A linear reduction in tensile strength and elastic

modulus with temperature was observed for PP-T, PP-G and

PPE/PS, for which Tg is not in the range of studied temperatures.
However, for PA66, two different slopes were identied for
variation of strength and stiffness at temperatures near and
above Tg.
2) Polypropylene with short glass bers was observed to have
higher tensile strength and lower elastic modulus as compared
to talc lled polypropylene. Talc has relatively high stiffness and
increases the crystallinity index of the material. However, it has
a weak interfacial bond with polypropylene and creates voids,
resulting in reduced strength.
3) Polyamide-based composite was found to be highly moisture
sensitive. However, full recovery of tensile properties was

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

163

Fig. 13. SEM micrographs of the glass ber on tension test fracture surface of (a) DAM at 23
C, (b) DAM at 85
C, (c) with 5.4 wt% moisture (saturated) at 23
C, (b) with 5.35 wt%
moisture (saturated) at 85
C, and (e) re-dried sample after absorption of 5.4 wt% moisture.

obtained by drying aged samples (aged in the laboratory environment for 14 months) at 80
C for 24 h in a vacuum chamber.
4) Fickian behavior was observed for water absorption of PA66
samples immersed in water at 23
C and 85
C. Maximum water
absorption was independent of temperature, while the rate of
water absorption increased dramatically with increasing temperature. Strength, elastic modulus and strain at failure were
considerably inuenced by the amount of absorbed water. A
parameter which incorporate both %wt of absorbed water and
aging time in the form of Mt t0.5 was used to correlate normalized elastic modulus and stiffness at different temperatures.
5) Water aging and also drying at elevated temperature increased
crystallinity and, in turn, increased both stiffness and strength of

PA66. For the aged samples (both in 23
C and 85
C water),

matrix yielding and plastic deformation were observed. In
addition, hygrothermal aging at 85
C was observed to considerably damage the ber/matrix interface.

Acknowledgements
Funding of this study was provided by General Motors. Technical
support of Dr. A.K. Khosrovaneh and Mr. T. Wang is appreciated.
Technical assistance of Dr. Kazem Majdzadeh Ardakani of the
Polymer Institute at the University of Toledo in assisting with
drying specimens and DSC analysis, is also highly appreciated.

164

M. Eftekhari, A. Fatemi / Polymer Testing 51 (2016) 151e164

References
[1] J.L. Thomason, The inuence of bre length and concentration on the properties of glass bre reinforced polypropylene: 5. Injection moulded long and
short bre PP, Compos. Part A Appl. Sci. Manuf. 33 (12) (2002) 1641.
[2] S. Mortazavian, A. Fatemi, Effects of ber orientation and anisotropy on tensile
strength and elastic modulus of short ber reinforced polymer composites,
Compos. Part B Eng. 72 (2015) 116.
[3] M. Eftekhari, A. Fatemi, Tensile, creep and fatigue behaviors of short ber
reinforced polymer composites at elevated temperatures: a literature survey,
Fatigue Fract. Eng. Mater. Struct. 38 (2015) 1395.
[4] M. De Monte, E. Moosbrugger, M. Quaresimin, Inuence of temperature and
thickness on the off-axis behaviour of short glass bre reinforced polyamide
6.6 e Quasi-static loading, Compos. Part A Appl. Sci. Manuf. 41 (7) (2010) 859.
[5] S. Mortazavian, A. Fatemi, Tensile and fatigue behaviors of polymers for
automotive application, Mater. Werkstofftechnik 46 (2) (2015) 204.

ne, A. Maazouz, A study of
[6] B. Mouhmid, A. Imad, N. Benseddiq, S. Benmedakhe
the mechanical behaviour of a glass bre reinforced polyamide 6,6: experimental investigation, Polym. Test. 25 (4) (2006) 544.
[7] J.M. Schultz, K. Friedrich, Effect of temperature and strain rate on the strength
of a PET/glass bre composite, J. Mater Sci. 19 (7) (1984) 2246.
[8] V. Di Liello, E. Martuscelli, G. Ragosta, A. Zihlif, Tensile properties and fracture
behaviour of polypropylene-nickel-coated carbon-bre composite, J. Mater
Sci. 25 (1) (1990) 706.
[9] S. Mortazavian, A. Fatemi, Effect of water absorption on tensile and fatigue
behavior of short glass ber polyamide-6 and short glass ber polybutylene
terephthalate, SAE Int. J. Mater Manuf. 8 (2) (2015).
gel, W. Grellmann, Inuence of hygrothermal
[10] T. Illing, M. Schoig, C. Biero
aging on dimensional stability of thin injection-molded short glass ber
reinforced PA6 materials, J. Appl. Polym. Sci. 132 (28) (2015).
[11] Z.A. Mohd Ishak, U.S. Ishiaku, J. Karger-Kocsis, Hygrothermal aging and fracture behavior of short-glass-ber-reinforced rubber-toughened poly(butylene
terephthalate) composites, Compos. Sci. Tech. 60 (6) (2000) 803.
[12] D. Valentin, F. Paray, B. Guetta, The hygrothermal behaviour of glass bre
reinforced Pa66 composites: a study of the effect of water absorption on their
mechanical properties, J. Mater. Sci. 22 (1) (1987) 46.
[13] M.P. Foulc, A. Bergeret, L. Ferry, P. Ienny, A. Crespy, Study of hygrothermal
ageing of glass bre reinforced PET composites, Polym. Degrad. Stabil. 89 (3)
(2005) 461.
[14] Z.A. Mohd Ishak, A. Arifn, R. Senawi, Effects of hygrothermal aging and a
silane coupling agent on the tensile properties of injection molded short glass
ber reinforced poly(butylene terephthalate) composites, Eur. Polym. J. 37 (8)
(2001) 1635.
[15] Z.A. Mohd Ishak, J.P. Berry, Effect of moisture absorption on the dynamic
mechanical properties of short carbon ber reinforced nylon 6, 6, Polym.
Compos. 15 (3) (1994) 223.

View publication stats

[16] Z.A. Mohd Ishak, J.P. Berry, Hygrothermal aging studies of short carbon ber
reinforced nylon 6.6, J. Appl. Polym. Sci. 51 (13) (1994) 2145.
[17] Z.A. Mohd Ishak, N.C. Lim, Effect of moisture absorption on the tensile properties of short glass ber reinforced poly(butylene terephthalate), Polym. Eng.
Sci. 34 (22) (1994) 1645.
[18] X.S. Bian, L. Ambrosio, J.M. Kenny, L. Nicolais, A.T. Dibenedetto, Effect of water
absorption on the behavior of E-glass ber/nylon-6 composites, Polym.
Compos. 12 (5) (1991) 333.
[19] M. Eftekhari, A. Fatemi, On the strengthening effect of increasing cycling
frequency on fatigue behavior of some polymers and their composites: experiments and modeling, Int. J. Fatigue 87 (2016) 153.
[20] ASTM D638-14, Standard Test Method for Tensile Properties of Plastics, ASTM
International, West Conshohocken, PA, 2014.
[21] ISO-527-1, Plastics. Determination of Tensile Properties, 2012.
[22] ASTM E646e07e1, Standard Test Method for Tensile Strain-hardening Exponents (n -Values) of Metallic Sheet Materials, ASTM International, West
Conshohocken, PA, 2007.
[23] H.G. Karian, Handbook of Polypropylene and Polypropylene Composites,
second ed., CRC Press, 2003.
[24] Y. Zhou, P.K. Mallick, Effects of temperature and strain rate on the tensile
behavior of unlled and talc-lled polypropylene. Part I: Experiments, Polym.
Eng. Sci. 42 (12) (2002) 2449.
[25] A.M. Zihlif, G. Ragosta, Mechanical properties of talc-polypropylene composites, Mater Lett. 11 (10e12) (1991) 368.
[26] H.E.H. Meijer, L.E. Govaert, Mechanical performance of polymer systems: the
relation between structure and properties, Prog. Polym. Sci. 30 (8e9) (2005)
915.
[27] J.A. Brydson, Polyamides and polyimides, in: J.A. Brydson (Ed.), Plastics Materials, seventh ed., Butterworth-Heinemann, Oxford, 1999, p. 478.
[28] H. Ishida, J.L. Koenig, The reinforcement mechanism of ber-glass reinforced
plastics under wet conditions: a review, Polym. Eng. Sci. 18 (2) (1978) 128.
[29] A. Fick, Ueber diffusion, Annalen der Physik 170 (1) (1855) 59.
[30] J. Crank, The Mathematics of Diffusion, second ed., Clarendon Press, Oxford,
1975.
[31] M. Broudin, P.Y. Le Gac, V. Le Saux, C. Champy, G. Robert, P. Charrier, Y. Marco,
Water diffusivity in PA66: experimental characterization and modeling based
on free volume theory, Eur. Polym. J. 67 (2015) 326.
[32] J.L. Thomason, G. Porteus, Swelling of glass-ber reinforced polyamide 66
during conditioning in water, ethylene glycol, and antifreeze mixture, Polym.
Compos. 32 (4) (2011) 639.
[33] J.L. Thomason, J.Z. Ali, The dimensional stability of glassebre reinforced
polyamide 66 during hydrolysis conditioning, Compos. Part A Appl. Sci.
Manuf. 40 (5) (2009) 625.
[34] E. Belmonte, M. De Monte, M. Quaresimin, C. Hoffmann, in: Proceedings of
16th European Conference on Composite Materials, Spain, Seville, 2014, pp.
22e26.