Materials and Design 30 (2009) 3489–3496

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Effects of material compositions on the mechanical properties of wood–plastic

composites manufactured by injection molding
Pei-Yu Kuo, Song-Yung Wang, Jin-Hau Chen, Huei-Chin Hsueh, Ming-Jer Tsai *
School of Forestry and Resource Conservation, College of Bio-Resource and Agriculture, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei, 10617 Taiwan, ROC

a r t i c l e i n f o a b s t r a c t

Article history: The objective of this study was to investigate the effects of material compositions on the mechanical
Received 13 January 2009 properties of wood–plastic composites (WPCs) manufactured by injection molding. Using a ratio of wood
Accepted 7 March 2009 flour/plastic matrix/MAPP (maleic anhydride polypropylene)/zinc stearate of 47:47:3:3, the tensile
Available online 17 March 2009
strength and modulus of rupture (MOR) of WPCs manufactured with LDPE (low-density polyethylene)
and PP (polypropylene) were found to be larger than those of LDPE and PP themselves. However, con-
Keywords: trasting findings were obtained when the polymer matrix was ABS (acrylonitrile-butadiene-styrene).
A. Composites
In comparison to the mechanical properties of RPP (recycled polypropylene) itself, the MOR increased
Polymer matrix
A. Natural materials
and the tensile strength decreased for WPCs manufactured with RPP.
C. Moulding The tensile strength, MOR, and storage modulus of WPCs made from PP mixed with 47% wood flours
E. Mechanical (<180 lm) and 3–4.5% MAPP were larger than those of the other WPCs manufactured in this study. How-
ever, the polymer damping peaks showed a contrary result.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction frames, roof shingles, and residential decking. Compared to timber,

WPCs exhibit greater durability, require less maintenance, absorb
It is well known that, over the past few decades, polymers have less moisture, and provide superior fungal resistance [11]. Wolcott
replaced many conventional materials, such as metal and wood, in [9] found that the addition of 40–50% wood improved thermal sta-
many applications. This is due to the advantages of polymers over bility, while the thermoplastic component improved moisture
conventional materials [1]. Recently, natural-fiber-reinforced poly- resistance and thermal formability. When exposed to moisture,
mers have attracted the attention of researchers because of their WPCs absorb less water at a slower rate, leading to superior fungal
advantages over established materials. Plant fibers are biodegrad- resistance and dimensional stability compared to timber [11].
able and readily available, and, compared with glass fibers, their Waterfront applications have also demonstrated that WPC materi-
attributes include low cost, low-density, lower abrasive nature, als exhibit improved durability with respect to checking, decay,
higher specific strength, and higher modulus of elasticity [2–5]. termites, and marine organisms compared to timber [12]. In addi-
However, poor compatibility with hydrophobic polymer matrices, tion, a wood filler improves thermal stability compared to other
a tendency to form aggregates during processing, and a low resis- polymer composites [9]. Ticky [13] indicated that WPCs are widely
tance to moisture, greatly reduce the potential of natural fibers to used in Europe, Japan, and North America because they are engi-
be used as reinforcements for polymers [6,7]. Karnani et al. [8] re- neered and manufactured to meet a range of applications. In recent
ported that the properties of wood fiber/polypropylene composites years, WPCs have been at the forefront of architectural concerns in
were very poor due to the absence of interfacial modifiers. Taiwan.
Wood–plastic composites (WPCs), defined as thermoplastics Numerous studies on the influences of the constituent plastic
reinforced with wood or other natural fibers, are principally pro- matrices and wood flours of WPCs have been carried out
duced from commodity thermoplastics such as polyethylene (PE), [10,12,14–16]. The properties of WPCs may be very poor due to
polyvinyl chloride (PVC), or polypropylene (PP) [9]. By adding bulk the absence of interfacial modifiers [8]. In these circumstances, a
vinyl polymers to the void spaces in wood, compression strength, coupling agent might be used to improve the properties. In this
hardness, and abrasion resistance are greatly improved [10]. Cur- study, the effects of material compositions (including different
rent uses of WPC materials include automotive trim, window plastic matrices, wood flours, and coupling agents) on the mechan-
ical properties of WPCs manufactured by injection molding have
been investigated. The results may provide information for esti-
* Corresponding author. Tel.: +886 2 33664641; fax: +886 2 23686335. mating optimal material compositions in the manufacture of
E-mail address: [email protected] (M.-J. Tsai). WPCs.

0261-3069/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2009.03.012
3490 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496

2. Materials and methods pressure: 417–973 kgf/cm2; (3) injection speed: 17.5–40.8 cm3/s;
(4) cooling time: 20–40 s; (5) screw rotation: 180 rpm; (6) clamp-
2.1. Materials ing temperature: 50–80 °C; (7) maximum clamping force: 50 ton;
(8) ejector: 2–3 times; and (9) screw stroke: 90–110 mm.
LDPE (low-density polyethylene; USI Corporation, Taiwan), PP
(polypropylene; ST868M, Taiwan Polypropylene Company, Tai- 2.3. Tensile strength
wan), RPP (recycled polypropylene; Shih-Jie Company, Taiwan),
and ABS (acrylonitrile-butadiene-styrene; PA-747S, Chimei Corpo- The tensile strength perpendicular to the surface was deter-
ration, Taiwan) were used as the plastic matrices. mined using five conditioned specimens of the manufactured
Three kinds of wood flours were selected for this study, namely WPCs with a tensile speed of 5 mm/min according to ASTM D638
recycled China fir (Cunninghamia lanceolata) supplied by Cheng- [17]. The test specimens were of Type IV, having a width of
Chang Corporation, Taiwan; recycled mixed-wood flours from Jap- 3.5 mm. The tensile strength (rt) was then calculated using the fol-
anese fir (Cryptomeria japonica), China fir, Taiwan red cypress (Cha- lowing formula:
maecyparis formosensis), and Taiwania (Taiwania crytomerioides); P
and commercial SPF (spruce, pine, and fir). Wood flours were di- rt ¼ ð1Þ
A
vided into 250–180 lm, 180–125 lm, and <125 lm grades, which
passed through 60, 80, and 120-mesh screens, respectively. The where P (N) is the tensile load and A (mm2) is the cross-sectional
flours were then dried for 24 h at 103 ± 2 °C in an oven. They were area.
left to cool naturally to room temperature inside the oven. They
were then removed from the oven and sealed in aluminum foil 2.4. Bending strength
bags prior to manufacturing WPCs for testing.
Three kinds of additives were used in this study, namely ZS According to ASTM D4761 – Method A [18], five specimens of
(zinc stearate) as lubricant; MAPP (maleic anhydride polypropyl- dimensions 130  13.5  3.5 mm were used for bending tests. A
ene) as coupling agent; and 8% NaOH as alkali solution. concentrated bending load was applied to the center with a span
of 24 times the thickness of the specimen. The proportional limit,
2.2. Mixing and injection molding ultimate load, and deflection were obtained from the load-deflec-
tion curves, and the bending modulus of rupture (MOR) was calcu-
The compositions of the tested WPCs are indicated in Table 1. lated as follows:
The constituent materials were first combined in a mixer according
to Table 1. The mixing speed, temperature, and time were 55 rpm, 3PL
MOR ¼ 2
ð2Þ
180 °C, and 5 min, respectively. Wood–plastic particles were pro- 2bh
duced using a pelletizer and then dried for 24 h at 103 ± 2 °C in where P (N) is the maximum loading, L (mm) is the span, b (mm) is
an oven prior to injection molding into tension and flexural bar test the width of the specimen, and h (mm) is the height of the
samples. specimen.
After drying, the wood–plastic particles were placed in an injec- The MOR increasing rates were evaluated from the following
tion molding machine (SM-50, Chen-Hsong Company, Taiwan) in formula:
order to manufacture tension and flexural bar test samples. The
processing conditions for the injection molding may be summa- MOR2  MOR1
MOR increasing rate ð%Þ ¼  100 ð3Þ
rized as follows: (1) nozzle temperature: 175–205 °C; (2) injection MOR1

Table 1
Composition of tested WPCs.

Specimen Code Plastic matrix Wood flour Additive Wood flour diameter
ABS 100 100% ABS – –
ABS 47 47% ABS 47% SPF 3% MAPP, 3% ZS <125 lm
LDPE 100 100% LDPE – –
LDPE 47 47% LDPE 47% SPF 3% MAPP, 3% ZS <125 lm
PEPP47 23.5%LDPE + 23.5% PP 47% SPF 3% MAPP, 3% ZS <125 lm
RPP 100 100% RPP – –
RPP 47 47% RPP 47% SPF 3% MAPP, 3% ZS <125 lm
PP 100 100% PP – –
PP 47 47% PP 47% SPF 3% MAPP, 3% ZS <125 lm
PPCW215 47% PP 47% CW 3% MAPP, 3% ZS 250–180 lm
PPCW152 47% PP 47% CW 3% MAPP, 3% ZS 180–125 lm
PPCW 47% PP 47% CW 3% MAPP, 3% ZS <125 lm
PPRW 47% RPP 47% RW 3% MAPP, 3% ZS <125 lm
PP 28 28.2% PP 65.8% SPF 3% MAPP, 3% ZS <125 lm
PP 38 37.6% PP 56.4% SPF 3% MAPP, 3% ZS <125 lm
PP 42 42.3% PP 51.7% SPF 3% MAPP, 3% ZS <125 lm
PP 66 65.8% PP 28.2% SPF 3% MAPP, 3% ZS <125 lm
MA 0 48.5% PP 48.5% SPF 0% MAPP, 3% ZS <125 lm
MA 1.5 47.75% PP 47.75%SPF 1.5% MAPP, 3% ZS <125 lm
MA 4.5 46.25% PP 46.25%SPF 4.5% MAPP, 3% ZS <125 lm
MA 6 45.5% PP 45.5% SPF 6% MAPP, 3% ZS <125 lm
MA 9 44% PP 44% SPF 9% MAPP, 3% ZS <125 lm
NaOH 47% PP 47% SPF 3% MAPP, 3% ZS, NaOH <125 lm

ABS, acrylonitrile-butadiene-styrene; LDPE, low-density polyethylene; PP, polypropylene; RPP, recycled polypropylene; SPF, spruce, pine and fir; CW, recycled China fir wood
flour; RW, recycled mixed-wood flours from Japanese fir, China fir, Taiwan red cypress, and Taiwania; MAPP, maleic anhydride polypropylene; ZS, zinc stearate.
 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496 3491

where MOR1 and MOR2 represent the MOR values of the tested plas-
tic matrices and the WPCs, respectively.

2.5. Dynamic properties

Dynamic-mechanical analysis (DMA) was carried out on a TA

Instruments 2980 analyzer. The dimensions of the test specimens
were 64  14  3.5 mm. A concentrated bending load was applied
to the center of the specimen. The frequency, heating rate, and
heating temperature were 1 Hz, 3 °C/min, and 100–150 °C (from
20 to 100–150 °C), respectively. This DMA yielded the parameters
E0 (storage modulus), E00 (loss modulus), and tan d (damping peaks).
By observing E0 , E00 , and tan d, the change of viscosity and elasticity
of the WPCs, as well as the compatibility between the wood flours
and plastic matrices, could be investigated.

2.6. Melt flow index

Fig. 1. Relationships between temperature and storage modulus (E0 ) of different
The melt flow index (MI) is a measure of the rate of extrusion of plastic matrices.
molten resins through a die of a specified length and diameter un-
der prescribed conditions of temperature, load, and piston position slowly with increasing temperature at first, then increased slightly
in the barrel as the timed measurement is being made. The melt at about 80 °C, but thereafter decreased rapidly at temperatures
flow index may be used to investigate the polymer flow properties above 80 °C, possibly due to the effect of secondary crystallization.
and its molecular weight. According to ASTM D1238 – Method B In our study, the mechanical properties of RPP100 were found
[19], the melt flow index was measured and used to investigate to be superior to those of PP100, presumably due to a higher
the effect of the addition wood flours on the flow properties of molecular weight of the former. Using the MI values to estimate
WPCs in our study. The melt flow index was calculated as follows: the molecular weight, the MI values of RPP100 and PP100 were
3.94 and 18 g/10 min (230 °C, 2.16 kg), respectively. The relation-
pc2 L ship between MI value and molecular weight was established by
MI ¼  600  D ð4Þ
T Bremner et al. [20] and may be expressed as follows:
where MI (g/10 min) is the melt flow index, c (cm) is the radius of
MI ¼ aM W 3:4
the piston, L (cm) is the length of the barrel, D (g/cm3) is the density,
and T (s) is the flow time through the barrel. Based on the a value of PP (1.57  1019 g  Da3.4  min1) accord-
ing to Nele et al. [21], the molecular weights of RPP100 and
3. Results and discussion PP100 were around 300,000 and 190,000, respectively. Such a dif-
ference in the molecular weights of RPP100 and PP100 could ac-
3.1. Properties of different plastic matrices count for the aforementioned difference in their mechanical
properties. Furthermore, in a plot of tan d vs. temperature, the Ta
Air dried densities, moisture contents, and mechanical proper- peak of RPP100 appeared broader and was shifted to higher temper-
ties of the different plastic matrices are listed in Table 2. The air ature in comparison with that of PP100 (Fig. 2). Fig. 2 shows that Tg
dried densities of the various plastic matrices are seen to decrease and Ta for PP100 were around 2 °C and 55 °C, respectively,
in the order: ABS100 (100% acrylonitrile-butadiene-sty- whereas the corresponding values for RPP100 were around 0 °C
rene) > RPP100 (100% recycled polypropylene) > LDPE100 (100% and 65 °C, respectively. McGrum et al. [22] reported that PP exhibits
low-density polyethylene) > PP100 (100% polypropylene). Due to three significant peaks, namely the c peak at around 80 °C, the b
non-polar hydrophobic properties, a low moisture content (0.09– peak (glass transition temperature, Tg) at around 10 °C, and the a
0.38%) was found for all of the tested plastic matrices. peak at around 100 °C. The temperatures of different peaks are
Table 2 indicates that the tensile strengths and MOR values of influenced by the molecular weight, molecular configuration, and
the various plastic matrices decreased in the order: heating rate in the dynamic mechanical analysis. Similar to our re-
ABS100 > RPP100 > PP100 > LDPE100. The differences in the sults, a tan d value of 0.086 at 20 °C was found by Mark [23].
mechanical properties of the various plastic matrices were statisti-
cally very significant (p < 0.01). Observing the E0 values, a similar 3.2. Effects of species and particle diameters of wood flours on the
trend was found (Fig. 1). Fig. 1 also shows that the E0 values of mechanical properties of WPCs
RPP100 and PP100 decreased rapidly when the temperature was
around 0 °C. The E0 value of LDPE100 decreased linearly with The effects of the species and the particle diameters of wood
increasing temperature. The E0 value of ABS100 decreased very flours on the mechanical properties of the WPCs are shown in

Table 2
Air dried density, moisture content and mechanical properties of different plastic matrices.

Specimen code Air dried density (g/cm3) Moisture content in air dry (%) Tensile strength (MPa) MOR (MPa)
LDPE 100 0.92 ± 0.00 0.09 ± 0.02 12.0a ± 1.3 19.9a ± 0.8
PP 100 0.91 ± 0.00 0.09 ± 0.00 27.0b ± 1.2 47.6b ± 0.4
RPP 100 0.96 ± 0.00 0.13 ± 0.01 30.1c ± 0.5 53.3c ± 0.4
ABS 100 1.02 ± 0.01 0.38 ± 0.01 42.2d ± 0.9 78.0d ± 1.6

LDPE100, 100% low-density polyethylene; PP100, 100% polypropylene; RPP100, 100% recycled polypropylene; ABS100, 100% acrylonitrile-butadiene-styrene; MOR, modulus
of rupture.
Means (suffixes a, b, c, and d) within a given column with the same letter do not significantly (p = 0.01) differ as determined by Tukey test.
3492 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496

0.14 WPCs made from wood flours of different particle diameters. How-
PP 100 RPP 100
0.12 ever, the tensile strength increased with decreasing particle diam-
eter. The tensile strengths of WPCs made from China fir flours with
0.1
particle diameters of 250–180 lm were significantly lower than
those of WPCs made from flours of particle diameter 180–
tan

0.08

0.06 T 125 lm or <125 lm. This may possibly have been due to a lack
Tg of bonding between the wood fibers and the plastic matrix because
0.04
of entanglement and agglomeration of longer wood fibers. In our
0.02 study, no significant differences in the MOR values of tested WPCs
0 made from China fir flours of different diameters were found. The
-20 0 20 40 60 80 MOR values determined in our study were larger than those deter-
Temperature( )
mined by Stark et al. [25]. This suggests that wood flours with a
particle diameter of <180 lm might be favorably used for WPC
Fig. 2. Relationships between temperature and tan d (damping peaks) of PP 100 and manufacture.
RPP 100.

3.3. Effects of different plastic matrices on the mechanical properties of

WPCs
Table 3. Using ratios of wood flour (<125 lm)/PP/MAPP/zinc stea-
rate of 47:47:3:3, the air-dried densities of China fir–plastic com-
Table 4 shows the effects of different plastic matrices on the
posites were lower than those of SPF–plastic composites. The
mechanical properties of WPCs. Using wood flour/plastic matrix/
highest densities were found in the group of recycled mixed-
MAPP/zinc stearate ratios of 47:47:3:3, the air dried densities of
wood–plastic composites because of the mixture of higher-density
WPCs made from PP, RPP, LDPE, and ABS were higher than those
wood species. However, the mixture of different wood species re-
of the respective plastic matrices (Table 2). This is because of the
sulted in a higher standard deviation. Compared to the WPCs man-
wood flours infilling into the voids in the plastic matrices. As a re-
ufactured from SPF and recycled mixed-wood species, the tensile
sult of the addition of polar hydrophilic wood fibers, the moisture
strengths and MOR values of WPCs made from China fir were low-
contents of the resulting WPCs (1.91–2.96%) were also higher than
er. Slaughter [12] reported that the MOR values of WPCs were be-
those of the respective plastic matrices (Table 2).
tween 23 and 60 MPa, and that the MOR value of WPC made from
Table 4 also shows the enhanced mechanical properties of the
maple (higher density) was greater than that of WPC made from
tested WPCs in comparison to those of the respective plastic matri-
pine (lower density). In addition, the compositions of different
ces. The tensile strength of the WPC made from LDPE (30.2 MPa)
wood species may influence the stretching of the ester bond with
was significantly higher than that of LDPE itself (12.0 MPa). How-
MAPP [24]. Due to the different wood fibers in WPCs manufactured
ever, the tensile strengths of the WPCs made from PP (27.2 MPa)
from recycled mixed-wood species and better bonding with MAPP,
and RPP (29.9 MPa) did not significantly differ in comparison to
their mechanical properties were superior.
those of PP (27.0 MPa) and RPP (30.1 MPa). Compared with the
In order to investigate the effects of the particle diameters of
tensile strength of ABS (42.2 MPa, Table 2), a significant decrease
the wood flours on the mechanical properties of WPCs, China fir-
in the tensile strength of the WPC made from ABS (23.0 MPa)
plastic composites with a ratio of wood flour/PP/MAPP/zinc stea-
was found. Observing the MOR values of the tested WPCs, those
rate of 47:47:3:3 were used. Table 3 indicates that there were no
of the WPCs made from LDPE (59.0 MPa), PP (56.8 MPa), and RPP
significant differences in the air dried densities among the tested

Table 3
Effects of species and diameter of wood flours on mechanical properties of WPCs.

Specimen Code Air dried density (g/cm3) Moisture content in air dry (%) Tensile strength (MPa) MOR (MPa)
PPRW 1.10 ± 0.10 2.93 ± 0.08 27.3b ± 0.4 57.1b ± 0.2
PP47 1.07 ± 0.00 3.00 ± 0.02 27.2b ± 0.1 56.8b ± 0.4
PPCW 1.05 ± 0.01 4.37 ± 1.32 26.5b ± 0.2 53.8a ± 0.9
PPCW152 1.05 ± 0.00 2.61 ± 0.07 26.3b ± 0.6 53.2a ± 0.8
PPCW215 1.06 ± 0.00 3.23 ± 0.04 25.2a ± 0.5 53.9a ± 2.4

For codes (PPRW, PP47, PPCW, PPCW152 and PPCW215), see Table 1.
Means (suffixes a and b) within a given column with the same letter do not significantly (p = 0.01) differ as determined by Tukey test.

Table 4
Air dried density, moisture content and mechanical properties of different WPCs made from different plastic matrices and increasing rate of mechanical properties of tested WPCs
in comparison to different plastic matrices.

Specimen Air dried density Moisture content Tensile strength MOR Increasing rate Increasing rate
code (g/cm3) in air dry (%) (MPa) (MPa) of tensile strength of MOR
LDPE 47 1.05 ± 0.01 2.96 ± 0.17 30.2b ± 0.3 59.0c ± 0.4 152% 196%
PP 47 1.05 ± 0.00 1.91 ± 0.61 27.2b ± 0.1 56.8b ± 0.4 0.93% 19.5%
RPP 47 1.06 ± 0.00 2.42 ± 0.02 29.9b ± 1.2 60.1c ± 0.3 0.98% 12.76%
ABS 47 1.17 ± 0.01 2.39 ± 0.03 23.0a ± 3.8 53.8b ± 1.3 45% 20.7%
PEPP 47 1.05 ± 0.00 2.16 ± 0.07 23.2a ± 0.1 47.3a ± 0.5 – –
Commercial WPC 1.15 ± 0.01 1.45 ± 0.05 23.7a ± 0.3 48.8a ± 0.4 – –

For codes (LDPE 47, PP47, RPP 47, ABS 47 and PEPP 47), see Table 1.
Means (suffixes a, b, and c) within a given column with the same letter do not significantly (p = 0.01) differ as determined by Tukey test.
 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496 3493

(60.1 MPa) were significantly higher than those of LDPE 25

(19.9 MPa), PP (47.6 MPa), and RPP (53.3 MPa) (Table 2). Compared PP(230ºC, 10kg)
with the MOR value of ABS (78.0 MPa, Table 2), a significant de- 20 RPP(230ºC, 10kg)
crease in the MOR value of the WPC made from ABS (53.8 MPa) LDPE(190ºC, 10kg)
was found. From the above results, we surmise that the wood fiber ABS(230ºC, 3.8kg)

MI(g/10 min)
15
was especially well-dispersed in the LDPE.
Table 4 also indicates that the mechanical properties of WPCs
10
made from LDPE/PP/wood flour/MAPP/zinc stearate with a ratio
of 23.5:23.5:47:3:3 were inferior to those of WPCs made from
LDPE (or PP)/wood flour/MAPP/zinc stearate with a ratio of 5
47:47:3:3. In other words, use of a mixture of different plastic
matrices was found to have a negative effect on the mechanical 0
30 40 50 60 70
properties of the WPCs.
In comparison to Fig. 1, positive effects on the mechanical prop- Wood flour content (%)
erties of the WPCs as a result of the addition of wood flours are ob- Fig. 4. Effect of wood flour content on melt index values of different WPCs.
served from the dynamic mechanical analysis, as shown in Fig. 3.
This is because the E0 value of the wood flour was larger than that
of the plastic matrix. Compared with the E0 value of the plastic ma- The MI value (200 °C, 5 kg) of ABS is 0.7. Relative to this value,
trix, an approximately tenfold higher E0 value at lower tempera- the MI values of WPCs made from ABS decreased due to the addi-
tures was found for WPCs made from LDPE because of the tion of wood flours. As a result, the mechanical properties of WPCs
rheology of the wood flour. In our study, the infilling of wood flours made from ABS were inferior. Chen [26] reported that the MI value
into plastic matrices was found not to have any enhancing effect on decreased due to poor flow properties of the plastic polymer fol-
the static mechanical properties. However, the wood flours were lowing the addition of wood flour. Couturier et al. [27] pointed
found to affect the dynamic properties of the WPCs. Fig. 3 also re- out that viscosity increased owing to the aggregation of wood fi-
veals a negative effect of using a mixture of different plastic matri- bers during the processing. Furthermore, the specific heat of wood
ces (PP and PE) on the static and dynamic properties of WPCs. An (0.39–0.66 cal/g °C) is greater than that of the plastic matrix
additional peak at 30 °C was found. It is possible that secondary (0.33 cal/g °C). As a result, the flow properties of wood–plastic
crystallization occurred and that this had a negative effect on the composite will be improved at higher temperatures. However,
static and dynamic properties of the WPCs. In comparison to com- the temperature in the manufacturing process of WPCs must not
mercial WPC, the static and dynamic properties of the WPCs pre- exceed 230 °C in order to prevent degradation of the wood flour.
pared in our study were superior. In summary, the MI value decreased and the mechanical properties
For the effects of the addition of wood flours to plastic matrices, of WPCs were inferior to those of the respective plastic matrices.
the results showed the tan d values of WPCs made from LDPE, PP, The relationships between the plastic matrix content and MI value
and RPP decreased in comparison to the values for LDPE, PP, and can be represented by positive linear regression formulae, and the
RPP. However, a contrary result was obtained when the polymer coefficient of determination (r2) values were found to be 0.91–0.99.
matrix was ABS. The results revealed a poor compatibility between For wood flour contents lower than 47%, the MI values of WPCs
the polar hydrophilic wood fiber and the non-polar hydrophobic made from RPP and PP were similar, but significantly higher than
ABS, with weak interfacial adhesion between them. The decrease those of WPCs made from LDPE and ABS (Fig. 4). However, the
in the tan d values was mainly due to an increase in E00 values. By MI values of WPCs made from LDPE, RPP, PP, and ABS were very
adding wood flour to plastic matrices, the E00 values are slightly in- low when the wood flour content was 65.8%. Due to poor flow
creased, but the tan d values are decreased. The tan d values de- properties, the use of a higher wood flour content is clearly not
crease in the order: LDPE > PP > RPP, whereas, the increasing rate conducive to the manufacture of WPCs.
of tensile strength and MOR was (in decreasing order)
LDPE > PP > RPP. 3.4. Effects of wood flour/plastic matrix ratio on mechanical properties
of WPCs

Table 5 presents the mechanical properties of WPCs with differ-

ent wood flour/plastic matrix ratios. The results show that the ten-
sile strengths of various WPCs decreased in the order:
PP42 > PP47 > PP100 > PP66 > PP38 > PP28. However, the specific
tensile strengths decreased with increasing wood flour content
due to weak interfacial adhesion between the wood fibers and
the plastic matrix. Thus, the specific tensile strengths of the various
WPCs decreased in the order: PP100 > PP66 > PP47 >
PP42 > PP38 > PP28. Table 5 also demonstrates that the MOR values
and specific MOR values of the various WPCs decreased in the or-
ders: PP47 > PP38 > PP42 > PP66 > PP28 > PP100 and PP47 >
PP66 > PP100 > PP38 > PP42 > PP28, respectively. Observing the ef-
fect of wood flour content on the E0 values of the WPCs, the increas-
ing percentage of E0 value was evident at higher temperature and
was not evident under 0 °C when the wood flour content was lower
than 28.2% (Fig. 5a). However, the E0 value was 1.5–2 times higher
than that of the plastic matrix when the wood flour content was
higher than 47%. The tan d value was also lower when the wood
Fig. 3. Relationships between temperature and storage modulus (E0 ) of different flour content was higher than 47% (Fig. 5b). This suggests that a
WPCs. wood flour content of 47% could be viewed as optimal for the man-
3494 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496

Table 5
Mechanical properties of WPCs with different wood flour/plastic matrix ratios.

Specimen Code Tensile strength (MPa) MOR (MPa) Specific tensile strength (MPa) Specific MOR (MPa)
PP28 24.0a ± 0.5 48.1a ± 2.7 21.3a ± 1.1 42.9a ± 2.4
PP38 25.6a ± 1.0 55.4cd ± 0.9 23.4a ± 0.9 50.8bc ± 0.8
PP42 27.7bc ± 0.8 52.7bc ± 0.4 26.0b ± 0.8 49.3b ± 0.4
PP47 27.2bc ± 0.1 56.8d ± 0.4 26.0b ± 0.1 54.2d ± 0.4
PP66 25.6ab ± 0.2 50.4ab ± 0.6 27.2b ± 0.2 53.6cd ± 0.7
PP100 27.0bc ± 0.2 47.6a ± 0.4 29.6c ± 1.3 52.3cd ± 0.5

For codes (PP28, PP38, PP42, PP47, PP66, and PP100), see Table 1.
Means (suffixes a, b, c, and d) within a given column with the same letter do not significantly (p = 0.01) differ as determined by Tukey test.

WPCs, their moisture contents increased. The tensile strengths

and MOR values of various WPCs coupled with MAPP decreased
in the orders: NaOH > MA 4.5 > PP 47 > MA 1.5 > MA 9 > MA
6 > MA 0 and NaOH > MA 6 > PP 47 > MA 4.5 > MA 1.5 > MA
9 > MA 0, respectively. Due to the contribution of better interfacial
modification between wood fiber and plastic matrix by the esteri-
fication reaction of MAPP, the mechanical properties of the tested
WPCs were clearly enhanced. Similar results have also been re-
ported by Gatenholm et al. [29], Kazayawoko et al. [30], and Nac-
htigall et al. [31]. In our study, the optimum enhancement of the
mechanical properties of the WPCs was attained when the MAPP
content was 3–4.5%. This is because the interfacial adhesion be-
tween the wood fiber and the plastic matrix became weaker at
higher MAPP contents. Moreover, the mechanical properties of
MAPP are inferior to those of wood fiber. Oksman and Lindberg
[32] also reported the optimum content of SEBS-MA (butylene-sty-
rene triblock butylene-styrene triblock) as an interfacial modifier
to be 4%.
Fig. 6 shows the effect of MAPP content on tan d value of WPC.
The tan d value was highest at lower temperature when MAPP
was not used in the manufacture of the WPC. On incorporating
MAPP in the manufacture of the WPC, the tan d value decreased
and the Tg peak moved slightly to higher temperature. This indi-
cated a better interfacial adhesion between the wood fiber and
the plastic matrix. However, the tan d value did not decrease with
increasing MAPP content. In our study, the lowest tan d value was
obtained by adding 3% MAPP in the manufacture of the WPC. On
adding more than 3% MAPP, the hard segment percentage of the
wood flour was decreased and thus the tan d value increased. Ash-
ida et al. [33] reported a better interfacial adhesion between wood
fiber and plastic matrix with a lower tan d value. In our study, the
Tb and Ta peaks appeared at around 10 °C and 100 °C, respec-
tively. This finding was also reported by Guo et al. [34]. Moreover,
another peak was found at 70 °C when MAPP was added in the
manufacture of WPC.
Using 3% MAPP and alkali-swollen wood flour for manufactur-
ing WPC, the tensile strength and MOR were 60% and 75% higher,
respectively, than those of WPC made from wood flour without
Fig. 5. Effect of wood flour ratio on (a) E0 value (storage modulus), and (b) tan d
NaOH swelling. This is because alkali-swelling led to a higher crys-
value (damping peaks) of WPCs.
tallization area of the wood fiber and a greater area for possible
reaction with MAPP. Consequently, interfacial adhesion between
the wood fiber and plastic matrix was improved. However, alka-
ufacture of WPCs in this study. Huda et al. [28] also reported that li-swelling requires more processing procedures.
the mechanical properties of WPCs were enhanced on increasing In an effort to improve the reaction between wood fiber and
the wood flour content up to 50%. plastic matrix, Rapp et al. [35] impregnated spruce wood samples
with a water-soluble melamine-formaldehyde (MF) resin over se-
3.5. Effect of the MAPP content on tensile strengths and MOR values of ven days. Using electron energy loss spectroscopy, they were able
WPCs to demonstrate penetration of the MF into the secondary cell wall
layers and the middle lamella. By means of UV-microscopy, Gindl
Table 6 indicates the effect of MAPP content on the tensile et al. [36] also found that water-soluble MF diffused well into
strengths and MOR values of WPCs. The results show that the air the secondary cell wall and the middle lamella. Nouwezem et al.
dried density decreased slightly with increasing MAPP content in [37] reported that an ozone pre-treatment renders wood (pine,
the tested WPCs. Due to hydrophilic properties of MAPP in the oak) more reactive to maleic anhydride and epibromohydrin. After
 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496 3495

Table 6
Effect of MAPP ratio on tensile strength and MOR of WPC.

Specimen code Air dried density (g/cm3) Moisture content in air dry (%) Tensile strength (MPa) MOR (MPa)
a
MA 0 1.05 ± 0.01 3.70 ± 0.45 21.1 ± 0.1 37.4a ± 0.4
MA 1.5 1.05 ± 0.00 1.87 ± 0.05 26.3c ± 0.6 48.1b ± 0.4
PP 47 1.05 ± 0.00 1.33 ± 0.12 27.2d ± 0.1 56.8c ± 0.4
MA 4.5 1.05 ± 0.00 1.87 ± 0.02 27.7d ± 0.2 53.8c ± 0.7
MA 6 1.04 ± 0.00 3.22 ± 0.80 24.7b ± 0.8 57.8d ± 1.6
MA 9 1.03 ± 0.01 3.23 ± 0.74 25.3bc ± 0.5 44.9b ± 1.4
NaOH 1.08 ± 0.00 1.34 ± 0.03 33.6e ± 0.9 65.7e ± 2.5

For codes (MA 0, MA 1.5, PP 47, MA 4.5, MA 6, MA 9, and NaOH), see Table 1.
Means (suffixes a, b, c, d, and e) within a given column with the same letter do not significantly (p = 0.01) differ as determined by Tukey test.

the tan d values were lowest when the SPF wood flour (<125 lm)
content was 47%. This suggests that a wood flour content of less
than 50% may be optimal in the manufacture of WPCs.

Acknowledgements

This investigation was supported financially by the National

Science Council of the Republic of China (NSC96-2628-B-002-
026-MY3).

References

[1] Nuñez AJ, Sturm PC, Kenny JM, Aranguren MI, Marcovich NE, Reboredo MM.
Mechanical characterization of polypropylene-wood flour composites. J Appl
Polym Sci 2003;88(6):1420–8.
[2] Nabi ZU, Hashemi S. Influence of short glass fibers and weld lines on the
mechanical properties of injection-moulded acrylonitrile-styrene-acrylate
Fig. 6. Effect of MAPP ratio on tan d (damping peaks) of WPCs. copolymer. J Mater Sci 1999;33(12):2985–3000.
[3] Ismail H, Nizam JM, Abdul Khalil HPS. The effect of a compatibilizer on the
mechanical properties and mass swell of white rice husk ash filled natural
rubber/linear low density polyethylene blends. Polym Test
reticulation of functionalized wood samples, thermosetting sheets 2001;20(2):125–33.
show greatly improved resistance to flexion rupture, higher Rock- [4] Wielage B, Lampke Th, Utschick H, Soergel F. Processing of natural-fibre
reinforced polymers and the resulting dynamic–mechanical properties. J Mater
well hardness, and improved hydrophobic character. Process Technol 2003;139:140–6.
[5] Georgopoulos STh, Tarantili PA, Avgerinos E, Andreopoulos AG, Koukios EG.
Thermoplastic polymers reinforced with fibrous agricultural residues. Polym
4. Conclusions
Degrad Stabil 2005;90(2):303–12.
[6] Coutinho FMB, Costa THS, Suarez JCM, Melo DP. Sawdust reinforced
The effects of material compositions (including different plastic polypropylene composites: a study of fracture behavior. Polym Test
matrices, wood flours, and coupling agents) on the mechanical 2000;19(6):625–33.
[7] Hristov VN, Lach R, Grellmann W. Impact fracture behavior of modified
properties of wood–plastic composites manufactured by injection polypropylene/wood fiber composites. Polym Test 2004;23(5):581–9.
molding have been investigated. Our results have shown that the [8] Karnani R, Krishnan M, Narayan R. Biofiber-reinforced polypropylene
mechanical properties and storage moduli (E0 ) of various plastic composites. Polym Eng Sci 1997;37(2):476–83.
[9] Wolcott MP. Wood–plastic composites. Encyclopedia Mater: Sci Technol 2001.
matrices decrease in the order: ABS100 > RPP100 > [10] Yildiz UC, Yildiz S, Gezer ED. Mechanical properties and decay resistance of
PP100 > LDPE100. Using a ratio of wood flour/plastic matrix/ wood–polymer composites prepared from fast growing species in Turkey.
MAPP/zinc stearate of 47:47:3:3, the tensile strength of the WPC Bioresour Technol 2005;96(9):1003–11.
[11] Clemons CM. Wood–plastic composites in the United States – the interfacing
made from LDPE (30.2 MPa) was significantly higher than that of of two industries. Forest Prod J 2002;52(6):10–8.
LDPE itself (12.0 MPa). However, the tensile strengths of the WPCs [12] Slaughter AE. Design and fatigue of a structural wood–plastic composites.
made from PP (27.2 MPa) and RPP (29.9 MPa) did not significantly Master thesis, Washington State University; 2004. p. 161.
[13] Ticky RJ. Industrial standard for wooden plastic in USA. Wood Industry
differ in comparison to those of PP (27.0 MPa) and RPP (30.1 MPa).
2004;59(8):348–9.
Compared with the tensile strength of ABS (42.2 MPa), a significant [14] Rangaraj S. Durability assessment and modeling of wood-thermoplastic
decrease in the tensile strength of the WPC made from ABS composites. Master thesis, Washington State University; 1999. p. 101.
[15] Chastagner MW. Slit die rheology of LDPE and ABS based wood plastic
(23.0 MPa) was found. Observing the MOR values of the tested
composites. Master thesis, Washington State University; 2005. p. 96.
WPCs, those of the WPCs made from LDPE (59.0 MPa), PP [16] Zawlocki MM. Characterization of wood–plastic composites by dissipated
(56.8 MPa), and RPP (60.1 MPa) were significantly higher than energy. Master thesis, Washington State University; 2005. p. 142.
those of LDPE (19.9 MPa), PP (47.6 MPa), and RPP (53.3 MPa). Com- [17] American Society for Testing and Materials (ASTM). D638. Standard test
method for tensile properties of plastics; 2004.
pared with the MOR value of ABS (78.0 MPa), a significant decrease [18] American Society for Testing and Materials (ASTM). D4761. Standard test
in the MOR value of the WPC made from ABS (53.8 MPa) was found. methods for mechanical properties of lumber and wood-base structural
The tensile strength, MOR, and E0 values of WPCs manufactured material; 2005.
[19] American Society for Testing and Materials (ASTM). D1238. Standard test
from PP mixed with 47% wood flour (<180 lm) and 3–4.5% MAPP method for melt flow rates of thermoplastics by extrusion plastometer; 2004.
were superior to those of the other WPCs manufactured in this [20] Bremner T, Rudin A, Cook DG. Melt flow index values and molecular weight
study. However, the polymer damping peaks (tan d) showed a con- distributions of commercial thermoplastics. J Appl Polym Sci 1990;41(7–
8):1617–27.
trary result. In the manufacture of WPCs from PP, with the contents [21] Nele M, Latado A, Pinto JC. Correlating polymer parameters to the entire
of MAPP and zinc stearate fixed at 3% each, the mechanical proper- molecular weight distribution: application to the melt index. Macromol Mater
ties, specific mechanical properties, and E0 values were highest and Eng 2006;291(3):272–8.
3496 P.-Y. Kuo et al. / Materials and Design 30 (2009) 3489–3496

[22] McGrum NG, Read BE, Williams G. An elastic and dielectric effects in polymeric [30] Kazayawoko M, Balatinecz JJ, Matuana LM. Surface modification and adhesion
solids. London: Wiley; 1967. p. 109. mechanisms in woodfiber–polypropylene composites. J Mater Sci
[23] Mark JE. Polymer data handbook. Oxford University Press; 1999. pp. 782– 1999;34(24):6189–99.
84. [31] Nachtigall SMB, Cerveira GS, Rosa SML. New polymeric-coupling agent for
[24] Kazayawoko M, Balatinecz JJ, Woodhams RT. Diffuse reflectance Fourier polypropylene/wood-flour composites. Polym Test 2007;26(5):619–28.
transform infrared spectra of wood fibers treated with maleated [32] Oksman K, Lindberg H. The influence of a SBS compatibilizer in polyethylene-
polypropylenes. J Appl Polym Sci 1997;66(1):1163–73. wood flour composites. Holzforschung 1998;52(6):661–6.
[25] Stark NM, Matuana LM, Clemons CM. Effect of processing method on surface [33] Ashida M, Noguchi T, Mashimo S. Effect of matrixes type on the dynamic
and weathering characteristics of wood–flour/LDPE composites. J Appl Polym properties for short fiber-elastomer composite. J Appl Polym Sci
Sci 2004;93(3):1021–30. 1985;30(3):1011–21.
[26] Chen J. Effect of rheology and coupling agent on wood plastic composites. [34] Guo C, Song Y, Wang Q, Shen C. Dynamic-mechanical analysis and SEM morphology
Master thesis, Tunghai University; 2000. p. 92. of wood flour/polypropylene composites. J Forest Res 2006;17(4):315–8.
[27] Couturier MF, George K, Schneider MH. Thermophysical properties of wood- [35] Rapp AO, Bestgen H, Adam W, Peek RD. Electron loss spectroscopy (EELS) for
polymer composites. Wood Sci Technol 1996;30(3):179–96. quantification of cell-wall penetration of a melamine resin. Holzforschung
[28] Huda MS, Drzal LT, Mohanty AK, Misra M. Chopped glass and recycled 1999;53:111–7.
newspaper as reinforcement fibers in injection molded poly(lactic acid) (PLA) [36] Gindl W, Zargar-Yaghubi F, Wimmer R. Impregnation of softwood cell walls
composites: a comparative study. Compos Sci Technol 2006;66(11– with melamine-formaldehyde resin. Bioresour Technol 2003;87(3):325–30.
12):1813–24. [37] Nouwezem R, Borredon ME, Parisi JP, Gaset A. Improvement of thermosetting
[29] Gatenholm P, Kubat J, Mathiasson A. Biodegradable natural composites 1. properties of lignocellulosic waste by treatment with ozone. Bioresour Technol
Processing and properties. J Appl Polym Sci 1992;45(9):1667–77. 1993;45(1):43–6.