Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

Advances and Challenges of Wood Polymer Composites

Roger M. Rowell

USDA, Forest Service, Forest Products Laboratory and Biological Systems Engineering,
University of Wisconsin, Madison, WI

ABSTRACT

Wood flour and fiber have been blended with thermoplastic such as polyethylene,
polypropylene, polylactic acid and polyvinyl chloride to form wood plastic composites (WPC).
WPCs have seen a large growth in the United States in recent years mainly in the residential
decking market with the removal of CCA treated wood decking from residential markets.
While there are many successes to report in WPCs, there are still some issues that need to
be addressed before this technology will reach its full potential. This technology involves two
different types of materials: one hygroscopic (biomass) and one hydrophobic (plastic) so
there are issues of phase separation and compatibilization. This technology also involves
two very different industries; wood and/or agricultural industries and plastic industries.
Processing the two phases presents issues centered on maximizing mixing while minimizing
damage to the biomass furnish. Melt flow index, processing temperatures, static electricity
and density are also issues for the plastic industry which is used to high flowing high
temperature processing conditions necessary when they either extrude plastics neat or use
inorganic fillers such as glass, calcium carbonate or talc. While the biomass in the
composite swells and shrinks due to moisture, the plastic phase swells and shrinks due to
temperature. Freezing and thawing cycles also cause problems as well as the effects of
microorganism attack and ultraviolet radiation. Dyeing, refinishing, color stability and
fasteners are also concerns. Solid extrusion, profile extrusion, co-extrusion, injection
molding and thermal molding are all being used to made different WPC products in different
parts of the world.

INTRODUCTION

The use of wood flour or fiber as fillers and reinforcements in thermoplastics has been
gaining acceptance in commodity plastics applications in the past few years. It is
interesting to point out that the use of the lignocellulosics in commodity thermoplastics to
reduce cost and/or to improve mechanical performance is not new, and there are plenty of
published papers, including patents dating back to the 1960's and 1970's. The resurgence
of interest in the 1990's is probably due to increasing plastic costs and environmental
aspects of using renewable materials.

The advantages of using a wood component in thermoplastic composites is that the

biobased resource is non abrasive, low in cost, widely available, sustainable, high filling
levels possible, high specific properties, lower density per weight of raw material, flexible and
recyclable (Caulfield et al. 2005). The physical form can vary from fine wood flour to wood
fibers (Jacobson et al. 1995). The cost and performance of the final composite products
dictates the form of the wood to be used in the plastics. The typical plastics used in this
technology are various types of polyethylene, polypropylene, and polyvinyl chloride.

Wood filled thermoplastics are being used in the auto industry, residential products, and
many other applications with growing markets. Smith and Walcott (2006) have pointed out
four factors that are driving the WPC decking market: quality/cost ratio, impact of treating

2
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

chemicals in the environment, builder acceptance, and effective marketing. There are,
however, several major challenges that remain in this field that need to be solved including
processing, use of additives, properties, and outdoor performance.

PROCESSING

Typical blending of wood with a thermoplastic involves the plastic-fillerheinforcement to be

shear mixed at temperatures above the softening point of the plastics. The heated mixture
is then typically extruded into "small rods" that are then cut into short lengths to produce a
conventional pellet. The pellets can then be used in typical injection or compression
molding processes.

Processing the wood without loss of fiber properties is a major issue. Table 1 shows the
loss of fiber length width and aspect ratio before and after processing oil palm fiber in a
Brabender Plastic Corder with polypropylene at 180 C. No matter what fiber length or width
is used before processing, the fiber loses length and width during the processing. Many of
the applications today use a biomass flour (aspect ration less than 10) rather than a fiber
(aspect ratio greater than 10) so degradation of fiber length is not an issue. For using
longer fibers, new technologies for mixing are needed to insure an even distribution of
biomass in the thermoplastic matrix without damage to the fiber.

The processing temperatures used in the plastic industry to are usually above the
decomposition temperature of the biomass materials. The plastic industry either does not
use fillers or fillers such as glass, calcium carbonate or talc fillers so processing
temperature can be very high to increase output. Temperatures above 180 C result in
decomposition of wood so processing temperatures lower that 180 C are needed if a wood
material is used as a filer. Thermoplastics with a high melt flow index and low softening
temperatures are usually used.

Table 1. Fiber length of oil palm before and after processing in a Brabender Plastic Corder
with polypropylene at 180 C
Average Fiber Length
Before Processing After Processing
Length Width Length Width
Aspect Ratio Aspect Ratio
(mm) (mm) (mm) (mm)
1.51 0.18 9.6 0.72 0.13 6.2
2.76 0.30 9.9 0.82 0.14 6.2
6.80 0.87 9.2 0.85 0.14 6.2

Static electricity and density are also issues for the plastic industry that is used to high
flowing high temperature processing conditions used when they either extrude plastics neat
or use inorganics as fillers. The density of wood materials is much lower than the inorganic
fillers used and feeding an extruder can be a problem. Fibers are usually fed into the
extruder in one of two ways. Pellets can be made of the wood material and fed into the
extruder along with the plastic pellets. A side stuffer can be added to the extruder line and
the fiber or flour can be added at that point. If a lose fiber or flour is used, there is usually a
build up of static electricity that has to be dealt with.

To reduce the cost of this blending process, direct injection molding can be done. The
direct injection molding process probably has limitations on the amount of filler/fiber that
can be used in the composite, and is also likely to be limited to particulate or shorter fiber.
The chemical characteristics of the surface and bulk properties of the wood are also

3
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

important in the blending with plastics. The ability of the matrix of the lignocellulosic
(hemicellulose and lignin) to soften at plastic processing temperatures may give these wood
materials unique characteristics to develop novel processing techniques.

Wood fiber-thermoplastic composites can be produced by extrusion, injection molding, or

thermoforming (compression molding). All three methods produce composites with
different properties. Figure 1, for example, shows the moisture sorption of extruded,
compression molded and injected molded wood flour-high density polyethylene (HDPE) as
compared to the moisture sorption of solid wood (Clemons 2002). The composites and
wood were placed in a constant 65% relative humidity room (27 C) and the weight gain
measured as a function of time. Solid wood sorbs moisture the fastest and reaches
equilibrium (9%) in about 10 days. Compression molded composites reach equilibrium
(4.1%) in about 5 weeks, extruded reach equilibrium (3.6%) in 9 weeks and injection
molded had a moisture content of less than 2% and had not reached equilibrium at the end
of the 9 week test. Moisture content as a function of time is an indication of the amount of
wood at or near the surface of the composite allowing moisture to be wicked into the
interior. Injection molded composites have a surface rich in thermoplastic that surrounds
and protects the wood from coming into contact with moisture.

Figure 1. Moisture sorption of highdensity polyethylene containing 50% wood flour injected molded,
extruded or compression molded and solid wood for comparison (Clemons 2002).

Figure 2. Profile extrusion versus solid extrusion

4
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

Figure 2 shows two different processing strategies for wood-thermoplastic products: profile
versus solid. More products can be made with the same quantity of material using profile
extrusion and the products are lighter and can be stronger. Figure 3 shows an example of
co-extrusion where a coating of vinyl is co-extruded over a wood fiber-thermoplastic core.

To reduce the weight and density of wood-thermoplastic composites. foaming or blowing

agents can be dissolved or finely dispersed in the thermoplastic before processing (Guo et
al. 2004). A large number of bubble nuclei are generated and uniformly dispersed in the
polymer matrix resulting in a light-weight composite when combined with a wood fiber.

Equipment development is going forward to produce more efficient mixing systems that
minimize fiber damage, faster processing through put, and new methods of product
formation.

Figure 3. Co-extrusion of a fiber-thermoplastic core with a vinyl surface

USE OF ADDITIVES

Blending of the plastics with the wood materials may require the use of many different types
of additives. Compatibilizers to improve the dispersion, flow and mechanical properties of
the composite are used when increased performance is required. Many of the applications
of wood materials in thermoplastics do not require a compatibilizer to improve mechanical
properties but other chemicals may be added to improve UV stability and/or stability to heat
or antimicrobials. antioxidants, colorants. or to improve impact strength. Processing aids or
lubricants are almost always used. Foaming agents can also be added to reduce the
density of the final product.

The most common compatibilizer used today is either maleic anhydride grafted
polypropylene (MAPP) or maleic anhydride grafted polyethylene (MAPE). The mechanism
of compatibilization is shown in Figures 4 and 5. First, the anhydride end of the MAPP or
MAPE reacts with a hydroxyl group on the surface of the wood material forming an ester
linkage. The PP or PE tail on the grafted MAPP or MAPE then become entangled with the

5
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

melted thermoplastic in the melt resulting in a mechanical link between the hydrophilic
lignocellulosic and the hydrophobic thermoplastic.

Improvements in mechanical properties using a compatibilizer are shown in Figure 6. The

bottom curve in the stress strain curve is PP alone. The dotted line is the curve for
non-compatibilized PP with 50% kenaf loading and the next to the top curve is for
compatibilized PP with 50% kenaf.

Figure 4. Mechanism of MAPP reacting with a lignocellulosic surface hydroxyl group

Figure 5. Entanglement of melted polypropylene with the reacted MAPP/lignocellulosic

Colorants are commonly used both for aesthetics and UV protection. For a consistent color
matching of the wood materials and the thermoplastic, it may be necessary to add the
colorant to both materials separately before extrusion. If the wood fiber is very light in color
and the thermoplastic dark, the composite will have light specks through out the composite.

Light stability in outdoor exposures is an area of considerable investigation (Rowell et al.

2000, Lundin 2001). Most WPCs tend to lighten over time (Falk et al. 2002). Some
manufacturers add pigments to slow this effect. Others add a gray pigment so that color
change is less noticeable, Still others co-extrude a UV-stable plastic layer over the WPCs.

6
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

In laboratory weatherometer tests, wood-PP composites lightened within 400 to 600 hours
of UV exposure. After 2000 hours of UV and water exposure, specimens were white and
the white color continued throughout the 10 mm specimen. This in-depth color change may
be due to stabilization of the free radicals formed from UV energy by the lignin in the wood
allowing the free radicals to penetrate deeper into the wood structure (Rowell et al. 2000).

The fire performance of WPC materials and products is just beginning to be investigated
(Malvar et al. 2001, Stark et al. 1997). These composites are different from many building
materials in that they can melt as well as burn, making testing for fire resistance difficult.

Figure 6. Kenaf filled polypropylene at different filling levels with and without MAPP

PROPERTIES

The properties of wood-thermoplastic composites depend on the volume fraction of wood in

the composite as well as processing temperature, additives, and type of thermoplastic used.
Table 2 shows some mechanical properties of aspen fiber-polypropylene composites varying
the volume fraction of the fiber with and without a compatibilizer (MAPP) (Sanadi et al.
1994). Flextural and tensile strength and modulus increase with increasing fiber content.
The fiber volume fraction has little effect on notched lzod but unnotched lzod goes down as
the fiber content increases. The addition of the compatibilizer (2%) increases flextural and
tensile strength and modulus but has little effect on notched lzod values.

OUTDOOR PERFORMANCE

The proof of any material is in its performance. Most of the applications for fiber-
thermoplastic composites have been for outdoor materials such as residential decking and
railing systems, windows, siding, roofing, door stiles, moldings, pallets, guard rails, sound
barriers, curbing and furniture (Clemons 2002, Smith and Wolcott 2006).

Outdoor applications of wood-thermoplastic composites require the materials to stand up to

a wide variety of environmental factors including moisture, UV radiation, high temperatures,
freeze-thaw cycles, mechanical action, and biological invasion. In almost all cases, more
than one of these factors are acting at the same time and in cycles. A residential deck
might be subject to all of these factors in a single year or less. Swelling, warping, twisting,
mold growth, decay, color changes, and connector failure are some of the problems often

7
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

encountered from normal outdoor use. For this reason, there has been a lot of
environmental testing done both under laboratory conditions and outdoor in/above ground
testing.

Table 3 shows the sorption of moisture as a function of time. Aspen fiber-polypropylene

specimens with varying volume fractions of fiber with and without a compatibilizer (MAPP)
were placed in a 90% RH room at 27 C and the increase in weight measured over time. As
the volume fraction of the fiber increases, the moisture sorption increases and is still
increasing after 200 days. There is a slight decrease in moisture sorption with the addition
of MAPP indicating that the compatibilizer may allow the PP to come into closer contact
with the fiber.
Table 2. Mechanical properties of aspen fiber-polypropylene composites

Flexural Flexural Tensile Tensile lzod lzod

Specimen strength modulus strength modulus notched unnotched
(MPa) (GPa) (MPa) (GPa) (J/m) (J/m)
PP 27.9 1.38 26.2 1.69 22.4 713.5
PP/MAPP 34.6 1.79 29.3 1.82 18.6 563.3
30A/70PP 49.5 4.12 29.3 4.52 24.8 101.7
30A/68PP/MAPP 60.2 3.82 44.9 4.10 21.1 128.3
40A/60PP 54.6 4.60 34.9 5.22 19.6 85.5
40A/58PP/MAPP 66.4 4.66 47.7 5.14 19.8 108.7
50A/50PP 50.2 5.48 28.4 5.81 26.4 67.1
50A/48PP/MAPP 75.7 5.88 53.1 6.68 21.9 98.5
60A/40PP 45.9 6.09 25.6 6.95 23.9 55.2
60A/38PP/MAPP 75.8 6.73 48.1 7.19 21.3 81.1

Table 3. Moisture sorption of aspen fiber- polypropylene composites

Weight Gain (%)

Aspen/PP/MAPP
25 D 50 D 75 D 100 D 150 D 200 D
0/100/0 0 0 0 0 0.2 0.4
30/70/0 0.7 1.4 1.7 2.1 2.4 2.8
30/68/2 0.7 0.7 1.1 1.5 1.5 2.2
40/60/0 0.7 1.4 1.7 2.0 2.4 3.0
40/58/2 0.4 1.2 1.5 1.9 2.7 3.5
50/50/0 1.3 2.0 2.6 3.6 4.3 5.3
50/48/2 1.5 1.8 2.2 2.9 4.0 5.1
60/40/0 3.7 4.5 5.6 6.0 6.3 6.7
60/38/2 1.6 2.2 3.5 4.4 5.1 6.0

Figure 7 shows the weight change in an oven dry to water soaked test for 60% aspen fiber
compounded with PP with and without MAPP. Each cycle represents 7 days of water
soaking and one day of oven drying at 105 C. The specimens without MAPP continue to
gain weight with each water soaking cycle but there is less change in weight in the
specimens containing MAPP.

Table 4 shows the influence of moisture on decay. Aspen fiber-polypropylene composites

with various volume fractions of fiber were subjected to a12 week ASTM standard soil block
test using the brown-rot fungus Gloeophyllum trabeum or the white-rot fungus Trametes

8
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

versicolor. Half of the specimens were places in test directly from injection molding and half
were soaked in water for 20 days before placing in test. The dry specimens were not
degraded by either fungus but the water soaked specimens were attacked by both. The
weight loss due to fungal attack increased with increasing fiber fraction. The compatibilizer
in each fraction set did reduce the fungal attack.

Figure 8 shows a railing in a state park in the Oregon made of wood-thermoplastic material.
Several types of micro-organisms are present growing on the surface of the composite.
The decking and rails are also warped due to thermal expansion.

Figure 7. Change in weight during a cyclic oven dried to water soaking of aspen fiber-polypropylene
composites 60% aspen fiber. 40% PP 60% aspen fiber, 38% PP, 2% MAPP

Figure 8. Growth of several types of organisms on plastic lumber in an outdoor exposure in Oregon

Figure 9 shows a badly warped deck in The Bahamas. The deck had also faded from its
original dark green color due to UV exposure. Table 5 shows laboratory data on loss in
flexural strength after 1000 and 2000 hours of UV exposure (N.M. Stark 2005). UV
exposure results in both strength loss and color fading.

9
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

There is also research going on to improve connections/bonding between fiber-

thermoplastic composite components, develop coating systems to change colors or
improve the degraded aesthetics of the products. and to improve all aspects of performance
in outdoor applications of wood-thermoplastic materials.

Table 4. ASTM 12 week soil block test on dry and water soaked aspen fiber-polypropylene composites

Composition Tested dry Tested after water soaking

G. trabeum T. versicolor G. trabeum T. versicolor
% wt loss % wt loss % wt loss % wt loss
30/70 0.2 0 2.4 1.3
30/68/2 0.2 0 1.6 0.3
40/60 0.3 0 3.6 2.1
40/58/2 0.2 0 3.3 1.8
50/50 0.3 0 4.6 2.3
50/48/2 0.2 0 5.3 3.0
60/40 0.8 0 4.7 2.0
60/38/2 0.6 0 5.9 1.7

Table 5. Loss of flexural strength due to UV exposure of a 50% wood flour 50% PE composite

Flexural Strength
Before exposure After 1000 hours After 2000 hours
40.2 37.9 (-2.3%) 31.2 (-9.0%)
N.M. Stark (2005)

Figure 9. Warping due to thermal expansion of a plastic lumber deck in The Bahamas

FUTURE CHALLENGES

There is still a great deal of research and development going on in many parts of the world in
the area of natural fiber/thermoplastic composites. Challenges in this research includes
developing better mixing systems that minimize damage to the wood fibers, developing
better compatibilization systems. developing faster and better extrusion/injecting/
10
Proceedings of the 8th Pacific Rim Bio-Based Composites Symposium

compression systems, improving impact strength, improving the properties of the wood
fraction through chemical, enzymatic and cold plasma modification to improve outdoor
performance, advancing new profile and co-extrusions systems, developing more intensive
testing procedures, and understanding performance through modeling.

REFERENCES

Caulfield, D.F., C. Clemons, R.E. Jacobson and R.M. Rowell 2005. Wood-thermoplastic
composites. In: Handbook of wood chemistry and wood composites, R.M. Rowell, ed,
Taylor and Francis, Boca Raton, FL, Chapter 13, 365-380.

Clemons, C. 2002. Wood-plastic composites in the United States: The interfacing of two
industries. Forest Prod. J. 52(6): 10-18.

Falk. R.H., Lundin. T., and Felton, C. 2002. Accelerated weathering of natural fiber-
thermoplastic composites: Effects of ultraviolet exposure on bending strength and
stiffness. Proceedings, Sixth lnternational Conference on Woodfiber-Plastic
Composites. Forest Products Society, Madison, WI, pp. 87-93.

Guo, G., G.M. Rizvi, C.B. Park and W.S. Lin 2004. Critical processing temperature in the
manufacture of fine-celled plastic/wood-fiber composite foams. J. Applied Polymer
Science 91: 621-629.

Jacobson, R.E., R. Rowell, D. Caulfield, and A.R. Sanadi 1995. United States based
agricultural "waste products" as fillers in a polypropylene homopolymer. Proceedings,
Second Biomass Conference of the Americas: Energy, Environment, Agriculture, and
Industry. August, Portland, OR, NREL/CP-200-8098. DE95009230, 1219-1227.

Lundin, T. 2001. Effect of accelerated weathering on the physical and mechanical properties
of natural-fiber thermoplastic composites. M.S. thesis. University of Wisconsin,
Madison.

Malvar, L.J., Pendleton, D.E., and Tichy, R. 2001. Fire issues in engineered wood
composites for naval waterfront facilities. Sampe J. 37(4):70-75.

Rowell, R.M.. Lange, S.E., and Jacobson, R.E. 2000. Weathering performance of plant-fiber
thermoplastic composites. Mol. Cryst. and Liquid Cryst. 353:85-94.

Sanadi, A.R., D.F. Caulfield, R.E. Jacobson, and R.M. Rowell 1994. Reinforcing
polypropylene with agricultural fibers. Proceedings: International Jute and Allied Fibre
Symposium on Biocomposites and Blends, New Dehli, India, p. 163-167.

Smith, P.M. and M.P. Walcott 2006. Opportunities for wood/natural fiber-plastic composites
in residential and industrial applications. Forest Prod. J. 56(3): 4-11, 2006.

Stark, N.M. Personal communication 2005.

Stark, N.M., White, R.H., and Clemons, C.M. 1997. Heat release rate of wood-plastic
composites. Sampe J. 33(5):26-31.

11
 Proceedings of the
8TH PACIFIC RIM BIO-BASED COMPOSITES
SYMPOSIUM
“Advances and Challenges in Biocomposites”

20 - 23 November 2006
Kuala Lumpur, Malaysia

Editors

Mohd Nor Mohd Yusoff

Koh Mok Poh
Mohd Dahlan Jantan
Wan Asma lbrahim
Mahmudin Saleh
Rafeadah Rusli
Rokiah Hashim
Jamaludin Kasim
Paridah Md Tahir
Sarani Zakaria

Forest Research Institute Malaysia

52109 Kepong, Selangor Darul Ehsan, Malaysia

November 2006