Compatibilizer selection to improve mechanical and moisture properties of extruded wood-HDPE...

Abstract

Copolymer additives can be useful to enhance the compatibility and interfacial adhesion between polar wood materials and nonpolar polyolefin. This improved interaction among materials can lead to improved performance attributes, especially for those properties influenced by the

interphase region. Although considerable scientific research has been conducted on compatibilizers for wood-polypropylene composites, much less work has addressed commercially viable formulations of wood and polyethylene. A literature review and comparative engineering analysis of the efficiency of different compatibilizers was conducted and used to guide material selection. Application methods, possible adhesion mechanisms, and performance are summarized. In this research, the efficacy of various forms of maleated polyethylene, maleated polypropylene and ethylene acrylic acid copolymer in a commercial-like extrusion process for wood flour-HDPE composites is presented. The various types of wood-HDPE formulations were extruded using a 55-mm conical twin-screw extruder with intermeshing counter-rotating screws. Mechanical properties of these composites were evaluated by static 3-point bending. The durability issues of the extruded composites related to prolonged exposure to moisture were also examined. The physical and mechanical properties of the extruded composites are significantly improved by the use of maleated polyethylene, MAPE-575 and maleated polypropylene, MAPP-950 copolymers. The effect of prolonged moisture exposure is also significant. About 20 to 30 percent reduction in strength was observed for various types of extruded composites.

**********

In the last decade, the use of wood materials as reinforcing fillers for thermoplastics has received considerable attention because wood materials have very attractive properties such as low density, high specific strength and modulus, low cost, renewable, biodegradable, and damage tolerance when compared with other reinforcing fibers (Czarenecki and White 1980, Dalvag et al. 1985, Sanschagrin et al. 1988). However, poor interfacial interaction limit both strength and moisture properties (Klason et al. 1984). Different chemical additives have been investigated to improve the compatibility between wood and thermoplastics (Woodhams et al. 1984, Dalvag et al. 1985, Maldas et al. 1989, Felix and Gatenholm 1991, Sanadi et al. 1994). In order to enhance the compatibility between wood materials and thermoplastics, recognition of better compatibilizers is a very important step to improve interfacial adhesion and thereby to develop an effective interface structure, which will improve physical and mechanical properties of the wood-thermoplastic composites.

Polyolefins are commonly used in wood-thermoplastic composites because their processing temperatures are below the degradation temperature (200 to 210[degrees]C) of the lignocellulosic materials. The typical commodity thermoplastics studied include polystyrene (PS), polyvinylchloride (PVC), polypropylene (PP), low-density polyethylene (LDPE), and high-density polyethylene (HDPE), with PP being the most common. However, the marginal processing temperature of PP (190 to 200[degrees]C), limits commercial use because when improperly processed it may harm the wood component. In contrast, the processing temperature of polyethylene (PE) is well below the degradation temperature of the lignocellulosic materials. PE is also widely available as recycled postconsumer products such as blow-molded containers and film stock. In fact, Mustafa and Balatinecz (1996) reported that PE comprises more than 70 percent of total plastic waste. These attributes attribute to PE's commercial success and suitability for wood-thermoplastic composites.

Maleated polypropylene (MAPP) has been well established as a compatibilizer for wood-PP composites (Woodhams et al. 1984, Kishi et al. 1988, Krzysik et al. 1991, Myers et al. 1991, Olsen 1991, Sanadi et al. 1994). Unlike polypropylene, useful compatibilizers for polyethylene are less documented. In addition, the current research addresses neither the component interactions that can commonly occur in multicomponent, commercial formulations, nor does it investigate the behavior of these components in processes that are similar to those used by commercial producers. This work is aimed at the engineering of compatibilized wood-HDPE composites. Therefore, the specific objectives of this paper are: (1) to review and analyze the engineering efficiency of the different compatibilizers for wood-HDPE composites, (2) to determine the influence of selected compatibilizers on the mechanical properties of extruded wood flour-HDPE composites using commercial methods, and (3) to investigate the effect of prolonged water exposure on durability of the extruded composites.

Review and comparative engineering analysis of the efficiency of compatibilizer systems

From literature it was found that the following compatibilizers were used for wood-PE composites. Their application methods, bonding mechanism, and performance are summarized below:

Organic peroxides

Peroxide can potentially initiate free radical reactions between cellulosic fibers and polyethylene matrix systems. The organic peroxides used to enhance interfacial adhesion between wood fibers and PE are benzoyl peroxide (BPO) and dicumyl peroxide (DCP).

Both BPO and DCP have been used as a pretreatment for wood fibers (Bataille et al. 1990) and matrix (Sapieha et al. 1990). In addition, they have been studied by direct addition to the wood-thermoplastic blend (Cousin et al. 1989, Bataille et al. 1990). It was found that direct mixing of peroxide is more effective than the fiber/matrix pretreatment (Bataille et al. 1990). Direct mixing of < 0.08 percent peroxide increased the yield stress of the LLDPE/cellulose composites by approximately 70 percent. In contrast, the HDPE/cellulose composites increased by only 15 percent. DCP is also found to be more effective than BPO at low peroxide levels.

Anhydrides

Esterification of wood materials with anhydrides provides a certain degree of hydrophobicity by blocking the hydroxyl groups. To accomplish this modification, an anhydride reacts with the hydroxyl group of wood materials to form an ester bond (Felix and Gatenholm 1991) and the [pi]-component of the unsaturated C=C may open and form new C-C bond with the thermoplastic backbone. Typical anhydrides are acetic anhydride (AA), maleic anhydride (MA), succinic anhydride (SA), and phthalic anhydride (PA). Anhydrides also grafted onto the thermoplastics backbone to impart polarity. Peroxide is usually used as initiator to graft anhydride onto the polymer backbone.

Raj et al. (1990a) and Raj and Kokta (1995a) precoated the wood fibers (CTMP aspen fibers) with MA (5% and 10% by weight of fiber), LLDPE/HDPE (5%), and a free-radical initiator (di-butyl peroxide) in a roll mill at 160[degrees]C. Precoating of fiber with MA, PE in presence of peroxide may produce esterified wood fiber, MA grafted PE and MAPE grafted wood fiber. Using fiber loadings between 10 percent and 40 percent, the precoated fibers were compounded with LLDPE/HDPE in an extruder. After grinding the compounded materials, dog-bone samples (tensile specimens) were compression molded to study the mechanical properties. However, they observed negligible improvement in strength. Raj and Kokta (1995a) reported that PA performed slightly better than MA for wood fiber-HDPE composites. Maldas and Kokta (1991) treated the wood fiber with MA (0% to 3% by weight of fiber), HDPE (5% by weight of fiber) and BPO/DCP (0% to 1% by weight of fiber). The MA-HDPE pretreated fibers showed a positive response with regard to mechanical properties of the compression molded wood fiber-HDPE composite specimens. They observed the performance of DCP is better than BPO as an initiator.

Maleated copolymers

As discussed in the previous section, precoating of wood fiber with MA and polyolefin in the presence of peroxide may produced maleated polyolefin and maleated polyolefin grafted wood fiber. However, commercially premanufactured maleated polyolefins are available in the market. Some research works also investigated premanufactured maleated polyolefins as compatibilizer in wood-thermoplastic composites.

Maleated polyolefins.--Maleated polyolefins contain bifunctional anhydrides and acids groups. The maleated polyolefin reacts with the hydroxyl groups of the wood materials to form an ester linkage and hydrogen bonding. Thus the MA portion of the maleated polyolefin is providing chemical covalent bonding and polar interaction with the wood materials. Maleated polyolefins differ by the MA content and distribution of MA groups along the polyolefin chain and also differ by the length/molecular weight of the polyolefin chain. Maleated polyolefins may also contain unbound oligomeric maleic anhydride and free maleic anhydride (Scott 2001). The polyolefin chain/backbone of the maleated polyolefins are speculated to provide chain entanglement with the thermoplastic matrix (Sanadi et al. 1997). Both maleated polypropylene (MAPP) and maleated polyethylene (MAPE) are commonly used.

Maleated polypropylene.--Using of MAPP as a compatibilizer was first reported by Dalvag et al. (1985), who added 2 to 6 percent MAPP (based on filler content) (Hercoprime A-35, supplied by Hercules) directly in the compounding process. Tensile test bars were subsequently formed by injection molding from the granulated compound. The strength of the wood fiber-HDPE composites increased approximately 27 percent. Maximum improvement of strength was observed with the addition of 3 percent MAPP (based on the filler content). Further addition of MAPP does not result in further improvements. Selke and Childress (1993) mixed MAPP (1% to 5% by weight of matrix) (Hercoprime, supplied by Himont) with the HDPE and wood fiber using a 38-mm corotating twin-screw extruder and observed approximately 50 percent improvement in tensile strength of subsequently compression molded wood fiber-HDPE composites. In addition, a 2-hour boiling water immersion procedure showed significant reduction in water sorption. Raj et at. (1989a) pretreated the wood fiber with MAPP (Epolene E 43, supplied by Eastman Chemical) before mixing it with LLDPE in the extruder. Again, compression-molded tensile specimens showed approximately 50 percent increase in strength.

Maleated polyethylene.--Sanadi et al. (1992) observed the improvement of the interfacial adhesion between wood and PE using a MAPE pretreatment of the wood. The maximum shear strength of a pull-out test was measured using a 2.14 mm diameter wood dowel. An approximate increase of 35 to 40 percent in shear stress was reported when the wood dowel was pretreated with a MAPE emulsion solution (1/50 v/v in water) for 15 minutes. In a separate study, pretreatment of maple wood fiber with MAPE (Fusabond[TM] D100, 10 percent by weight of fiber) enhanced the tensile strength of compression molded wood fiber-recycled PE composites (Maldas and Kokta 1994). MAPE also improved the mechanical properties of the kenaf fiber-recycled PE composites (Maldas and Kokta 1995). Balatinecz et al. (1999) used MAPE as interfacial compatibilizer in HDPE hybrid composites with wood flour and coal ash reinforcement. Khavkine et al. (2000) also used MAPE (Fusabond MB-110D, supplied by Dupont Canada Inc.) in wood flour-HDPE composites to investigate the effect of 2 hours water boiling, 5 boiling and freezing cycles, and fungal exposure on the composites. However, comparisons to composites produced without compatibilizers were not made.

Maleated Elastomer.--The linkage between the wood fibers and maleated elastomer (maleated styrene-ethylene/ butylenes-styrene, SEBS-MA) is almost similar to that of maleated polyolefins. The anhydride groups form ester linkage and hydrogen bond with the hydroxyl groups of the wood fibers. Oksman and Lindberg (1998) found that addition of 4 percent SEBS-MA triblock copolymer to wood flour-recycled LDPE compounded pellets increased the mechanical properties of injection-molded specimens by approximately 30 percent. However, tensile modulus decreased by about 37 percent. Unlike other maleated polyolefins, a gradual decrease in tensile strength was observed with further addition of SEBS-MA up to a level 10 percent.

Isocyanates

The isocyanate group (-N=C=O) is potentially reactive with the hydroxyl groups (-OH) present on the many of the wood polymers. The surface modified wood materials can react with the thermoplastics (free radical may create due to high temperature processing/compounding) (Raj et al. 1989a). Isocyanates that have been used as a compatibilizer between wood and PE are polymethylenepolyphenyl isocyanate (PMPPIC) (Raj et al. 1989a, Raj et al. 1990a, Raj et al. 1990b, Raj and Kokta 1995b), Toulene-2-4-diisocyanate (TDIC), and 1-6 hexamethylene diisocyanate (HMDIC) (Raj et al. 1990a).

Raj et al. (1989a, 1990a, 1990b) pretreated the wood fiber/ flour with PMPPIC (3% by weight of fiber) and a small amount of polymer (HDPE/LLDPE, 5%) in a roll mill at 150 to 160[degrees]C. The treated wood materials were compounded with HDPE/LLDPE in an extruder and necked tensile specimens were formed by compression molding. Pretreatment of both CTMP aspen fiber and bagasse fiber (Raj and Kokta 1995b) with PMPPIC increased the tensile strength of the compression molded HDPE composites about 112 percent and 48 percent, respectively. PMPPIC was found to be the most effective of the isocyanates studied (Raj et al. 1990a). In addition, the efficacy of PMPPIC was more pronounced when applied to the fiber as pretreatment rather than mixing with the polymer or directly mixed with the wood flour and polymer.

Silanes

The -Si[(OR).sub.3] portion of the silane, Y-[(C[H.sub.2]).sub.n]-Si[(OR).sub.3] or its hydrolyst product -Si[(OH).sub.3], caused by the moisture present in wood fiber, reacts with the hydroxyl groups of the wood polymers to form a covalent or hydrogen bond. The Y functionality (e.g., amino, mathacryloxy, vinyl) of the silane is selected to interact with the polymer matrix through formation of covalent bonds, hydrogen bonds, Van der Waals forces or acid-base interaction (Maldas et al. 1989).

The silanes used for wood-PE composites are vinyltri(2-methoxyethoxy) silane (Silane A-172, Union Carbide), [gamma]-methacryloxy-propyltrimethoxy silane (Silane A-174, Union Carbide), and [gamma]-amino-propyl-triethoxy silane (Silane A-1100). Wood fibers or flours were pretreated with silane in carbon tetrachloride solution (Raj et al. 1989a, Raj et al. 1989b, Oksman and Lindberg 1998). The amount of silane used was 1 to 4 percent (by weight of fiber). A small amount (0% to 5%) of dicumyl peroxide was also used to promote the polymerization of vinyl groups in the case of vinyl silanes. Silane (A-172 and A-174) pretreated wood flour showed a higher tensile strength of the compression molded LLDPE/ HDPE-wood flour composites (Raj et al. 1989a, Raj et al. 1989b, Raj et al. 1990b). As with the isocyanates, the effectiveness of silanes is more pronounced for LLDPE system than HDPE. The performance of the silane A-172 is slightly better than the silane A-174 (Raj et al. 1989b). In general, the effectiveness of silanes was observed to be less than that of maleated polyolefins and PMPPIC (Raj et al. 1989a, Raj et al. 1990b).

Monomers

Monomers are generally used to graft polymer chains to wood polymers by covalent bond. The grafted polymer chains act as a compatibilizer to enhance the interfacial adhesion. Liao et al. (1997) grafted acrylonitrile on wood fiber. Compression molded grafted wood fiber-LLDPE composites showed higher tensile strength compared to ungrafted wood fiber-LLDPE composites. They also observed that tensile strength of the composites increased with the increase of grafting ratio of the wood fiber. Modulus was found remained unchanged by grafting.

Titanates

Titanium di(dioctylpyrophosphate)oxyacetate (KR 138S, supplied by Kenrich PI) (Dalvag et al. 1985), TC-POT and TC-PBT (supplied by Shanghai Institute of Organic Chemistry, Academia Sinica) (Liao et al. 1997) were used as compatibilizers for wood fiber-PE composites. Wood fibers were treated by soaking in a 1 percent (Dalvag et al. 1985) or 3 percent (Liao et al. 1997) solution using methylchloride (Dalvag et al. 1985) or hexane (Liao et al. 1997) as solvent. Dalvag et al. (1985) observed a decreased in tensile strength of the titanium di(dioctylpyrophosphate)oxyacetate treated fiber-HDPE composites. However, Liao et al. (1997) reported slight increase in tensile strength of the TC-POT and TC-PBT treated fiber-LLDPE composites.

Stearic acid

Stearic acid is typically used as a dispersion aid rather than compatibilizer. In this case, increased fiber dispersion will increase mechanical properties of the composites. Raj and Kokta (1995) reported that the addition of 1 percent stearic acid slightly increases the tensile strength, modulus, impact strength and elongation at break of the HDPE composites. Dalvag et al. (1985) observed that stearic acid increases the impact strength of the HDPE composites, however, decreases the tensile strength. Woodhams et al. (1984) also observed the reduction of the interfacial shear stress in HDPE composites due to stearic acid (2% by weight of fiber).

Material selection

Table 1 summarizes the different types of compatibilizer for wood-polyethylene composites, their application methods and performance. This literature review was carried out to select commercially viable compatibilizers for commercially scenario extruded wood-polyethylene composites.

From the review of literature and above discussion, it appears that maleated copolymers, especially maleated polyethylene copolymers present the best option as commercial compatibilizers for wood-PE composites. Maleated polyolefin copolymers can be used directly in the formulation mixture instead of pretreatment or precoating, which is also cost effective. PMPPIC is also another good option after maleated polyolefins. However, it requires pretreatment of the reinforcing wood particle with PMPPIC prior to mix with the matrix. In almost all reported cases, tensile test specimens were formed by compression molding or injection molding precompounded materials. However, none of the reported studies demonstrated the performance of compatibilizer on materials produced with direct extrusion, as is most common for commercial wood-PE composite products.

In addition, most commercial wood-PE composites products are intended for exterior use. Common and emerging applications include deck surfaces, guardrails, window lineals, fencing, marine structures, siding, and exterior building trim. In all of these uses, durability under repeated or prolonged moisture contact is a necessity. Although considerable amount of work on compatibilizing wood-PE composites has been performed, none of the studies has characterized the material performance under prolonged moisture exposure.

Materials and methods

Materials

Wood flour (American Wood Fibers 4010) composed of maple wood (A cer spp.) was milled to a nominal 40 mesh, and dried to a MC of less than 1 percent immediately prior to extrusion. High-density polyethylene (Equistar Petrothene- LB 0100) was used as supplied in reactor flake powder with a melt flow index 0.40 g/10 minutes. Three commercial polyolefin copolymers (Table 2) were evaluated as potential compatibilizers: 1) maleated polyethylene (MAPE-575 and MAPE-573), 2) maleated polypropylene (MAPP-950 and MAPP-597), and 3) ethylene-acrylic acid (EAA-540). All compatibilizers were supplied by Honeywell and were used as received.

Wood flour-HDPE composite manufacturing

The wood-HDPE composite formulations were prepared by mixing the appropriate proportions of different material in a drum blender for 10 minutes. The ratio of wood flour to HDPE was 60:40. Compatibilizer was added in powder form directly to the formulation at a rate of 2 percent of the total formulation weight. The additional component was accommodated by reducing the amount of wood flour a proportional amount. It was assumed that lowering the amount of wood would give the conservative estimation of the properties.

The composites were direct extruded using a 55 mm conical twin-screw extruder with intermeshing counter-rotating screws using a solid profile slit die, measuring 6 by 0.5 inches. The temperature of the four-extruder barrel zones and screws were set at 145[degrees]C. The three die zones were set at temperatures of 149, 155, and 155[degrees]C moving downstream. The screw rotation rate and barrel vacuum were kept constant throughout the extrusion process at 12 rpm and -28 inches Hg, respectively. Upon exiting the die, the extrudate was cooled by a series of water jets that formed a continuous curtain of water (ca. 16[degrees]C) around the extruded board. After cooling the extruded materials were air-dried and planed to a nominal thickness of 0.25 inches for testing.

Physical and mechanical testing

Water absorption was tested in accordance with ASTM D1037 with the exception of specimen size and soaking time. For this research, specimens measuring 4 inches square were cut out from the extruded boards. Four replicate specimens were submerged in distilled water and soaked for 120 days at room temperature. The total specimen weight and thickness at five marked points were measured for each specimen prior to submersion and at weekly intervals thereafter. Fractional water absorption (WA) and fractional thickness swelling (TS) were calculated by using following equations:

WA = [W.sub.f] - [W.sub.i]/[W.sub.i] [1]

Where:

[W.sub.i] = weight of the samples before water soaking;

[W.sub.f] = weight of the samples after 120 days water soaking.

TS [T.sub.f] - [T.sub.i]/[T.sub.i] [2]

Where:

[T.sub.i] = thickness of the samples before water soaking; [T.sub.f] = thickness of the samples after 120 days water soaking.

Flexural performance of the materials was evaluated in three-point bending according to ASTM D790. For each material type, 12 6- by 0.5-inch specimens were tested using a screw-driven universal testing machine (Instron Model 4466). The length of the specimens is parallel to the extrusion direction. The bending tests were conducted using 4 inches free-span at a maximum strain rate of 0.1067 in/min. In addition, 12 replicate specimens were tested wet following 120 days of submersion in distilled water.

Results and discussion

Compatibilizers are used in wood-thermoplastic composites to alter the interaction of the different material components with the goal typically being increased mechanical performance. However, for hygroscopic materials such as wood, the addition of thermoplastics is often aimed at moisture exclusion. For this reason, the influence of various copolymers on both the moisture and mechanical properties is investigated.

Moisture properties

Water absorption.--The water absorption of the wood-HDPE composites was investigated as a function of copolymer type. Mean fractional water absorption (WA) after 120 days immersion is given in Table 3. Figure 1 illustrates the effect of compatibilizers on the water absorption rate ([M.sub.t]) and saturated moisture content ([M.sub.sat]) of the wood-HDPE composites. The untreated control rapidly absorbed larger quantities of water compared to the composites produced with maleated polyolefins. The addition of any of the commercial maleated polyolefins produced composites that exhibited decreased water absorption. The decrease is most pronounced with the MAPE-575 and MAPP-950, which are the high molecular weight MAPE and MAPP, respectively. The extent of water absorption of MAPE-575, MAPP-950 and MAPP-597 treated samples is about 57 to 62 percent less than that of untreated control samples after 120 days of water soaking. This may be due to the enhancement of the interfacial adhesion between the wood flour and HDPE and in-situ better encapsulation of the wood flour by HDPE. The water absorption values of MAPE-573 and EAA 540 treated samples are also less than the untreated samples. However, those values are much higher than that of MAPE-575, MAPP-950 and MAPP-597 treated samples.

[FIGURE 1 OMITTED]

Diffusion.--The moisture weight gain for composites produced with different types of compatibilizers was plotted against the square root of time in Figure 1. These plots suggest a Fickian behavior as evidenced by linear relationship prior to approaching saturation. It should be noted that the materials with high levels of moisture absorption (e.g., control, EAA) displayed some degree of non-Fickian behavior when approaching saturation. This slight deviation will, however, be neglected here for sake of analysis.

The diffusion behavior was characterized by Fick's Law (Ferderic et al. 1993, Khan and Rahman 1991) using a relation established by Springer (1988) to calculate the apparent diffusion constant [D.sub.A] from the slope of this linear region as:

[D.sub.A] = [pi] [[h/4[M.sub.sat]].sup.2] [[d[M.sub.t]/d[square root of (t)]].sup.2] [3]

Where: h is the thickness of the sample, [M.sub.sat] is the weight gain percentage at saturation, and d[M.sub.t]/d[square root of (t)] as the slope of the weight gain vs. square root of time relation. The equation is based on one-dimensional diffusion in a plate neglecting edge effects; Rao et al. (1988) proposed a geometric edge correction factor to calculate the true diffusion constant (D) considering moisture ingress through the specimen edges:

D = [D.sub.A]/[(1 + h/L + h/W).sup.2] [4]

where, L and W are length and width of the samples.

The average diffusion constants of the different composites are presented in Table 3 and range from 6.29e-7 to 8.07e-7 [mm.sup.2]/sec. Although these values vary some 28 percent across the formulation, there appears to be little relation with the coupling efficiency as revealed in the moisture performance. The lowest diffusion coefficients were found with the control and MAPP formulations, with the EAA formulation exhibiting the largest highest rates of diffusion. These findings seem to suggest that the coupling agents have the largest influence over the saturated MC ([M.sub.sat]) rather than the diffusion coefficient.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Swelling.--The addition of various copolymers had similar effect on the fractional thickness swelling (TS) as the water absorption. Untreated control samples had swollen 11.61 percent in thickness, whereas MAPE-575, MAPP-950, and MAPP-597 treated samples swelled only 4.59 percent, 4.33, and 5.27 percent, respectively after 120 days water soaking. These copolymer treatments represent almost 57 to 63 percent reduction in thickness swelling. Both an enhanced interfacial adhesion and the reduced moisture absorption could contribute to the decreased swelling of copolymer treated composites. The swelling coefficient ([beta] = TS/WA) was calculated to separate these effects (Table 3). The [beta] values for the various formulations are similar, ranging from 0.53 to 0.611. However, notice that those for the high molecular weight MAPE-575 and MAPP-950 represent the lowest values tested. Interpretation of these results indicate that a majority of the reduced moisture swelling results from a decreased moisture absorption, however, increased interfacial adhesion may contribute to some degree to both properties.

Mechanical properties

The maleated polyolefin compatibilizers have the potential to improve both the fiber-matrix interphase and the fiber dispersion. Whereas improved interphase structure will primarily influence failure properties, dispersion has also the potential to increase composite stiffness. In this study, the flexure behavior of the various materials was investigated. The flexural bars were failed in tension. Representative stress-strain curves are presented in Figure 2 and 3. Notice that the addition of some coupling agents increased both the slope and magnitude of the response (Fig. 2). In general, those copolymers that exhibited a positive effect on properties also decreased the strain to failure. The strain at failure of the untreated control dry samples was about 0.021. About 29 percent and 24 percent reduction in strain at failure was observed with the MAPE-575 and MAPP-950 treated dry samples respectively. The performance of the MAPP-597 is also similar to MAPP-950. There is no reduction in strain at failure was observed with the MAPE-573 and EAA-540 treated dry samples. The strain at failure became almost double for all different types composites due to 120 days water immersion. However, the strain at failure of the MAPE-575 and MAPP-950 treated wet samples is about 18 percent less than that of untreated control wet samples.

For all materials, the absorption of moisture decreased the strength and stiffness of the material. However, the magnitude of the wet properties was dependent on the copolymers used (Fig. 3).

Modulus of Elasticity.--The average modulus of elasticity (MOE) of composites produced, range from 3078 MPa to 3742 MPa as shown in Table 4. The composites produced with MAPE-575 and MAPP-950 displayed higher stiffness than the untreated control, representing an increase of about 22 percent for the MAPE-575 samples. In contrast, the MAPE-573, MAPP-597 and EAA-540 composites showed no significant difference with the control samples as shown in Table 5.

The effect of moisture on composite MOE was significant (p < 0.001). For all cases studied, the MOE of the water saturated samples were 40 to 45 percent lower than equivalent values in the dry condition. This decrease in properties for absorption of water could originate from two causes: (1) a decreased stiffness of the material components from absorbed moisture and (2) damage of the fiber interface from differential swelling.

Flexural Strength.--The average flexural strength of composites produced, range from 26.46 MPa to 37.56 MPa as shown in Table 4. The flexural strength values of the MAPE-575 and MAPP-950 treated dry samples are about 40 percent and 34 percent higher than untreated control dry samples. The effect of MAPP-597 treatment is also positive, the strength values are about 18 percent higher for MAPP-597 treated samples. The MAPE-573 and EAA-540 treated samples did not produce any significant improvement in strength compared to untreated control samples as shown in Table 5.

The effect of moisture on composite flexural strength was also significant (p < 0.001). The flexural strength of all materials tested decreased following a 120-day water soak. MAPE-575 and MAPP-950 treated wet samples retained about 80 percent of their dry strength, whereas untreated control samples retained only about 67 percent of their dry strength. This reduction may be due to the deterioration of the interfacial adhesion resulting from the swelling stresses gained by the wood flour during 120 days water soaking. The interesting finding is that the MAPE-575 and MAPP-950 treated wet (120 days water soaked) samples have higher (about 109%) flexural strength values than the untreated control dry samples.

Conclusions

The aim of this paper was to review the literature to screen out the better compatibilizer for wood-PE composites and to investigate their performance in commercial scenario extruded wood-HDPE composites. Literature showed that oraginic peroxides, anhydrides, maleated polyolefins, polymeric isocyanates, silanes, monomers, titanates, and stearic acid were used as compatibilizers for wood-PE composites. Among all those compatibilizers, maleated polyolefin copolymers and polymethylenepolyphenyl isocyanate were found to be better compatibilizers for wood-PE composites.

The inclusion of maleated polyolefin copolymers in the commercial scenario extruded wood-HDPE composites formulations improved the overall mechanical properties and dimensional stability. MAPE-575 and MAPP-950 copolymers showed superior performance than other copolymers for wood-HDPE composites. The effect of prolonged moisture contact was significant. A reduction in strength of around 20 percent and 33 percent was observed for MAPE-575 treated samples and untreated control samples respectively at around saturation. This is may be due to the moisture sorption and swelling induced stresses.

Literature cited

Balatinecz, J.J., M.I. Khavkine, S. Law, and V. Kovac. 1999. Properties of polyolefin composites with blends of wood flour and coal ash. In: Proc. of the 5th Inter. Conf. on Wood Fiber-Plastic Composites, ed. R.M. Rowell, pp. 235-240.

Bataille, P., P. Allard, P. Cousin, and S. Sapieha. 1990. Interfacial phenomena in cellulose/polyethylene composites. Polym. Compos. 11(5): 301-304.

Cousin, P., P. Bataille, H.P. Schreiber, and S. Sapieha. 1989. Cellulose-induced crosslinking of polyethylene. J. Appl. Polym. Sci. 37(10): 3057-3060.

Czarenecki, L. and J.L. White. 1980. Shear flow theological properties, fiber damage and mastication characteridtics of aramid, glass and cellulose fiber reinforced polystyrene melts. J. Appl. Polym. Sci. 25:1217-1244.

Dalvag, H., C. Klason, and H.E. Stromvall. 1985. The efficiency of cellulose fillers in common thermoplastics: Filling with processing aids and coupling agents. Inter. J. Polym. Mater. 11:9-38.

Felix, J.M. and P. Gatenholm. 1991. The nature of adhesion in composites of modified cellulose fibers and PP. J. Appl. Polym. Sci. 42:609-620.

Ferderic, P., L.A. Carlsson, and J.W. Gillespie, Jr. 1993. Marine environmental effects on polymer matrix composites. Composite Materials: Fatigue and Fracture, ASTM STP 1230, R.H. Martin, ed. American Soc. for Testing and Materials, 5:283-303.

Khan, M.A. and S. Rahman. 1991. Moisture sorption isotherms of wood and wood-plastic composites. Polym. Plast. Technol. Eng. 30:433-440.

Khavkine, M., M. Kazayawoko, S. Law, and J.J. Balatinecz. 2000. Durability of wood flour-thermoplastic composites under extreme environmental conditions and fungal exposure. Inter. J. Polym. Mater. 46: 255-269.

Kishi, H., M. Yoshiaka, A. Yamanoi, and N. Shiraishi. 1988. Composites of wood and PPs. Mokuzai Gakkaishi 34(2): 133-135.

Klason, C., J. Kubat, and H.E. Stromvall. 1984. The efficiency of cellulosic fillers in common thermoplastics: Filling without processing aids or coupling agents. Inter. J. Polym. Mater. 10:159-187.

Krzysik, A.M., J.A. Youngquist, and G.E. Myers. 1991. Wood-polymer bonding in extruded and non-woven mat composite panels. In: Proc. 1990 Wood Adhesives Syrup., Madison, Wisconsin. pp. 183-189.

Liao, B., Y. Huang, and G. Cong. 1997. Influence of modified wood fibers on the mechanical properties of wood fiber-reinforced polyethylene. J. Appl. Polym. Sci. 66:1561-1568.

Maldas, D. and B.V. Kokta. 1991. Influence of organic peroxide on the performance of maleic abhydride coated cellulose fiber filled thermoplastic composites. Polym. J. 23(10): 1163-1171.

-- and --. 1994. Recycled thermoplastic-woodflour composites. Inter. J. Polym. Mater. 26:69-78.

-- and --. 1995. Composite molded products based on recycled thermoplastcsand waste cellulosics.II. Kenaf fiber-recycled PE composites. J. Reinf. Plast. Compos. 14:458-470.

--, --, and C. Daneault. 1989. Influence of coupling agents and treatments on the mechanical properties of cellulose fiber-polystyrene composites. J. Appl. Polym. Sci. 37:751-775.

Mustafa, N. and J.J. Balatinecz. 1996. Plastics recycling issues. In: Proc. Wood fiber-plastic composites:virgin and recycled wood fiber and polymers for composites. Forest Prod. Soc., Madison, Wisconsin. pp. 157-160.

Myers, G.E., I.S. Chahyadi, C.A. Coberly, and D.S. Ermer. 1991. Wood flour/polypropylene composites: Influence of maleated polypropylene and process and composition variables on mechanical properties. Inter. J. Polym. Mater. 15:21-44.

Oksman, K. and H. Lindberg. 1998. Influence of thermoplastic elastomers on adhesion in polyethylene-wood flour composites. J. Appl. Polym. Sci. 68:1845-1855.

Olsen, D.J. 1991. Effectiveness of maleated polypropylenes as coupling agents for wood flour/polypropylene composites. ANTEC. pp. 1886-1891.

Raj, R.G. and B.V. Kokta. 1995a. Effect of aging cycle on mechanical properties of HDPE-pretreated wood fiber composites. Wood Fiber-Plastic Composites: Virgin and Recycled Wood Fiber and Polymers for Composites. Caulfield, D.F., R.M. Rowell and J.A. Youngquist, eds. pp. 235-239.

-- and --. 1995b. Surface-modified bagasse fiber, high-density polyethylene composites. Wood Fiber-Plastic Composites: Virgin and Recycled Wood Fiber and Polymers for Composites. D.F. Caulfield, R.M. Rowell and J.A. Youngquist, eds. pp. 220-226.

--, --, and C. Daneault. 1989a. Effect of chemical treatment of fibers on the mechanical properties of polyethylene-wood fiber composites. J. of Adhesion Sci. Tech. 3(1):55-64.

--, --, and --. 1990a. The use of isocyanate as a bonding agent to improve the mechanical properties of polyethylene-wood fiber composites. Inter. J. Polym. Mater. 14:223-234.

--, --, and --. 1990b. Wood flour as a low-cost reinforceing filler to polyethylene: Studies on mechanical properties. J. Mater. Sci. 25(3): 1851-1855.

--, --, D. Maldas, and C. Daneault. 1989b. Use of wood fibers in thermoplastics. VII. The effect of coupling agents in polyethylene-wood fiber composites. J. Appl. Polym. Sci. 37:1089-1103.

Rao, R.M., N. Balasubramanian, and M. Chanda. 1988. Factors affecting moisture absorption in polymer composites, Part I: Influence of internal factors. Environmental Effects on Composites Materials 3:75-87.

Sanadi, A.R., R.M. Rowell, and R.A. Young. 1992. Estimation of fiber-matrix interfacial shear strength in lignocellulosic-thermoplastic composites. In: Materials Interaction Relevant to Recycling of Wood-based Materials (Vol. 266), R.M. Rowell, T.L. Laufenberg and J.K. Rowell, eds. pp. 81-92.

--, R.A. Young, C. Clemons, and R.M. Rowell. 1994. Recycled newspaper fibers as reinforcing fillers in thermoplastics: Analysis of tensile and impact properties in PP. J. Reinf. Plast. Compos. 13(1):54-67.

--, D.F. Caufield, and R.E. Jacobson. 1997. Agro-fiber thermoplastic composites. In: Pap. and Composites from Agro-based Resources. R.M. Rowell, R.A. Young, J.K. Rowell eds.. CRC Lewis Press. Boca Raton, Florida. pp. 377-401.

Sanschagrin, B.S., T. Sean, and B.V. Kokta. 1988. Mechanical properties of cellulose fiber reinforced thermoplastics. J. Thermoplast. Compos. Mater. 1(4):184-195.

Sapieha, S., P. Allard, and Y.H. Zang. 1990. Dicumyl peroxide-modified cellulose/LLDPE composites. J. Appl. Polym. Sci. 41:2039-2048.

Scott, M.H. 2001. Not all maleated polyolefins are created equal. In: Proc. of the Soc. Of Plastics Engineers Annual Technical Conf., Vol. III-special area. Dallas, Texas. Paper #0172, pp. 2673-2676.

Selke, S.E. and J. Childress. 1993. Wood fiber/high-density polyethylene composites: Ability of additives to enhanced mechanical properties. Wood Fiber/Polymer Composites: Fundamental Concepts, Process and Materials Options. ISBN 0-935018-57-3. M.P. Wolcott, ed. Forest Prod. Soc., Madison, Wisconsin. pp. 109-111.

Springer, G.S. 1988. Environmental Effects. Environmental Effects on Composites Materials 3:-34.

Woodhams, R.T., G. Thomas, and D.K. Rodgers. 1984. Wood fibers as reinforcing fillers for polyolefins. Polym. Eng. Sci. 24(15): 1166-1171.

Mohammed Jahangir A. Chowdhury Michael P. Wolcott *

The authors are, respectively, Research Associate, Dept. of Wood Sci., Univ. of British Columbia, Vancouver, B.C., Canada (mchowdhu@interchange.ubc.ca) and Professor, Civil and Environmental Engineering, Washington State Univ., Pullman, Washington (Wolcott@wsu.edu). This paper was received for publication in May 2006. Article No. 10200.

* Forest Products Society Member.

Table 1.--Compatibilizers for wood-PE composites: their application
method and performance.

Compatibilizer                                         Fiber

                 Percent in    Application   Matrix
Type             formulation   method        polymer   Type

BPO              0.10          Mixing        LLDPE     Bleached pulp
DCP              0.021         Mixing        LLDPE     Bleached pulp
BPO              0.08          Mixing        HDPE      Bleached pulp
MA, BPO          2.00,0.04     Coating       HDPE      CTMP
MA, BPO          1.00,0.03     Coating       HDPE      Nutshell
MA, DCP          1.00,0.03     Coating       HDPE      Nutshell
MAPP             1.00          Mixing        HDPE      Wood flour
MAPP             3.60          Mixing        HDPE      Wood fiber
MAPE             4.00          Coating       LDPE      Wood flour
MAPE             4.50          Mixing        LDPE      Kenaf flour
SEBS-MA          4.00          Mixing        LDPE      Wood flour
PMPPIC           2.00          Coating       HDPE      Wood flour
PMPPIC           1.00          Coating       HDPE      CTMP
PMPPIC           1.00          Coating       LDPE      CTMP
PMPPIC           2.00          Coating       HDPE      Bagasse
HMDIC            1.00          Coating       HDPE      CTMP
TDIC             1.00          Coating       HDPE      CTMP
Silane A-172     1.20          Coating       LLDPE     CTMP
Silane A-174     1.20          Coating       LLDPE     CTMP
Silane A-172     1.20          Coating       HDPE      Wood flour
Silane A-174     1.20          Coating       HDPE      Wood flour
Silane A-174     0.30          --            LLDPE     Cellulose
  + BPO
Acrylonitrile    0.9           Grafting      LLDPE     Wood fiber
Titanate         0.9           Coating       LLDPE     Wood fiber

                                Property
Compatibilizer   Fiber       enhancement (%)

Type             Percent   Strength   Modulus   Reference

BPO              30         77         --       Cousin et al. 1989
DCP              33         78         --       Sapieha et al. 1990
BPO              --         10         --       Bataille et al. 1990
MA, BPO          40         44         -4       Raj et al. 1990a
MA, BPO          10        -10         16       Maldas and Kokta 1991
MA, DCP          30         65         33       Maldas and Kokta 1991
MAPP             30         25         --       Dalvag et al. 1985
MAPP             40         65         20       Selke and childress
                                                  1993
MAPE             40        116         40       Maldas and Kokta 1994
MAPE             45        135         48       Maldas and Kokta 1995
SEBS-MA          40         18        -37       Oksman and Lindberg
                                                  1998
PMPPIC           30         94         -6       Raj et al. 19906
PMPPIC           30        112         36       Raj et al. 1990a
PMPPIC           30         88          5       Raj et al. 1990a
PMPPIC           40         48         12       Raj and Kokta 19956
HMDIC            30         70         25       Raj et al. 1990a
TDIC             30         60         28       Raj et al. 1990a
Silane A-172     30         62         44.8     Raj et al. 19896
Silane A-174     30         37         13.5     Raj et al. 19896
Silane A-172     30         31        -24       Raj et al. 19906
Silane A-174     30         36        -27       Raj et al. 19906
Silane A-174     30         26          4.8     Bataille et al. 1990
  + BPO
Acrylonitrile    30         56         14       Liao et al. 1997
Titanate         30         23        -10       Liao et al. 1997

Table 2.--Compatibilizer description and supplier.

Compatibilizer            Description                      Supplier

Maleated polyethylene (MAPE-575)                           Honeywell
  Saponification number   32 to 36 mg KOH/g
  Viscosity               4200 cp @ 140[degrees]C
  Melt point              98[degrees]C
  Density                 0.92 g/cc
  MW (Mn)                 3600
Maleated polyethylene (MAPE-573)                           Honeywell
  Saponification number   3 to 6 mg KOH/g
  Viscosity               600 cp. @ 140[degrees]C
  Melt point              98[degrees]C
  Density                 0.92 g/cc
  MW (Mn)                 2500.00
Maleated polypropylene (MAPP-950)                          Honeywell
  Saponification number   40 to 45 mg KOH/g
  Viscosity               3000 cp. @ 190[degrees]C
  Melt point              139[degrees]C
  Density                 0.93 g/cc
  MW (Mn)                 8000
Maleated polypropylene (MAPP-597)                          Honeywell
  Saponification number   75 to 85 mg KOH/g
  Viscosity               300 to 800 cp @ 190[degrees]C
  Melt point              133[degrees]C
  Density                 0.94 g/cc
  MW (Mn)                 3100
Ethylene-acrylic acid copo(EAA-540)                        Honeywell
  Acid number             40 mg KOH/g
  Viscosity               575 cp. @ 140[degrees]C
  Density                 0.93 g/cc

Table 3.--Mean fractional water absorption (WA), thickness swelling
(TS), swelling coefficient ([beta]) and diffusion constant (D) values
for composites produced with various copolymers.

                                   MAPE

Property             Control       MAPE-575      MAPE-573

WA (%COV)            0.199 (0.1)   0.086 (3.4)   0.l35 (2.0)
TS (%COV)            0.116(l.1)    0.046 (8.3)   0.078 (0.3)
[beta]               0.548         0.535         0.574
D ([mm.sup.2]/sec)   6.70E-07      7.57E-07      7.56E-07

                     MAPP                        EAA

Property             MAPP-950      MAPP-597      EAA-540

WA (%COV)            0.077 (0.5)   0.082 (0.6)   0.l57 (2.6)
TS (%COV)            0.043 (0.5)   0.050 (5.3)   0.093 (0.2)
[beta]               0.562         0.611         0.595
D ([mm.sup.2]/sec)   6.27E-07      6.75E-07      8.07E-07

Table 4.--Mechanical properties of various wood-HDPE composites tested
at dry and wet conditions.

            Flexural strength               Modulus of elasticity

            Dry               Wet           Dry           Wet

Treatment                           (MPa)

Control     26.99 (2.4) (a)    18.2 (3.2)   3083 (7.5)    1644 (8.4)
MAPE-573    28.02 (2.8)       20.08 (3.2)   3141 (6.5)    1750 (14.2)
MAPE-575    37.56 (4.0)       29.43 (1.8)   3742 (7.0)    2133 (8.1)
MAPP-950    36.19 (l.5)       29.36 (2.1)   3463 (3.0)    2026 (3.8)
MAPP-597    31.82 (3.1)       20.08 (4.1)   3252 (6.2)    1910 (9.2)
EAA-540     26.46 (2.6)       18.11 (5.l)   3078 (6.8)    1716 (11.6)

(a) Values in parentheses = %COV.

Table 5.--Results of one way ANOVA carried out on dry wood-HDPE
Composite physical and mechanical properties between treatments.

Property     Treatment   Control    MAPE-573   MAPE-575

Flexural     Control     --         B (b)      A (a)
strength     MAPE-573    B          --         A
             MAPE-575    A          A          --
             MAPP-950    A          A          C
             MAPP-597    A          A          A
             EAA-540     NS         A          A
Modulus of   Control     --         NS         A
elasticity   MAPE-573    NS         --         A
             MAPE-575    A          A          --
             MAPP-950    A          A          C
             MAPP-597    NS         NS         A
             EAA-540     NS         NS         A
WA (e)       Control     --         A          A
             MAPE-573    A          --         A
             MAPE-575    A          A          --
             MAPP-950    A          A          C
             MAPP-597    A          A          NS
             EAA-540     A          B          A

Property     Treatment   MAPP-950   MAPP-597   EAA-540

Flexural     Control     A          A          NS (d)
strength     MAPE-573    A          A          A
             MAPE-575    C (c)      A          A
             MAPP-950    --         A          A
             MAPP-597    A          --         NS
             EAA-540     A          A          --
Modulus of   Control     A          NS         NS
elasticity   MAPE-573    A          NS         NS
             MAPE-575    C          A          A
             MAPP-950    --         C          A
             MAPP-597    C          --         NS
             EAA-540     A          NS         --
WA (e)       Control     A          A          A
             MAPE-573    A          A          B
             MAPE-575    C          NS         A
             MAPP-950    --         NS         A
             MAPP-597    NS         --         A
             EAA-540     A          A          --

(a) A = significantly different at p < 0.0001;

(b) B = significantly different at p < 0.001;

(c) C = significantly different at p < 0.05;

(d) NS = not significantly different (p > 0.05);

(e) WA = water absorption.