Field of the Invention

The present invention relates to a cross-linked plastic/natural cellulosic fiber composite, and more specifically to a composite wherein the plastic is cross-linked via a peroxide cross-linking initiator and having gypsum and/or Portland cement incorporated therein resulting in a composite which exhibits a significant reduction in water absorption. Background of The Invention

Wood plastic composites (WPC) are blends of cellulosic fibers such as wood flour, plastic (e.g. polyethylene, polypropylene or polyvinyl chloride), and other ingredients that are gaining prominence in the construction industry as a material for siding, decking, fencing, and other structural forms. One appealing aspect of these composites is that it is a means to recycle waste wood into useful products. Typically the wood flour is dry mixed with pulverized plastic along with other ingredients and then extruded at high temperatures (300° F to 400° F) during which time the wood mixes with the molten plastic. Upon exiting the extruder, the boards are then cooled quickly, usually with a water spray so that they retain their physical shape. High density polyethylene (HDPE) is typically used as the plastic in this application due to its high degree of crystallinity which provides good flexural properties.

Despite the rapidly growing use of WPCs, there are technical challenges to overcome for continued market growth. It was thought that these materials would not absorb a significant amount of water because it was believed that the continuous hydrophobic plastic phase would encapsulate the hygroscopic wood flour. Recent data from researchers in the industry have shown that this is not the case. In fact, water readily absorbs into the wood particles in these materials. This results in swelling of the fibers as water soaks into them. Absorption of water in a composite can lead to swelling, buckling, reduced creep resistance, and facilitate microbial growth resulting in increased fungal degradation . Wood fibers are polar (hydrophilic) whereas most polymers, especially thermoplastics, are non-polar (hydrophobic). This incompatibility can result in poor adhesion between polymer and wood fibers in WPCs. As a result, the mechanical properties, water resistance, and other properties are compromised. The use of compatibilizers to disperse the wood fibers into the polymer during extrusion to avoid poor melt strength of the wood composite extrudates has been proposed.

Modifications to the wood fiber, and the use of compatibilizers, coupling agents, and interfacial agents have been used to improve the compatibility and adhesion between the wood and plastic in the WPCs. US 3894975 and 3958069 describe an in-situ polymerization of wood fibers with maleic anhydride and styrene to prepare a wood-polymer composite. US 4851458 describes a pretreatment of cellulose fibers with an adhesion promoter. Other additives for improving the compatibility and adhesion of wood and plastic include: isocyanate bonding agents (US 4376144 and GB 2192398) and silane bonding agents (US 4820749 and GB 2192397).

Summary of The Invention The present inventor discovered that cross- linking the plastic and

incorporation of gypsum and/or Portland cement into WPC blends results in dramatically decreased water absorption by these materials. It is believed that this is due to the physical constraints imposed by the cross-links formed by the cross-linking agent. These cross-links create a tighter matrix around the wood fibers and greatly restrict their ability to absorb water and swell and growth. It is believed that the gypsum and/or Portland cement works synergistically with the peroxide cross-linker by the growth of gypsum crystals and/or cement in the constrained matrix. It was discovered that the inclusion of small amounts of a resin cross-linking agent such as a peroxide in combination with a gypsum and/or Portland cement additive would significantly reduce the water absorption of the resulting composite.

The invention relates to a composite material comprising a homogeneous distribution comprising:

a) 20 - 60 weight percent of one or more thermoplastics; b) 40 - 80 weight percent of natural cellulosic libers;

c) 2-25 weight percent gypsum and/or Portland cement; and

d) a cross-linking agent. The invention further relates to a process for reducing the water absorption of a thermoplastic/cellulosic fiber composite comprising adding to said

thermoplastic/cellulosic fiber composite, prior to or during processing, a cross-linking agent comprising a peroxide and an additive gypsum and/or Portland cement. Brief Description of The Drawings

Figure 1 is a graph of Weight Percent Water Absorption versus Weight Percent Gypsum Content

Figure 2 is a graph of Weight % water absorption versus Percent Gypsum.

Figure 3 is a graph of Density versus Weight % of Gypsum.

Figure 4 is a graph of MOR versus Weight % Gypsum.

Figure 5 is a graph of MOE versus Weight % Gypsum.

Detailed Description of The Invention

The invention relates to composite of a thermoplastic and natural cellulosic fibers with a polymeric cross-linking agent and an additive. Specifically, the cross- linking agent is a peroxide such as dialkyl peroxides, peroxyketals, peroxyesters, hydroperoxides, peroxycarbonates, diacyl peroxides and mixtures thereof. The peroxide cross-linking agent of the present invention is selected to provide efficient cross-linking for the thermoplastic component of the combination. For combinations which employ the preferred high density polyethylene, preferred cross-linking agents include peroxides such as VUL-CUP ^® (a, α'-bis (tert-butylperoxy)- diisopropylbenzene), DI-CUP® (dicumyl peroxide), LUPEROX® 801 (t-butyl cumyl peroxide), LUPEROX 101 (2,5-dimethyl -2,5-di(t-butylperoxy) hexane), and

LUPEROX 130 (2,5-dimethyl -2,5-di(t-butylperoxy) hexyne), all available from Arkema Inc. For some thermoplastic materials such as polypropylene and polyvinyl chloride, peroxides may require the use of a coagent in order to provide cross-linking in accordance with the present invention.

The additive of the present invention is gypsum and/or Portland cement. The gypsum and/or portland cement is believed to act synergistically with the peroxide cross-linking agent through the growth of gypsum crystals and/or cement in the constrained matrix which significantly reduces water absorption of the resultant WPC.

Gypsum consists of calcium sulfate dehydrate hydrate (CaSO ₄ ^• 2 H ₂ O).

When heated to temperatures greater than 150° C, it loses water and becomes calcium sulfate hemi-hydrate (CaSO ₄ ^{• 1} Zi H ₂ O), which is also known as Plaster of Paris.

When the hemi-hydrate form of calcium sulfate is exposed to water it reverts to the dihydrate and forms a network of interlocking crystals that make it useful for building applications such as stucco and drywall. In the subsequent discussion contained in this report, the material that was added to the wood-plastic composites was the hemi- hydrate form of calcium sulfate and is referred to as gypsum.

Portland cement is a blend consisting mainly of calcium silicates along with various iron and aluminum salts. When hydrated it forms an extensive crystalline network which also makes it useful as a building material. The thermoplastic matrix can be any thermoplastic including, but not limited to high density polyethylene, low density polyethylene, polypropylene, other olefin resins, polystyrene, acrylonitrile/styrene copolymers, acrylonitrile/butadiene/styrene copoloymers, ethylene/vinyl acetate copolymers, polymethyl methacrylate, vinyl chloride copolymers, polyvinyl chloride, chlorinated polyvinyl chloride, chlorinated polyethylene and mixtures thereof. Preferably the thermoplastic matrix is made up of high density polyethylene, low density polyethylene, or other olefin resins. Most preferably the thermoplastic is high density polyethylene or low density polyethylene. The thermoplastic matrix comprises from 20 to 60 percent by weight and preferably less than 50 percent by weight of the WPC. Percentages herein are in weight percent unless otherwise specified. While a WPC is generally referred to as a wood-polymer composite, it is envisioned that any cellulosic material, either natural or regenerated, may be used as the fibrous filler of the present WPCs. The cellulosic material may be a mixture of one or more materials including, but not limited to wood flour, wood fiber, and agricultural fibers such as wheat straw, flax, hemp, kenaf, nut shells, rice hulls and mixtures thereof. The cellulosic material may also be a pulped cellulosic fiber. The pulped cellulosic fiber may be made of fully or partially recycled materials. Typical cellulosic fibers contain 8%-12% moisture, therefore reducing the moisture content may be needed either by pre-drying the fibers or other methods known in the art. The cellulosic fiber is present in the composite at from 40 to 80 percent by weight, preferably from 45 to 80 percent by weight, more preferably greater than 50 percent by weight, and most preferably from 55 to 70 percent by weight of the composite. Wood polymer composites containing pulped cellulosic fiber may contain 10 to 90 weight percent of the thermoplastic and 10-90 weight percent of pulped cellulosic fiber.

The polymeric cross-linking agent is preferably added to the WPC at from 0.15 - 0.75, preferably 0.30 - 0.60 weight percent based on the weight of the total WPC formulation. The reduction in water absorption was found to occur with the addition of even small amounts of cross-linking agent. It is believed that with different cellulosic material and/or different cross-linking agent, the preferred amount of cross-linking agent may vary and could be easily determined.

Wood polymer composites are formed by blending the thermoplastic, cellulosic fiber and polymeric cross-linking agent, gypsum and/or Portland cement, and other additives in any order and by any method, and then either directly forming the mixture into a final article, or else forming the mixture into a form useful for further processing, such as pellets or a powder. In one embodiment, the wood polymer composite is formed by blending the thermoplastic matrix and any additives, including the polymeric cross-linking agent and gypsum and/or Portland cement and typical additives such as compatibilizers, lubricants, antioxidants, UV and heat stabilizers, colorants, impact modifiers, and process aids. The WPC may then be extruded directly into a final shaped article, or may be pelletized or ground to a powder prior to final use. A WPC made of the composition of the invention can be formed into a final article by means known in the art, such as by extrusion, injection molding, casting or combination thereof.

The WPC with cross-linking agent and gypsum and/or Portland cement described in the invention provides excellent flexural strength and modulus, and results in a decrease in moisture absorption compared to the WPC control without cross-linking agents. The WPC is useful in many applications, including, but not limited to outdoor decks, siding, fencing, roofing, industrial flooring, landscape timbers, railing, moldings, window and door profile, and automobile applications. The WPC may be foamed to produce a lighter and less expensive composite material.

Examples The compositional information and water absorption results were obtained from mixtures with and without a cross-linking agent and with and without gypsum and/or cement. The compositions that were mill-mixed and press-cured are shown in Table 1. From this data it can be seen that the simple blend of wood flour and high density polyethylene (HDPE) showed a weight gain of over 12% after four weeks of immersion in the water bath. Addition of Vul-Cup® 40KE, a peroxide cross-linking agent, to the formulation (Formulations 2 and 3) reduced the water absorption by nearly a half while formulations with gypsum alone (Formulations 4 and 5) or cement alone (Formulations 12 and 13) showed no improvement in the water absorption at the concentrations of those tests. When both gypsum and cement were added at low levels (about 5%) to formulations that contained 1% Vul-Cup 40KE (Formulations 6, 1, 14, and 15), there was no improvement in the water absorption relative to the formulation that contained 1% of the peroxide alone (Formulation 3). But it was observed that at higher loadings of gypsum the water absorption decreased markedly (Formulations 8, 9, 10, and 11). A plot of the water absorption with increasing gypsum content in the formulation is shown in Figure 1. This illustrates the beneficial impact that the peroxide/gypsum combination had on water absorption of the composite.

Table 2 summarizes the formulations that were included in a study of extruded boards. The boards of wood plastic composites were made using a 35 mm extruder. The components were dry mixed prior to adding to the extruder inlet arid then heated to 375° F during the extrusion process to decompose the peroxide and effect the cross- linking of the high density polyethylene. The first run is a control to give a baseline formulation without peroxide or gypsum. Run 2 contains only Vul-Cup® 40KE, a peroxide cross-linker, while the subsequent six runs have increasing amounts of gypsum along with the Vul-Cup 40KE. Table 3 contains the water absorption data for these samples as measured periodically over the course of 28 days (672 hours).

In order to measure the water absorption of the boards made from these formulations, the surface of the boards was shaved prior to immersion in water to prevent any plastic skin that might have formed at the surface immediately after extrusion from inhibiting the migration of water into the boards. It was immediately noticed that the boards that contained only Vul-Cup 40KE (Run 2) had a higher water absorption than the control without peroxide. This was surprising because previous data had shown that the addition of peroxides decreases water absorption in wood- plastic composites. In considering the data contained in Table 4, it can be shown that the density of the Run 2 samples were significantly lower than the control which indicates that the peroxide had caused some foaming of the board. It is believed that the voids formed by the volatile peroxide decomposition products provided a space for water to accumulate during the water absorption test, resulting in the higher water absorption. The amount of water absorbed did not improve when 1% gypsum was added to the formulation in Run 3, but when the gypsum content was > 2% the water absorption was significantly reduced with increasing gypsum content. The plot in Figure 2 shows that the water absorption of the formulation that includes 25% gypsum (Run 8) at 672 hours is reduced by 64% relative to the control (Run 1) and nearly 70% as compared to the formulation that contains only peroxide (Run 2).

Table 4

Results of Density and Flexural Tests for Composites Containing VuI-C up® 40 KE and Gypsum

Weight Percent of Component in Formulation

Parameter 1 2 3 4 5 6 7 8

Density

(g/cc) 1.15 1.07 1.08 1.09 1.15 1.17 1.20 1.25

MOE (psi) 564,469 395,626 408,163 435,242 515,622 511,222 501,538 496,624

MOR (psi) 3,858 3,058 3,080 3,279 3,963 3,S42 4,061 4,253

Figure 3 shows the effect of increasing gypsum content on density. Since gypsum has a density of 2.3 g/cm3 it was expected that the density of the overall composite would increase with the gypsum content. Figure 4 shows the effect of gypsum on the MOR. This indicates that increasing the gypsum content can have a beneficial impact on the rupture strength of these composites. It was observed that the MOR of the formulations gave a similar profile to the density plot. It is believed that the reason that MOR is lower for formulations that contain peroxide alone and peroxide plus low concentrations of gypsum is due to the porosity caused by the volatile peroxide decomposition products. Lowering the peroxide concentration is expected to minimize this effect. Figure 5 contains a plot of the MOE which is a measure of the stiffness of the extruded boards. It can be observed in this plot that the formulations that contain peroxide have a decreased MOE. It is known that peroxide cross-linking of HDPE reduces the stiffness because it interferes with reformation of crystalline domains within the polymer after curing at high temperatures. Adding gypsum to the composite appears to compensate for much of this loss in stiffness caused by the peroxide.