WO2002002886A1

WO2002002886A1 - Structural biocomposite materials, systems, and methods

Info

Publication number: WO2002002886A1
Application number: PCT/US2001/021052
Authority: WO
Inventors: Michael J. Riebel; Paul L. Torgusen
Original assignee: Phenix Biocomposites, Llc
Priority date: 2000-07-05
Filing date: 2001-06-30
Publication date: 2002-01-10
Also published as: AU2001270292A1; US20020100565A1

Abstract

A structural biocomposite material that incorporates small strands of agricultural straw, typically non-wood cellulosic straws, such as cereal grain straw.

Description

STRUCTURAL BIOCOMPOSITE MATERIALS, SYSTEMS, AND

METHODS

Statement of Related Applications This application claims the benefit of U.S. Provisional Application No.

60/215,946, filed July 05, 2000, which is incorporated herein by reference.

Field of the Invention The present invention relates to biocomposite materials, systems, and methods, particularly to materials made from agricultural straw. More specifically, the present invention relates to structural biocomposite materials that have applications in residential and commercial construction in addition to countertop, furniture, and other related industrial applications, as well as decorative applications.

Background The use of agricultural residue or "straw" in the production of biocomposite panels is known. The first development of these biocomposite panels used wheat straw and was originally formulated in Europe over three decades ago. In the early 1990's a renewed interest in this technology occurred due to the uncertainty of the wood supply as wood particleboard was placed on allocation during the late 1980's.

The early developments of such products typically used a cereal grain such as wheat straw as a filler. In the manufacture of cereal grain straw particleboard, an isocyanate resin is conventionally used to fuse the particles together. The advent of MDI (diphenylmethane diisocyanate) resin technology, for example, has allowed the use of a 100% cereal grain straw particleboard. Currently, there are commercial scale plants that will produce a "wheat board" using MDI resin technology. MDI is primarily used for agriculture particleboard as it has a very good affinity to bond most materials together including wood, agricultural fibers, and these materials to metal or plastic, for example. Such wheat board is not classified as "structural" material.

Structural material is defined as material that meets the mechanical and physical properties defined by the structural performance standards of the PFS Research Foundation entitled the PFS Research Foundation's Performance Standards for Policies for Structural-use Panels (PS-2), or equivalent grade requirements (e.g., CSA standards). These are standards of certification to assure that panel products satisfy the structural requirements of the application for which they are intended. Consequently, these standards are performance- based and not intended to address how panels are to be manufactured. Primary supporting documents of PS-2 are U.S. Department of Commerce Voluntary Product standards based on structural use panels such as structural use plywood, oriented strand board (OSB), and other biocomposite materials. The most widely used structural panels in home construction are plywood and OSB. OSB includes engineered, matt-formed, panel products made of strands, flakes, or wafers sliced from small diameter, round wood logs and bonded with an exterior-type binder under heat and pressure. Strand dimensions are predetermined and have a uniform thickness. The common strand geometries use a combination of strands up to 6 inches (150 mm) in length, 1 inch (25 mm) in width, and 0.025 inch (0.635 mm) to 0.035 inch (0.889 mm) in thickness. OSB panels consist of layered matts with the exterior or surface layers consisting of strands aligned in the long panel direction and inner layers consisting of cross- or randomly-aligned strands. The strength of OSB panels comes mainly from the uninterrupted wood fiber, interweaving of the long strands or wafers, and degree of orientation of strands in the surface layers. Binders are combined with the strands to provide internal strength and rigidity.

Plywood is a biocomposite panel derived from multi-layers of wood veneers that are glued together, typically in three or more layers. Plywood is a flat panel built up of sheets of veneer called plies, united under pressure by a bonding agent to create a panel with an adhesive bond between plies; it can be made from either softwoods or hardwoods. Plywood is always constructed with an odd number of layers with the grain direction of adjacent layers perpendicular to one another. Since layers can consist of a single ply or of two or more plies laminated such that their grain is parallel, a panel can contain an odd or even number of plies but always an odd number of layers. The outside plies are called faces or face and back plies; the inner plies are called cores or centers; and the plies with grain perpendicular to that of the face and back are called crossbands. The core may be veneer, lumber, or particleboard, with the total panel thickness typically not less than 1.6 mm (1/16 inch) or more than 76 mm (3 inches). The plies may vary as to number, thickness, species, and grade of wood. To distinguish the number of plies (individual sheets of veneer in a panel) from the number of layers (number of times the grain orientation changes), panels are sometimes described as three-ply, three-layer or four-ply, three-layer. The outer layers (face and back) and all odd-numbered layers

(centers) generally have their grain direction oriented parallel to the length or long dimension of the panel. The grain of even-numbered layers (cores) is perpendicular to the panel's length.

Generally, agricultural straw (e.g., cereal grain straw) has not been considered a suitable raw material for making structural materials. Cereal grain straw such as wheat, rice, and others have a unique strength to weight ratio and natural fiber geometry which make them desirable for making high strength biocomposite materials or biocomposite panels that meet or exceed structural performance ratings in accordance to US standards for building codes. Due to their unique geometries, chemical makeup, and bulk densities, however, traditional wood theory does not directly apply to the production of a biocomposite structural panel product.

To date only wood plywood or wood-derived OSB (oriented strand board) are the main structural composite panels that meet these structural code requirements. Research has been carried out using wood technology to create a structural straw panel in the past, but these materials have not met the specific code performance requirements discussed above. In addition, traditional wood processing equipment and processes used in the manufacture of structural wood products do not apply nor work for these unique fibers or straw.

Some limited basic home construction has used straw bales for construction where dense packing and size provide necessary strength and structural support; however, these projects do not meet the standard structural performance requirements of standard building codes. In fact, in many countries, the use of straw for construction is not permitted due to a common conception that straw is a poor building material.

U.S. Pat. No. 5,498,469 (Howard et al.) discloses a thin panel of compressed non- woody lignocellulosic material made by mixing short straw pieces with a binder. The panel was used as a core layer or core stock in a plywood laminate; thus, a thin layer of straw panel was sandwiched between two stronger wood sheets of plywood. Although this thin panel appears to perform its intended function, it does not have sufficient strength as a structural board.

Other methods of making panels using agricultural fiber and devices for making such panels are disclosed in, for example, U.S. Pat. Nos. 5,730,830 (Hall), 5,729,936 (Maxwell), and 5,728,269 (Kohno et al.).

U.S. Pat. No. 5,656,129 (Good et al.) discloses a method of refining wheat straw into fibers by cutting the straw to a length of two to four inches, wetting the straw, softening the straw by subjecting the straw to pressurized steam, and refining the softened straw in a pressurized mechanical refiner to produce fibers capable of being used in the manufacture of cellulosic board products. The straw fibers can be combined in any proportion to other fibers, such as wood fibers, and used in known dry, wet-dry, and wet board manufacturing processes to produce softboard, medium-density fiberboard, and hardboard products.

U.S. Pat. No. 5,932,038 (Bach et al.) discloses a method of fabricating a straw panel, board, or beam using straw split with rollermill technology typically used in grinding grain. The rollermill included two closely spaced shear rollers, each being substantially the same size and having a diameter of 200 mm to 800 mm. The straw was split in two pieces lengthwise with a preferred length comparable to wood OSB (preferably, 50 mm to 100 mm long). This process is not desirable at least because it is difficult and/or expensive to scale up to a commercial level, it creates an irregularly shaped strand that does not allow for good surface contact between strands, and it requires that a panel having a higher density than wood be made to meet structural requirements. The strands produced by this process are not desirable at least because they are not of a regular, flat shape and do not lay well together with high surface area contact between strands. Furthermore, upon pressing into a panel, they tend to form curved strands that do not lay flat. Also, the strands described in U.S. Pat. No. 5,932,038 (Bach et al.) must be oriented such that the longitudinal axes of the straw are aligned.

Commercially, wheat has been used to make particleboard where the wheat is ground into particles, with fine particles on the surface (e.g., 20 mesh (0.037 inch/0.95 mm opening) to 60 mesh (0.01 inch/0.25 mm opening)) and larger particles in the core (e.g., 10 mesh (0.075 inch/1.905 mm opening) to 20 mesh (0.037 inch/0.95 mm opening)). Such particleboards have not been found to meet structural material requirements.

Thus, there is still a need for materials, particularly structural materials, and methods that utilize agricultural straws, particularly non-wood cellulosic straws such as cereal grain straws. There is also a need for materials and methods that are more cost effective than wood OSB and more environmentally friendly.

Summary of the Invention

The present invention provides a structural biocomposite material, typically in the form of a multi-layer construction, that incorporates small strands of agricultural straw, typically non- ood cellulosic straws, such as cereal grain straw. The strands are preferably straight and flat. Because these strands are flat, there is a high level of surface area contact between adjacent strands. Furthermore, they can form an interwoven or interlocking structure that is believed to contribute to the unexpected mechanical and physical properties. Significantly, the material typically exhibits mechanical and physical properties that meet or exceed plywood or OSB products produced using wood flakes or wood veneers.

As used herein, "flat" strands have a substantially planar geometry, and "straight" strands are generally rectangular in shape with substantially parallel sides (the longer dimension), although the ends (the shorter dimension) may not be parallel. Thus, there may be slight variations from perfect planarity and perfect linearity in the shapes of "fiat" and "straight" strands according to the present invention. In one aspect, the present invention provides biocomposite material in the form of structural panels. These panels can be used in manufacturing a wide variety of products, including a substrate for laminates. Significantly, preferred panels (e.g., boards, beams, or nonflat or shaped objects) of the present invention have a strength that exceeds the strength of the straw panels described in U.S. Pat. No. 5,932,038 (Bach et al.) as shown in Figure 2 of Bach et al. by the modulus of rupture (MOR) and modulus of elasticity (MOE) values that are below that required for structural materials (at the same strand lengths). In addition to the long strands, a high density (e.g., about 6.6 pounds per cubic foot, 692 kilograms per cubic meter (692 kg/m³)) is required for the Bach et al. oriented split strawboard to achieve structural grade property requirements, as shown in Bach, "Structural Board Manufactured from Split Straw," Forest Products Research Society, May 19-20, 1999.

The biocomposite material of the present invention includes non- wood cellulosic straw, such as cereal grain straw, and a binder. Preferably, the binder is an isocyanate binder. The strands of the agricultural straw are typically of a sufficient size and geometry that they can be laid in a flat interwoven or interlacing pattern, and oriented if desired. Preferably, the strands are oriented in a parallel fashion in a face layer of a biocomposite panel, although orientation is not required to obtain structural grade material. Such products, as well as others, can be made due to the unique methods of making the strands of agricultural straw and biocomposite panels described herein. In one embodiment, a panel includes flat non-wood cellulosic strands and a resin, wherein at least about 40 wt-% (preferably, at least about 50 wt-%) of the panel includes flat non- wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

Herein, the particle size distributions are described in terms of the "mesh" of a screen, which correspond to an average opening size. For example, material that has a particle size distribution of about 4 mesh to about 12 mesh is material that passes through a 4 mesh screen, which corresponds to an average 0.21 inch or 5.46 mm opening, and is retained on a 12 mesh screen, which has a 0.06 inch or a 1.52 mm opening.

The structural biocomposite panel can be a homogenous construction, a two-layer construction (i.e., a construction with two general regions of different particle sizes), a three-layer construction (i.e., a construction with three general regions, e.g., two surface regions and a core region of a different particle size material), or more.

Preferably, the flat non-wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) form surface regions of the panel. More preferably, the panel includes a core that includes non- wood cellulosic strands having a different particle size than the non-wood cellulosic strands at the surfaces.

The non- wood cellulosic strands preferably include strands of cereal grain straw, such as wheat, oat, rice, barley, millet, rye, and combinations thereof. Preferably, the cereal grain straw is wheat. The resin is preferably an isocyanate resin. Alternatively, the resin can be an acid-catalyzed resin.

In another embodiment, a structural biocomposite panel of the present invention includes flat non- wood cellulosic strands and a resin, wherein at least about 40 wt-% of the panel includes flat non- wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm), and further wherein the panel includes surface regions that include the flat non- wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a core including non-wood cellulosic strands having a smaller particle size than the non-wood cellulosic strands at the surfaces.

In another embodiment, a structural biocomposite panel includes flat non-wood cellulosic strands and a resin, wherein the panel includes surface regions that include non- wood cellulosic strands and a core including non-wood cellulosic strands having a smaller particle size than the non-wood cellulosic strands at the surfaces.

In yet another embodiment, a structural biocomposite panel includes flat non- wood cellulosic strands and a resin, wherein at least about 40 wt-% of the panel includes flat non- wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length:width:thickness of about 100:10:1, and aparticle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

The present invention also provides a sample of flat non- wood cellulosic strands. Such a sample includes at least about 75 wt-% having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length: width:thickness of about 100: 10: 1 , and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

In another aspect, the present invention provides a method of producing strands of straw that are used in making the biocomposite material. According to a preferred method of the present invention, the straw is cut and sized using an impact classification milling (ICM) process that forms a unique geometry and particle size range that allows the strands to be laid in a flat interwoven or interlacing pattern. Advantageously, these strands are produced from cereal grain straw without the use of shear rollers, which give non-straight, non-flat, strands of a wide particle size distribution. In a preferred embodiment, the method includes: providing non-wood cellulosic straw; impact milling the non-wood cellulosic straw into strands; and classifying the strands to form a sample of flat non- wood cellulosic strands that include at least about 75 wt-% of the strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm). Preferably, the steps of impact milling and classifying are repeated. Preferably, providing non- wood cellulosic straw includes providing a bale of non- wood cellulosic straw and reducing the bale, using, for example, rotary slicing optionally in combination with classifying. Preferably, the step of classifying after impact milling can include, for example, air density classifying, rotary screemng, or a combination thereof. The method preferably includes subsequently drying the strands and optionally classifying the dried strands.

In one embodiment, a method provides a sample of flat non- wood cellulosic strands wherein at least about 75 wt-% of the strands have a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm). In another embodiment, a method provides a sample of flat non- wood cellulosic strands wherein at least about 75 wt-% of the strands have an average ratio of length: width:thickness of about 100:10:1. In still another embodiment, a method provides a sample of flat non-wood cellulosic strands wherein at least about 75 wt-% of the strands have a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

The present invention also provides methods of preparing a structural biocomposite panel. In one embodiment, the method includes: providing non- wood cellulosic strands coated with a resin; forming a matt having larger strands on the surfaces of the matt and smaller strands toward the core; and compressing the matt to form a structural biocomposite panel. Preferably, prior to compressing, the method includes applying soap, wax, or oil to the matt. If desired, the matt is compressed on a screen.

In one preferred embodiment, the matt is formed using a reversed windformer to place larger strands on the surfaces of the matt and smaller strands toward the core. Preferably, the windformer includes an orienting device for orienting the strands closer to the surface. In another preferred embodiment, the matt is formed using a reversed gradient screen former to place larger strands on the surfaces of the matt and smaller strands toward the core. Preferably, the reversed gradient screen former includes an orienting device for orienting the strands closer to the surface. Using the methods of fabricating a panel, the strands of cereal grain straw can be placed by size and geometry for optimal mechanical strength performance. That is, a matt (and resultant panel) of the present invention can include at least two layers (i.e., regions), which can differ by the particle size of the strands in the layers. Alternatively, the matt (and resultant panel) can have a substantially uniform distribution of strands.

For example, in certain embodiments, a biocomposite panel of the present invention includes a uniform distribution of particle sizes of strands throughout its thickness. In certain other embodiments, a biocomposite panel includes a gradation of particle sizes of strands throughout its thickness. In still other embodiments, a biocomposite panel includes face layers (surfaces) of strands of similar particle sizes and geometries and a core (third layer) of strands of a relatively smaller particle size, which may or may not be oriented. The core can also have a different density than that of the face layers (e.g., the face layers can have a higher density than the core layer, thereby providing a density gradient that provides additional mechanical strength). The face layers can have a higher percentage of resin than the core for enhanced moisture resistance and mechanical strength.

In another embodiment, a method of preparing a structural biocomposite panel includes: providing non-wood cellulosic strands coated with a resin; forming a matt having a substantially uniform distribution of strands; wherein at least about 75 wt-% of the strands have a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length:width:thickness of about 100:10:1, and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm); and compressing the matt to form a structural biocomposite panel. In still another embodiment, a method of preparing a structural biocomposite panel includes: providing non-wood cellulosic straw; impact milling the non- wood cellulosic straw into strands; classifying the strands to form a sample of flat non-wood cellulosic strands including at least about 75 wt-% of the strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm); coating the strands with a resin; forming a matt that includes the resin-coated strands; and compressing the matt to form a structural biocomposite panel. Preferably, the matt has larger strands on the surfaces of the matt and smaller strands toward the core, which can be accomplished using, for example, a reversed windformer or a reversed gradient screen former, either of which can optionally include an orienting device for orienting the strands closer to the surface.

Brief Description of the Drawings Figure 1 is a chart of MOE values of various panels including two panels of the present invention ("Structural Wheat 1" and "Structural Wheat 2") compared to wood OSB and the Canadian standard for wood OSB ("O-2");

"Para" is parallel to the panel length; "Perp" is perpendicular to the panel length. Figure 2 is a chart of MOR values of various panels including two panels of the present invention ("Structural Wheat 1" and "Structural Wheat 2") compared to wood OSB and the Canadian standard for wood OSB ("O-2");

"Para" is parallel to the panel length; "Perp" is perpendicular to the panel length. Figure 3 is a representation of the volume of one OSB wood strand compared to about 160 flat strands made according to a method of the present invention.

Figure 4 is a representation of the number of OSB wood strands in a surface region of a panel prior to pressing relative to the number of flat strands made according to a method of the present invention.

Figure 5 is a schematic of the process for preparing strands and panels according to the present invention. Figure 6 is a schematic of a rotary sheer with optional retention screen for the initial bale reducing.

Figure 7 is a schematic of a rotary screen classifier and an air classifier in combination with an impact mill with hammers having sharp knife-like leading edges and a screen.

Figure 8 is a cross-section of a straw showing the angles at which the straw tubes are cut using an impact milling process compared to a rollermilling process.

Figure 9 is a graph of particle size distributions of strands made according to a process of the present invention compared to material made by a conventional process.

Figure 10 is a bar chart of particle size distribution of strands made according to a process of the present invention compared to material made by a conventional process. Figure 11 is a bar chart of MOE values of panels of the present invention made with various resin levels compared to the Canadian standard for randomly formed wood OSB ("R-l OSB").

Figure 12 is a bar chart of MOR values of panels of the present invention made with various resin levels compared to the Canadian standard for randomly formed wood OSB ("R- 1 OSB").

Figure 13 is a bar chart of water immersion values of panels of the present invention made with various resin levels compared to wood OSB.

Figure 14 is a schematic of a reversed gradient screen system for panel formation. Figure 15 is a schematic of a reversed windforming system for panel formation.

Figure 16 is a chart of MOE values of various panels including one panel of the present invention ("Structural Wheat 3") compared to conventional oriented split strawboard (OSSB) described in Bach, "Structural Board Manufactured from Split Straw," Forest Products Research Society, May 19- 20, 1999, and the Canadian standard for wood OSB ("O-2"); "Para" is parallel to the panel length; "Perp" is perpendicular to the panel length.

Figure 17 is a chart of MOR values of various panels including one panel of the present invention ("Structural Wheat 3") compared to conventional OSSB described in Bach, "Structural Board Manufactured from Split Straw," Forest Products Research Society, May 19-20, 1999, and the Canadian standard for wood OSB ("O-2"); "Para" is parallel to the panel length; "Perp" is perpendicular to the panel length.

Detailed Description of Preferred Embodiments

The present invention provides biocomposite materials, systems, and methods. Significantly, the biocomposite materials are made from agricultural straw in the form of very small, flat, and straight strands. Advantageously, the biocomposite materials of the present invention can be used as structural materials.

The structural materials, typically in the form of panels (e.g., boards, beams, or nonflat or shaped objects such as chair components, automotive components, etc.), can be used for a variety of applications. For example, the panels of the present invention can be used as prepared (e.g., in an unsanded form) as a replacement for OSB or plywood construction panels in home building or industrial usage. A sanded form of the panels of the present invention has a high degree of surface integrity that allows them to be replacements in sanded plywood applications. Furthermore, they can be laminated, as for example, by a medium density overlay for painting. As defined above, structural materials meet the mechanical and physical properties defined by the structural performance standards of the PFS Research Foundation entitled the PFS Research Foundation's Performance Standards for Policies for Structural-use Panels (PS-2) or equivalent grade requirements (e.g., CSA standards). For example, referring to Figures 1 and 2, graphs are shown comparing the modulus of rupture (MOR) and modulus of elasticity (MOE) in both the parallel and perpendicular directions to the panel length (or relative to the longitudinal axis strand) of two different panels of the present invention compared to wood OSB that meets or exceeds the PS-2 standard. Also shown for comparative purposes are the MOR and MOE standards for CSA0437.0-93, the Canadian standard referred to herein. Thus, the material of the present invention meets or exceeds the MOR and MOE values for both PS-2 and CSA standards (generally, an MOR of at least about 4000 psi (27.6 MPa) and an MOE of at least about 700,000 psi (4827 MPa)).

Although the inventors do not wish to be bound by theory, it is believed that this is because the strands formed from the agricultural straw used to make the panels are formed into flat, generally rectangular pieces with straight sides. Because of this very uniform geometry, the strands can be layered in an interwoven, interlacing manner to form a large number of layers, particularly within the surface regions of a panel.

These strands are generally of a much smaller size than those included in wood OSB. For example, as shown in Figure 3, one OSB wood strand, which is shown as an oval, but can be of a variety of shapes, is the same volume as about 160 strands of the present invention. These smaller strands used to prepare panels of the present invention are described in greater detail below. Preferred wood OSB flake thickness is about 0.030 inch (0.762 mm) with a strand volume of about 0.12 cubic inch (1.96 cubic centimeters (cm³)), whereas a preferred strand of the present invention is about 0.01 inch (0.254 mm) thick with a strand volume of about 0.0007 cubic inch (0.011 cm³). Thus, each surface region of a 23/32-inch (18.2-mm) wood OSB panel is approximately five strands thick.

These surface regions (also referred to herein as layers, although they are not preformed layers laminated together or layers with a distinct dividing line) provide the mechanical strength. By comparison, the surface regions of a panel of the present invention are each over 15 strands thick, arranged in an interwoven or interlacing fashion. This is shown in Figure 4, which shows the relative number of layers of strands prior to pressing. This differential results because the strands used to prepare the panels of the present invention are flat, straight, and are long and narrow with a length, width, and thickness ratio of about 100:10:1 as compared to the shape of an OSB wood strand, which varies in thickness, is not uniform in shape, and has a length, width, and thickness ratio averaging 12:3:1. The strands of the present invention are also very different from the strands of U.S. Pat. No. 5,932,038 (Bach et al), which would generally be in the shape of a half circle (a straw split in half) and are of various lengths, and typically not straight nor flat strands.

The flat and straight shape of the strands of the present invention allows for high surface contact between strands and an interweaving or interlacing effect. In addition to having significantly more layers in both major surface regions, as compared to wood OSB, the bulk densities of the strands of the present invention are lower than wood flakes. This allows for a higher compression ratio (e.g., 10:1) of these surface layers that yield higher mechanical performance in the structural biocomposite panel compared to wood strands (e.g., 5:1). Also, significantly, compared to wood OSB, the materials of the present invention typically have higher water resistance because of the combination of resin and a naturally more water-resistant agricultural material. Furthermore, there is good bonding between the strands and the resin because there is relatively high contact area between the individual strands, which results in high strength. This is discussed in greater detail below with respect to the resin.

Biocomposite panels of the present invention can have a wide range of thicknesses and densities. Typically, for structural panels, the thickness of a panel is about 0.25 inch (6.35 mm) to about 1.5 inches (38 mm), although panels in a wide variety of thicknesses can be made if desired. Typically, panels can be made according to the invention in a wide variety of densities. For example, densities ranging from about 35 pounds per cubic foot (pcf) (561 kilograms per cubic meter (kg/m³)) to about 55 pcf (881 kg/m³) can be achieved, with a preferred range of about 40 pcf (641 kg/m³) to about 50 pcf (801 kg/m³). The surfaces can have higher densities than the cores, which can provide additional strength. In a preferred construction, the two surface layers equal 60% of the weight of the biocomposite and the balance of the weight (40%) is in the core. Panels of the present invention can be of uniform or non-uniform construction and have uniform or non-uniform properties throughout any one panel for the desired effect. For example, in certain embodiments, a biocomposite panel can include a uniform distribution of particle sizes of strands, geometries of strands, resin contents, and/or densities throughout its thickness. In certain other embodiments, a biocomposite panel can include a gradation of particle sizes of strands, geometries of strands, resin contents, and/or densities throughout its thickness. In certain preferred embodiments, the panels are in a multi-layer construction. Multi-layered constructions can include layers formed by differing particle sizes of the strands, differing densities, differing resin contents, etc. For example, a panel can include a core of one or more materials having a finer (i.e., smaller particle size) and/or geometry and at least one face layer, and preferably both face layers, of one or more materials having a higher aspect ratio and size than that of the core material. The material can be oriented or not. Preferably, the strands of the face layers are oriented, although this is not a requirement for structural grade material. A panel can include a core having a different density than that of the face layers (e.g., the face layers can have a higher density than the core layer, thereby providing a density gradient that provides additional mechanical strength). Also, the surface layers can have a higher percentage of resin than the core for enhanced moisture resistance and mechanical strength.

Agricultural Strands

The agricultural strands used in the biocomposite materials of the present invention are "non-wood cellulosic straw," which includes cereal grain straws as well as other non-wood cellulosics. As used herein, the terms "cereal grain straw" and "cereal straw" refer to the stems from a wide variety of grain crops commonly used for food and feed sources. Some examples of such cereal grain straws include, but are not limited to, wheat (summer, winter, durum, semolina, etc.), oat, rice, barley, millet, rye, the triticum genus of cereal grasses, prairie grasses of the plains states, flax, and cannola. The term "non- wood cellulosic straw" also includes other cellulosic materials that are cereal strawlike in structure, such as bamboo, as well as soybean, knaf, hemp, and sugarcane straw.

The non-wood cellulosic straws can be formed into the desired geometry and particle size range by a number of methods, including, for example, hammer milling, chopping, knife refining, and grinding. Preferably, the strands are formed by an impact classification process, which is described in greater detail below. This process of the generation of these small, linear (i.e., straight), flat strands are less expensive to produce than wood OSB strands. Thus, the resultant biocomposite panels can be of lower cost than traditional structural wood composite panels.

Individual strands used to prepare the materials of the present invention are generally two-dimensional, preferably sliced, classified, strands (i.e., not split or chopped). They are extremely flat and thin. The thickness is generally very uniform, with a variation of typically less that 0.001 inch (0.025 mm), as compared to wood OSB strands that can vary well over 0.02 inch (0.508 mm) across a strand. In addition the edge profile of an individual strand is defined and does not round off as do wood OBS strands.

Individual strands used to prepare the materials of the present invention are flat and straight (i.e., linear). Preferably, they are of a length of no greater than about 2 inches (50 mm), and more preferably, no greater than about 1.5 inch (38 mm). Preferably, they are of a length of at least about 0.25 inch (6.35 mm), and more preferably, at least about 0.5 inch (12.7 mm). Preferably, they are of a width of no greater than about 0.1 inch (2.54 mm), and more preferably of a width of at least about 0.05 inch (1.27 mm). Alternatively stated, the strands are of a size sufficient to pass through no larger than a 4 mesh (5.46 mm opening) screen and be held on no smaller than a 12 mesh (1.52 mm opening) screen. The thickness of the strands is dependent on the type of straw.

It is believed that the ratio of length to width (i.e., aspect ratio) contributes to the structural and physical properties of the material. For particularly preferred results, it is believed that the average ratio of length to width to thickness is about 100:10:1. The geometry of the straw used in making the materials described in U.S. Pat. No. 5,932,038 (Bach et al.) is a strand that is "split" in half, thus being half of a cylinder (i.e., curved), which is typically what results from rollermilling. In addition, these strands are preferably 2 inches (50 mm) to 4 inches (100 mm) in length and are "crimped" by the means of their rollermill process. Thus, they are not typically flat nor straight, thus requiring a significantly higher density and/or strand lengths to achieve structural grade performance. For example, the preferred lengths of the strands of U.S. Pat. No. 5,932,038 (Bach et al.) are at least 40 mm for structural MOR and MOE values (as shown in Figure 2 of U.S. Pat. No. 5,932,038). Furthermore, the split straws must be oriented, such that the longitudinal axes of the straws are aligned, to obtain structural grade materials.

It is also believed that the particle size range contributes to the structural and physical properties of the material. For particularly preferred results, it is believed that the range of particle size is about 4 mesh to about 12 mesh, which translates to about a 0.21 inch/5.46 mm opening to about a 0.06 inch/1.52 mm opening, and more preferably, about 4 mesh to about 10 mesh, which translates to about a 5.46 mm opening to about a 1.9 mm opening. As used herein, "particle size" refers to material that passes the larger mesh screen and is retained on the smaller mesh screen.

In sum, "optimal" strands have a length of at least about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) with a preference of about 0.75 inch (19 mm) to about 1.5 inch (38 mm), a width of at least about 0.005 inch (0.127 mm) and no greater than about 0.1 inch (2.54 mm) with a preference of about 0.050 inch (1.27 mm), an average ratio of length:width:thickness is about 100:10:1, and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

Significantly, over 160 strands of agricultural straw described above have the same weight as one strand of wood OSB. In addition these strands have a higher tensile strength than wood. For multi-layered biocomposites, the faces of a biocomposite panel are made of at least about 15 layers of these strands. Core material can have a wider range of strand geometry and particle sizes than the face material. This can result from the addition of fine material left over from the impact classification processing, as described in greater detail below. Face material that is smaller in size or not optimal as defined above can be detrimental to overall mechanical and physical performance of the biocomposite. Preferably, for good mechanical and physical properties, the face layers typically have at least about 50%, and more preferably, at least about 80%» material within the ranges described above.

Preparation of Agricultural Strands

Suitable strands are produced using a rotary slicing process, or alternative bale-reducing method followed by an impact classification milling (ICM) process, wherein cereal grain straw is sliced and fragmented by impact into a range of strands having a desired mesh size, density, and physical geometry. The ICM process produces a narrower aspect ratio strand at a specific length than do conventional process technologies that grind or split the straw. Such conventional processes cannot control the length, uniformity, or geometry of the strands to the degree that the ICM process can. Thus, conventional process technology produces a wide range of strand sizes and geometries that are not sorted or classified within the process. The ICM process produces an optimal strand geometry, being of a narrow aspect ratio and a specific flat planar geometry, as described above. Advantageously, this provides for increased fiber weaving and interlacing of the surface layers of the biocomposite. This effect makes it possible to achieve a structural mechanical performance using much shorter fibers than wood or other non-wood materials. The basic process is shown in Figure 5. Processing of large compacted bales of straw (e.g., wheat or rice) can be done using a bale reduction process, such as a slicing or bale-breaking process.

The slicing process involves a rotary cutter or rotary impact drum that both cuts and breaks the strands through impact means. This machine, shown in Figure 6, uses a rotating drum 10 with sharp cutter knives 12 attached to the rotating drum 10. The sharpness, length, and speed of the knives reduces the strands to specific lengths whereas the speed of the rotary action controls the breakage of the strands. These knives can have either a triangular or square shape with a sharp tip or sharp edge on the leading edge of the cutter. Slicing of a bale in this manner reduces the straw to a strand geometry and size at or larger than the optimal geometry and size to reduce undesired fines. Typically, the strands are cut into lengths of about 2 inches (50 mm) to about 6 inches (150 mm).

Preferably, a screen 14 is mounted on the back-side of the drum to increase retention of the strands in the rotary cutter for better control of size and geometry of the strands, and to perform the first level of classification, although a screen is not required. Preferably, the screen is mounted within about 0.25 inch (6.35 mm) of the knives on the drum and follows the contour of the drum covering up to about one-half (typically, about one-quarter to about one-half) of the circumference of the drum. The screen size is typically between about a 0.75-inch (19-mm) hole size to about a 4-inch (101.6-mm) hole size with optimal being about a 2-inch (50-mm) hole size. Such rotary cutters are commercially available without these screens from companies such as Vermeer (Pella, I A) and Deweze (Harper, KS).

The process subsequently involves impact classification milling, which involves screening or classification and impact milling, as shown Figure 5. The first part of the process uses a rotary screen classifier 20 and/or an air density classifier 30, which classifies the fibers primarily by width or arc angle of a tube, as shown in Figure 7. The strands of suitable size are removed and the material that is longer, wider, and inclusions are transferred to an impact mill 40. Unwanted fines can be removed at this point.

In a rotary screen classifier 20 there can be one or more screens of different mesh size (e.g., a 4 mesh rotary screen 22 and a 10 mesh rotary screen 24). The material that is retained on screen 22 is transferred to the impact mill 40 along line 26, the material that is retained on screen 24 is of desirable particle size and is transferred to a dryer along line 27, as shown in Figure 5, and the fines and dust are removed at 28. In an air density classifier 30, straw input is carried by air from air source 32 along lines 34, 36, and 38. The larger, heavier particles will drop in a shorter distance (e.g., along line 38) whereas the smaller finer particles will travel further (e.g., along line 34) and the material of a desirable particle size will travel, for example, along line 36. These portions can be collected and transported to their desired locations for drying, further impact milling or storage.

An impact mill 40 is similar to a standard hammer mill. This milling process is different from standard hammer milling, however, by reconfiguring the hammers 42 to have a sharp knife-like leading edge with a screen 44 concave in the mill, as shown in Figure 7. Other modifications may also be possible for the same results. The screen size is typically between about a 0.75- inch (19-mm) hole size to about a 2-inch (50-mm) hole size with optimal being about a 1-inch (25.4-mm) hole size. Material can be transferred back and forth between the rotary or air density classifier and the impact mill to reduce the strands to the desired particle size. That is, the strands are separated by mesh size and materials larger than desired are resliced and impact reduced in this looping system. Typically, using this process, the strands are cut into arc angles of 90 degrees or less, as shown in Figure 8. This is compared to how a straw tube is sliced in half using a rollermill, for example.

Traditional processes for wood or non- wood biocomposites use various forms of grinders or hammer mills to reduce fiber into a plurality of geometries typically very small with a wide range of sizes. In addition, such traditional processes create significant amounts of fines or dust, which is detrimental to manufacturing any higher performance panel and creates a waste problem within the operation. For a comparison of particle size distributions of a conventional particle reduction process versus the process of the present invention, see Figures 9 and 10.

For example, U.S. Pat. No. 5,932,038 (Bach et al.) discloses the use of grinding or hammer milling processes to grind a wide distribution of strands to produce a range of strands from over 4 inches (102 mm) to very fine micron size dust. This material with such a wide range of particle sizes is passed through a rollermill to break the tubes of straw in half creating a long half circle strand that does not lend itself for multi-layered surface biocomposites, as produced in the methods of the present invention. That is, the strands of U.S. Pat. No. 5,932,038 (Bach et al.) are generally not screenable, nor uniform, and are generally not straight. In contrast, the strands of the present invention are flat and straight, and thus, better able to be oriented and layered, which can provide better performance.

Compared to the structural biocomposite panels of the present invention, the homogeneous structural boards of U.S. Pat. No. 5,932,038 (Bach et al.) have poorer performance (e.g., MOR, MOE) at the same strand length. Furthermore, the grinding or rollermilling processes used in U.S. Pat. No. 5,932,038 (Bach et al.) do not generally allow for the use of straws other than cereal grain straws (i.e., nontubular straws). For example, soy straw would not readily split using these processes as they require a tubular geometry that can be flattened and split. Thus, the methods of the present invention allow for a greater input of materials (all non-wood cellulosic materials).

Particle reduction processing using a rotary cutter and impact classifier is preferably optimized to create a particle size distribution wherein the majority of the strands are of the desired geometry and particle size. Any strands larger than desired can be reprocessed. Significantly, the use of these two systems reduces the amount of strands that are finer or smaller than the desired geometry and particle size, thereby reducing the amounts of fines and dust created.

However, if desired these strands can be further classified by a variety of screening methods, particularly for removal of the relatively small amount of fines. Two forms of classification processes are preferred for production and sorting of strands. Rotary screening or air density classification processing works well to optimize distribution and efficiency. By selecting a mean strand size and geometry to be slightly larger than the optimum strand size and geometry, the amount of fines or dust created in the process is reduced. By classification, the optimum strands are separated and the larger material is reprocessed, thereby contributing to the reduction in the amount of fines and dust. Although fines are acceptable as a relatively small percentage of the core of the structural biocomposite panels, their inclusion is not recommended as they can diminish optimal biocomposite performance. Dust, however, is undesirable. Thus, the process of the present invention is optimized for minimal dust creation.

Rotary screening is preferably carried out using a rotary drum screener due to the linear nature of the fiber. Even through flat bed screeners may work, the rotary drum system optimizes the geometry and efficiency of the screen process. Traditional flat bed screeners classify particles that are typically uniform geometries. The strands of the present invention are long and narrow, and thus have a tendency to slide across a screen if not fluidized. To fluidize the fibers and optimize the distribution, a rotary system places the strands in the air to allow the strands to hit the screen in a random flow whereas a bed screener allows strands to flow across parallel to the screen. The tighter the distribution of optimal size, the better the properties of the structural biocomposite panel.

The unique geometry of the strands of the present invention lends itself to classification by airflow and density. The use of air classification to separate particles are used in other industries but only with very fine particles, typically less than 5 microns. A wind-forming machine used in the classification process described herein is manufactured by Schenck Company, Frankfurt, Germany, and modified by retrofitting to reverse the direction of the air flow and installing separate slots for each desired particle size. This process is preferred over rotary screeners. In addition to optimal classification, this form of classification can also be used for removal of foreign objects such as rocks, dirt, metal, wheat or rice seeds, and other inclusion that would be detrimental to the end biocomposite.

Thus, in the present invention an ICM process uses larger internal classification screens in conjunction with secondary classification by rotary or wind (air) density classification to optimize strand geometry and maximize the number of strands produced in the desired size region. Length of the strands is controlled by the primary slicer/ breaker and the hole size of the ICM milling process. Width is controlled by the speed of the ICM process and rotary or air density classification. Geometry is controlled by the balance of the process speeds, and relationship between screens in the rotary slicer and ICM process. The graph shown in Figure 10 shows strand distribution of the present ICM/Classification process compared to conventional methods.

Referring again to Figure 5, the strands are then dried while material outside of the desired particle size range is reprocessed or discarded. The straw typically has an initial moisture content of about 10 wt-% to about 30 wt-%. This is preferably reduced to less than about 10 wt-% and more preferably, to about 6 wt-% or less. For desired results, the moisture content of the strands upon combination with the resin system is preferably about 2 wt-% to about 10 wt-%, and more preferably about 3 wt-% to about 6 wt-%.

Subsequent to drying, additional classification can be optionally carried out using conventional methods, such as screening, as shown in Figure 5. Depending on the panel being made, various material fractions can be formed. For example, fines (e.g., material of about 30 mesh (0.024 inch 0.6 mm opening) to about 60 mesh (0.01 inch/0.25 mm opening) and dust (e.g., material of less than 60 mesh (0.01 inch/0.25 mm opening) and finer) can be removed at this stage, prior to blending with a resin system, as shown in Figure 5.

Resin System

The strands within the structural biocomposite panels, particularly the strands within the surface layers, as opposed to the resin, provide the greatest contribution to the strength of the panels. The resin bonds the strands together but does not provide a significant contribution to the strength. Resins useful in the resin system include, for example, thermosetting resins. A wide variety of resins can be used, such as an isocyanate, epoxy, phenolic, melamine, or urea-containing binder. Preferably, the thermosetting resin is an organic isocyanate, and more preferably an aromatic isocyanate. Preferably, the resin system includes one or more isocyanate resins.

A wide variety of isocyanate resins can be used. Suitable isocyanates include, but are not limited to, the aromatic isocyanates 4,4-diphenylmethane diisocyanate (MDI), toluene isocyanate (TDI), xylene diisocyanate (XDI), and methaxylene diisocyanate (MXDI). Preferably, the isocyanate is 4,4-diphenyl methane diisocyanate (polymeric MDI, which is available from ICI polyurethanes under the RUBINATE trademark and from Bayer under the MONDUR trademark). Many other isocyanates are commercially available from BASF and Dow, for example.

Although it is not preferred, an "acid-catalyzed resin system" as disclosed in WO 00/25998 can be used in one or more layers of the biocomposite panel. Such a system includes a resin (i.e., one or more resins) and an acid catalyst (i.e., one or more acid catalysts) for curing the resin. Examples or suitable resins for this embodiment include, but are not limited to, "phenolic resins," such as phenol formaldehyde and phenol-melamine formaldehyde, melamine, melamine formaldehyde, melamine-urea formaldehyde, melamine-urea-phenol formaldehyde, urea formaldehyde, or a combination of these resins. Examples of phenolic resins include those available as PMF 9707 (ARC Resins Corp., Longueuil, QC, Canada) and CRC 153 (Capital Resin Corp., Columbus, OH). Acids useful in the acid-catalyzed resin system include organic acids and mineral acids. Examples include, but are not limited to, formic acid, fumeric acid, sulfuric acid, as well as aromatic sulfonic acids such as benzenesulfonic acid, phenolsulfonic acid, and toluenesulfonic acid. In a preferred embodiment, the acid is formic acid, which can be obtained under the tradename ARC CATALYST 9700 (ARC Resins Corp.).

The types and amounts of resin (and acid) employed in the resin system typically depend upon the type of non- wood cellulosic material selected and the final biocomposite panel attributes desired. The types and amounts ofresin (and acid) will effect the cure rate of the resin and bond quality of the panel produced. Typically, a desired rate of cure is within a range of about 10 seconds to about 25 seconds per millimeter of final panel thickness. Resins employed in biocomposite panels of the invention are typically present in an amount of about 2 weight percent (wt-%) solids to about 10 wt-% solids, and preferably, about 2 wt-% to about 6 wt-%, based on the total amount of solids in a biocomposite panel. The resin amounts may be adjusted throughout the biocomposite independently for the surface layers and core layers. Higher amount ofresin can be used in the surface as compared to the core to add both mechanical and physical water resistance performance. Varying the resin percentage does not significantly affect the strength of the panels, as evidenced by the MOE and MOR values shown in Figures 11 and 12. Increasing the resin percentage increases the water resistance as evidenced by the data shown in Figure 13. At most resin levels, the water resistance is better for the panels of the present invention compared to wood OSB.

Preferably, MDI is used as the resin, although extenders such as soybean protein resins, vegetable oils or petroleum oils, modified waxes, etc., can be used. Additional additives may be added to the resin or to the strands to impart fire resistance (e.g., fire retardants such as borates), additional mold resistance (e.g., fungicides, biocides), termite resistance (e.g., insecticides), color (e.g., dyes or pigments), etc. Also, mold release agents, such as waxes and oils, can be added that also provide additional water resistance.

Preparation of Biocomposite Panels

As described above, a biocomposite panel according to the invention can be homogeneous with respect to types of resins, types of agricultural straws, and particle size distribution of the strands. The outer (i.e., face or surface) layer(s) (i.e., region(s)) of the structural biocomposite panels preferably include material of a larger particle size than is used in the inner (i.e., core) layer(s) (i.e., region(s)).

Forming of the structural biocomposite panels is preferably accomplished in a matt-forming machine that creates a loose matt of layered materials prior to thermal pressing. Uniform or nonuniform matts can be formed depending on the desired end use. Matts are preferably formed in two layers in which these layers are formed in gradients classified by strand geometry and density. The goal is to place the optimally sized strand closest to the surface and strands shorter than optimal near the core of the matt. Three layer matts may also be formed in which a third mechanical former can be used to lay a separate core layer to speed production rates of forming.

As shown in Figure 5, in a preferred embodiment, a bottom or first layer of strands is formed using a first matt former, an optional core or second layer of strands is formed using a second matt former, and a top or third layer of strands is formed using a third matt former. These matt formers can be any of a variety of matt formers, such as a reversed gradient screen forming system, shown in Figure 14, and a reversed wind forming system, shown in Figure 15. Gradient screen forming and orienting uses a unique system to classify and form the strands into a matt with the largest strands on the surface of the composite. Figure 14 shows a reversed gradient screen former 50 that includes a large box 52 which contains three screens, although any number of screens can be used. As exemplary, screen 53 is a 4 mesh (5.46 mm) screen, screen 54 is a 6 mesh (3.35 mm) screen, and screen 55 is a 10 mesh (1.905 mm) screen, of which the largest mesh screen is on the leading edge of the matt formation process. These screens may independently vibrate or move to assist in flowing the strands through the screens. A series of moving paddles 58 also assists in moving the strands forward to the screens and assists in allowing the strands to flow through the screens. A series of orienting disks 59 can be placed under the largest screen section which deposits the longest fibers on the surface. These disks can be 3/8 inch (9.53 mm) apart with a diameter of approximately 6 inches (15.2 cm). A second reversed gradient screen former, which is a mirror image of the reversed gradient screen former 50 shown in Figure 14, can be used to apply a second "layer" on the first "layer" applied by the first gradient screen former, thereby forming a gradient of strand length throughout the biocomposite. In this resultant gradient, the largest strands are on the surface and lighter or small strands are wind classified to the core. Even though two gradient screen formers can be used to create the two "layers," the biocomposite panel exhibits a continuous gradient of strand lengths and does not have a visible or mechanical layering effect.

Alternatively, a reversed windformer can be used to make a matt. Windformers are used in the forming of conventional wood particleboard matts for surface layers of three-layer particleboard biocomposites. The windformer system uses finer wood materials and classifies them in the surface layer placing the finest materials on the surface to create a smooth surface for optimal laminating conditions. The panel making process of the present invention uses a conventional windformer modified in at least two areas, as shown in Figure 15, to form a reversed windformer 60. First, the system is modified by reversing the wind (air) flow 62 to allow the larger strands to settle first along line 64, intermediate strands along line 66, and the finest strands settle along line 68. By decreasing air flow in the reverse direction, the strands can be laid flat to optimize performance of the biocomposite surface layer. Second, the area where the strands are laid allows for placement of a series of disks 69 at a 3/8 inch (9.53 mm) interval to align most of the strands in the parallel direction of the board for increased mechanical strength. A second reversed windformer, which is a mirror image of the windformer 60 shown in Figure 15, can be used to apply a second "layer" on the first "layer" applied by the first reversed windformer, thereby forming a gradient of strand length throughout the biocomposite. In this resultant gradient, the largest strands are on the surface and lighter or small strands are wind classified to the core. Even though two wind formers can be used to create the two "layers," the biocomposite panel exhibits a continuous gradient of strand lengths and does not have a visible or mechanical layering effect.

Prior to the thermal pressing of the matt a fine mist of wax, oil, soap or other water resistant liquid can be applied to the surfaces of the matt to impart additional surface water resistance and/or mold release. These surface liquids saturate into the surface of the matt and are cured within the thermal pressing process.

The matts are compressed under thermally heated platens. By controlling the closure speed and time, heat transfer and temperature, and degas cycle, the mechanical and physical properties of the structural biocomposite panels can be controlled within a specified range. Typical pressing temperatures range from about 250°F (121 °C) to about 450°F (232°C) with a more preferred range from about 320°F (160°C) to about 400°F (204°C). Typical pressing pressures range from 250 psi (1.7 MPa) to 750 psi (5.2 MPa) depending on the final desired panel density. The preferred pressure is about 500 psi (3.4 MPa). Closing speeds of the thermal press range from about 15 seconds to about 120 seconds depending on the surface density specification. Total cycle times or time under heat and pressure range from about 2 minutes to about 6 minutes depending on thickness and density. For example, a % inch (19 mm) panel typically has a curing rate under these typical pressures and temperatures of about 10 to about 18 seconds per millimeter thickness of the final panel. To optimize production or reduce cycle time within this process a fine woven steel screen is used beneath the matt and optionally on top of the matt. This reduces water vapor pressures in the matt allowing for faster degasing times during press release and offers advantages in compression of the matt. Cereal straw typically has different thermal dynamic issues as compared to higher density wood strands, thus these parameters with screens offer advantages in thermal transfer issues related to processing straw. Without these screens, cycle times for straw biocomposites typically exceed 18 seconds per millimeter and can be as long as 30 seconds per millimeter. The structural biocomposite panels of the present invention can be used sanded or unsanded. A sanded biocomposite panel of the present invention can have applied thereto a laminated medium or high density overlay by means of a secondary thermal fused pressing process. These overlays can be used to produce a structural paintable grade of panels for usage in the transportation industry. Other laminates include metal, melamine foil or a high pressure laminate (e.g., FORMICA).

Examples The invention will be further described by reference to the following detailed examples, which are exemplary and not intended to limit the invention.

Preferred Process

Referring to Figure 5, straw bales, typically large square bales, weighing approximately 1300 pounds (590 kg) are run through a primary slicer or bale breaking system, which opens the bales, slices the individual straws into strands, although tubular pieces can be formed, and reduces the size of the strands/tubular pieces. Typically, they are reduced to a length of about 1 inch (25.4 mm) to about 5 inches (127 mm), although there may be a percentage of finer materials. The preferred method of slicing uses a drum of 20-inch (500 mm) diameter with multiple cutting knives placed in a helical pattern around the drum and a retention screen covering at least the back ^lA of the drum. This drum spins at a speed ranging between about 200 revolutions per minute (rpm) and about 1500 rpm and uses between about 20 horsepower to about 100 horsepower to process about 5 tons (4500 kg) to about 10 tons per hour. This is significantly lower than the high horsepower requirement needed to generate wood OSB strand, which is done with a wood strand flaker equipment with horsepowers typically over 500 horsepower to create equal product flow.

In a second step, the material is classified using a rotary or air density classification system. This process classifies the strands primarily by width or arc angle of a tube. Strands of the desired size and geometry are separated and removed to bins whereas material that is longer, wider, and inclusions will go to the next step of the process. Unwanted fines can be removed at this point.

In a third step, the material is further reduced in particle size using an impact mill. This mill includes a combination of knives or sharp hammers in conjunction with a classification screen. This process controls two variables. The speed of rotation of the knives or hammers controls the impact and the breakage of the tubes into specific widths or arc angles. The classification screen controls both retention and length of the strands. This is connected with the classification system used in the second system. That is, the output of the impact mill flows directly back to the classification system for further classification, preferably to reduce the amount of fines generated and tighten the particle size distribution range.

The strands are then dried using high volumes of air and low degree of heat due to the thermal dynamic diffusion properties of these strands. This is clearly a significant cost advantage over wood OSB that requires extreme temperatures and high energy costs. Wood drying under these conditions emits large quantities of VOCs, and particulate emissions whereas non- wood cellulosic materials do not emit VOCs and relatively small amounts of particulate material. Drying is done in simple rotary dryer manufactured by companies such as MEC Corp., Minneapolis, KS. The dried material is then further classified if desired. This process separates the optimal strands for placement in a surface storage bin and the finer or less optimal strands are removed and placed in a core storage bin to be used for the core of the biocomposite if desired. This classification step, however, is not required with reversed windforming or gradient screen forming.

A resin, with or without various additives, is then blended with the strands. Typically, an MDI-containing resin is used in an amount of about 2 wt- % to about 10 wt-% solids, based on the total weight of the structural biocomposite panel. Typically, about 2 wt-% to about 3 wt-% resin content has proven to meet PS-2 Structural Performance Requirements.

A matt is then formed of the resin-strand mixture as a single layer, two layers, or three layers, for example. Typically, all of these matts have the optimal strands on the surface. A single layer matt can be formed using a simple mechanical forming head with optional orienting discs. A two-layer matt can be formed using a reversed windforming system or a reversed gradient screen forming head that places the optimal strands on both surfaces of the biocomposites. These matts are sprayed with a mold release, which is a water based wax, soap or oil applied directly to the surface of the matt or to a screen caul system by means of individual spray head similar to an automotive paint spraying system. Amounts of mold release or water resistant additives are applied sufficient to create unique water resistant or water proof surfaces on the structural biocomposite panel.

The matts are then compressed to form biocomposite panels. This process uses a high-pressure press with thermally heated platens. By controlling the closure speed and time, heat transfer and temperature, and degas cycle the mechanical and physical properties can be controlled within a specified range.

Example 1: Preparation and Testing of Wheat Panels (Figures 1 and 2)

The purpose of this example is to compare the mechanical properties (specifically bending - MOE and MOR) of a 'Structural Wheat Straw Board' (SWSB) of the present invention, wherein coarse longer wheat strands are used on the surface of the panel to improve strength and stiffness, with that of a wood strand Oriented Strand Board (OSB). A further comparison was made with a performance standard used for typical wood strand Oriented Strand Board (OSB).

Strand Formation. Wheat straw bales were first reduced with the Arasmith grinder (Arasmith Co., Rome, GA) that had a screen with 3-inch x 4- inch (76-mm x 102-mm) rectangular perforations. Resultant strands were further refined with a hammermill (Champion Corp., Mason City, IA) that had a screen with 1.0-inch (25.4-mm) diameter perforations. The resultant strands were then sent into the dryer where moisture was reduced to about 5 wt-% and collected.

Screening was done on the furnish using a Rotex screener (Rotex Screening Company, Cincinnati, OH) with 4, 10, and 60 mesh screens. Strands passing through the 4 mesh screen with a 0.215-inch (5.46-mm) opening but retained on the 10 mesh screen with a 0.075-inch (1.905-mm) opening were used for the surface layers of the panels. Strands passing through the 10 mesh screen but retained on the 60 mesh screen with a 0.01-inch (0.25-mm) opening were used in the core layer of the panel. One set of panels ("Structural Wheat 1") utilized core strands that had been further screened on the Sweco circular screener (Sweco Corp., Florence, KY) to further remove strands below a 30 mesh screen with a 0.0269-inch (0.683-mm) opening.

Resin and Blending. The resin used was an isocyanate resin available under the trade designation M20S PMDI from BASF. All wheat strands had water and resin applied to them in a ribbon paddle mixer. Water and resin was sprayed into the closed mixer with conventional air atomizing spray heads. Water was added to achieve approximately 9.0 wt-% moisture content on the surface layers and 7.0 wt-% in the core layer. Resin levels were targeted at 7.0 wt-% for the surface layers and 4.0 wt-% for the core layers. No other additives were used. Matt Forming. Matts were formed by hand leveling predetermined amounts of blended strands in a 24-inch x 24-inch (610-mm x 610-mm) square frame. Alignment of the surface strands was achieved by inserting a removable jig into the square frame onto which strips of metal had been welded parallel to each other with an approximate 0.25-inch to 0.375-inch (6.35-mm to 9.53-mm) gap between strips. Core layer strands were spread randomly without the use of the alignment jig. A 30/40/30 (weight percent) surface/core/surface ratio was used. Predetermined amounts were based on a dry solids weight basis from moisture content tests run after blending.

Press Parameters. Matts were pressed using a PHI press (PHI Co., Oakland, CA). Target density of the panels was 41 pounds per cubic foot (lb/ft ) (657 kg/m³) for Structural Wheat 1 and 43 lb/ft³ (689 kg/m³) for Structural Wheat 2. Pressures, closing speeds, compression time, and degas parameters remained constant for all sets.

Press temperature: 410°F (210°C)

Closing Time: 35 seconds

Compression time: 300 seconds (includes closing time) Degas time: 45 seconds

Maximum pressure 500 psi (3.45 MPa)

After the panels were removed from the press they were rough trimmed to 24 inches x 24 inches (61 cm x 61 cm) and hot-stacked. Panels were left overnight to cool.

Testing. After overnight cooling the panels were further trimmed to 20 inches x 20 inches (50.8 cm x 50.8 cm) and weighed and measured and the density was calculated. Individual samples for MOE/MOR tests were cut from the wheat panels parallel and perpendicular to the alignment of the wheat surface strands. The Wood OSB was purchased from a local lumber store and individual test samples were cut as per the standard cut pattern in the ANSI A208.1-1999 Particleboard publication.

MOE/MOR Testing was performed as per ASTM D 1037 - 96. Any deviations are noted. It is noted that these test methods may or may not be included, or may differ, from the set of tests typically performed on OSB or similar structural products as specified from country to country, however, they do provide for practical comparative results during product development.

No environmental preconditioning was performed on any of the test samples. All individual test samples were weighed and measured and the density was calculated prior to testing.

Sample Code# Test Direction Parallel or Perpendicular to strand alignment

Structural Wheat 1 para Parallel Structural Wheat 1 perp Perpendicular

Structural Wheat 2 para Parallel Structural Wheat 2 perp Perpendicular

Wood PS2-92 para Parallel Wood PS2-92 perp Perpendicular

O-2 OSB para Parallel OSB Grade Property Requirement Values from Table B4 of O437.0-93 (A Canadian Standard)

O-2 OSB perp Perpendicular OSB Grade Property Requirement Values from Table B4 of O437.0-93 (A Canadian Standard) Test Results. Results in graphical format are shown in Figures 1 and 2. These figures demonstrate the material of the present invention meets or exceeds the MOR and MOE values for both PS-2 and CSA standards (generally, an MOR of at least about 4000 psi (27.6 MPa) and an MOE of at least about 700,000 psi (4827 MPa)).

Example 2: Preparation and Screening of Strands (Figures 9 and 10)

Wheat straw bales were reduced using a rotary drum grinding system. For this test, an Arasmith rotary grinding system described in Example 1 was used. This material was then dried to an average moisture content of about 5 percent on a dry weight basis.

The material was processed in the first section of the ICM process in classification screens with three screen decks of 4 mesh, 10 mesh, and 60 mesh. This system was available from Rotex Screening Company (Cincinnati, OH). This system had a bed size of 5 feet by 12 feet (1.5 meters x 3.7 meters) and was set at the maximum angle determined by Rotex. The material of 4-10 mesh (5.46 mm-1.905 mm) was filtered off of the system whereas the material greater in size than 4 mesh (5.46 mm) was sent through a secondary impact milling process. Material finer than 10 mesh (1.905 mm) was separated out for usage in the core of the biocomposite. The material processed through the impact mill was sent back through the classification system and blended with the primary flow. Samples from the 4-10 mesh (5.46-mm- 1.905 mm) flow of strands were taken over a 1-hour period. The impact milling process used a Sprout Bauer impact mill (Sprout

Bauer Company, Marcy, PA). For this test, standard hammers were used with the mill set at 1750 rpm and a flow rate less than ^lA of the designed flow rate of the mill. In addition, the mill had a retention plate with 5/8-inch (15.88-mm) holes for this test. Material strands from the 4-10 mesh (5.46 mm-1.905 mm) stream were collected and fiber distribution was measured using standard ASTM screens. The testing system was a laboratory set-up of a WS Tyler Sieve Shaker using USA Standard Test Sieves. The material was screened for 3-5 minutes that simulated the retention time on the primary classification machine. This lab tester was set up with a series of screens that represent a range from 4 mesh (0.187 inches/4.75 mm opening for this particular 4 mesh screen) to 60 mesh (0.01 inch/0.25 mm opening). After the set screening times, each layer was weighed and graphed, resulting in the data in Figures 9 and 10. These graphs show that through this process the formation of the optimum strands described herein can be maximized with little waste.

Example 3: Preparation of Wheat Panels Varying Resin Level (Figures 11-13) The purpose of this example is to compare the mechanical (specifically bending - MOE and MOR) and moisture resistance (specifically 24-hour Immersion) properties of a 'Structural Wheat Straw Board' (SWSB) of the present invention, wherein coarse longer wheat strands are used on the surface of the panel to improve strength and stiffness, when resin levels have been varied. A further comparison was made with a performance standard used for typical wood Oriented Strand Board (OSB). A further comparison of moisture resistance (24-hour Immersion) was made with a typical wood strand Oriented Strand Board (OSB). Four (4) sets of laboratory test panels were produced with 'Random' (non-aligned to an axis) face and core layers with varying levels of resin.

Wheat strands were prepared and screened as described in Example 1 except no strands between 30 mesh (0.6 mm opening) and 60 mesh (0.25 mm) were removed. The resin used was an isocyanate resin available under the trade designation M20S PMDI from BASF. All wheat strands had water and resin applied to them in a ribbon paddle mixer. Water and resin was sprayed into the closed mixer with conventional air atomizing spray heads. Water was added to achieve approximately 9.0 wt-% moisture content on the surface layers and 7.0 wt-% in the core layer. Resin levels were varied between the sets of panels. No other additives were used. Resin Levels

Sample No. Surface Percent Core Percent

9%/5% 9.0 5.0

7% 4% 7.0 4.0

5%/3% 5.0 3.0

3%/2% 3.0 2.0

Matt Forming. Matts were formed by hand leveling predetermined amounts of blended strands in a 24-inch x 24-inch (610-mm x 610-mm) square frame. Each layer of strands was spread randomly. A 30/40/30 (weight percent) surface/core/surface ratio was used. Predetermined amounts were based on a dry solids weight basis from moisture content tests run after blending.

Press Parameters. Matts were then pressed using a PHI press. Target density of the panels was 43 lb/ft (689 kg/m ). Pressures, closing speeds, compression time, and degas parameters remained constant for all sets.

Press temperature: 410°F (210°C)

Closing Time: 30 seconds Compression time: 180 seconds (includes closing time)

Degas time: 45 seconds

Maximum pressure 500 psi (3.45 MPa)

After the panels were removed from the press they were rough trimmed to 24 inches x 24 inches (610 mm x 610 mm) and hot-stacked. Panels were left overnight to cool.

Testing. After overnight cooling the panels were further trimmed to 20 inches x 20 inches (508 mm x 508 mm), weighed, and measured and the density was calculated. Individual samples for MOE/MOR and 24-hour Immersion tests were cut from the wheat panels. The Wood OSB was purchased from a local lumber store and individual test samples were cut as per the standard cut pattern in the ANSI A208.1-1999 Particleboard publication.

MOE/MOR and 24-hour Immersion Testing was performed as per ASTM D 1037 - 96. Any deviations are noted. It is noted that these test methods may or may not be included, or may differ, from the set of tests typically performed on OSB or similar structural products as specified from country to country, however, they do provide for practical comparative results during product development.

Sample Code# Surface Resin% Core Resin%

9%/5% 9.0 5.0

7% 4% 7.0 4.0

5%/3% 5.0 3.0

3%/2% 3.0 2.0

Sample Code# Description

R-l OSB Random OSB Grade Property

Requirement

Values from Table B4 of O437.0-93 (A Canadian Standard)

Wood PS2-92 Typical Wood OSB Test Results. Results in graphical format are shown in Figures 11-13. These figures demonstrate that varying the resin percentage does not significantly affect the strength of the panels, as evidenced by the MOE and MOR. Also, increasing the resin percentage increases the water resistance. Furthermore, at most resin levels, the water resistance is better for the panels of the present invention compared to wood OSB.

Example 4: Preparation of Wheat Panels and Testing in Comparison to Oriented Split Strawboard (Figures 16 and 17)

The purpose of this example is to compare the mechanical properties (specifically bending - MOE and MOR) of a 'Structural Wheat Straw Board' (SWSB) of the present invention, wherein coarse longer wheat strands are used on the surface of the panel to improve strength and stiffness, with that of a wheat Oriented Split Strawboard (OSSB). A further comparison was made with a performance standard used for typical wood Oriented Strand Board (OSB).

Strand Formation. Wheat straw bales were first reduced with the Arasmith grinder that had a screen with 3-inch x 4-inch (76-mm x 102-mm) rectangular perforations. Resultant strands were further refined with a Champion hammermill that had a screen with 1.0-inch (25.4-mm) diameter perforations. The resultant strands were then sent into the dryer where moisture was reduced to about 5 wt-% and collected. Screening was done on the furnish using a Rotex screener with 4, 10, and 60 mesh screens. Strands passing through the 4 mesh screen with a 0.215- inch (5.46-mm) opening but retained on the 10 mesh screen with a 0.075-inch (1.905-mm) opening were used for the surface layers of the panels. Strands passing through the 10 mesh screen but retained on the 60 mesh screen with a 0.0099-inch (0.25-mm) opening were further screened on the Sweco circular screener to further remove strands below a 30 mesh screen with a 0.0269-inch (0.683 mm) opening. These -10, +30 strands were used in the core layer of the panels.

Resin and Blending. The resin used was an isocyanate resin available under the trade designation M20S PMDI from BASF. All wheat strands had water and resin applied in a ribbon paddle mixer. Water and resin was sprayed into the closed mixer with conventional air atomizing spray heads. Water was added to achieve approximately 10.0 wt-% moisture content on the surface layers and 6.0 wt-% in the core layer. Resin levels were targeted at 7.0 wt-% for the surface layers and 4.0 wt-% for the core layers. No other additives were used.

Matts were formed as described in Example 1.

Parameters. Matts were then pressed using a PHI press. Target density of the panels was 40 lb/ft³ (641 kg/m³). Pressures, closing speeds, compression time, and degas parameters remained constant for all sets.

Press temperature: 410°F (210°C) Closing Time: 30 seconds

Compression time: 180 seconds (includes closing time)

Degas time: 45 seconds

Maximum pressure 500 psi (3.45MPa)

Testing. After overnight cooling the panels were further trimmed to 20 inches x 20 inches (508 mm x 508 mm), weighed, and measured and the density was calculated. Individual samples for MOE/MOR tests were cut from the wheat panels parallel and perpendicular to the alignment of the wheat surface strands.

Sample Code# Test Direction Parallel or Perpendicular to strand alignment

Structural Wheat 3 para Parallel Structural Wheat 3 perp Perpendicular

*OSSB 4/41 para Parallel OSSB 4/41 perp Perpendicular

*OSSB 4/43 para Parallel OSSB 4/43 perp Perpendicular

*OSSB 4/47 para Parallel OSSB 4/47 perp Perpendicular

O-2 OSB para Parallel OSB Grade Property Requirement Values from Table B4 of O437.0-93 (A Canadian Standard) O-2 OSB perp Perpendicular OSB Grade Property Requirement

Values from Table B4 of O437.0-93 (A Canadian Standard)

*NOTE: The numbers in the codes for the OSSB samples denote the resin percentage and the density in pounds per cubic foot, hence '4/41 ' has 4.0% resin at 40.7 lb/ft³ (652 kg/m³), '4/43' has 4.0% resin at 43.2 lb/ft³ (692 kg/m³), and '4/47' has 4.0% resin at 46.6 lb/ft³ (747 kg/m³). This is compared to the average panel density of Structural Wheat 3 is 41.0 lb/ft³ (657 kg/m³).

Test Results. Results in graphical format are shown in Figures 16-17. These figures demonstrate that better performance can be achieved using shorter strands and lower densities than obtained with oriented split strawboard described in Bach, "Structural Board Manufactured from Split Straw," Forest Products Research Society, May 19-20, 1999.

All patents, patent applications, and publications are incorporated by reference herein as though individually incorporated by reference. Various modifications and alterations of this invention will become apparent to those skilled in the art without departing from the scope and spirit of this invention, and it should be understood that this invention is not to be unduly limited to the illustrative embodiments set forth herein.

Claims

What Is Claimed Is:

1. A structural biocomposite panel comprising flat non- wood cellulosic strands and a resin, wherein at least about 40 wt-% of the panel comprises flat non-wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

2. The structural biocomposite panel of claim 1 wherein the flat non-wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) form surface regions of the panel.

3. The structural biocomposite panel of claim 2 comprising a core comprising non- wood cellulosic strands having a different particle size than the non- wood cellulosic sfrands at the surfaces.

4. The structural biocomposite panel of claim 1 comprising a homogenous construction.

5. The structural biocomposite panel of claim 1 comprising a two-layer construction.

6. The structural biocomposite panel of claim 1 comprising a three-layer construction

7. The structural biocomposite panel of claim 6 wherein the flat non- wood cellulosic sfrands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) form surface regions of the panel.

8. The structural biocomposite panel of claim 1 wherein the non-wood cellulosic strands comprise strands of cereal grain straw.

9. The structural biocomposite panel of claim 8 wherein the cereal grain straw is selected from the group consisting of wheat, oat, rice, barley, millet, rye, and combinations thereof.

10. The structural biocomposite panel of claim 9 wherein the cereal grain straw is wheat.

11. The structural biocomposite panel of claim 1 wherein the non- wood cellulosic strands have a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm).

12. The structural biocomposite panel of claim 1 wherein the non-wood cellulosic strands have an average ratio of length: width:thickness of about 100:10:1.

13. The structural biocomposite panel of claim 1 wherein at least about 50 wt-% of the panel comprises flat non- wood cellulosic sfrands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

14. The structural biocomposite panel of claim 13 wherein the non- wood cellulosic strands have a particle size distribution of about 4 mesh (5.46 mm) to about 10 mesh (1.905 mm).

15. The structural biocomposite panel of claim 1 wherein the resin comprises an isocyanate resin.

16. The structural biocomposite panel of claim 1 wherein the resin comprises an acid-catalyzed resin.

17. The structural biocomposite panel of claim 1 which is sanded.

18. The structural biocomposite panel of claim 17 further comprising a laminate comprising metal, melamine foil, high pressure laminate, or a medium or high density overlay.

19. A structural biocomposite panel comprising flat non-wood cellulosic sfrands and a resin, wherein at least about 40 wt-% of the panel comprises flat non-wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm), and further wherein the panel comprises surface regions comprising the flat non-wood cellulosic sfrands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm) and a core comprising non-wood cellulosic strands having a smaller particle size than the non- wood cellulosic sfrands at the surfaces.

20. The structural biocomposite panel of claim 19 wherein the non- ood cellulosic strands have an average ratio of length: width:thickness of about 100:10:1.

21. A structural biocomposite panel comprising flat non- wood cellulosic strands and a resin, wherein the panel comprises surface regions comprising non-wood cellulosic sfrands and a core comprising non-wood cellulosic strands having a smaller particle size than the non- wood cellulosic strands at the surfaces.

22. A structural biocomposite panel comprising flat non-wood cellulosic strands and a resin, wherein at least about 40 wt-% of the panel comprises flat non-wood cellulosic strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length: width:thickness of about 100:10:1, and aparticle size distribution of about 2 mesh (6.35 mm) to about 12 mesh (1.52 mm).

23. A sample of flat non- wood cellulosic sfrands comprising at least about 75 wt-% of the strands have a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length: width:thickness of about 100:10:1, and aparticle size disfribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

24. A method of preparing flat non-wood cellulosic strands, the method comprising: providing non-wood cellulosic straw; impact milling the non-wood cellulosic straw into strands; and classifying the strands to form a sample of flat non- wood cellulosic strands comprising at least about 75 wt-% of the strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm).

25. The method of claim 24 wherein impact milling and classifying are repeated.

26. The method of claim 24 wherein providing non- wood cellulosic straw comprises providing a bale of non- wood cellulosic straw and reducing the bale.

27. The method of claim 26 wherein reducing the bale comprises rotary slicing.

28. The method of claim 27 wherein reducing the bale comprises rotary slicing and classifying.

29. The method of claim 24 wherein classifying after impact milling comprises air density classifying, rotary screening, or a combination thereof.

30. The method of claim 24 wherein the sample of flat non-wood cellulosic strands comprises at least about 75 wt-% of the strands having a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm).

31. The method of claim 24 wherein the sample of flat non- wood cellulosic strands comprises at least about 75 wt-% of the strands having an average ratio of length: width:thickness of about 100:10:1.

32. The method of claim 24 wherein the sample of flat non- wood cellulosic strands comprises at least about 75 wt-% of the sfrands having a particle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm).

33. The method of claim 24 wherein the non- wood cellulosic strands comprise strands of cereal grain straw.

34. The method of claim 33 wherein the cereal grain straw is selected from the group consisting of wheat, oat, rice, barley, millet, rye, and combinations thereof.

35. The method of claim 34 wherein the cereal grain straw is wheat.

36. The method of claim 24 further comprising drying the strands.

37. The method of claim 36 further comprising classifying the dried sfrands.

38. A method of preparing a structural biocomposite panel, the method comprising: providing non-wood cellulosic strands coated with a resin; forming a matt having larger strands on the surfaces of the matt and smaller strands toward the core; and compressing the matt to form a structural biocomposite panel.

39. The method of claim 38 wherein the matt is formed using a reversed windformer to place larger strands on the surfaces of the matt and smaller strands toward the core.

40. The method of claim 39 wherein the reversed windformer includes an orienting device for orienting the sfrands closer to the surface.

41. The method of claim 38 the matt is formed using a reversed gradient screen former to place larger strands on the surfaces of the matt and smaller strands toward the core.

42. The method of claim 41 wherein the reversed gradient screen former includes an orienting device for orienting the sfrands closer to the surface.

43. The method of claim 38 wherein the matt comprises at least two layers.

44. The method of claim 38 further comprising applying soap, wax, or oil to the matt prior to compressing.

45. The method of claim 38 wherein the matt is compressed on a screen.

46. A method of preparing a structural biocomposite panel, the method comprising: providing non-wood cellulosic sfrands coated with a resin; forming a matt having a substantially uniform distribution of strands; wherein at least about 75 wt-% of the strands have a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm), a width of about 0.005 inch (0.127 mm) to about 0.1 inch (2.54 mm), an average ratio of length:width:thickness of about 100:10:1, and aparticle size distribution of about 4 mesh (5.46 mm) to about 12 mesh (1.52 mm); and compressing the matt to form a structural biocomposite panel.

47. The method of claim 46 further comprising applying soap, wax, or oil to the matt prior to compressing.

48. The method of claim 46 wherein the matt is compressed on a screen.

49. A method of preparing a structural biocomposite panel, the method comprising: providing non-wood cellulosic straw; impact milling the non-wood cellulosic straw into sfrands; classifying the strands to form a sample of flat non- wood cellulosic strands comprising at least about 75 wt-% of the strands having a length of about 0.25 inch (6.35 mm) to no greater than about 2 inches (50 mm); coating the sfrands with a resin; forming a matt comprising the resin-coated strands; and compressing the matt to form a structural biocomposite panel.

50. The method of claim 49 wherein forming a matt comprises forming a matt having larger strands on the surfaces of the matt and smaller sfrands toward the core.

51. The method of claim 50 wherein the matt is formed using a reversed windformer to place larger strands on the surfaces of the matt and smaller strands toward the core.

52. The method of claim 51 wherein the reversed windformer includes an orienting device for orienting the strands closer to the surface.

53. The method of claim 50 the matt is formed using a reversed gradient screen former to place larger sfrands on the surfaces of the matt and smaller strands toward the core.

54. The method of claim 53 wherein the reversed gradient screen former includes an orienting device for orienting the sfrands closer to the surface.

55. The method of claim 50 wherein the matt has a substantially uniform distribution of strands.

56. The method of claim 50 further comprising applying soap, wax, or oil to the matt prior to compressing.

57. The method of claim 50 wherein the matt is compressed on a screen.

58. The flat non-wood cellulosic strands preparable by the method of claim

24.

59. The panel preparable by the method of claim 38.

60. The panel preparable by the method of claim 46.