Multi-layer lignocellulosic products which are produced employing adhesive bonding agents and lignocellulosic particles have been known for many years. Adhesive bonding agents such as polymeric resins have been used in the formation of particulate wood products such as strandboard, particularly oriented strand board ("OSB"), waferboard, chipboard, particleboard and fiberboard. A resin such as an aldehyde and or an isocyanate resin exhibits excellent strength capabilities and is capable of bonding together the lignocellulose particles which form these particulate wood products.

Conventional mills which manufacture lignocellulosic particleboard, such as OSB, screen the smallest strands out of a particle furnish stream, typically employing a 1/4"screen. The 1/4"screen is used to remove about 15-20% of the strands from stream, and the"fines"which are segregated are burned as fuel.

Some mills employ a modified or two-stage screening strategy. The minus 1/499 fines are subsequently passed over a smaller mesh screen, and only the dust that passes though this screen is burned. The intermediate size strands are typically returned to the core layer. While this strategy helps improve the product yield, it leaves a considerable amount of small strands in the face layer and large strands in the core layer.

Some mills are operating without any strand size classification. In these mills, it is difficult to form uniform mats on the forming line because of the high level of dust in the furnish, and the large range of size in the strands passing through the forming heads. This strategy makes it especially difficult to make thicker high-strength products because of the poorly formed mat, and the high level of smaller strands and dust in all layers. Product yield can also be adversely

affected if the product density must be increased to compensate for a poorly formed mat or a subpar quality of strands in the face layers.

Accordingly, a need exists for a method and a product which overcomes the above-described problems.

SUMMARY OF THE INVENTION It has now been discovered that, in accordance with the present invention, a method can be provided for manufacturing multi-layer lignocellulosic products which exhibit a substantially lower density and a similar strength and stiffness as a multi-layer lignocellulosic product produced from a conventional multi-layer lignocellulosic mat. Stated another way, it has now been discovered that, in accordance with the present invention, a method can be provided for manufacturing multi-layer lignocellulosic products that exhibit a substantially higher strength and stiffness properties than a multi-layer lignocellulosic product at a similar density which is produced from a conventional multi-layer lignocellulosic mat.

This multi-layer lignocellulosic product comprises at least one core layer including a fraction of smaller lignocellulosic particles separated by size from a total fraction of lignocellulosic particles comprising a fraction of smaller and a fraction of larger lignocellulosic particles. It also includes at least one face layer including the fraction of larger lignocellulosic particles separated from the total fraction of lignocellulosic particles. The layers comprising the fraction of smaller lignocellulosic particles and the fraction of larger lignocellulosic particles are bonded together with an adhesive bonding material to form the multi-layer lignocellulosic product.

There are certain criteria which can be employed in defining a preferred quality of separation by size of the respective fractions of smaller and larger lignocellulosic particles when the present invention is utilized. The specific terms which characterize this preferred quality of separation are"Small Particle Separation Factor"and"Large Particle Separation Factor". Small Particle Separation Factor is defined as the weight percentage of lignocellulosic particles

in the Core Furnish that pass through a screen with 1/4"openings divided by the weight percentage of lignocellulosic particles in the Face Furnish that pass through a screen with 1/4"openings. Large Particle Separation Factor is defined as the weight percentage of lignocellulosic particles in the Face Furnish that are retained on a screen with 3/4"openings divided by the weight percentage of lignocellulosic particles in the Core Furnish that are retained on a screen with 3/4"openings.

A Small Particle Separation Factor and/or a Large Particle Separation Factor value of about 1 would indicate that no initial separation or classification by size of the lignocellulosic material has occurred. In the method of this invention, the Small Particle Separation Factor and/or a Large Particle Separation Factor with respect to the fraction of smaller lignocellulosic particles and the fraction of larger lignocellulosic particles, respectively, produced by the subject initial separation, is preferably at least about 2, more preferably at least about 2.5, and most preferably at least about 3.

The multi-layer lignocellulosic product exhibits a more uniform surface appearance than a multi-layer lignocellulosic product produced from a conventional multi-layer lignocellulosic mat which is not produced having the fraction of larger lignocellulosic particles in the face layer and the fraction of smaller lignocellulosic particles in the core layer.

In conventional methods for forming products from lignocellulosic particles, the presence of long particles and fines in the furnish results in a very broad particle size distribution which can cause problems in forming a uniform density mat. Conversely, by conducting the subject initial separation into respective fractions of larger and smaller lignocellulosic particles, a substantially narrower particle size distribution results which facilitates forming a mat having a more uniform density.

The subject invention, reduces the cost of manufacturing by achieving increased structural strength properties, while also maximizing the product yield and maintaining a lower product density. These benefits can be accomplished when the larger lignocellulosic particles are used in the face layer, and the smaller lignocellulosic particles are used in the core layer.

Another problem that this size segregation strategy solves is the difficulty that is experienced in trying to produce a uniform density mat composed of widely varying lignocellulosic particles sizes and the presence of dust. This is a severe predicament in mills that are using the prior strategies or have no plan at all to solve this problem. This invention supplies the face forming equipment and the core forming equipment with a much narrower lignocellulosic particles size distribution, making it possible to better design the forming heads for a specific size furnish and ensure a substantially more uniform mat formation. This promotes a reduction in product density and in strength variations within a finished product.

The present invention relates to a method for producing a multi-layer lignocellulosic product. More specifically, the method comprises providing lignocellulosic particles for use in producing the multi-layer lignocellulosic product. Preferably, the lignocellulosic particles are selected from a group consisting of lignocellulosic strands, lignocellulosic chips, lignocellulosic wafers and lignocellulosic fibers.

These lignocellulosic particles are initially separated by size into a plurality of fractions of lignocellulosic particles including a fraction of larger lignocellulosic particles and a fraction of smaller lignocellulosic particles.

Typically, the step of initially separating the lignocellulosic particles by size comprises screening of the lignocellulosic particles by size. However, it is recognized that the separation step of this invention can be effected using various types of equipment and is not limited to any particular device or machine.

In some instances, the step of initially separating the lignocellulosic particles by size can comprise providing a screening apparatus, introducing the lignocellulosic particles into the screening apparatus, passing the fraction of smaller lignocellulosic particles through the screening apparatus, and retaining the fraction of larger lignocellulosic particles which do not pass through the screening apparatus. A preferred screening apparatus for use in the method of this invention comprises a plurality of screens forming openings there between of about % 2" to 1 1/4'', more preferably 5/8"to 1", through which pass the fraction of smaller lignocellulosic particles for separating the fraction of smaller lignocellulosic

particles and the fraction of larger lignocellulosic particles. However, separation can be conducted using an apparatus other than a screening apparatus.

There is a problem of optimizing the moisture content of the lignocellulosic particles employed to form the product. Many factors can influence the optimum moisture content which include press design, the adhesive resin system used, the product thickness, etc. A further important factor which affects optimization of moisture content is the broad size variation of the lignocellulosic particles employed in forming the subject product. The differing sized lignocellulosic particles in the furnish exiting the dryer contain a wide- ranging spectrum of moisture contents making it difficult for uniform drying thereof to occur. In a furnish having a mixture of larger and smaller particles, conventional drying technology must take into account the fact that the larger particles have a significantly higher moisture content than the smaller particles.

In the method of the present invention, the lignocellulosic particles can be dried prior to initially separating of the lignocellulosic particles by size. Thus, the moisture content in the respective larger and smaller fractions created by the initial separation of the lignocellulosic particles will have a much narrower variation with respect to moisture content due to the fact that they have a much narrower variation in size distribution particularly as compared to the above-described conventional product manufacturing methods. More specifically, the lignocellulosic particles, prior to initial separation by size, can be dried to an average moisture content of not more than about 6 %, preferably not more than about 8 %, more preferably not more than about 10 %, most preferably not more than about 12 %, based on the weight of the lignocellulosic particles. In this case, the higher moisture content of the larger size fraction to be employed in the face layer, and the lower moisture content of the smaller fraction to be employed in the core layer, are of considerable benefit to the product manufacturer.

Alternatively, the fraction of larger lignocellulosic particles and the fraction of smaller lignocellulosic particles can be dried, subsequent to initially separating the lignocellulosic particles by size, and prior to treating the lignocellulosic particles with an adhesive bonding material. This substantially

reduces the presence of higher moisture content larger particles in the core layer where they can cause excessive internal pressure which can lead to product blowouts in formation press process. In this case, the fraction of smaller lignocellulosic particles are typically dried to an average moisture content of not more than about 6 %, preferably not more than about 8 %, more preferably not more than about 10 %, most preferably not more than about 12 %, based on the weight of the lignocellulosic particles. On the other hand, the fraction of larger lignocellulosic particles can be employed at a higher moisture level, namely, an average moisture content of preferably not more than about 10 %, more preferably not more than about 12 %, most preferably not more than about 15 %, based on the weight of the lignocellulosic particles.

In a preferred form of this invention, a fraction of fine lignocellulosic particles of substantially lesser average size than other particles in the fraction of smaller lignocellulosic particles are removed therefrom. In this case, the fraction of fine lignocellulosic particles which will preferably pass through a standard 14 Mesh sieve.

Next, the fraction of larger lignocellulosic particles and the fraction of smaller lignocellulosic particles are treated with an adhesive bonding material.

The adhesive bonding material can comprise any one of an isocyanate polymer, an aldehyde resin, an aldehyde resin-latex copolymer, and an isocynate resin-latex copolymer. Preferably, the adhesive bonding material comprises a diisocyanate and/or an aldehyde resin.

A multi-layer lignocellulosic mat is then formed. The mat has at least one core layer comprising the fraction of smaller treated lignocellulosic particles and at least one face layer comprising the fraction of larger treated lignocellulosic particles. The fraction of smaller treated lignocellulosic particles and the fraction of larger treated lignocellulosic particles are bonded together in the multi-layer lignocellulosic mat to form the multi-layer lignocellulosic product of this invention.

The multi-layer lignocellulosic product can exhibit a substantially lower density, a more uniform surface appearance, and similar strength and stiffness

properties as a multi-layer lignocellulosic product produced from a conventional multi-layer lignocellulosic mat which is not formed having the fraction of larger lignocellulosic particles in the face layer and the fraction of smaller lignocellulosic particles in the core layer. The subject multi-layer lignocellulosic product preferably has a thickness of from about 1/4"up to 1 1/4", more preferably from about 7/16"up to 1".

Regarding the density of the multi-layer lignocellulosic product, small decreases in product density will result in significant cost savings due to the large volume of product which is manufactured. Thus, the product of the subject invention can be formed have a density which is typically at least about 5 % lower, preferably at least about 8 % lower, and more preferably at least about 10 % lower, than the density of the conventional multi-layer lignocellulosic product.

Various physical properties of the multi-layer lignocellulosic product exhibit significant improvement. For example, the dry modulus of elasticity parallel (MOE) of the product herein is preferably at least about 10 % higher, more preferably at least about 15 % higher, and most preferably at least about 20 % higher, at the same density, than the MOE of a multi-layer lignocellulosic product produced from a conventional multi-layer lignocellulosic mat which is not formed having the fraction of larger lignocellulosic particles in the face layer and the fraction of smaller lignocellulosic particles in the core layer. Furthermore, the dry modulus of rupture parallel (MOR) of the product herein is preferably at least about 10 % higher, more preferably at least about 15 % higher, and most preferably at least about 20 % higher, at the same density, than the MOR of a multi-layer lignocellulosic product produced from a conventional multi-layer lignocellulosic mat which is not formed having the fraction of larger lignocellulosic particles in the face layer and the fraction of smaller lignocellulosic particles in the core layer. The test designation for MOR and MOE is the method described in ASTM D-1037, Section 11-20. There are also minor modifications to the sample size testing which is set forth in APA Standard PRP-108, Test Method S-14 for MOR & MOE test specimens.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic flow diagram illustrating a preferred method for forming a multi-layer lignocellulosic product of the present invention.

FIG. 2 is a schematic flow diagram illustrating another preferred method for forming a multi-layer lignocellulosic product of the present invention.

FIG. 3 is a block diagram illustrating the experiments used for forming a 38 lb/ft multi-layer lignocellulosic product of the present invention.

FIG. 4 is a graphical diagram illustrating the Dry MOR for 34-41 lb/ft3 density multi-layer lignocellulosic products of the present invention as compared to conventionally produced multi-layer lignocellulosic products of comparable density.

FIG. 5 is a graphical diagram illustrating the Dry MOE for 34-41 lb/ft3 density multi-layer lignocellulosic products of the present invention as compared to conventionally produced multi-layer lignocellulosic products of comparable density.

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT Referring now to FIGS. 1 and 2, novel methods of segregating and utilizing lignocellulosic particles for manufacturing particleboard products, such as strandboard, are provided that solves all of these problems.

For instance, a strandboard furnish can be initially separated into approximately equal portions with a device that separates the strands based on size, such as a screen device, more particularly a rotary screen device. Typically, a 3/4"screen size is used to accomplish this initial separation.

The larger size strand fraction is used to produce the face layer (s) of a strandboard panel. The smaller size strand fraction from this initial separation is used to form the core layer (s).

In the preferred mode, the smaller strands are further separated to remove a smaller portion of dust, typically not more than about 5%, preferably not more than about 4%, more preferably not more than about 3%, and most preferably not

more than about 2%, of the total lignocellulosic particle furnish. The dust, if removed, would normally be used as a source of plant fuel.

The separation (s) can be performed either after the drying operation or before. Performing the separation before the drying step makes it possible to optimize the core & face dryers for the strand size, thereby providing maximum flexibility to independently establish improved moisture content control in the face and core layers. In this case the fraction of larger lignocellulosic can be dried to a higher moisture content than the fraction of smaller lignocellulosic particles. This bifurcated drying is not available when both larger and smaller fraction are dried as a single entity.

Performing the dust removal before the drying step also produces a preferred fuel for wet fuel energy systems. Performing the dust removal step after the drying step produces a preferred fuel for a dry fuel burner. The option of screening after the dryers is illustrated in FIG. 1, and the option of performing both separations before the dryers is illustrated in FIG. 2.

In a typical method ("Method") of this invention for forming a lignocellulosic product ("Product") having a multi-layer structure, for example, a first layer of lignocellulosic particles and combined adhesive bonding material is generally laid down on a formation surface such as a caul plate, belt or screen.

This first layer is termed a"face mix"and forms a face of the finished product.

Then, a second layer of lignocellulosic particles and adhesive bonding material is deposited in one or more steps. This second layer is termed a"core mix"since it will form the core of the finished product.

Following this, a third layer of lignocellulosic particles and adhesive is layed down on top of the core mix of the second layer. This third layer is also a face mix and will form the opposite face of the finished product. The three layers which are deposited on the caul plate are termed the"mat." The mat, including the caul plate, is loaded into a formation press, and a pressing operation is carried out in order to form the desired product. During typical pressing operation, the mats are heated to an elevated temperature as they

are being compressed. The exact conditions utilized in the pressing and heat curing of the mat can, of course, be easily selected by one skilled in the art depending, of course, upon the desired characteristics of the final product.

As demonstrated in the experimental results that follow, this invention teaches a method of producing a multi-layer lignocellulosic product, more particularly OSB with improved properties, and/or a lower density, while also utilizing a greater fraction of lignocellulosic furnish, as compared to conventional manufacturing methods. The starting material in all experiments conducted was a sample of dry aspen OSB furnish. Another wood species preferred for this use is southern yellow pine. The finished product was 23/32"OSB in all cases, using PF resin in the Face and MDI resin in the core. More details regarding materials and methods are summarized hereafter.

In a set of experiments, 38 lb/CF density aspen OSB panels were made using four methods to illustrate the method and product of the invention. This set of experiments proved that not only the yield, but also the quality of OSB could be improved.

The four methods ("Methods 1-4") were conducted as illustrated in FIG. 3.

These methods are as follows: Method 1 is a control experiment. This Method 1 represents the most commonly practiced strand utilization strategy. The entire lignocellulosic particulate furnish was screened on a 1/4"screen. The larger size accept fraction represented about 80-85 % of the starting lignocellulosic particulate material, and was split 50/50 without any classification with respect to particle sized for use as core and face furnish. About 15-20 % of the starting material, i. e., the reject fraction, was not used in making the OSB. In a commercial setting, this smaller sized furnish would be used a low value fuel to produce thermal energy for the OSB mill. Method 1 was conducted to produce a final product at a density of 38 lb/ft3.

Method 2 applies the subject invention with dust removal from the core furnish. 50% of the lignocellulosic particulate furnished was retained on a 3/4" screen (larger size strand fraction) and used as furnish for the face layers. The

smaller sized strand fraction was then passed over a 20 Mesh screen to remove a small amount of dust (precisely, 2% of the original furnish). The material that passed through the %"screen but was retained on a 20-mesh screen was used as furnish for the core layer. Method 2 was also at a 38 lb/ft3 density.

Method 3 is another method of performing this invention, except without dust removal from the core layer furnish. 50% of furnish was recovered on a 3/4" screen (larger size strand fraction) and used as furnish for the face layers. The remaining 49% of the furnish passed through the %"screen (smaller size strand fraction) and was used as furnish for the core layer. Method 3 was also at a density of 38 lb/ft3.

Method 4 is a second control that represents a situation where no strand size segregation (screening) equipment is employed.

Comparing Method 3 to Method 1, it was found that Method 3 used 100% of the OSB furnish. This strategy produced a higher quality product in every measure, as compared to Method 1 (which uses only 80% of the furnish). MOR, MOE, IB, water absorption and thickness swell were all as good or better using Method 3 than using Method 1. Thus about 20% more OSB would be produced from the same supply of furnish in Method 3 versus the conventional strategy of Method 1, and the quality of the product would be as good or better.

The mechanical properties for lignocellulosic product formed via Methods 2 and 3 were substantially better than those of both controls. Water absorption MOE and MOR were better when Methods 2 and Method 3 were employed, than when Method 1 or Method 4 were employed. This was a very unexpected result.

The method of the present invention can improve not only the product yield, but also the mechanical strength and visual appearance of the subject lignocellulosic product.

The experiments described in flow diagram 1 and flow diagram 4 of FIG. 3 are directed to conventional screening methods. The experiments set forth in flow diagrams 2 and 3 of FIGS. 3 relate to the methods of the present invention.

All of the experiments employed as the starting material aspen strands from the Louisiana Pacific mill at Tomahawk, Wisconsin, which have been dried to a moisture content of about 7%.

In the experiments described in flow diagram 1 of FIG. 3, the aspen strands were first separated using vibrating screen system having 1/4"openings. The vibrating screen system used for these experiments was manufactured by the BM&M Partnership of Surrey, British Columbia, Canada. The 1/4"aspen skand fraction passing through the vibrating screen system was considered to be waste.

The + 1/4XS accepts aspen strand fraction was then divided into two substantially equal portions, one of which was employed as the furnish for face layer, and the other as the furnish for core layer of a subsequently formed multi-layer OSB product.

In the experiments described in flow diagram 2 of FIG. 3, the aspen strands were first separated using a vibrating screen system having 3/4"openings.

The-3/4"aspen strand fraction passing through the screen system was introduced to a 20 mesh screen. The fines which passed through the 20 mesh screen were considered to be waste. The + 3/4"accepts aspen strand fraction was employed as the furnish for face layers of a subsequently formed OSB multi-layer product. The -3/4"accepts aspen strand fraction, after rejecting the fines as waste, was employed as the furnish for core layer of a subsequently formed OSB multi-layer product.

In the experiments described in flow diagram 3 of FIG. 3 the aspen strands were first separated using a vibrating screen system having 3/4"openings. The + 3/4"accepts aspen strand fraction was employed as the furnish for face layers of a subsequently formed OSB multi-layer product. The-3/4"accepts aspen strand fraction was employed as the furnish for core layer of a subsequently formed OSB multi-layer product.

In the experiments described in flow diagram 4 of FIG. 3, the aspen strands were used as received without any screening for respective core and face layers of a subsequently formed multi-layer product.

In each of the above-described experiments, the strands which formed the core layer of each lignocellulosic multi-layer mat ("Mat") which made up the Product tested were blended with an MDI adhesive resin bonding materials. In each of the above-described experiments, the strands which formed the face layers of each Mat which made up the Product tested were blended with phenol- formaldeyde resin adhesive bonding materials. The amount of phenol- formaldeyde resin in each of the two outer face layers mixes was about 3.8 % by weight, and the amount of the MDI resin in the inner core layer mix was about 2.5 % by weight. A dosage of E-Wax in an amount of about 1.0 % by weight was also added. The Mats when formed had a weight of 18.0 pounds (38 lbs/cu. ft.).

The Mats were pressed at a plate temperature of about 420 to 430 degrees F. The total press cycle time was about 314 seconds (about 14 seconds to close the press, 270 seconds of press time, and 30 seconds of decompression time). The pressing was conducted in an Alberta Research Council Model No. 2842-01-98, 450 Ton Lab Press. The press plates are 39"X 39"in size. The product thereby formed was an 34"X 34"panel having a nominal thickness of 23/32".

The Products formed as described above were then evaluated for MOE and MOR according to standard methods specified by ASTM D-1037, Section 11-20 with minor modifications to the sample size testing as set forth in APA Standard PRP-108, Test Method S-14 for the MOE & MOR test specimens.

A further set of experiments as also conducted relating to furnish segregation and fines removal for multi-layer lignocellulosic products having a density of from about 34-41 lbs/ft3. In the control experiments, no particle segregation by size of the face or core furnish was conducted. In the comparable experiments involving size segregation of the particles in the furnish, a 50/50 segregation was carried out, the larger segregated 50% fraction going to the face layers, and the small segregated 50% fraction going to the core layer.

Three sets of control experiments and three sets of corresponding experiments employing the method of the present invention were performed. The first set involved using 100% of an Aspen particle furnish (approximately 6% moisture). The second set used approximately 96% of the furnish, a 14 Mesh

fines portion being removed. The third set used approximately 80-85% of the furnish, affines portion being removed. 34"X 34"panels having a thickness of 23/32"were produced as described above except that 3.5% of pMDI was used as the adhesive bonding material in each of the core and face layers, and the E-wax dosage was 1.2%.

FIGS. 4 and 5 are graphical diagram illustrating the Dry MOE and MOR values for 34-41 lb/ft3 density multi-layer lignocellulosic products of the present invention using the furnishes described above for the first, second and third sets of experiments. These results are graphically compared to conventionally produced multi-layer lignocellulosic products of comparable density using the furnishes described above for the first, second and third sets of experiments. In both FIGS.

4 and 5, the solid symbols were used to designate the results of the present invention, and the open symbols of the same shape and size were used to designate the corresponding controls.

Based on FIGS. 4 and 5, it is concluded that the segregation of the furnish as described above causes the MOE and the MOR of the product to increase by 20-40% over the entire 34-41 lb/ft3 density range, regardless of the level of fines removal. Furthermore, other physical properties of the product (not shown herein) remain intact and are not adversely effected by furnish segregation.

The term"multi-layer lignocellulosic product", as used herein, describes a number of lignocellulosic board products. Examples of products formed of lignocellulosic particles are particleboard, chipboard, waferboard, fiberboard, and strandboard, particularly oriented strandboard.

The multi-layer lignocellulosic products of this invention can be prepared by application of an adhesive bonding material to lignocellulosic particles, wood chips and wood stands, wood chips and wood fibers which are formed into layers.

Similarly, the method of the present invention and its attendant advantages can be achieved with respect to various forms of the above-described lignocellulosic particulate starting material, but wood strands are preferred.

The adhesive bonding system of the present invention generally comprises an isocyanate polymer and/or an aldehyde polymer resin. The adhesive bonding

system can also be an isocyanate/latex copolymer or a phenol-formaldehyde/latex copolymer. The polymers which form the adhesive bonding system are typically in liquid form so that they can be applied directly to a surface of a layer of lignocellulosic material, although a powdered form of this resin can readily be employed. The polymer resins can be combined together prior to their application.

The aldehyde polymer resins can comprise thermosetting resins such as phenol-formaldehyde, resorcinol-formaldehyde, melamine-formaldehyde, urea- formaldehyde, modified lignosulfonates, urea-furfural and condensed furfuryl alcohol resins. The phenolic component can include any one or more of the phenols which have heretofore been employed in the formation of phenolic resins and which are not substituted at either the two ortho-positions or at one ortho-and the para-position, such unsubstituted positions being necessary for the polymerization reaction. Any one, all, or none of the remaining carbon atoms of the phenol ring can be substituted. The nature of the substituent can vary widely, and it is only necessary that the substituent not interfere in the polymerization of the aldehyde with the phenol at the ortho-and/or para-positions. Substituted phenols employed in the formation of the phenolic resins include: alkyl- substituted phenols, aryl-substituted phenols, cyclo-alkyl-substituted phenols, alkenyl-substituted phenols, alkoxy-substituted phenols, aryloxy-substituted phenols, and halogen-substituted phenols, the foregoing substituents containing from 1 to 26 and preferably from 1 to 12 carbon atoms. Specific examples of suitable phenols include: phenol, 2,6 xylenol, o-cresol, m-cresol, p-cresol, 3,5- xylenol, 3-4-xylenol, 2,3,4-trimethyl phenol, 3-ethyl phenol, 3,5-diethyl phenol, p- butyl phenol, 3,5-dibutyl phenol, p-amyl phenol, p-cyclohexyl phenol, p-octyl phenol, 3,5-dicyclohexyl phenol, p-phenyl phenol, p-crotyl phenol, 3,5-dimethoxy phenol, 3,4,5-trimethoxy phenol, p-ethoxy phenol, p-butoxy phenol, 3-methyl-4- methoxy phenol, and p-phenoxy phenol.

The aldehydes reacted with the phenol can include any of the aldehydes heretofore employed in the formation of phenolic resins such as formaldehyde, acetaldehyde, propionaldehyde, furfuraldehyde, and benzaldehyde. In general, the

aldehydes employed have the formula R'CHO wherein R'is a hydrogen or a hydrocarbon radical of 1 to 8 carbon atoms. The most preferred aldehyde is formaldehyde.

The isocyanate polymer may suitably be any organic isocyanate polymer compound containing at least 2 active isocyanate groups per molecule, or mixtures of such compounds. Generally, the isocyanate polymers employed in the method of this invention are those which have an isocyanato group functionality of at least about two. Preferably, this functionality ranges from 2.3 to 3.5 with an isocyanate equivalent of 132 to 135. The isocyanato functionality can be determined from the percent available NCO groups and the average molecular weight of the isocyanate polymer composition. The percent available NCO groups can be determined by the procedures of ASTM test method D1638.

The isocyanate polymers which can be employed in the method of the present invention can be those that are typically employed in adhesive compositions, including typical aromatic, aliphatic and cycloaliphatic isocyanate polymers. Representative aromatic isocyanate polymers include 2,4-tolylene diisocyanate, 2,6-tolylene diisocyanate, 4,4'-methylene bis (phenyl isocyanate), 1,3- phenylene diisocyanate, triphenylmethane triisocyanate, 2,4,4'triisocyanato- diphenyl ether, 2,4-bis (4-isocyanatobenzyl) phenylisocyanate and related polyaryl polyiscocyanates, 1,5-naphthalene diisocyanate and mixtures thereof.

Representative aliphatic isocyanate polymers include hexamethylene diisocyanate, xylylene diisocyanate, 1,12-dodecane diisocyanate and lysine ethyl ester diisocyanate. Representative cycloaliphatic isocyanate polymers include 4,4'- methylenebis (cyclohexyl isocyanate), 1,4-cyclohexylene diisocyanate, 1-methyl- 2,4-cyclohexylene diisocyanate and 2,4-bis (4-isocyanatocyclhexylmethyl) cyclohexyl isocyanate. The isocyanate polymer is typically applied in its liquid form. Generally, when a phenol-formaldehyde resin is used as the phenolic resin it is present in the adhesive composition used in the method of the present invention within the range of about 50 to 90% by weight, preferably within the range of about 60 to 80% by weight of the total amount of adhesive. Generally, the isocyanate polymer is present in an amount of about 10% to 50% isocyanate

polymer, preferably 20 to 40%, by weight. When the adhesive bonding system is used according to these percentages, one achieves a commercially attractive combination of desired board properties and economic advantages.

The formation of the layers of lignocellulosic material from lignocellulosic involve the application of an adhesive bonding composition to the lignocellulosic particles with subsequent application of heat and pressure to form the layers into its desired consolidated configuration. It should be appreciated that the adhesive composition can be applied to the lignocellulosic particles in any conventional means, such as spray coating of the adhesive composition onto the lignocellulosic particles.

Having described and illustrated the principles the invention in a preferred embodiment thereof, it should be apparent that the invention can be modified in arrangement and detail without departing from such principles. We claim all modifications and variations coming within the spirit and scope of the following claims.