Note: Descriptions are shown in the official language in which they were submitted.
WO 95/04779 PCT~US94/09090
~168664
BIOCOMPOSITE MATERIAL AND METHOD OF MAKING
S Field of the Jnvention
The present invention relates to biocomposite materials,
particularly to a thermoset biocomposite particulate m~tt~ri~l made from a
ground leguminous crop and cellulosic material that forms a structurally rigid
thermoset biocomposite material upon fusing the particulate material. The
10 resultant material emulates the aesthetic characteristics of stone and can be engineered to meet the physical characteristics of wood.
I~ark~rollnd of the Invention
With growing concerns and pressures to develop products out
15 of renewable resources, various efforts have been directed at developing
useable products from agricultural crops. The majority of these efforts have
been directed at producing replacements for petrochemical-based plastics and
fuels as well as wood-based panels, for example. Furthermore, with growing
concerns and pressures to find alternative uses for waste matter, various
20 efforts have been directed at developing useable products from recycled
newsprint and other paper stocks. The majority of these efforts have been
directed at producing building insulation materials, fiberboard, particle board,and the like.
Insulation materials ~lepa,ed from dry, shredded cellulosic
25 materials or multiple layers of newspaper are described in U.S. Patent Nos.
4,184,311 (Rood) and 4,300,322 (Clark), and Australian Application Serial
No. 36603/84 (Hartlett et al.). Stable blocks formed of shredded newspaper
and 0.5-3% of a ligninsulfonate binder are described in U.S. Patent No.
4,148,952 (Nelson et al.). These blocks are used to consolidate waste paper
30 to facilitate h~n~11ing and transport to locations where the blocks are shredded.
Numerous processes are known for producing structural
building materials, e.g., fiberboard, panel board, particle board, from waste
paper. For example, U.S. Patent No. 4,111,730 (Balatinecz) discloses paper
flake board prepared from shredded waste paper, e.g., flakes of about 0.5-2
WO 95/04779 ~ ~ 6 g 6 ~ 4 PCT/US94/09090
inches by 1-6 inches, and a synthetic petrochemical-based thermosetting resin,
e.g., urea-formaldehyde or phenol-formaldehyde resin. The resin is used in an
amount to provide 6-15% resin solids. The moisture content of the paper
flakes during blending with the resin is m~int~ined at less than 12%. To
5 enhance the surface quality of the rçslllting paper flake board and to enhancestrength and ~tiffnes~, a mixture of wood flour or cellulosic matter and a
formaldehyde-cont~ining resin is coated on the surface.
U.S. Patent No. 5,011,741 (Hoffman) discloses a multiple ply
paper product, e.g., linerboard, produced from layers of Kraft pulp, i.e.,
10 pulped wood chips, and pulped nt;w~ufilll, i.e., newsprint subjected to
chemical dispersants used in conventional pulping processes. The pulped
newsprint is substantially free of fines and fibers having a length less than 50microns. It is impregnated with cooked cationic starch to enhance internal
strength and bonding between the layers.
U.S. Patent Nos. 4,994,148 (Shetka) and 5,064,504 (Shetka)
also disclose methods for creating molded structural blocks from a n~w~lhll
slurry which is formed in a screen wall molding chamber of a press. The
newsprint or other cellulosic product is combined with a sufficient amount of
water, e.g., a 50:50 mixture, to pulp the paper into a flowable form. A
20 plaster, cement, or latex binder can also be added to the aqueous pulp slurry.
The resultant slurry is then poured into the molding chamber. Curing of the
blocks is effected through air drying.
U.S. Patent No. 3,718,536 (Downs et al.) discloses a composite
board formed from a mass of shredded paper cont~ining a thermoplastic
25 polymer such as polyethylene, polypropylene, polyvinyl chloride, and the like.
The composite board is made of individual pieces, i.e., sheets or shreds, of
paper, arranged in an overlying and overlapping configuration, which is
distinguished from composite boards made from fibrous materials prepared by
wet or dry processes. In the process, the paper pieces are coated with the
30 thermoplastic polymer, which binds the pieces together. Thus, cross-sections
of the composite board reveal l~min~tion of paper plies surface-bonded, i.e.,
contact-bonded, together by the thermoplastic polymer. Such materials
WO 95104779 ~ 1 ~ 8 ~ fi ~ PCT/US94/09090
cont~ining thermoplastic polymers are nnll~e~ble in building products,
however, because of fire regulations. This is because thermoplastic materials
soften upon exposure to elevated tempeldlu,~s and thereby lose their structural
integrity.
All previously known panel materials cont~inin~ cellulosic
materials from waste paper, paper products, or pulp waste and/or agricultural
products, unless separately l~min~ted with a decorative finish l~min~te, have
uniformly demonskated undesirable properties. For example, they generally
possess dull grey or matte grey colors without any distinctive or aesthetically
appealing p~ttern~. Furthermore, they generally lack the skength, stiffne~,
hardness, and durability of structural grade building materials. Also, they are
generally too porous for many applications.
The lack of materials cont~ining renewable natural resources
with such desirable characteristics is believed due to a variety of factors,
including, for example: (1) the lack of processing means for repeatedly
producing an aesthetically pleasing texture and coloration of the fini~hecl
material; (2) the lack of processing means for repeatedly producing cellulose-
based material having the skength and stiffness of skuctural grade materials;
(3) shredding and repulping of waste paper reduces the length of the cellulose
fibers that can compromise the tear and shear skength of the finished
material; (4) costly de-inking and ble~ching of the cellulosic material, which
reduces the competitiveness of the material; and (5) the potential presence of
undesired cont~min~nt~ in the waste paper stock that can deleteriously effect
surface finish, bonding, rigidity, and skuctural integrity of the product.
Sl~mm~ry of the Tnv-~ntior
In appreciation of the above and to overcome the shortcomings
of the prior art, the present invention provides fiber-reinforced protein-based
biocomposite particulate material, i.e., discrete particles, made from a legume-based thermosetting resin and cellulosic material. It also provides structurallyrigid biocomposite pressure-formed materials produced from the particulate
material upon fusing the particulate material under elevated pressure and
WO 95/04779 ~ ~ 6 8 6 6 g PCT/US94/09090
temperature. The individual particles, which can also include a colorant,
m~int~in clearly defined and distinct boundaries once fused. This results in
the formation of a pressure-formed material that can exhibit a colorized
pattern resembling granite and other natural stones. The invention also
S discloses the processes necessary to fabricate the particulate material and
products thel~fio.,l, i.e., board stock and other p-es~we-formed materials.
The particulate material and resultant p-es~we-formed materials
contain the legume-based resin and fibrous cellulosic material in amounts such
that the ratio of cellulose solids to resin solids is about 0.8:1.0 to about
1.5:1.0, preferably about 0.8:1.0 to about 1.3:1Ø Particularly preferred
pLes~w~-formed materials also include a secondary thermosetting binder, such
as an isocyanate, preferably an aromatic isocyanate, for enhanced mechanical
and physical plope.lies.
The discrete fiber-reinforced protein-based particles initially
formed have a moisture content of about 55-75% and a particle size no
greater than about 0.5 inch (1.3 cm). For particular advantage in the pressure-
forming process, the moisture content of these discrete particles is reduced to
less than about 20%, preferably to less than about 15%, and more preferably
to less than about 12%. For forming plt;s~we-formed products using the
secondary thermosetting binder, e.g., an aromatic isocyanate, the moisture
content is most preferably about 6-8% to provide particularly enhanced
mechanical and physical plol)ellies. These properties are further enhanced
through the use of a water-repellant sizing agent in combination with the
secondary thermosetting binder.
The discrete biocomposite particles are capable of forming a
rigid biocomposite material having a Modulus of Rupture of about 1000-
10,000 pounds per square inch (psi) and a Modulus of Elasticity of about
100,000-1,000,000 psi (when tested in accordance to ASTM:D1037-91).
Typically, with the use of the secondary thermosetting binder (e.g., an
aromatic isocyanate) and the water-repellant sizing agent (e.g., a water-solublewaxy material), the rigid biocomposite ~lc;~w~-formed material has a
Modulus of Rupture of greater than about 3500 psi, preferably greater than
WO 95/0477~ ~ ~ 6 8 6 6 i PCT/US94/09090
about 4000 psi, and more preferably greater than about 4500 psi, and a
Modulus of Elasticity of greater than about 500,000 psi, preferably greater
than about 600,000 psi, and more preferably greater than about 700,000 psi.
The present invention is also directed to methods of preparing
5 these materials. These methods involve: pl~ing an aqueous legume-based
resin having a pH of about 10-14, preferably cont~ining a colorant; and
combining a fibrous cellulosic material with the aqueous legume-based resin
in an amount and manner effective to form discrete biocomposite particles
having a moisture content of about 55-75%, a particle size of no greater than
10 about 0.5 inch (1.3 cm), and a ratio of cellulose solids to resin solids of about
0.8:1.0 to 1.5:1Ø These particles, preferably having a moisture content of
less than about 20%, can then be fused into a rigid plessule-formed material
by pressing the dry biocomposite particles under an elevated temperature and
pressure, preferably a temperature of about 250-340F (121-171C), and a
15 pressure of about 450-750 psi. In particularly pl~fc.~ed embo~limçnte the
dried particles are coated, e.g., spray-coated, with the secondary thermosettingbinder prior to fusing the particles into a rigid biocomposite ~res~ e-formed
material.
Various advantages and distinctions of the present invention
20 will become more a~uc;l~l from the following detailed description, including
the figures and examples.
Brief nes~ription of the Dlawi~e
Figure 1 is a photograph of the particulate material of the
25 present invention, in a high moisture-content state and in a dry state.
Figure 2 is a photograph of the granular structure and stone-like
appearance of a section of board stock prepared from the particulate material
of Figure 1.
Figure 3 is a block diagram of a method of m~nufa~turing the
30 particulate material and a pressure-formed board stock of the present
invention.
WO 95/04779 . PCT/US94/09090 --
~68~ 6
Figure 4 is a block diagram of an alternative method of
m~nllf~turing the particulate material and an extruded product of the present
invention.
Figure 5 is a schematic of a press.
Figure 6 is a schematic of a continuous extruder station.
nPt~iled Description
Biocomposite Material
The present invention provides a thermoset biocomposite
10 particulate material made from a ground legurninous crop and cellulosic
material that forms structurally rigid thermoset biocomposite materials upon
fusing the particulate material under p~ . The individual particles are
fiber-reinforced protein-based biocomposite particles that can also include a
colorant if desired. Once fused into a structurally rigid product, either
15 through colllplession molding or extrusion molding, for example, the particles
m~int~in clearly defined and distinct boundaries forming irregularly shaped
domains. Thus, particles of varying shades and colors can be combined to
form products, e.g., board stock, that exhibit colorized patterns resembling
granite and other natural stones of an igneous nature or natural wood tones
20 and/or grains such as burled wood. That is, a granular appearance is exhibited
which in one instance closely resembles granite. Other ~ye~dllces are
~tt~in~hle upon varying the concentrations of the sized particulates and
colorants, however. Figure 1 is a photograph of the particulate material of the
present invention, in a nigh moisture-content state and in a dry state, and
25 Figure 2 is a photograph of the granular structure and stone-like appearance of
a section of board stock prepared from the particulate material of Figure 1.
Such pressure-formed materials, however, exhibit physical
properties more like wood than stone. That is, they exhibit densities,
structural integrity, rigidity, and m~chin~hility that is more characteristic of30 natural wood than stone. Furthermore, the strength, stiffness, density,
hardness, and durability of the pressure-formed materials of the present
invention can be much greater than particle board or other wood composite or
WO 95/04779 2 1 1~ ~ 6 ~i 4 PCT/US94/09090
cellulose-based composite board products, and equivalent to structural grade
wood-based flake board.
Thus, the present invention provides irregularly patterned,
multi-colored, i.e., granite-like, rigid biocomposite pressure-formed materials.5 Such granite-like patterns have never before been accomplished in composite
materials, particularly composites cont~ininp; cellulosic materials such as
new~ . Thus, the biocomposite materials of the present invention result
from an appreciation of the process of forming natural stone and the
relationship of human ~ ;e~ions of natural stone. That is, natural stone is
10 formed from densely packed random crystals of variously colored minerals,
which, over time in response to specific physical laws, combine and tend
toward a minimum energy system. Such a process parallels recent
underst~n~linp;~ of the theories of ordered chaotic systems. Human eye-brain
perceptions, in contrast, are learned. Thus, any recognition and appreciation
15 that a material is "natural" or "artificial" is formed upon a mental comparison
to learned perceptions. Because the artificially produced materials of the
invention closely ~l,pl~oxi,l,ate the natural random crystal ordering of igneousstone, particularly granite, the resultant material is perceived to be stone.
The density and m~hin~hility characteristics of the rigid
20 pressure-formed granite-like materials are comparable to wood, such as oak,
maple, walnut, tropical hardwoods, etc., and often better than wood-like
materials, such as flake or particle board. For example, the stock can be
sawed, routed, planed, sanded, and fini~hecl like cabinet-grade wood. The
material accepts nails and screws in a similar fashion, generally without
25 fracture or splitting. The material also has edges that can be machined and
finished as do natural wood products, as opposed to particle or flake boards
which require the use of expensive edging of wood inlay strips. The material
may also be used as part of large or complex assemblies or structures
constructed of a number of individual panels or pieces joined together with
30 conventional cabinetry techniques. The limited porosity of the pressure-
forrned biocomposite material also renders it suitable for vacuum
impregnation with acrylics, epoxy resins, varnishes or other surface treatments
a
to increase ~e suitability of ~e m~t~i~l for b~ in~ ve or s~uctural
purposes.
The biocomposite particulate m~t~n~l of ~e ~ inve~tioIL
and the ~lllt~nt pressure ~rmed pro~cts, iIlclude ~m~7y ~ e
5 natural resources, i.e., C~ O.~iC m~t~l, such as ~cled ,~:~a~, and a
proteiIl-based resin ~el ~m ~ound legllminous m~t~i~l, such as soy
flour. Ihat is, ~e particulate m~t~ l is prim~r~ly a c~lhll-l~e~ ,eed
vege~ble protein-based t~ setting resin syster~ This protein-based resin
(also 1~ L1~1 to herein as a legume-based resin) forms ~e m~i~ i.e., ~e
10 prLm~y binding agent, of the biocoll~osiLe ~ 1 of the present inver~ion.
whether in the form of particulate m~t~ or pressure-formed m~t~i~l
Plel~l~ly, on a dry weight basis, the Iatio of cellulose to resin solids, e.~.,
paper to soy flour, is about 0.8:1.0 to about 1.5:1.0, and more ~.ef~.~ly about
1.0:1.0 to abou~ 13:1Ø Thus, the amount of the th~mose~n~ resin solids
15 relative to the total amount of resin solids and cellulose solids in the
particulate m~t~l prior to pressure-~rming is ~ Ld~ly about 40-56%, and
more preferably about 43-50%.
Upon sectioning the ~ d composite par~icles ~l~ed
from waste ll~ws~. and soy flour at the most ~ led mf)isture content
and ratio of resin solids to cellulose solids, the particles appear to contain
pieces of paper fully impre~t~ with resin such that individual fibers of the
paper are resin coated (or m~e ~ ;ately, fused with resin). Tha~ is, a
new composite m~t~ iS ~ ed rather ~an a m~t~i~l that is produced by
simply gluing together paper pieces by an a~siv~ ~rough contact bonding.
Upon formatian, ~e par~cula~e m~t~ pl~ )ly co~ s
about 55-75% mois~e, i.e., water, and more ~l~f~ably about 59~7/O
moisture, based on ~e total w~t weight of the particles. As used herein, this
m~t~ l is refared to as ~e "high moisture~..~ particulate m~t~n~l or
particles. These par~icles are ~ypically in the form of soft, pliable, tacl~,
30 iIregularly shaped lumps or balls, although individual fiber-like particles can
also be forme~ They have a particle size (as ~i.v(r-~ " ,;"~ by the largest
1im~n~ion of ~e particle) of no greater than a~out 0.5 inch (1.3 cm),
~EN13E~ SHE~
WO 9~/04779 ~ l ~i 8 ~ 6 4 PCT/US94/09090
and often no greater than about 0.38 inch (0.97 cm). Typically, particles
larger than this do not generally process well, e.g., dry or press well. These
discrete particles are formed subst~nti~lly simultaneously from an
agglomerated mass of cellulosic material and legume-based resin under
5 a~ o~liate processing conditions, as described below. They have a relatively
dry-feeling or semi-dry feeling even though they contain a large amount of
water, e.g., about 55-75% total water content, which is believed to be bound
within the particles such that it cannot be readily squeezed out under hand
pressure as water is from a sponge. These high moisture-content particles
10 have sufficient internal bond strength to exist as discrete particles. Thus, they
can be handled relatively easily in bulk m~nllf~turing processes without
significantly sticking together and agglomerating into larger particles.
Distinct advantages have been obtained upon forming such high
moisture-content particles. For example, the unique "granite-like" appearance
15 of the pressure-formed products of the present invention results from the fact
that the composition and process described herein forms composite particles
of this type. Although particles collt~ ing about 55-75% water are capable of
forming pressure-formed products, particular advantage is realized if the
moisture content is within a range of about 59-67%. That is, when the
20 moisture content of the originally formed particles is about 59-67%,
particularly desirable pressure-formed products with respect to their
mechanical properties (e.g., high modulus of rupture, high modulus of
elasticity, high hardness) and physical properties (e.g., low water absorption),are obtained, as illustrated by Example 2. Although the inventors do not
25 intend to be held to any particular theory, it is believed that this optimum
moisture content provides substantially complete impregnation of the
cellulosic material, e.g., paper flakes, by the protein-based resin such that all
the fibers are integrally associated or "fused" with the resin. If less than
about 59% water is present in the high moisture-content particles, the
30 cellulosic material is not fully impregnated and cellulose fibers protrude from
the pliable balls forming "fuzzy" extensions. There may even be pieces of
uncoated paper. If greater than about 75% water is used in the p~epald~ion of
WO 95/04779 ~ 6 ~ PCT/US94/09090
the particles, a slurry generally results from vhich particles are not formed.
Furthermore, at such a high water content, the soy resin is diluted to the
extent that the interparticle bond strength is reduced considerably. Thus~ thesehigh moisture-content particles are not simply surface-coated pieces of paper,
nor are they particles of pulped paper.
Typically and preferably, prior to fusing into a rigid pressure-
formed material, the moisture content of the high moisture-content
biocomposite particulate material is reduced to less than about 20%,
preferably less than about 15%, more preferably less than about 12%, and
most preferably about 3-12%, based on the total weight of the biocomposite
particles. If the moisture content is too low, however, the biocomposite
particles do not generally bind well upon fusing them together under elevated
pressures and temperatures, without the addition of a secondary thermosetting
binder. For certain applications, the moisture content is narrowly tailored to
about 8-11%, whereas in other applications the moisture content is narrowly
tailored to about 6-8%. Generally, the biocomposite particles cont~inin~ 6-8%
moisture are ideal for preferred processes wherein the particles are coated
with a secondary thermosetting binder, e.g., an aromatic isocyanate.
During the drying process, not only is water removed but the
protein-based resin is at least partially cured. In this way, the particulate
material has internal bond strength. Furthermore, these particles are capable
of being bonded or fused together under heat and pl~ U~. This is because
the viscoelasticity can be altered with elevated temperatures and pl~5~ult~S
such that the particles are plasticized and flowable but not meltable, i.e., thethermoset resin/cellulose composite particles are plastically deformable but notsufficiently thermoplastic to melt. It is also believed that the thermoset resinfilrther cures such that there are resin-resin interactions between the particles,
although not to such an extent that the particles flow together and lose their
distinct boundaries.
These dried biocomposite particles are generally platy and
irregularly shaped, although they can be spherical in shape. Upon losing
water, there is a significant reduction in the volume of the particles. Thus~ the
WO 95/0477~ ; 8 ~ 6 I PCT/US94/09090
particle size (as determined by the largest dimension of the particle) is
typically no greater than about 0.5 inch (1.3 cm), and often no greater than
about 0.38 inch (0.97 cm). Although the particle size is generally greater than
about 0.05 inch (0.13 cm), there can be small fiber-like particles, i.e., small
S cellulosic fibers fused with the legurninous resin. Typically, however, the
fraction of fines, i.e., dried biocomposite particles having a particle size less
than about 0.05 inch (0.13 cm), is less than about 5% by weight. The specific
gravity of the dried particles is typically about 0.7-1.0, preferably about 0.9-0.95.
The cellulosic material can be any cellulosic-cont~ining
material, such as newspaper, glossy paper, uninked paper, office paper,
computer paper, phone books, coated papers, cellulose insulation (coated or
uncoated), Kraft paper, agricultural fiber (e.g., hammer-milled corn stalks),
pulped wood fiber, paper mill waste sludge, etc. Such cellulosic materials are
15 useful if they contain fibrous extensions of the type that result from shredding
or hammer milling. Thus, useful cellulosic materials are those that are
fibrous, i.e., have fibrous extensions. That is, smooth cellulosic material, such
as non-bond typing paper that has been cleanly cut into pieces is generally not
as useful as that which has been h~mmer milled. Of the cellulosic materials
20 listed above, the ~erelled materials are newspaper, uninked paper, office
paper, computer paper, uncoated cellulose insulation, and paper mill waste
sludge. More preferably, the cellulosic material is shredded or h~mmer-milled
newspaper of the type that is used as insulation. The cellulosic material can
optionally include a fire retardant, e.g., boric acid, in an amount up to about
25 15% by weight of dry cellulosic material.
Depending upon various appearance and mechanical attributes
desired for the finished end product, the particle size and shape of the fibrouscellulosic material can vary. Generally, the cellulosic material includes a
broad distribution of sizes (e.g., individual fibers up to l-inch (2.54-cm) strips
30 of paper) and shapes (e.g., round, elongated, or irregularly shaped pieces ofpaper). Preferably, the particle size (as determined by the largest dimension)
of the majority (i.e., greater than 50%) of the cellulosic material is less than
W0 9s/04779 ~ ~81~6 ~ 12 PCT/U594/09090
about l inch (2.5 cm), more preferably less than about 0.5 inch (1.3 cm)~ and
most preferably about 0.06 inch (0.16 cm) to about 0.38 inch (0.95 cm). For
example, a typical cellulosic particle used in making the products of the
present invention has nominal (limencions of approximately 0.25 inch (0.64
S cm) by O.S inch (1.3 cm). The aspect ratio, i.e., length to width, of the bulkof the particles of cellulosic material is typically no more than about 3: l andcan be about l: l, although fibers can be present which can have a quite large
aspect ratio. Thus, cellulosic material used in the present invention can
include cellulose particles or flakes admixed with loosely compacted cellulose
10 fibers or fines, i.e., dust-sized particles. Preferably, in such an admixture,
advantage is realized in processin~ if the fines are below about 10% of the
total weight of the material.
Although the fibrous cellulosic material, e.g., shredded
newsprint, can be used by itself as the reinforcing material for the legume-
lS based resin, other fibrous materials can be included to provide certaindesirable attributes. For example, chopped fiberglass, spun plastics, or other
fibrous additives of ayyroyl;ate fiber length can be incorporated into the
biocomposite particulate material of the present invention or into the pressure-formed rigid biocomposite material of the present invention. Depending upon
20 the fibrous material, the fibers can be randomly dispersed or mixed in various
fashions to precletermined ~ nments. For example, where a directional fiber
layering is used, a board stock can be developed from multiple 1~min~tions
having specific axial alignments of the added fiber. The derived board stock
can thereby exhibit preferential bending, shaping, shear or tear characteristics.
25 Alternatively, 1~min~tions of random fiber alignment can be laid down one
upon another. In all cases, however, a stone-like appearing board stock is
obtained with improved structural integrity.
The ground leguminous material that is used in the formation of
the legume-based resin includes any of the nitrogen-fixing crops, such as
30 beans, peas, clover, and alfalfa, for example, that is ground to a particulate
material, e.g., a meal or flour-like material. Preferably, the leguminous
material is an edible leguminous crop, i.e., a pulse crop, such as soybeans,
WO 95/04779 ............................. ~ PCT/US94/09090
2 1 ~
13
pinto beans, peas, pelagis, etc. The leguminous material can be in various
forms and sizes. For example, the leguminous material can be in the form of
ground whole beans (including the hulls, oil, protein, minerals, etc.), a meal
(extracted or partially extracted), a flour (i.e., generally cont~ining less than
5 about 1.5% oil and about 30-35% carbohydrate), or an isolate (i.e., a pure
protein flour cont~ining less than about 0.5% oil and less than about 5%
carbohydrate). As used herein, "flour" includes within its scope material that
fits both the definitions of flour and isolate. Preferably, the leguminous
material is in the form of a flour, at least because the ~ e-formed
10 products produced from a flour, as opposed to a meal, has more desirable
physical properties (e.g., water absorption) as well as mechanical properties
(e.g., modulus of rupture and modulus of elasticity).
Preferably, the leguminous material has a particle size (as
determined by the largest riimen.cion) of less than about 0.1 inch (0.25 cm),
15 and more preferably less than about 0.05 inch (0.125 cm). If the particle size
is much larger than this, the leguminous material is not sufficiently soluble
and the resultant pressure-formed products have lessened mechanical
properties, e.g., board strength, and less visual perfection. Furthermore, the
time required to solubilize the material is undesirably long. Although material
20 such as a meal can be used, wherein about 75% does not pass through a 28
mesh screen (i.e., 0.0334 inch, 0.08 cm) and about 5% passes through a 60
mesh screen (i.e., 0.0167 inch, 0.04 cm), a flour is more preferred because of
its generally smaller particle size distribution, as illustrated in Example 6.
That is, the most preferred ground leguminous material has a m~iml-m
25 particle size of that of a flour, i.e., about 0.005 inch (12.7 x 10-3 cm). There
does not appear to be a minimurn particle size requirement for the ground
leguminous material; however, the particle size of commercially available
soybean flour is generally less than about 0.003 inch (7.6 x 10-3 cm). For
example, for commercially available soybean flour, greater than about 92%
30 passes through a 325 mesh screen, which corresponds to a particle size of
about 0.003 inch (7.6 x 10-3 cm).
WO 95/04779 ~ 8~6 6 ~ PCT/US94/09090
Preferably, the legurninous material has a dispersible protein
content of about 20-100 pdi (i.e., protein dispersion index), and more
preferably about 70-100 pdi, as illustrated by Example 5. Most preferably,
the ground leguminous material is a soybean flour having a dispersible protein
5 content of about 70-95 pdi. This material represents a desirable balance
between residual carbohydrate concentration and the soluble protein level, i.e.,water-soluble protein. A 90 pdi soy flour contains soluble protein in an
amount of about 90% of the total available protein. A typical 90 pdi soy
flour contains, by total weight, a soluble protein level of about 50-55%, a
10 residual carbohydrate level of about 30-40%, and less than about 1.5% oil,
with the remainder being hulls, water, and ash, including minerals (about 5-
10%). A suitable 90 pdi soy flour cont~inin~ 50% protein, 40% carbohydrate,
5% water, 4% ash, and 1% fat can be obtained from Honeymead Products
Company (Mankato, MN).
The legume-based resin, i.e., the matrix of the biocomposite
material of the present invention, is prepared by combining the ground
leguminous material with a highly ~lk~line aqueous solution cont~ining an
alkaline dispersing agent, such as a strong inorganic or organic base. The
base is preferably a strong inorganic base, such as KOH, NaOH, CaOH,
NH40H, or combination thereof. Preferably, the base is KOH or NaOH, more
preferably KOH. Most preferably, the pH is adjusted to the ~plul~l;ate level
using both KOH and CaOH, at least because potassium and calcium ions tend
to inhibit moisture transport and thus impart a measure of moisture reci~t~nee
to the final product. An amount of base is used to provide a resin having a
pH of about 10-14, preferably about 10-13, and more preferably about 12-13.
A pH lower than about 10 generally does not provide particulate material of
the desired consistency. This ~lk~line-dispersed protein-based, i.e., legume-
based, resin preferably contains about 58-92% water, and more preferably
about 72-83% water, based on the total weight of the resin.
The ground leguminous material is used in the protein-based
resin in an amount effective to produce a generally smooth resin having a
viscosity of about 37,000-640,000 centipoise (cps), preferably about 104,000-
WO 95/04779 ~ 36 6~ PCT/US94/09090
184,000 cps at 65F (18C~. Preferably, the ground leguminous material is
used in an amount of about 15-40% by weight based on the total weight of
the resin. More preferably, it is used in an amount of about 20-40%, and
most preferably about 30-35%. Depending upon the nature of the cellulosic
5 m~teri~l and the plo~e~lies desired in the ples~ule-formed products, the
amount of ground leguminous material can be varied, although the viscosity
should be m~int~ined within the desired range (i.e., 37,000-640,000 cps),
which is significantly higher than the viscosities of typical protein glues (i.e.,
500-75,000 cps) used in the production of particle board, etc.
This resin is a therrnosetting resin, i.e., it solidifies irreversibly
when heated, typically through mechanical bonding and chemical cross-linking
reactions. This makes the resultant pressure-forrned products useable for
possible structural building materials because they m~int~in structural integrity
in a high temperature or fire situation compared to thermoplastic resin
15 compositions. The resin can be stored for subsequent use, but the shelf life is
typically limited to less than about 12 hours, and often to less than about 1
hour if cross-linking agents are used.
The resin can also include a water-dispersible colorant that can
withstand the processing conditions discussed below, e.g., telllpt;ldlules as
20 high as 450F (232C) and a pH within a range of about 10-14. That is, the
colorant must be able to m~int~in color f~tnp~s under heat, I~les~ule, and high
~lk~line conditions. Furtherrnore, the colorant should be W-stable and
capable of being uniformly dispersed in a highly ~lk~line aqueous solution.
Also, each colored particulate m~tPri~l must not significantly bleed or transfer25 color to adjacent colored particulate material during processing into the
pressure-formed products. These requirements are generally met by
commercially available inorganic or organic colorants, i.e., dyes or pigments,
used in composite materials.
Suitable inorganic colorants are generally metal-based coloring
30 materials, such as ground metal oxide colorants of the type commonly used to
color cement and grout. Such inorganic colorants include, but are not limited
to: metal oxides such as red iron oxide (primarily Fe2O3), yellow iron oxide
WO 95/04779 ~ i 6 ~ PCT/US94/09090
16
(Fe2O3-H2O), titanium dioxide (TiO2), yellow iron oxide/titanium dioxide
mixture, nickel oxide, m~ng~n~se dioxide (MnO2), and chromium (III) oxide
(Cr2O3); mixed metal rutile or spinel pigment~ such as nickel antimomy
lita~ l rutile ({Ti,Ni,Sb}02), cobalt alllnnin~te spinel (CoAl204), zinc iron
S chromite spinel, m~n~nese antimony titaniurn rutile, iron titanium spinel,
chrome antimony titanium rutile, copper chromite spinel, chrome iron nickel
spinel, and m~ng~nese ferrite spinel; lead chromate; cobalt phosphate
(Co3(PO4)2); cobalt lithium phosphate (CoLiPO4); mAn~nese ammonium
pyrophosphate; cobalt m~gnesium borate; and sodiurn alumino sulfosilicate
10 (Na6Al6Si6O24S4). Suitable organic colorants include, but are not limited to: carbon black such as lampblack pigment dispersion; xanthene dyes;
phthalocyanine dyes such as copper phthalocyanine and polychloro copper
phthalocyanine; quinacridone pigments including chlorinated quinacridone
pigments; dioxazine pi~ment~; anthroquinone dyes; azo dyes such as azo
15 naphthalenedisulfonic acid dyes; copper azo dyes; pyrrolopyrrol pigments; andisoindolinone pi~ment~ Such dyes and pigments are commercially available
from Mineral Pigments Corp. (Beltsville, MD), Shephard Color Co.
(Cincinnati, OH), Tamms Tn~ tries Co. (Itasca, IL), Huls America Inc.
(Piscataway, NJ), Ferro Corp. (Cleveland, OH), Engelhard Corp. (Iselin, NJ),
20 BASF Corp. (P~ippally, NJ), Ciba-Geigy Corp. (Newport, DE), and DuPont
Chemicals (Wilmington, DE).
The colorant is typically added to the legume-based resin
without further processing and in an amount suitable to provide the desired
color. Preferably, the colorant is present in the particulate material in an
25 amount no greater than about 15% by weight of the legume-based resin, more
preferably no greater than about 10%, and most preferably no greater than
about 3%.
The legume-based resin can also include a defoaming agent, a
fining agent, or a combination thereof. Each of these additives is used in the
30 resin in an amount to provide desirable characteristics to the resin and
particulate material produced therefrom. For example, a fining agent, i.e.,
dispersing agent, can be used to improve the texture, i.e., smoothness, as well
WO 95/04779 ~ 1 6 8 ~ 6 ~ PCT/US94/09090
as surface tension, viscosity, and flow properties of the resin. The fining
agent also aids in dispersing the resin components, particularly the colorant,
adjusting the viscosity of the resin, and can even act as an adhesion promoter.
A defoarner, i.e., defoaming agent, can be used to release ~nl-dp~ed air, which
S can adversely affect the strength of the pressure-formed materials. Preferably,
a defoaming agent and a fining agent are used alone, or in combination, in an
amount of no greater than about 8%, and more preferably no greater than
about 5%, by weight of the resin. That is, whether used alone or in
combination, the total amount of defoaming agent plus fining agent is no
10 greater than about 8%. Suitable defoaming agents include, but are not limitedto, pine oil and silicone defoamers. Suitable fining agents include, but are notlimited to, sodium silicate and CoCob, i.e., ground corn cobs.
The legume-based resin used to prepare the composite
particulate material can also contain minor amounts of other additives, such as
15 inert fillers, latex emulsions or waxy materials, cross-linkers, thermoplastic
adhesive binders, thermosetting adhesive binders, fire retardants, fungicides,
catalysts, and insecticides. Each of these additives is used in the resin in an
amount to provide desirable characteristics. For example, a filler can be used
in an amount to add bulk and reduce the porosity of the resultant pressure-
20 formed product. Suitable fillers include, but are not limited to, wood flour,bentonite, kaolin, dust, and recycled biocomposite m~t~ l of the present
invention. A latex emulsion or waxy material can be used in the resin
formulation in an amount to increase the water resistance of the particle.
Alternatively, or additionally, a waxy material can be applied to the dry
25 biocomposite particulate material. Suitable latex emulsions include, but are
not limited to, latex emulsions available under the tradename FULATEX PD-
0512, which is a blend of polyvinyl acetate and a latex emulsion from H.B.
Fuller Company (St. Paul, MN), styrene-butadiene resin, and any latex
emulsion typically used in water-based paints. Suitable waxy materials
30 include, but are not limited to, slack wax and wax emulsions available from
H.B. Fuller Company (product number RM0255) and Hercules Incorporated
(Wilmington, DE, product number 2100P). A cross-lirlker can be used in the
WO 95/04779 PCT/US94/09090--
2~6~66~
18
resin in an amount to strengthen the internal bond strength of the individual
particles. Suitable cross-linkers include, but are not limited to, formaldehyde-based latex resins suitable for use as cross-linkers available under the
tradename FULLREZ (product number WB-2523) from H.B. Fuller Company
5 (St. Paul, MN), dialdehyde starch, and ammonium chloride. A thermoplastic
adhesive binder can be used in the resin in an amount to increase the strength
and hardness characteristics of the individual particles, although it is preferred
to avoid the use of thermoplastic materials in the products of the present
invention. Suitable thermoplastic a&esive binders include, but are not limited
10 to, polyvinyl acetate ("PVA"). This can be added alone, or in combination
with a latex emulsion as is done in the examples in the form of H.B. Fuller
Colllpally's FULATEX PD-0512. A fire retardant can be used in the resin in
an amount effective to reduce the fl~mm~bility of the cellulosic material.
Alternatively, or additionally, the fire retardant can be added directly to the
15 cellulosic material. When added to the resin, however, a fire retardant such as
boric acid aids in dispersing the dry ingredients. Suitable fire retardants
include, but are not limited to, boric acid and ammonium sulfate.
Preferably, these additives are used alone or in combination in
an amount of no greater than about 20%, and more preferably no greater than
20 about 10%, by weight of the resin. Thus, the resin and res--lt~nt particulate material and pressure-formed products contain no more than about 20%
thermoplastic materials, such as thermoplastic adhesive binders. Preferably,
they contain no greater than about 10% and more preferably no greater than
about 5%. Most preferably the resin, particulate material, and pressure-
25 formed products of the present invention are substantially free ofthermoplastic materials, particularly thermoplastic adhesive binders.
The legume-based resin and cellulosic material are combined in
a manner to form the high moisture-content particulate material described
above. Upon drying, the particulate material can be stored for an indefinite
30 period of time before being formed into the pressure-formed products. The
particulate material can be formed into rigid materials under elevated
WO 95/04779 ~ 16 ~ ~ 6 Q PCT/US94/09090
19
pressures and temperatures. This can be done without any additional
themosetting binders other than the legume-based resin itself.
For particularly advantageous results the dry particulate
material is preferably coated with a secondary thermosetting binder, such as
5 an isocyanate, phenolic, melamine, or urea-co~ g binder. Preferably, the
secondary thermosetting binder is an organic isocyanate, and more preferably
an aromatic isocyanate. The isocyanate provides greater mechanical
properties, e.g., stiffnes~ and strength, to the pressure-formed products. For
example, it decreases the amount of cupping and warping of the pressure-
10 formed material. As shown in Examples 7 and 8, the use of isocyanate canincrease the modulus of elasticity by about 30-40% and the modulus of
rupture by about 60-70%.
The isocyanate also provides greater dimensional stability to the
rigid biocomposite pressure-formed material of the present invention. For
15 example it is useful in significantly lowering the water absorption of the
pressure-formed products, particularly when used with a sizing agent as
discussed below. As shown in Example 8, the water absorption can be
decreased by as much as 75%. Such products can be used in applications
where a high degree of ~limen.~ional stability is required, e.g., as flooring or20 countertop material. Furthermore, the isocyanate reduces the processing time
and energy consu~ lion, thereby making the pressure-forming process
significantly more efficient.
It is believed that this is because the strength of the
interparticle bonding is significantly improved with the addition of the
25 isocyanate binder to the dry particulate material. Although the inventors do
not wish to be held to any particular theory, it is believed that this is due toan increased number of bonding interactions. Not only are there interactions
between the protein-based resin and the cellulose as in the nonisocyanate-
cont~ining pressure-formed products, but it is believed that there are
30 interactions between the protein-based resin and the isocyanate binder and
between the cellulose and the isocyanate binder. Thus, in essence the
secondary thermosetting binder interacts with the dried particles cont~inin~ at
WO 95/04779 . ~ , PCT/US94/09090
8~4~
least partially cured protein-based resin to create a dual resin system, which
provides greater advantage than either used alone.
Suitable isocyanates include, but are not limited to, the
aromatic isocyanates 4,4-diphenylmethane diisocyanate (MDI), toluene
5 isocyanate (TDI), xylene diisocyanate (XDI), and methaxylene diisocyanate
(MXDI). Preferably, the aromatic isocyanate is MDI. The isocyanate can be
used in an amount of about 2-20%, preferably about 2-10%, based on the total
weight of the dry biocomposite particles. Preferably, the aromatic isocyanate
is used in an amount of about 5-8%. Advantageously, no hardeners or buffers
10 are required to be used with the aromatic isocyanate, thereby reducing any
potential for incompatibility with other additives used in the compositions of
the present invention.
As discussed in Example 8 below, control of the moisture
content of the particles is particularly important when an isocyanate is used.
15 For example, if MDI is added to the wet particles, through addition to the
resin prior to addition of the cellulose, no significant advantage is realized in
the mechanical properties of the resultant ples~u~e-formed products.
However, if the MDI is added to the particles that are dried to a moisture
content of less than about 20%, preferably less than about 15%, more
20 preferably less than about 12% (often about 3-12%), and most preferably
about 6-8%, significant advantage is realized in mechanical properties as well
as physical properties.
In addition to, or as an ~Itern~tive to, the secondary
thermosetting binder, a sizing agent having water-repellant properties can be
25 added to the dry particulate material prior to being fused together into a
pressure-formed product. The use of a sizing agent, particularly when used in
combination with the secondary thermosetting binder, significantly improves
the dimensional stability of the pressure-formed panels when they are
subjected to water and other liquids or vapors, as illustrated in Example 8.
30 Suitable sizing agents include, but are not limited to, slack wax, agricultural
oils, modified agricultural oils, and wax emulsions as described above for use
in the resin formulation. Preferably, the waxy or oily material is a wax
WO 95/04779 ~ 16 ~ 6 ~ A PCTIUS94/09090
emulsion. The water-soluble waxy material, i.e., wax emulsion, can be used
in an amount of about 0.5-2.5%, based on the total weight of the dry
biocomposite particles. Preferably, the waxy emulsion is used in an amount
of about 0.9-1.5%. Either or both of these components, i.e., isocyanate binder
and sizing agent, alone or in combination, can be coated on the dry particulate
material by spray coating or mechanical mixing, for example.
The mechanical and physical properties of the pressure-formed
materials of the present invention can be readily controlled and engineered as
desired. That is, depending on the additives, resin level, water content of the
particles, the applied pressure, and the heating schedule, for example, the
density, surface hardness, bending modulus, and tensile and compressive
strength of the ~les~ule-formed products can be varied from values
comparable to a representative medium-density fiberboard to values exceeding
those of oak and even apitong (a dense tropical hardwood). Thus, a tiffnecs
and strength exceeding that required of structural grade particle board and
other building panels may also be obtained. For example, MDI-coated
particles can be pressed into panels that have strength and stiffness plopl_.Lies
that are equivalent to, or better than, high density particle board of the type
used in the m~nnf~cture of furniture.
Depending on the various components, amounts, and processing
parameters chosen, values for Modulus of Rupture ("MOR") can range from
about 1000 psi to about 10,000 psi, and values for Modulus of Elasticity
("MOE") can range from about 100,000 psi to about 1,000,000 psi. The
density can also be tailored to a range of about 45-100 pounds per cubic foot
(i.e., the specific gravity to a range of about 0.72-1.60). Particularly preferred
pressure-formed materials have a density of about 70-90 pounds per cubic
foot (i.e., a specific gravity of about 1.1-1.45). Typically, the preferred higher
MOR, MOE, density, and specific gravity values can be obtained through the
use of high resin levels and/or high pressures. The amount of water
absorption as determined by the percentage of edge swell in a 24-hour water
immersion test can vary from about 3.5-40%. Typically, the preferred low
values can be obtained through the use of a secondary thermosetting resin,
8 ~ ~ 4 ~
WO 95/04779 . PCT/US94/09090
e.g., MDI, and/or a sizing agent applied to the dry particulate material prior to
pressure forming. Although the most preferred pressure-formed materials, e.g.
panels, of the present invention have low water absorption values, e.g., less
than about 15% edge swell, those having high water absorption values can
5 still have advantageous mechanical properties (e.g., high MOR, MOE, and
density) and thus be useable.
Generally, MOR values of the rigid biocomposite pressure-
formed material of the present invention are greater than about 2000 psi,
preferably greater than about 2800 psi, and more preferably greater than about
10 3000 psi. The MOE values are generally greater than about 200,000 psi,
preferably greater than about 350,000 psi, and more preferably greater than
about 450,000 psi. Rigid biocomposite pressure-formed material prepared
according to the processes and fonn~ tions of the present invention can attain
such MOR and MOE values when therm~lly pressed at a telllpcldLIlre of about
15 320F (160C) and a pressure of about 520 psi, for about 13-14 minutes.
For the rigid biocomposite pressure-formed material prepared
according to the processes and formulations of the present in which a
secondary thermosetting resin, such as an aromatic isocyanate, is used, the
MOR values are generally greater than about 3500 psi, preferably greater than
20 about 4000 psi, and more preferably greater than about 4500 psi. The MOE
values are generally greater than about 500,000 psi, preferably greater than
about 600,000 psi, and more preferably greater than about 700,000 psi. Rigid
biocomposite pressure-formed material co~ lillg a secondary thermosetting
resin can attain such MOR and MOE values when therm~lly pressed at a
25 temperature of about 320F (160C) and a pressure of about 520 psi, for about 8-9 minlltes
Process for Preparation of Biocomposite Material
In general, the pressure-formed products of the present
30 invention are made by a process that involves fusing the dry biocomposite
particulate material described above into a rigid thermoset biocomposite
material. Preferably, the process involves five separate and distinct stages:
WO 95/04779 ~ ~ 6 ~ 6 ~ ~ PCT/US94/09090
(1) obtaining the required raw materials; (2) preparing the legume-based resin,
i.e., the biocomposite matrix material; (3) ~lepa~ g the high moisture-content
biocomposite particulate material; (4) reducing the moisture content of the
high moisture-content biocomposite particulate material, optionally con~ining
5 an admixture of separate colorized batches of the particles; and (5)
compacting and further curing the dry particulate material, optionally coated
with a secondary thermosetting binder. The resulting material, e.g., board
stock or shaped object, is suitable for further forming and fini.~hing steps, e.g.,
structure fabrication, surface fini~hinp;, mechanical shaping, etc.
Referring to Figure 3, a generalized block diagram is shown of
the presently preferred process used to fabricate board stock from cellulosic
material. This process is described using recycled newsprint and soybean
flour. It is to be understood, however, that the process can be modified to use
any cellulosic material and any leguminous material as described above. The
15 preferred process is in broad terms carried out in five distinct stages to
produce a raw board stock or shaped product. The raw stock is then finished
with ~plol)-;ate sizing, shaping, and surface fini~hin~ steps, if desired, to
produce a final finished product. If used as a l~rnin~te or molding accent, a
finished board stock would t~vpically undergo further plel)aldlion for
20 ~tt~chment to furniture or possibly ceiling or wall panels.
During the first or initial stage, the required raw materials are
obtained, and preprocessed if n~cess~ry. Such preprocessing can involve
selecting, sizing, grading, sorting, and storing in applol,l;ate bins in a form
ready for processing in the following stages. These materials typically include
25 a cellulosic material, a ground leguminous material, e.g., soybean derivative,
optional colorants, water, and miscellaneous optional additives. Each of the
materials is described in more detail above.
During this initial stage, the cellulosic material is typically
prepared by dry shredding and/or hammer milling waste paper, such as
30 newsprint or other uncoated printing papers, to desired particulate sizes andrelative size concentrations. Alternatively, commercially available loose
cellulose fiber insulation can be obtained and used directly. The shredded
WO 9~/04779 ~ 1~ 8 ~ 6 ~ PCT/US94/09090
24
paper is commingled as necessary with other materials to impart certain
desired properties to the finished product, e.g., fire retardance via the addition
of boric acid.
The ground soybean material is typically prepared by grinding
de-hulled whole beans in a conventional grinding process which operates at
below 160F (71C) to a granularity in the range of 100-325 mesh, preferably
200-325 mesh. As desired, the beans may be pre-washed and may be stripped
of natural oils. If the oils are removed, this can be achieved using
conventional mechanical or chemical extraction techniques, such as one of the
"cold process" oil extraction processes. Alternatively, commercially available
soybean flour can be obtained and used directly. The ground soybeans,
preferably soybean flour, is not otherwise washed or treated.
The protein-based resin, i.e., biocomposite thermosetting matrix
material or adhesive binder, is separately prepared in stage 2 by mixing the
processed soybean derivative, i.e., ground soybeans, which can be in the form
of a meal or flour, preferably a flour, with a highly ~Ik~line aqueous solution.Added to the resin in suitable quantity and as desired or necessary are minor
amounts of various additives, such as colorants, defoaming agents, fining
agents, thermoplastic adhesive binders, inert fillers, latex emulsions or waxy
materials, etc. These additives impart various desirable properties to the
finished product and/or facilitate the removal of water and other liquids in
later stage proces.sing. Moisture r~si~t~nce, fire retardance, mold and fungal
resistance, elasticity, and surface hardness represent some of the
aforementioned desirable properties.
The ingredients that are used to prepare the resin can be
combined in any desired order. Generally, however, it is desirable for the soy
flour to be added to water and for any cross-linkers used to be added toward
the end of the mixing process. Preferably, any cross-linkers used are added
within about one hour before the cellulosic material and the resin are
combined in stage 3. The water used can be tap water, distilled water, and
condensate from other portions of the process. It can even be milk, which is
primarily casein in water. The temperature of the water can vary depending
WO 95/04779 ~ ~ ~ 8 ~ ~ 4 PCT/USs4/OsOso
upon the desired mixing rate and viscosity of the resin; however, it is
advantageous for the temperature of the water to be at least about 50~
(10C), and preferably about 60-70F (16-21C). Higher temperatures are
possible for faster reaction times; however, they are generally undesirable due
to higher energy consumption.
The resin ingredients are generally blended for a sufficient
period of time to form a homogeneous batter-like consistency. Any
conventional batch mixing techniques and batch-type mixers can be used.
Preferably, any vertical or horizontal paddle-type mixer, can be used. A
paddle-type mixer is typically used to nonturbulantly mix viscous materials.
It typically contains one or more paddles rotating in a fixed shell having a
rounded bottom and straight sides that provide gentle lifting and good
circulation without deterioration of the material being stirred. The use of a
mixer equipped with slow-moving paddles that gently lift, fold, and agitate the
material, in conjunction with a means for shearing the material to ensure
thorough mixing is more l,ler~lled. This type of mixer is capable of high
shear and slow agitation, which avoids mechanical break down of the protein
molecules. Most plerelled for prep~i~lg the resin of the present invention is a
paddle-type mixer equipped with shear bars and chopper blades.
For colored materials, colorants are selected according to the
desired color pattern prescribed for the finished product and added during
processing of the resin. The eventual coloration of the produced products is
obtained upon mixing various proportions of differently colored particulate
material derived at stage 3. That is, during stage 3, one or more
monochromatic batches of biocomposite particulate material is prepared from
the preprocessed components, i.e., the shredded paper and the legume-based
resin. A three color batch is depicted, wherein the different style lead lines
indicate the mixing of the three separate colorized batches.
During stage 3, the legume-based resin prepared in stage 2 is
- 30 combined with the ground or milled cellulosic material in a manner to
produce agglomeration and particulate forrnation. That is, mixing conditions,
e.g., relative amounts of solids to liquid and the mixing action, are designed
WO 95/04779 ~ iC~ PCT/US94/09090
26
such that the cellulosic material and legume-based resin are blended together
into a substantially homogeneous agglomerated mass having a fiber-reinforced
batter-like consistency. That is, the mixing conditions are designed to ensure
the formation of soft, pliable, tacky, discrete, irregularly shaped lumps or balls
5 substantially simultaneously from the agglomerated mass. Typically, this
involves using a sufficient amount of water in the blend of resin and
cellulosic material such that the desired discrete particles are formed. If too
much water is used, relative to the amount of solids, e.g., paper plus soy
flour, the individual pieces of the cellulosic material tend to lose their identity
10 through the formation of a fibrous slurry or the mixture agglomerates into a
large mass without the formation of discrete particles. If too little water is
used, "fuzzy" particles are formed and/or unblended paper particles remain.
Preferably, the water content of the blend, i.e., the legume-based resin
including any optional ingredients and cellulosic material, should be about 55-
15 75% for the discrete particles to form. It is this particle formation that isimportant to the formation of the unique look, as well as the physical and
mechanical characteristics, of the pressure-formed products. The water
content is determined primarily by the resin, although the water content of the
blend of resin and cellulose can be modified by the addition of more water, or
20 more cellulosic material, to produce the desired biocomposite particulate
material.
Upon combining the resin, which is at a pH of about 10-14,
with the cellulose, e.g., paper, the pH generally drops by at least about I pH
unit, preferably to less than about 1 1, more preferably to less than about 10,
25 and typically to a range of about 7-10. Although the inventors do not wish tobe held to any particular theory, it is believed that the alkaline resin causes
the protein molecules to hydrolyze and open, i.e., unfold. As the cellulose
becomes intermingled with the protein molecules and the pH lowers, a linking
between the cellulose and protein occurs, which is believed to occur through
30 both mechanical and chemical interactions. This is believed to be a
significant contributing factor to the internal bond strength within the
particles.
-
-
~ WO 95104779 2~ ~8 B ~4 PCTlUSs4/09090
27
Generally, the desired particulate material can be obtained by
combining, i e., blending or mixing, the legume-based resin and cellulosic
material in a ratio of about 5 parts resin to about 1 part cellulose by weight.
Preferably, however, an amount of resin and cellulose are combined such that,
5 on a dry weight basis, the ratio of cellulose to resin solids, e.g., paper to soy
flour, is about 0.8:1.0 to about 1.5:1.0, preferably about 1.0:1.0 to about
1.3:1.0, as illustrated by Example 4. The blending time can vary depending
on the moisture content of the blend and the particle size desired. Typically,
however, the particles are formed from the agglomerated mass in less than
10 about 10 minlltes.
Typically, the formation of the discrete particles also involves
blending under shearing conditions to form a uniforrn mixture of the cellulosic
material and the resin, which initially agglomerates into one large ball and
then under the shearing forces breaks into individual, i.e., discrete, particles.
15 Thus, the use of a high shear mixer is preferred, although other mixers, suchas ribbon mixers and batch-type mixers, may be able to produce the desired
particles, i.e., the irregularly shaped, tacky, balls referred to herein as the high
moisture-content particles, under certain conditions. In particularly preferred
embodiments, the mixer used to prepare the resin has paddles, shear bars, and
20 chopper blades. A continuous feed mixer, such as a ribbon blender, may also
be used to advantage to continuously blend each separately colored batch of
feedstock. Appropriate proportioning controls are required to assure the
proper compositions of the feedstock ingredients.
Once prepared, each of the separate batches of high moisture-
25 content particles can be stored for up to about 12 hours until combined at
stage 4, although it is desirable to combine them almost immediately upon
forrnation. The separate colorized batches are combined or admixed at stage
4 in various prescribed proportions to provide a desired particulate material
mixture that will yield a reproducible "chaotic" color pattern in the finished
30 product. Three batches are shown in Figure 3. Each batch contains a single
monochromatic colorant. More or less batches of the same or different
volumes may be mixed as necessary to obtain the proportions required to
WO 95/04779 ~ PCTtUS94/09090
achieve a desired type and color of board stock. The color pattern ensuing in
the derived board stock is entirely determined by the relative proportions of
each of the particulate material batches in the admixture. For example, to
provide a board stock product that virtually duplicates the appearance of
natural "red granite," separate red, black, and white batches are prepared. The
individual white, red, and black batch stocks are then mixed in weight
percentage ratios of 25:37.5:37.5 to form the material used to form a pressure-
formed product in stage 5. For "green granite," a 50:50 mixture of green and
black particulate material is used. For "black granite," only black particulate
material is used.
Mixing of the batches of colored particulate material can be
carried out in a ribbon blender, a screw mixer, a tumble mixer, or any other
means typically used in blending particulate material streams, e.g., free-
flowing solids. Preferably, a ribbon blender is used to form a generally
uniform distribution of the differently colored particles. It is significant to
note that there is not discernible color transfer from one particle to another
during this blending process. This is at least because stable colorants are
chosen such that they do not significantly bleed under these blending
conditions or the subsequent processing conditions. Typically, this blending
process of stage 4 does not significantly alter the particle size of the
particulate material; however, the mechanical mixing process can reduce the
size of large particles and break up aggregates of particles. This homogenized
mixture is then typically screened to remove any rem~ining large clumps or
aggregates of material that may have formed in the mixing and blending
processes of stages 3 and 4. These clumps can be broken apart into smaller
particles or discarded as deemed appropriate. Preferably, a screening device
having a mesh size of 0.5 inch (1.3 cm) or smaller is used. Typically, a
vibrating angled diamond screen having a 0.38 inch (0.95 cm) screen size is
used.
Upon blending the colored batches at stage 4, a mixture of
relatively dry-feeling or semi-dry feeling ball-shaped particles is obtained,
although the particles contain a large arnount of water, e.g., about 55-75%
-
WO 9~/04779 ~ l ~i 8 6 6 4 PCT/US94/ogogo
29
total water content. Although the consistency of the mixture could be referred
to as being similar to coleslaw in that coleslaw contains "wet" particles, the
consistency is more like a flocculant, and is significantly different from the
aqueous slurries or adhesive-coated paper flakes that are prepared in most
5 conventional recycled newsprint processes.
This admixture of dirr~ ly colored particulate materials is
preferably subjected to a drying step, wherein it is partially dried for a
sufficient amount of time and at a sufficient t~ aL-I-e to reduce the
moisture content to less than about 20%, preferably to less than about 15%,
10 and more preferably to less than about 12%, based on the total weight of the
particulate material. While a small quantity of moisture is generally necessary
to permit the reaction of the resin with the fibrous cellulose and promote
preferential bonding, extraneous water hampers later curing and can increase
production costs. Although advantage is achieved by drying the combined
15 colored batches of particulate material, this is not necess~.ily a requirement.
That is, each batch can be partially dried prior to mixing them together.
Drying of the high moisture-content biocomposite particles can
be achieved using any of a variety of energy sources, i.e., microwave energy,
electrom~gnetic radio frequency energy, indirect infrared radiation, solar
20 energy, or thermal energy, for a sufficient period of time to reduce the
moisture content to the desired level. Preferably, the admixed particulate
material of stage 4 is dried with air before being pressure-formed at stage 5.
This can be accomplished, for example, by passing dry heated air through a
layer of the particulate material supported by an open metal mesh belt.
25 Although an air temperature of about 140-180F (60-82C) can be used to dry
the particulate material, it is advantageous to use a slightly higher
temperature, i.e., a temperature of about 175-375F (79-191C), and even as
high as 500-1000F (260-538C). The surface temperature of the particles
should not be so high, however, that the particles are scorched. That is, the
30 temperature of the particles themselves should not exceed 210F (99C),
although the air temperature may be significantly higher.
WO 95/04779 . . PCT/US94/09090
2 ~ 4
The residence time of the particles at the elevated temperature,
i.e., the exposure time, is typically no greater than about 30 minutes,
preferably it is about 10-20 minutes. The drying process should be
sufficiently gradual, however, so that case hardening does not occur, wherein
S a moisture-impervious shell forms at the surface of the particles and kaps
moisture within the particles. Slow drying also allows more uniform drying
of both large and small particles in the mixture. If a belt dryer is used, a total
heated zone residence time of about 5-15 minutes for any individual particle
is a~lopliate for an air temperature of about 200-2602F (93-127C). The
10 material is cooled prior to transfer to storage bins or to the next stages of the
process. Once dried, the particulate m~teri~l is significantly reduced in
volume and appears as relatively hard, colored, solid particles of irregular
shape with numerous rigid protruberances. In this form, the particulate
material is very stable, i.e., it has an indefinite shelf life.
The reduction in the moisture content of the high moisture-
content biocomposite particulate material permits a more rapid heating in the
molding press, without creating significant quantities of steam or liquid. For
example, a cured one-inch thick board can be formed during the hot press step
of Figure 3 upon applying 300-550 psi to the pre-dried feedstock with press
20 platens held at 280-320F over a period 5-30 minl~tes Without pre-drying the
particulate material, several additional min-ltes would be required with
consequent greater energy usage. Furthermore, unless the particulate material
is pre-dried, a "skin effect" can occur which tends to harden and seal the outersurfaces of the board stock and prevent the release of intern~l moisture,
25 thereby causing blow-outs in the material. Edges may also char or become
brittle to the point where additional mechanical processing is required to
remove the unusable edges. Drying the particulate material, preferably with
hot air, is therefore desired for a production process which incorporates
conventional radiant heating or thermal curing equipment. Regardless of the
30 compacting and curing process, however, drying the particulate material with
air provides advantages to the overall process.
-
WO 95/04779 2 ~ ~ 8 $ 6 4 PCT/US94/09090
. s s - . ~
31
Alternatively, drying can be achieved by compressing or
preforming the particles in the presence of heat, although this is not preferredbecause of the danger of causing blow^outs as discussed above. Final water
removal may occur at the "hot press" step or stage 5 of ~igure 3 (or in the
5 second roller compression step of stage 5 in the continuous extrusion process
of Figure 4).
At stage 5, the ples~ e-formed biocomposite material is
produced in one method by com~lessillg and curing the dry particulate
material at an elevated pressure and tell~eldl lre in a hot press. ~s used
10 herein, "elevated" refers to lenlp~ldLIlres and pressures above ambient
conditions, i.e., room temperature and atmospheric pressure. The press may
be either closed or open as in a caul or screen system, and the material may
be pressed to stops or thickness, or pres~L~e. If an open system is used, cold
pre-pressing may be necessary for h~ndling purposes. Alternatively, the
I5 particulate material can be roll pressed, reference Figure 4, or extruded in a
screw extruder, reference Figure 5. Preferably, the m~t~p~l is compacted and
heated in a one-step pressing operation to compress and further cure the
particulate material.
Prior to compressing the particulate m~teri~l, it is preferably
20 and advantageously coated with a secondary thermosetting binder, i.e., an
aromatic isocyanate, a sizing agent for water repellency, or a combination
thereof. This can be accomplished by blending the dry biocomposite particles
with the secondary thermosetting binder and/or sizing agent using a blender,
such as a continuous or batch-type ribbon blender or a batch-type or
25 continuous drum blender and coating, preferably spray coating using high
ples~ e purnps, air atomizers, mechanical atomizers (e.g., a spinning disc
atomizer), or a combination thereof. The secondary thermosetting binder
interacts with the dried particles cont~inin~ a protein-based resin that has been
at least partially cured in the drying step of the process. This creates a dual
30 resin system, i.e., a partially cured protein-based resin and an isocyanate resin,
which is believed to provide greater advantage than either used alone.
WO 9~;/04779 PCT/US94/09090
68~64
32
Final curing during the hot press step of Figure 3 may be
achieved in a thermal, microwave, or radio frequency heated environment.
Although such heating appliances are reasonably tolerant of moisture, energy
requirements are reduced if unnecessary moisture is first removed as discussed
S above. Typical press pressures are about 200-3000 psi, preferably about 300-
1000 psi, and more preferably about S00-600 psi. Although the pressure can
be applied in a progressive manner, typically it is more efficient to simply
apply a constant ples~ule for a specific period of time. For heated platen
systems, the platen temperature during the pressure phase is preferably about
10 250-340F (121-171C), while the ples~u~e is preferably about 450-750 psi.
Typical values are 500 psi and 320F (160C). It is to be appreciated that the
compacted particulate material can either be alternately heated and
co~ essed or simultaneously heated and compressed.
The duration of the final cure exposure and applied thermal
15 energy are selected in relation to the thickness and desired density of the
board stock being prepared. For a nominal board stock of 1 inch thickness,
an exposure time in the range of 3 to 5 minutes and 2 to 5 kilowatts per
square foot at microwave frequencies up to 2.5 GHz produces suitable board
stock. Although higher frequency microwaves may provide certain
20 advantages, most commercial grade ovens operate at lower RF frequencies of
13.6, 27, 44, and 100 MHz. The lower RF frequencies are equally effective
to obtain final curing. Using an electrically heated thermal platen press, an
exposure time of about 10 minutes and 10- 15 watts per square inch at 510 psi
and 320F (160C) produces suitable board stock. A thermally heated platen
25 press is particularly advantageous at least because it is economical.
Depending upon the pressure-formed product being prepared, a
suitable oven or heated press is selected that is capable of compacting the
particulate material to the desired size and providing a preferred throughput
rate and energy consumption. The particular heated, pour batch molding press
3~ or "hot press" used in the process of Figure 3 is shown at Figure 5 in cross-section view. This heated molding press is suitable for hot pressing a single
piece of board stock in the form of panels.
~ WO 95/04779 PCT/US94/09090
216~64
33
The body of the press mold is constructed of suitable heavy
metal side and bottom walls l. The walls 1 must be capable of withst~n(ling
the maximum molding pressure, and may or may not be lined with a porous
member 2 to t~c~ilit~te removal of extracted liquid, such as a screen, although
5 this is not necessary. A movable top wall or colllpres~ion plate 3 is providedalong the top of the mold and is coupled to hydraulic means (not shown for
convenience and clarity, but similar to the piston 20 described below) for
directing the plate 3 to compress any feedstock material contained in the mold
cavity 4.
The compression plate 3 includes a heated platen, which could
be heated electrically or by stearn or hot oil, or an RF or microwave plate
electrode 5 which is connected to a suitable coaxial cable, waveguide, or
electric wire 6. If RF or microwave energy are used, this coaxial cable or
waveguide 6 is fed through a connector 7 from an RF or microwave generator
15 10. Also, the center conductor 8 (for RF or microwave energy) of the cable 6
passes through an electrical and thermal insulator plate 9 which separates the
compression plate 3 from the electrode 5. Any substitute plate 5 should be
capable of providing adequate heating to the dry particulate material.
Once the particulate material is compressed, heated, and cured
20 to a final form as shown at 11, the material is extracted from the mold cavity
4 via the piston 20. The piston 20 is secured to a plate 12 along the bottom
wall 1 of the mold. An additional (optional) porous member 13 may be
placed below the member 2 and between the formed board stock 11 and the
plate 12 to facilitate removal of released moisture, although predrying the
25 particulate material is preferred to avoid the release of large amounts of water
during this compression process. Either or both the bottom and top plates 12
and 5 may also contain an array of holes or channels 14 to permit the
extraction of released moisture via insulating tubing 15 and a vacuum or other
suitable suction pump 16. Again this is not necessary when the preferred dry
30 particulate material is used.
With attention next directed to the extrusion process of Figure
4, a generalized block diagram is shown of a continuous production process,
WO 95/04779 PCT~US94/09090 --
8~ ~ 34
rather than a batch process. Such a process and related production equipment
can be constructed to provide board stock thicknesses varying from 0.125 inch
(0.318 cm) to 2 inch (5.08 cm) thickness and would typically include an
extrusion molder and a continuous feed oven or heating assembly. The
5 configuration of a screw type extrusion apparatus 30 which could be used in
the extrusion process is shown at Figure 6.
With the exception of the final stage 5, each of the process
stages of Figures 3 and 4 are the same, use comparable equipment and
generally follow the procedures outlined above. That is and as in Figure 3, a
10 mixer is provided at stage 2 to receive a~proll.;ate solids and liquids from the
raw m~teri~l storage hoppers or liquid supply lines of stage 1. Motorized
paddles or augers mix the separate cellulosic material and adhesive binder
ingredients to a pler~ d con~i.ctency before transferring the ingredient
mixtures to a mixing station at stage 3 for ~lcpa~ g each of the colorized
15 particulate material batches.
In lieu of separate mixing stations at stages 2 and 3, a single
mixer may be provided that includes a hopper. The hopper can contain a
suitable supply of shredded cellulosic m~teri~l. The hopper may also include
a shredder mech~ni~m for final shredding of the cellulosic material to an
20 ~pLopl;ate particulate size or shredding an additive pulp stock mixed with the
base pulp stock. Once al)plopl;ate qn~ntities of the resin and cellulosic
material are ~Amitted to the mixer, the materials are blended with the colorant
and any additional additives, such as strengthening fibers, etc., are admixed.
The batches of monochromatic colored particulate material are
25 next admixed at stage 4 at either a different mixer station or combined usingthe mixer of stage 3. The produced mixture is next fed through a screw drive
extruder 30, such as shown at Figure 6. The feedstock is particularly received
at a material intake 32 and lower lying screw 34, which is driven by a drive
system 35, the feedstock is progressively compressed, extraneous liquids are
30 removed and the feedstock is shaped by forcing the feedstock through an
ap~lopl;ately configured extrusion die 36. The shape of the die orifice 38 can
be varied as desired. For a board stock, a wide, shallow height rectangular
~16~6~
Wo 95/04779 PCT/US94/09090
orifice 38 is used. The tenlpe.dLu,~ of the particulate material along the path
of the screw 34 can be varied with provided heaters which surround the screw
34 and control signals applied at temperature control lines 40.
Upon exiting the extrusion die 36, the extruded material is
5 typically admitted to a sequential arrangement of shaping rollers, such as roll
formers or pinch rollers, which may include or be separated by intervening
RF, microwave, or thermal heating appliances. The particulate material is
progressively compressed as it is further heated. Final compression and
forrning occurs at the outboard section of shaping rollers. Residual moisture
10 and vapor is bled off at each roller section prior to the shaped particulate
material being final cured. Final curing can occur in a larger oven appliance
which may surround the rollers or at a separate oven appliance which cures
the formed particulate material to proper composition and dryness.
An equilibrating process, such as a post-curing or kiln drying
15 process, may be incorporated into either of the processes of Figures 3 and 4
for acclim ~ting the pressure-formed material to the ambient environment. In
this way, the internal stresses resulting from l~les~ul~-forming can be
gradually relieved and the entire volume of material allowed to come to
equilibrium with ambient conditions. Whether the particulate material is
20 formed by compression at Figures 3 and 5 or extrusion at Figures 4 and 6, it
typically will retain less than 20%, and preferably less than 12% moisture.
An equilibrating process can be incorporated into either of the
processes of Figures 3 and 4 for acclim~tinF the pressure-formed material to
the ambient environment. In this way, the int~ l stresses resulting from
25 ~le;,~ule forming can be gradually relieved and the entire volume of materialallowed to come to equilibrium with ambient conditions. This equilibrating
can either be accomplished by post-conditioning or kilning.
In post-conditioning, the panels are surface-planed immediately
after pressing while they are still hot. The planed panels are stacked, i.e.,
30 piled one on top of another, while still hot and covered with an in~ tin~
counter weight flat panel material. The panels are stored in this fashion for atleast about one week prior to sizing and fini~hing.
WO 9~/04779 PCT/US94/09090
~ 6~6~
36
Kilning, i.e., kiln conditioning, is typically performed over 6 to
24 hours at an oven temperature of about 130-210F (54-99C), preferably
about 150-180F (65-82C), with humidity levels up to about 75% relative
humidity. The board stock material is typically constrained in a horizontal
S position during this conditioning process. The oven parameters are then
gradually reduced to ambient conditions over several days to several weeks.
Once dried, the pressure-formed material is preferably stored in a controlled
humidity environment to m~int~in 6-8% moisture in order to enhance long-
term dimensional stability. Kilning is typically not used in the process in
which the secondary thermosetting resin, e.g., aromatic isocyanate, is used.
At the fini~hing stage, the pl~s~ule-forrned product is surface
finished, sized, and final shaped according to various market requirements by
s~n~ling, planing, sawing, shaping or other a~ropliate machining. The
surface may also be coated or impregnated with preferred sç~l~nt.s, lacquers,
varnishes, or the like. By bonding individual board stock components
together, complex shapes and assemblies can be forrned. Composites with
other materials can also be obtained, such as by inlay. Seamed assemblies
preferably are formed from stock prepared from common color batches to
minimi7:~ any color differentiation.
A desirable attribute of the board stock produced by either of
the processes of Figures 3 or 4 is that the material is relatively dense and thus
can be finished to forrn relatively smooth edges and surfaces.
Correspondingly, the edges accommodate conventional finger, or butt jointing
and other jointing techniques to enable the production of larger, complex
pieces from multiple smaller pieces. Also, because of the unique
homogeneous color patterns exhibited throughout an entire volume of the
produced stock, essentially seamless joints are obtained. Complex shapes,
such as corner molding, contours and orthogonal jointed corners can also be
formed with invisible seams at the juncture of two pieces of the same
material.
A further attribute of derived board stocks is that compositional
or inlaid board stock can be fabricated, using the batch processing method. In
WO gs/~4779 37 P~/US94/09090
such instances and for the process of Figure 3, separately prepared inlay
members, such as commemorative plates, name tags or the like, and which
ca~ comprise almost any organic material (e.g., wood or plastics), or inorganic
material (e.g., metal) can be inserted into the press mold prior to stage 5.
Once the feedstock is added and compressed during the molding process, it
forms around and in intim~te contact with the inlay material. Upon final
curing, a strong contact and bond is made between the inlay and board, and
the inlay becomes an integral part of the resulting board stock.
Still another attribute of the invention and using the process and
equipment of Figures 3 and 5 is the ability to machine the upper and/or lower
plates either in cameo or intaglio with a desired figure or shape. When the
particulate material is pressed in the mold, the board stock will retain an
image of the pattern provided in the mold plate either intaglio or cameo,
depending upon whether the mold plate pattern is raised or inset. Upon
curing the raw product or board stock, the pattern is retained in fine detail.
The following examples are offered to further illustrate the
various specific and ~,cre.lcid embodiment~ and techniques. It should be
understood, however, that many variations and modifications may be made
while rem~inin~ within the spirit and scope of the present invention. As used
in the examples, and elsewhere herein, percentages are weight percentages
unless otherwise indicated.
F~perimental EYs~nlplesF~mple 1
Process for Production of l~id ~iocomposite Materi~l
Table I, below, summarizes a preferred process sequence and
recipe for the ~ule~dldlion of a single monochromatic batch of feedstock
through stages 2 and 3 described above. This process and recipe are referred
to as "A" in the following examples.
,
WO 95/04779 ~ ~ 6 8 6 6 4 PCT/US94/09090
38
TAP~.F. I
Si~le Color~nt Particulate Material Preparation
Process A
Tr~redientAmount (pounds)
Step 1: Combine Water (60-70 F) 175
Ground soybean 97
Pine oil 3
Mix 3 minntec or until smooth
Step 2: Add Water (60-70F) 169
Mix 2 minutes or until smooth
Step 3: Add 25% Potassium or Sodium 14
hydroxide solution
Mix 1 minute
Step 4: Add ~' brand sodium 25
silicate
Step 5: Add Soy resin from steps 1-4117
Polyvinyl acetate 24
Mix 1 minute
Step 6: Add H.B. Fuller RM-0255 8
wax emulsion
Mix 1 minute
Step 7: Add H.B. Fuller WB-2523 8
("FULLREZ") formaldehyde-
based latex resin
Mix 1 minute
Step 8: Add Desired metal oxide
colorant
Mix 1-2 minutes
Step 9: Combine Resin from steps 1-85 parts by weight
(83% by
weight)
Cellulose material1 part by weight
(17% by
weight)
Mix until particulate material forms (2-10 minutes)
WO 95/04779 - PCT/US94/09090
~1 ~g66~
39
Step 10: Dry the particulate material to a moisture content of about 12-
15%
Modest variations of the formulation of Table I has been shown
to produce final pressure-formed materials, e.g., board or shaped stock or
5 panels, with more stable or optimized mechanical properties. Tables Il and
III, below, have each been shown to produce l,les~ -formed panels with
improved properties of hardness and moisture sensitivity.
TABLE II
10 Altern~-tive Parti-~nl~t~ Forlnulation
Ingredient % Wet-Weight
Water 55-9
Cellulose (paper) 16.9
Soy flour 15.8
Sodium silicate4.1 (50% in water)
Sodium hydroxide2.3 (50% in water)
Lime 1.9
Styrene-butadiene latex 1.6
Tint 0.7
Pine Oil 0.5
H.B. Fuller WB-2523 0.3
(FULLREZ)
TOTAL 1 00.0
WO 95/04779 ~ ~;8~ PCT/US94/09090
TABLE III
Particulate Formulation-WaterResistant
Ingredient % Wet-Weight
Water 54.8
Cellulose (paper) 17.7
Soy flour 15.4
Sodium silicate4.0 (50% in water)
Sodium hydroxide3.1 (45% in water)
Lime 1.9
Styrene-butadiene resin1.6
Tint 0.7
Pine Oil 0.5
H.B. Fuller WB-2523 0.3
TOTAL 1 00.0
Table IV demonstrates comparable values of wood based
20 products to measured values of ~ -formed samples of one-inch
thickness, which were produced generally in accord with the above-listed
mixtures (i.e., Process A and variations thereof listed in Tables II and III).
WO 95/04779 41 I'CT/US94/09090
o ~
o
~ 3 o ~ ~ o o
Z v~ o
V~
~ " _ X ~ o~
o, ~
~ e ~ Z ", ~ O ~ I I A
O o
_~ O O o
O ~ ~q g O o O o O o
o
O ~ ~ o l-- ~~o O ~
p ~ ~ ~ o~ , o
I_ C5
e~ = o ~ C ~~o
~ o ~ ~ ~ ~ V ~
Z ~ ~ 3 ~ ~ 3 o ~
WO 95/04779 PCT/US94/09090
6~.~
42
TART,~ V
Particulate Material Preparation
Process B
Tn~redient Amol-nt (polln~
Step 1: Combine Water (60-70 F) 680
Colorant 46
Pine Oil 9
Mix 1 minute
Step 2: Add slowly H.B. Fuller PD512 71
(ca. 20 min.) (latex emulsion with PVA)
Mix 1-2 minutes once completely added
Step 3: Add Honeymead 90 pdi 280
with mixing soy flour
Mix with chopping 2 minl1te~
Step 4: Add Water (60-70F) 238
with mixing
Mix with chopping 2 minlltes
Step 5: Add Lime water solution 103
with mixing (50 lbs. in 13 gallons water)
Mix with chopping 2 minutes
Step 6: Add 45% Potassium 33
with mixing hydroxide solution
Mix with chopping S minutes
Step 7: Add 'N' brand sodium 72
WO 95/04779 ;~ 16 8 6 q 4 PCT/US94/09090
43
with mixing silicate
Mix with chopping 2 minutes
Step 8: Add H.B. Fuller WB-2523 5
with mixing ("FULLREZ")
Mix with chopping 2 minutes
Step 9: Add Hammer-milled 330
nontreated inked newspaper
Mix until particulate material forms (6-10 minutes)
Step 10: Combine separate batches of colored particulate material
prepared in steps 1-9
15 Step 1 1: Dry the particulate material to a moisture content of about 10-
15% (ca. 50 minlltPs)
The materials used were obtained from the following suppliers:
potassium hydroxide, Hawkins Chemical Distributors (Minneapolis, MN);
20 lime, Mi~si~ pi Lime MR200 distributed by Hawkins Chemical Co.; aqueous
dispersion of PVA and a latex emulsion (50-52.5% PVA), H.B. Fuller
Company; "N" brand sodium silicate, Hawkins Chemical; pine oil, Union
CamLp Co. distributed by Hawkins Chemical Co.; 90 pdi Soy flour,
Honeymead Co.
wo 95/04779 .i, ~ PCT/uSs4/09090
æ ~
44
TART,F VI
Particul~tP Material Preparation
Process C
Tru~redient ~mount (pounds)
Step 1: Combine Water (60-70 F) 814
Colorant 23
Pine Oil 9
Mix 1 minute
Step 2: Add Honeymead 90 pdi 280
with mixing soy flour
Mix with chopping 2 minlltes
Step 3: Add Water (60-70F) 193
with mixing
Mix with chopping 2 minutes
Step 4: Add Lime water solution 206
with mixing (50 lbs. in 26 gallons water)
Mix with chopping 2 min~lt~s
Step 5: Add 45% Potassium 33
with mixing hydroxide solution
Mix with chopping 5 minutes
Step 6: Add ~' brand sodium 72
with mixing silicate
Mix with chopping 2 minutes
WO 9~/04779 ; ~ ~ ~ 6 ~ 6 6 4 PCT/US94/09090
Step 7: Add H.B. Fuller WB-2523 5
with mixing ("FULLREZ")
Mix with chopping 2 minlltec
Step 8: Add Hammer-milled 340
nontreated inked newspaper
Mix until particulate material forms (6-10 minutes)
Step 9: Combine separate batches of colored particulate material
prepared in steps 1-8
Step 10: Dry the particulate material to a moisture content of about 5-
7%
(ca- 50 minllte~)
The particulate material made by Process C is typically further
blended with a wax emulsion and MDI for desired results.
20 The Fffects of Water Cortent Of Soy l~e~in~ or M~hanical and Pllysical
Proper~ies of panel~
The amount of water that is added to the soy-based resin may
influence several factors in the production of ~res~ -formed panels. Some
of these include adhesive viscosity, which det~rrnin~s the amount of paper
25 particles covered by the resin, the depth of color of the dried particles, drying
rates, and particle geomet~y. To determine the impact of varying the moisture
content of the particulate material, six batches were produced using Process B
adjusted proportionately to a laboratory scale (Aquasperse II liquid lamp black
colorant obtained from Huls Inc., Piscataway, NJ was used). The only item
30 varied between the batches was the amount of water added. This produced
wet particles with the specified target moisture contents. These moisture
contents were 55%, 59%, 63%, 67%, 71%, and 75%. The viscosity of each
WO 95/04779 ~ 3 ~ PCT/US94/09090 --
46
resin mix was measured prior to the addition of paper. All batches were dried
to approximately 11% moisture content prior to pressing.
Four test panels per moisture content were pressed on a 2 ft. x
2 ft. laboratory hot press. Platen te~ dlul~ was 315F (157C), panels were
5 pressed at a constant 512 psi for a total press time of approximately 13.5
mimlte~ Press time varied due to slight fluctuations in mat moisture content
prior to pressing. Target panel specific gravity was 1.30 coming out of the
press.
Three of the four panels produced per moisture content level
10 were used for mechanical and physical panel property testing. From each
panel, five static bending and five water absorption/thickness swell samples
were obtained. Hardness samples were taken from the ends of tested static
bending specimens. All testing was done in accordance with ASTM D-1037,
Standard Methods of Evaluating the Properties of Wood-Base Fiber and
15 Particle Panel Materials. Descriptive statistics were generated for all
properties evaluated at the various target wet particle moisture contents.
The results of the viscosity measurements are as follows:
55% 640,000 cps
59% 440,000 cps
63% 184,000 cps
67% 104,000 cps
71% 74,670 cps
75% 37,330 cps
The viscosity directly impacts the coatability of the paper with
25 the soy adhesives. The higher viscosity resins exhibited reduced coverage of
the paper in the blending process. This could negatively affect the resulting
panel properties. Conversely as the viscosity was lowered due to the addition
of water the color of the blended material was enhanced. This enhanced color
held through the process. The resulting pressed panels had different shades of
30 black depending upon the wet particle moisture content. The drier particles
were light in color while the high moisture-content particle produced very
deep-black panels.
21~6~
WO 9~/o477g PCT/US94/09090
47
The viscosity also affects the bulk density of the final dried
particles. As the viscosity was lowered due to increased water in the blend
the bulk density of the particles also decreased. This is due to the increased
shearing action that takes place in the blender at the lower viscosities. The
5 average panel specific gravity by wet particle moisture content is as follows: 55% 1.31
59% 1.33
63% 1.33
67% 1.33
71% 1.31
75% 1.36
The results of mechanical testing are presented in Table VII.
The panels made with the lowest wet particle moisture content exhibited the
poorest strength and stiffness. This is due to the poor coverage of the soy
15 resin on the paper particles. The numerous uncoated paper particles that are
present in the final press panel do not provide significant strength or stiffn~ss
to the panel. This resulted in the 55% wet particle moisture content panels
having poorer mechanical L,-opcllies when colllp~ed to the other moisture
contents. As the viscosity of the resin decreased, the coverage of the paper
20 particles increased. However, there was no detectable difference in strength
or stiffness between the wet particle moisture content ranges of 59% to 75%.
The hardness values are also presented in Table VII. As the
bulk density of the dried particles decreased with the higher wet particle
moisture content, the hardness values had a tendency to increase. This was
25 caused by the finer particles that make up the panel being able to con~ t;ss
tighter than larger particles. This in turn made for a less porous, harder
surface.
TAT~T,T~' VII
Selected Mechanical Properties
% Moisture Content MOR (psi) MOE (psi) Hardness (lbs)
of Particles Prior
to Drying
2,351 296,912 2,070
WO 95/04779 ~ - ' PCT/US94/o9090
~1~8~ 48
59 2,965 430,019 2,573
63 2,704 383,793 2,372
67 2,431 315,575 2,498
71 2,724 572,831 4,610
2,344 357,228 3,194
These values represent the mean of 15 test specimens.
The results of the physical property testing is presented in
Table VIII. The panels m~nl-f~ctured with the 55% wet moisture content
10 particles exhibited poor two-hour and twenty-four hour edge swell
characteristics. This was due to the poor soy resin coverage previously
mentioned. Furthermore, the results for the higher wet moisture content
particles, 71% and 75%, exhibited extremely poor short term and long term
edge swell characteristics. The twenty-four hour results for these two
15 combinations were actually worse then indicated because the mean is based on
the number of samples left to measure after twenty four hours of submersion.
With the 71% test samples, eight fell apart in the water and with the 75% test
samples, three fell apart in the water. Thus, with higher water content, the
soy resin does not have sufficient bonding strength due to resin washout (i.e.,
20 dilution) to hold the panel together when the panels are immersed in water.
The moisture content that exhibited the overall best two-hour and twenty-four
hour edge swell values was 67%.
TARRF VT~
Selected P~lysical Properties
% Moisture Content Percent Percent
of Particles Prior Two-Hour Twenty-Four
to Drying Edge Swell Hour Edge Swell
16.2 43.7
59 11.7 40.1
63 19.4 52.3
67 8.2 41.4
71 14.9 67.7
12.0 57.3
These values represent the mean of 15 test specimens.
Combination 71% had 8 samples fall apart and combination 75% had 3
WO 95/0477~ 21 ~ 4 PCT/US94/09090
49
samples fall apart during testing.
F~nple 3
The Fîfects of ~dditiorl~l Paper or Physical ~nd Mech~nical Properties of
Panels
The addition of extra paper and water to the recipe detailed in
Process B (cont~ining PVA) produced a less dense panel, while reducing the
paper to resin ratio resulted in lighter panels and reduced panel cost per
square foot. The arnount of paper used was increased to 1-1/2 times the
amount norrnally used. The amount of water added was adjusted accordingly
to keep the total water content in the batch at 61%. The total volume of the
resin was adjusted to a laboratory scale. The particles were dried to 11%, and
pressed in a 9-3/4 inch x 4 inch mold. Only two panels were produced and
were tested in the usual manner.
With the larger amount of paper, the paper did not coat as well
as usual. Therefore, it was mixed longer to allow better coating. The
par~icles were long and slender as opposed to the round shape that is usually
produced. No noticeable differences were noted during pressing. Mechanical
properties of the test panels showed little variation from standard panels. A
slight decrease in water absorption properties was noted.
S~mrle ~ ~ H~r(lness %Wt. G~in %Vol. Swell
453,379 3251 4157 19.83 22.65
2 528,765 3185 4056 18.34 21.84
Panel density only dropped slightly, from 1.3 down to 1.21.
This could be the reason for poorer absorption properties. The dried particle
density was about 0.97 compared to 0.93 for standard particles. Dried particle
density was measured by volurne displacement. A known mass of particles
was put into a liquid that had a specific gravity of approximately 0.68 g/cc,
and the volume the particles displaced was recorded. The reason for the
lower density of the higher paper content particles is the longer mixing time
and the smaller particle size. When round particles are produced with mostly
a paper core, air pockets are trapped within the particles resulting in a less
WO 9~/04779 21 6 ~ 6 6 4 PCT/US94/09090
dense particle. When mixed longer, particles keep breaking down and air
pockets get filled with resin, producing a more dense particle.
.E~a~e 4
Evalualion of ~e ~alio of p~pr 'to Soy l~in ~ the Pn~ on of Panels
S The ratio of cellulose m~t~ l to resin influences the
m~h~nical and physical properties of colI~o~ile panels. If there is not
enough cellulose m~t~ l within the composite panel, then the strength and
stiffil~.c~ will be l~king On the other hand, if the cellulose material content
is too high in the cornposite panel given the same resin lo~(lin~, then the
10 tiim~n~ional stability of the panel is l~l~king Three half batches were made in
the 60 cu. ft. mixer using the recipe detailed above in Process B (c~."l~i";.,~
PVA). The ratio of paper to soy resin was varied for the three rnixes. The
ratioswereas follows: 0.8:1.0, 1.3:1.0, and 1.5:1Ø Themixedm~teri~l was
dried to ~ ~lely 11% moisture content prior to pressing.
The panels were pressed in a 2 ft. by 2 ft. hot press. Platen
temperature was 312F (156C) and a total press tirne of a~n~ ely 13.5
mimlt~.c. Press tirne was varied slightly to compensate for h~n~ in furnish
moisture content. Target specific gravity of the panels coming out of the
press was 1.30. The panels were tested for streng~, stiffn~, hardness, and
20 ~lim~n~ional stability. Descriptive st~ti~ti~ were generated for the p~u~llies
tested for each ratio.
The paper in the mi~ure ~,~al,_d with the 1.5:1 paper to soy
resin ratio did not coat well, whereas in the other rnixtures, the paper coated
well. As the paper ratio increased the bulk density of the dried particles
25 decreased. No problems were encountered during pressing of the three
different ratios.
The following is a summary of the m~h~nical properties of the
panels produced in this study. The values listed represent the mean of 15 test
specimens.
WO 95/04779 ~ 16 8 & ~ 4 PCTtUS94/09090
Ratio paper/ Modulus Modulus
soy adhesive of Elæticity (psi) of Rupture Hardness (lbs)
(psi)
0.8:1.0 263,958 2,225 2,241
1.3:1.0 529,553 3,014 3,525
1.5:1.0 525,240 2,909 3,384
The 0.8:1.0 ratio had significantly (P < 0.05) lower strength,
stiffnP~, and hardness due to lack of sufficient cellulosic material present
15 within the composite panel. The cellulosic material provides int~n~l support
to the structure of the panel. The following is a sumrnary of the physical
properties of the panels produced in this study. The values listed represent
the mean of 15 test S~Cil~lellS.
Paper:Soy Resin 2-Hour 24-Hour
20 Ratio Edge Swell (/O) Edge Swell (/O)
0.8: l .0 7.29 37.26
1.3:1.0 10.32 35.87
1.5:1.0 22.02 55.48
Ihe 1.5:1 ratio had significantly (P < 0.05) higher two hour and
twenty-four hour edge swell values than the other two con~i~Lions. The
high amount of fiber present in this ratio did not receive adequate coating by
the soy resin. lherefore, there was insufficient int~n~l bond strength of the
composite panel to prevent it from swelling considerably when s~melged in
35 water. lhere was no significant (P < 0.05) diLL~lel,ce between the two lower
ratios used in this study.
This example shows that the ratio of paper to soy resin solids is
important in the production of the panels according to the present invention.
If the arnount of paper is too low, the panels produced do not have sufficient
40 strength and ~l i Ir"~ If the amount of paper is too high, the panels do not
WO 95/04779 ; ~ PCT/US94/09090
~g`~6~
52
have sufficient water resistance.
~n~e 5
Fvalua~on of P~'tein Dispelsion ~dex
This t;~)ClilllCllL was c(m~ cte~l to c~ e the level of
5 soluble protein required in the soy flour. Soy flours with a pdi of 20 and 70
were ev~ t~1 R~trhPs of composite particles were m~mlf~rtllred in a
laboratory 60 cu. ft. mixer using the recipe detailed in Process B (co"l;l;";l~gPVA) scaled down pl~ ionally, but s~lkstit~lting the various pdi flours for
the standard 90 pdi flour used. The particulate material was dried to
10 a~loxill~Lely 11% moisture content prior to pressing
Panels were pressed in a 2 ft. by 2 ft. hot pre~ss. Platen
temperature was about 312F (156C). The total press time was about 13.5
mimlte~ Press time varied due to slight flllrtll~tions in the moisture content
of the particulate material. The target specific gravity of the panels coming
15 out of the press was 1.30. The panels were te~sted for strength, stiffnP~,
hardne~ss, and ~1imPmional stability. Descriptive statistics were generated for
each plo~lly evaluated.
There were no problen~s encountered in either m~mlf~rtllring of
the composite particles or the pre~ssing of the panels. The following is a
20 summary of the mPrll~nical properties of the panels produced in this study.
The values are the mean of 15 test ~cilllells.
PDI of Modulus of Modulus of Hardness
Flour Elasticity (psi) Rupture (psi)
301,960 2,554 2,784
266,504 2,485 2,416
The mrrh~nical properties of the panels produced in this study
compared similarly to the m~rh~nical properties obtained with the 90 pdi soy
flour. There was no significant (P < 0.05) difference in any of the mechanical
pl`u~t;llies tested between the 20 pdi and 70 pdi soy flour. The following is a
35 summary of the physical properties of the panels produced in this study. The
wo 95/04779 21 68 6 6~ PCT/US94/09090
values listed are the mean of 15 test specimens.
PDI of 2-Hour 24-Hour
Flour Edge Swell (/O) Edge Swell (/O)
5.95 30.65
4.84 24.01
The 20 pdi soy flour had significantly (P c 0.05) higher twenty-
four hour edge swell values than the 70 pdi soy flour. The effect of the
protein providing the int~n~l bonding strength that ~nh~nrP~ the dimensional
stability is evident. The 70 pdi soy flour exhibited rlim~n.~ional stability
15 values similar to the 90 pdi flour.
Thus, a wide range of protein dispersion index soy flours can
be used to produce the composite particles of the present invention. The
lower pdi flours produced panels with similar mPr~nical p~ ies as panels
produced using 90 pdi flour. However, for dimensional stability, higher pdi
20 levels are required.
e 6
~nn o~ (~n~e SoyMeal
It was the objective of this study to (lt~tr~ e if a soy meal
could be used in place of soy flour to produce the composite particles that
25 make up the panels of the present invention. A 60 cu. ft. batch was
m~mlf~*lred in the laboratory blender using the recipe detailed in Process B
;l Ig PVA) adjusted a~l~L~liately in volume. The only variation from
the recipe was the sukstihltion of soy meal for the soy flour. The particulate
material was dried down to approximately 11% moisture content prior to
30 pressing.
Panels were pressed in a 2 ft. by 2 ft. hot press. The press
t~ ,dLure was 312F. Total press time was approximately 13.5 mimltt ~.
Press time may have varied due to slight fluctuations in moisture content of
the particulate material. Target specific gravity out of the press was 1.30.
35 The panels were tested for strength, stiffn~, hardness, and lim~n~ional
WO 95/04779 ~16 ~ 6 6 ~5 PCT/US94/09090
54
stability. Descriptive statistics were generated for the properties tested.
After the addition of the soy meal to the mixture, the resin was
noticeably thicker and more chunky than the usual recipe. After the paper
was added there was no noticeable difference in the particles than the typical
S recipe used. No problems were encountered during the pressing of the panels.
The average specific gravity of the panels m~mlf~ lred was 1.25. The
following is a summary of the data collected from test specimens. The values
listed are the mean of 15 test specimens.
Modulus of Modulus of Hardness 2-Hour
10 Elasticity (psi) Rupture (psi) (lbs) Edge Swell (%)
523,823 2,266 3,067 31.80
The panels produced in this study had m~h~nical properties
equivalent to panels m~mlf~*lred with soy flour. However, twenty-four hour
edge swell data could not be collected because the specim~n~ inte~rated in
the water.
~ 7
Use of ~.~o~; ~ Based Binrlf r
Composite particles were prepared using recipe B (co"~;"i"g
PVA) adjusted to a laboratory scale, and dried to 6.5% moisture. Some of
these composite particles were pr~ -formed as is, and others were coated
25 with MDI (50 grams per 500 grams of particulate). Panels were pressed from
this material using a te~ re of 320F (160C), a pressure of 520 psi, for
6 mimlt~ (Sample 1 = uncoated particles and Sample 2 = MDI-coated
particles). Panels were pressed at this same temperature and pressure but for
half the time, i.e., 3 mim~t~s (Sample 3), or for 6 mimlt~ at the same
30 temperature and a pressure of 780 psi (Sample 4). The mP~h~nical properties
of these panels are listed below.
wo 95/04779 216 8 6 6 4 PCT/US94/09090
Sarnple Modulus of Modulus of Hardness (lbs)
No. Elasticity (psi)Rupture (psi)
1 444,000 3,126 3,019
2 691,000 4,911 4,330
3 291,000 1,576 571
4 797,000 6,027 4,447
These results indicate that the addition of MDI results in
15 improved m~h~nical p~u~lLies relative to the same particles pressed into
panels without MDI. Furlh~rmore, these results show that increased pressure
and time of pressing results in improved m~h~nical pl~u~lies as well.
~e 8
Use of l~o~ ~Bas~ P~inder ~ml W~ Fm~ ion
Composite particles were blended using the recipe detailed in
Process C. The conl~sil~ particles were dried in a belt dryer from
al~loxill~lely 61% moisture content to a 6% target moisture content. The
particles were then blended with various levels of MDI and wax emulsion.
The MDI used in this study was Rubinate 1840 from ICI Polyurethanes (West
25 Deptford, NJ). It is a urethane-based th~rmosetting clear, brown liquid with a
viscosity below 500 cps at 77F (25C). Ihe wax emulsion was 2100P from
Hercules Incol~ul~led (Minneapolis, MN). Particles (150 Ibs) were blended
with the various co~ il~lions of MDI and wax emulsion outlined below in a
batch-type blender. Once the MDI and wax had been applied to the particles,
30 the mixture was blended an additional two mim t~s to ensure even mixing and
distribution of the components. The amounts of MDI used in this study were
3.5/4 5.0/4 7.5/4 and 10%. The amounts of wax emulsion used in this
study were 1.0% and 1.5%. Each MDI-loading was evaluated with both wax
levels.
If the moisture content of the particles was below the target
6/4 then additional make-up water was applied during the blending process.
This was done to target the pre-press mat moisture content at approximately
WO 95104779 ~ 4 PCT/US94/09090
56
8%. The blended material was transferred to a loader-hopper and an even mat
was laid down in the forming box. The target thickness of the panels coming
out of the press was 0.90 inches with a target density between 1.25 and 1.30.
The pressing pararneters used to achieve these targets were platen temperature
5 of 325 F (163C) at full pressure of 525 psi until the desired thickness was
obtained. The pressure was then reduced sufficiently to hold the panels at the
desired thickness. The total press time was 8.5 mimTt~: 8.0 mimlt~ pressure
cycle and 0.5 mimlt~ degas cycle. Panels coming out of the press were hot
stacked for at least twenty-four hours before cutting into test samples.
Two sets of co~ ~;son panels were m~mlf~ red. One set
using the recipe detailed in Process B (c(.~ it~g the PVA), and the other set
using particles from this same recipe (c( nt~ining PVA) and then blended with
5% MDI resin. These comparison panels were pressed using the same
L~ outlined above.
From the pressed panels, five were randomly selected to test for
m~ nical and physical panel p~ Lies. From the selected panels five static
bending samples were cut to ~let~rmin~ strength and stiffn~ Also, five
water absorption and thickness swell sarnples were cut from three of the
selected MDI/wax panels and five of the latex and latextMDI panels. Testing
20 was done in accordance with procedures set by ASTM Standard ~1037,
Standard Methods of Evaluating the Properties of Wood-Based Fiber and
Particle Panel Materials.
Descriptive statistics were generated for each combination of
resin loading, wax loading, and ~ro~lly tested. The data was analyzed using
25 classical statistical techniques for significant differences between groups for
the various properties tested.
While moisture content of wood-based materials is a concern in
the m~mlf~(*lre of composite panels, ~--l~allies typically tly to run as dry as
economically feæible to ensure fæt press times and increæed throughput of
30 the mills. However, the moisture content of the particles before and a~er
blending is important in the m~mlf~ re of panels when using MDI resin.
The moisture content of the composite particles coming out of the dyer should
WO 95/04779 ;~ i 6 8 6 6 ~ PCT/US94/09090
57
~l~r~l~ly be in the range of 6% to 7%. This is needed to ensure that the
post blending moisture content of the particles is very close to 8% before
pressing If the moisture content of the particles is low, compaction of the
mat in the press is inadequate be~ause the panel can not reach either the target5 thickness or target specific gravity. This produces panels with very porous
surfaces. If the moisture content is high, then the excess water in the panel isconverted into large amounts of steam under the pressing conditions of
pressure and heat and causes numerous blow-outs of the panels coming out of
the press. Tl~ r~l~, using particles with either too low or too high of a
10 moisture content during pressing with MDI is undesirable.
The results of the m~rh~nir~l property testing are presented in
Table IX The combination cn~ g just the soy adhesive with the
PVA/latex had significantly (P < 0.05) lower modulus of rupture (strength)
than the colnl~il~lions blended with MDI adhesive. There wæ no significant
15 (P < 0.05) differences d~tecte~l in strength due to increasing MDI levels.
However, there wæ a general trend: as the MDI addition levels increæed, the
strength also increased.
The combination c~,,l;.lll;ll~ just the soy resin with the
PVA/latex had significantly (P < 0.05) lower m~ lc of elasticity (stiffnr-cc)
20 than the coll~il~lions blended with the MDI. As with the streng~, there was
no significant (P < 0.05) di~ ces l1ete~teA in stiffin~.c due to increasing
MDI levels. However, the same trend was present: as the MDI addition rate
increased, the stiffnP-ss also increased.
The strength and stil~r~ of the panels m~mlf~rtl-red using
25 MDI are sukst~nti~lly improved over the recipe that used the PVA/latex
without MDI. The panels produced in this study with MDI exhibit strength
and sti~ properties that are equivalent or better than high density particle
koard of the type used in the m~nllf~rture of commercial and residential
furniture.
The increase in strength and stifl~ due to increasing rates of
MDI will allow the biocomposites to be tailored to desired strength and
stiffil~ss properties for specific end uses. The strength and stiffilrc~ of
WO 95/04779 ~16 8 6 ~ 4 PCT/US94/09090
58
production panels using the addition rates of MDI from this study would tend
to be higher given a slightly higher target specific gravity of 1.3 and less
variability that is associated with a continuous system versus a batch system.
The results of the physical proper~y testing is presented in
5 Table X The panels produced from just the soy resin system c~ g the
PVA/latex had significantly (P < 0.05) higher edge swell and water absorption
than the ~u~ ~lions m~m-f~ red with the MDI resin. There was no
significant (P<0.05) differences in either edge swell or water absorption
~l~tect~l between con~ ions due to increasing MDI addition rates or
10 between the two wax levels. No trends were ~1ete~ as with the strength and
stiffnt-~ Only at the 3.5% addition rate was there any improvement in
dim~l~sional stability due to increasing wax addition from 1.0% to 1.5%.
From the table we can see that the wax has taken the place of the PVA latex
as the sizing agent in the panels without adversely affecting ~limpn~iona
15 stability.
Unlike conventional pressing of wood-based composites where
MDI addition rates can vary up to 20-30%, the panel m~kin~ process of the
present invention typically has an upper limit to the amount of MDI that can
be added without r~ ing adverse effects on the panels coming out on the
20 press. In the process, the target specific gravity is extremely high, 1.3 to 2.0
times higher than conventional m~mlf~*lred wood-based composite panels.
As stated before, this high specific gravity is needed to ensure low porosity ofthe panel surfaces. However, these tight panels leave very little void-space
where the excess MDI can squeeze into. During the pressing operation using
25 particles blended with the high loading rate of 10% MDI all free void space
in the panel was filled and with some panels the excess MDI squeezed out.
By filling all the free void space within the panel there are no areas or
avenues for the steam that is generated in the pressing operation and CO2
formed during the isocyanate curing process to escape. This cause numerous
30 blow-outs in the panels when they come out of the press, similar to what one
would find if the moisture content of the particulate material was too high.
This problem was not encountered during pressing of panels using the
WO 95/04779 ~ 6 8 6S4 PCT/US94/09090
59
particles blended with the 7.5% addition rate. This .eug~s that the upper
limit on MDI loading is about 10%.
Other conclusions that can be drawn from this study inr.lu-le:
the press times using the recipe without PVA, but with the MDI/wax
5 blending systems are s~lbst~nti~lly lower than the press times used m~king
panels with the recipes c( "~ ,;"g PVA. Given the m~mlf~tl rin~ process
used in this exarnple and desired pLu~LLie_, the upper limit of the amount of
MDI that can be blended with the composite particles is about 10%, although
this can be about 20% under other con-liti~-n.e. By va~ying the amount of
10 MDI coated on the dry particles, the m~h~nical and physical p~u~llies of the
r~ellltin~ panels can be ~l~.ej~l~l to meet specific uses. At loading of 5% or
above of MDI, there is no substantial gain in the .~ ional stability when
the amount of wax emulsion used is increased ~om 1.0% to 1.5%.
WO 9~/04779 ~ - 6 8 ~ 6 ~ PCTrUS94/09090
TABIE VIII
Select~slMe~l~r~ ~,lies of p~ r~l---~ vu~ vanous
Co.~ lio~ of ~i~ and S~ ~e~
P~nel Type Specific Modulus of Twenty-Four
Gravity Rupture (psi) Rupture (psi)
LatexC~y n 25 25 25
mean 1.23 2,809 411,513
L~ex/ 5.0% n 23 23 23
~DDI mean 1.33 4,737 547,861
3.5% ~DD~ n 25 25 25
1.0% Wax mean 1.25 3,894 647,595
3.5% ~DD~ n 30 29 30
1.5% Wax mean 1.26 3,871 576,761
5.0% ~DDI n 25 25 25
1.0% Wax mean 1.21 4,033 535,627
5.0% ~DD~ n 20 20 20
1.5% Wax mean 1.25 4,369 645,858
7.5% ~DD~ n 20 20 20
1.0% Wax mean 1.23 4,508 672,700
7.5% ~DD~ n 25 25 25
1.5% Wax mean 1.20 4,208 672,043
10% ~DD~ n 25 25 25
1.0% Wax mean 1.25 4,703 746,482
10% ~DDI/ n 20 20 20
1.5% Wax mean 1.28 4,744 719,268
~6g~4
WO 95/04779 PCTAUS94/09090
61
T~R~,F, 1
Scle~l Pkvsical Properties Qf Panels lV~ r-~l V~fi~ Vanous
- Co.. ~l.. ~l;-n~; of l~in~: and si7i~ ~ents
.,
Panel Type Specific Twenty-Four Twenty-Four
Gravity HourPercent Hour Percent
F~ge Swell Wei~t ~in
Latex Ck~y n 18 18 18
mean 1.25 40.6 30.9
Latex / 5.0% n 24 24 24
~DDI mean 1.28 10.3 9.6
3.5% ~DD~ n 15 15 15
1.0% Wax mean 1.27 15.6 11.9
3.5% ~DD~ n 15 15 15
1.5% Wax mean 1.29 12.6 9.7
5.0% ~DDI n 15 15 15
1.0% Wax mean 1.24 12.4 10.9
5.0% ~DD~ n 15 15 15
1.5% Wax mean 1.27 11.8 9.5
7.5% ~DD~ n 10 10 10
1.0% Wax mean 1.26 9.3 8.5
7.5% ~DD~ n 15 15 15
1.5% Wax mean 1.27 9.3 8.5
10% MDI/ n 11 11 11
1.0% Wax mean 1.32 9.8 8.1
10% ~DD~ n 15 15 15
1.5% Wax mean 1.31 10.4 8.1
WO 95/04779 ~ ~ 6 8 61~ 4 PCTIUS94/09090
62
FYSn11T~ 9
p~nrl~ l\~e F~m Soy Flo~ ~n(l T imP
This study was cnntll-cte~l as a comparison to evaluate the use
of lime and soy flour in the production of a soy resin. The recipe used to
5 m~mlf~r*lre the soy resin was a combination of 8 kg lime, 148 kg water, and
49 kg 90 pdi soy flour, which was combined with 46 kg h~mmrr milled
nullll~led inked nt;w~er. The material was dried to ~roxill~lely 11%
moisture content prior to pressing Platen ten~ re was about 312F
(156C). The total press time was about 13.5 mimltes. Press time varied due
10 to slight fluctuations in the moisture content of the particulate material. The
target specific gravity of the panels coming out of the press was 1.30. The
panels were tested for strength, stiffn~, hardness, and tiimrn~ional stability.
Descriptive statistics were generated for each plu~ ly ev~lu~trA
There were no problems encountered in either m~mlf~rtl~ring of
15 the composite particles of the pressing of the panels. The following is a
sllmm~ry of the mrrh~nical and physical p~ ies of the panels produced in
this study. The values listed are the mean of 15 test specimens.
Modulus of Modulus of 2-Hour
Fl~cticity (psi) Rupture (psi) _dge Swell (/O)
217,559 1,080 16.82
The panels had considerably less streng~ and stiffnP-~ than
panels made from colll~osiLe particles m~nnf~r*lred using the typical soy
resin recipe. The (limrn~ional stability of the panels produced in this study
were extremely poor. The test samples ~ inte~rated in the water during
testing. Therefore, 2~hour edge-swell data was lm;~ 1e. This simple
30 recipe of lime, soy flour, and water produced particles that lack both internal
bonding capabilities as well as particle to particle bonding capabilities.
WO 95/04779 .~ . PCT/US94/09090
63
~ e 10
lion of Bioco~ ial U~ PeaResin
The resin-based biocomposite particles of the present invention
uses protein-derived systems. Although the most pl._rcllcd protein is derived
5 from soy flour, other legumes can be used, in~ ling a wide variety of beans
and peas. Exemplary of this is pea flour, which was ground in a h~mm~r mill
from dried yellow split peas to a c--n~i~t~n~y finer than 325 mesh.
Particulate material was made using pea flour in place of soy
flour in the recipe detailed in Process B (cf)~ g PVA) reduced to a
10 laboratory scale. Four panels were pressed in a 2 ft. by 2 ft. hot press.
Platen temperature was 312F (156C) and a total press time of approxirnately
13.5 mimlt~. Press times varied slightly to compensate for cll~n~e~ in
particulate material moisture content. Target specific gravity coming out of
the press was 1.3. The panels were tested for strength, stiffn~ and
15 ~ onal stability. Descriptive statistics were gen~l~led for each property
tested.
There were no noticeable diff1culties encountered in using the
pea flour in substitution of the soy flour during mixing of the composite
particles. However, after the drier the particles had a noticeably lower bulk
20 density than particles m~mlf~lred with the soy flour. The low bulk density
of the composite particles caused some difficulty in pressing, and resulted in agreater degree of variation in density along and across the panels than when
so~ flour was used. This was most likely due to the fine grind of the pea
flour. This problem could be corrected by adjusting the grinding par~m~
25 to produce the same mesh size flour as the currently used soy flour.
The following s~ n~;s the m~h~nical properties obtained
from the panels produced in this cx~c~ ent. The numbers listed represent
the mean of 15 test specimens.
WO 95/04779 2~ 1 r PCT/US94/09090 --
64
Modulus of Rupture (psi) Modulus of Elasticity (psi)
2,787 424,951
The strength and stiffn~ee of the panels produced with pea flour
were very similar to panels produced with soy flour. Typical strength and
stiffn~Aes of panels produced using the recipe detailed in Process B with soy
10 flour is 2,809 psi and stiffnPee is 411,513 psi (see Example 7).
The following Sl~ l~ the physical pro~llies obtained
from the panels produced with pea flour. The values listed are the mean of 8
test speCiml?ne
2-Hour 24-Hour
Edge Swell (/O) Edge Swell (/O)
24.5 54.2
The edge swell values are slightly higher than panels produced
under the same conditions using soy flour. Typical twenty-four hour edge
swell is 40.6% (see Example 7). This increase in edge swell values is
believed to be due to the high variability of the panel density. Wlth a more
25 con~i.et~nt density around the target of 1.3, the edge swell values of the panels
made with pea flour should be very col~ 1e to panels produced using soy
flour, and are clearly suitable for interior use applications.
Test 1\~
The following test methods were used to obtain the above-listed
30 results Typically, the samples were tested at their target in-service moisture
content of ~ l~tely 7%. Test specimens were removed at random from
the lot of material tested.
Static ~en~ling (MOR ~ MOF~
The strength, i.e., modulus of rupture ("MOR"), and stiffness,
wo 95/04779 2 1 ~ ~ 6 ~ ~ PCT/US94/09090
i.e., modulus of elasticity ("MOE"), were evaluated using the static bending
test detailed in ASTM:D1037-91, Sections 11-20. A number of 3-inch x 10-
inch test specimens were removed from each test panel, and placed in a
universal testing ~ i"~ A span of 8 inches and a constant cross head speed
5 was applied to the test specimens until failure. Load vs. deflection data was
recorded and the MOR and MOE were calculated for each specimen and
averaged for each condition tested.
Hal I 1, Il'~x
The hardness was (l~tP~min~A using the method detailed in
ASTM:D1037-91, Sections 7~80. A number of 3-inch x 6-inch test
specimens were removed from each test panel, and placed in a universal
testing m~ ine using a~lupliate fixtures. A modified Janka Ball (0.4375-
inch rli~m~ter) wæ penetrated into the specimen at a ~l~L~l~ rate to a
15 deflection of 0.10 inch. The load vs. deflection data was recorded and the
equivalent Janka Ball hardness values were calculated and averaged for each
condition being tested.
Water Absorption ~ntl Thi~kn~ Swellin~
The water absorption and thickness swelling test was conf1ucte 1
using the method detailed in ASTM D1037-91, Sections 100-106. A number
of ~inch x 6-inch test specimens were measured for original thickness and
weight and then sul~me,ged ho,i~ lly under 1 inch of water. Changes in
weight and thickness were measured after 2 hours and 24 hours. The changes
in thickness and weight were calculated and averaged for each condition being
tested.
JntP~ Bond
The int~l bond testing was conducted using the test method
detailed in ASTM: D1037-91, Sections 28-33. A number of 2-inch by 2-inch
test speciments were removed from the test panels and glued to al~rol~,iate
test fixtures. The glued-up sample was placed in a universal test m~ ine and
WO 95/04779 PCT/US94/09090 ~
~ 6~6~
66
a c )n~t~nt load applied until failure. Maximum obtained load was recorded
and corresponding internal bond values calculated.
Mo;~ re Content and Den~i~/Specific Gravity
S The moisture content and specific gravity were cletPrmined
using the method detailed in ASTM D1037, Sections 126-127. Specimens
were obtained from the various test samples for static ben-ling The
specimens were weighed and measured at their in-service moisture content of
approximately 7% for density. These spP~im~n~ were then oven-dried at
103C to practical equilibrium, weighed, and compared to the initial weight to
cl~l~ .",;1l~ actual moisture content.
Viscosity
Viscosity of various soy adhesive mixtures was measured using
15 a Brookfield model DV-I+ viscometer. A number of 100-ml beaker samples
of the soy adhesive in its form prior to the addition of cellulose was tested at60-70F (16-21C) using a number 6 spindle. The rçslllting c~llLi~ise
readings were recorded and averaged for each condition calc -l~te~l
Although the invention has been described with respect to
20 various presently preferred mixtures and production equipment, it is to be
appreciated that still other methods and mixtures may be suggested to those
skilled in the art. Accordingly, it is cnntPrnplated that the foregoing
description should be interpreted to include all those equivalent embo lim~nt~
within the spirit and scope thereof.
