Fire-retardant wood and wood composites, and compositions useful for making same

Abstract

An amino-functionalized polyglycerol phosphate flame retardant can be synthesized through polycondensation of glycerol and phosphoric acid, followed by crosslinking with one or more amino functionalized compounds such as ethylene diamine or polyethyleneimine. The formulation is optionally enhanced with halloysite nanotubes or other clay minerals, which not only act as direct flame retardants but can also serve as carriers for flame-retardant compounds, aiding in controlling the migration of the flame retardants, and enhancing their long-term effectiveness. The flame-retardant is preferably combined with a water-resistant glue, allowing its application as a durable coating on wood substrates. The coating provides fire, water, and biological resistance.

Description

FIRE-RETARDANT WOOD AND WOOD COMPOSITES,AND COMPOSITIONS USEFUL FOR MAKING SAMEGOVERNMENT SUPPORTThis invention was made with government support under grant 22-DG-l 1083150-201 awarded by the United States Department of Agriculture / Forest Service. The United States Government has certain rights in the invention.PRIORITY CLAIM

[0001] The benefit of the 02 February 2024 filing date of United States provisional patent application serial number 63 / 549,183 is claimed under 35 U.S.C. § 119(e) in the United States, and is claimed under applicable treaties and conventions in all countries.TECHNICAL FIELD

[0002] This invention pertains to enhanced fire resistance in wood and wood-based materials through treatment with compositions comprising amine-modified polyglycerol phosphate compounds.BACKGROUND ART

[0003] Wood treatments for enhanced fire resistance have a long history. However, there have been no major breakthroughs for decades. The two most common prior approaches have been water-soluble penetrating formulations, and intumescent coatings. Penetrating fire retardant (FR) formulations have typically comprised primarily ammonium phosphate, along with boric acid or another simple form of borate.

[0004] Phosphate-based FRs cause treated combustible organic materials to char at a lower temperature than adjacent, untreated materials (e.g., by sacrificing the top layer of a substrate). The charring process can help decrease the evolution of flammable gases, and to reduce the heat that is released. The resulting char also helps to insulate the remainder of the material.

[0005] Intumescent FR coatings function by rapidly expanding in response to heat. Intumescent coatings generally comprise three main components: char formers, charringcatalysts, and blowing agents. When an intumescent coating is exposed to heat, the charring catalyst promotes char formation. At the same time, the blowing agent releases non-flammable gases, which expand the soft char and form a fine-textured carbon char foam that insulates the substrate.

[0006] Engineered wood products (EWPs) are composites in which various materials adhere to one another, such as wood veneers, wood particles, wood fibers, and wood strands. EWPs can be flexible; they are available in a wide range of thicknesses and sizes; and they can readily be cut, drilled, routed, jointed, glued, and secured. One type of EWP, wood oriented strand board (OSB), is expected to experience a compound annual growth rate of 5.6 percent worldwide, reaching a valuation of around $15 billion by 2027. Applications for OSB include flooring, sheathing, web material for I-joists, structural insulated panels, industrial containers, furniture, and many other uses. The sustainability of and further market expansion for OSB depend on the development of improved green technologies (e.g., formaldehyde-free adhesives), and newer product lines (e.g., fire-resistant OSB). Fire retardant-treated oriented strand board (FRTOSB) could potentially replace more expensive FR-treated plywood for firerated applications, such as structural sheathing on exterior and interior walls, pitched roofs, flat roof decks, and floors.

[0007] US Pat. No. 6713168 discloses a fire retardant oriented strand board composite material comprising a mixture of wood strands, an organophosphorus ester, a polymeric binder resin, and a wax.

[0008] US Pat. No. 7371787 discloses composites containing, inter alia, a fire retardant known in the art, such as for example a phosphate / borate composite.

[0009] US Pat. Appl. Publ. No. 2015 / 0020476 discloses fire resistant coatings for wood products, comprising an aromatic isocyanate, castor oil, intumescent particles, and a fire retardant. The fire retardant is selected from the group consisting of disodium ocataborate tetrahydrate, Colemanite, Ulexite, Aluminum trihydrate, Magnesium hydroxide, Hydromagnesite, and Hunitite.

[0010] US Pat. Appl. Publ. No. 2022 / 0177638 discloses flame-retardant polyurethane or polyurethane / polyisocyanurate foam materials containing an isocyanate-reactive component, a propellant, a catalyst, optionally an additive, and a flame retardant with an isocyanate component and (hydroxymethyl)phosphonate and optionally the dimer thereof.

[0011] US Pat. Appl. Publ. No. 2021 / 0399360 discloses a composite material made of a porous matrix material and a filling material, the filling material being inserted in pores of the matrix material, the filling material being a phase change material, the phase change materialhaving a melting temperature in a range of 15° C to 200° C, wherein the matrix material contains graphite with intercalated acid.

[0012] J. Pretula et al., J. Poly. Sci., Part A: Poly. Chem. 2014, 52, 3533-3542 discloses polyglycerol phosphates prepared by polycondensation of glycerol and phosphoric acid.

[0013] Unfortunately, most penetrating FRs do not work well with engineered wood products (EWPs) such as OSB. Most such materials are not fully compliant with modem building codes. Many are cost-prohibitive, especially for large scale use. Treating wood strands prior to manufacturing OSB can adversely affect glue strength, leading to long-term durability issues and liability issues. Post-treating of manufactured OSB with water-soluble penetrating formulations leads to excess swelling of OSB and strength reduction, and is not a good solution to the problem.

[0014] There remains an unfilled need for more cost-effective and environmentally - friendly compositions and methods to produce FR treated wood and wood materials, including OSB and other Engineered Wood Products.SUMMARY OF THE INVENTION

[0015] We have discovered novel amino-functionalized polyglycerol phosphate (APGP) compounds that are useful as flame retardants for coating wood and wood products. The novel FR coating may be synthesized, for example, by polycondensation of glycerol and phosphoric acid, followed by crosslinking with amino-functionalized compounds such as ethylenediamine (EEI) or polyethyleneimine (PEI). Halloysite nanotubes (HNT), bentonite, or other clays such as kaolinite or montmorillonite (MMT) can optionally be added to the formulation — both as a flame retardant, and also as a carrier for other flame-retardant compounds to impede the migration of the flame-retardant molecules (for example, encapsulated inside HNT lumens) to reduce the effects of long-term aging of the FR products. Lumens in certain clays, such as halloysite, can help encapsulate payload compounds. Nanoplatelet structures in other clays, such as bentonite, kaolinite, or montmorillonite, can help reduce heat transfer into the interior of the wood or wood composite. The fire retardant may optionally be combined with a glue, such as a water-resistant aliphatic glue, which can be cured for bonding the fire-retardant formulations to the wood substrate surface. The FR treatment can be used as a surface coating for wood materials and wood composite materials not only to provide fire resistance, but also resistance to water and biological organisms. In many embodiments, wood treated with the novel FR treatment demonstrates improved mechanical properties, such as enhanced flexural strength or modulus, due to the reinforcing effects of the FR additives and their interaction withthe wood matrix. In some cases, however, the mechanical properties may decline slightly, particularly at higher concentrations of fillers (e.g., bentonite or other clays). Even in the latter case, the significant increase in fire resistance from the novel FR treatment will often more than compensate for a slight decrease in mechanical properties. Serendipitously, wood that has been treated for enhanced fire resistance in accordance with the present invention also has greater resistance than untreated wood to wood-destroying insects such as termites, at least in many instances.

[0016] In the novel formulation, the phosphate compound acts as a char former. The charring process plays an important role, both in reducing the evolution of flammable gases, and in reducing the overall heat released. The char helps insulate the rest of the material, decreases the temperature, and leads to lower release of flammable gases. The charring process also releases water vapor and CO2 gas, which also help to inhibit combustion.

[0017] Glycerol is a triol with molecular formula CsHsCh. Each of the three carbon atoms in glycerol is bonded to a separate hydroxyl ( — OH) group. When glycerol reacts with phosphoric acid in the presence of a catalyst, a high molecular weight polyglycerol phosphate (PGP) is produced by polycondensation. The hydroxyl groups from the glycerol react with acidic hydrogen atoms from phosphoric acid, forming ester bonds. The resulting polyglycerol phosphate is characterized by a networked structure, in which glycerol units are interconnected through ester bonding with phosphate groups to form a complex, stable molecular structure. The molar ratio of glycerol to phosphoric acid in polyglycerol phosphate (PGP) is preferably from 0.25 to 2.

[0018] Polyethyleneimine (PEI) is a polymer with multiple amine groups (-NH2). The amine groups are highly reactive. The amine groups can displace some of the unreacted hydroxyl groups from the glycerol units, leading to the formation of secondary amine linkages; or they can react with unreacted acidic hydrogens from the phosphate groups within the polyglycerol phosphate. Amine-phosphate crosslinking enhances structural integrity within the polymer structure, and also introduces nitrogen into the composition, which is beneficial for flame retardation. The nitrogen atoms can release non-flammable gases when heated, particularly nitrogen and ammonia; and can also promote the release of carbon dioxide and water vapor; as well as various mixtures of the above. Phosphorus can promote char formation. Both effects help to reduce flammability. In addition to its role in flame retardance, PEI also has antimicrobial properties. The amine groups in PEI interact with microbial cell membranes, disrupting their function, and leading to inhibition of or death of many microorganisms.

[0019] Halloysite is a naturally-occurring aluminosilicate nanotube mineral. Halloysite (A12Si2O5(OH)4*2H2O) is a two-layered aluminosilicate, having a predominantly hollowtubular structure with lumen dimensions in the submicron range. Halloysite is chemically similar to kaolin. The neighboring alumina and silica layers, along with their waters of hydration, curve and form multilayer tubes due to a packing disorder. Halloysite is mined as a raw mineral commercially. Halloysite particles typically vary between 1-15 microns in length, with lumens 10-150 nm in inner diameter, depending on the particular deposit from which the halloysite is mined. The nanoscale structure of halloysite nanotubes (HNTs) optionally helps to improve mechanical properties of the formulation, to reduce gas permeability, to increase thermal stability, and to improve flame retardancy. HNTs also help mitigate the migration of other flame-retardant molecules encapsulated in the HNT lumen, improving long-term aging properties.

[0020] A water-resistant glue such as an aliphatic glue — e.g., a cross-linked polyvinyl acetate, (PVAc) — acts as an optional bonding agent for the fire retardant coating onto wood substrates. PVAc is a nontoxic, thermoplastic adhesive typically prepared by polymerization of vinyl acetate. PVAc is commonly prepared by aqueous emulsion polymerization. The adhesive sets upon diffusion of water into the wood, followed by in situ coalescence of the PVAc. Other adhesives known in the art can also be used, for example, aromatic adhesives, epoxy resins, or polyurethanes.

[0021] The formed APGP with the optional bonding agent is spread onto a wood surface, where it is preferably cured with heat to form a coating on the wood surface, which for many applications is often a uniform coating. Phosphate-based FRs function by promoting the charring of combustible, treated organic materials at a lower temperature than the adjacent untreated materials — i.e., by sacrificing a top layer of substrate. The charring process helps to decrease the evolution of flammable gases, and to reduce the heat released. The resulting char also helps insulate the underlying material.BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 depicts a synthesis of a polyglycerol phosphate from phosphoric acid and glycerol.

[0023] Figure 2 depicts a synthesis of an amino-functionalized polyglycerol phosphate from polyglycerol phosphate and polyethyleneimine.

[0024] Figure 3 depicts the preparation of a dry, powdered form of an amino-functionalized polyglycerol phosphate.MODES FOR CARRYING OUT THE INVENTIONExample 1: Synthesis of an amino-functionalized polyglycerol phosphate (APGP) flame retardant

[0025] Phosphoric acid (PA, 85.0% in water), N-butyl alcohol (n-BA), glycerol (GL), and Tin (II) chloride (SnCh) were purchased from Sigma Aldrich (Burlington, MA, USA). PVAc- based wood glue was purchased from a local store. Halloysite nanotubes (HNTs) were supplied by Sigma Aldrich (Burlington, MA, USA) with an average tube outer diameter of 50 nm, and average lumen inner diameter of 15 nm. The typical specific surface area of the halloysite was 65 m2 / g.

[0026] To prepare the polyglycerol phosphate (PGP), glycerol (GL) and phosphoric acid (PA) were combined in a molar ratio of 0.5: 1 or 1 : 1. See Figure 1. Tin (II) chloride (SnCh) was used as a catalyst, at 0.1% by mass of the total reactants. (In other embodiments, the SnCh can be from 0.01% to 1% by mass.) The reaction was carried out with vigorous stirring under reflux at 100°C in a flask with an azeotropic receiver. The pressure was gradually decreased to remove any water produced as a condensation by-product. The reaction continued for 4 hrs. to the desired degree of polymerization. The reaction mixture containing PGP was then cooled to room temperature to form a more viscous material.

[0027] The synthesized polyglycerol phosphate (PGP) was placed in a 250 ml roundbottom flask equipped with a magnetic stirrer. The PGP was heated to 95°C under reflux, after which polyethyleneimine (PEI), approximately 5% by weight of the PGP, was carefully added to the mixture with continuous stirring. The mixture was then allowed to react for 4 hours, with temperature maintained at 95°C with constant stirring; followed by cooling to room temperature. The resulting product, an amino-functionalized polyglycerol phosphate (APGP), was then collected as a viscous solution for subsequent use. See Figure 2.Example 2: Forming an APGP-HNT FR Coating

[0028] Various concentrations of halloysite nanotubes (HNT), specifically 0.0%, 2.5%, 5%, and 10% by mass, were added to the previously-prepared, amino-functionalized polyglycerol phosphate (APGP). The mixture was sonicated for 10 minutes until a homogeneous suspension had formed.Example 3: Characterization of the APGP FR Coating

[0029] Fourier Transform Infrared Spectroscopy (FTIR) was used to study the PGP andAPGP materials using a Bruker FTIR analyzer (Tensor-27, Bruker Optics Inc., Billerica, MA).Spectra were taken from 4000 to 600 cm1at a resolution of 4 cm for a total of 32 scans. Broad bands around 3500 cm1were attributed to the stretching vibration of -OH groups; the intensity of this peak decreased in the APGP due to the replacement of OH groups by PEI. A peak at 2319 cm1was assigned to P-OH groups. The appearance of new peaks at 1518 cm1in the APGP indicated amide linkages, specifically C-N stretching vibrations in secondary amides, which confirmed the crosslinking reaction with PEI. A peak around 1100 cm1for PGP and APGP was assigned to P=O stretching vibrations in phosphate groups.

[0030] NMR spectroscopy was performed with a Bruker Avance 400 instrument (Karlsruhe, Germany). The 'H NMR and31P NMR spectra of PGP and APGP revealed significant chemical shifts in polyglycerol phosphate (PGP) and its amino-functionalized derivative (APGP). In the 'H NMR spectra, peaks in the range 3.20-3.50 ppm were attributed to hydrogen atoms in P-OH groups, while those at 2.86-2.96 ppm corresponded to exocyclic hydrogen atoms linked to the cyclic structure of PGP. A notable shift at -0.44 ppm in the31P NMR spectra, indicating phosphate groups forming C-O-P bonds, suggested alterations from crosslinking with PEI, perhaps further influenced by NH4+cations in the structure.

[0031] Thermogravimetric analyses (TGA) were conducted with a Q50 Analyzer (TA Instruments Inc., New Castle, DE) under a nitrogen atmosphere. The temperature ranged from 30 to 600 °C, at a heating rate of 10 °C / min, using a 5 mg sample. The effect of HNT content on APGP was observed. For APGP itself, there are three thermal degradation processes, occurring around 50-160°C, around 160-450°C, and around 450-600°C, respectively; the second process (160-450°C) is the dominant one. The first stage at 50-160°C was ascribed to the release of small molecules from thermal decomposition of amino groups, and was accompanied by a small mass loss. When the temperature increased to 160-450°C, APGP decomposed to phosphoric acid derivatives that promoted the formation of molten char. The third stage at 450-600°C was ascribed to degradation of the char layer at high temperature. The residue from APGP after heating to 600°C retained about 35% of the initial mass.

[0032] When HNT was introduced (at levels of 0, 2.5, 5, and 10 % total mass), the thermal stability increased. The residue remaining after heating to 600°C then increased to 42, 45, 51, and 53 %, respectively.

[0033] The thermal behavior of pristine wood and FR-coated wood is shown in Table 1. The nomenclature “FRn” in Table 1 refers to a APGP fire retardant layer having n% HNT. For example, FR2.5-Wood is wood coated with an APGP formulation having 2.5% HNT by mass.

[0034] In the early stages of thermal decomposition, both uncoated and FR-coated wood lost weight as the temperature increased. The initial 5% weight loss for all samples was attributed primarily to moisture evaporation, from the loss of both free and minimally-boundwater molecules. In the FR-coated samples, particularly as HNT content increased, the first notable weight loss peaks, those characteristic for pristine wood, declined so much that they were nearly absent. As the temperature increased further, a second significant weight loss event occurred around 300°C for pristine wood, and for the FR-coated wood around 277.8°C. This loss was attributed primarily to the breakdown of cellulose and lignin. The third weight loss peaks began above a temperature of 400°C: For the pristine wood, TPW3 and PW3 were found to be 409 °C and 0.77 % / °C. But for the FR-coated wood, especially with increasing HNT content, the third weight loss peak occurred at higher temperatures due to the higher thermal stability afforded by the FR-HNT coating. For FRIO-Wood, for example, the third peak was observed at 570°C, at a rate 21% lower than that for pristine wood.Table 1. TG test results: weight loss for various treated and untreated materials.

[0035] Differential scanning calorimetry (DSC) was carried out with a DSC QI 00 instrument from TA Instruments, USA. Around 10 mg of sample was placed in a DSC cell. Each sample was heated from -80 to 250 °C at a heating rate of 10 °C min ' . The DSC data showed the thermal behavior of FR-coated OSB samples, comparing their performance before and after exposure to fire. For the FRO-wood, a pronounced exothermic peak was observed around 251 °C, which is characteristic of wood decomposition. As the HNT content increased in the FR coatings, the exothermic peak shifted to higher temperatures (265°C and 266°C), indicating delayed onset of thermal degradation. This shift demonstrated the effectiveness of the FR coating in enhancing thermal stability of the OSB samples. After combustion, DSC data for the char residue for all samples exhibited a smoother heat flow profile, indicating the formation of a stable protective char layer. The FR-treated samples remained thermally stable post-burning, confirming the durability of the char layer and its contribution to the fireretarding properties for the OSB.Example 4: Preparation of FR coating on Wood OSB

[0036] Oriented Strand Board (OSB) sheathing materials were purchased from a local supplier. The materials were cut into small specimens, suitable for various testing protocols, including assessments of surface morphology, Limiting Oxygen Index (LOI), cone calorimetry, direct flaming, and combustion tests on scale-model structures. Each sample was carefully coated with one of the flame-retardant compositions. The coating was applied using a brushing technique, ensuring a consistent and controlled rate of coating to achieve uniform coverage and adherence of the fire-retardant formulation to the OSB surfaces.Example 5: Characterization of FR Coated OSB

[0037] Scanning electron microscopy (SEM) was used to study the coated OSB samples, using a Quanta™ 3D DualBeam™ FEG-FIB-SEM with an accelerating voltage of 20 kV. All samples were deposited on a carbon ribbon on an aluminum rod prior to observation, and then sputter-coated with platinum for 2 min. The SEM observations highlighted significant differences in surface morphology between the pristine OSB wood and the FR-coated wood. The pristine OSB wood displayed its inherent fibrous structure, characterized by distinct vascular channels and pores. These features, essential for the wood's biological functions, also contribute to its vulnerability to ignition and combustion. By contrast, the FR-coated wood showed substantial alteration in surface morphology. The flame-retardant coating formed a consistent layer over the wood surface, obscuring the underlying cellular structures. A uniform coating layer is believed to assist in protecting against fire. Among other things, the coating impedes direct interaction between the wood on the one hand; and oxygen and heat on the other hand. Furthermore, the analysis suggested that the coating had penetrated the wood’s porous structures, indicating strong adhesion and suggesting that the flame-retardant properties might also extend underneath the surface for enhanced protection during prolonged fire exposure.

[0038] A Dynisco Limiting Oxygen Index (LOI) chamber was used to test the combustion behavior of several FR-coated OSB samples, following the protocols of ASTM D2863. The samples used for these tests were each 80 mm x 10 mm x 4 mm. Three samples were tested for each group. Prior to testing, the specimens were conditioned at 23°C and 50% relative humidity for 24 h. The Limiting Oxygen Index (LOI) data in Table 2 demonstrate the superior flame resistance of the flame retardant-coated wood as compared to that for pristine wood. The FR-coated wood showed a remarkable LOI value of 40%, significantly higher than the 25% value for pristine wood. This substantial increase in LOI characterizes the FR-coated wood as a slow-burning material. In burning tests, for pristine wood a flame starting at one end of a sample reached the 5-cm mark after about 43-47 seconds, while the FRO-Wood displayedexceptional resistance, never burning as far as the 5-cm line even at a high oxygen concentration of 40%, underscoring the effectiveness of the flame-retardant treatment.Table 2: Limiting oxygen index (LOI) and burning time for pristine wood and FR- coated wood.Sample 1 2 3 4 5 Test24 25 25 25 25 Oxygen index %Pristine wood-b 47 37 43 47 Burn time(c)30 34 36 38 40 Oxygen index %FRO-wood-b -b -b -b -b Burn time(c)30 35 40 42 - Oxygen index %FR2.5-wood-b -b -b -b - Burn time(c)30 35 40 42 - Oxygen index %FR5-wood-b -b -b -b - Burn time(c)30 35 40 42 Oxygen index %FRIO-wood-b -b -b -b - Burn time(c)bdid not burn to 5-cm line.ctime to bum to 5 cm.

[0039] The flammability of pristine OSB wood and of the FR-coated wood samples was tested in a direct flame test. The pristine OSB wood ignited within 30 ± 3 seconds, and the flame continued with a long flame length after the igniter had been removed. The flame retardant-coated wood, on the other hand, did not start to ignite until 480 ± 5 seconds, and the flame was extinguished within 60 seconds after the igniter had been removed. The test results showed that the FR-coated wood remained mostly intact, with the formation of a layer of char on its surface. The char acted as a barrier to further ignition, forming a layer that separated the wood from the air. Lacking adequate oxygen, the wood underneath the separated layer was unable to smolder. Additionally, the char layer also helped to reduce the burning process, further inhibiting ignition.

[0040] A cone calorimeter (Model 82121, Fire Testing Technology, West Sussex, UK) was used to measure the heat release rate (HRR) and total heat release (THR) of the coatings according to IS05660-2002, with a sample size of 100 mm x 100 mm x 4 mm. Table 3 shows cone calorimeter data for various wood samples, including pristine and flame retardant-coatedwood, showing significant improvements in fire resistance for the novel treatments. The time to ignition (TTI) for the coated wood significantly increased: Pristine wood ignited at 40.03 seconds, and FRIO-Wood at 93.57 seconds. The Peak Heat Release Rate (PHRR) decreased notably for the coated samples: Pristine wood had a PHHR of 301 ± 10 kW / m2, while FR10- Wood had a substantially lower rate of 200 ± 8 kW / m2. The reduced heat release correlated with slower, less intense burning, especially in samples having higher HNT content. The Total Heat Release (THR) varied across samples, but generally showed a decreasing trend for the treated wood samples. The Total Smoke Release (TSR) increased, due to changes in the combustion process from the flame retardants. The increased residue weight for the treated wood samples, e. g., 26.11% for FRIO-Wood, indicated more effective char formation, enhancing protection against fire. The Maximum Average Rate of Heat Emission (MARHE), a parameter that predicts large-scale combustion behavior, decreased from 148.7 kW / m2for pristine wood to 129.8 kW / m2for FRIO, indicating that the FR treatments effectively slowed combustion, reduced heat release, and improved fire resistance.Table 3Sample TTI (s) PHRR THR TSR MARHE residue(kW / m2) (MJ / m2) (m2 / m2) (kW / m2) (wt %)Pristine wood 40.03 301 ± 10 81 ± 2.0 723.1 148.7 18.4FRO-Wood 89.01 284 ± 12 85 ± 1.0 1153.8 142.7 23.7FR2.5-Wood 99.40 282 ± 9 79 ± 1.5 1032.0 141.3 24.5FR5-Wood 95.02 264 ± 7 83 ± 2.5 1089.0 141.4 25.1FRIO-Wood 93.57 200 ± 8 74 ± 1.0 915.4 129.8 26.11

[0041] Small model houses are made from both uncoated and APGP-coated OSB. These models are then subjected to direct flames to evaluate the overall effectiveness of the APGP treatment in a simulated construction scenario.

[0042] We examined the morphology of char residues by SEM-EDS analysis following a combustion test. The images revealed clear differences between the char of pristine wood and that of FR-coated wood. The pristine wood char was characterized by a cracked and porous structure, which is disadvantageous in a fire as it facilitates the penetration and propagation of heat and smoke, potentially exacerbating fire intensity and spread. By contrast, the FR-coated wood’s char layer was considerably more compact and uniform, attributes that are beneficial in mitigating fire damage. The dense char layer acts as an effective thermal barrier, slowing the transfer of heat into unburned wood, and obstructing the passage of smoke and volatileorganic compounds that fuel combustion. A continuous char layer is beneficial, as a continuous char layer impedes the formation of paths through which flames can travel and attack the material's core. EDS maps complemented these observations, demonstrating a homogeneous distribution of phosphorus (P) and nitrogen (N) within the char layer. These elements presumably originated from the APGP constituents in the FR treatment, which contribute to char formation and stability.

[0043] Termite resistance properties of control OSB and FR OSB were measured according to the AWPA El standard. Table 4 summarizes termite mortality and weight loss for various oriented strand board (OSB) samples, including a control sample, FRO-Wood, and FRIO-Wood. There was very substantial termite mortality (100%) for both the FRO- and FRIO coated wood samples. By comparison, the untreated control sample showed only 5% termite mortality. Additionally, the weight loss due to termite activity was notably lower in FRO-Wood (6.64%) and FRIO-Wood (11.31%) as compared to the control (22.05%). The novel flame retardant not only enhances fire resistance but also, as an unexpected but highly desirable property, also synergistically contributes to termite resistance. Without wishing to be bound by this hypothesis, the termite resistance is perhaps attributable to the polyethyleneimine (PEI) component of the FR coating; PEI possesses antimicrobial and antibacterial properties.Table 4 Efficacy of Flame-Retardant treatments in OSB on termite mortality and weight LossSamples Termite mortality (%) Wood weight loss %Control 5 22.05FRO-Wood 100 6.64FRIO-Wood 100 11.31Example 6: Preparation and Testing of a Powdered form of an APGP

[0044] In some embodiments, it is convenient to prepare the APGP in the form of a dry powder. The powder can be dissolved in water to form a liquid FR-treatment solution. The powder can also be used in combination with other powder materials, such as for example various clays or resins (e.g., powdered phenolic formaldehyde, or polymeric methylene diphenyl diisocyanate). A powdered material can be added to wood composite formulations for making wood-based composites such as OSB. A powder can also be used in water-based paints or other materials for enhanced flame resistance.

[0045] An APGP flame retardant was synthesized through a two-step process. In the first step, glycerol (GL) and phosphoric acid (PA) were mixed in molar ratios of 0.5: 1 or 1 : 1, with 0.1% tin (II) chloride (SnCE) added as a catalyst. The mixture was heated under reflux to 100°C for 4 hours, during which time the pressure was gradually reduced, to remove water as the PGP formed. In the second step, the PGP was heated to 95°C, and PEI was introduced at 5% by weight. The mixture was stirred continuously for 2 hours for proper functionalization. After cooling, the APGP was obtained as a viscous liquid. Finally, ammonium hydroxide (NFEOH) was added to reach pH 6 to neutralize acidic phosphate groups, resulting in stable ammonium phosphate salts. The APGP product was purified by extraction with anhydrous ethanol to remove impurities and was then subsequently dried, yielding a white solid.

[0046] FTIR spectra confirmed the functional groups present in the synthesized flame retardant. A broad peak at 3200-3050 cm1was attributed to O-H and N-H stretching vibrations, indicating the presence of hydroxyl and amino groups. A peak at about 2800 cm1corresponded to C-H stretching vibrations from the aliphatic chains in the polyglycerol backbone. A peak at 1200 cm1was assigned to the P=O bond of the phosphate group, while peaks at 1049 cm1and 949 cm1corresponded to P-O-C and P-O-H stretching, respectively. Infrared spectra confirmed the successful incorporation of phosphate and amino groups onto the polyglycerol framework, verifying the intended chemical structure of the synthesized flame retardant.

[0047] The1H NMR spectrum provided further confirmation of the APGP chemical structure. Signals in the range 5 3.0-3.8 ppm corresponded to the CH and CH2 groups of the polyglycerol backbone, indicating the presence of the glycerol-derived structure. The methylene (CH2) and methine (CH) protons indicated successful functionalization with hydroxyl and amino groups. A distinct D2O solvent peak was observed at 5 ~4.6 ppm, typical for NMR spectra in deuterium oxide (heavy water). The1H NMR results were consistent with the FTIR Results, further confirming the successful synthesis of APGP having the intended chemical structure. The31P NMR spectrum confirmed successful incorporation of phosphate groups onto the polyglycerol backbone. The spectrum revealed two distinct peaks: the P2 phosphorus atom from the phosphine oxide group appeared at 0.8 ppm, while the Pl phosphorus atom from the O=P-ONH4 group was observed at 0.1 ppm. These well-defined chemical shifts provided clear evidence of phosphate functionalization, and the formation of ammonium phosphate groups in the synthesized amino-functionalized polyglycerol phosphate flame retardant.

[0048] Several 2 wt% flame retardant solutions were prepared, incorporating BT nanoparticles at 0, 1%, 2%, and 3% by weight, designated F, FIB, F2B, and F3B, respectively.Each mixture was ultrasonicated for 10 minutes to allow the homogeneous dispersion of BT nanoparticles within the FR solution. The amino groups in APGP facilitate strong electrostatic attractions with BT nanoparticles and hydrogen bonding with the negatively charged surfaces of the BT nanoparticles, significantly improving the dispersion stability of the suspension.Example 7: Vacuum-pressure impregnation process

[0049] Flame-retardant wood composites were fabricated by impregnating wood with a FR / BT nanoparticle suspension. Natural wood (NW) samples measuring 80 mm x 10 mm x 4 mm were placed in an impregnation chamber and subjected to a vacuum of -0.098 MPa for 30 minutes to remove air trapped within the wood’s macropores. Next the FR / BT suspension was introduced into the chamber under vacuum, and the system was then pressurized to 60 psi for one hour to allow deep penetration of the solution into the wood structure. The pressure was then gradually released, and the samples were removed from the chamber. The samples were then thoroughly washed to remove any residual solution on the surface, and weighed to determine the change in their weight from the impregnation process. The wood samples were designated as NW, NW-F, NW-F IB, and NW-F2B for natural wood (control) and treated wood with 2 wt% flame retardant and BT nanoparticles at 0, 1%, 2%, and 3% by weight, respectively.

[0050] Scanning electron microscopy (SEM) of cross-sections of (untreated) natural wood samples clearly revealed the natural cellular structure of the wood, characterized by well-defined, open lumens and intact cell walls. Following impregnation with the 2 wt% FR solution, SEM images indicated that the cell lumens were partially filled, presumably from deposition of the FR material within the porous structure of the wood. For the FR / bentonite- treated wood samples, the SEM images demonstrated greater filling of the cell lumens as compared to wood treated with FR alone. The bentonite particles, identified by their distinct morphology, were seen dispersed throughout the wood matrix. Energy-dispersive X-ray spectroscopy (EDS) analysis of the FR / bentonite-treated wood revealed the presence of carbon (C), oxygen (O), silicon (Si), calcium (Ca), aluminum (Al), nitrogen (N), and phosphorus (P), which are the primary components of bentonite and the FR agent. The uniform distribution of these elements confirmed the effective penetration of the FR agent into the wood matrix.

[0051] XRD analysis revealed structural changes in natural wood following treatment with FR and different concentrations of bentonite (0, 1, 2, or 3%). The natural wood showed a characteristic peak for cellulose at 29 = 22.5°, indicating a crystalline cellulose structure, as well as a peak at 16.7°, corresponding to amorphous regions of lignin and hemicellulose. Following treatment with 2 wt% FR, the intensity of the 22.5° peak decreased slightly,reflecting partial disruption of cellulose crystallinity from interactions between the fire retardant and the wood matrix, e.g., the formation of phosphate-based complexes. Adding bentonite further reduced the crystallinity slightly, as shown by decreased peak intensity at 22.5°; while new peaks appeared at higher angles (26.5°, 29.0°, 30.9°, and 34.7°), corresponding to the (110), (210), (124) and (004) planes of bentonite. Calculated crystallinity index (CI) values for NW, NW-F, NW-F1B, NW-F2B, and NW-F3B bentonite samples were 50.6%, 49.9%, 48.3%, 55.4 %, and 54.8%, respectively.

[0052] The TGA and DTG data highlighted the effects of the FR / BT treatments on the thermal stability of wood. Representative TGA and DTG data are summarized in Table 5. NW underwent a two-stage thermal degradation process. The first stage (200-350°C) corresponded to the decomposition of hemicellulose and cellulose, while the second stage (350-500°C) was associated with lignin degradation, consistent with previous studies on wood thermal behavior. Table 5 shows that NW exhibited higher decomposition temperatures (Ts wt%, Tiowt%, and Tso wt%), consistent with the natural degradation pathway in the absence of any fire-retardant agents. However, the residual mass at 600°C for untreated wood was significantly lower (10.9%), due to its poor thermal stability and high flammability. The treated wood samples (NW-F, NW-F IB, NW-F2B, and NW-F3B), displayed different thermal degradation responses. The initial decomposition temperatures (Ts wt% and Tio wt%) were slightly lower, presumably due to the catalytic role of the fire-retardant agents in promoting early-stage char formation. The catalytic effect may contribute to an altered degradation pathway, facilitating the formation of thermally stable char layers. The treated samples also showed a notable increase in char residue at 600°C, with the residual mass increasing progressively with higher bentonite concentrations, reaching 26.8% for NW-F3B. The reduction in maximum degradation temperature (TmaX) for the treated samples, particularly NW-F3B, emphasized the effectiveness of the treatment in shifting thermal decomposition toward earlier stages, while promoting the retention of carbonaceous residues. Incremental improvements in char yield with increased bentonite content suggested the role of bentonite in stabilizing the thermal decomposition process and in enhancing fire resistance.Table 5: TGA data for APGP-NB wood compositesSamples T s wt% T iowt% T sowt% TmaxChar Residue at 600 °C(°C) (°C) (°C) (°C) (wt%)NW 160 250 320 352 10.9NW-F 150 235 315 288 5.37NW-F1B 191 247 295 293 19.5NW-F2B 192 245 305 296 25.1NW-F3B 204 240 310 286 26.8

[0053] XPS analysis gave insights into the surface chemistry of NW, NW-F, NW-F1B, NW-F2B, and NW-F3B. The C Is spectrum of NW showed characteristic peaks at 284.8 eV (C-C / C-H), -286 eV (C-O), and -288 eV (O-C=O), corresponding to aliphatic carbon, carbon singly bonded to oxygen, and carbonyl / ester groups, respectively — which are typical of cellulose, lignin, and hemicellulose. In the 2% NW-F sample, an additional peak at -286.5 eV (C-N and C-O) indicated the successful incorporation of APGP FR into the wood matrix. The O ls spectrum of NW exhibited peaks at -532-533 eV (C-OH and C-O-C), confirming the presence of hydroxyl and ether groups in cellulose and lignin. For the NW-F sample, a new peak at -534 eV (P=O) confirmed the presence of phosphate groups, which are desirable for enhancing flame retardance. In the NW / 2%FR / 3% bentonite sample, the O 1 s spectrum showed a shoulder at -531 eV (Si-O), confirming the successful introduction of bentonite, while the Si 2p (-103-104 eV) and Al 2p (-74-75 eV) peaks reflected the silica and alumina contributions from bentonite. The N Is spectrum of the NW / 2%FR sample revealed peaks at -400 eV (N-H) and -401 eV (NFF ), confirming the presence of amino functionalities from the flame retardant. The P 2p spectrum showed a peak at -133 eV (P-O / P=O), confirming the presence of phosphate, which promotes char formation during combustion. These observations demonstrated that the flame retardant and bentonite treatments significantly altered the surface chemistry of the wood.

[0054] The XPS results complemented the XRD observations, in which bentonite peaks were detected, and LOI measurements, in which a significant improvement in fire resistance with the addition of bentonite was seen. The presence of Si and Al peaks in the XPS spectra confirmed the uniform distribution of bentonite within the wood structure, as also seen in the SEM-EDS analysis.

[0055] Measurements of mechanical properties showed that the flame-retardant treatment slightly improved mechanical properties of the wood; but that increasing bentonite contentled to a decline in both flexure stress and modulus. While bentonite enhanced flame retardancy, its concentration should be optimized to preserve the wood mechanical properties.

[0056] We compared Limiting Oxygen Index (LOI) values for untreated wood, wood treated with flame retardant (FR), and wood treated with FR and varying concentrations of bentonite. The LOI is the minimum oxygen concentration required to sustain combustion, and is one measure of a material’s fire resistance. The untreated wood sample had a relatively low 22% LOI, reflecting high flammability and low oxygen demand for combustion. Impregnation of FR (without bentonite) into the wood significantly enhanced fire-retardant properties, increasing the LOI to 28%. For wood impregnated with both FR and bentonite, the LOI values progressively increased at higher bentonite concentrations, up to 32% with 2% bentonite. Above this concentration, no further major increases in LOI were observed, presumably due to saturation of the wood structure, limiting the incremental value of further impregnation with bentonite. These measurements highlighted the synergistic effect of FR and bentonite in improving the fire resistance of wood. The untreated wood was heavily charred during these tests, while the treated wood samples showed significantly less damage, especially at higher bentonite concentrations.

[0057] The complete disclosures of all references cited in this specification are hereby incorporated by reference. Also incorporated by reference is the complete disclosure of the priority application, United States provisional patent application serial number 63 / 549,183. In the event of an irreconcilable difference, the present disclosure shall control over an incorporation by reference.

Claims

What is claimed:

1. An APGP; wherein APGP denotes an amino-functionalized polyglycerol phosphate; and wherein said APGP comprises polyglycerol phosphate crosslinked with a polyamino compound.

2. The APGP of Claim 1, wherein said polyamino compound is selected from the group consisting of polyethyleneimine, ethylenediamine, triethylenetetramine, tetraethylenepentamine, diethylenetriamine, propylene diamine, and mixtures thereof.

3. The APGP of Claim 1, wherein said polyamino compound comprises 1% to 5% of said APGP by mass.

4. The APGP of Claim 1, wherein the molar ratio of glycerol to phosphate in said polyglycerol phosphate is from 0.25 to 2.

5. An aqueous solution of the APGP of Claim 1.

6. The APGP of Claim 1, wherein said APGP is a solid-phase powder.

7. A composition comprising the APGP of Claim 1, additionally comprising clay particles, wherein said clay particles are 1% to 10% by mass of said APGP.

8. The composition of Claim 7, wherein said clay particles comprise halloysite nanotubes.

9. The composition of Claim 8, wherein said halloysite nanotubes have an inner diameter between 30 nm and 200 nm, and a length between 1 pm and 15 pm.

10. The composition of Claim 7, additionally comprising a glue.

11. The composition of Claim 10, wherein said glue is an aliphatic glue.

12. A composition comprising the APGP of Claim 1, additionally comprising bentonite, kaolinite, montmorillonite, graphene oxide, a layered double hydroxide, zinc borate, or a mixture thereof.

13. An article of manufacture comprising a wood component and a fire retardant; wherein said wood component comprises native wood or an engineered wood product; and wherein said fire retardant comprises the APGP of Claim 1; wherein said article of manufacture has substantially greater resistance to fire than an otherwise identical wood component lacking said fire retardant.

14. The article of manufacture of Claim 13, wherein said fire retardant additionally comprises clay particles, wherein said clay particles are 1% to 10% by mass of said APGP.

15. The article of manufacture of Claim 14, wherein said clay particles comprise halloysite nanotubes.

16. The article of manufacture of Claim 15, wherein said fire retardant additionally comprises a glue.

17. The article of manufacture of Claim 13, wherein said article of manufacture has substantially greater resistance to termites than an otherwise identical wood component lacking said fire retardant.

Images