BACKGROUND OF INVENTION The reaction scheme to form a 1,3-benzoxazine is typically shown in the literature with a monohydric phenol, formaldehyde, and a simple amine. Dihydric or polyhydric phenols are conventionally used in benzoxazines to form difunctional or polyfunctional benzoxazines, which was anticipated to increase the molecular weight of the reaction product. Physical properties of the polybenzoxazines above their glass transition temperature would supposedly be dramatically increased with difunctional benzoxazines due to crosslinking and/or much higher molecular weights.

U. S. Patent 2,098,869 by Harmon and coworker used a 1: 2: 1 molar ratio (phenol : formaldehyde: aliphatic amine) composition to form an acid-soluble, heat hardening resin. In the patent the reaction of the components was continued until the resin separates out or until the mixture becomes too stiff to stir efficiently. In the claims a process is claimed including steps wherein the reaction is continued until the resinous mass thickens, grinding the cooled brittle resin with water to slurry, filtering, washing and drying. In the reaction scheme shown in the patent the reactants went directly to phenols connected with Mannich bridges without showing a 1,3-benzoxazine monomer.

SUMMARY OF INVENTION Liquid or low viscosity 1,3-benzoxazine monomers are shown and claimed. They are formed in the benzoxazine ring forming reaction of a monohydric phenol; an aldehyde such as formaldehyde; and simple amines such as linear or branched aliphatic amines, cyclic aliphatic amines, and aromatic (optionally meta substituted) amines. Benzoxazine monomers that are liquids at 23-25 °C were not available from conventional dihydric phenols that are used commercially to form benzoxazines.

Much progress has been made optimizing reaction conditions for forming difunctional benzoxazines and polymerizing them into useful polybenzoxazines with excellent thermal stability, good physical properties, and high char yield. However the conventional difunctional benzoxazines could not be prepared as liquids at room temperature. The technology advances for difunctional benzoxazines have been applied to monofunctional benzoxazines (from monohydric phenols) in this disclosure. The monofunctional benzoxazines unexpectedly were found to produce useful polymers for applications. It was presumed, due to side reactions that occur during polymerization via ring-opening, that monofunctional benzoxazines could not be polymerized to form anything other than oligomers and low molecular weight polymers. It is shown that benzoxazine monomers from aromatic amines or relative pure benzoxazine monomers from aliphatic amines can be thermally or cationically polymerized to result in useful polymers. The polybenzoxazines from aromatic amines have useful properties both below and above their glass transition temperature, which is usually indicative of crosslinked networks.

DETAILED DESCRIPTION OF THE INVENTION Benzoxazine monomers that are liquid at 23-25 °C or low viscosity at 100 °C can be formed from the reaction of monohydric phenols, aldehydes such as formaldehyde. and primary amines. The primary amines can be linear or branched alkyl amines with from 1 to 12, more desirably 1 to 8, and preferably 1 to 6 carbon atoms; cyclohexylamine; aromatic, heteroaromatic, or substituted aromatic amines of 6 to 15 carbon atoms. In some embodiments it is preferred to exclude benzoxazines made from methyl amine and cyclohexylamine or alkyl amines in general. The substituted aromatic amines include aromatic amines substituted in one or both of the meta positions with a hydroxyl group, an alkoxy group of 1 to 4 carbon atoms or an alkyl group of 1 to 4 carbon atoms. The heteroaromatic amines include those amines with heteroatoms such as nitrogen and sulfur in an aromatic ring with or without substitution. Examples of the amines include methylamine, cyclohexylamine, aniline, meta-toluidine, 3,5 xylidine, 1 (or 2)-aminopyrimidine, 2-aminothiazole, 4-aminotriazole, 2-thiazoline, and 2-aminopyridine. A preferred aldehyde is formaldehyde. Preferred monohydric phenols include an unsubstituted phenol.

The mole ratio of primary amines to the reactive hydroxyl groups on the phenol is desirably 1: 1. The mole ratio of aldehyde to reactive hydroxyl groups on the phenol is desirably 2: 1. These ratios are based on a single reactive group on each molecule. The stoichiometry can vary by less than 20 mole percent, more desirably less than 10 mole percent, and preferably less than 5 mole percent. The reasons one might vary from the preferred ratios is to compensate for impurities, poor reactivity of one component relative to the others, volatility of one component, etc. and are known to those skilled in the art.

The aldehyde, monohydric phenol, primary amine (mole ratios about 2: 1: 1) generally react according to the equation below to generate benzoxazine monomer.

The benzoxazine monomer can ring open to form a Mannich base structure as shown below. The Mannich base can form a Mannich bridge structure by attaching itself to an ortho or para position on a phenol or an ortho or para position to the oxygen on a 3,5-benzoxazine.

Thus the generation of Mannich bridge structures builds the molecular weight of polymers from benzoxazines. The Mannich bridge can also degrade to give off methyiene and the starting amine.

The monohydric phenol can have from about 6 to about 12 carbon atoms but the smaller less substituted monohydric phenois are preferred such as a simple phenol with only 6 carbon atoms. Any substituents on the phenol are desirably not present in the ortho and para positions with respect to the hydroxyl group as these sites are used in building the molecular weight of the polybenzoxazine. While the substituents can be anything that does not interfere with the formation of a benzoxazine therefrom, one or more alkyl and alkoxy substituents are preferred. It is acknowledged that the benzoxazine ring is generally lost in the ring opening polymerization.

The aldehyde can have from 1 to 6 carbon atoms but formaldehyde is the preferred aldehyde. It is available as a solution in water or as paraformaidehyde, which breaks down into formaldehyde.

For solventless benzoxazine polymerizations the paraformaidehyde is usually used as the release of water can help drive the reaction to completion.

The benzoxazine monomers can be formed by ring formation in a compatibilizing solvent or in solventless synthesis. Both are disclosed in the literature. It is desirable to synthesize the benzoxazine monomer by a procedure that results in predominantly benzoxazine rings with little or no oligomeric material as the oligomers can cause the material to be a solid at 23-25 °C. Desirably the benzoxazine ring content is at least 30 or 50 mole percent based on the total possible moles of monomer, more desirably at least 75 or 85 mole percent and preferably at least 90 mole percent of the theoretical amount of benzoxazine rings for the particular reactants. Desirably the oligomer content is low. The oligomer, byproducts, and unreacted reactants would make up the residual of the theoretical benzoxazine rings Desirably the 1,3-benzoxazine monomer as produced has a viscosity of less than 100 Pa. s at 100 °C or 10,000 Pa. s at 23-25 °C, more desirably less than 1000 Pa. s and preferably less than 100 Pa. s at 23-25 °C using a dynamic mechanical spectrometer with a 200 g-cm force rebalance transducer, 50 mm parallel plate fixture with a gap of 0.5 mm and shear rate of 0.01 to 100 s-'.

While benzoxazines from monohydric phenols are preferred due to their lower viscosity, benzoxazines formed from a mixture of monohydric and multihydric phenols having the specified low viscosities or mixtures of the preferred benzoxazines from the monohydric phenol with benzoxazines from multihydric phenol, with the proviso that they have the above specified low viscosities or the reduced viscosities later

specified. Further polymers from mixtures of benzoxazines having the specified viscosities are desirable.

These monomers are very useful as liquid precursors or easily moldable resin precursors to polybenzoxazines. The monomers can be applied to various surfaces or transported to molds much more easily than solid monomers. They can be used with high filler loadings with good processability. In adhesive or electronic potting applications where wetting the substrates is critical they can achieve that goal at significantly lower processing temperatures than benzoxazines that are solid at or near room temperature.

While the liquid monomers is a significant improvement over solid monomers, they would be of little use unless they can be converted to polymers with useful properties. It has previously been assumed that due to numerous side reactions that monofunctional benzoxazines could not be ring-opening polymerized into high molecular weight or crosslinked polymers with useful engineering applications. In this study it was unexpectedly found that monofunctional benzoxazines could be ring-opening polymerized to yield polymers with equivalent or better physical properties to those achieved with difunctional benzoxazine monomers. The physical properties include Tg, modulus at or above the Tg, char yield, etc. The monomers can be thermally or cationically polymerized. The monofunctional benzoxazine monomers produced from aromatic amines are preferred due to foaming difficulties between 250 and 300 °C when benzoxazines from aliphatic amines are heated.

The possibility of forming high polymers from mono-oxazine benzoxazines opens the possibility of forming both polybenzoxazines from mono-oxazine benzoxazines and blends of multi-oxazine ring benzoxazine monomers with mono-oxazine benzoxazine monomers. The blends are anticipated to have low viscosities, e. g. less than 80,50, or 10% of the viscosity of the multi-oxazine ring benzoxazine monomer at

a temperature 50°C above liquification temperature, when the amounts of the mono-oxazine benzoxazine and multi-oxazine benzoxazine are properly selected. Thus others would not have used mono-oxazine benzoxazines in blends with multi-oxazine benzoxazines due to a molecular weight reducing effect, attributed to monofunctional benzoxazines.

The liquid or low viscosity nature of the benzoxazine monomers and blends thereof made with this technology could also advantageously be used in blends with other thermosetting resins to prepare polymer blends with improved properties. They are very useful in these applications due to their low viscosities at mixing temperatures of 25 °C to 100 °C. This facilitates making blends without causing crosslinking of the thermosetting resin. Blends could be prepared with epoxy resins, phenolic resins, unsaturated polyester resins, vinyl ester resins, urethane resins, bismaleimide resins, imide resins isocyanate resins, or cyanate ester resins. Ternary blends could also be prepared such as blends of benzoxazine monomers with a) epoxy and phenolic resins or b) epoxy and bismaleimide resin. Combinations of thermosetting resins with benzoxazine monomers could also be made.

The desired benzoxazine polymers are those from the benzoxazine monomers which are liquids (e. g. viscosities less than 10,000,1,000 or 100 Pa. s at 25 °C). These monomers are characterized by the inclusion of monohydric phenol as a portion or all of the total phenolic molecules used and the repeat units are characterized by the presence of repeating units derived from a monohydric phenol as opposed to repeating units derived from a difunctional or higher functionality phenol.

The thermal polymerization can be conducted at any temperature capable of opening the benzoxazine ring. As specified below some chemicals such as cationic initiators can lower the polymerization temperature. Generally pure thermal polymerization is accomplished at

temperatures between 100 and 300 °C more desirably between 140 and 280 °C and preferably from about 150 to about 250 °C. Polymerization times vary with temperature. Often an incremental temperature increase is used such as 160 °C for 60 min, 180 °C for 40 min and 200 or 240 °C for 5 to 30 minutes.

In a prior application the use of cationic initiators/catalysts was found to reduce the temperature at which polymerization occurred.

These cationic species include cationic initiators known to the art for polymerizing unsaturated monomers. These include H2SO4, HCI04, BF3, AlCl3, t-BuCl/Et2AlCl, AlBr3TiCl4,I2,SnCl4,WCl6,AlEt2Cl,AlBr3, PFs, VCI4, AlEtCI2 and BF3Et20. Preferred initiators inciude PCls, PCI3, POCI3, TiC'51 SbCls, (CôHs) 3C+ (SbCI6)-, and metallophorphyrin compounds such as aluminum phthalocyanine chloride. Methyl tosylate, methyl triflate, and triflic acid appear to function as cationic initiators/catalyst for the polymerization of the benzoxazine monomers. Desirably about 0.001 to about 50 mole percent initiator based upon the monomer and more desirably from about 0.1 to about 10 mole percent initiator is used for the cationic polymerization of benzoxazines. These initiators can decrease the polymerization temperatures between 10 and 170 °C.

The polybenzoxazines from liquid benzoxazines are desirably crosslinked to an extent such that less than 30, less than 20 and more desirably less than 10 weight percent of the polymer can be extracted from samples using tetrahydrofuran Soxhlet extraction.

A process is also described for polymerizing a liquid or low viscosity benzoxazine into a polybenzoxazine via ring opening polymerization. This process is facilitated due to the convenience of using a liquid monomer that can be polymerized without undue evolution of volatile components. The ability of these monomers to polymerize in the presence of oxygen is quite an advantage over unsaturated monomers whose polymerization is hindered by oxygen, often requiring

purging with argon or nitrogen prior to polymerization. The polymerization process can be practiced as part of a mechanized production line with the application of the liquid benzoxazine monomers (optionally with a cationic initiator) and then the thermal activation of polymerization to a polymer with high thermal stability.

The polymers of this invention are useful as adhesives, coatings, molding compositions, in electrical components, and in flame resistant articles or laminates.

Examples Monofunctional 1,3-benzoxazine monomers were made from monohydric phenol (Phe), formaldehyde, and five different amines. The amines were methyl amine (m), cyclohexylamine (cyha), aniline (a), m-toluidine (mt), and 3,5-dimethyl xylidine (35x). The benzoxazines will be identified by the initias assigned to represent the phenol and the amine. As formaldehyde is present in all the benzoxazines and cannot be used to distinguish the samples it is not used in identifying the samples. A difunctional benzoxazine from a dihydric phenol (bisphenol A) was prepared and polymerized. It was used for comparison purposes.

The materials used to prepare the monomers are available from a variety of commercial sources such as Aldrich and Fisher Scientific. The monohydric phenol used in the experiments was 99% pure.

Formaldehyde was used in the synthesis of Phe-m and it was 37 weight percent active in water. Paraformaldehyde was used in the solventless synthesis of Phe-cyha, Phe-a, Phe-mt, and Phe-35x. The amines were 98 to 99 weight percent pure except for the methyl amine which was a 40 weight percent solution in water.

The Phe-a was prepared by mixing 0.2 moles of the phenol, 0.4 moles of the formaldehyde, and 0.2 moles of aniline (a) in a 400 mL beaker at room temperature. Immediately thereafter the beaker was

immersed in a 100 °C oil bath using a magnetic stirrer to mix the ingredients. The reaction was continued for 20 minutes after the temperature reached equilibrium. The resulting yellow liquid at room temperature was then dissolve in ethyl ether or chloroform and washed in a separatory funnel with a 2N NaOH solution several times until the aqueous base layer was colorless after shaking with the product. The purpose of washing was to remove unreacted material or oligomeric phenolic structures. The ether or chloroform solution was then washed with deionized water until the water pH was about 7. The product was dried over sodium sulfate overnight. The ether or chloroform was then removed with a rotary evaporator. The resulting light yellow liquid at room temperature was further dried in a vacuum oven at 40 °C for 2 hours to remove residual solvent from the purification process.

The molecular structures and purities of all the benzoxazine monomers were verified by'H NMR. The molecular weight of the oligomers and polymers were determined by size exclusion chromatography (SEC). The curing profile of the monomers was measured using differential scanning calorimetry (DSC). The DSC samples were run in a hermetic aluminum pan at a heating rate of 10 °C/min with a nitrogen purge flow of 80 mL/min. A second DSC run was performed to check the glass transition temperature assigned to the sample. The thermal stability of the cured samples was measured by thermal gravimetric analysis (TGA). The temperature was ramped at 20 °C/min under a 90 mL/min nitrogen purge.

Steady shear viscosity measurements were performed on a Rheometric Dynamic Mechanical spectrometer (RMS-800) using a 200 g-cm torque force rebalance transducer. Liquid samples were transferred directly onto a 50 mm parallel plate fixture with a gap of approximately 0.5 mm. Void-free solid disks were prepared by preheating the solid samples in a RTV silicon rubber mold. 25 mm

parallel plates were used for the solid sample measurements. The temperature for the viscosity measurements was varied to accommodate solid samples and the testing shear rate was varied from 0.01 to 100 s'.

Dynamic mechanical spectra of the polymer samples were obtained on the RMS-800 spectrometer using a 2000 g-cm toque force rebalance transducer. Specimens of approximately 50x12x3 mm were tested in a rectangular torsion fixture. A 0.1 % strain was applied sinusoidally after a strain sweep was conducted to ensure that the applied strain was within the linear viscoelastic limit. The test frequency was 6.28 rad/sec (1 Hz) and the temperature was increased from 25 °C into the rubbery plateau region of each material.

Stress relaxation experiments were performed immediately after the temperature sweep experiments. After the temperature equilibrated at 50 °C and 100 °C above the Tg, a 0.2% strain was applied and the relaxation of the shear modulus was monitored for 700 seconds.

As shown in Table I below all of the monofunctional benzoxazine monomers made from unsubstituted monohydric phenol were liquid at room temperature. In fact the Phe-m sample had such a low viscosity that it was decided to measure the viscosity at-40 °C. A difunctional phenol reacted with the aniline and formaldehyde to form a benzoxazine was a solid at 23-25 °C and had a viscosity of 5.5 Pa. s at 100 °C. The viscosities of the benzoxazines from aliphatic amines were generally lower than those from aromatic amines.

TableI Viscosity BenzoxazineMonofunctional Monomers Monomer Viscosity (Pa. s) Phe-m 5. 6a Phe-cyha 1. ob Phe-a 4. 6b Phe-mt 15 b Phe-35x43"

a. measured at-40 °C b. measured at 23-25 °C In U. S. Patent 2,098,869 to Harmon and coworkers a reaction product from phenol, formaldehyde and aliphatic amines was reported.

Their goal was to make an acid-soluble, heat hardening resin. To differentiate those resins from the benzoxazine resins in this application, experiments following the patented procedures were religiously followed. The phenol-formaldehyde-methyl amine based resin made according to the experimental procedure was a white solid at room temperature with a melt viscosity of 3.5 x104 Pa. s at 100 °C. NMR analysis of the resin of the patent revealed that only 11 mole % of the theoretical benzoxazine ring content was present and 74 mole % of the product was oiigomeric material. Apparently the reaction conditions result in predominantly oligomeric product which is a solid.

To better characterize the relative amounts of monomer and oligomer that are present in liquid and solid benzoxazine resins an experiment was conducted to generate incremental decreases in the monomer content and increases in the oligomer content. Phe-a was the monomer used. The Phe-a resin was partially cured in an oven at 160 °C for different time intervals. The Phe-a resin was evaluated for benzoxazine ring content and viscosity after the partial curing. The relationship between ring content and viscosity is given in Table II. It is

seen that with 61 % ring content the viscosity is high, with ring content of 52% or less the viscosity couldn't be measured at room temperature but had to be measured at 50 or 100 °C, and with 30 mole % ring content the resin appears to be a solid at room temperature. Even though the high oligomer resins are solids at room temperature the viscosity drops rapidly with increasing temperature.

Table II Relationship Between Phe-a Resin Ring Contents and Viscosities Benzoxazine ring (Pa.s)Viscosity (% of theoretical) 95 4. 7a 778.3a 70 88a 67250a 61800a 52 5. 6 30 1 7. 5c I a. measured at 23-25 °C b. measured at 50 °C c. measured at 100 °C The polymerization behavior of all the synthesized monofunctional monomers was studied by DSC. The non-isothermal thermograms are summarized in Table IIl. The Phe-m and Phe-cyha exhibited noisy baselines when the scanning temperature exceeded 250 °C. This result may be due to volatile aliphatic amines evolving from these two cured systems. The higher Tg of the Phe-m and Phe-cyha may reflect the formation of phenolic methylene linkages. It is generally believed that Mannich bridge decomposition reactions that evolve aliphatic amines form methylene linkages which are stiffer linkages than Mannich bridges and result in higher Tg values.

Table III Calorimetric Data for the Polymerization Reaction Phe-m Phe-cyha Phe-a Phe-mt Phe-35x Exothermic peak °C 212 237 206, 284 217 207 Heat of reaction (J/g) 285 214 327, 22 333 353 | Tg (°C) b 2nd DSC run 192 225 142 122 130

Each of the aforementioned polymer products from DSC non- isothermal cure were taken out of the DSC pan after two DSC runs in order to check their solubility in THF, thereby roughly estimating whether or not the samples had been crosslinked during the DSC cure process. Four of the samples showed excellent resistance to both cold and hot THF for prolonged time. The Phe-cyha sample was the only exception and it showed partial solubility in THF after 24 hours.

The Phe-a monomer was cured under several different ring- opening polymerization conditions. The thermal polymerization in air circulating oven conditions are shown in Table IV. The Tg values of the samples in Table IV were 96,124,136, and 165 °C respectively.

Further DSC runs on these samples indicated they were not completely cured. The thermal stability of these polymers were investigated by TGA. All the poly (Phe-a) investigated by TGA were generally thermally stable up to 300 °C, which is approximately 50 °C higher than that of their difunctional counterpart. The char yield of the poly (Phe-a) (cured at 160 °C for 1 h, 180 °C for 40 min, and 200 °C for 15 min from Table IV) was about 7 weight percent greater than the char yield of a difunctional benzoxazine polymer counterpart. This indicates superior char properties for the polymer derived from a monofunctional monomer.

TableIV Different Thermal Curing Conditions Poly (Phe-a) 160 °C for 1 h 180 °C for 40 min Poly (Phe-a) 160 °C for 1 h 180 °C for 40 min 200 °C for 15 min Poly (Phe-a) 160 °C for 1 h 180 °C for 40 min 220 °C for 15 min Poly (Phe-a) 160 °C for 1 h 180 °C for 40 min 240 °C for 15 min

A comparison of the properties of a Phe-a polymerized under thermal and methyl tosylate initiated curing is shown in Table V. The methyl tosylate was used at the 5 mole percent level based on the monomer. Both specimen were cured in a U-shaped silicone rubber spacer between two surface-treated glass plates. The thermal curing conditions were 160 °C for 1 h, 180 °C for 40 min, and 200 °C for 15 min. The methyl tosylate cured sample was cured at 150 °C for 0.5 h and 180 °C for 1 h. The methyl tosylate initiated cure occurred with an exothermic peak at 146 °C and a second peak centered at 178 °C. As shown in Table III the majority of the exothermic peak for the thermal polymerization of Phe-a occurs at 206 °C.

Table V Physical Properties of Thermally and Cationically Cured Resins Poly (Phe-a) Methyl tosylate initiated poly (Phe-a) y-transition (°C)-52-55 Tg (°C) 120 126 Storage modulus G'at 25 (°C) (Pa) 2.4 x 109 2.3 x 109 Plateau modulus G'at 170 (°C) (Pa) 1.3 x 106 1. 1 x 106 Crossiink density (10z3 mol, cmtcalc 0. 4 0. 3

The crosslink density was estimated from the equilibrium storage modulus in the rubbery region using the kinetic theory of rubber elasticity with the equation Pe = Ge/RTe in which + is referred to as the front factor and is typically assumed to be 1, R is the gas constant, and Te is taken as the temperature 50 °C above Tg from which the Ge values are obtained.

Stress relaxation measurements have been reported previously for difunctional based polybenzoxazines but the results indicated some relaxation of stress. This prompted speculation that the polymers from difunctional benzoxazines were not truly crosslinked materials but possibly only hyperbranched. Given the moderate Tg and a high onset of degradation of the Phe-a based polybenzoxazine of this study, it was decided to measure the stress relaxation of the polymer both at 50 and 100 °C above its Tg. The Phe-a system used for this study did not show any measurable relaxation over a 700 second testing span at both temperatures. The facts of THF insolubility of the polybenzoxazine from Phe-a and no measurable stress relaxation at temperatures above the Tg suggest this polymer is chemically crosslinked.

While in accordance with the patent statutes the best mode and preferred embodiment have been set forth, the scope of the invention is not limited thereto, but rather by the scope of the attached claims.