GB1575366A - Process for producing particulate resoles from aqueous dispersion - Google Patents

Description

(54) PROCESS FOR PRODUCING PARTICULATE RESOLES FROM AQUEOUS DISPERSION (71) We, UNION CARBIDE CORPORATION, a Corporation organised and existing under the laws of the State of New York, United States of America, of 270 Park Avenue, New York, State of New York 10017, United States of America do hereby declare the invention, for which we pray that a patent may be granted ot us, and the method by which it is to be performed, to be particularly described in and by the following statement:- The invention relates to a process for producing particulate solid, heatreactive, substantially water-insoluble resoles.

Although solid, fusible, heat-reactive phenolic resole (one-step) resins have been produced by essentially the same process for more than half a century, the manufacture of these resins suffers from poor product control and poor batch-tobatch reproducibility.

To make a resole, phenol is reacted with a molar excess of formalin, using a base catalyst. Under alkaline conditions, phenol and formaldehyde condense to form methylols:

Further condensation results in the formation of methylene ethers and methylenes:

Because of the polyfunctionality of the system and because of the high formaldehyde to phenol mole ratio (typically 1.3 to 1.8), condensation to these latter structures cannot proceed very far before gelation will occur.

In the historical resole process, phenol, formalin and the base catalyst are held at 70" to 1000C. for a predetermined time to convert substantially all of the formaldehyde to methylol. Then the batch is dehydrated at a low temperature in a reduced pressure distillation step so that the 100% solids resin may be advanced at an elevated temperature to the desired degree of condensation. A hot plate gel time test (at 1 500C) is run on samples taken at regular intervals, and the time of arrival at a pre-selected endpoint is estimated. At this time the molten resin is discharged as rapidly as possible into water-jacketed resin coolers. It has been observed that the gel times of samples taken at the start of the discharge, at the midpoint, and at the end of the discharge differ considerably since the hot resin in the still continues to advance before it can be completely discharged. The final resole resin is an amorphous brittle solid having a glass transition temperature not much above room temperature.

From this description of the historical resole process, it is apparent that the large scale batch production of solid resole resins involves a considerable amount of operator skill and art if either too "green" or gelled batches are to be avoided.

On the other hand, two-step resins (novolaks) are incapable of gelation and, therefore, are much less critical to make. It is readily understandable that, wherever possible, two-step resins, plus hexa (i.e., hexamethylenetetramine) to provide cure, are preferentially employed commercially. There are certain applications, however, for which the characteristics of the one-step resins are preferred. One such application area is that of the pulverised resins where a two-step resin/hexa system is undesirable because of the volatility of hexa. A resole produced direct in particulate form might also be of interest in these application areas for economic reasons; the grinding operation adds significant cost per pound of pulverized resins.

Further economic incentive for providing an improved particulate resole resin process can be found in the lower cost of formaldehyde than that of phenol (as well as a lesser rate of recent increase), so that resoles currently have a raw material cost advantage over novolaks.

The published literature has from time to time mentioned the production of powdered or particulate phenolic resins from dispersions, usually aqueous dispersions, by removal of the dispersing medium. For instance, there are casual references to the drying of aqueous suspensions or dispersions of phenolic resins to leave a dry, powdered residue (see, e.g. Dent et al., United States Patent Specification No, 2,069,178). Cheetham, in United States Patent Specification No.

1,976,433, discloses the joint precipitation of a phenolic resin and a filler to produce an intimate mixture of the resin and filler. A number of patents disclose the spray drying of aqueous phenolic resin mixtures to produce a dry, but usually water-soluble, phenolic resin. Suitable examples are United States Patent Specifications Nos. 2,186,369, 2,675,336 and 2,489,336. Wismer et al., in United States Patent Specification No. 3,850,868, disclose the production of inert phenolic resin beads by reacting phenol with from 1 to 3 moles of formaldehyde in an aqueous basic medium to form a prepolymer, blending the prepolymer with a protective colloid such as polyvinyl alcohol, fully polymerizing the prepolymer by acidifying and heating the mixture under agitation to form the inert beads, and removing the water. Kooguchi et al., in United States Patent Specification No. 3,743,619 describe the production of novolak powders by removing the water from an aqueous dispersion of a novolak. Spitzer, in Swiss Patent Specification No. 153,843 (Chem.

Abs. 1933, 27, 1113), proposed to precipitate plastic masses from colloidal solutions by the addition of a ureaformaldehyde hydrosol. Bresser, in British Plastics and Moulded Products Trader, August 1932 4, 105, 112, reported the preparation of phenolic resin precipates by dilution, "salting out", or acidification of an aqueous phenolic resin suspension containing a dispersing agent or protective colloid. (In this respect, Bresser cited D. R. P. 516, 677.) In United States Patent Specification No. 1,807,545, Ostersetzer et al disclose the production of powders by reacting a phenol and formaldehyde in an alkaline aqueous medium in the presence of a protective colloid such as gum arabic, followed by precipitating a powder from the reaction mixture. Example 2 of the patent, wherein the basic catalyst was a mixture of sodium hydroxide and ammonia, was followed as closely as possible. (This was made difficult by the lack of detail in the Example relating to endpoint and reaction cycle). Using reaction conditions (15 minutes at 850C.) so as to obtain a resin having applicant's typical 150"C. hotplate gel time endpoint (60 to 90 seconds), an emulsion of a liquid phenolic resin was made that passed through a filter. In a repeat experiment using a longer reflux time (1.5 hours at 1000C.) a filterable but gelled product was produced which nevertheless sintered on storage.

Other prior art discloses the preparation of aqueous phenolic resin dispersions utilizing various interfacial agents. (This art does not teach the isolation of the dispersed reactive resin by removing the water.) Examples include United States Patent Specifications Nos. 3,823,103, 3,666,694 and 3,862,060.

The present invention provides a process for the production of particulate solid heat-reactive, substantially water-insoluble resoles. The process comprises the steps of: (a) reacting formaldehyde, a phenol comprising a trifunctional phenol, a tetrafunctional phenol, or a mixture thereof, and an effective amount (as hereinafter defined) of hexamethylenetetramine, a compound containing amino hydrogen (as hereinafter defined) or a mixture thereof, in an aqueous medium in the presence of an effective amount (as hereinafter defined) of a protective colloid which colloid is present at the beginning of the reaction, at a temperature of from 70" to 1000C, for a period of time sufficient to produce an aqueous dispersion of a particulate solid, heat-reactive (as hereinafter defined), substantially waterinsoluble thermosetting resole: (b) cooling the reaction mixture to a temperature below 40"C on completion of the reaction; and (c) recovering the particulate resole from the aqueous dispersion.

The following definitions apply in this invention: Particulate solid - This term refers to a finely divided solid having a maximum particle size of the order of 500 microns. Preferably, the average particle size will range from 5 to 30 microns. By "solid", is meant a resin having a softening point (glass transition temperature) above 30"C., and preferably above 40"C.

Heat-reactive - This refers to a phenolic resin that is capable of polymerization at a usefully practical rate when subjected to elevated temperatures, e.g., temperatures above 100"C. Ordinarily, these resins will advance slowly, but measurably, even at lower temperatures.

Phenolic resin - a condensation product of a phenol and an aldehyde, usually formaldehyde.

Resole - a phenolic resin produced using at least equimolar amounts, and usually more, of aldehyde. Base catalysts are ordinarily used in making resoles. As used in the art, "resole" refers to phenolic resins that contain useful reactivity, as opposed to the fully polymerized or cured resins.

Substantially water-insoluble - This refers to a phenolic resin that, while it may very well contain a finite proportion of water-soluble components (such as unreacted phenol or mononuclear methylol phenols), is capable of existing as a separate phase in an aqueous medium.

Protective colloid - A composition that is capable of promoting the production of, or of maintaining, a phenolic resin-in-water dispersion wherein the phenolic resin is the dispersed phase and the aqueous medium is the dispersing or continuous phase.

Aqueous dispersion -- A used herein, an aqueous dispersion is a dispersion wherein the dispersing or continuous phase is composed primarily of water, although some organic material will be dissolved in the aqueous phase and some water will be contained in the dispersed resin phase.

Amino hydrogen -- Refers to a hydrogen atom bonded directly to an amino nitrogen atom (including that in ammonia). As used herein, the term excludes amidohydrogen atoms wherein the nitrogen atom is alpha to a carbonyl group.

In the first step of the process of the invention, formaldehyde is reacted with a phenol an(t hexamethylenetetramine or a compound containing at least one amino hydrogen, in an aqueous medium in the presence of a protective colloid. The formaldehyde is most conveniently and economically employed in the form of the aqueous solution known as "formalin", although other forms, such as paraform, can be used.

The process also uses a trifunctional phenol, such as phenol (monohydroxybenzene), o-, m- orf-cresol or resorcinol, or a tetrafunctional phenol; such as, 2,2-bis(4-hydroxyphenyl)-propane ("bisphenol-A"), and mixtures thereof.

Para-substituted phenols such as p-cresol, p-chlorophenol, and p-t-butylphenol, can also be used, but only as a fraction of the phenol part of the formulation. When p-tbutylphenol was used with phenol in amounts of at least about 50 mole per cent or more of the phenol component, a non-thermosetting product was produced. The reason for this failure to produce a thermosetting product is not known; it can only be postulated that the p-alkylphenols lack sufficient reactivity toward formaldehyde under the conditions of the process of this invention. Compared to the conventional resole process (in the melt, at 1000C. or higher, and at a pH of 8), our conditions are quite mild (in the presence of 30 by weight or more water with 30 to 50% being preferable, at 850C. or even lower, at a pH of about 7). In fact, theoretically, even a p-alkylphenol/formaldehyde/ammonia (or hexa) formulation should yield a thermosetting product because of the polyfunctionality contributed by the ammonia. Routine experimentation will suffice to determine the maximum proportions of difunctional phenol that can be used in particular cases, while still producing a thermosetting product, but generally less than 50 mole% of a difunctional or monofunctional phenol is used, based on the total amount of phenol present.

The third reactant is hexa or a compound that contans at least one amino hydrogen. Specific illustrative examples include ammonia, hexa (the two are complete equivalents in this invention), and aliphatic amines such as methylamine, dimethylamine, ethylene diamine, and monoethanolamine. Ammonia and hexa are preferred. The other amines mentioned act as plasticizers and yield products that tend to sinter. When used, they will usually be employed in minor amounts in combination with ammonia and/or hexa as a means for increasing flow.

The reactants are employed in proportions such that the resulting resole is a thermosetting composition. The maximum utilizable proportion of methylene is 0.5 methylene per ring hydrogen in the phienol that is reactive with formaldehyde. (One equivalent of methylene is provided by each mole of formaldehyde, and six equivalents of methylene are provided by each mole of hexa. The reactive ring hydrogens are those that are ortho and para to the phenolic hydroxyl.) The operative proportions of the reactants are much broader, however, and actually cannot be stated with much precision. For instance, a large excess of methylene can be present in the reaction mixture, but the excess will either be extracted (as free formaldehyde) during subsequent washing of the resole, or it will be lost during cure. Resoles containing excess formaldehyde will also have an increased tendency to sinter. If insufficient methylenes are present in the reaction mixture, a thermosetting product can still be produced, but part of the phenol will remain unreacted. Much of the free phenol can be extracted from the product by washing, but it would have to be recycled to avoid waste. In almost all cases, the reaction mixture will contain from about 0.25 to about I methylenes per reactive hydrogen in the phenol. As a guide, when phenol (monohydroxybenzene) is used, preferred proportions will be from about 1.25 to about 1.75 equivalents of methylene per mole of phenol.

As a general rule, the amine compound is used in an amount sufficient to impart a substantially hydrophobic character to the resole. As will be discussed more fully below in the experimental section, most of the amino compound reactant is apparently incorporated in the product in the form of a benzylamine, a certain minimum proportion of which appears to be required in order to impart the desired substantially water-insoluble property to the product. This minimum proportion will vary somewhat, depending on such factors as nature of reactants and proportion of methylene to phenol. When hexa is used with phenol, the minimum is about 6 parts of hexa per 100 pars, by weight, of phenol. When 5 parts of hexa were used in a similar formulation, the resole coalesced during filtering.

When hexa is used with bisphenol-A, a minimum of about 3 parts of hexa per 100 parts of bisphenol-A should be used. A convenient way to determine the approximate minimum proportion needed in particular cases, is to vary the proportion of amine compound in a series of batches to determine at what point the product becomes sufficiently hydrophobic to be readily separable from the aqueous phase by filtration. Those products that are insufficiently hydrophobic will coalesce to an unfilterable mass (in extreme cases), or, in borderline cases, will excessively agglomerate during filtration or drying. (This test is valid to give the approximate minimum of amine needed, even though a means other than filtration, e.g., centrifugation, is used in separating the aqueous phase from the particulate resole. Slightly more hexa might be needed when centrifugation is used because filtering is usually a gentler treatment.) There is also a maximum amount of amine compound that should be used.

Above a certain proportion of amine compound, a rapid initial reaction is obtained, but the reaction product cannot be advanced beyond a certain point, as is evidenced by failure to obtain gelation on the 150"C. hotplate after 5 minutes or more. (From one to one and one-half minutes is the usual 150 .C. gel time target. In particular cases the precise maximum will have to be determined by routine experimentation. When hexa is the amine compound, at least one mole of free formaldehyde (as distinguished from the methylenes provided by hexa) should be provided per equivalent of amino nitrogen. When ammonia is used, at least 2-1/2 moles of free formaldehyde should be used per mole of ammonia.

The amount of amine compound that should be used will be termed herein as an "effective amount", which is defined as being above the minimum and below the maximum proportions described above.

The process of the invention also uses a protective colloid. Materials that have been found to be operative as protective colloids include gum arabic, gum ghatti, hydroxyalkyl guar gum (e.g., the ethylene oxide or propylene oxide adducts of guar gum having a minimum degree of hydroxyalkylation of about 5 weight per cent), partially hydrolyzed polyvinyl alcohol, hydroxyethyl cellulose, carboxymethyl cellulose, soluble starch, and agar. Gum arabic, gum ghatti, hydroxypropyl guar gum, and partially hydrolyzed polyvinyl alcohol have given the best results. The commercially available polyvinyl alcohols that gave the best results were those having a molecular weight of the order of 10,000 and a degree of hydrolysis of about 88 per cent (remainder of the units in the polymeric chain being vinyl acetate units). Higher molecular weight versions of partially hydrolyzed polyvinyl alcohol (about 96,000 and about 125,000 m.w.) yielded coarser particle size dispersions, and in one instance, a dispersion too poor to permit isolation of a particulate solid (the resin coalesced). Lower molecular weight versions (about 3000 m.w.) of 88 per cent hydrolyzed polyvinyl alcohol, and a 99 per cent hydrolyzed 10,000 molecular weight, version of polyvinyl alcohol, were unsatisfactory in that the reaction mixtures utilizing them coalesced upon cooling.

The protective colloid is used in an effective amount, that is, in an amount sufficient to promote the formation of and/or stabilize a phenolic resin-in-water dispersion. It is sufficient that the dispersion be stable while the reaction mixture is being agitated to the degree ordinarily associated with the production of phenolic resins. It is desirable for the dispersion to settle on standing within a reasonable time (e.g., within a few minutes), in order to facilitate separation of the particulate resin phase from the aqueous phase. Of course, when the resin phase settles, it should not coalesce to a non-particulate mass. This capability is a function of both the substantial water-insolubility of the resin and the protective colloid.

It is desirable to use the protective colloid in a minimum effective amount, because when excessive amounts are used the dispersion may take an impractically long time to settle on standing. While the precise proportions needed may vary somewhat from one case to another, and can be determined by routine experimentations, a minimum effective amount will normally be from 0.1 to 0.5 weight per cent, based on weight of phenol. It is rare that the protective colloid will be used in amounts of more than about 2 weight per cent, based on weight of phenol. A preferred proportion is about I weight per cent, based on weight of phenol.

The reaction is carried out in an aqueous medium. The minimum amount of water that should be used is that proportion which will permit the formation of a phenolic resin-in-water dispersion. While the minimum amount of water needed will vary somewhat from one formulation to another, as a rule at least about 50 parts by weight of water should be provided per 100 parts by weight of phenolic resin. (The weight of phenolic resin is equal to the weight of phenol, formaldehyde and amine charged to the reaction, minus the amount lost as water of condensation). A moderate excess over the minimum amount of water is recommended. There is no advantage to using a large excess (in fact, it is undesirable because reactor productivity will be reduced), although the process will be operable with a large excess of water. Preferably the aqueous dispersion continues from 30 to 50% by weight of water.

The reaction of step (a) can be carried out by charging all the components to a conventional reaction vessel (such as the type that has been used for over 50 years in making phenolic resins), and heating the reaction mixture, while agitating, to an elevated temperature. It is probably preferable to charge the protective colloid at the beginning of the reaction, although it could be added later during step (a), if desired, up to the point that molecular weight build-up to water insoluble products occurred, as evidenced by clouding of the reaction mixture. A reaction temperature of from 80" to 850C. has been found to be convenient, although a temperature of from 70" to 1000C. could be used in some cases. At lower temperatures, below, for example, about 75"C., the reaction becomes impracticably slow. At temperatures much above about 90"C., it may be too difficult to control the reaction in lafge scale batch equipment. A few minutes after reaching reaction temperature, the mixture will become cloudly. The progress of the condensation reaction can be followed by periodically withdrawing samples and testing the samples for the 150"C. hotplate gel time (a procedure known to all phenolic resin chemists). The desired gel time may vary, but it will usually be reached in from 60 to 90 minutes at from 800 to 850C. The temperature can be maintained during the reaction by standard heating or cooling means, as required.

In commercial size batches, vacuum reflux is a convenient way to maintain the desired temperature.

When the desired gel time has been reached, the mixture is cooled to below 40 C. to quench the reaction.

In step (c) of the process, the particulate solid resole is recovered from the aqueous dispersion product of step (a). The dispersion can be diluted with more water, cooled to from 200 to 300 C.; and filtered to produce a wet filter cake. (In commercial scale operation, centrifugation may be preferable to filtration.) The filter cake can then be washed with pure water (at 200C to 300C.,) filtered again, and then dried. During the final drying step, gentle agitation is desirable in order to deter agglomeration of the particles. Fluidized bed drying has been found to be an excellent way to effect the final drying step.

The resulting particulate resole is often a free-flowing powder, preferably having an average particle size of less than 100 microns in diameter, that has excellent resistance to sintering. The powder can be used for the classical phenolic resin applications such as fiber bonding, molding, and adhesives.

The invention will now be further described in the following Examples.

EXAMPLE 1.

This Example illustrates a general procedure for carrying out the process of the invention, and the effect of varying the reaction time in step (a).

A "standard formulation" was employed for many of the experiments reported herein. (The formulation was selected for convenience, and is not necessarily an optimized formulation for any particular end-use application). The components of the reaction mixture are shown in Table I: TABLE I Parts :buy Grams for a Moles per Component Weight 3-liter batch mole phenol Phenol 100 700 1 40% aqueous formaldehyde 90 630 1.1:3 Water 146 1020 Hexa 9 63 0.06 Protective colloid 1 7 The formulation contained 1.49 equivalents of methylene per mole of phenol and 0.24 equivalents of amino nitrogen per mole of phenol.

The experimental procedure is the following (using gum arabic as the protective colloid): The phenol, formalin and water were weighed directly into a 3-liter flask which was equipped with a motor-driven paddle stirrer, a thermometer, and a reflux condenser. After adding the hexa and gum arabic (both as dry powders), the batch was heated cautiously to 850C.: this required about 45 minutes. Within less than one minute at 850C., the originally clear solution suddenly became very opaque; from this point on the batch had the appearance of a dispersion. At this 3-liter scale the condensation was exothermic enough that some external cooling was required for about 40 minutes to keep the temperature from rising above 85"C.; at a l-liter scale the same reaction appeared to be much less exothermic. Sixty minutes after the cloud point, a 400-ML sample was removed from the batch without interrupting the agitation; it was cooled to about 40"C. and then poured into 1200 ML of tap water at room temperature stirred with a magnetic bar. Four more 400 ML samples were taken at 75, 90, 120, and 150 minutes, and the reaction again quenched by cooling and diluting.

In the subsequent work-up, each of the diluted samples was stirred for a few minutes and then allowed to settle. With gum arabic as the interfacial agent, settling was surprisingly rapid. The supernatant liquid was decanted and discarded (after filtering off a small amount of non-settled solid to keep the experiment as quantitative as practical). The solid left behind was washed with another 1200 ML of tap water at room temperature. After a second and a third decantation and a fourth wash, the solid was filtered off and the wet solid spread out to dry. In this experiment and others like it, a small amount of sintering occurred during the room temperature (about 30"C.) drying. It was found that turning over the wet powder during the first hour or two of drying helped to minimize sintering. In the experiment being described here, it was noted that the first sample (100-sec Gelt Time) filtered somewhat more slowly than the others and sintered a bit more. After removing the last sample (which analysis later showed to be completely gelled), the flask was cleaned by flushing with cold water; the glassware was essentially clean except for a very small amount of resin film on the agitator shaft.

As anticipated, the last sample was insofuble and infusible. The fourth sample was barely fusible at 1500C. (gelling quickly) and merely swelled in solvents. The other 3 samples were completely soluble and fusible. Table II summarizes the properties of these products along with the same properties measured on commercial one-step resins selected as reference materials. It was concluded that the three fusible particulate solids bracketted the useful one-step resin range.

A comment should be made about the plate flow data of Table II. In the usual plate flow test, 0.5 g of powdered resin is pressed to a pill. In the case of the experimental products these did not sinter sufficiently to make a coherent pill; it was necessary to increase the pressure (by increasing the pill weight to 0.8 g) to do so. This was interpreted to be an indication of the protective action of gum arabic.

Making a coherent pill from gum ghatti-based resins was even more of a problem.

In the plate flow test, the pressed pill is placed on a glass plate, which is then placed in a horizontal position in a 125"C. oven. After 3 minutes, the plate is tilted down 45" from the horizontal. The length the resin flows before gelling is then measured. Plate flow in phenolic resins is a function of degree of cure and the presence or absence of plasticizing agents such as water and free phenol. The plate flow test measures gravity induced flow. It is not unusual for a phenolic resin to have zero plate flow, but at the same time, for the same resin to have sufficient flow to be useful in end-use applications wherein the resin is subjected to heat and pressure during cure.

TABLE II Chacterization of One-step Resins 90-Factor/9#Part Hexa GPC Analysis Sample 85 C. Gel Time 125 C. Free No. Reaction %Nb at 150 C Plate Flow Phenol An Aw Aw/An 1 60 min. 2.41 102 sec. 111, 100mm 10.4% 41.0 85.7 2.09 2 75 2.34 70 53, 58 9.9 44.2 100.3 2.26 3 90 2.31 42 31, 28 8.5 45.1 104.1 2.30 4 120 2.34 ------------ Only slightly fusible; merely swells in solvent-----------5 150 2.28 ------------------ Comple tely infusible, insoluble ------------------- Resole Aa - 45 34, 34 2,6 54.3 120.3 2.21 Resole Ba - 63 85, 91 9.3 43.1 91.9 2.13 NOTES: a. Recent commercial batches of Resole A (Ba(OH)2/hexa resole) and of Resole B (hexa-catalyzed resole) chosen as reference materials. b. Calculated from hexa charge: 2.46%N. c. An number average molecular length; Aw weight average molecular length; Aw/An = dispersity, a measure of molecular weight distribution. For phenolic resins, the corresponding molecular weight figures are about 10 times the GPC values.

The particle size of the powder did not appear to be related to easily crushed. The use of a fluidized bed dryer has virtually eliminated agglomeration during the drying step.

In the water-washing step of the experiments reported here, after reaching the gelt time end-point, the dispersion, originally at about 40% per cent solids, was diluted with about 3 times its volume of water at 20 -30 C., to quench the condensation and also to make decantation or filtration easier. The first wash was followed by three more washes of the same volume. These washes add up to a considerable volume which would become of concern to any scale-up of the process. Further work has shown that one, or at most two, washes using one volume (instead of three) of additional water per wash, is probably sufficient in almost all cases to remove the lower molecular weight, water-soluble organic materials that promote sintering in the final product. The water from the first wash can be concentrated and recycled to the reaction in order to lessen the loss of raw materials.

In the standard formulation, it is beneficial to cool the dispersion to a temperature below about 30"C. before filtering. This facilitates filtering, apparently by cooling the resin particles below the temperature at which they tend to agglomerate. Little advantage is gained by cooling below 20 C. The temperature below which filtering is facilitated may very well differ somewhat from one formulation to another, but it can readily be determined by routine experimentation. Unless otherwise stated, the filtering step in the experiments reported below was carried out at about 25 C.

EXAMPLE 2.

Samples of particulate resole made by the formulation and procedure of Example I were evaluated in a particle board molding formulation. The results are shown below in Table III, with a commercial pulverized phenolic resin shown for comparison. The commercial resin was a one-step, two-step resin mixture. To the best of applicants' knowledge, there are no pulverized pure resole phenolic resins that are commercially available.

DEMONSTRATION OF THE UTILITY OF PARTICULATE RESOLE THE MIX, parts by weight A B C D Commercial Resin 2 4 - - Particulate Resole - - 2 4 Coarse Wood Flour 106 106 106 106 RESULTS A B C D Torsion Sheer (psi) 6 Min. Cure 185 + 26 300 # 32 182 # 22 255+25 10 Min. Cure 217+13 337+28 208*12 287 +17 Dry Flex. Str. (psi) 2.5 Min. Cure 2529 # 273 4620 # 466 3690 # 190 4782 # 513 5 Min. Cure 3945+388 5726+386 3724*284 4800+352 Wet Flex. Str. (psi) 2.5 Min. Cure 543 # 123 1533 # 158 648 # 354 1278 # 432 5 Min. Cure 950+158 12650+734 802*126 1598+218 All cures were at approximately 165 C. for the lengths of time noted. Wet flexural strength data were obtained after a 16-hour soak.

EXAMPLE 3.

In this experiment, the following procedure was employed using the standard formulation: All of the reactants were weighted directly into a creased (Morton) flask; the order of addition is not considered to be critical. The flask was equipped with a motor-driven stirrer, a reflux condenser and a thermometer. The batch was heated to 850C. and held at this temperature by cooling or heating as required. (In a 35-gallon still run the desired 851l0C. temperature was held easily by vacuumreflux.) Very soon after reaching 85"C. the batch, originally a clear solution, clouded as the condensate reached the water-insoluble stage. After about 30 minutes beyond the cloud point, the degree of resin advancement was ascertained by measuring the gel time on a 150"C. steam hot plate; in the work reported here a 75-second gel time was the target.

At the extrapolated time, the condensation reaction was quenched by cooling the flask contents with a cold water bath; in the pilot plant run noted above, vacuum reflux cooling was very effective. After cooling to about 30"C., the batch was poured into 500 parts by weight of cold water contained in a beaker and stirred with a magnetic stirring bar. After stirring 10 minutes, the slurry was filtered through coarse paper (Whatman [Registered Trade Mark] No. 1) in a large Buchner funnel with the aid of modest suction. When the application of full house vacuum failed to wring any more water from the filter, the wet filter cake was coarse-ground by pushing it through an 18-mesh screen, dried with air in a Pfaltz and Bauer fluidized bed drier at ambient temperature for 30 minutes (one-liter batch size; l hour for a two-liter batch) and the dry powder passed through a 25mesh screen to break up any agglomerates that might be present. Excellent results were later obtained by fluidized bed drying for 30 minutes at 600 C. This drying cycle invariably reduced the water content of the powder to less than 1 per cent, and usually to less than 0.5 per cent. When drying at elevated temperature, the powder should be cooled before being discharged from the fluidized bed, in order to avoid sintering.

Some 22 materials were substituted for the gum arabic of the standard formula, each at the 2 parts level. It was easy to differentiate the protective colloids from the materials that did not exhibit protective colloid properties. The latter resulted in gross separation into a fluid aqueous phase and a viscous resin phase as the condensation proceeded beyond the cloud point. The protective colloids, on the other hand, gave fine particle size dispersions (the whiteness of which gave some clue to the particle size) which could be filtered and dried.

Three gum arabic substitutes, gum ghatti, ELVANOL 51-05 and JAGUAR (Registered Trade Mark) HP-I, gave small particle size, very free-flowing powders which were at least the equivalent of gum arabic-based products. The appearance, plate flow and reactivity of these products is summarized in Table IV. The resoles based on gum ghatti and JAGUAR HP-I fused and were reactive at elevated temperatures but showed no (gravity induced) plate flow. That this lack of flow was due to more than fine particle size was suggested by the finding of some flow at essentially the same particle size with ELVANOL 51-05 as the colloid.

Four other materials (hydroxyethyl cellulose, carboxymethyl cellulose, soluble starch, and agar) did not show the very fine particle size and free-flowing character of the products of Table IV, but gave an adequate dispersion and good filterability, but the products tended to be lacking in sinter resistance. With some further work, optimized formulas might be found for these materials. Our results are summarized in Table V.

Table VI lists the other 15 materials which showed no protective colloid properties at the 2 parts level.

TABLE IV Protective Colloids Showing Excellent Performance in Particulate Resole Preparationa Sample Protective Colloidb Gel Time Plate Flow Other Observations 1 Gum Arabic (1 Part) 77 sec. 40 min. Range of particle sizes, very free-flowing 2 Gum Ghatti (1 Part) 79 0 Small particle sizes, very free-flowing 3 ELVANOL 51-05 78 44 Small particle size, very free-flowing 4 JAGUAR HP-1 80 0 Small particle size, very free-flowing NOTES: a. Standard formulation and process. Protective colloids at 2 parts per 100 phenol (by weight), except as indicated. b. Identity of materials.

Gum Arabic - Polysaccharide of natural origin, imported from Africa.

Gum Ghatti - Polysaccharide of natural origin, imported from India.

ELVANOL 51-05 Polyvinyl alcohol, 88% hydrolized, low molecular weight (about 10,000) JAGUAR HP-1 Hydroxypropylated guar gum containing about 15 weight per cent propylene oxide units.

TABLE V Protective Colloids Showing Promise in Particulate Resoles Sample Protective Colloida Gel Timeb Plate Flowb Other Observations 5 CELLOSIZE WP-09 77 sec. 73 min. The best results of this group 6 CMC - - Sintered on standing overnight 7 Soluble Starch - - Sintered overnight 8 Agar - - Sintered overnight NOTES: a. Identity of protective colloids: CELLOSIZE WP-09 - low molecular weight hydroxyethyl cellulose (UCC).

CMC - carboxymethyl cellulose (Hercules).

Soluble Starch - (Merck [Registered Trade Mark] reagent grade).

Agar - polysaccharide derived from seaweed (MC & B). b. All four products handled well during the condensation, on cooling and diluting, and during filtration and drying. On standing, however, all sintered, #5 slowly over a period of several weeks and #6, 7 and 8 on standing overnight.

TABLE VI Materials Showing No Activity as Protective Colloids in Standard Particulate Resole Formula Sample Material 9 EEA-9500, NH4+ salt Ammonium salt of ethylene acrylic acid polymer 10 POLYOX* N-10 Poly(ethylene oxide), 100,000 MW 11 CARBOWAX* 6000 Poly(ethylene oxide), 6,000 MW 12 JAGUAR Guar gum, saccharide of domestic origin 13 Casein Protein 14 Gelatin Protein 15 PVP Polyvinyl pyrrolidone 16 ACRYSOL Polyacrylic acid, Na salt 17 CYANAMER P-250 Polyacrylamide 18 L-77 Silicone fluid 19 L-540 Silicone surfactant 20 L-5340 Silicone surfactant 21 TERGITOL* NP-44 Nonionic surfactant 22 EAA-9500, Na+ salt Sodium salt of ethylene/ acrylic acid copolymer 23 ELVANOL 50-42 Polyvinyl alcohol, 88 per cent hydrolized, high molecular weight (about 96,000) *Registered Trade Mark. None of these experiments yielded a filterable product. - Elvanol 5105 is no longer available commercially. GELVATOL (Registered Trade Mark) 2030 (Monsanto)* and VINOL 205 (Airco)* are commercially available, partially hydrolyzed polyvinyl alcohols that work equally well in the process. About l part per hundred of phenol of these materials is recommended.

EXAMPLE 4.

Using the procedure of Example 3, several nitrogen compounds were evaluated as substitutes for hexa in the formulation: 100 Parts - Phenol 120 Parts - 40% Formalin * GELVATOL 20-30 and VINOL 205 are 88 per cent hydrolyzed polyvinyl alcohols having molecular weights of about 10,000.

(0.257 Equiv.) - Nitrogen Compound 1.0 Part - Gum Arabic (To 50% Conc.) - Water The results of this study are summarized in Table VII.

TABLE VII Hexa Replacements in Particulate Resoles Samples N-Compound Parts Results 1 29% Aq. NH3 15.0 Equivalent to hexa in all respects.

2 40% CH3NH2 19.9 Filterable but sintered on storage; has high gel time, long flow.

3 40% (CH3)2NH 28.8 Filterable but sintered on storage; high gel time, long flow.

Nitrogen found: 2.59%, calc d: 2.77%.

4 Ethylene Diamine 7.7 Filtered with difficulty; sintered readily; has long flow, normal gel.

5 Monoethanolamine 23.5 Filtered with great difficulty, sintered very quickly.

6 Urea 7.7 Failed to yield a water-insoluble condensate after 2 hours/100 C.

7 p-Phenylene Diamine 13.9 Failed to advance to the water insoluble state.

8 Diethylene Triamine 8.7 Filtered with difficulty, coalesced to a sticky mass, 9 Melamine 10.8 Still water-soluble after 1.5 hours/reflux.

10 Guanidine Carbonate 10.4 No apparent reaction.

11 Dicyandiamide 7.2 Heat-reactive but not filterable.

12 Ammonium carbonate-hydrate 14.6 Liberated CO2 and gave a normal Particulate Resole.

13 Ammonium Acetate 19.8 Liberated at least some acetic acid; Product was not filterable.

14 64% Aqueous Hydrazine 6.4 Slow reaction to an apparent thermoset product.

Some of the lower alkylamines -- especially methylamine and dimethylamine - were surprisingly effective in promoting filterability in spite of their effect in reducing the functionality of the condensation system. NMR analysis indicated that nitrogen was predominantly in the form of a benzyl tertiary amine. As a class, the alkylamines showed the plasticizing effect of the alkyl groups in a tendency to sinter on storage and, in extreme cases, in fusion during isolation. The use of a polyamine did not correct this nor did an arylamine. Neither urea nor melamine proved to be a useful modifier but, rather, appeared to foster water solubility. Only water-soluble products were made when urea/formalin/hexa/gum arabic or melamine/formalin/hexa/gum arabic systems were condensed in aqueous media.

The finding that substitution of ammonia NH, with any organic groups detracted from performance suggested that only the other nitrogen hydride, hydrazine, H2N-NM2 might function as a hardening modifier. At the concentration noted above, however, a phenol/formalin/hydrazine reaction produced a water-insoluble product, but it could not be brought to a thermosetting stage in a reasonable time.

Both methylamine and dimethylamine produced a very fast initial reaction as judged by the exotherm and by testing for hotplate gel time - but the isolated product in each case had a longer gel time and a longer flow than expected on the basis of standard particulate resole experience. This behaviour was explained as a consequence of the lower functionality of these amines compared to ammonia.

This reasoning prompted the consideration of ammonia/dimethylamine (or hexa/dimethylamine) blends as a route to increasing flow. Some encouragement was found for this idea, but increased flow was invariably accompanied by lessened sinter resistance.

EXAMPLE 5.

CPC analysis data indicated that the particulate resoles are fully comparable to conventional resoles in molecular weight, in spite of the very mild conditions of the process -- about an hour at 85"C., diluted with at least an equal volume of water, and at a pH of about 7. Samples of particulate resoles and of a general purpose conventional resole containing 6 parts hexa were submitted for NMR analysis. The analyses, summarized in Table VIII, show that the particulate resoles have high tribenzylamine and methylene bridge contents and somewhat lower methylol and methylene ether contents. It was concluded that the subtle differences in composition between the conventional resole and the standard (9part hexa) particulate resole must be at least partially responsible for the difference in sinter resistance.

TABLE VIII Methylene Distribution in Resoles; Summary of NMR Analysis Methylene Per Hundred Moles Phenol Conventional Particular Resoles of Resole Invention (METHYL AMINE) Methylol 38 30 37 28 Hemi-formal 0 5 3 10 Methylene Ether 10 6 9 2 Methylene Bridge 35 47 39 48 Tert-Benzylamine 40 41 51 59 Sec-Benzylamine 3 4 0 3 Oxazine 1 4 11 0 127 137 150 150 RESOLES STUDIED: Conventional resole, containing 6 parts hexa and a formaldehyde factor yielding 1,27 methylene/phenol; Particulate Resole based on "standard" formula; hexa reduced to 6 parts; and CH3NH2 for NH3 of standard.

STRUCTURES:

METHYLOL HEMIFORMAL METHYLENE ETHER

METHYLENE TERT BENZYL- SEC. BENZYL- BENZOXAZINE AMINE AMINE EXAMPLE 6.

Using the procedure of Example l, a series of particulate resoles were made with different formaldehyde proportions. The proportion of reactants, and yields, gel times, and plate flows of the products are shown in Table IX: TABLE IX The Effect of Formalin Factor Formula Weights Usedb Product 150 C 125 C.

Sample Form Fact.a Phenol 40% CH2O HEXA H2O G.A.d Yield Gel Plate Flow 1 70 200 140 18 292 2.0 220 89 100 2 90 200 180 18 292 2.0 222 81 74 3 110 200 220 18 292 2.0 225 72 62 4 120 200 240 24 292 2.0 240 78 49 5c 110 200 220 18 292 2.0 230 65 6c 110 200 220 18 (c) 2.0 254 58 NOTES: a. Formaldehyde factor, parts of 40% CH2O per 100 parts phenol. b. All reactions carried ont in one-liter flasks. c. In this pair of experiments the first wash (2470 grams) of Exp. 5 was concentrated to 330 grams by boiling off water.

The concentrate was substituted for make-up water of Exp. 6. d. Gum arabic.

EXAMPLE 7.

The formulation of Table X was used to produce an example of a bisphenol-A one-step resin in particulate form. The procedure used was analogous to that described in earlier Examples. As in the case of phenol, reaction was at 85 C. and a 70-second gel time was chosen as the end point. The recovery of the solid product proceeded exceptionally well. The product was not further characterized than noted in Table X.

TABLE X Bisphenol-A One-Step Resin Formula 200 grams Bisphenol A 160 40% Formalin 16 Hexa 320 Water 2.0 Gum Arabic: Characterization 78 sec. Gel Time (1500C.) ?aMM. Plate Flow (125 C.)- NOTE: a. The plate flow value could not be determined because the sample did not adhere to the glass plate but slid off. The product, how ever, did flow at 1250C.

EXAMPLE 8.

Particle Size Distribution The particle size distribution of several particulate resoles was determined, using a Sharples Micromerograph (Registered Trade Mark). (A micromerograph is an instrument that is used to measure the particle size distribution in the 1 to 100 micron range by a sedimentation technique.) The particulate resoles were made by the technique of Example 3, using the standard formulation. The particulate resoles of this invention were compared with a commercial pulverized one-step/two-step phenolic resin. The results are displayed below in Table XI: TABLE XI Particle Size Distribution 2 parts hydroxy Diameter, I part gum propyl guar part Commercial Microns arabic gum PVAa Resin < 100 92% 96% 96% 100% < 50 80 93 92 100% < 30; 60 69 74 92 < 20 48 50 55 64 < 15 32 36 38 40 < 10 14 18 18 12 < 5 3 4 2 4 Average 21 20 18 18 Diameter, Microns a. Polyvinyl alcohol, 88 per cent hydrolyzed, 10,000 molecular weight.

EXAMPLE 9.

R esorcinol-based Particulate Resole Because of the position of its hydroxyl groups, resorcinol is an especially reactive phenol. The successful preparation of a resorcinoUformalin/hexa resole and its isolation as a particulate solid is evidence for the excellent product control inherent in the Particulate Resole process. The following charge was placed in a 1000-ml flask: 200 grams Resorcinol 200 Formalin (40% aqueous CH2O) 20 Hexa 2 Gum Arabic 146 Water The resorcinol was predissolved in the water and the other reactants added. An exotherm carried the temperature to 550C. Five minutes later, at 600 C., the batch clouded. Checking the dispersion on the 1500C. hotplate it was noted that the gel time was already very short. The batch, therefore, was cooled, poured into 1000 grams of water, filtered and dried at ambient temperature. The dry product was a fine red-brown powder. When tested on the 150"C. hotplate, the powder fused and then very quickly set to a film which had no flow; gelation was too rapid to estimate a gel time. As in the case of conventional phenolics, resorcinol would find practical utility as a minor modifier for a phenol-based Particulate Resole.

EXAMPLE 10.

The use of Paraform as a source of formaldehyde in the preparation of our standard particulate resole The following were weighed into a 1000-ml 3-neck creased flask: 200 grams Phenol 80 Paraform (90% CH2O) 18 Hexa 2 Gum Arabic 284 Water The batch was heated to 850C.; a clear solution formed at 750C. Nine minutes after reaching 85"C., the batch clouded as the insoluble resin stage was reached. After an hour at 85"C., a gel time of 127 seconds on the l500C. hotplate was measured; a 97 second gel time was recorded at 85 minutes. At 87 minutes thereaction was stopped by cooling to 300 C. These gel time data indicate that the reaction was slightly slower than when aqueous formalin is used; it was assumed that the extra time was required to hydrolyze the paraform to its reactive form of methylene glycol .

Work-up of the product was by our standard methods of diluting with 1000 grams of water, filtering, and drying. The dry powder showed at 1500C. hotplate gel time of 76 seconds and a 1250C. plate flow of 76 mm. The product appeared to be equivalent to that made from 40% formalin.

On the other hand, trioxane did not react with phenol plus hexa and gum arabic in an aqueous medium; it was concluded that the tri(oxymethylene) structure was too stable to yield formaldehyde under these reaction conditions. In another experiment, phenol, paraform and hexa were reacted at reflux in (95%) ethanol; reaction did occur but at a very slow rate. In another variation, phenol, formalin and hexa were reacted in acetone as the solvent; actually, the reactions medium was a mixture of acetone and water. Again, a very slow rate was noted. It was concluded that the aqueous medium of the Particulate Resole process plays a very specific role in the fast reaction rates found in this process.

EXAMPLE 11.

Hexa versus fixed alkali catalysts The hexa in our standard Particulate Resole formula serves two purposes, acting as a catalyst to speed up the resin-forming reaction and as a modifier of the product made, insuring filterability from the aqueous medium and also playing a role in the sinter resistance of the dry powder. The latter effect is thought to be the result of a lesser tendency of hexa-modified phenolics to absorb atmospheric moisture, thereby aiding sintering.

In spite of the fact that hexa must be a very weak base, hexa has been shown to produce a resole resin at a much more rapid rate than the strong base NaOH. The following experiment (using the standard procedure) established the equivalency of hexa and aqueous ammonia plus formalin: 200 grams Phenol 240 Formalin (40% CH2O) 30 29% NH3 (aq.) 220 Water 2 Gum Arabic After 80 minutes at 85"C., a 75-second gel time (150"C. hotplate) was found; workup gave a product indistinguishable from that made with hexa.

In the following experiment, the above formula was used again except that the ammonia was replaced with enough NaOH to ionize 5% of the phenol and to raise the pH to about 8 (compared to 7.2 with ammonia or hexa): 200 grams Phenol 240 Formalin 8 25% NaOH (aq.) 260 Water 20 Vino1205 This batch was also held at 850C., and sampled for the hotplate gel test: At 1.0 hour 186 secs. at 1500C 1.5 ,, 158 2.0 ,, 144 2.5 ,, 129 3.0 ,, 125 3.5 ,, 111 3.75 ,, 90 4.0 ,, 72 Also, in contrast to the ammonia/hexa catalysis, precipitation first occurred at 3.5 hours (instead of < 5 minutes). After this 4-hour reaction, the dispersion was acidified to a pH of 5.8 with dilute phosphoric acid; acidification resulted in breaking of the dispersion, so that the product was in the form of a soft lump of resin, rather than a particulate solid.

In another experiment, barium hydroxide was employed with polyvinyl alcohol substituted for gum arabic: 200 grams Phenol 222 Formalin (40%) 5 Barium Hydroxide Monohydrate 20 Gelvatol 20-30 132 Water The phenol, formalin, and barium catalyst were held at reflux for 30 minutes, after which the colloid and water was added. After acidification with dilute H2SO4 to a pH of 5.0-5.5, the batch was held at 850C. for a 1500C. hotplate gel time of 90-100 seconds. This required 3.3 hours. After cooling to 250C. and stopping the stirring, the "emulsion" separated into two phases, a resinous and an aqueous phase. Thus, the PVA is not a suitable protective colloid for the dispersion which, instead, uses 2 parts gum arabic (per 100 parts phenol) plus 0.5 part guar gum (a thickener), as is taught in the Harding patent, U.S. Patent No. 3,823,103. On the other hand, fine particle size emulsions have been made with 5 or more parts of polyvinyl alcohol when the resole formula contained hexa as modifier/catalyst. The latter is illustrated by the following experiment: 200 grams Phenol 180 Formalin (40% CH2O) 18 Hexa 20 Vinol 205 184 Water A 70-minute reaction at 85 C. was sufficient to reach a 71-second gel time. That this amount of PVA had produced a very fine dispersion was evident from the intense whiteness of the emulsion, from complete inability to filter the solid product, from a very limited amount of settling during a centrifugation attempt and from inspection under the microscope. The latter showed the largest particle size to be about 2 microns in diameter, with the bulk of the product at 1 micron or less.

In another experiment NaOH was added to the standard formulation: 200 grams Phenol 180 Formalin (40%) 8 25% NaOH (aq.) 18 Hexa 2 Gum Arabic 184 Water When the above reactants were held at 850C., a cloud was noted only after 40 minutes: this did not develop into the normal precipitation of a water-insoluble product even though held for another hour at 850C. Gel Times also fell at a very slow rate. Acidification of the NaOH to a pH of 7 resulted in the separation of two liquid layers. Thus, the addition of NaOH to the standard formula inhibited the resin-forming reaction; this behaviour has not been explained.

There is no advantage to a particulate solid if this particulate nature is not maintained on storage. Thus, although a conventional phenolic resole can be ground to a pulverized resin, such a material is not offered commercially because the ground resin will maintain its particulate form at ambient temperature for only a limited period of time. The process by which a powder fuses to a larger mass, thereby losing its particulate nature, is known as sintering.

Conventional phenolic resoles are amorphous solids that have softening temperatures a bit above room temperature. More precisely, these resins have glass transition temperatures in the neighborhood of 30"C. The Particulate Resoles of this invention have shown glass transition temperatures as high as 42"C. when thoroughly dry. These resins have been found to resist sintering even at 400---450C.

Furthermore, the Particulate Resoles of this invention have been dried at 600 C. in a fluidized bed drier (that is, at zero-gravity conditions) with no apparent loss of their particulate character. This "extra" protection has been ascribed to the colloid used in the resin manufacture. It has also been shown that incomplete drying of the Particulate Resole will lower the glass transition temperature and adversely affect resistance to sintering. We believe that the water content, as measured by Karl Fischer analysis, of the dry resins should be kept at 1.0% maximum and preferably at 0.5% maximum.

We have devised an Accelerated Sinter Test for the Particulate Resoles which employs a test tube having an inside diameter of 23 mm and a cylindrical brass insert having a diameter of 21 mm and weighing 160 grams. A 2 to 3 centimeter height of powder is poured into the test tube, the weight inserted, and the assembly placed in a 40"C. or a 32"C. oven in an upright position, usually overnight. (These dimensions and weight translate to a force of 100 pounds per square foot or approximately the weight of a stack of four bags of resin.) On removal from the test the resin powder is examined and one of three classifications selected: Sinter-Resistant -- Tapping the test tube is sufficient to restore the free-flowing character of the dry powder.

Slight Sintering -- A very light crust just below the weight is easily broken, exposing the original free-flowing powder.

Sintered -- Some degree of fusion has occurred so that the material no longer possesses its particulate nature;

TABLE XII Sinter Test Results Formulation, Parts by Weight 125 C. 150 C. Sinter Test Plate Flow, Gel time, Run No. Phenol(a) Formalin Hexa Colloid(b) mm. sec. 40 C. 32 C, Control(c) - - - - 56 58 Failed Failed 1 100P 90 9 1 58 90 Failed Passed 2 90P 90 9 1 58 75 Sl. Sint, Passed 3 80P 90 9 1 40 80 Sl. Sint. Passed 4 20B/80P 90 9 1 62 80 Failed Passed 5 100P 96 7 1 47 72 Failed Passed 6 80P 90 9 1 133 101 Failed Passed 20t 7 100P 80 9 1 69 80 Sl. Sint. Passed 8 80P 90 9 1 64 78 Failed Passed 20T 9 100P 90 9 .75 90 84 - Passed 10 100P 90 9 .75 115 79 - Passed 11 100P 90 9 1 40 77 Passed 12 100P 90 9 1 52 90 Passed 13 100P 90 9 1 19 55 Passed (a) P=phenol; B=bisphenol-A; T=p-t-butyl phenol (b) VINOL 205, except Run No. 11, which was gum arabic (c) Commercial Pulverized one'-step, two-step phenolic resin The Particulate Resoles of this invention have been noted to possess plate flow values which are somewhat lower than might be required for specific applications.

Although certain processing parameters, such as reaction time, can be manipulated to increase plate flow, greater flow has been accompanied by loss of sinter resistance, a valued property of the Particulate Resoles. It has been found that both increased flow and sinter resistance may be obtained if a small fraction - 2 to 25% by weight (preferably 5 to 10%) - of trifunctional or tetrafunctional phenol is replaced with a less functional phenol., that is an ortho- or para-substituted phenol.

Some examples of suitable, commercially available difunctional phenols are:

p - Cresol CK3 9 0H OH o - Cresol g HH3 3 p-tertiary-Butylphenol (CH3)3 C z-OH p-Phenylphenol < - OH H3 p-Cumylphenol 3 p-Nonylphenol C9H e OH p-Octylphenol C' 8h7 + H p-Dodecylphenol C12 H 25 Styrylphenol t CHH CH 3 Some Examples of monofunctional phenols include:

2,4-Xylenol 9H s OH OH3 2,6-Xylenol CH t H3 4R 2,6-Di-t-butylphenol (CH3)3- )C-(CH ) 2,4-Di-t-butylphenol OR (CH3)3- C C(CH ) 33 Table XIII summarizes the results found when 2,5 or 10 weight % para-tertiarybutylphenol was incorporated into a Particulate Resole. Up to a 7-fold increase in plate flow was realized, sinter resistance on storage was retained, but the average particle size was increased to an undesirable level. The latter was eliminated when a large amount of water was omitted from the formulation, raising the dispersion "solids" during reaction from 41% to 52%, 60% or 66%. The higher reaction solids is of interest not only to control flow but also because of the potential increase in still productivity (as high as 60%) and its favorable effect on economics. This formulation change, therefore, has also been extended to short-flow as well as longflow Particulate Resoles.

Table XIV summarizes the result of a series of three preparations of short-flow Particulate Resole made at a calculated 66% concentration of resin product and of three preparations of long-flow Particulate Resole, together with characterization data. The data shown are the averages of each set of three runs plus the calculated standard deviations.

Table XV is similar to Table XIV except that the condensations were carrried out at 52% solids.

As noted above, ortho- and para-alkyl- or aryl-substituted phenols may be used to realize higher flow Particulate Resoles. The examples given in Tables Xl II, XIV and XV were limited to para-tertiary-butylphenol because of its ease of handling, ready availability and high purity. Cresols were also tested experimentally. These cresols included a pure ortho-cresol and an industrial grade cresol which is a blend consisting of about 45% meta-cresol plus phenol (that is, the tri-functional fraction) and about 55% ortho- and para-cresol (the difunctional fraction).

EXAMPLE 12.

190.0 grams Phenol 10.0 Cresol Blend 180.0 40% Formalin 18.0 Hexamethylenetetramine 3.0 Vinol 205 184.0 Water TABLE XIII Para-Tertiary-Butylphenol as a Particulate Resole Modifier

Ratio of Run Phenol:p-t- Reaction No. Butylphenol Time 150 Gel 125 P.F. H2O Phenol #n #w #w/#n Size 1 90/10 80 min. 85 sec. 132 mm. 1.29% 6.0% 40.3 76.6 1.90 coarse 2 90/10 90 69 79 0.54 5.5 42.4 86.6 2.04 coarse 3 95/5 80 65 86 2.0 5.6 43.1 87.8 2.04 coarse 4 95/5 80 68 81 1.14 5.5 44.4 89.6 2.02 coarse 5 98/2 80 67 65 - - - - - fine coarse 6 100/0 80 53 18 0.71 5.4 43.2 87.9 2.04 fine TABLE XIV Short and Long Flow Particulate Resole at 66% Solids: Formulas and Characterization Data

Product Type Short Flow Long Flow Formula -Phenol 100. Parts 95. - Parts t-Butylphenol - 5.0 " 50% Formalin 72. " 72. " Hexa 9.0 " 9.0 " VINOL 205 1.0 " 1.0 " Reaction Cycle 850C/65 Minutes 850C/65 Minutes Characterization 1500C Gel Time 71 # 11 sec. @ 105 + 6 sec 125 C Plate Flow 44 # 8 mm 88 # 10 mm Moisture Content 0.33 # 0.12% 0.0 # % Free Phenol 5.4 # 0.1% 5.5 # 0.2% #n 42.2 # 1.5 39.7 # 0.7 #w 87.3 # 5.7 75.8 # 3.0 #w/#n 2.07 # 0.06 1.91 # 0.04 Sinter Test at 350C Passes Passes TABLE XV Long and Short Flow Particulate Resole at 52% Solids:Formulas and Characterization Data

Product Type Short-Flow Long Flow Formula Phenol 100. Parts 95. Parts t-Butylphenol - 5.0 ", 40% Formalin 90 " 90. ,, Hexa 9.0 " 9.0 " VINOL 205 1.0 " 1.0 " Water 30.0 " 30.0 " Reaction Cycle 850C/65 Minutes 850C/65 Minutes Characterization 1500C Gel Time 65 # 2 sec 88 # 1 sec 1250C Plate Flow 40 # 8 mm 98 + 4 mm moisture Content 0.0 # % 0.33 # 0.32% Free Phenol 6.1 # 0.4% 6.4 # 0.2% #n 40.9 # 0.6 39.1 # 0.2 #w 84.0 # 2.4 77.8 # 3.3 #w/#n 2.05 # 0.04 1.97 # 0.05 Glass Trans. Temp. 40 + 30C 37 # 3 C Sinter Test at 35 C Passes Passes The above charge in a 1-liter flask was heated to 85 and held for 80 minutes past the cloud point. The product was worked up by cooling, diluting with 1000 grams of water, filtering and drying. Measured; 22 mm plate flow, 43 sec gel time at 150 C.

The increase in flow is low because of the 45% trifunctional content of the cresol blend.

EXAMPLE 13.

In the next preparation, the cresol blend was doubled in the charge shown below: 180.0 g. Phenol 20.0 " Cresol Blend 180.0 ,, 40% Formalin 18.0 " Hexamethylenetetramine 2.0 " Vinol 205 184.0 " Water The batch was held at 850C. for 80 minutes after the cloud point. It was then cooled, water-washed, filtered and dried. Plate flow was 88 mm and gel time at 1500C. was 68 sec.

EXAMPLE 14.

A highly purified ortho-cresol was obtained from the Aldrich Chemical Company was used in the charge below: 190.0 g. Phenol 10.0 ,, o-Cresol 180.0 ,, 40% Formalin 18.0 ,, Hexamethylenetetramine 2.0 ,, Vinol 205 184.0 ,, Water After an 80-minute reaction and work-up in the standardized manner, the product showed a plate flow of 63 mm and a gel time of 65 sec.

EXAMPLE 15.

291.0 g. Phenol 9.0 ,, 2,6-di-t-Butylphenol 216.0 ,, 50% Formalin 27.0 ,, Hexamethylenetetramine 3.0 ,, Vinol 205 The above charge in a 1-liter flask was heated to 850C. and held at 850C. for 65 minutes beyond the cloud point. The product was washed up by cooling and diluting with 1500 grams of water. The product was filtered and dried. The measured plate flow was 120 mm. and the gel time at 1500C. was 83 seconds. The product passed the sinter test at 360 C.

Claims (20)

WHAT WE CLAIM IS:

1. A process for producing a particulate solid, heat-reactive, substantially water-insoluble resole, which process comprises the steps of: (a) reacting formaldehyde, a phenol comprising a trifunctional phenol, a tetrafunctional phenol, or a mixture thereof, and an effective amount (as hereinbefore defined) of ,an amine compound consisting of hexamethylenetetramine, a compound containing amino hydrogen (as hereinbefore defined), or a mixture thereof, in an aqueous medium in the presence of an effective amount (as hereinbefore defined) of a protective colloid which colloid is present at the beginning of the reaction, at a temperature of from 70" to 100"C, for a period of time sufficient to produce an aqueous dispersion of a particulate solid, heat-reactive (or hereinbefore defined), substantially waterinsoluble, thermosetting resole; (b) cooling the reaction mixture to a temperature below 40"C. on completion of the reaction; and (c) recovering the resole from the aqueous dispersion.

2. A process as claimed in claim 1 wherein the amine compound is ammonia.

3. A process as claimed in claim l or claim 2 wherein the phenol is monohydroxybenzene, 2,2-bis(4-hydroxyphenyl)-propane, m-cresol, resorcinol, or a mixture thereof.

4. A process as claimed in claim 1 or claim 2 wherein the phenol is a mixture of a trifunctional or tetrafunctional phenol and less than about 50 mole% of a difunctional or monofunctional phenol, based on the total amount of phenol present.

5. A process as claimed in claim 4 wherein the trifunctional phenol is monohydroxybenzene and the difunctional phenol is p-t-butylphenol.

6. A process as claimed in claim 4 wherein the trifunctional phenol is monohydroxybenzene and the difunctional phenol is para- or ortho-cresol.

7. A process as claimed in claim 4 wherein the tetrafunctional phenol is 2,2bis(4-hydroxyphenyl) propane.

8. A process as claimed in any one of the preceding claims wherein a minor molar proportion of the phenol used comprises a para-substituted phenol.

9. A process as claimed in any one of the preceding claims wherein the formaldehyde is in the form of formalin.

10. A process as claimed in any one of the preceding claims wherein the protective colloid is gum arabic, gum ghatti, hydroxyalkyl guar gum, or partially hydrolyzed polyvinyl alcohol.

11. A process as claimed in any one of the preceding claims wherein the protective colloid is present in an amount of about l weight percent, based on the weight of the phenol.

12. A process as claimed in any preceding claim wherein the aqueous dispersion contains from 30 to 50% by weight of water.

13. A process as claimed in any one preceding claims wherein step (a) is carried out at a temperature within the range of from 75" to 900 C. to a predetermined end point, whereupon the aqueous dispersion is rapidly cooled at a temperature below 40"C.

14. A process as claimed in any one of the preceding claims wherein the resole is recovered by diluting with water at from 200 to 300 C., filtering or centrifuging the resole, and drying the filtrate or centrifugate, optionally with agitation.

15. A process as claimed in claim 14 wherein drying is effected in a fluidized bed.

16. A process as claimed in claim 1 substantially as hereinbefore described.

17. A process as claimed in claim 1 substantially as hereinbefore described in any one of the specific Examples.

18. A resole when produced by a process as claimed in any one of the preceding claims.

19. A resole as claimed in claim 18 having an average particle size of less than 100 microns in diameter.

20. A bonded fibrous material, a molded article or an adhesive composed of or containing a resole as claimed in claim 18 or claim 19.