This invention relates to chlorination, particularly to chlorination of organic

5 compounds having acidic protons. The invention also relates to subsequent reaction of certain chlorinated products, more particularly the use of these products to alkylate aromatic compounds.

Products of compounds such as fluorene in which the acidic protons have been replaced by chlorine undergo alkylation reactions which are useful in preparing other

10 compounds such as bis(hydroxyphenyl)-fluorenes and bis(amιnophenyl)fluorenes and other functionally substituted aromatic compounds which in turn find applications as monomers for high performance polymers It is particularly important that the chlorination be specific to produce the desired isomers, preferably in the substantial absence of other chlorination products. For instance, in the case of chlorination of fluorene to produce 9,9 dichlorofluorene

15 it is very important that chlorination of the aromatic rings bo avoided

Ida Smed ley reported a preparation of 9,9-dichlorofluorene in 1905 from heating fluorenone and a slight excess of phosphorus pentachlorιde (J. Chem. Soc. 87, 1249 (1905). Smedley did not report a yield, but did mention that the product contained fluorenone and required recrystallization from benzene Ray et al. repeated Smedley's method and attained a

20 66 weight percent yield. (J. Amer. Chem. Soc, 70, 1954 (1948).) Ray et al report that the pure product is quite stable if protected from moisture, but that samples of impure material decomposed within a week to give a sticky green-yellow mass with the sharp odor of hydrogen chloride.

Chlorinations of compounds having acidic protons and a molecular structure

25 which can delocalize the electron density of the conjugate base such as fluorene using common chlorination agents such as chlorine, sulfuryl chloride, N-chlorosuccinimide and phosphorus pentachloride are generally disadvantageous because the products of such reactions exhibit substitution on the aromatic rings, instead of substitution of the acidic protons Therefore, it is

notfeasibleto prepare 9,9-dichlorofluorene from fluorene using conventional chlorination technology.

A procedure for the preparation of 9,9-dichlorofluorene directly from fluorene, without fluorenone as an intermediate, was reported by Reeves et al. in Israel J. Chem 26, 225, (1985). Tetrabutylammonium bromide was used as a phase transfer catalyst to chlorinate such compounds as fluorene, phenyipropanone, acetophenone, 1-chloroacetophenone, p-methoxyacetophenoπe, benzoin ethyl ether, p-nitroacetophenone, deoxybenzoin, and xanthene using carbon tetrachloride in an organic phase as chlorine source with an aqueous hydroxide phase. It was reported ^' that use of potassium carbonate in the aqueous phase in the attempted chlorination of p-nitroacetophenone resulted in no reaction. Using this reaction for the chlorination of fluorene. Reeves etal. reported a 51.9 percent yield of 9,9-dichlorofluorene. Other reaction conditions and time for the chlorination of fluorene are not given. A 57 percent yield was reported by Reeves for production of xanthone from xanthene using the procedure. It would be desirable to have a selective process for chlorinating compounds which have acidic protons and a molecular structure which can delocaiize the electron density of the conjugate base such that replacement of the acidic hydrogens is the predominant reaction but in greater yields and/or shorter reaction times than those achieved when following the process reported by Reeves etal.

9,9-Bis(hydroxyphenyl)fluorene is typically prepared from fluorenone such as by reaction of fluorenone with phenol in the presence of such compounds as beta- mercaptopropionic acid and anhydrous hydrogen chloride (P.W Morgan, Macromolecules.3, 536, (1971); or in the presence of zinc chloride and anhydrous hydrogen chloride (U.S. Patent 4,467,122). Alternatively, fluorenone has been converted to 9,9-dichlorofluorene and subsequently reacted with phenol to produce the 9,9-bis(hydroxyphenyl)fluorene, such as by the reactions reported by Smedley in J. Chem. Soc.87, 1249 (1905). All these reported methods involve use of fluorenone to prepare 9,9-bis(hydroxyphenyl)fluorene.

It would be desirable to have a process for preparation of such bisphenols as 9,9-bis(hydroxyphenyI)fluorene without using fluorenone as a reactant because preparation of fluorenone involves loss of starting materials, additional steps which may be time-consuming, and optionally, use of unpleasant starting materials. For instance, fluorene may be converted to fluorenone by use of sodium dichromate to achieve a 60-70 percent yield. A higher yield is reported by AIneri, etal. Tetrahedron Letters, 24,2117 (1977), but requires 24 hours. A multiphase system involving an organic phase, an aqueous sodium hydroxide phase and a catalyst of elemental carbon and phase transfer catalyst has also been reported in U.S. Patent 4,297,514 (K. Ma) but has the disadvantage of handling a solid and separating a product from it. In each instance the fluorenone product must be isolated and purified before subsequent reaction. It would be desirable to avoid such extra steps.

The invention is a process for chlorinating at least one compound having acidic protons and a molecular structure which can delocalize the electron density of the conjugate base comprising contacting said compound with at least one perchloroalkane and aqueous base in the presence of a phase transfer catalyst which is a tetraalkylammonium hydroxide The process of the invention is particularly useful for chlorinating compounds having acidic protons and a molecular structure which can delocalize the electron density of the conjugate base compounds such as fluorene, which with conventional chlorination procedures will not replace the acidic protons with chlorine.

When used to prepare such dichloro compounds as 9,9-dichlorofluorene, the process leads to an especially preferred process of alkyl ating phenols, phenolics, aromatic amines, alkylaromatics and other such aromatic compounds to give valuable polymers, oligomers, intermediates and monomers, particularly, bis(hydroxyphenyl)fluorene.

The process of the invention is useful for chlorinating compounds which have acidic protons and a molecular structure which can delocalize the electron density of the conjugate base. It has been stated by Reutov et. al. (O. A. Reutov, I. P. Beletskaya and K. P. Butin, CH-ACIDS, Pergamon Press, New York, N.Y., 1978.) that "almost any organic compound can ionize in solution to give carbanions, that is, negatively charged species whose charge is totally or more often partially localized on one of the carbon atoms." When certain substituents are part of the hydrocarbon structure and are bonded to a saturated carbon atom which also bears hydrogen atoms, these hydrogen atoms are relatively acidic. Examples of such substituents are unsaturated functional groups such as vinyl, nitro, carbonyl, cyano, sulfone, or phenyl groups. The inductive electron withdrawing ability and the ability of these substituents to delocalize the negative charge remaining when a proton has been removed are responsible for the acidity of these carbon-hydrogen bonds. These compounds are often referred to as active methylene (-CH ₂ -) or active methine (-CH-) compounds. Active methylene compounds are preferred for use in the practice of the invention; more preferred are compounds having an active methylene group adjacent to at least one vinyl, nitro, carbonyl, cyano, sulfone, cyclopentadiene, or phenyl group, most preferably adjacent to at least two such groups which may be the same or a combination thereof. Exemplary of such compounds are fluorene, ring- substituted fluorenes, indene, xanthene, anthrone, phenalene, chromene, acetone, acetophenone, deoxybenzoin, phenylacetonitrile, cyclopentadiene, dihydroanthracene, 1- phenyl-2-propanone, alkylpyridines, alkylpyrazines, alkylquinolines, alkylisoquinolines, alkylquinoxalines, alkylquinazolines, and alkylcinnolines. The process of the invention is particularly useful for compounds for which the replacement of the acidic protons with chlorine is not easy under conventional chlorination conditions including fluorene, indene, xanthene, anthrone and the like, preferably fluorene and its derivatives which are ring- substituted, most preferably fluorene. Such target compounds are unsubstituted or inertly substituted, that is having substituents which do not undesirably interfere with the

chlorination or subsequent reactions. Such substituents include alkyl, halo, nitro, cyano, carboxyl, thio, sulfoxϊde, sulfone, carbonyl, ether, and aryl groups, as well as other substituents not having a hydroxyl, primary or secondary amino, or mercapto group. Preferably the compounds have from 5 to 30 carbon atoms and more preferably at least one aryl group which is preferably carbocyclic, preferably of from 6 to 20 carbon atoms or heterocyclic of from 5 to 20 carbon atoms and at least one oxygen, sulfur, nitrogen, selenium, silicon, or other heteroatom. The target compound is chlorinated by contacting it with a perchloroalkane such as carbon tetrachloride, hexachloroethane, or benzotrichloride as the chlorine source. Carbon tetrachloride is the preferred chlorine source and is used herein to exemplify perchloroalkanes, but not to limit the process thereto. The perchloroalkane is suitably used in any amount which provides sufficient chlorine for the reaction, and may also be present in an amount sufficient to dissolve the compound being chlorinated (target compound). It is, however, unnecessary that there be sufficient perchloroalkane to dissolve the target compound. When the compound to be chlorinated has a low solubility in the perchloroalkane, it is preferableto use a solvent miscible in the perchloroalkane which dissolves significant amounts of the target compound. Preferably the perchloroalkane is used in an amount from 1:1 to 100: 1 based on the molar concentration of reactant (target compound), more preferably from 2: 1 to 50: 1 , most preferably from 2:1 to 10:1 based on the molar concentration of the target compound.

When an additional solvent is used, it is preferably one which is miscible with the perchloroalkane and which dissolves the target compound and, conveniently, is not undesirably affected by the reaction conditions. Such solvents include methylene chloride, ethyl benzene, cumene, chlorobenzene, tetrahydrofuran and the like. Such a solvent is conveniently used in an amount sufficient to obtain the maximum concentration of the target compound but not so little that the product would precipitate from the reaction mixture. The target compound is contacted with the perchloroalkane in the presence of a base strong enough to deprotonate the target compound, that is, capable of forming the conjugate base of the target com pound. Such bases include inorganic and organic hydroxides and any other strong bases compatible with water, preferably alkali metal hydroxides or tetraalkylamrnonium hydroxides more preferably alkali metal hydroxides, most preferably sodium hydroxide. Alkali metal hydroxides are preferred because they have good solubility in water and relatively low equivalent weight. Sodium hydroxide is more preferred because of commercial availability. The base is advantageously in aqueous solution because of ease of removal from product The solution is suitably of a concentration sufficient to promote the reaction at a desirable rate, preferably from 10 percent to 80 percent, more preferably from 20 percent to 50 percent, most preferably from 30 percent to 40 weight percent base in water. Sodium hydroxide solutions of 40 percent and above often result in emulsions which are difficult to handle. The aqueous solution of base and perchloroalkane are suitably present in any ratio sufficient to promote the reaction at a desirable rate. A desirable rate is generally

one sufficient to complete the reaction in the desired time, but insufficient to cause excessive or uncontrollable exothermic heating of the reaction mixture.

Contrary to the teachings of Reeves et al., Israel J. Chem. 26, 225, (1985) wherein a large excess of base was used with a tetrabutylammonium bromide phase transfer catalyst, in the process of the invention it is surprisingly observed that much less than an equivalent of base is needed. The preferable amount of base as a function of the concentration of the target compound is 0.001 to 1000, more preferably from 0.01 to 100, most preferably from 0.1 to 10 molar ratio. Less than a stoichiometric amount of base is preferred because it leaves more room in the reactor to make product and there is less base to dispose of after the reaction. Because the target compound is not sufficiently soluble in the aqueous base, a phase transfer catalyst is used. Surprisingly good yields and low reaction times are noted when the phase transfer catalyst is a tetraalkylamrnonium hydroxide such as tetrabutylammonium hydroxide, tetramethylammonium hydroxide, tetraethylammonium hydroxide, tetrapropylammonium hydroxide, benzyltrimethylammonium hydroxide, tributylmethylammonium hydroxide and the like, or a tetraalkylphosphonium hydroxide such as tetrabutylphosphonium hydroxide, tetrapropylphosphonium hydroxide, (together tetraalkylonium hydroxide), preferably the phase transfer catalyst is a tetraalkylamrnonium hydroxide wherein all alkyl groups have from 1 to 20 carbon atoms and are non-aromatic, more preferably the tetraalkylamrnonium hydroxide is tetrabutylammonium hydroxide or tributylmethylammonium hydroxide, most preferably tetra-n-butylammonium hydroxide because this catalyst brings the reaction to completion in the shortest time with the least amount of catalyst relative to the target compound.

The phase transfer catalyst is suitably present in any amount sufficient to give a desired rate of reaction, advantageously at least 0.0001 mole ratio, preferably from 0.0001 to 1 , more preferably from 0.001 to 0.1, most preferably from 0.001 to 0.05 molar ratio based on the number of moles of the target compound because this amount gives an acceptable rate of reaction and using more generally costs more and makes purification of the product more difficult. While the hydroxide phase transfer catalyst is optionally admixed with other phase transfer catalysts, for example the halide salts, the phase transfer catalyst is preferably present in the hydroxide salt form in at least the concentrations noted.

The concentrations indicated are preferably that of the phase transfer catalyst in the form of the hydroxide salt. Although theoretically phase transfer catalysts having other anions could convert into the hydroxide when hydroxide ions are present, such a conversion is not generally observed. Quaternary ammonium salts are true ionic species in aqueous media and behave as salts much like the alkali halides ^τ hus, in an aqueous solution, the quaternary ammonium salts are present as ion pairs and freely undergo ion exchange with other ions in solution. When a quaternary ammonium salt is employed as a catalyst in a two phase reaction system, the concentration of that salt in the separate phases is dependent on the relative

solubility of the salt in each phase. When phase transfer catalysts having two or more types of anion are present, the relative concentrations of quaternary ammonium salts may be calculated from the extraction constants for the salts and solvents of interest. Extraction constants (EQX) for systems Q+aq + X ^~ _a q = QXorg (concentrations of quaternary ions in aqueous phase + anion in aqueous phase is in equilibrium with quaternary salt in organic phase) are defined by

E _Q χ = (QX)org/CQ ⁺ )aq (X-)aq

where Q+ is the quaternary ammonium cation, X- is the anion, and QX is the quaternary ammonium salt of interest. When two ions are present, the extraction equilibrium K defined as

K = EQY/EQX = (QY)org(X-)aq / .QX)org(Y-)aq

(where X-and Y-are different anϊons) defines the relative amounts of the quaternary ammonium saltsthatwill be present in the organic phase. DehmLow (E.V. DehmLow, M. Slopianka, and J. Heϊder, Tetrahedron Lett., 1977, 2361.) has measured this value by equilibrating tetra n-butylammonium chloride with 50 percent NaOH solution and found that of the tetra n-butylammonium ion present in the organic phase only 4.2 percent was present as the hydroxide form, the remainder being present as the chloride. The extraction constants for tetra n-butylammonium chloride (EQX = 1.00), tetra n-butylammonium bromide (EQX = 48.5) and tetra n-butylammonium hydroxide (EQX = 0.01) are reported by Gustavi (K. Gustavi and G. Schill, Acta Pharm. Suec.,3259, (1966) in A. Brandstrom, Principles of phase transfer catalysis by quaternary ammonium salts, in "Advances in Physical Organic Chemistry," Vol. 5, V. Gold, Ed., Academic Prss, London and New York, 10977, page 281). Using these values we may calculate that when tetra n-butylammonium bromide is equilibrated with 50 percent NaOH only 0.086 mole percent of the tetra n-butylammonium ion present in the organic phase is present as the hydroxide form, the remainder being present as the bromide.

Thus, since the hydroxide form of the phase transfer catalyst is observed to be, very surprisingly, more effective than the bromide form, it is evident that the hydroxide form is preferably present in a concentration greater than 0.09 mole percent, more preferably greater than 0.1 mole percent, most preferably greater than 1 mole percent of the total phase transfer catalyst. Even more preferably at least 10 percent of the phase transfer catalyst present is in the hydroxide salt form. Also, the chlorination process of the invention preferably takes place in the presence of insufficient bromide or other ion that would extract the tetraalkylamrnonium ion into the organic phase to reduce the concentration thereof in the aqueous phase below that concentration achieved by the hydroxide salt at the preferred concentrations. Most preferably, the reaction takes place in the substantial absence of bromide ion -that is in the absence of added bromide anion.

Conveniently, the compound to be chlorinated is dissolved in the perchloroalkane, to which are added the aqueous base and phase transfer catalyst either sequentially in either order, simultaneously but separately or in admixture to form a reaction mixture. This order is convenient because it is observed that the solution of the compound in perchloroalkane is conveniently purged, for example with an inert gas such as nitrogen, helium, argon, neon, or hydrogen to remove oxygen to avoid production of an oxidized target compound as a by-product. Alternatively, the reagents are suitably mixed in any order such that all reactants are present at one time. The reaction mixture is preferably agitated by any means effective to maximize the surface area of the immiscible phases so that the reactants in each phase are repeatedly brought together.

When ketone products are not desired it is often preferable to exclude oxygen from the reaction. Oxygen is suitably excluded by any means within the skill in the art such as by maintaining a nitrogen blanket over the reaction mixture, such as by nitrogen sparging. Other inert gasses or the vapors of highly volatile organic compounds may be employed. Any reaction conditions under which the chlorination takes place are suitable, but preferred temperatures are from 0°Cto 100°C, more preferably from 15°Cto 80°C, most preferably from 25°C to 40°C because at these temperatures the reaction proceeds rapidly and there is little degradation of the catalyst. Any effective pressure is suitable, at or near atmospheric pressure is generally convenient. High pressure is not harmful. Lower pressures are limited by the vapor pressures (boiling points) of the solvents employed.

Good mixing is important for rapid reaction. For instance at a mole ratio of sodium hydroxide to fluorene of 10: 1 ; mole ratio of tetrabutylammonium hydroxide to fluorene of 0.02: 1.0; mole ratio of carbon tetrachloride to fluorene of 2: 1 and 25 weight percent fluorene in methylene chloride at 30°C for the indicated times, the following table indicates the importance of stirring on yield of 9,9-dichlorofluorene (9,9-DCF).

Thus, for relatively shorter reaction times, relatively faster mixing is preferred. While mixing is difficult to quantify, in a situation with relatively constant viscosities, power per unit volume (watts per liter) is indicative of the amount of mixing. These values were obtained using a Lightnin™' LabMaster ll" ^* Model TSM2010 Mixer commercially availablefrom Mixing Equipment Company, Avon Division, a unit of General Signal which directly measures the watts input into the mixer. Thus, in the practice of the invention, mixing preferably involves use of at least 0.8 W/L, more preferably at least 15.0 W/L, most preferably at least 100 W/L. Such mixing is suitably accomplished by any means within the skill in the art such as by rotary, static (for example recirculating, for example by pump) or other mixing.

The reaction is preferably carried out using non-metallic vessels and equipment, that is not having exposed metals, because metals such as iron (including steel, even stainless steels such as those designated as 304 or 316 stainless steel), nickel and titanium are observed to inhibit the reaction. The term non-metallic vessels and equipment is used to include vessels and equipment lined with non-metallic materials such as polymers (including plastics, resins and glass). Thus the reaction preferably occurs in the substantial absence of such metals, that is

in the absence of sufficient metal to undesirably inhibit the reaction, more preferably in the absence of other than incidentally present (not deliberately added) metals particularly iron, including 304 stainless steel and 316 stainless steel. These metals are believed to inhibit the tetraalkylamrnonium hydroxides; thus use of additional tetraalkylamrnonium hydroxide to replace that which is inhibited permits reaction in the presence of metals.

The product can be isolated by means within the skill in the art, preferably by washing the solution with water to remove catalyst, then evaporating the solvent. Products are usually solids and are optionally purified by crystallization.

The reaction is allowed to go to a predetermined degree of completeness, advantageously to completion as determined by cessation of an increase in concentration of product. At temperatures such as 30°C, completion is observed after 1 minute to 3 hours depending on catalyst concentration, caustic concentration, and degree of agitation or mixing.

The catalyst (tetraalkylonium hydroxide) and/or the base (inorganic or organic hydroxides) are, optionally, conveniently recycled to prepare gem dichloro compounds through many reaction cycles with no loss in efficacy. The catalyst is easily recovered from the reaction mixture after completion of the reaction by means known to those skilled in the art, such as extraction with water or other immiscible solvent having good solubility for the tetraalkylonium hydroxide, or alternatively by contacting the reaction solution with an acidic ion exchange resin to retain the catalyst as a salt followed by regeneration of the tetraalkylonium hydroxide by contacting the ion exchange resin with an aqueous hydroxide solution. In either case, the catalyst is conveniently isolated by evaporative removal of the solvent or is simply used without isolation if the concentration and the solvent are appropriate for the desired reaction. Reuse of the base is, for instance, accomplished by phase separation of the organic and aqueous phases after completion of the reaction and admixing or contacting fresh organic reaction mixture with the separated aqueous phase. Catalyst, either fresh or recovered, is then supplied and the reaction repeated. Recycle of catalyst and/or base is a major advantage since it reduces the amount of raw materials needed with corresponding reduction of waste to dispose.

When the chlorination process is used to prepare a dichloro compound, preferably 9,9-dichlorofluorene, dichlorocyclopentadiene, 1 ,1-dichloroindene, 9,9- dichloroxanthene, 9,9-dichlorothioxanthene, 1 ,1-dichlorophenalene, 1 1 ,1 1-dichioro-4,5- methylenephenanthrene, p-biphenylyldiphenyidichloromethane, dichlorophenylpropanone, 4,4-dichloro-4,H-chromene, dichlorodeoxybenzoin, dichloroacetophenone, 1,1- dichloroacetone, more preferably 9,9-dichlorofluorene, it is particularly beneficial to react the dichloro compound with a compound having an activated (electron rich) aromatic structure such as a phenol, an aniline, a phenolic, a polyphenolic, an aromatic hydrocarbon such as toluene, anisole, indene, xylene, ethylbenzene, dimethoxybenzene, thiophene, furan, pyrole and the like. The term "dichloro compound" includes compounds having at least one gem-

dichloro group (two chlorine atoms on the same carbon) including such compounds as tetrachloroanthracene, as produced by chlorination of dihydroanthracene by practice of the invention.

Dichloro compounds such as 9,9-dichlorofluorene, are reactive in alkylation and can be reacted with aromatic compounds to form for example 9,9-diarylfluorenes where aryl substituents replace the chlorine atoms of the dichloro group(s). For simplicity, this aspect of the invention is explained in terms of aromatic derivatives of 9,9-dichlorofluorene, but the invention is not limited thereto and is applicable to all dichloro compounds such as are prepared by the process of the invention. Such compounds can be reacted with any aromatic compound which is reactive toward electrophilic aromatic substitution. These include aromatic compounds substituted with activating groups such as alkoxy, alkyl, hydroxy, or amino groups, as well as aromatic compounds substituted with weakly deactivating groups such as halo, ester, ketone, anhydride, or haloalkoxy groups. Reaction of for example 9,9-dichlorofluorene with two equivalents of an activated aromatic substrate can form for example a 9,9-diarylf luorene compound. For less activated aromatic substrates it is advantageous to use an excess of the aromatic substrate to produce for example a 9,9-diarylf luorene compound. Reaction of dichloro compounds such as 9,9-dichlorofluorene with an approximately equimolar amount of asuitably reactive aromatic substrate produces a polymer. To avoid crosslinking through the aromatic portion of a dichloro compound during the alkylation reaction with a dichloro compound to form monomers, olϊgomers, and polymers, the aromatic substrate should be more activated to alkylation than the dichloro compound itself. Activated aromatic compounds such as those having ether, hydroxyl or amine substituents can react with such compounds as 9,9-dichlorofluorene without the addition of a Lewis acid catalyst. Although the reaction proceeds without added catalyst, less activated aromatic compounds react more efficiently with the addition of an acid catalyst. Suitable protic and Lewis acid catalysts include, butare not limited to, HCI, AICl ₃ , FeCl ₃ , SbCI ₅ , H ₂ S0 ₄ , CH ₃ S0 ₃ H, EtAlCI ₂ (ethyl aluminum chloride), BF ₃ , ZnCI ₂ , GaC^,- calcined sulfate salts of Fe, Zn, Co, Mn, and Cu; AICI ₃ -CH ₃ N0 ₂ , SnCl ₄ , T1CI4, the metal alkanoates commonly known as paint driers such as iron naphthenate, zinc octoate, cobalt naphthenate, tin octoate, and similar such compounds, and polymeric acid catalysts including ion exchange resins, fluorine-containing sulfonic acid catalysts, acidic clays, zeolites, oxides of aluminum and silica. Lewis acid catalysts are optionally generated in-situ by reaction of active metals such as Al, Zn, Fe with HCI either added or formed as a byproduct of the reaction. Examples of aromatic compounds which can be alkylated with such compounds as 9,9-dichlorofluorene include toluene, xylene, 2-amiπophenol, ethyl benzene, indene, benzocyclobutane, anϊsole, phenol, aniline, 2-bromotetraf luoroethoxybenzene, bromobenzene, chlorobenzene, fluorobenzene, phenyl acetate, acetophenone and the like. Examples of aromatic compounds which can be alkylated with such compounds as 9,9- dichlorofluorene to form polymers include phenyl ether, phenyl carbonate,

dimethoxybenzene, diphenyl amine, benzene, xylene, durene, and substituted derivatives thereof where the substituents do not deactivate the aromatic compounds such that no polymer is formed. The dichloro and aromatic compounds are suitably unsubstituted or inertly substituted, that is having substituents which do not undesirably interfere with the alkylation. Such substituents include alkyl, alkoxy, halo, nitro, cyano, carboxyl, thio, sulfoxide, sulfone, carbonyl, ether, aryl, ester, anhydride, and ketone groups.

Any compound reactive with the dichloro compound is suitably reacted therewith. Exemplary phenolic compounds include any phenolic or thiophenolic compound which reacts with the dichloro compound, including alkylphenols, cresol, chlorophenol, propylphenol, 2,6-dimethylphenol, naphthol, dichlorophenol, phenylphenol, resorcinol, catechol, hydroquinone, aminophenols, hydroxybiphenyl, hydroxyacetophenone, allylphenols, dialkyl phenols, thiophenols, nitrophenols, halophenols, naphthols, hydroxybiphenyls, nonylphenol and ethylphenol, preferably o- and m-cresols, 2,6-dimethylphenol, o- and m- chlorophenols, 2-naphthol, 1-naphthol, 2-methylthiophenol, 2-nitrophenol, 3-nitrophenol, 4- nitrophenol, 2-aminophenol, 3-aminophenol, 4-aminophenol, 2,6-dimethylphenol, 2,6- dichlorophenol, 3,5-dichlorophenol, and pyrogallol, more preferably phenol, aminophenol, methoxyphenol, o-hydroxyacetophenone, 2,6-dimethylphenol, 2,6-dichlorophenol, 3,5- dichlorophenol, and pyrogallol, most preferably (unsubstituted) phenol, 2,6-dimethylphenol, 2,6-dichlorophenol, 3,5-dichlorophenol, and pyrogallol. Alkyl and dialkyl phenols preferably have alkyl groups of from 1 to 50 carbon atoms, more preferably from 1 to 10 carbon atoms, and are suitably cyclic, straight chain or branched.

Exemplary aniline compounds include unsubstituted aniline, N-alkylanilines, alkylanilines, dialkylanilines, o-, m-, p-phenylenediamine, chloroanilines, phenylanilines, N- methylaniline, methylaniline, 2,6-dimethylaniline, 2-chloroaniline, toluenediamine, methylenedianiline, polymeric methylenedianiline, 2,6-dichloroaniline, and ethylaniline. Exemplary aromatic hydrocarbon compounds include toluene, benzene, ethylbenzene, biphenyl, xylene, trimethylbenzene, durene, napthalene, indene, benzocyclobutane, diethyl benzene, dialkyl benzenes, furans (particularly methyl and chloro substituted) and thiophenes (particularly methyl and chloro substituted). Exemplary aromatic ether compounds include anisole, methylanisole, phenyl ether, dimethoxybenzene, biphenyl ether and naphthyl ether.

Exemplary halogenated aromatic compounds include chlorobenzene, fluorobenzene, and bromobenzene.

Exemplary aromatic ester compounds include phenyl acetate, phenyl carbonate, methyl benzoate, methylsalicylate, and phenyl benzoate.

Other exemplary aromatic compounds include acetophenone, and phthalic anhydride, as well as heterocyclic compounds including furan, methyl furan, chlorofuran,

benzofuran,thiophene, methylthiophene,chlorothiophene, pyrrole, methyl pyrrole, and chloropyrrole.

Derivatives of the listed compounds such as the acetamide or maleimide derivative of aniline or the listed anilines are similarly useful. For instance, the N- phenylmaleimide is useful to prepare bismaleimides which are useful as monomers for certain addition polymers. Similarly, the acetamide derivative of aniline may be used in place of aniline.

Alternatively, the dichloro compounds may be reacted with compounds whjch have at least two aromatic rings connected by a bond or a bridge of one atom which is optionally substituted, as illustrated by: aromatic-X-aromatic wherein "aromatic" stands for any aromatic group, including any suitable for use in the aromatic compounds not having a bridge, and X stands for a bond or any bridging group such as -0-, -S-, -CO-, -CH ₂ -, -NH-, -PR-, -S0 ₂ -, -SO-, -Se-, -SiR ₂ -, and the like, which are also optionally substituted, for example -NR-, -CR ₂ -, -CHR-, where R is any group, preferably an optionally inertly substituted hydrocarbyl group, more preferably a hydrocarbyl or fluorocarbyl group of from 1 to 6 carbon atoms. Such compounds are advantageously substituted such that dϊalkylation of the dichloro compounds onto these compounds occurs with the formation of a 5- or 6-membered ring, preferably between aromatic rings, more preferably including the bridging groupXandthe carbon of a gem-dichloro group inthedichloro compound. Forthis type of cyclization, it is preferred that substitution on the aromatic compound be such that alkylation ortho to the -X- group is favored. For example, where the presence of -X- would favor alkylation ortho or para to -X-, such as wherein -X- is -0-, -S-, or a bond, it is advantageous for the compound to be substituted in the para position with an activating or deactivating group such that alkylation ortho to -X- becomes the favored reaction. Compounds of this type include ,4'-thiodipheπol, 4,4'-oxydianiline,4,4'-thiodianiline, and 4,4'-diphenic acid. It is also advantageous to not only be substituted in the para position, but also to have additional substituents which further activate the position ortho to -X- to alkylation. Compounds of this type include 3,3',4,4'-tetraaminobiphenyl, 3,3'-dimethoxybenzidine, 3,3'-dihydroxy-4,4'- diaminobiphenyl, and 3,3'-dimethylbenzidine. Where such additional substituents do not activate the position ortho to -X-to alkylation, it can be advantageous to have additional substituents such that alternate positions which would have been active toward alkylation are already substituted and therefore, blocked. Compounds of this type include tetrabromobϊsphenol A. When the nature of the -X- group is such that alkylation ortho to this group is not favored, such as wherein -X- is -CO-, -S0 ₂ -, or -SO-, it is advantageous that the compound be substituted meta to -X- with groups which will activate the position ortho to -X- to alkylation. Compounds of this type include 3,3'-diamino-4,4'-diphenylsulfone, 3,3'- dimaleimido-4,4 ^* -diphenylsulfone, and 3,3'-diaminobenzophenone. Exemplary compounds for

the formation of such cyciized products include bismaleimide (1 ,1 '(methylenedi-4, 1 - phenylene)bismaleimide), 4,4'-diphenic acid, biphenyl tetracarboxylic dianhydride, 4,4'- diamino-3,3'-dihydroxybiphenyl, 3,3'-diamino-diphenylsulfone, 3,3'-dimaleimido- diphenylsulfone, 3,3'-4,4'-tetraaminobiphenyl, 3,3'-diaminobenzophenone, bis(4- fluorophenyOmethane, o-tolidine (3,3'-dimethylbenzidine), bisphenol F, tetramethylbisphenol F, 4,4'-thiodiphenol, 4,4'-oxydiphenol, 4,4'-oxydianiline, 4,4'-oxydiacetanilide (for example made by reaction of 4,4'-oxydianiiine with acetyl chloride), 4,4'-thiodianiline, 4,4'- thiodiacetanilide (for example made by reaction of 4,4'-thiodianiline with acetyl chloride), 3,3'- dimethoxybenzidine, 3,3'-dimethylbiphenyl, and 1 ,1-bis(3,4-dimethylphenyl)ethane. While cyclization advantageously takes place under the reaction conditions useful for alkylation (reaction of dichloro and aromatic compounds), especially when those conditions are acidic; additional acid is useful when the conditions are less acidic, particularly when the aromatic compound is amine-substituted, for example an aniline, and when little or no acid catalyst is used. The dichloro compounds are also reactive with aromatic compounds substituted such that cyclization to form 5- or 6-membered rings occurs during or subsequent to the alkylation reaction. Exemplary compounds for the formation of such cyciized products include resorcinol, hydroquinone, bisphenol A, bi phenol, m- and p- phenylenediamine, p- aminophenol, m-aminophenol and mixtures thereof. In the discussions of aromatic compounds, most illustrations of aromatic rings have been phenyl groups, however, while phenyl rings are preferred for their wide availability, the rings are suitably fused and/or heterocyclic rings such as furans, thiophenes and pyrrole. The aromatic compound preferably has from 5 to 20 ring atoms, at least 4 of which are preferably carbon atoms, with heteroatoms in the ring(s) suitably any atom which forms a ring compound with carbon, but preferably selected from O, S, N, and phosphorus, all optionally substituted such as S0 ₂ , NH, -PR ₂ or -P( = 0)R ₂ , where R is alkyl or alkoxy preferably of from 1 to 10 carbons. More preferably the ring has from 6 to 13 ring atoms, all most preferably carbon.

As one example of the use of multicyclic aromatic compounds, the dichloro compounds are optionally reacted with polyphenolic material such as novolac resins produced as reaction products of such aromatic hydroxyl compounds as phenol, cresol or xylenol with an aldehyde, preferably formaldehyde, to produce resinous products. The novolacs are advantageously kept at stoichiometric equivalent or excess to the dichloro compounds. These modified polyphenolics are useful as curing agents for polyepoxy compounds, and as starting materials for producing polyepoxies and polycyanates. To modify the reactivity, processability, or thermal and mechanical properties of the resultant products, dichloro-compounds and aromatic compounds are optionally reacted in suitable ratios to form oligomeric products which are useful for further reaction to form polymeric materials or which are optionally terminated with the same or different aromatic

compounds. By increasing the ratio of the dichloro compound to aromatic compound greater than that ratio sufficient to replace both chlorine atoms by aromatic compound, oligomers terminated with end groups containing a resulting mono-chloro compound are produced. For example, reaction of 9,9-dichlorofluorene with phenyl ether in a molar ratio of 3:2 will produce a mixture of oligomers wherein the terminal fluorene groups retain a 9-chloro functionality. Reacting this oligomeric mixture with an excess of phenol will produce a phenolic oligomer suitable for conversion into a polymer. Reacting the 9-chloro terminated oligomeric mixture with, for instance, a mixture of 9,9-dichlorofluorene and phenyl carbonate in 1 : 1 : 1 molar ratio will produce a block copolymer. Reacting the 9-chloro terminated oligomeric mixture with, for instance, benzocyclobutane, will produce a benzocyclobutanyl-capped oligomer suitable for useϊn forming benzocyclobutanyl polymers. Similarly, these oligomeric compounds can be further reacted by processes taught herein or within the skill in the art. Exemplary of aromatic compounds that can be used to form oligomers when reacted with dichloro compounds are phenyl ether, anisole, dimethoxybenzene, xylene, mesitylene, methylanisole, phenol, cresol, bisphenol A, bisphenol F, biphenol, phenol-aldehyde novolac resins, bisphenol S, biphenyl, phenylphenol, phenylthϊoether, and the like, and mixtures thereof.

Combinations of aromatic compounds are useful to react with the dichloro compounds. Advantageously, when combinations of aromatic compounds are reacted with the dichloro compounds, they have similar reactivity with the dichloro compounds such that the products of such reactions have approximately equal amounts of each aromatic compound distributed therein. For instance, benzocyclobutane and N-phenylmaleϊmide are reacted with 9,9-dichlorofluorene to produce mixtures of 9-(benzocyclobutanyl)-9- (maleϊmidophenyl)f luorene, 9,9-bis(benzocyclobutanyl)f luorene, and 9,9-bis(4- maleimidophenyOfluorene. Similarly, o-allylphenol and N-phenyimaleimide are reactive with 9,9-dichlorofluorene to produce an analogous mixture of compounds. Advantageously, when the aromatic compounds have dissimilar reactivities or when certain predetermined products or mixtures are desired, product mixture is controlled by stoichiometry of the mixture of aromatic compounds reacted or by use of sequential reactions. Sequential reactions are particularly exemplified by the reaction of one mole of an aniline with a dichloro compound according to the practice of the invention, which reaction produces a monoalkylation product; the monoalkylation product is then reacted with another mole of aromatic compound, for example a phenol. Alternatively, the monochloro compound may be isolated and used in a separate application. For example, reaction of 9,9-dichloroxanthene with one equivalent of benzene produces 9-chIoro-9-phenylxanthene, a deoxynucleoside 5'-0-protecting reagent. Control by stoichiometry of reactants is exemplified by reaction of a dichloro compound with a mixture of one mole each of two aromatic compounds (illustratively referred to as A and B) to produce a product mixture of alkylation products with two moles of B (BB), with two moles of A (AA), and with a mole of each (AB). When even distribution of products is desired and the

reactivities are dissimilar, those skilled in the art realize that use of an excess of the less reactive aromatic compound (the amount of excess determined by the difference in reactivities) will result in a more evenly distributed product mixture than will use of equi molar quantities of the aromatic compounds. Other predetermined proportions of products are obtained by use of proportions of the aromatic compounds determined by their relative reactivities and the predetermined desired proportions.

The reaction between the dichloro compound and the aromatic compound (together, reactants) is suitably conducted with any reactant ratios and under any conditions in which the compounds react, but, hen monomeric products are desired, preferably the aromatic is present in an amount sufficient to consume the chloro compound but insufficient to make isolation of the product difficult, preferably from 100: 1 to 1 : 100, more preferably from 50:1 to 1 : 1 , most preferably from 4: 1 to 1 : 1 based on equivalents of chlorine to be replaced. When the aromatic compound is an aniline, the most preferred ratio is from 4: 1 to 2: 1 moles of the aniline to moles of dichloro compound. When preparation of a polymer or oligomer is desired, reactant ratios are suitably any sufficient to produce polymer or oligomer but insufficient to produce mostly monomeric compound, preferably from 0.5: 1 to 1 :0.5 more preferably from 0.9: 1 to 1 :0.9 based on equivalents of chlorine to be replaced.

The reactants are suitably used neat or in solution. When used neat, they are preferably liquids, but may alternatively be used in the solid state by means within the skill in the art such as by mixing or grinding the finely divided solids together in a mill or blender under conditions of high shear. Especially when at least one reactant is solid, a solvent is used, suitable solvents include any solvent for at least one of the reactants, preferably for both, and preferably which does not interfere with the reaction or react with either reactant. Preferred solvents include carbon tetrachloride, chloroform, methylene chloride, ethylene dichloride, trichloroethane, tetrachloroethane (any common chlorinated solvents); aromatic hydrocarbons such as benzene, toluene, ethyl benzene; chlorinated aliphatic compounds; aromatic ethers such as phenyl ether; ethers, ketones, esters, amides, sulfoxides, alcohols, alkanoic acids, halogenated aromatics such as chlorobenzene; tetrahydrofuran; acetic acid and nitriles such as acetonitrile, more preferably dichloromethane, carbon tetrachloride, ethylbenzene, toluene or chlorobenzene.

While suitable reaction conditions include any effective conditions, including temperatures of from subambientto several hundreds of degrees centigrade; preferably temperatures used are less than the boiling points of any reactants under the pressure used. Conveniently, when a solvent is not used, the temperature is sufficient to allow at least one reactantto be liquid. Preferred temperatures range from -30°Cto 100°C, more preferably from 0°Cto 70°C, most preferably from 20°C to 50°C, particularly when a catalyst is used. When no additional catalyst is used, higher temperatures are generally advantageous. For instance, when no catalyst is used and the aromatic compound is an aniline, temperatures are preferably

at least 40°C, more preferably from 40°C to 200°C, most preferably from 50°C to 150°C. At these temperatures, reaction times are preferably at least 1 hour, more preferably from 1 to 8 hours, most preferably from 3 to 7 hours. The pressure is not critical, but conveniently ranges from 0.1 to several hundreds of atmospheres(10 to several ten thousands kPa), more preferably from 1 to 50 (100-5,000 kPa), most preferably from 1 to 200 atmospheres (100-20,000 kPa). It is generally desirable to run the reaction with less active compounds at the higher limits of the indicated ranges and the more active compounds at the lower limits of the indicated temperature and pressure ranges.

By controlling reaction temperatures, one can achieve substitution of one chlorine on the dichloro compound with an aromatic compound, leaving a remaining chlorine atom unreacted. For instance, when an excess of aniline is reacted with 9,9-dichlorofluorene at a temperature of 60°C, an exotherm to 110°C is observed, which exotherm corresponds to formation of 9-aminophenyl-9-chlorofluorene. Continued heating at 130°C results in further alkylation to form 9,9-bis(aminophenyl)fluorene. When isolation of a mono-substituted compound is desired, the reaction is advantageously run in a non-solvent for a compound or adduct that binds hydrochloric acid produced in the mono substitution. For instance, when the aromatic compound is an aniline, the hydrochloride of the aniline forms; therefore, when the reaction is run in a non-solvent for the aniline hydrochloride, such as monochlorobenzene, dichlorobenzene, toluene, or xylene, the aniline hydrochloride precipitates, removing the hydrochloride which is believed to otherwise act as catalyst for the substitution of the remaining chlorine atom by a second molecule of aromatic compound. It is observed that after precipitation of the aniline hydrochloride, even additional heating does not result in formation of the disubstituted product, for example bis(aminophenyl)fluorene. The mono-substituted product thus obtained is useful for instance in reactions with aromatic compounds different from that used to prepare the mono-substituted product; for instance, 9-aminophenyl-9- chlorof luorene is reactive with aromatic compounds such as phenol under conditions discussed herein to produce disubstituted compounds such as 9-aminophenyl-9- (hydroxyphenyl)fluorene.

Although the reaction is autocatalytic because the HCI produced by the reaction is an effective catalyst, any acid catalyst, advantageously HCI or any other hydrogen halide, may be added to the aromatic compound as a catalyst for the reaction, particularly as a catalyst for the thermodynamically favored (generally the para, para substituted aromatic) product. Use of a catalyst permits reaction at temperatures lower than would be effective without a catalyst. The temperatures depend on the reactivity of the aromatic compound with the dichloro compound; for instance in reactions with 9,9-dichlorofluorene, without catalyst, phenol will react at 0°C, aniline at 25°C and toluene at 70°C The catalyst is preferably added before the addition of the dichloro compound to the aromatic compound because addition of the acid catalyst is more useful before the concentration of HCI produced in the reaction has reached a

desired catalytic concentration. For instance, while the reaction of aniline with 9,9- dichlorofluorene at temperatures of more than 20°C does not require a catalyst, reaction of 9,9- dichlorofluorene with phenol preferably involves a catalyst to proceed at atmospheric pressure and temperatures of less than -20°C. That catalyst, however, is preferably a protic acid such as a hydrogen halide (preferably used at pressures greater than atmospheric pressure), or methanesulfonic (MSA), sulfuric, toluenesulfonic, hydroxybenzenesulfonic, trifluoromethanesulfonic, acetic, haloacetic, oxalic acid or mixtures thereof, at atmospheric pressure. By way of contrast, reaction of 9,9-dichlorofluorene with hydrocarbons such as benzocyclobutane and toluene preferably involve a Lewis acid catalyst, preferably AICI ₃ - CH ₃ N0 ₂ , SbCIs, FeCI ₃ , ZnCI ₂ , and more preferably ferric chloride or SbCIs in the case of benzocyclobutane to allow reaction to occur at less than 60°C. Anisole reacts with 9,9-dichloro without catalysts at 80°C, and at room temperature to 40 ^C C with Lewis Acid catalysts such as ZnCI or FeCI ₃ . Where, such as in the case of benzocyclobutane, the HCI produced can have a detrimental effect on the organic substrate, it is advantageous to sparge the HCI from the reaction with an inert gas such as nitrogen. When a hydrogen halide is used, it is advantageous to conduct the reaction under a pressure greater than atmospheric by the use of a hydrogen halide, for example hydrogen chloride. The pressure of the hydrogen halide may vary from 1 to 1000 (100-100,000 kPa) atmospheres, preferably from 10 to 100 (1000-10,000 kPa) atmospheres. Catalyst concentration affects the distribution of isomers in a product. For instance, in the case of the reaction of phenol with 9,9-dichlorofluorene, methanesulfonic acid (MSA) is an effective catalyst in concentrations of from 1 percent to 1000 percent based on the dichloro compound, but at the lower concentrations of from 1 percent to 10 percent, the ortho, para- isomer of bis(hydroxyphenyl)f luorene is formed along with the para, para- isomer. When the para-, para- isomer is preferred, the concentration of MSA is preferably from 15 percent to 1000 percent, more preferably from 20 percent to 100 percent based on the dichloro compound. In general, at atmospheric pressure and at temperatures of from 0° to 170°, and for a time sufficient to isomerize the product to the thermodynamically favored isomer, increasing the concentration of MSA from 1 percent to 1000 percent, increases the concentration of the para-, para- isomer relative to the concentration of ortho-, para- isomer of bis(hydroxy- phenyl)fluorene. Similar effects are found in other alkylations of the dichloro compounds. Similarly, when HCI or other hydrogen halide is the catalyst, increasing the pressure of the hydrogen halide from atmospheric pressure to 100 atmospheres (10,000 kPa) also increases the concentration of the thermodynamically favored para, para-, isomer relative to the concentration of the ortho, para-, isomer of bis(hydroxyphenyl)fluorene. Acid can be used to isomerize less thermodynamically favored products to more thermodynamically favored products during and/or after reaction of the dichloro compounds with the aromatic compounds.

Methods of recovering product alkylated aromatic compounds from reaction mixtures are within the skill in the art. Conveniently, when there is little excess aromatic compound remaining in the reaction mixture, a crystallization solvent is added to the reaction mixture to precipitate product. Advantageously, when excess reactant aromatic compound is present, it is removed by means within the skill in the art before a crystallization solvent is added to precipitate the product. Convenient crystallization solvents include hydrocarbons such as pentane and hexane; aromatic hydrocarbons such as toluene and ethylbenzene; chlorinated aliphatics such as chloroform and carbon tetrachloride; ketones such as acetone and methyl ethyl ketone; and esters such as diethyl carbonate and mixtures of these solvents. In the case of amine derivatives such as 9,9-bis(aminophenyl)f luorene, basic solutions such as, advantageously aqueous, solutions of sodium hydroxide or sodium bicarbonates are useful crystallization solvents. Crystallization is advantageously enhanced by cooling of the reaction mixture before or after addition of the crystallization solvent. Cooling alone is sometimes sufficient to cause precipitation of product without addition of crystallization solvent. Before crystallization, hydrogen chloride, if present, is optionally, but preferably, removed by means within the skill in the art such as distillation. The precipitate in each case is advantageously washed with a non-solventthereforto remove remaining reactants.

Avoiding addition of water or other material which would result in an additional waste to dispose is advantageous; therefore, methods of recovery which avoid using waterto wash or otherwise isolate the product are of particular interest. Such methods include use of hydrogen halides, preferably hydrogen chloride, as acid because it can be removed by such means as vaporization at reasonable temperatures without a water wash which is generally advantageous to remove such acids as sulfuric acid and methanesulfonic acids. Solid acids such as clay and polymer acids are also removable without water washing. When approximately stoichiometric amounts of dichloro compound and aromatic compound are reacted, especially to produce 9,9-bϊsarylfluorenes such as 9,9-bis(hydroxyphenyl)f luorene at temperatures less than 40°C in such solvents as methylene chloride, ethylbenzene, toluene, cumene, carbon tetrachloride, hexane, heptane, or other alkanes, the product precipitates without addition of other materials; thus, these conditions, too, are preferred for avoiding unnecessary waste disposal. The precipitate is recovered by means such as filtration and, optionally, recrystallization. Use of carbon tetrachloride as solvent in the reaction of an excess of such aromatic compounds as phenol with such dichloro compounds as 9,9-dichlorofluorene results in a precipitate identified as an 1 : 1 : 1 adduct of product: carbon tetrachloride: aromatic compound which can be recovered by such means as filtration, washing with a non-solvent for the product such as methylene chloride, a ketone or other chlorinated solvents to remove the aromatic compound and carbon tetrachloride or heating to remove carbon tetrachloride with washing to remove aromatic compound and, optionally, recrystallization from a suitable solvent, such as those suitable for washing.

In the alkylation of the dichloro compounds, isomers of the alkylation product are formed. For instance the reaction of phenol with 9,9-dichlorofluorene, both ortho, para- and para, para- isomers of the bis(hydroxyphenyl)fluorene are formed. Analogous ortho, para- and para, para- isomers of alkylation products of other dichloro compounds are also formed. Conversion of ortho, para- isomers to para, para- isomers is advantageous and can be accomplished by increased time in contact with an acid such as the acid catalyst which is optionally used in the reaction of the dichloro compound with the aromatic compound, an ion exchange resin, or a hydrogen halide, for example liquefied HCI, preferably an acid which is easily removed such as a polymeric acid (including ion exchange resin) or an acid that is easily vaporized such as a hydrogen halide or the acid present in the reaction mixture, removal of which would not add additional steps to the overall process. Frequently, there are also products representing the addition products of the dichloro compound with one mole of the aromatic compound and with product diaryl compound. Such byproducts are conveniently reacted with additional dichloro compound and/or additional aromatic compound to convert them into the desired product. Conveniently, these conversions of byproducts including ortho, para- product is accomplished by removal of the desired (para, para-) product and addition of additional aromatic compound and/or dichloro compound to the remaining reaction mixture; preferably the aromatic compound is added to the mixture including the byproducts, the mixture is heated (with additional acid if needed) to rearrange the isomers, then additional dichloro compound is used. In a continuous process or a process having sequential batches, advantageously a recycle process is used in which additional aromatic compound, additional dichloro compound and, optionally, additional acid are added to an initial reaction mixture, then desired product is removed as additional byproduct mixture is added

The process of the invention facilitates the preparation of such compounds as 9,9- bis(4-hydroxyphenyl)xanthene, 9,9-bis(2,3-diaminophenyl)fluorene, 9,9-bis(4-carboxyphenyl)- fluorene, 9,9-bis(4-acetylphenyl)fluorene, 9,9, 10,10-tetrakis(4-hydroxyphenyl)-9,10-dihydro- anthracene, 10,10-bis(4-hydroxyphenyl)anthrone, 9,9-bis(methylphenyl)fluorene, 9,9-bis(4- methoxyphenyl)f luorene, 1,1 '-(9H-fluoren-9-ylidenedi-4,1-phenylene)bismaleimide, and spiro[9H-fluorene-9,9'-[9H]xanthene]-3',6'-diol, previously available from ketones. The process of the invention also facilitates preparing a number of novel compounds including 9,9-bis(4-ethylphenyl)fluorene, 9,9-bis(4-ethenylphenyl)f luorene, 9,9- bis(4-ethynylphenyl)f luorene, 9,9-bis(3,4-dimethylphenyl)fluorene, 9,9-bis(2,3- dimethylphenyl)fluorene, 9-(3,4-dimethylphenyl)-9-(2,3-dimethylphenyl)-fluorene, 9,9-bis(3- amino-4-hydroxyphenyl)fluorene, 9,9-bis(4-amino-3-hydroxyphenyl)f luorene, 9-(3-amino-4- hydroxyphenyl)-9-(4-amino-3-hydroxyphenyl)fluorene, 9,9-bis(1 ,3-isobenzofurandion-4- yl)fluorene, 9,9-bis(1,3-isobenzofurandion-5-yl)fluorene, 9-(1 ,3-isobenzofurandion-4-yl)-9-(1 ,3- isobenzofurandion-5-yl)fluorene, 9,9-bis(benzocyclobutanyl)fluorene, 9,9-bis(4- halophenyl)fluorene, 9,9-bis(dicarboxyphenyl)fluorene, 9,9-bis(dihydroxyphenyl)fluorene, 9,9-

bis(dimethylphenyl)fluorene, spiro[9H-fluorene-9,9'-[9H]xanthene]-2',7'-diol, spiro[9H- fluorene-9,9'-{9H]carbazine]-3',6'-diol, spiro[9H-fluorene-9,9'-[9H]carbazine]-3',6'-diamine, spϊrotθH-fluorene-θ^'-øHlcarbazinel^' '-dϊamine. spirofθH-fluorene-g^'-pHlxanthene]- 2',7'-dicarboxylic acid, spiro[9H-fluorene-9,9'-[9H]xanthene]-3',6'-diamine, 2',7'- diacetylspiro[9H-fluorene-9,9'-[9H]xanthene], spiro[9H-f luorene-9,13'-[13H]-6-oxapentacene]- 2 ^, ,10 ^, -diol,spiro[9H-fluorene-9,13'-[13H]-6-oxapentacene]-3' ,9 ^, -diol, 3',6 ^, -diaminospiro[9H- fluorene-g.g'-thϊaxanthenel-IO'^O'-dϊoxide. spirofgH-fluorene-g^'fgH OH]- dihydroanthracene]-2 ^r ,7'-bismaleimide, 10-oxo-spiro[9H-fluorene-9,9'[9H,10H]- dihydroanthracenej-S'^'-diamine^' -dimethyl-spiro^H-fluorene-g.g'-OHlxanthenel^' '- dicyano-spiro[9H-fluorene-9,9'-t9H]xanthene], 2',7 ^r -diformyl-spiro[9H-fluorene-9,9'- t9H]xanthene], 2,7-diamϊno-3,6-dihydroxy-9,9'-spirobifluorene, 2,7-diamino-3,6-dimethyl-9,9'- spirobifluorene,spiro[9H-fluorene-9,9'[9H,10H]-dihydroanthra cene]-2',7'-diamine, 2',3',6 ^, ,7'- tetraaminospiro[9H-fluorene-9,9'-thiaxanthene]-l0',10'-dioxi de, spiro[9H-fluorene-9,9'- [9H]xanthene]-2',3',6,7'-tetraamine, 2,3,6,7-tetraamino-9,9 ^r -spirobifIuorene, 2,7-diamino-9,9'- spirobif!uorene-3,6-dithiol, 2,7-bis(1-methyI-1-(4-hydroxyphenyl)ethyl)spiro[xanthene-9,9 '- f luorene], 2,7-bis(4-hydroxyphenyl)sptro[xanthene-9,9'-f luorene], 1 ,3,6,8,10,10- hexamethylspϊro[dihydroanthracene-9,9'-f luorene]-2,7-diol, 1 ,3,6,8-tetrabromo- 10,10- dϊmethylspiro[dihydroanthracene-9,9'-fluorene]-2,7-dϊol, 1,3,6,8- tetramethylspiro[dihydroanthracene-9,9'-fluorene]-2,7-diol for example prepared from 9,9- dichlorofluorene reacted with ethylbenzene; styrene; ethylbenzene; 1,2-dimethyl benzene; o- amino phenol, phthalic anhydride, benzocyclobutane, halobenzene (for example bromo-, chloro-, iodo- or fluoro-benzene), phthalic acid, dϊmethylbenzene, catechol, hydroquinone, 3- aminophenol, p-phenylenediamine, m-phenylenediamine, 4-hydroxybenzoicacid, 4- aminophenol, 4-hydroxyacetophenone, 2,6-naphthalenedioI, 2,7-naphthalenediol, 3,3'- diaminodiphenylsulfone, 1,1'-(methylenedi-4,1-phenylene)bismaleimide, 4,4'-diamino- benzophenone, p-cresol, 4-cyanophenoI, 4-hydroxybenzaldehyde, 4,4'-diamino-3,3'-dihydroxy- biphenyl, o-tolidine, methylenedianiline, 3,3',4,4'-tetraaminodiphenylsulfone, 3,3',4,4'- tetraaminodiphenyletherorS diaminophenoI^^'^^'-tetraaminobiphenyl^^'- diaminobiphenyI-3,3 ^* -dithioI, bisphenol A, biphenol, tetramethyl bisphenol A, tetrabromobisphenol A, tetramethyl bisphenol F, respectively. The compounds are useful as monomers in polymers such as polyesters, polycarbonates, epoxy resins, polyamides, polyimides, polybenzoxazoles, polybenzimidazoles, benzocyclobutane polymers, polybenzthiazoles, polyquinoxalϊnes, and as intermediates for preparing monomers. For instance 9,9-bis(4-ethylphenyl)fIuorene is useful to prepare 9,9-bis(4-ethenylphenyl)fluorene by reactions analogous to those used to prepare styrene from ethylbenzene, and 9,9-bis(3,4- dimethyIphenyl)fluorene is useful to prepare 9,9-bis(1 ,3-isobenzofurandion-5-yl)f luorene by reactions analogous to those used to prepare phthalic anhydride from xylene. The polymers are prepared by means within the skill in the art. Additionally, the process of the invention

wherein mono-substitution of the gem-dichloro compound is achieved is particularly useful to prepare such aniline derivatives as 9-(4-aminophenyl)-9-chlorofluorene, and the alkyl derivatives thereof including the N-alkyl and ring-substituted derivatives such as 9-(4-(N- methylaminophenyl))-9-chlorofluorene, 9-(4-amino-3-methylphenyl)-9-chlorof luorene, 9-(4- amino-3-ethylphenyl)-9-chlorofluorene, 9-(4-amino-3-chlorophenyl)-9-chlorofluorene, 9-(4- amino-4-methylphenyl)-9-chiorofluorene, 9-(4-amino-2-ethylphenyl)-9-chlorofluorene, 9-(4- amino-2-chlorophenyl)-9-chlorofluorene, and mixtures thereof. Such compounds are particularly useful for further alkylation to form compounds having mixed aromatic substituents replacing the chlorines of the dichloro compounds. While products of the invention are generally useful in forming condensation polymers when they contain at least two reactive functional groups (for example hydroxyl, amine, sulfide, acid, acid halide, anhydride, or aldhehyde groups) or in addition polymers when they contain at least one carbon to carbon unsaturation, they can also be converted to additional compounds having useful functional groups. For instance phenol- and/or amine- containing products, especially phenol- and/or amine- containing derivatives of 9,9- dichlorofluorene are useful for conversion into epoxy resins by means within the skill in the art such as those described in "Epoxy Resins, Chemistry and Technology" C. A. May, Yoshio Tanaka, Marcel Dekker, Inc., NY (1973) and "Handbook of Epoxy Resins" H. Lee, K. Neville, McGraw Hill, NY (1967). Similarly, the phenol-containing products are useful for conversion into cyanate resins by means within the skill in the art such as those disclosed in Angew. Chem. Int'l Ed. 6, 206 (1967), E. Grigat and R. Putter; U.S. Patent No. 4,1 10,364 (1978) M. Gaku, K. Suzuki, K. Nahamichi; U.S. Patent No. 4,060,541 (1977) Rudolf Sunderman; U.S. Patent No. 3,994,949

(1976) Karl-Heinrich Meyer, Claus Burkhardt, Ludwig Bottenbruch; U.S. Patent No. 4,046,796

(1977) (Gunther Rottloff, Rudolf Sundermann, Ernest Grigat, Rolf Putter); and U.S. Patent No. 4,028,393 (1977) (Gunther Rottloff, Rudolf Sundermann, Ernest Grigat, Rolf Putter). The phenol-containing compounds are also useful as agents for partial advancement or hardening of epoxide-containing compounds. The amino-containing products are also useful as curing agents for epoxide-containing compounds.

The following examples are given to illustrate, but not limit the invention. In the examples, all parts, ratios and percentages are by weight unless specified otherwise. Examples of the invention (Ex.) are designated numerically, while comparative samples (C.S.) are designated alphabetically.

EXAMPLE 1 : Preparation of 9,9-Dichlorofluorene from Fluorene The reactor was a 500 mL 3-neck round bottomed flask equipped with a magnetic stir bar, nitrogen purge and thermometer. The reactor was flushed with nitrogen, and a solution of fluorene (6.00 g, 0.036 mole) and carbon tetrachloride (CCI ₄ ) (669.48 g, 4.35 mole, 420 mL) was charged to the reactor followed by NaOH (50 percent solution in water, 6.00 g,

0.075 mole, 4.0 mL, 3.00 g dry weight). The stirrer was started and the speed adjusted to 500 RPM (revolutions per minute). The mixture was stirred with a subsurface nitrogen sparge. The temperature of the reaction solution was 28°C. The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 1.9 g, 0.0029 mole, 1.9 mL), was added at once. The reaction was followed by gas chromatography (GC) on a GC (commercially available from Varian Associates underthe trade designation Varian 3700 equipped with a 15 meter by 0.53 mm Megabore (Trademark of JδtW Scientific) capillary column coated with a 1 micron film of polydimethylsiloxane commercially available from J&W Scientific underthe trade designation DB-1 as the stationary phase and a flame ionization detector (FID)(commercϊaIly available from Varian Associates underthe trade designation Varian 3700) with conditions of 250°C at the injector, 350°C at the detector, 130°C of the column for the first minute then programmed to rise 3° per minute to 160°C and to hold that temperature for one minute. The reaction mixture was sampled after 20 minutes and analyzed by GC which shows that the reaction mixture contains 18.46 percent fluorene, 81.54 percent 9,9-dichlorofluorene. No other product was evident. The reaction was sampled and analyzed periodically over the next three hours. After stirring overnight no fluorene remains according to the GC analysis; results were shown in Table 1. Stirring was stopped, the phases were allowed to separate, and the aqueous phase was removed from the reactor. The reaction mixture was filtered through alumina and the CCI4 was removed on a rotary evaporator leaving 8.40 g of light yellow crystals, 99 percent of theory. NMR and GC-mass spectral analyswas ofthismaterial shows itto be identical with that of a known sample of 9,9-dichlorofluorene.

TABLE 1 CHLORINATION OF FLUORENE BY CARBON TETRACHLORIDE via PHASE TRANSFER CATALYSIS

*C1-C14 is a chloro-fourteen carbon compound foun in the product of photochlorination of fluorene.

The data in Table 1 shows that chlorination of fluorene by the process of the invention leads to 9,9-dichlorofluorene and not 2,7-dichlorof luorene or the Cl-Cι compounds detected in photochlorination of fluorene. The product was also observed to have a mass spectrum corresponding to that of a known sample of 9,9-dichlorofluorene with base peak at 199 atomic mass units (AMU) and parent ion at 234 AMU. The nuclear magnetic resonance (NMR) spectrum of the product was also consistent with that of a known sample of 9,9-dichlorofluorene with peaks at 7.81 - 7.84 ppm (2H), 7.59 - 7.63 ppm (2H), 7.37 - 7.47 ppm (4H) relative to TMS (tetramethylsilane) in CDCI ₃ solution. The C ¹ 3 NMR shows peaks at 146.69, 136.48, 130.67, 129.041. 124.60, 120.14, 82.93 relative to tetramethylsilane.

COMPARATIVE SAMPLE A: Reaction Carried Out as Described by Reeves et al.

The reactor was a 500 mL 3-neck round bottomed flask equipped with a magnetic stir bar, nitrogen purge and thermometer. The reactor was flushed with nitrogen, and a solution of fluorene (6.00 g, 0.036 mole) and carbon tetrachloride (669.48 g, 4.35 mole, 420 mL) was charged to the reactor followed by NaOH (50 percent solution in water, 6.00 g, 0.075 mole, 4.0 mL, 3.00 g dry weight). The stirrer was started and the speed adjusted to 500 rpm. The mixture was stirred with a subsurface nitrogen sparge. The temperature of the reaction solution was 28°C. The catalyst, tetrabutylammonium bromide (0.93 g, 0.0029 mole), was added at once.

The reaction was followed by gas chromatography (GC) as in Example 1. The reaction mixture was sampled after 40 minutes and analysis shows 98.17 percent fluorene, 1.83 percent 9,9-dichlorofluorene. The reaction was sampled and analyzed periodically overthe next thirty six hours. After three hours, only 10.29 percent of the fluorene had been converted to 9,9-dichlorofluorene compared to 95.46 percent in the previous example (Example 1). At thirty six hours, 15.16 percent of the fluorene remained unconverted in the reaction mixture. This data showsthattetrabutylammonium bromide was not as effective a catalyst for this reaction as wastetrabutylammonium hydroxide.

EXAMPLE 2: Use of Less than an Equivalent of Sodium Hydroxide in the Chlorination of Fluorene

The reactor was a 12 inch (30.48 cm) section of 2 inch (5.08 cm) inside diameter pipe made from fluorocarbon polymer commercially available from E.l. du Pont de Nemours & Co. under the trade designation Teflon PFA which was swaged to a 0.5 inch (1.27 cm) tee at the lower and upper ends. To the bottom-most leg of the lower tee was joined a stopcock which can be used to drain the reactor's contents. To the other leg of this tee was attached the suction inlet of a pump (a magnetically driven centrifugal pump commercially available from March Manufacturing Inc. underthe trade designation March model, MDX-MT3 rated at 28.39 liters/minute at zero head). The outlet of the pump was plumbed to a heat exchanger through which the reaction mixture passes and which, in turn, was connected to the top-most leg of the upper tee on the reactor. The other leg on the upper tee was used as a port to charge reactants to and vent purge gas from the reactor. A nitrogen sparge was provided at the inlet of the heat exchanger. This design results in continuously impinging the organic phase at a high velocity into the aqueous phase, achieving good interfacial contact. A thermocouple was provided*at the discharge port of the pump for measuring the temperature of the reaction. The reactor was flushed with nitrogen. Then a solution of fluorene (16.62 g, 0.1000 mole) and carbon tetrachloride (149.60 g, 0.9726 mole, 93.85 mL) was charged to the reactor foil owed by NaOH (30 percent solution in water, 1.33g, 0.010 mole, 1.00 mL, 0.40 g dry weight). The pump was started and the temperature adjusted to 30° C. The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 1.28 g, 0.0020 mole, 1.28 mL), was added at once. The mixture was circulated with a subsurface nitrogen sparge. The temperature of the reaction solution was 30°C.

The reaction was followed by gas chromatography (GC) by the procedure of Example 1. Analysis after 60 minutes showed 15.46 percent fluorene, 84.09 percent 9,9- dichlorofluorene.

This result shows that less than an equivalent of sodium hydroxide was effective in achieving chlorination bythe process of the invention.

EXAMPLE 3: Use of Benzyltrimethylammonium Hydroxide as Catalyst in the Chlorination of

Fluorene

The reactor described in Example 2 was flushed with nitrogen, and a solution of fluorene (49.87 g, 0.3000 mole) and carbon tetrachloride (448.79 g, 2.9177 mole, 281.55 mL) was charged to the reactor followed by NaOH (30 percent solution in water, 4.00 g, 0.030 mole, 3.00 mL, 1.20 g dry weight). The pump was started and the temperature adjusted to 30°C. The catalyst, benzyltrimethylammonium hydroxide (40 percent solution in water, 2.82 g, 0.0060 mole, 2.66 mL) was added at once. The mixture was circulated with a subsurface nitrogen sparge. The temperature of the reaction solution was 30°C. The reaction was followed by gas chromatography as in Example 1. After 60 minutes analysis showed 84.34 percent fluorene, 14.20 percent 9,9-dichlorofluorene.

This example shows that benzyltri methyl ammonium hydroxide was an effective catalyst for this reaction.

EXAMPLE 4: Use of a Ten Percent Concentration of Sodium Hydroxide in Chlorination of Fluorene The reactor was a 1000 mL cylinder 4 inches in diameter (100 mm) by 5.5 inches high (140 mm) equipped with a 2 inch (50 mm) diameter turbine impeller driven by a vertical shaft. Stirring rate was measured by a tachometer. Temperature was controlled by a 10 foot (3.048 m) by 0.25 inch (0.635 cm) external diameter coil immersed in the reaction medium through which coolant was pumped maintained at a constant temperature by a circulating refrigerated/heated bath.

The temperature was measured by a thermocouple inside a thermowell which runs the entire depth of the reactor. The reactor was also equipped with a nitrogen inlet which was used to maintain a nitrogen atmosphere above the reaction solution. The entire apparatus was constructed of fluorocarbon resin commercially available from E.l. du Pont de Nemours &

Co. underthe trade designation Teflon PFA.

The reactor was flushed with nitrogen. Then a solution of fluorene (14.96 g,

0.090 mole) and carbon tetrachloride (134.64 g, 0.8753 mole, 84.47 mL) was charged to the reactor followed by NaOH (10 percent solution in water, 359.99 g, 0.90 mole, 324.32 mL,

36.00 g dry weight). The stirrer was started and the speed adjusted to 3000 rpm. The coolant was admitted to the coils, and the temperature of the reaction solution was adj usted to 30°C.

The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 1.14 g, 0.0018 mole, 1.16 mL), was added at once. The reaction mixture was sampled after 1 minute and analyzed by GC according to the method described in Example 1 ; it shows 94.89 percent fluorene, 4.50 percent 9,9-dichlorofluorene and 0.25 percent 9-fluorenone. After 15 minutes, analysis showed 94.26 percent fluorene, 4.56 percent 9,9-dichlorofluorene and 0.30 percent 9-

fluorenone. After an additional three hours, analysis showed 88.78 percent fluorene, 8.52 percent 9,9-dichlorofluorene and 0.76 percent 9-fluorenone.

This example shows that even 10 percent sodium hydroxide was effective in this reaction.

EXAMPLE 5: Use of Twenty Percent Sodium Hydroxide in the Chlorination of Fluorene

The reactor described in Example 4 was flushed with nitrogen. A solution of fluorene (25.56 g, 0.1538 mole) and carbon tetrachloride (230.08 g, 1.4958 mole, 144.34 mL) was charged to the reactor followed by NaOH (20 percent solution in water, 307.59 g, 1.5380 mole, 252.13 mL, 61.52 g dry weight). The stirrerwas started, and the speed adjusted to 3000 rpm. The coolant was admitted to the coils and the temperature of the reaction solution was adjusted to 30°C The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 1.97 g, 0.0031 mole, 1.98 mL) was added at once. After 1 minute, analysis by the procedure of Example 1 shows 39.65 percent fluorene, 60.04 percent 9,9-dichlorofluorene, and 0.32 percent 9-fluorenone. After 15 minut.es, analysis showed 1.44 percent fluorene, 98.21 percent9,9- dichlorof luorene and 0.34 percent 9-fluorenone. After a three hour reaction time analysis showed 0.97 percent fluorene, 98.66 percent 9,9-dichlorofluorene and 0.37 percent 9- fluorenone. This result indicates that twenty percent aqueous sodium hydroxide was effective.

EXAMPLE 6: Use of Thirty Percent Sodium Hydroxide in the Chlorination of Fluorene

The procedure of Example 5 was repeated except that a solution of fluorene (33.24 g, 0.200 mole) and carbon tetrachloride (299.20 g, 1.94 mole, 187.20 mL) was charged to the reactor followed by NaOH (30 percentsolution in water, 266.66 g, 2.00 mole, 200.50 mL, 80.00 g dry weight) and the catalyst was tetrabutylammonium hydroxide (40 percent solution in water, 2.54 g, 0.0040 mole, 2.57 mL). After 1 minute, analysis showed the reaction mixture contains 18.14 percent fluorene, 81.34 percent 9,9-dichlorofluorene, and 0.51 percent 9-fluorenone. After 15 minutes, analysis showed 1.41 percent fluorene, 98.39 percent 9,9-dichlorofluorene and 0.21 percent 9-fluorenone. After an additional one hour, 45 minutes, analysis showed 0.52 percent fluorene, 98.66 percent 9,9-dichlorofluorene and 0.56 percent 9-fluorenone. This example showsthat30 percent sodium hydroxide was effective in the process of the invention.

EXAMPLE 7: Use of 40 percent Sodium Hydroxide in Chlorination of Fluorene

The procedure of Example 5 was repeated except that a solution of fluorene (33.24 g, 0.200 mole) and carbon tetrachloride (299.20 g, 1.94 mole, 187.20 mL) was charged to the reactor foil owed by NaOH (40 percentsolution in water, 203.64 g, 2.00 mole, 139.86 L, 80.00 g dry weight) and the catalyst was tetrabutylammonium hydroxide (40 percent solution in water, 2.54 g, 0.0040 mole, 2.57 mL). After 1 minute, analysis showed that the reaction

mixture now contains 17.22 percent fluorene, 81.50 percent 9,9-dichlorofluorene, and 1.10 percent 9-fluorenone. After 15 minutes, analysis showed 0.95 percent fluorene, 98.69 percent 9,9-dichlorofluorene and 0.35 percent 9-fluorenone. After an additional one hour, 45 minutes, analysis showed 0.01 percent fluorene, 99.07 percent 9,9-dichlorofluorene and 0.92 percent 9- 5 fluorenone.

This example shows that 40 percent sodium hydroxide was effective in this reaction.

EXAMPLE 8: Use of Fifty Percent Sodium Hydroxide in Chlorination of Fluorene 10 The procedure of Example 5 was repeated except that a solution of fluorene

(33.24 g, 0.200 mole) and carbon tetrachloride (299.20 g, 1.94 mole, 187.20 mL) was charged to the reactor followed by NaOH (50 percent solution in water, 160.00 g, 2.00 mole, 106.67 mL, 80.00 g dry weight) and the catalyst was tetrabutylammonium hydroxide (40 percent solution in water, 2.54 g, 0.0040 mole, 2.57 mL). After 1 minute, analysis showed 6.58 percent fluorene, 15 92.38 percent 9,9-dichlorofluorene, and 1.03 percent 9-fluorenone. After 15 minutes, analysis showed 0.92 percent fluorene, 98.59 percent 9,9-dichlorofluorene and 0.49 percent 3- fluorenone. After an additional one hour, 45 minutes, analysis showed 0.01 percent fluorene, 92.13 percent 9,9-dichlorofluorene and 7.86 percent 9-fluorenone.

This example shows that 50 percent sodium hydroxide was effective in this 0 reaction.

EXAMPLE 9: Use of Methylene Chloride as an Alternative Solvent with a Stoichiometric amount of CCI ₄ .

The reactor described in Example 4 except with a baffle affixed to the immersed 5 coil was flushed with nitrogen, and a solution of fluorene (47.47 g, 0.2856 mole), methylene chloride (142.40 g, 1.6767 mole, 107.47 L) and carbon tetrachloride (87.85 g, 0.571 1 mole, 55.1 1 mL) was charged to the reactor foil owed by NaOH (30 percent solution in water, 380.75 g, 2.8556 mole, 286.28 mL, 114.23 g dry weight). The stirrerwas started and the speed adjusted to 3000 rpm. The coolant was admitted to the coils and the temperature of the reaction solution 0 was adjusted to 30°C. The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 3.633 g, 0.0057 mole, 3.67 mL), was added at once.

Analysis by the procedure of Example 1 at one minute showed 3.35 percent fluorene, 95.76 percent 9,9-dichlorofluorene, and 0.89 percent 9-fluorenone. After 15 minutes, analysis showed 0.78 percent fluorene, 99.05 percent 9,9-dichlorofluorene and 0.17 percent 9- 5 fluorenone. After an additional hour, analysis showed 0.00 percent fluorene, 99.02 percent 9,9- dichlorofluorene and 0.65 percent 9-fluorenone.

This example shows that good results were obtained using alternative solvents such as methylene chloride with only a stoichiometric amount of CCI ₄ .

EXAMPLE 10: Use of Cumene as an Alternative Sol vent with a Stoichiometric amount of

CCI ₄ . The procedure of Example 9 was repeated except that a solution of fluorene (33.24 g, 0.200 mole) cumene (188.38 g, 1.5672 mole, 218.04 mL) and carbon tetrachloride (61.53 g, 0.400 mole, 38.60 mL) was charged to the reactor followed by NaOH (50 percent solution in water, 160.0 g, 2.00 mole, 103.9 mL, 80.00 g dry weight); and the catalyst was tetrabutylammonium hydroxide (40 percentsolution in water, 2.59 g, 0.004 mole, 2.57 mL). After 1 minute, analysis showed 0.92 percent fluorene, 97.05 percent 9,9-dichlorofluorene, and 2.03 percent 9-fluorenone. After 15 minutes, analysis showed 0.92 percent fluorene, 97.27 percent 9,9-dichlorofluorene and 1.81 percent 9-fluorenone. This data showed that the reaction was essentially complete within one minute under these conditions.

This example shows that good results were obtained using alternative solvents such as cumene with only a stoichiometric amount of CCI4.

EXAMPLE 11 : Use of Ethylbenzene as an Alternative Solvent with a Stoichiometric amount of CCI4. The procedure of Example 9 was repeated except that a solution of fluorene (33.24 g, 0.200 mole), ethylbenzene (188.38 g, 1.7743 mole, 217.28 mL), and carbon tetrachloride (61.53 g, 0.400 mole, 38.60 mL) was charged to the reactor followed by NaOH (50 percentsolution in water, 160.0 g, 2.00 mole, 103.9 mL, 80.00 g dry weight); and the catalyst was tetrabutylammonium hydroxide (40 percentsolution in water, 2.59 g, 0.004 mole, 2.57 mL). After 1 minute, analysis showed 1.23 percent fluorene, 97.78 percent 9,9-dichlorofluorene, and 0.59 percent 9-fluorenone. After 15 minutes, analysis showed 1.29 percent fluorene, 97.84 percent 9,9-dichlorofluorene and 0.87 percent 9-fluorenone.

This example shows that good results were obtained using alternative solvents such as ethylbenzene with only a stoichiometric amount of CCI4.

EXAMPLE 12: Use of Ethylbenzene as an Alternative Solvent with a Stoichiometric amount of CCI4 with 30 percent Sodium Hydroxide.

The procedure of Example 9 was repeated except that a solution of fluorene (33.24 g, 0.200 mole) ethylbenzene (188.38 g, 1.7743 mole, 217.28 mL) and carbon tetrachloride (61.53 g, 0.400 mole, 38.60 mL) was charged to the reactor followed by NaOH (30 percent solution in water, 266.67 g, 2.00 mole, 200.5 mL, 80.00 g dry weight); and the catalyst was tetrabutylammonium hydroxide (40 percentsolution in water, 2.59 g, 0.004 mole, 2.57 mL). The reaction mixture was sampled after 1 minute and periodicallythereafterforthe next30 minutes and analyzed by GC according to the procedure in Example 1. GC analysis at one minute showed 25.65 percent fluorene, 74.35 percent 9,9-dichlorofluorene, and 0.0 percent 9-

fluorenone. GC analysis at 15 minutes showed 1.28 percent fluorene, 98.72 percent 9,9- dichlorofluorene, and 0.0 percent 9-fluorenone. GC analysis at 30 minutes showed 1.19 percent fluorene, 98.81 percent 9,9-dichlorofluorene, and 0.0 percent 9-fluorenone.

This result shows that ethylbenzene was effective as a solvent when used with a stoichiometric amount of carbon tetrachloride with 30 percent sodium hydroxide.

Example 13: Continuous Preparation of 9,9-dichlorofluorene

This reaction was carried out in a reactor constructed from a 2 inch (5.08 cm) diameter pipe of fluorocarbon resin commercially available from E.I. du Pont de Nemours & Co. under the trade designation Teflon PFA. The reactor contains 6 stirred sections each 1.75 inches (4.45 cm) long separated by horizontal spacers 0.25 inch (0.64 cm) thick which were perforated with eight 0.25 inch (0.64 cm) diameter holes which allow communication between the stages. Centered within each stage was an impeller mounted on a vertical drive shaft constructed of type 316 stainless steel which was 0.375 inches (0.954 cm) tall, by 0.625 inches (1.651 cm) in diameter. An air driven motor drives the drive shaft at a constant speed of 1500 rpm. Each stirred section was approximately 100 mL in volume. The top of the reactor was equipped with ports for the introduction of nitrogen and venting the same such that an inert atmosphere can be maintained during the course of the reaction. The upper-most stirred section or stage contained a port for the introduction of reactants and a thermowell for measuring the temperature of the reactor's contents. There were additional thermowells in stage four from the top and just below the sixth stage. Below the sixth stage from the top there was a twelve inch (30.48 cm) long section which was unstirred and acts as a quiet zone so that the organic and aqueous phases can disengage or phase separate. At the bottom of this reactor was a tee which was connected to a bottom drain on one leg so that the enti re contents of the reactor can be removed, and to an overflow tube on the other leg which can be adjusted to control the liquid level in the reactor. The product solution can be continuously drawn off this overflow at a rate equivalent to that which the feed solution is introduced to the first stage of the reactor. The reactor was purged with nitrogen and then charged with a volume of carbon tetrachloride (400 mL) such that its level just comes to the bottom of the sixth stage. NaOH solution (50 percent by weight, 12.0 moles, 480 g dry weight, 960 g solution weight, 627.45 mL) was then charged to the reactor, which fills all six of the stirred zones. The stirrer was started and its speed was adjusted to 1500 rpm. Fluorene (3.0 moles, 498.66 g) dissolved in carbon tetrachloride (64.84 moles, 9973.20 g, 6256.71 mL) was fed into the first reactor stage through a metering pump at a rate of 19 mL/min. At the same time, tetrabutylammonium hydroxide (0.03 moles, 7.78 g dry weight, 19.46 g as 40 percent aqueous solution) was fed through a separate metering pump into the first reactor stage at a rate of 0.05 mL/min. The product solution was collected at the overflow and analyzed by gas chromatography (GC) on a gas chromatograph commercially available from Varian Associates under the trade designation

Varian 3400 GC equipped with a 30 meter by 0.53 mm Megabore (Trademark of J&W Scientific) capillary column coated with a 1 micron film of polytrifluoropropyl-co-dimethylsiloxane commercially available from J&W Scientific under the trade designation DB-210 as the stationary phase and a flame ionϊzation detector (FID) commercially available from Varian Associates underthe trade designation Varian 3400.

When fluorene was no longer detected in the effluent stream the product was fed to a wash column which was identical in design to the reactor column. The product solution was fed to the first stirred stage at a rate of 10 mL/min and water was fed at the sjxth stirred stage at a rate of 20 mL/min. The organic solution containing the 9,9-dichlorofluorene was collected at the overflow of the wash column and then passed through a column of molecular sieves (4A size, commercially available from Linde Division, Union Carbide Industrial Gasses Inc.) which lowered the water content of the stream as measured by Karl-Fisher titration from 211 ppm to 11.4 ppm. The total product solution collected in this fashion amounts to 12,004 g which was 5.8 weight percent 9,9-dichlorofluorene and 0.05 weight percent fluorenone.

EXAMPLE 14: Large Scale Preparation of Bis(hydroxyphenyl)f luorene

Fluorene (800 lb, 363.63 kg) was dissolved in methylene chloride (2722 lb, 1237.27 kg) in a tank. Then an amount of carbon tetrachloride (CCI ₄ ) stoichiometric with the fluorene (1488 lb, 676.36 kg) was added to the resulting solution and mixed thoroughly using a retreated blade agitator. The resulting solution was charged via nitrogen pad into a 1000 gallon (3785 liter) glass lined reactor containing (4963 lb, 2256 kg) of a 30 weight percent aqueous solution of sodium hydroxide (NaOH) which was at a temperature of 25°C. The reactor was maintained at a temperature of 25°C by a flow of a 50 weight percent aqueous solution of ethylene glycol through a steel jacket around the reactor. The reactor was purged with nitrogen. After all of the fluorene solution was charged to the reactor and phase separation occurs (about 10-15 minutes), agitation was begun. Eitherthe aqueous phase orthe organic phase can be a continuous phase.

While the temperature was maintained at 25°Cto avoid catalyst deterioration, (12.5 lb, 5.68 kg) of tetra-n-butylammonium hydroxide in aqueous solution (40 percent) was fed to the reactor at a rate of 2.84 kg/hr over a period of 2 hours. Agitation and cool i ng were continued for one hour, after which agitation was stopped. This was to assure that all of fluorene has been reacted. Reaction of all fluorene was confirmed by GC analysis of the reaction mixture, then the phases were allowed to separate with the organic phase on the bottom of the reactor.

The organic phase was removed at a rate of 1500 Ib/hr (681.81 kg/hr) using a centrifugal pump until a small rag layer (where phases were incompletely separated) remains. The organic phase was then washed with 1200 Ib/hr (545 kg/hr) portions of water three times

with separation of water from the organic phase each time. The organic phase was found to contain 23 weight percent dichlorofluorene (DCF) in methylene chloride with some CCI ₄ and chloroform present. That phase was then stored and mixed with subsequent batches of essentially the same composition prepared by the same process. Optionally, the phase could be used immediately to prepare bis-(hydroxyphenyl)fluorene. Repeating the synthesis of DCF offers an opportunity to reuse the NaOH solution. Optionally, the organic phase could be dried using a molecular sieve column to remove water.

After a desired number of batches of DCF were produced, the NaOH solution was pumped from the reactor to the wash section. The sodium hydroxide solution was then treated with sufficient HCI (hydrochloric acid) to neutralize the NaOH.

The reactor and associated piping were flushed with methylene chloride. Fifty five gal (2081) of molten phenol at 50°C was transferred to a reactor using a nitrogen pad. During transfer, the phenol was maintained under a nitrogen atmosphere and weighed such that 232 kg (25 mole percent excess based on 9,9 dichlorofluorene) were transferred into the 1000 gal. (3785 1) reactor which was also purged with nitrogen.

Methanesulfonic acid (MSA) (93 lbs, 42.2 kg) was added to the reactor as the catalyst for the phenolation process. For the first batch, the reactor did not contain the recycled material. After the first batch, the reactor content contained the recycled material which consisted of chloroform, methylene chloride, p,p-BHPF, o,p-BHPF, 3,2-fluorene-phenol adducts, MSA and phenol. The entire content of the reactor was circulated around the reactor using the pump at the bottom of the reactor and the agitator inside the reactor.

A 1207.27 kg portion of the DCF in methylene chloride was added to the recirculation loop of the reactor containing phenol at a rate of 603.63 kg/hr, over a period of 2 hours while the reactor was maintained at 15°C by cooling in the jacket of the reactor. The pressure remained atmospheric. Precipitation of product p,p-bis(hydroxyphenyl)fluorene (p,p- BHPF) was noted after half of the DCF was added.

The addition of DCF was completed after 2 hours, precipitation of product p,p- bis(hydroxyphenyl)f luorene (p,p-BHPF) was noted after half of the DCF was added, but the contents of the reactor were allowed to digest (remain at the same temperature with stirring) over the period of 1 hour at 40°C. Then the reaction mixture was allowed to cool down to 10°C over a 1 hour period. HCI produced during the first two hours of the reaction was vented into a scrubber containing 5001 of 15 weight percent aqueous NaOH. Heat was removed from the scrubber using an external heat exchanger to cool the circulated sodium hydroxide. After 95 percent of the HCI was removed by venting it to the scrubber, then sufficient nitrogen was introduced from the bottom of the reactor to remove HCI. When removal of HCI was complete as determined by the pH of the solution not being acidic, the remaining contents of the reactor (hereinafter, reaction mixture) were transferred to a nolding tank using a pump. The reaction mixture was a slurry.

After the entire contents of the reactor have been transferred (130 gal, 492.05 i), the slurry of crystalline product in reaction mixture was transferred to a pressure ilter unit commercially available from Rosenmund, Inc. where a pressure of 15 psi (103.41 kPa) was applied using nitrogen pressure. The amount transferred was controlled by weighing the slurry feed tank before and after transfer using a commercial weight cell.

During transfer to the filter unit, a drain valve was closed; when transfer was complete and no malfunction of the filter was noted, the valve was opened and a pressure of 35 psig, (241.29 kPa) of nitrogen was applied such that a filtrate containing methylene chloride, excess phenol, carbon tetrachloride, chloroform, p,p-BHPF; o,p-BHPF and phenol-f luorene adducts was collected in a check tank.

The filtrate was transferred from the check tank to a batch distillation unit to remove a mixture of chloroform and methylene chloride, which mixture was suitable for recycle to wash the filter cake. Before the distillation started, 232.25 kg phenol with 42.27 kg methanesulfonic acid were added to the batch distillation pot. Then the distillation of chloroform and methylene chloride was started. After ail of the chloroform and all of the methylene chloride had been removed, the remaining filtrate (about 15 percent by weight) was transferred to a waste storage tank, the rest of the mixture was maintained at a temperature of 70°C and a pressure of 103.41 kPa under nitrogen for a period of one hour to isomerize the pheπol-f luorene adducts and o,p-BHPF to p,p-BHPF. A filter cake of BHPF forms on the filter and was washed using the recovered methylene chloride/chloroform mixture by closing the drain valve, charging the mixture to the filter apparatus at a rate of 10 gal/min (37.85 l/min) until a total of 100 gal (378.51) was charged, opening the valve and applying a pressure of 35 psig (241.29 kPa) of nitrogen. The mixture was pushed through the filter cake and collected then recycled to be used with the isomerization mixture. The recycled methylene chloride (MeCI ₂ )/chloroform solution was sent back to the batch distillation pot. The washing step above was repeated two more times: first with the recycled MeCI ₂ solution, and then with fresh MeCl ₂ solution; only a trace of MSA remained in the filter cake.

The wet filter cake was found to contain 45 weight percent methylene chloride and was suitablefor use in a process in which methylene chloride was suitable, or can be washed with water to displace most of the methylene chloride by making a slurry of it with water one or more times. The water preferably contains 1 percent Na ₂ C0 ₃ and was at 70°C, then the filter cake was preferably washed with pure water at 70°Cto remove all residual MSA in the wet cake. Optionally, and alternatively, the filter cake was dried, first by using 30 psig (207 kPa) steam to strip MeCI ₂ from the cake then using a pressure of 35 psi (241.29 kPa) nitrogen at a temperature of at least 50°C blowing through the filter cake to dry the water from the cake. The pressed pressure filter had a mechanical agitator arm to stir up the cake and break up the clump to aid in the drying process.

The filter cake was optionally slurried back in water solution before the drying step to transfer out of the filter apparatus or was optionally dried and transferred out as a solid.

The BHPF was produced in 80 percent yield based on DCF and has a melting point of 223-225°C and the purity of 99 percent as determined by HPLC analysis when prepared as described, without the isomerization step. With the recycle of the adducts and isomers back to the isomerization steps, the overall yield of the process was 95 percent.

Those skilled in the art will recognize that a number of variations on these processes are within the scope of the invention. For instance, phenol may be added in solution (for instance, in methylene chloride) or as a solid. DCF can be added to phenol or other phenolic solution in the phenolation reaction in the form of a solid. The product p,p-BHPF can be recrystallized in methylene chloride. The chlorination reaction can be carried out in a continuous reactor instead of a batch reactor. The phenolation reaction can be carried out in a continuous reactor. Product BHPF can be recrystallized either in addition to or alternative to filtration; alternatively, the BHPF can be used without drying if used in a system where methylene chloride is an acceptable solvent. Similarly, solid separation of the reaction product slurry is optional and can be accomplished by any means within the skill in the art such as basket centrifuge, solid bowl centrifuge, other forms of solid separation and the like. A filter cake can also be washed in a slurry wash and/or displacement wash using fresh or recycled methylene chloride or other non-solvent; a slurry wash would involve stirring the filter cake with the non-solvent until a slurry is formed, and removal of said non-solvent, for instance by filtration or other solid separation technique.

EXAMPLE 15: Preparation of 9,9-Bis(4-methylphenyl)fluor ene; Alkylation of DCF onto Toluene using FeCI ₃ as Catalyst

Dichlorofluorene (10.0 g, 0.042 mole) (DCF) prepared as in Example 13 and ferric chloride (0.1 g, 0.0006 mole) were weighed into a 50 mL 2-necked flask fitted with a stirbar, nitrogen inlet, and thermometer. Toluene (50 mL) was added and the mixture was stirred and heated with a heating mantle. The mixture rapidly became dark red, and began to evolve HCI. At a temperature of 40°C, the solution vigorously evolved HCI. Analysis by GC (using a gas chromatograph commercially available from Varian Associates under the trade designation Model 3700, with a 30 meter column coated with a 1 micron layer of polydimethylsiloxane commercially available from J&W Scientific under the trade designation DB-1 ) at this point showed almost all of the DCF had reacted, and several heavy products were formed. The mixture was then heated at 50°C for 1 hour, after which, the solution was black in color. Analysis of the mixture by GC showed all the DCF has been reacted. The mixture was worked up by washing with water and then diluting the mixture (solution) with pentane, which caused the precipitation of most of the product as a tan powder. This powder was

filtered from the solution. The remaining mother liquor was then evaporated, and the resulting solid was slurried in pentane and filtered. The remaining mother liquor was evaporated to give an oily yellow solid. Weight of precipitated product was: first crop, 6.92 g; second crop, 2.64 g; total, 9.56 g (65 percent of theoretical); weight of yellow solid, 2.19 g, (15 percent of theoretical weight).

The H-NMR was consistent with a sample of 9,9-bis(methylphenyl)f luorene (MPF) having peaks relative to tetramethylsilane (TMS) at δ 2.1 - 2.5 (m, 6H, CH ₃ ), 6.6 - 7.9 (m, 16H). C- ^• 13 NMR: δ 151.63, 145.41, 143.17, 140.15, 138.44, 136.09, 129.15, 128.89, 128.33, 127.95, 127.79, 127.41, 126.12, 125.42, 120.21, 119.69, 65.42, (p,p isomer, quaternary C) 64.90 (o,p isomer, quaternary C), 21.59 (CH ₃ ),21.08 (CH ₃ ).

As determined by gaschromatography/mass spectroscopy (GCMS) the product mixture was: 77.3 percent bis(4-methyiphenyl)fluorene, 16.4 percent (4-methylphenyl)(2- methylphenyOfluorene, 6.2 percent bis(2-methylphenyl)f luorene. Primary peaks on GCMS for each compound were: (in atomic mass units, AMU, with percentage of height of base (largest) peak at 100 percent in parenthesis after the AMU) in parenthesis afterthe AMU : for bis(4- methylphenyDfluorene: 347(27.3); 346(100.0); 331(24.6); 255(17.3); 239(13.7); 65(14.0); for(4- methylphenyI)-(2-methylphenyl)fluorene: 347(28.2); 346(100.0); 331(16.9); 255(18.5); 253(11.3); 65(12.37); and for bis(2-methylphenyl)f luorene: 347(27.5); 346(100.0); 331(21.5); 255(18.0); 65(14.5).

EXAMPLE 16: Preparation of 9,9-Bis(4-methoxyphenyl)fluorene; Alkylation of DCF onto

Anisole using Sulfonic Acid Polymer Catalyst Dichlorofluorene (0.5 g, 0.002 mole) was weighed into a 50 mL 2-necked flask fitted with a stirbar, nitrogen inlet, and septum. Anisole (1.08 g, 0.01 mole) was added, and a GC (using a gas chromatograph commercially available from Varian Associates underthe trade designation Model 3700, with a 30 meter column coated with a 1 micron layer of polydimethylsiloxane commercially available from J&W Scientific under the trade designation DB-1) was taken of the resulting mixture as a standard. Activated Dow Fluorinated Sulfonic Acid (DFSA) pellets prepared by the process disclosed in U.S. Patent 4,791,081 and available from The Dow Chemical Company underthe trade designation XU 40036.01, heated at 170°Cfor 24 hours under vacuum (< 10 mm Hg) to activate, then stored under nitrogen, were added to the mixture. Then the mixture was heated in a water bath. At a temperature of 50°C, the pellets began to turn purple. When the mixture had reached 60°C, the solution had taken on a purple cast. The mixture was heated to 80°Cover 2 hours, at which time the solution was a yellow color, and the color of the DFSA beads was a light reddish color. Analysis of the mixture by GC shows that all the DCF had been reacted. The solution was decanted, and the excess anisole was removed under vacuum. The product, a thick yellow oil, was analyzed by H and C-13 NMR (Nuclear Magnetic Resource). The pattern observed in the aromatic region of the H-NMR was

consistent with para substitution of the anisole. The C-13 shows 12 carbons for the main product (a small amount of a byproduct was seen at the base of these peaks) which was the number of distinct carbon signals that would be predicted for the desired product. H-NMR: delta 3.79 (s, 6H, OCH ₃ ), 7.89-6.86 (m, 16 H), all from a standard of tetramethylsilane. C-13 NMR: delta 158.3, 151.9, 140.0, 138.2, 129.2, 127.6, 127.2, 126.1, 120.4, 1 13.6, 64.2 (quaternary C), 55.2 (OCH ₃ ) from a standard of tetramethylsilane.

EXAMPLE 17: Polymerization of DCF with Di phenyl carbonate (DPC) using TiCI ₃ Catalyst

DCF (9,9-Dichlorofluorene, 1.18 g, 0.005 mole) and DPC (diphenylcarbonate, 1.07 g, 0.005 mole) were weighed into a 25 mL two-necked flask fitted with a stirrer, heating mantle, and thermometer. Chloroform (4 mL) was added, followed by a catalytic amount of TiCI ₃ (0.03 g, 0.0002 mole) and the mixture was stirred and heated to 60°C over a period of 4 hours. After 30 minutes, the mixture had turned a dark red color, and was evolving HCI. After being stirred for 4 hours from the addition of the catalyst, the mixture solidified and forms brown paste. The paste was dissolved in dichloromethane (except some insoluble portions), and then diluted with acetone. This caused the precipitation of the product as a brown powder which was filtered from the yellow solution. The melting point of the powder was >280°C.

EXAMPLE 18: Polymerization of DCF with Diphenyloxide (DPO) using ZnCI ₂ as Catalyst.

DCF (9,9-Dichlorofluorene, 5.88 g, 0.025 mole) was weighed into a 150 mL resin kettle along with diphenyl oxide (DPO) (4.28 g, 0.0251 mole) and chloroform (10 mL). The kettle was fitted with a condenser and mechanical stirrer, and a starting GC (using a gas chromatograph commercially available from Varian Associates under the trade designation Model 3700, with a 30 meter column coated with a 1 micron layer of polydimethylsiloxane commercially available from J&W Scientific under the trade designation DB-1 ) was taken as a reference. The mixture was stirred and heated in a water bath (40-50°C). A small amount of ZnCI ₂ was added (about 0.02 g) and the mixture was stirred and heated at 50-70°C. After 30 minutes of heating the mixture had turned a dark green color, and was evolving HCI. Shortly thereafter, a thick solid paste was deposited on the walls of the kettle. This thick paste did not dissolve when 100 mL chloroform was added. The paste was triturated with acetone, which caused the solid to turn off-white in color. The off-white solid was filtered from the acetone and dried. The acetone was evaporated to yield a greenish semi-solid. The off-white polymer weighed 5 grams (60 percent of theoretical) and had a melting point (under a pressure of 68,000 kPa) of 250°C. The polymer was pressed into a thin, clear film at 250°C and 10,000 psig (700 kg/cm2). The film was brittle indicating low molecular weight. The Tg of the polymer was measured at 179°C (onset) by DSC (Differential Scanning Calorimetry).

EXAMPLE 19: Reaction of 9,9-Dichlorofluorene with Phenol to Prepare 9,9-

Bϊs(hydroxyphenyl)fluorene This reaction was carried out in a reactor constructed from fluorocarbon resin commercially available from E.l. du Pontde Nemours & Co. underthe trade designation

TEFLON PFA in the form of 2 inch (5.08 cm) diameter pipe 12 inches (30.48 cm) in length. At the top of the reactor was a 1/2 inch (1.27 cm) diameter port containing a ball valve through which phenol was added to the reactor. Also attached to the top of the reactor was a nitrogen purge line as well as a vent line (attached to a sodium hydroxide scrubber). Attached two inches (5.08 cm) above the bottom of the reactor was (1) a thermowell and (2) an inlet line which served as a point of HCI injection and/or sampling of the reaction mixture. Contained in the reactor was a star-shaped magneticstir bar which, when acted upon by an external magnetic stir plate, provided agitation to the reaction mixture. Approximately six inches (15.24 cm) from the bottom ofthe reactor was attached a feed line through which a solution of 9,9- dichlorofluorene (DCF) (or other dichloro compound) was passed into the reactor from an external holding tank. The reactor was wrapped with an electric heating tape attached to a variable voltage controller which serves to regulate the temperature ofthe reaction mixture. Molten (60°C) phenol (94.1 g, 1.0 mole) was poured into the reactor and the ball valve closed. While the phenol was stirred, the variable voltage controller was adjusted to maintain the phenol temperature between 40 and 45°C. A sample of 63.9 grams of a solution of DCF in carbon tetrachloride and containing 6.45 g (0.027 mole) of DCF, were placed in the DCF holding tank. When reactor temperature was stable, anhydrous HCI was passed slowly into the phenol until reactor pressure was approximately 15 psig (pounds per square inch gauge) (102 kPag). At this point, HCI addition was ceased and nitrogen pressure was used to force the DCF solution from the holding tank into the reactor. Total time for addition of the DCF solution was approximately 30 seconds. Reactor pressure was maintained at approximately 20 psig (136 kPag) by adjusting flow through the vent line with a needle valve. After 2 hours reaction time, the reactor contents were removed and quantitatively analyzed by reverse phase liquid chromatography (HPLC). Selectivity to p,p-BHPF was found to be 70 percent.

EXAMPLE 20: Effect of Phenol: DCF Molar Ratio

The procedure of EXAMPLE 19 was repeated except that 509.7 g of the DCF solution (0.23 mole DCF) were added to the reactor and reactor pressure was 90 psig (612 kPag). After 2 hours reaction time, the reactor contents were removed and treated with 167 g ϊsopropyl alcohol, then quantitatively analyzed by reverse phase liquid chromatography (HPLC). Selectivity to p,p-BHPF was 54 percent.

EXAMPLE 21 : Effect of Using Methanesulfonic Acid (MSA) as Catalyst

The procedure of EXAMPLE 20 was repeated except 100 g (1.06 mole) of phenol and 5 g methanesulfonic acid (MSA) were placed in the reactor and 267.8 g of a DCF/carbon tetrachloride solution (7.9 percent weight/weight in DCF) were added to the reactor and no HCI was passed into the reactor. After approximately 2 hours reaction time, the temperature of the reaction mixture was increased from 40-45°C to approximately 70°C and maintained at these conditions for another 17 hours. The reaction mixture was quantitatively analyzed by reverse phase liquid chromatography (HPLC). Selectivity to p,p-BHPF was 86 percent.

EXAMPLE 22: USE of MSA at Atmospheric Pressure

A sample of 25.0 g (0.266 mole) phenol and 7.5 g methanesulfonic acid (MSA) were stirred in a 3-necked 250 mL round-bottom flask equipped with a Dewar condenser containing dry ice. A solution of DCF was prepared by adding 25.03 g (0.106 mole) DCF to a mixture containing 39.9 g methylene dichloride and 20.1 g chloroform. When the temperature of the phenol/MSA mixture was approximately 30°C, the DCF solution was added to the flask, via dropping funnel, over a 20 minute period. After 2 1/2 hours reaction time, the reaction mixture was heated with a heating mantle to a temperature of 38-40°C and maintained at this temperature for 3 hours. The solution was then allowed to cool overnight. The reaction mixture was filtered and the filter cake washed with methylene chloride to yield a white solid which, after drying to constant weight at 60°C, yields 20.8 g of a white solid. HPLC analysis indicated the white solid to be 96 percent p,p-BHPF by peak area. Quantitative analysis of the filtrate by HPLC revealed 10.5 g p,p- BHPF to be dissolved in the filtrate. Overall selectivity to p,p-BHPF was 84 percent.

EXAMPLE 23: Effect of Water

The procedure of EXAMPLE 22 was repeated except a) 22.0 g (0.234 mole) phenol was used, b) the methylene chloride was saturated with deionized water prior to preparing the DCF solution and c) reaction temperature was 25°C. After 2 hours selectivity to p,p-BHPF was 62 percent.

EXAMPLE 24: Effect of Sulfuric Acid

A sample of 40.0 g (0.425 mole) phenol was added to a 250 mL Erlenmeyer flask containing a magnetic stir bar and fitted with a thermometer. With stirring, 5.6 g (0.06 mole) 96 percent sulfuric acid and 0.1 mL (1.15 millimole) β-mercaptopropionic acid were added to the phenol. A sample (25.0 g 0.106 mole) DCF was added as a solid to the phenol/acid mixture over a 20 minute period during which the temperature of the reaction mixture never exceeded 50°C. After 2 hours reaction, the reaction mixture was dissolved in isopropyl alcohol and quantitatively analyzed by HPLC. Selectivity going to p,p-BHPF was 70 percent.

EXAMPLE 25: Effect of Tem erature

A sample of hot phenol (249.2 g, 2.65 mole, at a temperature of 70°C) was placed in a 1 liter, 3-necked round bottom flask fitted with a thermometer, distillation arm and a dropping funnel. While the contents were stirred and heated to approximately 100°C, anhydrous HCI was bubbled into the phenol. A sample of (250 mL) of a DCF/CCI4 solution which contains 0.089 mole of DCF (as determined by HPLC analysis) was charged to the dropping funnel. When the temperature of the phenoI/HCI solution has stabilized at 98 to 100°C,the DCF solution was slowly dripped into the phenol/HCI solution. As the DCF solution was added, a distillate was collected from the reaction mixture. When all ofthe DCF had been added, the HCI flow was discontinued and the contents ofthe flask were allowed to cool. After 15 hours total reaction time, the reaction mixture was quantitatively analyzed by HPLC. Selectivity was 66 percent going to p,p-BHPF.

EXAMPLE 26: Use of Trϊf luoromethanesulfonic (Triflic) Acid as Catalyst

The procedure of EXAMPLE 25 was repeated except that (a) 0.5 m L of triflic acid was added to the flask in lieu of HCI, and (b) 750 mLof the DCF/CCI4 solution were placed in the dropping funnel. Selectivity was 50 percent relative to p,p-BHPF.

EXAMPLE 27: Use of Trifluoromethanesulfonic Acid as Catalyst

To a 100 mL 3-necked flask equipped with a thermometer, magnetic stir bar, condenser and a nitrogen inlet through which a positive nitrogen sweep was maintained, 32.4g of phenol was added. The temperature was adjusted to 41 °C and nitrogen was swept through the system for ten minutes. Atthis point, triflic acid (0.1 mL) was added to the flask. Solid crystals of 9,9-dichlorofluorene, obtained by evaporating the solvent from the DCF/CCI ₄ solution used in EXAMPLE 13, were added overthe next hour. The temperature ofthe reaction mixture was maintained at40°Cfor four more hours and then sampled for HPLC analysis which shows a 91 percentyield.

EXAMPLE 28: Use of Ethyl Acetate as Solvent

The procedure of EXAMPLE 18 was repeated except 1) a mixture of 91 g ethyl acetate and 100 g (1.06 mole) phenol were added to the reactor, 2) 24 g (0.102 mole) DCF were dissolved in 123 g ethyl acetate and placed in the DCF holding tank, 3) the DCF solution was added overa 5 minute period, 4) HCI addition was continued during DCF addition, reaction temperature was allowed to vary from room temperature to 41 °C and 5) reactor pressure was not controlled (never exceeds 22 psig (149.6 kPag)). After 16 hours reaction time, the reaction mixture was collected and quantitatively analyzed by HPLC. Selectivity to p,p-BHPF was 53 percent.

EXAMPLE 29: Use of Isopropanol as Solvent

A sample of 100 g isopropyl alcohol and 208 g (2.21 mole) phenol were combined and saturated with anhydrous HCI in a 500 mLerlenmeyer flask. A solution of 24 g (0.102 mole) DCF in 200 g carbon tetrachloride was added to the stirred phenol/alcohol solution over a 50 minute period during which the reaction mixture was continuously sparged with anhydrous HCI. After 95 minutes reaction time, the reaction mixture was collected and quantitatively analyzed by HPLC. Selectivity to p,p-BHPF was 69 percent.

EXAMPLE 30: Effect of Pressure, Temperature, and Time on Selectivity

The procedure of EXAMPLE 18 was repeated except temperature was controlled at 60-70 ^c C and reactor pressure was controlled at 80-90 psig (612 kPag) and reaction time was 15.5 hours. Selectivity to p,p-BHPF was 85 percent.

EXAMPLE 31 : Use of a Recirculating Reactor

The basic reactor configuration was 1 ) a 2 inch (5.08 cm) diameter pipe of fluorocarbon resin commercially available from E.I. du Pont de Nemours & Co. under the trade designation Teflon PFA which served as the mixing tank, 2) a heat exchanger of the same material 3) a feed tank in which a solution of 9,9-dichlorofluorene was stored prior to starting the phenolation reaction and 4) a pump which continually circulated the reactor contents through the heat exchanger and mixing tank.

A solution of 57 g (0.6 mole) of phenol in approximately 236 mL of CCI ₄ was placed in the reactor mixing tank. After the reactor was sealed, the pump was energized, and the solution was allowed to circulate through the system. At this point an ethylene glycol/water mixture commercially available from The Dow Chemical Company under the trade designation Ambitrol ^* " 50 which has been cooled to 10°C, was admitted to the heat exchanger to maintain a reactor temperature of 17°C. Upon stabilization of the CCU _/ phenoi solution temperature, anhydrous HCI was admitted into the headspace of the mixing tank, and reactor pressure was regulated at 10 psig (68 kPag) by means of a flow meter on the vent line. A solution of CCl ₄ /9,9-dichloro fluorene (DCF) was prepared by dissolving 14.1 g (0.06 mole) of DCF in approximately 88 mL of CCI4. The solution was charged to the DCF feed tank and pressured to 30 psig (204 kPag) with nitrogen. A flow meter connecting the DCF feed tank to the suction of the circulating pump was then opened to admit the DCF/CCI4 solution into the reactor system at a rate of 3.1 g/minute. Samples of the reaction mixture were periodically removed via the sampling line and analyzed by reverse phase liquid chromatography (HPLC) to determine extent of reaction and distribution of reaction products. After 23 hours reaction time, the reaction mixture was analyzed by HPLC. p,p-BHPF comprised 44 percent of all reaction products as measured by peak area.

EXAMPLE 32: Effect of Phenol Concentration and Temperature

The procedure of EXAMPLE 31 was repeated except: a) 191.3 g (2.04 mole) phenol was added to the reactor followed by 180 mL CCI4, b) 250 mL of a DCF/CCI4 solution containing 0.089 mole of DCF (as determined by HPLC analysis) was charged to the DCF feed tank, c) reactor temperature was controlled at 64°C. After 6 hours, the reaction mixture was analyzed by HPLC; analysis showed p,p-BHPF comprised 49 percent of all reaction products as measured by peak area.

EXAMPLE 33: Use of Glacial Acetic Acid as Solvent

The procedure of EXAMPLE 31 was repeated except that 100 g (1.1 mole) of phenol and 100 mL of glacial acetic acid were placed in the reactor mixing tank. The reactor was sealed, the pump was energized, and the solution was allowed to circulate through the system. Atthis point an ethylene glycol/water mixture commercially available from The Dow Chemical Company underthe trade designation Ambitrol'" 50, which has been cooled to 25°C, was admitted to the heat exchanger to maintain a reactor temperature of 30°C. Upon stabilization ofthe phenol temperature, anhydrous HCI was injected into the suction ofthe circulation pump until reactor pressure was 20 psig (136 kPag), at which point the HCI flow was discontinued. A solution of CCl4/9,9-dichloro fluorene (prepared by dissolving 24 g (0.1 mole) of DCF in approximately 60 mL of CCI4 and 200 mL glacial acetic acid) was charged to the DCF feed tank and pressured to 40 psig (272 kPag) with nitrogen. A flow meter connecting the DCF feed tank to the suction of the circulating pump was then opened to admit the DCF/CCI4 solution into the reactor system at a rate of 5.5 g/min. The reaction mixture was analyzed by HPLC. Selectivity to p.p-BHPF was 27 percent.

EXAMPLE 34: Use of Catalyst Dissolved in DCF

A sample of 24.0 g (0.102 mole) DCF was dissolved in 88 g ethyl acetate. This solution was sparged with anhydrous HCI until 13.1 g HCI has dissolved in the solution. A sample of 297 g (3.16 mote) molten (60°C) phenol was placed in a 1-L Erienmeyer flask and stirred. The ethyl acetate/DCF/HCI solution was added slowly to the phenol. HPLC analysis indicated 84 percent of product peak area was that of p,p-BHPF.

EXAMPLE 35: Reverse Addition: Phenol into DCF

The procedure of EXAMPLE 34 was repeated except that the ethyl acetate/DCF/HCI solution was placed in a 500 mL Erienmeyer flask and stirred while HCI was sparged through the solution. Molten (60°C) phenol was added to the ethyl acetate/DCF/HCI solution over a 1 hour period. After 4.3 hours, HPLC analysis showed p,p-BHPFto be 74 percent ofthe products' peak areas.

EXAMPLE 36: Use of Propylene Carbonate as Solvent

A sample of 1.1 g (0.012 mole) phenol was dissolved in 1.0 g propylene carbonate/0.1 mL β-mercaptopropionic acid (BMPA). A solution of 1.25 g (0.0053 mole) DCF in 2 g methylene chloride/1 g chloroform was added to the propylene carbonate/phenol/BMPA solution. Anhydrous HCI was slowly sparged into the mixture to initiate reaction. After 1.5 hours, HPLC analysis showed p,p-BHPF to constitute 62 percent of the products' peak areas.

EXAMPLE 37: Direct Addition of Phenol to a Chlorination Reaction to Obtain p,p-BHPF The reactor was a 1000 mL cylinder 4 inches in diameter (100mm) by 5.5 inches in height (140mm) equipped with a 2 inch (50mm) diameter turbine impeller driven by a vertical shaft. Stirring rate was measured by a tachometer. Temperature was controlled by a 10 foot (3.048 m) by 0.25 inch (0.635 cm) external diameter coil immersed in the reaction medium through which coolant was pumped maintained at a constant temperature by a circulating refrigerated/heated bath.

The temperature was measured by a thermocouple inside a thermowell which runs the entire depth of the reactor. The reactor was also equipped with a nitrogen inlet which was used to maintain a nitrogen atmosphere above the reaction solution. The entire apparatus was constructed of polytetrafluoroethylene/copolyheptafluoropropyl trifluorovinyl ether commercially available from E.I. du Pont de Nemours & Co. underthe trade designation Teflon PFA .

The reactor was flushed with nitrogen and a solution of fluorene (33.24 g, 0.200 mole) and carbon tetrachloride (299.20 g, 1.9451 mole, 187.70 mL) was charged to the reactor followed by NaOH (30 percent solution in water, 53.33 g, 0.40 mole, 40.10 mL, 16.00 g dry weight). The stirrer was started, and the speed adjusted to 4000 rpm. The coolant was admitted to the coils, and the temperature of the reaction solution was adjusted to 30°C. The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 2.59 g, 0.004 mole, 2.57 mL) was added at once. The reaction mixture was sampled after 30 minutes and analyzed by GC according to the procedure of EXAMPLE 1 ; analysis shows 0.38 percent fluorene, 98.74 percent 9,9-dichlorofluorene, and 0.88 percent 9-fluorenone. Phenol (41.41 g, 0.44 mole, 38.70 mL) in 42.46 mL of CCI4 was added to the reactor and stirring was continued. After 30 minutes the reaction mixture was sampled and analyzed by reverse phase liquid chromatography (HPLC) as in EXAMPLE 19 which showed the composition of the mixture to then be 41.70 percent 9,9-bis(4-hydroxyphenyl)fluorene, 2.45 percent o,p-BHPF, 5.42 percent 9-fluorenone, and 4.76 percent 9,9-dichlorofluorene.

EXAMPLE 38: Addition of Phenol as Phenolate in Aqueous Solution Directly to the

Chlorination Reaction to Obtain the p,p-BHPF. The reactor described in EXAMPLE 37 was flushed with nitrogen, and a solution of fluorene (33.24g, 0.200 mole), ethylbenzene (188.38 g, 1.7743 mole, 217.28 mL) and carbon tetrachloride (61.53 g, 0.400 mole, 38.60 mL) was charged to the reactor followed by NaOH (50 percentsolution in water, 160.0 g, 2.00 mole, 103.9 mL, 80.00 g dry weight). The stirrerwas started, and the speed adjusted to 3000 rpm. The coolant was admitted to the coils, and the temperature of he reaction solution was adjusted to 30°C The catalyst, tetrabutylammonium hydroxide (40 percentsolution in water, 2.59 g, 0.004 mole, 2.57 mL), was added at once. The reaction mixture was sampled after 1 minute and periodically thereafter for the next 30 minutes and analyzed by GC according to the procedure used in EXAMPLE 1. The results of these analyses were shown in Table 2.

Table 2

The data in Table 2 show that the reaction was essentially complete within one minute. GC analysis at one minute showed 1.23 percent fluorene, 97.78 percent 9,9- dichlorofluorene, and 0.59 percent 9-fluorenone. After 30 minutes, phenol (37.64 g, 0.40 mole, 35.18 mL) in 10.39 mL of 50 percent NaOH (0.40 mole, 16.00 g dry weight) was added to the reactorand stirring continued. After 30 minutes analysis showed 76.97 percent DCF remaining and 1.96 percent BHPF. The mixture was allowed to stir overnight; then analysis showed 100.0 percent 9,9-bis(4-hydroxyphenyl)f luorene.

EXAMPLE 39: Preparation of Bis(aminophenyl)fluorene into a two-neck round bottomed flask provided with a thermometer and a condenser were placed 9,9-dichlorofluorene (10.0 g, 0.0425 mole) and aniline (50.0 g, 0.537 mole). The resulting mixture was a slowly heated over a period of 30 minutes to 60°C in an oil bath with stirring using a magnetic stirrer. A rapid reaction took place with an exotherm (—110°C) and resulted in complete disappearance of dichlorofluorene and appearance of one product as confirmed by GC/MS (gas chromatographic mass spectrometry), showing a primary peak at 254-257 AMU,to be 9-chloro-9-(aminophenyl)fluorene. Further heating at 135°C for three hours resulted in complete conversion of the monoamine to the diamine. Excess aniline was flash distilled under vacuum, and the resulting residue was washed with 5 percent sodium hydroxide, filtered, and washed with hexane to give 14.0 g (94 percent yield) of the diamine, which was found to be a mixture of 92.23 percent p,p'-, 2.17 percent o,p- and 4.59 percent n,p- isomers(gas chromatography); m.p. 234-235°C. The mass spectrum showed a primary peak at 348 AMU.

The diamine was recrystallized by pouring it into boiling toluene and cooling the resulting mixture to precipitate pure p,p'-isomer , m.p. 234-235°C; NMR ( • H and ¹ 3Q data was consistent with 9,9-bis(p-aminophenyl)fluorene structure. The 1 H NMR of the compound in d6- DMSO (deutero-dimethylsulfoxide) showed peaks at δ 4.90 (singlet, NH ₂ , 2H), δ 6.40 (doublet, aromatic, 4H), δ 6.75 (doublet, aromatic, 4H), δ 7.2-7.3 (multiplet, fluorene, 6H), δ 7.8-7.9 (multiplet, fluorene, 2H), all from TMS (tetramethylsilane) standard.

When the procedure was repeated except that heating at 130°C was continued fora period of 7 hours rather than 3 hours, the product was 98 percent p,p'- and 2 percent N,p- isomer. This data indicated that heating converts the N,p- isomer to the p,p' isomer.

EXAMPLE 40: Effect of Recycling Byproducts to Increase para, para-BHPF

In this EXAMPLE, recycle was illustrated by 5 batch reactions. Reaction #1 : To a 100 mL glass flask equipped with a thermometer, stirring paddle, and a cooling /heating glycol jacket were added 15.98 grams (0.170 moles) of phenol in 8.60 grams of chloroform plus 3.07 grams of methanesulfonic acid (MSA) to form a solution. This solution was stirred and kept at 24°C. To this solution were added 18.04g (0.0768 moles) of

9,9-dichlorofluorene (DCF) in 18.1 grams chloroform plus 38.4 grams methylene chloride over a period of 75 minutes. Afterthe DCF was added, the temperature was 27°C. Thattemperature was maintained for 60 minutes, then the solution was cooled to 10°C which was maintained 60 minutes- Crystals form and were collected on a glass filter frit and washed with a total of 100 mL of methylene chloride and then a total of 100 grams of warm water. The resulting 10.7 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 - 225°C. This melting point indicated thatthe isomer was p,p-BHPF. The f iltrate/methylene chloride wash was used for the recycle reaction below.

Reaction #2: Recycle: Eighty-four percent ofthe filtrate/methyl ene chloride wash from Reaction #1 was added back to the glass reactor used in Reaction #1 with 7.22 grams (0.0767 moles) of phenol and 0.49 grams of MSA. The resulting solution was heated to 70°C which was maintained for one hourwhile most of the methylene chloride and chloroform were removed by vacuum distillation. Then the remaining mixture was cooled to 24°C and 5 g of methylene chloride were added. Then the mixture was seeded with 0.005 g BHPF crystals to promote crystallization. After 10 minutes, 8.89 grams (0.0378 moles) of DCF dissolved in a mixture of 9.0 g CHCI3 and 19 g CH ₂ CI ₂ were added over a period of 108 minutes. The temperature was raised to 40°C and maintained at that temperature for 60 minutes, then cooled to 10°Cand maintained at that temperature for one hour. The resulting crystals were collected on a glass filter frit and washed as in Reaction #1. The resulting 13.1 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 - 225°C. The filtrate/methylene chloride wash was used for the next recycle reaction, Reaction #3.

Reaction #3 - Recycle: Eighty-five percent ofthe filtrate/methylene chloride wash from Reaction #2 was added back to the glass reactor used in Reactions #1 and #2, with 8.36 grams (0.0888 moles) of phenol and 0.46 grams of MSA. The resulting solution was heated to 70°Cand maintained at that temperature for one hourwhile most of the methylene chloride and chloroform were removed by vacuum distillation. Then the solution was cooled to 24°C, and 5 g of methylene chloride were added. Then the resulting mixture was seeded with 0.005 g BHPF crystals. After 10 minutes, 10.23 grams (0.0435 moles) of DCF in the same solvent mix as Reaction #2were added overa period of 130 minutes. The temperature was raised to 40°C and maintained at thattemperature for 60 minutes, then cooled to 10°Cand maintained at that temperature for one hour. The resulting crystals were collected on a glass filter frit and washed as in Reaction #1. The resulting 11.9 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 -225°C. The filtrate/methylene chloride wash was used for the next recycle reaction, Reaction #4.

Reaction #4 - Recycle: Eighty-five percent ofthe filtrate/methylene chloride wash from Reaction #3 was added back to the glass reactor used in Reactions #1 and #3 with 7.92 grams (0.0842 moles) of phenol and 0.46 grams of MSA. The resulting solution was heated to

70°C and maintained at that temperature for one hourwhile most of the methylene chloride and chloroform were removed by vacuum distillation. Then the solution was cooled to 24°C and seeded with 0.005 g BHPF crystals. After 10 minutes, 9.54 grams (0.0406 moles) of DCF in the same solvent mix used in Reaction #2 were added over a period of 1 10 minutes. The temperature was raised to 40°C and maintained at that temperature for 60 minutes, then cooled to 10°Cand maintained at that temperature for one hour. The resulting crystals were collected on a glass filter frit and washed as in Reaction #3. The resulting 1 1.4 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 - 225°C. The filtrate/methylene chloride wash was used for the next recycle reaction, Reaction #5.

Reaction #5 - Recycle: Eighty-five percent of the filtrate/methylene chloride wash from Reaction #4 was added back to the glass reactor used in Reactions #1 and #4 with 7.58 grams (0.0805 moles) of phenol and 0.45 grams of MSA. The resulting solution was heated to 70°C and maintained at that temperature for one hourwhile most of the methylene chloride and chloroform were removed by vacuum distillation. Then the solution was cooled to 24°C, and 5 g of methylene chloride were added. Then the mixture was seeded with 0.005 g BHPF crystals. After 10 minutes, 9.21 grams (0.0392 moles) of DCF in the same solvent mix as Reaction #2 were added over a period of 1 15 minutes. The temperature was raised to 40°C and maintained at that temperature for 60 minutes, then cooled to 10°C and maintained at that temperature for one hour. The resulting crystals were collected on a glass filter frit and washed as in Reaction #1. The resulting 1 1.0 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting poi nt of 223 - 225°C.

The results of Reactions #2 - #5 shows the usefulness of a recycle process to convert the byproducts of the reaction to BHPF using MSA as the catalyst. A steady state was produced where the conversion rate to BHPF was 95 percent, [(moles of BHPF/moles of DCF feed -15 percent, which was removed) X 100 = 95 percent]

EXAMPLE 41 : Recycle to form p,p-BHPF Exemplified in 3 Reactions

Reaction #41 : 1 To a 100 mL glass flask equipped with a thermometer, stirring paddle, and a cooling /heating glycol jacket were added 15.98 grams (0.170 moles) of phenol in 8.60 grams of chloroform plus 3.07 grams of methane sulfonic acid (MSA). The resulting solution was stirred and heated to 68°C. To the solution, were added 18.04 g (0.0768 moles) of dichlorofluorene (DCF) in 18.1 grams chloroform plus 38.4 grams methylene chloride over 60 minutes. After the DCF was added, the temperature was kept at 68°C and maintained at that temperature for 30 minutes then slowly dropped to 45°C over a period of 20 minutes. Then the solution was seeded with 0.005 g BHPF crystals. The temperature was decreased to 10°C while 26 grams of methylene chloride were added. The temperature was maintained at 10°C for 60 minutes. The resulting crystals were collected on a glass filter frit and washed with 100 g

methylene chloride and then with 100 g warm water. The resulting 12.9 grams of BHPF were determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 - 225°C. The filtrate/methylene chloride wash was used for the recycle reaction, Reaction #41:2. Reaction #41 :2 - Recycle: To the glass reactor described in Reaction 41 : 1 was added the filtrate/methylene chloride wash from Reaction #41 :1 with 8.12 grams (0.0863 moles) of phenol to form a solution. The solution was stirred and heated to 68°C To the solution, were added 10.04 g (0.0427 moles) of DCF in 10.0 grams chloroform plus 21.4 grams methylene chloride over 67 minutes. Afterthe DCF was added, the temperature was maintained at 68°C and maintained at thattemperature for 50 minutes then slowly dropped to 45°C Then, 5 g of methylene chloride were added and the mixture was seeded with 0.005 g BHPF crystals. The temperature was decreased to 10°C white 25 grams of methylene chloride were added. The temperature was maintained at 10°C and maintained at thattemperature for 60 minutes. The resulting crystals were collected on a glass filter frit and washed with methylene chloride and then warm water as in Reaction 4-1 :1. The resulting 10.6 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals had a melting point of 222 - 225°C. The filtrate/methylene chloride wash was used for the recycle reaction, Reaction 41:3.

Reaction #41 :3 - Recycle: To the glass reactor used in Reactions #41 : 1 and #41 :2 was added the filtrate/methylene chloride wash from Reaction #41 :2 with 5.72 grams (0.0608 moles) of phenol to form a solution. This solution was stirred and heated to 68°C and maintained at that temperature for 90 minutes. The temperature was decreased to 30°C, then 7.04g (0.0300 moles) of DCF in 7.14 grams chloroform plus 15.1 grams methylene chloride were added over a period of 90 minutes. Afterthe DCF had been added, the temperature was lowered to 10°Cand maintained at 10°Cfor60 minutes. The resulting crystals were collected on a glass filter frit and washed with methylene chloride and then warm water as was Reaction 41 : 1. The resulting 10.0 grams of BHPF was determined to be 99 percent pure by HPLC analysis, and the crystals have a melting point of 223 - 225°C.

The results of Reactions #41 :2 - # 41 :3 show that conversion of byproduct to BHPF takes place and thatrecycle is, therefore, useful.

EXAMPLE 42: Use of Tetraalkylamrnonium Hydroxide as Base in Chlorination of Fluorene

The reactor used in EXAMPLE 1 was flushed with nitrogen. A solution of fluorene (12.50 g, 0.0752 mole), ethylbenzene (112.50 g, 1.0596 mole, 129.76 mL) and carbon tetrachloride (23.14 g,0.150 mole, 1.451 mL) was charged to the reactor. The stirrerwas started and the speed adjusted to 500 rpm. The temperature ofthe reaction solution was 30°C The catalyst, tetrabutylammonium hydroxide (40 percentsolution in water, 24.39 g, 0.0376 mole, 24.15 mL) was added at once. The reaction mixture was sampled after 5 minutes and periodically

thereafterforthe next 30 minutes and analyzed by GC by the procedure described in EXAMPLE 1. The results ofthese analyses were shown in Table 3. The reaction was essentially complete within thirty minutes. GC analysis at five minutes showed that the reaction mixture contains 33.26 percent fluorene, and 66.74 percent 9,9-dichlorofluorene. After 30 minutes GC analysis showed that the reaction mixture contains 0.14 percent fluorene, and 99.86 percent 9,9- dichlorofluorene.

Table 3

This example shows that organic bases like tetraalkylamrnonium hydroxides were useful in chlorination processes of the invention.

EXAMPLE 43: Use of an Ion Exchange Catalyst

To a 100 mL flask was added 20 grams (0.2125 moles) of phenol plus 13 grams of CCU and 6.67 grams of a dried acid ion exchange resin commercially available from The Dow Chemical Company underthe trade designation MSC-1. The flask was heated to 40°C, and 5.15 grams (0.0219 moles) of DCF in 60.8 grams of CCI ₄ /CHCI3 were added over a one hour period. The temperature was raised to 60°C and maintained for a period of one hour. HPLC analysis indicated 84 percent selecti ity to p,p-BHPF.

EXAMPLE 44: Use of an Acid Clay Catalyst

To a 100 mL flask was added 20 grams (0.2125 moles) of phenol plus 14 grams of CC1 ₄ and 6.00 grams of a dried clay acid catalyst commercially available from Harshaw/Filtrol under the trade designation Filtrol-22. The flask was heated to 40°C, and 5.07 grams (0.0216 moles) of DCF in 59.9 grams of CCI_t/CHCI ₃ were added over a one hour period. The temperature was kept at 40°C for 20 minutes then cooled to 24°C. The BHPF crystals were collected by filtration, leaving a filtrate, and dried at 40°C for 16 hours under a vacuum of 28 in Hg (6.7 kPa) to give 5.7 grams of product. The catalyst was washed with acetonitrile. HPLC

analysis ofthe acetonitrile plus the filtrate shows 1.10 grams of BHPF inthe solution. Total selectivity was 90 percentto p,p-BHPF.

EXAMPLE 45: Use of a Fluorocarbon Sulfonic Acid Catalyst To a 100 mLflask was added 18.3 grams (0.1945 mol) of phenol plus 18.5 grams of

CCUand 10.0 grams of fluorocarbon sulfonic acid catalyst (0.139 meq/g) prepared by the process disclosed in U.S. Patent 4,791,081 and available from The Dow Chemical Company underthe trade designation XU-40036.01 (DFSA). The flask was heated to 40°C and 4.95 grams (0.0210 moles) of DCF in 58.5 grams of CCI4/CHCI3 were added over 7.5 hours. HPLC analysis indicated 90 percent selectivity to p,p-BHPF.

EXAMPLE 46: Effect of Metals on Chlorination

The reactor was a 500 mL 3-neck round bottom flask equipped with a magnetic stir bar, nitrogen purge and thermometer. The reactor was flushed with nitrogen and a solution of fluorene (15.00 g, 0.090 mole) and carbon tetrachloride (234.89 g, 1.52 mole, 147.36 mL) was charged to the reactor followed by NaOH (30 percentsolution in water, 6.65 g, 0.050 mole, 5.00 mL, 2.00 g dry weight). The stirrerwas started, and the speed adjusted to 500 rpm. The mixture was stirred with a subsurface nitrogen sparge. The temperature ofthe reaction solution was 27°C. The catalyst, tetrabutylammonium hydroxide (40 percent solution in water, 4.95 g, 0.0076 mole, 4.90 mL) was added at once. The reaction was followed by gas chromatography (GC) as in Example 1. After 5 minutes, GC analysis showed that the reaction mixture contained 74.95 percent fluorene, and 25.05 percent 9,9-dichlorofluorene. The reaction was sampled and analyzed periodically over the next several hours until no fluorene remains according to the GC analysis. This procedure was repeated with the addition of 10 g of each ofthe metals indicated in Table 4to the flask before any ofthe other reagents. The results were tabulated in Table 4. The 304-stainless steel and 316-stainless steel were in the form of washers whereas the other metals were in the form of cuttings produced from drilling operations. These cuttings had much greater surface area than the washers but still had less inhibitory effect on the reaction than the stainless steel washers.

Table 4 Percent 9 , 9-DCF FORMED IN PRESENCE OF METAL

*SS = Stainless Steel

15 In all cases tabulated where a metal was present, the reaction does not go to completion even after 17 hours; whereas, the reaction in glass with no metal present was 99.95 percent complete at three hours and all fluorene was converted at 17 hours. The large surface area ofthe cuttings was believed to contribute to mixing thereby resulting in higher initial rates of reaction for the mixtures where cuttings were present. 0

EXAMPLE 47: Use of Methylene Chloride as Solvent in Alkylation and Isolation without

Water Wash A sample of 23.1 g (0.246 mole) phenol was placed in a 500 mL 3-necked round bottomed flask equipped with a magnetic stir bar, thermometer and dropping funnel. A ⁵ solution prepared from 25.03 g (0.107 mole) DCF and 59 g methylene chloride was placed in the dropping funnel. While the phenol was still fluid (at 35°C), drop-wise addition of the DCF solution was started and continued over a period of 90 minutes until all the solution had been added. Reaction temperature was maintained at room temperature (about 25°C) during the course ofthe reaction. After 3 hours total reaction time, the reaction mixture was filtered to 0 yield a solid which, after rinsing with methylene chloride and drying to constant weight at 60°C, weighs 20.8 g. HPLC analysis indicated the white solid to be greater than 95 percent p,p- BHPF by peak area. Quantitative analysis of the filtrate by HPLC indicated 3.5 g p,p-BHPF still dissolved in the filtrate. Selectivity to p,p-BHPF was 66 percent.

5 EXAMPLE 48: Non-aqueous Isolation of Alkylation Product where Chloroform was

Alkylation Solvent To a 100 mL flask was added 6.2 grams (0.66 moles) of phenol plus 3.35 grams of CHCI ₃ (chloroform). The flask was cooled to 11°C, and 6.0 grams (0.0255 moles) of DCF in 3.14

grams of CHCf ₃ and 13.1 grams of CH ₂ CI ₂ was added within 5 seconds. The temperature was kept at 1 TCfor 4 hours, then resulting BHPF crystals were collected on a glass filter frit, washed with 9 grams of CH ₂ CI ₂ and dried to give 4.63 grams of BHPF.

EXAMPLE 49: Continous Method for the Preparation of 9,9-Dichlorofluorene with

Recovery and Recycle ofthe Phase Transfer Catalyst and Sodium Hydroxide Part A:

To the reactor described in Example 13 which has been purged with nitrogeri was charged a volume of carbon tetrachloride (400 mL) such that its level just came to the bottom of the sixth stage. NaOH solution (25 percent by weight, 6.14 moles, 245.6 g dry weight, 983 g solution weight, 774.0 mL) was then charged to the reactor, and filled all six ofthe stirred zones. The stirrerwas started and its speed was adjusted to 1500 rpm. Fluorene (0.185 moles, 30.75 g) dissolved in carbon tetrachloride (2.24 moles, 344.8 g, 225.35 mL) was fed into a vertical catalyst saturator which was a cylinder 1 inch (2.54 cm) in diameter by 12 inches (30.48 cm) long containing the catalyst solution (tetrabutylammonium hydroxide (0.132 moles, 34.30 g dry weight, 85.75 g as 40 percent aqueous solution)) such thatthe feed solution fell through the aqueous catalyst solution before entering the first reactor stage, using a metering pump at a rate of 5.5 mL/min. The product solution was collected at the overflow and analyzed by gas chromatography (GC) according to the procedure of Example 13. When fluorene was no longer detected in the effluent stream, the product was fed to a wash column which was identical in design to the reactor column. The product solution was fed to the first stirred stage at a rate of 5.5 mL min and water was fed at the sixth stirred stage at a rate of 11.0 mL/min. The organic solution containing the 9,9-dichlorofluorene was collected at the overflow ofthe wash column and then passed through a column of molecular sieves (4A size, commercially available from Linde Division, Union Carbide Industrial Gasses Inc.) which lowered the water content of this stream as measured by Karl-Fisher titration from 208 ppm to 6.3 ppm. The total productsolution collected inthis fashion amounts to 1026 g which was evaporated to dryness on a rotary evaporator, leaving 42.52 g of very light yellow crystals, 9,9-dichlorofluorene (97.76 percent ofthe theoretical yield of 43.49 g). The aqueous solution from the wash column was evaporated on the rotary evaporator (40°C/10 mm of Hg (1.32 kPa)) leaving 33.4g of a clear oil which crystallizes (long needles) on cooling (97.4 percent recovery of the tetrabutylammonium hydroxide charged (34.30 g dry weight)).

Recycle run #1.

The tetabutylammonium hydroxide recovered from the above reaction (33.4 g, 0.129 moles) was dissolved in 57 mL water and charged to the catalyst saturator.

To the reactor which had been purged with nitrogen was charged a volume of

5 carbon tetrachloride (400 mL) such that its level just came to the bottom of the sixth stage. NaOH solution (the same solution used in the previous run (Example 49), Part A), 25 percent by weight, 6.14 moles, 245.6 g dry weight, 983 g solution weight, 774.0 mL) was then charged to the reactor, and filled all six of the stirred zones. The stirrer was started and its speed adjusted to 1500 rpm. Fluorene (0.185 moles, 30.75 g) dissolved in carbon tetrachloride (2.24 moles,

10 344.8 g, 225.35 mL) was fed into the catalyst saturator containing the catalyst solution using a metering pump at a rate of 5.5 mL/min. The product solution was collected at the overflow and washed and dried as in Part A. The total product solution collected in this fashion amounted to 1056 g which was evaporated to dryness on a rotary evaporator leaving 42.66 g of very light yellow crystals, 9,9-dichlorofluorene (98.1 percent of the theoretical yield of 43.49 g).

15 The aqueous solution from the wash column was evaporated on the rotary evaporator (40 C/ 10 mm of Hg

(1.32 kPa)) leaving 32.7 g of a clear oil which crystallizes (long needles) on cooling (98.0 percent recovery ofthe tetrabutylammonium hydroxide charged (33.40 g dry weight).

0 Recycle run #2.

The tetrabutylammonium hydroxide recovered from the Recycle run #1 reaction (32.7 g, 0.126 moles) was dissolved in 55 mL water and charged to the catalyst saturator.

To the reactor which has been purged with nitrogen was charged a volume of carbon tetrachloride (400 mL) such that its level just came to the bottom of the sixth stage. 5 NaOH solution (the same solution used in Recycle run #1 , 25 percent by weight, 6.14 moles, 245.6 g dry weight, 983 g solution weight, 774.0 mL) was then charged to the reactor, and filled all six ofthe stirred zones. The stirrer was started and its speed adjusted to 1500 rpm. Fluorene (0.185 moles, 30.75 g) dissolved in carbon tetrachloride (2.24 moles, 344.8 g, 225.35 mL) was fed into the catalyst saturator containing the catalyst solution using a metering pump at a rate of 0 5.5 mL/min. The product solution was collected at the overflow, washed and dried as in Part A. The total product solution collected in this fashion amounts to 1006 g which was evaporated to dryness on the rotary evaporation leaving 42.23 g of very light yellow crystals, 9,9- dichlorofluorene (97.1 percent of the theoretical yield of 43.49 g).

The aqueous solution from the wash column was evaporated on the rotary 5 evaporator (40°C/10 mm (1.32 kPa)) leaving 32.5 g of a clear oil which crystallized (long needles) on cooling (99.3 percent recovery of the tetrabutylammonium hydroxide charged (32.70 g dry weight).

Recycle run #3.

The tetrabutylammonium hydroxide recovered from the above reaction (32.5 g, 0.125 moles) was dissolved in 55 mL water and charged to the catalyst saturator.

To the reactor which had been purged with nitrogen was charged 400 mL carbon tetrachloride such that its level just came to the bottom of the sixth stage. NaOH solution (the same solution used in Recycle run #2, 25 percent by weight, 6.14 moles, 245.6 g dry weight, 983 g solution weight, 774.0 mL) was then charged to the reactor, and filled all six ofthe stirred zones. The stirrerwas started and its speed adjusted to 1500 rpm. Fluorene (0.185 moles, 30.75 g) dissolved in carbon tetrachloride (2.24 moles, 344.8 g, 225.35 mL) was fed into the catalyst saturator containing the catalyst solution using a metering pump at a rate of 5.5 mL/min. The product solution was collected atthe overflow and washed and dried as in Recycle run #2. The total product solution collected in this fashion was 1046 g which was evaporated to dryness on the rotary evaporator leaving 41.11 g of very light yellow crystals, 9,9-dichlorofluorene (94.5 percent of a theoretical yield of 43.49 g). The aqueous solution from the wash column was evaporated on the rotary evaporation (40°C/10 mm of Hg (1.32 kPa)) leaving 30.3 g of a clear amber oil (93.2 percent recovery of the tetrabutylammonium hydroxide charged (32.5 g dry weight)).

These results show that both the phase transfer catalyst and base can be recovered and/or reused (recycled) in the practice ofthe invention.

EXAMPLE 50: Alkylation of Dichlorofluorene onto Xylene

Ferric chloride (0.03 g, 0.2 mmol) was weighed into a 50 mL3-necked flask fitted with a stirbar, nitrogen inlet, thermometer, and condenser with drying tube. A mixture of dichlorofluorene (4.7 g, 20 mmol) in o-xylene (anhydrous, 20 mL) was added by syringe. The addition was exothermic atthe beginning, raising the temperature from 25 to 30°C The reaction was heated by means of a heating mantle to a temperature of 40-50 ^c C while a continuous stream of nitrogen was bubbled through the mixture. The solution rapidly exhibited a dark red color, and began to evolve HCI. The mixture wasanalyzed by the procedure of Example 15 one hour after addition was complete and showed formation of a single product appearing as a single peak at 13.58 minutes. This material was worked up by diluting with methylene chloride, washing the resulting solution with water and 1 M HCI, and then evaporating the solvent. The resulting tacky semisolid was then boiled in ethanol (100 mL)to precipitate a pale yellow solid. This solid was flushed through alumina with methylene chloride (100 mL), and the methylene chloride was quickly evaporated, leaving the product as a pale yellow semicrystalline solid, yield 7.47 g, 99.7 percent of theoretical. Analysis ofthe product by GC/MS showed 98.3 percent ofthe product as a single peak with 1.7 percent of the product as an isomer. The spectral peaks were: GC/MS: 375 (29.76); 374(100.00); 360 (18.1); 359 (37.14); 269 (10.75); H-NMR: delta 2-2.5 (12 H), 6.8-7.9 (14 H).

EXAMPLE 51 : Alkylation Aniline

The procedure of Example 39 was repeated except thatt aniline was used as the aromatic compound to produce 9,9-bis(4-aminophenyl)fluorene (BAPF) in the number of equivalents indicated. By products such as, 9,9-bis(4-aminophenyl)fluorene (abbreviated BAPF) and N,p BAPF rearranged to p,p-BAPF. Formation of BAPF was followed by gas chromatography on a 15 m capillary column. Results are shown in Table 5 indicate isomerization of by products to p,p-BAPF.

Table 5 BAPF PREPARATION

he data in this table shows that over the indicated periods of time, product was formed and byproducts rearranged to the desired p,p-isomers.

The data in this table shows that over the indicated periods of time, product was formed and byproducts rearranged to the desired p,p-isomers.

EXAMPLE 52: Alkylation of Dichlorofluorene onto Benzocyclobutane

The reaction was run in the same manner as in Example 50 except that antimony pentachloride dissolved in dichloromethane was used as the catalyst. The crude product was dissolved in carbon tetrachloride and treated with decolorizing carbon. The solvent was removed by rotary evaporation to produce a light tan foam which was crushed to a powder.

Analysis ofthe powder by LC (liquid chromatography) showed itto be a mixture of 9,9- bis(beπzocyclobutanyI)fluorene (about 58 percent), a 3:2 adduct, 9-benzocyclobutanyl-9-((9- benzocyclobutanylfluoren-9-yl)benzocyclobutanyl)fluorene (about 27 percent) and 15 percent higher oligomers. Proton NMR showed broad overlapping singlets at delta 3-3.2 from TMS (tetramethylsilane) (CH ₂ "s ofthe cyclobutane rings) and a complex pattern of multiplets at delta 7-7.9 (aromatic protons) in a ratio of 1 :2, aliphaticto aromatic. Carbon NMR showed three signals for aliphatic CH ₂ at 29.35, 29.53, and 29.81 ppm (relative to TMS) and three signals ^' for quaternary aliphatic carbons at 66.16, 66.33, and 66.40 ppm. The material was cured at 160°Cfor one hour and 210°Cfor 12 hours. The resulting amber-colored plaque showed a 2 percent loss in weight when heated to 400°C.

EXAMPLE 53: Alkylation of Dichlorofluorene onto p-Cresol

Dichlorofluorene (23.66 g) was weighed into a 500 mL four-necked round- bottomed flask equipped with a magnetic stir bar, heating mantle, thermometer, nitrogen inlet and drying tube. Molten p-cresol (114g, about 40°C) was added rapidly and the mixture immediately turned purple and began to vent HCI. The mixture was stirred for one hour under a sweep of nitrogen. Analysis ofthe mixture by GCshowed formation of a single product. After stirring for an additional hourduring which much ofthe product precipitated, the excess p-cresol was removed by vacuum distillation. The residue was recrystallized from carbon tetrachloride to yield 9,9-bis(2-hydroxy-5-methylphenyl)fluorene in greaterthan 90 percent yield. H-NMR (relative to TMS) delta 1.93 (s, 6H, CH ₃ ), 6.41 (s, 2H), 6.59-6.74 (AB, 4H), 7.17-7.32 (m, 4H), 7.76-7.95 (m, 4H), 8.81 (s, 2H, OH). C-NMR: (ppm) 151.76, 150.56, 137.63, 129.18, 125.59, 125.40, 125.32, 124.54, 124.48, 117.99, 113.83, 60.37 (quaternary C), 18.75 (CH ₃ ). This bisphenol can be cyciized to the spirocyclic ether, 2',7'-dϊmethyl-spiro[9H-fluorene-9,9'- [9Hlxanthene byrefluxing in toluene with catalytic amounts of triflic, toluenesulfonic, or methanesulfonic acid, or by refluxϊng in acetonitrile/water. The spirocyclic ether was easily separated from the starting bisphenol by slurrying mixtures ofthe two in a nonsolventforthe spiro-ethersuch as acetonϊtrϊle followed by filtration. H-NMR delta 2.02 (s, 6H, CH ₃ ), 6.16-6.17 (m, 2H), 6.96-7.0 (m, 2H), 7.0-7.4 (m, 8H), 7.80-7.84 (m, 2H). C-NMR: (ppm) 154.99, 149.37, 139.57, 132.24, 128.81, 128.28, 127.67, 127.60, 125.65, 124.35, 119.8, 116.33, 54.22 (quaternary C), 20.56 (CH ₃ ).

EXAMPLE 54: Alkylation of Dichlorofluorene onto Resorcinol

Dichlorofluorene (23.51 g) was added to a solution of resorcinol (110.11 g) in 500 mL of dry acetonitrile with stirring. The mixture was-held at about 40°C and swept with nitrogen for 4 hours. Analysis ofthe reaction mixture by LCshowed formation ofthe bisphenol 9,9-bis(2,4-dihydroxyphenyl)fluorene as the major product (about 60 percent) with the balance asthespirocyciized bisphenol ether spirotfluorene-g^'-xanthenej-S'jδ'-diol (about 25 percent)

and higher oligomers. The excess resorcinol was removed by vacuum distillation. Recrystallization of the residue from acetonitrile afforded a purified sample of the bisphenol. H-NMR (relative to TMS) (in DMSO): delta 5.88-6.40 (m, 6H), 7.15-7.30 (m, 4H), 7.75-7.78 (m, 4H), 8.84 (s, 2H, OH), 8.93 (s, 2H, OH). C-NMR (ppm): 158.8, 156.77, 153.55, 139.63, 127.37, 126.94, 126.40, 1 19.97, 106.56, 105.36, 103.56, 61.44 (quaternary C). The crude mixture was converted to the spirocyclic bisphenol ether, spiro[fluorene-9,9'-xanthene]-3',6'-diol by treatmentwith acid or refluxing in toluene with acid catalyst as in Example 52. H-NMR (DMSO): delta 6.3-6.9 (m, 5H), 7.3-8.2 (m, 8H), 9.8 (s, 2H).

EXAMPLE 55: Alkylation of Dichlorofluorene onto Hydroquinone

The reaction was carried out using the same procedure as in Example 53, substituting hydroquinone for resorcinol. The spirocyclic bisphenol ether, spiro[fluorene-9,9'- xanthene]-2',7'-diol was isolated as an off-white powder. H-NMR (DMSO): delta 5.9-6.95 (m, 4H), 7.38-8.25 (m, 10H), 9.2 (s, 2H)

EXAMPLE 56: Alkylation of Dichlorofluorene onto a Mixture of Benzocyclobutane and Phenyl Ether.

The reaction was carried out using the same procedure as Example 51 , substituting a solution of dichlorofluorene, benzocyclobutane and phenyl ether (2:6: 1 molar ratio) in dichloromethane for the solution of dichlorofluorene and benzocyclobutane. The reaction mixture was flushed through neutral alumina to remove antimony salts and the product was isolated from the solution by rotary evaporation. The resulting brittle orange foam was crushed to a powder, dissolved in dichloromethane and treated with decolorizing carbon. Evaporation of the solvent yielded a yellow powder. H-NMR:delta 2.95-3.2 (broad overlapping singlets, CH ₂ 's), 6.82-8.0 (m, aromatic H) in a ratio of 1 :5.

EXAMPLE 57: Preparation of Spiro[fluorene-9,9'-xanthene]-2 ^, ,7'-dicarboxylic acid; Oxidation of 2',7'-Dimethylspiro[fluorene-9,9'-xanthene]

Cobalt acetate dihydrate (0.25 g), 2", 7'-dimethylspiro[fluorene-9,9'-xanthene] (1.8 g), o-dichlorobenzene (10 mL), and acetic acid (8 mL) were transferred into a 25 mLthree- necked round-bottomed flask equipped with a magnetic stirbar, condenser, air inlet, and thermometer. Hydrobromic acid (30 percent by weight in acetic acid, 0.2 mL) was added by syringe and the mixture was stirred and heated to 120-125°C as a stream of air was rapidly bubbled through the solution. The progress of the reaction was followed by LC analysis. Heating and aeration of the reaction were continued for 48 hours. The reaction mixture was then cooled and washed with water to remove the acetic acid. The organic layer was extracted with aqueous sodium hydroxide (1 M) to remove the product acid as its sodium salt. Acidification of the base layer with concentrated HCI caused precipitation of the product ,

spϊro[fluorene-9,9'-xanthene]^7'-dicarboxylic acid, as a yellowish powder, which was filtered off, washed with water and dried. H-NMR (DMSO): delta 6.88-6.89 (m, 2H), 7.05-7.11 (m,2H), 7.19-7.24 (m, 2H), 7.35-7.45 (m, 4H), 7.80-7.84 (m, 2H), 7.96-8.03 (m, 2H), 10.55 (bs, 2H). C-13 NMR(DMSO), (ppm): 166.09 (acid C=0), 154.19, 153.35, 139.02, 129.95, 129.22, 128.92, 128.58, 126.51, 125.23, 124.29, 120.81, 117.34, 53.14 (quat. C).

EXAMPLE 58: Preparation of 9,9-Bis(3,4-diaminophenyl)fluorene; Alkylation of 9,9- Dichlorofluorene with o-Phenylene Diamine

Into a three-necked 100 mL round-bottomed flask was placed o-phenylene diamine (26 g). The flask was then equipped with a drying tube, magnetic stirbar, thermometer, powder addition funnel, and nitrogen inlet. The o-phenylene diamine was then stirred and heated to 105°C and solid 9,9-dichlorofluorene was added to the melt. An immediate reaction occurred as evidenced by a color change to dark brown , formation of a copious precipitate and an increase in temperature to 110-120°C. The temperature was then increased to 130°Cand held there for 4 hours during which time the mixture became more liquid. The reaction was then allowed to cool and the excess o-phenylene diamine was removed by pouringthe reaction mixture into ethanol and filtering. The resulting dark brown solid was recrystallized from toluene to yield 9,9-bis(3,4-diaminophenyl)fluorene. H-NMR (DMSO): delta 4.97 (bs,8H, NH ₂ ), 6.07-6.11 (m, 2H), 6.31-6.42 (m, 4H), 7.17-7.33 (m, 6H), 7.76- 7.79 (m, 2H). C-13 NMR(DMSO) (ppm): 152.55, 139.71, 136.29, 134.37, 132.64, 127.56, 127.17, 126.40, 120.37, 118.36, 117.69, 115.63, 115.11 , 64.4 (quat. C).

Example 59: Preparation of 9,9-Bis(4-amino-3-ethyl phenyl)

The procedure of Example 39 was repeated except that 0.537 mole of 2- ethylaniline was used in place ofthe aniline and, after the stirring at 60 °C, the temperature was raised to 175 °C 9,9-Bis(4-amino-3-ethyIphenyl)fluorene (abbreviated BEAPF) , had a molecular of 404 and melting point of 191-192 °C, and was prepared in 92 percent yield. Results are shown in Table 6.

Example 60: Alkylation of 2-Ethylaniline

The procedure of Example 59 was repeated. Afterthe reaction temperature was raised to 175 °C, the formation of 9,9-bis(4-amino-3-ethylphenyl)fluorene isomers (abbreviated BEAPF) was followed by gas chromatography on a 15 m capillary column. The results are shown in Table 6.

Table 6

BEAPF PREPARATION

*Formation of o,p-isomer was not observed

The data in this table shows that over the indicated periods of time, product was formed and byproducts rearranged to the desired p,p-isomers.

Example 61 : Preparation of 9,9-Bis(N-methyl-4-aminophenyl)

The procedure of Example 39 was repeated except that 0.537 mole of N- methylaniline was used in place of the aniline and, afterthe stirring at 60 °C, the temperature was raised to 135 °C. 9,9-Bis(N-methyl-4-aminophenyl)f luorene (abbreviated BNMAPF) , had a molecular of 376 and melting point of 203-204 °C, and was prepared in 85 percent yield. Results are shown in Table 7.

Example 62: Alkylation of N-methylaniline

The procedure of Example 59 was repeated using 0.537 mole of N-methylaniline. Afterthe reaction temperature was raised to 135 °C, the formation of 9,9-bis(N-methyl-4- aminophenyl)fluorene (abbreviated BNMAPF) was followed by gas chromatography on a 15 m capillary column. Thedata in Table 7 shows that over the indicated periods of time, product was formed and byproducts such as o,p-BNMAPF rearranged to the desired p,p-ϊsomer.

Table 7 BNMAPF PREPARATION

The data inthistable shows that over the indicated periods of time, product was formed and byproducts rearranged to the desired p,p-isomers.

Example 63: Preparation of 9,9-Bis(4-amino-3-chlorophenyl)

The procedure of Example 39 was repeated except that 0.537 mole of 2- chloroaniline was used in place ofthe aniline and, after the stirring at 60 °C, the temperature was raised to 175 °C. 9,9-Bis(4-amϊno-3-chlorop ^" henyl)fluorene (abbreviated BACPF) had a molecular weight of 404 and melting point of 235-236 °C, and was prepared in 94 percent yield.Results are shown in Table 8.

Example 64: Alkylation of 2-chloroaniline

The procedure of Example 63 was repeated. Afterthe reaction, temperature was raised to 175 °C, the formation of 9,9-bis(4-amino-3-ethylphenyl)fluorene (abbreviated BACPF) was followed by gas chromatography on a 15 m capillary column. Results are shown in Table 8.

Table 8 BACPF PREPARATION

The data in this table shows that over the indicated periods of time, product was formed and byproducts rearranged to the desired p,p-isomers.