Prior Art Methyl-or ethyl-substituted phenolic compounds that can be selectively oxidized according to the present invention are known to the present art and may be represented by the general formulae: wherein: X is CH3 or CH2CH3, Each of R, to R4 is H, CI, Br. C,-C, o alkyl, or C1-Cno alkoxy, R, and R2, R2 and R or R3 and R4 together may be- (CHn--. wherein n is 3 to 6, provided only one of R, and R2, R2 and R ?, or R3 and R4 is- (CH2) n- Specific examples of phenolic compounds include p-cresol, o-cresol, p- ethylphenol, o-ethylphenol, 2,6-dichloro-p-cresol, 2,6-dichloro-p-ethylphenol, 2- bromo-p-cresol, 2,4-xylenol, 3,4-xylenol, 2,6-di-tert-butyl-p-cresol, 2,6-tert- amyl-p-cresol, and 2-methoxy-4-methyl phenol.

The hydroxy-substituted aromatic afdehydes or ketones thus produced are known compounds, which are either used as intermediates in the preparation of pharmaceuticals, agricultural chemicals, polymeric resins and other industrial chemicals, or are themselves used for a variety of purposes, for example, as artificial flavors.

The direct formation of hydroxybenzaldehydes by the oxidation of the corresponding alkylphenol is known in the art. Processes for the said transformations may be categorized into three groups. These are: (i) The oxidation of the acetate derivative of the phenol in carboxylic acid media, as for example described in JP63154644 (1986) and 62242644A (1986).

(ii) The oxidation of the phenolate ion in alkaline alcoholic solutions, as for example described in NL9200968-A (1992), JP2172940 (1988), 2172941 (1988), and 2172942 (1988), US 4429163 (1984), 4453016 (1982), 4471140 (1984), 4748278 (1988), 4929766 (1989), and 5130493 (1991), and EP 323290 (1987).

(iii) The oxidation of the alkylphenol in carboxylic acid media, with or without the addition of co-solvents and/or supplementary reagents, over noble-metal catalysts, as for example described in JP7879832 (1976), and US 4915875 (1990).

The difficulties in oxidising methyl or ethyl groups on aromatic rings that also contain a free hydroxy group selectively, are well known in the art. Oxidation of the unprotected alkylphenol in carboxylic acid media is problematic in view of the formation of many unwanted by-products. In particular, phenolic systems substituted in either the para-or ortho-positions, or both, are known to undergo coupling, polymerization and ring opening in addition to oxidation. E Fache et a/ ("Oxidation of alkylphenols to hydroxybenzaldehydes", in The roots of organic development, J-R Desmurs, and S Ratton (Eds.), Elsevier, Amsterdam, pp. 380-390 (1996)) teaches that by-product formation during these reactions may be reduced by keeping the concentration of substrate in reaction mixtures low. Experiments by the inventors of the present application confirmed these teachings and have shown that acceptable selectivities to hydroxybenzaldehydes can only be achieved by keeping substrate loading below 5% (mass/vol.).

Oxidations in alkaline media suffer two major shortcomings, namely the production of large amounts of inorganic salts as by-products, and inherently unsafe operating conditions. In US 4748278 (1988) it is shown that salt formation may be minimized by isolating the aldehyde product as the salt of the base used during oxidation procedures, since in certain cases such a salt is required for consequent reactions. However, due to the wide flammable limits and low flash point of methanol, oxidations carried out in this manner are inherently unsafe and virtually impossible to operate on a large scale.

SUMMARY OF THE INVENTION According to the present invention, a method of oxidising an alkyl phenol of the formula (I) or (II) to a corresponding hydroxybenzaldehyde or hydroxy- acetophenone, includes the steps of producing a reaction medium of the alkyl phenol in a solvent which comprises a dihydric alcohol and is acidic or alkaline and reacting the alkyl phenol with oxygen in the presence of an oxidation catalyst.

The solvent is either acidic or alkaline. When alkaline, a base will generally be present, and when acidic, an acid such as an organic acid will generally be present. An example of a suitable base is a hydroxide, and an example of a suitable acid is acetic acid.

The solvent may include a co-solvent. When a co-solvent is present, it is preferably water, particularly when the solvent is alkaline.

The desired reaction product of the oxidation is a hydroxybenzaldehyde or hydroxyacetophenone. Partially oxidised products may also be produced during the oxidation reaction. These partially oxidised products can be subjected to further oxidation, alone or with addition alkylphenol, to produce further hydroxybenzaldehyde or hydroxyacetophenone. The process may be a continuous one in that the desired reaction product is removed from the reaction medium, further alkylphenol added to the reaction medium, with the addition of further solvent or co-solvent if needed, while the oxidation reaction occurs continuously.

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION The inventors of the present application have studied the formation of by- products during the catalytic air oxidation of alkylphenols in the presence and absence of base, especially at high substrate loadings. In almost all cases, the formation of by-products can be traced to the high reactivity of the primary oxidation product, a cyclohexadienone intermediate, (for example 4- methylene-2, 5-cyclohexadienone or 2-methylene-3, 5-cyclohexadienone during the oxidation of p-and o-cresol, respectively) particularly towards nucleophilic addition on the methylene group. In all of the cases where alkylphenols have been oxidised successfully to either the corresponding hydroxybenzaldehyde or hydroxyacetophenone, the primary oxidation product is stabilized, either by forming a benzylic acetate (4-, or 2-hydroxybenzyl acetate for p-and o-cresol, respectively) when oxidations are carried out in acetic acid/acetate ion mixtures, or the mono-methoxy ether (4-, or 2-hydroxybenzyl methyl ether for p-and o-cresol, respectively) when oxidations are carried out in alkaline methanolic solutions. In the case of the benzyfic acetate derivative, the inventors have found that the benzylic acetate is quite resistant towards further oxidation, and unless an equilibrium with the corresponding benzylic alcohol is established, low reaction rates are achieved. In addition, the benzylic acetate, and/or benzylic alcohol with which it may exist in equilibrium, can readily act as nucleophile towards the cyclohexadienone intermediate to give coupling products, particularly on the benzylic carbon, even in the absence of catalysis.

The inventors have also found that coupled products may undergo further reactions such as migration and elimination under the influence of catalysts to give a variety of by-products. In alkaline methanolic solutions, nucleophilic attack by the benzyl methyl ether on the cyclohexadienone intermediate is less favorable and higher selectivities to desired products can be achieved.

However, methanoi is a somewhat weaker nucleophile than the acetate ion, with the result that the alkylphenol, or the alkylphenolate ion, succeed in competing with methanol for the cyclohexadienone intermediate to form phenoxy ethers as by-products.

The inventors have found that the successful and selective oxidation of alkylphenols to the corresponding hydroxybenzaldehydes, or hydroxy- acetophenones, is dependent upon the successful stabilization of the cyclohexadienone intermediate. Such stabilization prevents side reactions of the type discussed above, but does not interfere with the oxidation of the stabilized intermediate to desired products. More particularly, the inventors have found that this is achieved with use of dihydric alcools as a solvent. The process of the invention offers numerus advantages over processes currently known in the art. It is, for example, an advantage of the present invention that alkylphenols can be oxidized with high selectivity to desired products in the presence or absence of base.

It is a further advantage of the present invention that, when the oxidation is carried out in the presence of base, the base loading can be reduced up to six times compared to other processes known in the art.

It is an advantage of the process of the present invention that oxidations can be carried under much safer conditions compared to similar oxidations in methanolic solutions presently known in the art.

The present invention, therefore, provides an improved process for the production of hydroxybenzaldehydes and hydroxyacetophenones by means of the catalysed oxidation of the respective alkylphenol, or mixture of alkylphenols, in the liquid-phase and using molecular oxygen as the primary oxidant. The process of the invention will now be described in detail. o-and p-Methyl-or ethyl-substituted phenolic compounds can be selectively oxidized according to the present invention. The amount of alkylphenol used in the process of the invention is generally in the range 0,05 to 0,60 times by mass the mass of reaction solvent, and preferably in the range 0,1-0,5 times by mass the mass of reaction solvent.

Products that may be isolated from the oxidation reaction medium are known to the present art and may be represented by the general formulae : wherein: X is CHO or COCH3,.

Each of R, to R4 is H, CI, Br, C,-C, o alkyl, or C,-C, o alkoxy, R, and R2, R2 and R3 or R3 and R4 together may be-(CH2) n-, wherein n is 3 to 6, provided only one of R, and R2, R2 and R3 or R3 and R4 is- (CH2) n-.

For oxidations carried out in the presence of a base, the reaction product may be in the form of a salt, for example the sodium, potassium or ammonium salt.

The oxidation reaction is carried out in the liquid-phase in the presence of a solvent which is or contains a dihydric alcohol and which is acidic or alkaline.

According to the invention, dihydric alcools of choice are ethylene glycol, 1,2- propanediol, 1,3-propanediol, 1,4-butanediol, and the like. It is possible, and in many cases advantageous, to use a co-solvent with the alcohol to promote oxygen solubility, to improve mass transfer, to reduce the viscosity of the reaction mixture, and to aid product isolation. The co-solvent may be selected from a wide range of potential solvents based on solubility in the dihydric alcool, stability towards oxidation under the reaction conditions employed, and the absence of interference with any step during the oxidation or product isolation procedures. Examples of suitable co-solvents include water, which is preferred, Cl-ces aliphatic carboxylic acids, Cl-C5 aliphatic alcohols, and the Ci-Cg aliphatic esters of C2-Cs aliphatic carboxylic acids. Specific examples of suitable co-solvents include water, acetic acid, propionic acid, ethyl acetate, propyl acetate, glycol diacetate, methanol, ethanol, propanol and butanol. No specific limitation is placed on the number of co-solvents used in combination with the dihydric alcool, nor on the amount of co-solvent used.

Generally, however, a co-solvent will be present in an amount of 1 to 60% (v/v) relative to the dihydric alcohol.

The primary oxidant for the process of the invention is molecular oxygen, and may be supplie to the oxidation reactor in the gaseous form as pure oxygen, air, or oxygen diluted with a suitable inert gas. In all cases, the reactor off-gas may be further diluted with inert gas. A preferred inert gas is nitrogen. The amount of oxygen in the reaction zone, or passed through the reaction zone, should at least be equal to the theoretical amount required to oxidize the alkyl constituents to the desired extent, and will generally be in excess of that theoretical amount.

Any type of base having a greater basicity than the alkylphenol of the types described above may be used in the process of this invention to provide addition protection of the phenolic group. The actual use of a base is not a requirement of the method of the present invention, and depends, for example, on the type of catalyst used, the actual solvent composition, and whether the neutral compound, or its salt, is required as the desired product. Examples of bases are metal hydroxides such as sodium hydroxide, potassium hydroxide, lithium hydroxide, magnesium hydroxide, calcium hydroxide and aluminium hydroxide, metal alkoxides such as the alkoxides of metals mentioned above, and where the alkoxide is, for example, methoxide, ethoxide or isopropoxide, metal amides such as the amides of the metals mentioned above, and where the amide may, for example, be ethylamide, diethylamide or diisopropylamide, and quaternary ammonium hydroxides such as ammonium hydroxide, tetramethylammonium hydroxide, tetraethylammonium hydroxide, and tetrabutylammonium hydroxide. Preferred bases are the metal hydroxides, alkoxides and quaternary ammonium hydroxides.

The amount of base, when used, will generally be present in an amount such that there is a minimum ratio of alkylphenol: base of 1: 0,1. An alkylphenol: base ratio in the range of 1: 0,1 to 1: 5 is preferable. The lower range 1: 0,1 to 1: 2 is preferred so as to minimise base catalyse reaction of the dihydric alcohol. All these ratios are on a mass basis.

A wide range of oxidation catalysts, or catalyst mixtures, can be employed for the oxidation of alkylphenols in the presence of a dihydric alcool, or a mixture of dihydric alcohol and a co-solvent. Many catalysts are capable of the desired transformation, and are known in the art. Such catalysts may contain one or more of the following active metals: cobalt, copper, iron, manganese, ruthenium, rhodium, iridium, palladium, platinum and cerium. Catalysts may be added to the reaction medium in the form of soluble metal compounds such as metal halides, for example metal chlorides and metal bromides, and the metal salts of inorganic and organic acids, for example metal nitrates, metal sulfates, metal acetates and metal propionates. Catalysts may also be added to the reaction medium as insoluble compounds, for example as metal hydroxides, metal oxides, or as metal compounds suspended on a suitable support.

Suitable supports should have a large surface area, preferably surface areas in the range 10-1500 m'/g or more. Examples of suitable supports include activated charcoal, silica, zeolites, alumina and clays. In addition to the"active metals", catalysts on insoluble supports may also contain one or more addition metals which function as promoters and/or stabilizers for the active metals. The use of promoter metals are known in the art, and include metal ions such as potassium, calcium, cesium, barium, tin, antimony, lead and bismuth.

The amount of active metal used is not limited in any specific way, provided that the active metal is present in sufficient quantities to catalyze the desired reaction. When soluble or insoluble metal compounds are used, the metal compounds are added in their pure forms to the reaction mixtures. It is preferred that the active metal be present in the reaction mixture in an amount not less than 0.0001 moles per mole of substrate, but more specifically in the range 0.001-0.10 moles active mezcal per mole substrate. For catalysts supported on solid supports, the metal loading on the support should preferably be in the range 0.5-20 % by weight of the supported catalyst, and more preferably, in the range 1-10 % by weight of the supported catalyst.

The preferred range of active metal loading of supported catalyst in reaction mixtures are the same as specified for soluble or insoluble metal compound catalysts.

The process of the invention may be carried out at elevated temperature.

Preferred temperatures are from 20°C to approximately 300°C, and more specifically in the range 30-250°C. Reactions may also be performed under applied pressure, particularly at higher reaction temperatures, in order to increase the solubility of oxidant in the reaction mixture, to increase the rate of reaction and to keep the reaction mixture in the liquid-phase. No particular restriction is placed on the pressure of the oxygen containing gas used for oxidation reactions, but a pressure between 1 and 100 bar (1 x 105 Pa and 1 x 107 Pa) is preferred.

To ensure adequate reaction rates, it is preferred that reaction medium be subjected to mechanical mixing such as, for example, mechanical stirring of the reaction medium, using a blowing technique or using a mixing nozzle. The degree of mixing should be such that further increases in the degree of mixing have no influence on the rate of oxidation achieved.

Due to the relative instability of the desired reaction products, and due to the high boiling point of the preferred reaction solvent/s, the preferred method of product isolation is crystallization. The catalyst component may be removed from the reaction medium before the desired product is isolated. Catalyst may be removed by conventional methods such as filtration from hot reaction mixtures. It may be desirable to dilute the reaction mixture with water prior to filtration and/or crystallization. When the reaction mixture is diluted with water prior to catalyst filtration, it is desirable to heat the reaction mixture to the oxidation temperature prior to filtration.

In contrast to other procedures for product isolation known in the art, it is not necessary in the process of the present invention to remove reaction solvent prior to the crystallization process. Upon dilution, if required, the reaction mixture may simply be cooled, optionally accompanied with seeding, to effect crystallization. Dilution with water is preferred. The amount of water added to the reaction medium depends upon the initial water content of the reaction solvent. A total water content of between 5-200% (vol./vol.) with respect to the alcoholic solvent is preferred.

After crystallisation, the desired product may be separated by means of filtration, and if required, washed before drying. If the product isolated is in the form of a salt, for example the sodium or potassium salt of the hydroxybenzaldehyde or hydroxyacetophenone, the salt may be used without further treatment, or the hydroxybenzaldehyde/hydroxyacetophenone may be liberated by treatment with acid.

The mother liquor from the catalyst/product separation stage may be recycle back to the oxidation reactor without further treatment other than the removal of water to the desired levels, if required.

The process of this invention is now described in further detail below with illustrative examples which are not intended to limit the scope of the invention in any way whatsoever.

Oxidation reactions were carried out in a jacketed slurry reactor, which consisted of an 8-cm-i. d., fully baffled, mechanically agitated glass reactor equipped with a reflux condenser. A 4.5-cm-diameter four-bladed stainless steel impeller was used for agitation. The reaction temperature was kept constant by circulating heating fluid, maintained at a constant temperature in a constant temperature bath, through the external heating/cooling jacket of the reactor. Oxygen was supplie to the reaction mixture by means of an open- ended glass tube positioned just above the impellers.

In the examples which follow,"substrate"refers to the starting phenolic compound.

Examples 1 (a) and 1 (b) To the oxidation reactor was supplied (a) methanol (100 cm3), p-cresol (21.6g), NaOH (21.0g) and oxygen (40 cm3 min') whilst stirring at a rate of 1000 rpm at a temperature of 60°C, and (b) ethylene glycol (100 cm3), p-cresol (21.6g), NaOH (21.0g) and oxygen (40 cm3 min') whilst stirring at a rate of 1000 rpm at a temperature of 100°C. The catalyst used in both cases was Cocu2.6H2O (0.42g). The respective reactions were allowed to proceed for (a) 8 hours to give p-cresol 5.0%), 4-hydroxybenzaldehyde (19.3g; 79.0%), and 4- hydroxybenzyl methyl ether 3.0%), and (b) 18 hours to give p-cresol 1.34%), 4-hydroxybenzaldehyde (11.47g, 46.9%) and 0- (4- hydroxybenzyl) ethylene glycol (15.5g; 46.0%).

Examples 2 (a), 2 (b), 2 (c), 2 (d), 2 (e) and 2 (f) The experiment described in Example 1 (b) was repeated, but using (a) none (0), (b) half (1/2), (c) one (1), (d) two (2), (e) three (3) and (f) five (5) equivalents of NaOH per mole of substrate. HPLC analysis of the six reaction mixtures showed (a) no conversion of substrate, (b) 94.0% conversion of substrate, and 6.4% 4-hydroxybenzaldehyde, 6.0% 0- (4-hydroxybenzyl) ethylene glycol, and (c) 100.0% conversion of substrate, and 25.9% 4-hydroxybenzaldehyde, 2.2% 0- (4-hydroxybenzyl) ethylene glycol, and (d) 100.0% conversion of substrate, and 80.6%-4-hydroxybenzaldehyde, 3.6% 0- (4-hydroxybenzyl)-ethylene glycol, and (e) 97.54% conversion of substrate, and 38.9% 4- hydroxybenzaldehyde, 52.9% 0- (4-hydroxybenzyl) ethylene glycol, and (f) 51.9% conversion of substrate, and 4.8% 4-hydroxybenzaldehyde and 44.2% 0- (4-hydroxybenzyl) ethylene glycol.

Examples 3 (a) and 3 (b) The experiment described in Example 1 (b) was repeated, but using (a) one (1) and (b) three (3) equivalents of KOH per mole of substrate. HPLC analysis of the two reaction mixtures showed (a) 94.5% converston of substrate, and 7.4% 4-hydroxybenzaldehyde, 54.2% 0- (4-hydroxybenzyi) ethylene glycol, and (b) 100.0% conversion of substrate, and 71.2% 4-hydroxybenzaldehyde, 2.75% 0- (4-hydroxybenzyl) ethylene glycol.

Example 4 The experiment described in Example 1 (b) was repeated, but using propylene glycol instead of ethylene glycol as reaction solvent. HPLC analysis of the reaction mixture showed a 95.2% conversion of substrate, with 23.0% 4- hydroxybenzaldehyde and 45.9% 4-hydroxybenzyl 2-hydroxypropyl ether.

Examples 5 (a) and 5 (b) The experiment described in Example 1 (b) was repeated, but using (a) ten (10) and (b) thirty (30) percent (by mass) water in the ethylene glycol. HPLC analysis of the two reaction mixtures showed (a) 97.5% conversion with 39.9% 4-hydroxybenzaidehyde, and 54.2% 0-(4-hydroxybenzyl) ethylene glycol, and (b) 80.5% conversion with 49.1% 4-hydroxybenzaldehyde and 43.5% 0- (4- hydroxybenzyl) ethylene glycol.

Examples 6 (a), 6 (b), 6 (c), 6 (d), 6 (e) and 6 (f) The experiment described in Example 1 (b) was repeated, but using (a) 0.42g CoCi2'6H20, (b) 1.34 cobalt ion-exchanged Montmorillonite K10, (c) 0.25g CoO8, (d) 1.04g Co-oxide supported on silica, (e) 0.25g CoO and (f) 0.25g Co304. The reactions were allowed to proceed for six hours. HPLC analysis of the six reaction mixtures showed (a) 60.8% conversion of the substrate, and 27.9% 4-hydroxybenzaldehyde, and 23.2% 0- (4-hydroxybenzyi) ethylene glycol, and (b) 36.3% conversion of the substrate, and 7.1 % 4- hydroxybenzaldehyde, and 18.3% 0- (4-hydroxybenzyl) ethylene glycol, and (c) 44.1% conversion of the substrate, and 10.5% 4-hydroxybenzaidehyde, and 21.2% 0- (4-hydroxybenzyl)-ethylene glycol, and (d) 49.6% conversion of the substrate, and 3.7% 4-hydroxybenzaldehyde, and 43.7% 0- (4- hydroxybenzyl) ethylene glycol, and (e) 66.7% conversion of the substrate, and 9.3% 4-hydroxybenzaldehyde, and 55.2% 0- (4-hydroxybenzyl) ethylene glycol, and (f) 47.7% conversion of the substrate, and 15.5% 4-hydroxybenzaldehyde and 21.4% 0- (4-hydroxybenzyl) ethylene glycol.

Examples 7 (a) and 7 (b) The experiment described in Example 1 (b) was repeated, but using (a) 0.5g 5% palladium on activated carbon and (b) 0.5g lead promoted 5% palladium on activated carbon. HPLC analysis of the two reaction mixtures showed (a) 55.3% conversion of the substrate, and 38.9% 4-hydroxybenzaldehyde, and 15.6% 0- (4-hydroxybenzyl) ethylene glycol, and (b) 35.8% conversion of the substrate, and 21.5% 4-hydroxybenzaldehyde and 13.2% 0- (4- hydroxybenzyl) ethylene glycol.

Experiment 8 (a), 8 (b), 8 (c) and 8 (d) The experiment described in Example 1 (b) was repeated, but using (a) 0.25g, (b) 0.50g, (c) 0.75g and (d) 1.00g Co (II) oxide supported on silica. The reactions were allowed to proceed for six hours. HPLC analysis of the four reaction mixtures showed (a) 49.6% conversion of the substrate, and 3.7% 4- hydroxybenzaldehyde, and 43.7% 0- (4-hydroxybenzyl) ethylene glycol, and (b) 59.8% conversion of the substrate, and 7.6% 4-hydroxybenzaldehyde, and 49.6% 0- (4-hydroxybenzyl) ethylene glycol, and (c) 71.7% conversion of the substrate, and 10.6% 4-hydroxybenzaldehyde, and 58.8% 0- (4- hydroxybenzyl)-ethylene glycol, and (d) 70.24% conversion of the substrate, and 9.8% 4-hydroxybenzaidehyde and 58.5% 0-(4-hydroxybenzyi) ethylenë glycol.

Example 9 To a Buchi pressure glass reactor of 1000cm3 capacity, equipped with a heating/cooling jacket, mechanical overhead stirrer, baffles, thermocouple, gas inlet and pressure release valve, was added ethylene glycol (500cm3), cobalt (II) chloride hexahydrate (2.1g), p-cresol (109.1g), NaOH (40.0g) and the vessel pressurized to 5 bar with dioxygen. Stirring was commenced at a rate of 1200rpm and the temperature set at 80°C. The reaction was allowed to proceed for 4 hours to give a 93.4% conversion of substrate, and 45.1% 4- hydroxybenzaldehyde and 41.7% 0- (4-hydroxybenzyl) ethylene glycol.

Examples 10 (a), 10 (b), 10 (c) and 10 (d) The experiment described in Example 1 (b) was repeated, but using (a) 21.6g o-cresol, (b) 21.6g m-cresol, (c) 24.4g 4-ethylphenol and (d) 21.6g of a 52: 48 commercial mixture of p-cresol and m-cresol. HPLC analysis of the four reaction mixtures showed (a) 28.5% conversion of the substrate, and 11.3% 2- hydroxybenzaldehyde, and 13.6% 0- (2-hydroxybenzyl) ethylene glycol (b) no conversion of the substrate, (c) 15.9% conversion of the substrate, and 10.7% 4-hydroxyacetophenone, and 3.6% 0- (4-hydroxybenzyl)-a-methyl ethylene glycol, and (d) p-cresol (trace amounts), m-cresol (100%), 4- hydroxybenzaldehyde (94.4%), and 4-hydroxybenzyl glycol ether (4.2%) (percentages are selectivities based on m-and p-cresol, respectively).

Examples 11 (a) and 11 (b) The experiment described in example 5 was repeated with the following changes: the water was replace with acetic acid, no base was used in the reaction and the catalyst used was 5% Pd on activated carbon. HPLC analysis of the two reaction mixtures showed (a) 89.1% conversion of substrate, and 42.7% 4-hydroxybenzaldehyde and 50.5% 0- (4-hydroxybenzyl) ethylene glycol when using a 10: 90 mixture of acetic acid and ethylene glycol, and (b) 73.8% conversion of substrate, and 71.4% 4-hydroxybenzaldehyde and 17.8% 0- (4- hydroxybenzyl) ethylene glycol when using a 30: 70 mixture of acetic acid and ethylene glycol.

Examples 12 (a) and 12 (b) To the oxidation reactor was supplie 15% water in ethylene glycol (100 cm3), p-cresol (20.0g), NaOH (24.4g), Co304 (0.5g) and oxygen (40 cm3 min'') whilst stirring at a rate of 1000 rpm at a temperature of 110°C. The reactions were allowed to proceed for (a) 14 hours and (b) 16 hours, respectively. HPLC analysis of the two reaction mixtures showed 100.0% conversion of the substrate, and (a) 95.56% 4-hydroxybenzaldehyde, and 2.0% 0- (4- hydroxybenzyl) ethylene glycol, and (b) 98.1 % 4-hydroxybenzaldehyde, and 0.9% 0- (4-hydroxybenzyl) ethylene glycol.

Example 13 The experiment described in Example 12 (b) was repeated, but using CoO (0.5g) as catalyst. HPLC analysis of the reaction mixture showed 99.8% conversion of the substrate, 91.6% 4-hydroxybenzaldehyde and 5.6% 0- (4- hydroxybenzyl) ethylene glycol.

Example 14 The experiment described in Example 12 (b) was repeated, but at 100°C. HPLC analysis of the reaction mixture showed 96.4% conversion of the substrate, 71.4% 4-hydroxybenzaldehyde and 19.8% 0- (4-hydroxybenzyl) ethylene glycol.

Example 15 (a) and 15 (b) The experiment described in Example 12 (b) was repeated, but by using (a) 30% and (b) 40% water in ethylene glycol (100cm3). HPLC analysis of the reaction mixtures showed (a) 99.6% conversion of-the substrate, 95.9% 4- hydroxybenzaldehyde and 2.9% 0- (4-hydroxybenzyl) ethylene glycol, and (b) 97.5% conversion of the substrate, 89.3% 4-hydroxybenzaldehyde and 5.5% 0- (4-hydroxybenzyl) ethylene glycol.

Example 16 The experiment described in Example 12 (b) was repeated, but using 25. Og p- cresol and 30.5g NaOH. HPLC analysis of the reaction mixture showed 92.8% conversion of the substrate, 50.2% 4-hydroxybenzaldehyde and 44.3% 0- (4- hydroxybenzyl) ethylene glycol.

Examples 17 (a), 17 (b) and 17 (c) The experiment described in Example 12 (b) was repeated. HPLC analysis of the reaction mixture showed 98.7% theoretical yield. The reaction mixture was divided into three parts and to each part enough deionized water was added to ensure a (a) 3: 20, (b) 1: 2 and (c) 1: 1 water to ethylene glycol ratio. The catalyst for each part was removed by filtration, and the mixture cooled to-17'C, seeded and the sodium salt of 4-hydroxybenzaldehyde allowed to crystallize while unreacted substrate and intermediates remain in solution. The solidified product was filtered and vacuum-dried at room temperature to give (a) 0.0% isolated yield, (b) 39.7% isolated yield and (c) 98.1% isolated yield at >99% purity (purities determined by HPLC analysis).

Examples 18 (a), 18 (b) and 18 (c) The experiments described in Example 17 were repeated, but to each part enough deionized water was added to ensure a (a) 1: 1, (b) 3: 2 and (c) 2: 1 water to ethylene glycol ratio. HPLC analysis of the reaction mixture showed 96.7% theoretical yield. Isolation of the sodium salt of 4-hydroxybenzaidehyde gave (a) 91.64% isolated yield at >99% purity, (b) 92.84% isolated yield at 97.9% purity and (c) 85.1% isolated yield at 97.6% purity (purities determined by HPLC analysis).

Example 19 The experiment described in Example 1 (b) was repeated five times, but by using 15% water ethylene glycol and 1.09 Co304 as catalyst for all of the reactions. The catalyst was recovered and charged to the next reaction without regeneration. Reactions were allowed to proceed for 11.5 hours.

Results from HPLC analysis of the reaction mixtures are indicated in Table 1.

Table 1 4-HBA = 4-hydroxybenzaldehyde; 4-lziBEG = 0- (4-hydroxybenzyl) ethylene glycol.

Example 20 The experiment described in Example 1 (b) was repeated, but by using 15% water ethylene glycol and 1. 0g C0304 catalyst. The reaction was allowed to proceed for 11.5 hours. HPLC analysis indicated 57.1% conversion of the substrate, and 76.1% 4-hydroxybenzaldehyde and 19.7% 0- (4- hydroxybenzyl) ethylene glycol. The sodium salt of 4-hydroxybenzaldehyde was isolated according to the experiment described in Example 17 (c) and in 94.8% isolated yield and 98.7% purity. After removal of the water from the resulting mixture, 15% deionized water was added and the mixture used in a consecutive identical reaction with the addition of p-cresol (21. 6g), NaOH (6.2g) and C0304 (1.0g). HPLC analysis indicated 97.3% conversion of the substrate, and 82.9% 4-hydroxybenzaldehyde and 19.2% 0- (4- hydroxybenzyl) ethylene glycol.