PROCESS FOR THE PRODUCTION OF ACETIC ACID AND/OR ACETIC ANHYDRIDE The present invention relates in general to a carbonylation process and in particular to a carbonylation process for the production of acetic acid and/or acetic anhydride in the presence of a Group VIII metal carbonylation catalyst, a halogen- containing co-catalyst, and a finite concentration of water.

Processes in which carboxylic acids having n+1 carbon atoms are produced by carbonylating an alcohol having n carbon atoms or a reactive derivative thereof in the presence of a Group VIII metal carbonylation catalyst, a halogen-containing co-catalyst and a finite concentration of water have been known for some time. In particular processes in which acetic acid is produced by the carbonylation in the liquid phase of methanol and/or a reactive derivative thereof in the presence of a rhodium-or an iridium-containing carbonylation catalyst, an iodine-containing co-catalyst, a finite concentration of water, and optionally, a promoter are operated on a commercial scale.

Typically Howard et al in Catalysis Today, 18 (1993), 325-354 describe rhodium-and iridium-catalysed carbonylation of methanol to acetic acid. The continuous rhodium-catalysed, homogeneous methanol carbonylation process is said to consist of three basic sections; reaction, purification and off-gas treatment.

The reaction section comprises a stirred tank reactor, operated at elevated temperature and pressure, and a flash vessel. Liquid reaction composition is withdrawn from the reactor and is passed through a flashing valve to the flash tank where the majority of the lighter components of the liquid reaction composition (methyl iodide, methyl acetate and water) together with product acetic acid are vaporised. The vapour fraction is then passed to the purification section whilst the liquid fraction (comprising the rhodium catalyst in acetic acid) is recycled to the reactor (as shown in Figure 2 of Howard et al). The purification section is said to

comprise a first distillation column (the light ends column), a second distillation column (the drying column) and a third distillation column (the heavy ends column) (as in Figure 3 of Howard et al). In the light ends column methyl iodide and methyl acetate are removed overhead along with some water and acetic acid. The vapour is condensed and allowed to separate into two phases in a decanter, both phases being returned to the reactor. Wet acetic acid is removed from the light ends column as a side draw and is fed to the drying column where water is removed overhead and an essentially dry acetic acid stream is removed from the base of the distillation zone. From Figure 3 of Howard et al it can be seen that the overhead water stream from the drying column is recycled to the reaction section.

Heavy liquid by-products are removed from the base of the heavy ends column with product acetic acid being taken as a side stream.

Carbonylation processes for the production of a carboxylic acid in the presence of rhodium catalysts are described, for example, in US-A-3769329; GB- A-1468940; GB-A-1538783 and EP-A-0087070.

Carbonylation processes for the production of a carboxylic acid in the presence of iridium catalysts are described, for example, in GB-A-1234121; US-A- 3772380; DE-A-1767150; EP-A-0616997; EP-A-0618184; EP-A-0618183; and EP-A-0657386.

The production of acetic anhydride by carbonylation is known from, for example, GB-A-1468940 which discloses the two-stage production of an anhydride of a monocarboxylic acid by reacting a carboxylate ester satisfying the formula RCOOR or an ether satisfying the formula ROR with an acyl halide satisfying the formula RCOX, formed in situ or in a separate stage, under substantially anhydrous conditions, wherein X is iodide or bromide, the Rs may be the same or different and each R is a monovalent hydrocarbyl radical or a substituted monovalent hydrocarbon radical wherein the or each substituent is inert. The acyl halide may be produced by carbonylation of a halide satisfying the formula RX at super atmospheric pressure, R being as hereinbefore defined, and the carbonylation may be effected in the presence as catalyst of a Group VIII noble metal, and optionally a promoter. According to GB-A-1468940 it is important that the carbonylation reaction should be carried out under substantially anhydrous conditions. There is no specific mention in GB-A-1468940 of the production of acetic anhydride by the carbonylation of any ester other than methyl acetate.

It is also known that acetic anhydride can be produced with or without the

net co-production of acetic acid. Thus, EP-A-87870 for example, discloses a process for the production of acetic anhydride with or without the net co- production of acetic acid from methanol and carbon monoxide in a series of esterification, carbonylation and separation steps comprising:- (1) reacting methanol with recycle acetic acid in an esterification step to form an esterification product containing predominantly methyl acetate, water and optionally unreacted methanol, (2) removing part of the water from the esterification product, (3) reacting the esterification product still containing water with carbon monoxide in a carbonylation step in the presence as catalyst of free or combined metallic carbonylation catalyst and as promoter of free or combined halogen to form a carbonylation product containing acetic acid and acetic anhydride, (4) separating the carbonylation product by fractional distillation into a low boiling fraction containing carbonylation feed and volatile carbonylation promoter components, acetic acid and acetic anhydride fractions, and a higher boiling fraction containing carbonylation catalyst components, (5) recycling the low boiling fraction containing carbonylation feed and carbonylation promoter components and the higher boiling fraction containing carbonylation catalyst components to the carbonylation step, and (6) recycling at least part of the acetic acid fraction to the esterification step.

As the carbonylatable reactant in carbonylation processes for the production of acetic acid there is used methanol and/or a reactive derivative thereof which is, for example, either methyl acetate, dimethyl ether, or methyl iodide.

Since methyl acetate is formed in situ by the reaction of methanol with acetic acid present in the liquid reaction composition either by way of solvent for the carbonylation or by way of the product of carbonylation, methyl acetate is a carbonylatable reactant of choice for the production of acetic acid. As the carbonylation reactant in the production of acetic anhydride by carbonylation there is generally used methyl acetate.

We have now found that the use of methyl esters having a boiling point higher than methyl acetate as carbonylatable reactant can lead to advantage in the production of acetic acid and/or acetic anhydride.

Accordingly the present invention provides a process for the production of acetic acid and/or acetic anhydride which process comprises carbonylating with carbon monoxide in a liquid reaction composition in a carbonylation reactor at least

one methyl ester of an aliphatic carboxylic acid having a boiling point equal to or greater than the temperature of the carbonylation reaction in the presence of a Group VIII metal carbonylation catalyst, a hydrocarbyl halide co-catalyst and in the presence or absence of water.

In the absence of water in the liquid reaction composition in the carbonylation reactor the carbonylation product comprises acetic anhydride. In the presence of at least a finite concentration of water, that is a water concentration in the liquid reaction composition of at least 0.1 % by weight, the carbonlylation product comprises acetic acid. Water may be introduced to the carbonylation reactor together with or separately from other components of the reaction composition. Water may be separated from other components of the reaction composition withdrawn from the reactor and may be recycled in controlled amounts to maintain the concentration of water in the liquid reaction composition.

Suitably, the water concentration in the liquid reaction composition for the production of acetic acid may be in the range from 0.1 to 15% by weight, preferably below 11% by weight, more preferably below 7% by weight.

Esters suitable in the process have a boiling point equal to or greater than the carbonylation reaction temperature. Typically the reaction temperature is at least 100°C.

Suitable methyl esters having a boiling point higher than methyl acetate may be selected from:- (A) esters having the formula: R [C02CH3] n (I) wherein R is selected from (i) C2 to C30 aliphatic hydrocarbyl groups and substituted derivatives thereof. R may also comprise an aromatic functionality provided the aromatic group and the ester functionality is separated by at least one aliphatic carbon moiety, for example methyl phenylacetate.

(B) esters having the formula: Ru, X (O) n (OCH3) p (II) wherein R is an organic group, X is either S, Se, P or As, m is 0,1 or2, n is either 1 or 2; and

p is 1,2 or 3 depending upon the valency of X, i. e. when m = 0, n = 2, p = 2, X = S or Se when m = 2, n = 1, p = 1, X = P or As whenm=l, n=l, p=2, X=PorAs, etc.

(C) polymeric materials having a pendent methyl ester group; and (D) methyl esters of inorganic oxo-acids.

As regards (A), in the formula (I), the C2 to C3o hydrocarbyl group (i) may suitably be an alkyl group or an alkenyl group. Suitable alkyl groups may be linear or cyclic, branched or unbranched groups. Suitably the alkyl or alkenyl group may contain from 5 to 20 carbon atoms. Examples of suitable olefinically unsaturated esters include maleic acid esters, acrylic acid esters, and itaconic acid esters.

Preferably n in the formula (I) is greater than 1, for example 2. Suitable esters of formula (I) wherein n is greater than 1 are for example dimethyl adipate, dimethyl succinate and dimethyl glutarate. An advantage of employing, for example, dimethyl succinate is that because of its higher density and hence higher moles of methyl per unit volume than methyl acetate it is possible to use less of it, thereby allowing more space in the reactor for acetic acid or acetic anhydride. Additionally an advantage of the esters is their reduced vapour pressure allowing a higher partial pressure of CO in the reactor. Suitable aromatic groups include C6-aryl groups which may be substituted with up to five substituents. Suitable substituents of the group R in the formula (I) include alkyl, aryl, alkoxy, aroxy or carboxylate, and hydroxyl groups. The preferred ester is where R is C3 and n is 2, namely dimethyl glutarate. In particular the use of dimethyl glutarate provides the advantage that it does not precipate out of solution.

As regards (B), in the formula (II), R may be independently the same as in the formula (I), including groups of the formula OR'wherein R'is a group as defined aforesaid.

As regards (C), examples of suitable polymeric materials include polymeric esters having a repeat unit which is: either [CH (C02CH3) CH2] m (III) or [CH (CO2CH3) CH (CH3)] m (IV) wherein in the formulae (III) and (IV) m is in the range 50 to 500.

As regards (D), examples of suitable methyl esters of inorganic oxo-acids include trimethyl phosphate, trimethyl borate, dimethyl carbonate, and dimethyl sulphate.

Examples of suitable methyl esters having a boiling point equal to or greater than the carbonylation reaction temperature include methyl hexanoate, methyl octanoate, methyl decanoate, methyl-3-hydroxybenzoate, dimethyl adipate, dimethyl succinate and dimethyl glutarate.

The methyl ester may comprise substantially the whole of the carbonylatable reactant or it may be replaced in part by, for example, one or more of methanol, methyl acetate, and methyl iodide.

Suitably, the methyl ester is present at a concentration of from 0.1 to 40 weight percent, preferably from 10 to 30 weight percent.

As mentioned hereinabove the replacement of methyl acetate by a methyl ester having a boiling point equal to or greater than the reaction temperature in the liquid carbonylation reactor composition can offer several advantages. Firstly, reducing the amount of volatile components in product distillation, for example with reference to Howard et al in the flashing section of the reactor section and the first distillation column of the purification section, will simplify the plant and its operation and thereby reduce costs. Secondly, by decreasing the partial pressure of ester reactant in the reaction mixture a higher partial pressure of carbon monoxide can be utilised, thereby giving higher catalyst activity at a given temperature, or a lower total reaction pressure may be employed if desired.

The carbonylation is accomplished in the presence of a Group VIII metal carbonylation catalyst. Suitable metals of Group VIII include for example nickel, cobalt, and the noble metals. Preferably the carbonylation catalyst is a noble metal of Group VIII of the Periodic Table of the Elements. As the Group VIII noble metal there may be used iridium, osmium, platinum, palladium, rhodium or ruthenium, preferably rhodium or iridium. Rhodium is the catalyst of choice for the production of acetic anhydride under anhydrous conditions, whereas both rhodium and iridium are preferred for the production of acetic acid in the presence of water.

The catalyst may comprise any metal compound which is soluble in the liquid reaction composition. The catalyst may be added to the liquid reaction composition in any suitable form which dissolves in the liquid reaction composition or is convertible to a soluble form.

Examples of suitable rhodium-containing compounds which may be added

to the liquid reaction composition include [Rh (CO) 2Cl] 2, [Rh (CO) 2I] 2, [Rh (Cod) CI] 2, Rh (III) chloride, Rh (III) chloride trihydrate, Rh (III) bromide, Rh (III) iodide, Rh (III) acetate, Rh dicarbonlylacetylacetonate, RhCl3 (PPh3) 3 and RhCl (CO) (PPh3) 2. Suitably the rhodium catalyst concentration in the liquid reaction composition is in the range from 50 to 5000 ppm by weight of rhodium, preferably from 100 to 1500 ppm.

Examples of suitable iridium-containing compounds which may be added to the liquid reaction composition include IrCl3, IrI3, IrBr3, [Ir (CO) 2I] 2, [Ir (CO) 2CI] 2, [Ir (CO) 2Br] 2, [IrCCC]-H\[IrCCCBr-lT,[WOW,[Ir(CH3)l3 (CO) 2] H+, Ir4 (CO) l2, IrCI3.3H20, IrBr3.3H20, Ir4 (CO) l2, iridium metal, Ir203, Ir02, Ir (acac) (CO) 2, Ir (acac) 3, iridium acetate, [Ir30 (OAc) 6 (H20) 3] [OAc] and hexachloroiridic acid [H2IrCI6], preferably, chloride-free complexes of iridium such as acetates, oxalates and acetoacetates which are soluble in one or more of the carbonylation reaction components such as water, alcohol and/or carboxylic acid.

Particularly preferred is green iridium acetate which may be used in an acetic acid or aqueous acetic acid solution. Suitably the iridium catalyst is present in the liquid reaction composition at a concentration in the range from 400 to 5000 ppm by weight of iridium, preferably from 700 to 3000ppm.

Alternatively, the catalyst may be attached to a suitable carrier, for example by complexing with a suitable ligand.

It is preferred to use one or more promoters for the carbonylation catalysts in the process of the invention. The choice of promoter will depend to some extent on the particular noble metal used as catalyst.

When a rhodium carbonylation catalyst is used the promoter is suitably selected from the group consisting of iodide salts of alkali and alkaline earth metals of which lithium iodide is preferred, quaternary ammonium iodides, quaternary phosphonium iodides, and phosphine oxides. The promoter may be present up to its limit of solubility. Alternatively, ruthenium and/or osmium may be used as promoters as described, for example, in EP-A-0728727.

When an iridium carbonylation catalyst is used the promoter is suitably a metal selected from ruthenium, osmium, tungsten, rhenium, zinc, cadmium, indium, aluminium, gallium and mercury, and is preferably selected from ruthenium and osmium. Ruthenium is the more preferred promoter for iridium catalysts.

Preferably, the promoter is present in an effective amount up to the limit of its solubility in the liquid reaction composition. The promoter is suitably present in

the liquid reaction composition at a molar ratio of promoter: iridium of [0.5 to 15]: 1. A suitable promoter concentration is 400 to 5000 ppm by weight of promoter metal.

Suitable hydrocarbyl halide co-catalysts include alkyl halides. Preferably the alkyl halide is an iodide. Most preferably, the hydrocarbyl halide is methyl iodide. Suitably the concentration of methyl iodide in the liquid reaction composition is in the range from 1 to 30% by weight, for example from 5 to 20% by weight.

The carbon monoxide reactant may be essentially pure or may contain inert impurities such as carbon dioxide, methane, nitrogen, noble gases, water and Cl to C4 paraffinic hydrocarbons. The presence of hydrogen in the carbon monoxide feed and generated in situ by the water gas shift reaction is preferably kept low in the production of acetic acid as its presence may result in the formation of hydrogenation products. Thus, for acetic acid production the amount of hydrogen in the carbon monoxide reactant is preferably less than 1 mol %, more preferably less than 0.5 mol % and yet more preferably less than 0.3 mol % and/or the partial pressure of hydrogen in the carbonylation reactor is preferably less than 1 bar partial pressure, more preferably less than 0.5 bar and yet more preferably less than 0.3 bar. The presence of hydrogen may, on the other hand, be beneficial in the production of acetic anhydride. The partial pressure of carbon monoxide in the reactor is in the range greater than 0 to 40 bar, typically from 4 to 30 bar.

The total pressure of the carbonylation reaction is suitably in the range 10 to 200 barg, preferably 15 to 100 barg, more preferably 15 to 50 barg. The temperature of the carbonylation reaction is suitably in the range 100 to 300°C, preferably in the range 150 to 220°C.

The process of the present invention may be performed as a batch or a continuous process, preferably as a continuous process.

Acetic acid product may be recovered for example from the liquid reaction composition by withdrawing vapour and/or liquid from the carbonylation reactor and recovering acetic acid from the withdrawn material. Preferably, acetic acid is recovered from the liquid reaction composition by continuously withdrawing liquid reaction composition from the carbonylation reactor and recovering acetic acid from the withdrawn liquid reaction composition by one or more flash and/or fractional distillation stages in which the acetic acid, is separated from the other components of the liquid reaction composition such as rhodium or iridium catalyst,

methyl iodide co-catalyst, promoter, unreacted higher boiling methyl ester, water and acetic acid solvent which may be recycled to the reactor to maintain their concentrations in the liquid reaction composition. To maintain stability of the catalyst during the acetic acid product recovery stage, water in process streams containing carbonylation catalyst for recycle to the carbonylation reactor should be maintained at a concentration of at least 0.5% by weight.

Acetic anhydride may be recovered in similar manner except for the non- participation of water in the recovery process.

The process of the present invention will now be illustrated by reference to the following Examples.

General reaction method A 300cm3 zirconium autoclave, equipped with a stirrer and a liquid injection facility, was used for a series of batch autoclave experiments. The autoclave was pressure tested to 40 barg with nitrogen and then flushed three times with carbon monoxide up to 10 barg. An initial charge consisting of an ester (approx. 50.0g), acetic acid (approx. 58.0g), methyl iodide (approx. 9.8g) and water (approx. 10.0g), was placed into the autoclave, which was then re-purged with carbon monoxide to 4 barg and vented slowly so as not to lose any volatiles.

Then carbon monoxide (approx. 4-5 barg.) was placed in the autoclave which was then heated with stirring (1500 rpm) to 190°C. A catalyst solution was primed into the liquid injection line with (approx. 1.35g) of H2IrCI6 solution (22.26% Ir w/w), water (approx. 5.0g) and acetic acid (approx. 5.0g) and injected to the hot autoclave to bring the autoclave pressure to 22 barg. The reaction rate was monitored by drop in carbon monoxide pressure from a ballast, typically charged to 70 barg. The autoclave pressure and temperature were maintained at a constant 22 barg and 190°C throughout the reaction by pressure and coolant control valves.

The reaction was terminated when the drop in ballast pressure became less than 0.1 barg per 5 minutes. After cooling and carefully venting the autoclave the liquid components were discharged and analysed for liquid products and by-products.

Comparative Experiment A A baseline experiment was performed with the autoclave charged with methyl acetate (48.1g, 649.3mmol), acetic acid (68.99g, 1148.8mmol), water (9.94g, 0.55mmol), methyl iodide (8.84g, 62.3mmol). The iridium catalyst solution (1.457g, 0.79mmol) with acetic acid (6.5g, 108mmol) and water (6.5g, 360mmol).

The rate of reaction, based on carbon monoxide uptake started at a rate of 17.3 moWhr and steadily declined until virtually all the methyl acetate was consumed.

This is not an example according to the present invention as methyl acetate was used as the methanol source and not an alternative higher boiling ester as described.

Comparative Experiment B Experiment A was repeated except that the charge consisted of methyl-3- hydroxybenzoate (49.68g, 327mmol), acetic acid (58.71g, 977mmol), water (10.51g, 584mmol) and methyl iodide (9.91g, 70mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.35g), water (5.0g, 278mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate, based on carbon monoxide uptake was 4.4mol/Vhr. Acetic acid was the major liquid product detected (>99%). This demonstrates an example of an aromatic, substituted methyl ester as a substitute for methyl acetate.

Example 1 Experiment A was repeated except that the charge consisted of methyl hexanoate (49.91g, 383mmol), acetic acid (57.9g, 964mmol), water (10.27g, 570mmol) and methyl iodide (9.87g, 69mmol). The iridium catalyst solution consisted of H2IrCI6 solution (1.34g), water (5.0g, 278mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate, based on carbon monoxide uptake was 8.56mol/Vhr. Acetic acid was the major liquid product detected (>99%). This demonstates an example of a linear C6 methyl ester.

Example 2 Experiment A was repeated except that the charge consisted of methyl octanoate (49.06g, 310mmol), acetic acid (58. Ig, 967mmol), water (10.65g, 591mmol) and methyl iodide (9.89g, 69.6mmol). The iridium catalyst solution consisted of H2IrCI6 solution (1.368g), water (5.0g, 278mmol) and acetic acid (6.1g, 101mmol). The initial reaction rate, based on carbon monoxide uptake was 8.9moWhr. Acetic acid was the major liquid product detected (>99%). This demonstrates an example of a linear C8 methyl ester as a substitute for methyl acetate.

Example 3 Experiment A was repeated except that the charge consisted of methyl decanoate (49.76g, 267mmol) acetic acid (68.9g, 1147mmol), water (10. lg,

561mmol) and methyl iodide (9.86g, 68.9mmol). The iridium catalyst solution consisted of H2IrC16 solution (1.42g), water (6.0g, 333mmol) and acetic acid (5.0g, 83mmol). The initial reaction rate, based on carbon monoxide uptake was 8.24moVVhr. Acetic acid was the major liquid product detected (>99%). This demonstrates an example of a linear Clo methyl ester as a substitute for methyl acetate.

Example 4 Experiment A was repeated except that the charge consisted of dimethyl adipate (50.01g, 287mmol), acetic acid (58.1g, 967mmol), water (10. lg, 558mmol) and methyl iodide (9.81g, 69mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.40g), water (4.89g, 272mmol) and acetic acid (5.95g, 99mmol). The initial reaction rate, based on carbon monoxide uptake was 13.9moVUhr. Acetic acid was the major liquid product detected (>99%). This demonstrates an example of a linear C6 dimethyl ester as a substitute for methyl acetate.

Example 5 Experiment A was repeated except that the charge consisted of dimethyl succinate (49.81g, 341mmol), acetic acid (64.09g, 1067mmol), water (5.06g, 281mmol) and methyl iodide (9.78g, 69mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.374g), water (5.0g, 278mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate, based on carbon monoxide uptake was 14.8moVVhr. Acetic acid was the major liquid product detected (>99%). This demonstrates an example of a linear C4 dimethyl ester as a substitute for methyl acetate.

Example 6 Experiment A was repeated except the charge consisted of dimethyl glutarate (51.97g, 324mmol), acetic acid (76.2g, 1268mmol), water (10.27 g, 559mmol) and methyl iodide (8.91g, 63mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.374g), water (5.97g, 332mmol) and acetic acid (6.5g, 108mmol). The initial reaction rate based on carbon monoxide uptake was 16.4 mol/Uhr. Acetic acid was the major product detected (>99%). Unlike other high boiling diesters no precipitation of the resultant diacid occurred on cooling.

This demonstrates an example of a linear Cs dimethyl ester as a substitute for methyl acetate.

Example 7 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (64.0g, 398.6mmol), acetic acid (64.1 lg, 1068mmol), water (10.7g, 594mmol) and methyl iodide (8.91g, 63mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.3708g), water (6.0g, 333mmol) and acetic acid (6.0g, 100mmol). The reaction rate at 30% wt (610mmol) ester, based on carbon monoxide uptake, was 17.7 mol/Uh. Acetic acid was the major liquid product detected (>99%). This demonstrates that dimethyl glutarate can be used as a 40% wt charge replacement for methyl acetate in a carbonylation reaction without precipitation of the resultant glutaric acid at the end of the reaction.

Example 8 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (47.65g, 297.2mmol), acetic acid (79.97g, 1333mmol), water (9.99g, 555mmol) methyl iodide (8.93g, 63mmol) and Ru (CO) 4I2 (1.4608g, 3.1mmol). The iridium catalyst solution consisted ofH2IrCI6 solution (1.3708g), water (6.05g, 336mmol) and acetic acid (5.99g, 99.8mmol). The initial reaction rate based on carbon monoxide uptake, was 20.2 moUl/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that ruthenium carbonyl iodide species promotes the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate. This also demonstrates that addition of ruthenium promoter to a dimethyl glutarate carbonylation further reduces the liquid and gas by-products.

Example 9 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (47.53g, 297mmol), acetic acid (85.51g, 1425mmol), water (4.73g, 263mmol) methyl iodide (8.92g, 62.8mmol) and Ru (CO) 4I2 (1.4588g, 3.1mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1.3571g), water (6.04g, 336mmol) and acetic acid (5.99g, 99.8mmol). The initial reaction rate based on carbon monoxide uptake, was 36.45 mol/l/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that ruthenium carbonyl iodide species at reduced water further promotes the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate. This also demonstrates reduced gaseous by- products compared to experiment A, Table 1.

Example 10 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (25.95g, 162mmol), methyl acetate (24g, 324mmol), acetic acid (74.0g,

1233mmol), water (10. Og, 555mmol) methyl iodide (8.9g, 62.6mmol) The iridium catalyst solution consisted of H2IrCl6 solution (1.30g), water (6.0g, 333mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 14.4 moVVh. Acetic acid was the major liquid product detected (>99%).

This demonstrates that dimethyl glutarate can be used as a co-feed with methyl acetate for iridium catalysed carbonylation with no significantly adverse effect on the rate of reaction compared to example7. This also demonstrates that a 50% replacement of MeOAc with dimethyl glutarate can reduce the liquid and gaseous by- products compared to experiment A.

Example 11 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (5.19g, 32.4mmol), methyl acetate (43.2g, 584mmol), acetic acid (72.0g, 1200mmol), water (10.02g, 557mmol) methyl iodide (8.9g, 62.6mmol) The iridium catalyst solution consisted of H2lrCI6 solution (1.37g), water (6.04g, 335mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 15.2 mol/Uh. Acetic acid was the major liquid product detected (>99%).

This demonstrates that dimethyl glutarate can be used as a low concentration co- feed with methyl acetate for iridium catalysed carbonylation with no significantly adverse effect on the rate of reaction. This also demonstrates that a 10% replacement of MeOAc with dimethyl glutarate can reduce the liquid and gaseous by- products compared to experiment A.

Example 12 Experiment A was repeated except that the charge consisted of glutaric acid (5.19g, 32.4mmol), methanol (19.25g, 601mmol), acetic acid (47.73g, 795mmol), methyl iodide (18.8g, 132mmol) The iridium catalyst solution consisted of H2IrCl6 solution (1.379g), water (4.55g, 253mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 14.5 moVVh. Acetic acid was the major liquid product detected (>99%). This demonstrates that a charge of glutaric acid and methanol gives approximately the same rate of reaction as dimethyl glutarate, experiment 7. This shows that any conjugate acid of any claimed methyl ester can be used in the carbonylation process to form the ester, and associated equilibrium derivatives, insitu by combination with methanol. This further demonstrates that the reduction in liquid and gaseous by-products is also approximately the same as with pre-formed dimethyl glutarate, experiment 7, Tablel.

Example 13 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (52.02g, 325.1mmol), acetic acid (75.7g, 1262mmol), water (10. Olg, 556mmol) methyl iodide (8.92g, 62.8mmol) and Ru (CO) 4I2 (4.41g, 9.4mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1. 3688g), water (6.03g, 335mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 33.3 mol/l/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that six equivalents of ruthenium carbonyl iodide species to iridium catalyst, promotes the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate and significantly reduces the liquid by-products.

Example 14 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (52.02g, 325.1mmol), acetic acid (76.0g, 1267mmol), water (1.25g, 69mmol) methyl iodide (10. Og, 70.4mmol) and Al (OAc) 2 (OH) (1.26g, 7.8mmol).

The iridium catalyst solution consisted of H2IrCl6 solution (1. 362g), water (6.0g, mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 3.6 mol/l/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that aluminium hydroxide di-acetate can be used as an additive for the iridium catalysed carbonylation of dimethyl glutarate.

Example 15 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (52.0g, 325mmol), acetic acid (76.0g, 1267mmol), water (1.25g, 69mmol) methyl iodide (8.9g, 62.6mmol) and GaI3 (3.55g, 7.9mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1. 358g), water (6.01g, 334mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 24.6 mol/l/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that, Gallium iodide promotes the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate and significantly reduces the liquid by-products.

Example 16 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (52.0g, 325mmol), acetic acid (76.0g, 1267mmol), water (10. Olg, 556mmol) methyl iodide (8.9g, 62.7mmol) and InI3 (4.1g, 8.3mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1. 36g), water (6.0g, 333mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 24.4 moUl/h. Acetic acid was the major liquid product detected (>99%). This

demonstrates that indium promotes the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate.

Example 17 Experiment 7 was repeated except that the charge consisted of dimethyl glutarate (52.02g, 325.1mmol), acetic acid (75.7g, 1262mmol), water (10. Olg, 556mmol) methyl iodide (8.92g, 62.8mmol) and Os3 (CO) 12 (4.41g, 9.4mmol). The iridium catalyst solution consisted of H2IrCl6 solution (1. 36g), water (6.0g, 335mmol) and acetic acid (6.0g, 100mmol). The initial reaction rate based on carbon monoxide uptake, was 28.75 mol/l/h. Acetic acid was the major liquid product detected (>99%). This demonstrates that osmium carbonyl species promote the rate of reaction of iridium catalysed carbonylation of dimethyl glutarate.

Tablel. ____ Liquid and gas By-product analysis Experiment Total propionic hydrogen Carbon dioxide Methane acid (ppm) (mmol) (mmol) (mmol) A 504 10. 6 8. 0 12.9 6 180 5. 3 4. 2 6.1 7 351 7. 3 8. 0 11.3 8 120 2. 3 5. 4 4.8 9 477 2. 4 4. 2 4.5 10 330 4. 9 4. 2 5.7 11 149 5. 6 3. 9 6.0 12 184 5. 2 5. 9 7.2 13 43 2. 2 3. 9 7.4 14 <30 0. 3 1. 5 2.6 15 954 0. 2 5. 3 5.6 16 379 10. 9 7. 8 9.3 17 624 0.2 5.0 9.0

Comparative Experiment C A baseline experiment was performed using rhodium catalyst instead of an iridium based catalyst as per examples 1-17. The experiment was performed in an autoclave charged with methyl acetate (47.77g, 645 mmol), acetic acid (57. 5g, 958mmol), water (14.24g, 791 mmol) and methyl iodide (25.52g, 180 mmol). The

rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.1525g, 0.39 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 10.14 moWh and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is not an example according to the present invention as methyl acetate was used as the methanol source and no high boiling methyl ester in any proportion was used.

Example 18 Experiment C was repeated except that the charge consisted of dimethyl glutarate (50. 5g, 315 mmol), acetic acid (63.2g, 1053mmol), water (14.25g, 792 mmol) and methyl iodide (25.76g, 181 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0. lSg, 0.385 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 10.0 mol/l/h and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention of diethyl glutarate being used as the methanol source compared to experiment B. The rate and the liquid and gaseous by-products are comparable to experiment B, table 2.

Example 19 Experiment 18 was repeated except that the charge consisted of dimethyl glutarate (53.52g, 334 mmol), acetic acid (48.5g, 808mmol), water (15.21g, 845 mmol), methyl iodide (27.0g, 190 mmol) and lithium iodide (8.01g, 60 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.1514g, 0.389 mmol) with acetic acid (15.01g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 10.75 mol/l/h and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention of diethyl glutarate being used as the methanol source compared to experiment B. This demonstrates that the liquid and gaseous by-products are reduced when LiI and dimethyl glutarate are used in combination.

Example 20 Experiment 18 was repeated except that the charge consisted of dimethyl glutarate (36.03g, 225 mmol), acetic acid (83.0g, 1383 mmol), water (11.12g, 618

mmol), methyl iodide (18.54g, 130 mmol) and lithium iodide (16. Og, 120 mmol).

The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.1521g, 0.39 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 6.4 moVUh and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention of dimethyl glutarate being used as the methanol source compared to experiment B. This demonstrates that the liquid and gaseous by-products are reduced when LiI 10% wt and low MeI with dimethyl glutarate are used in combination.

Example 21 Experiment 18 was repeated except the charge consisted of dimethyl glutarate (33.4g, 209 mmol), acetic acid (83.34g, 1389 mmol), water (10.29g, 572 mmol), methyl iodide (25.07g, 176 mmol) and triphenylphosphine oxide (16.03g, 57.7 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.1513g, 0.39 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 8.9 mol/Uh and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention of dimethyl glutarate being used as the methanol source compared to experiment B. This demonstrates that the liquid and gaseous by-products are reduced when triphenylphosphinel0% wt with dimethyl glutarate are used in combination.

Comparative Experiment D A baseline experiment was performed using rhodium catalyst promoted with five equivalents of a ruthenium carbonyl complex. The experiment was performed in an autoclave charged with methyl acetate (32.5g, 645 mmol), acetic acid (74.5g, 1242mmol), water (14.0g, 777 mmol), methyl iodide (25.04g, 176 mmol) and [Ru (CO) 4I2] (1.82g, 3.9 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.1524g, 0.39 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 13.8 mol/l/h and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is not an example according to the present invention as methyl acetate was used as the methanol source and no high boiling methyl ester in any proportion

was used.

Example 22 Experiment D was repeated except the charge consisted of dimethyl glutarate (34.0g, 212 mmol), acetic acid (79.26g, 1321mmol), water (14.0g, 777 mmol), methyl iodide (26.82g, 189 mmol) and [Ru (CO) 4I2] (1.79g, 3.8 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.15 lg, 0.388 mmol) with acetic acid (15g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 18.5 mol/l/h and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention and demonstrates the use of dimethyl glutarate with a rhodium/ruthenium catalyst system which gives increased rates and reduced liquid by-products compare to experiment C, table 2.

Example 23 Experiment 22 was repeated except the charge consisted of dimethyl glutarate (34.0g, 212 mmol), acetic acid (78.02g, 1300mmol), water (14. Olg, 777 mmol), methyl iodide (27.01g, 190 mmol) and [Ru (CO) 4lu (3.60g, 7.6 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0. 15g, 0.386 mmol) with acetic acid (15.01g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 21.5 mol//h and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention and demonstrates the use of dimethyl glutarate with a rhodium/ruthenium catalyst system, at a ratio of 1: 10, which gives increased rates and reduced liquid by-products.

Example 24 Experiment 22 was repeated except the charge consisted of dimethyl glutarate (68.22g, 426 mmol), acetic acid (53.02g, 884mmol), water (7.3g, 405 mmol), methyl iodide (26.184g, 189 mmol), [Ru (CO) 4I2] (0.72g, 1.54 mmol) and lithium iodide (0.1 lg, 0.82 mmol). The rhodium catalyst solution comprised of [RhCl (CO) 2] 2 (0.151g, 0.388 mmol) with acetic acid (14.99g, 250 mmol).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 12.2 moWh and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is an example according to the present invention and demonstrates the

use of dimethyl glutarate with a rhodium/ruthenium/lithium catalyst system, at a ratio of 1: 2: 1, which gives reduced liquid by-products compare.

Experiment E A baseline experiment was performed using rhodium catalyst supported on a quarternised Purolite 4-VP resin. The experiment was performed in an autoclave charged with methyl acetate (25.045g, 338 mmol), acetic acid (31.45g, 524mmol), water (8.506g, 472 mmol), methyl iodide (10.355g, 73mmol). The rhodium catalyst comprised of [RhCl (CO) 2] 2 (0.064g, 0.16 mmol) supported on 13.906g of methyl iodide quarternised Purolite 4-VP resin (dry weight before quarternisation: 7.758g).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 12.58 mol/l/h (at 30 wt% MeOAc) and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected (>99%).

This is not an example according to the present invention as methyl acetate was used as the methanol source and no high boiling methyl ester in any proportion was used.

Example 25 Experiment E was repeated except the charge consisted of dimethyl glutarate (27.035g, 169 mmol), acetic acid (31.335g, 522mmol), water (8.578g, 476 mmol), methyl iodide (10.16g, 71.5mmol). The rhodium catalyst comprised of [RhCl (CO) 2] 2 (0.064g, 0.16 mmol) supported on 17.989g of methyl iodide quarternised Purolite 4- VP resin (dry weight before quarternisation: 7.801g).

The rate of the reaction, based on carbon monoxide uptake started at the rate of 7.75 mol//h (at 30 wt% MeOAc) and slowly declined until virtually all the methyl acetate had been consumed. Acetic acid was the major liquid product detected ( >99%).

This is an example according to the present invention and demonstrates that high boiling esters such as dimethyl glutarate can be carbonylated readily by resin supported catalysts Table2 Liquid and gas By-product analysis Experiment Total propionic hydrogen Carbon dioxide Methane acid (ppm) (mmol) (mmol) (mmol) C 68 2. 85 1. 5 0.1 18 <30 3. 5 1. 4 0.15 19 145 1. 7 0. 53 0.16 20 82 1. 6 0. 49 <0.1 21 32 1. 5 0. 5 0.2 D 44 2. 6 1. 1 0.22 22 <30 2. 8 1. 0 0.4 23<302. 60. 80.5 24 <30 <0. 1 0. 43 0.4 E 156 4. 51 2. 63 1.0 25 <30 1. 98 0.99 0.52

Note: Total propionic acid analyses of less than 30ppm are below the detection limit for the combined propionic acid precursors e. g. ethyl iodide, ethyl acetate, acetaldehyde, diethylglutarate and mono-ethylglutarate.