NL8301891A - METHOD FOR PREPARING ACETIC ACID. - Google Patents

Description

Method for the preparation of acetic acid.

This invention relates to a process for the preparation of acetic acid by carbon addition of methanol.

Acetic acid has been known as an industrial chemical for many years and large amounts are used in the preparation of various products. Proposals for preparing carboxylic acid by the action of carbon monoxide on alcohols (carbonylation) have been described, inter alia, by Reppe et al. In U.S. Pat. No. 2,729,651 and by Holmes in U.S. Pat. Nos. 4,133,963 and 4,218,340. Very high pressures were needed in 10 previous carbonylation proposals. Carbonylations operating at lower pressures have also been suggested. French patent 1,573,130 discloses, for example, the carbonylation of methanol and mixtures of methanol with methyl acetate in the presence of compounds of Group VIII noble metals, such as platinum, iridium, palladium, osmium and ruthenium, and in the presence of bromine or iodine under lower pressures than those required by Reppe and Holmes. According to U.S. Pat. Nos. 3,769,329 and 3,772,380, acetic acid is prepared from the same starting materials in the presence of iridium or rhodium and bromine or iodine. U.S. Pat. Nos. 3,689,533 and 3,717,670 describe a vapor phase process for the preparation of acetic acid using various supported rhodium-containing catalysts. However, all these lower pressure processes require precious noble metals.

More recently, Belgian Pat. No. 860,557 describes the preparation of carboxylic acids by carbonylation of alcohols using nickel as the catalyst, which is strengthened by a compound of trivalent phosphorus and an iodide. Here one has a low-pressure carbonylation without precious metal. This method has an effect, but there is room for better yields of the desired acid.

An improved process is also described in US patent application 219,786 of December 24, 1980. This discloses the preparation of acetic acid by carbonylation of methanol in the presence of molybdenum nickel or tungsten nickel as catalyst in the presence of iodide and in the presence of an organophosphorus or organo nitrogen compound as a promoter.

Now the surprising finding was made that the carbonylation can proceed significantly faster if there is a mixture of carbon monoxide and hydrogen in which the ratio between the partial stresses of hydrogen and carbon monoxide is between 0.05: 1 and 0.4: 1, better between 0.15: 1 and 0.30: 1 and best kept between 0.2: 1 and 0.25: 1.

Thus, according to the invention, methanol is carbonylated in the presence of molybdenum-nickel or tungsten-nickel as a catalyst, in the presence of an iodide and of an organophosphorus or organo-nitrogen compound as a promoter, with such amounts of carbon monoxide and hydrogen as mentioned above. give ratio between the partial stresses between the indicated limits.

The reaction is carried out under elevated pressure with generally partial CO stresses between 1 and 140 bar, preferably between 1 and 70 bar (0.1 to 14 MPa and 7 MPa, respectively). 25 But C0 voltages up to 350 bar and even up to 700 bar (35 resp.

70 MPa) can also be used. The total pressure is, of course, that of the carbon monoxide, along with the pressure required to keep the other substances in liquid phase. As is known with carbonylations, the reaction rate increases with the partial CO voltage, but surprisingly it has now been found that at a certain partial CO voltage the reaction is surprisingly accelerated further if hydrogen is also present, and in such an amount that the ratio between the partial voltages of hydrogen and carbon monoxide is within the above-mentioned limits. With less hydrogen, one has no significant effect and at higher ratios, one does not get more effect.

A wide temperature range, from 25 ° to 350 ° C, can be used, but preferably the temperature is between 100 ° and 250 ° C and most preferably between 125 ° and 5 225 ° C. Lower temperatures can be used but lead to lower reaction rates and higher temperatures can be used but do not give any particular advantage. The reaction time is not a parameter of this method and depends mainly on the temperature used, but representative residence times are from 0.1 to 20 hours.

The final reaction mixture will normally contain volatile compounds such as hydrocarbon iodide, unreacted alcohol, the acid to be prepared and optionally the corresponding esters and / or ethers; after separating the acid, all these volatile components can be recycled to the reaction. After the desired residence time has elapsed, the reaction mixture is separated into its constituents, usually by distillation. Preferably, the reaction mixture is distilled, for example in a fractional distillation column, into a series of columns which works well enough to separate the volatiles from the acid to be prepared and also to prepare the acid from the less volatile catalyst and the promoter components. The boiling points of the volatiles are far enough apart that normal distillation does not present a particular problem. Likewise, the higher boiling organic components can be easily distilled from the metal catalyst and an optional organic promoter in the form of a relatively low volatile complex. The catalyst thus recovered and the promoter, including the iodide, can then be combined with new amounts of alcohol and carbon monoxide and reacted to a new amount of carbon.

Although not necessary, the reaction can proceed in the presence of an organic solvent or diluent. Since methanol has a relatively low boiling point, the presence of a higher boiling solution or dilution makes 8301891 4.

agent, preferably acetic acid or the corresponding ester such as methyl acetate, it is possible to set more moderate total pressures. Alternatively, the solvent or diluent can be any organic solvent that is inert under the reaction conditions, such as hydrocarbons (eg, octane, benzene, toluene, xylene, and tetralien), halohydrocarbons (eg, trichlorobenzene), and carboxylic acids or esters thereof (eg Methoxyethyl acetate) etc. Mixtures of solvents can also be used, such as mixtures of methyl acetate and acetic acid. If a carboxylic acid is used, it is preferably acetic acid, because this is already found in the reaction system. If one uses a solvent or diluent that. it is not already automatically present in the reaction mixture, preferably one having a boiling point sufficiently different from the boiling points of the components of the reaction mixture is taken so that it is easy to separate, as will be understood by a person skilled in the art.

The reaction is best carried out in the presence of a small amount of water, such as between 2 and 8%, preferably between 4 and 6%, based on the total weight of the reaction mixture. As described in co-pending patent application no., The co-use of water has a beneficial effect on the system.

The carbon monoxide is suitable as a substantially pure substance such as is commercially available, but inert diluents such as carbon dioxide, nitrogen, methane and noble gases may be present if desired. The presence of such inert diluents does not affect the carbonylation, but their presence naturally increases the total pressure if one wants to achieve the partial CO stress.

The catalyst components can be used in any suitable form, so both as an element and with higher valence. For example, nickel and molybdenum or tungsten can be used as finely divided metals, but also as organic or inorganic compounds. Representative compounds are the carbonates, oxides, hydroxides, bromides, iodides, chlorides, oxyhalides, hydrides, lower alkaxides (methoxides) and phenoxides and the carboxylates derived from a 1 to 2 carbon atoms such as the acetates, butyrates, decanoates , laurates, benzoates, etc. Likewise, complexes of the catalytic metals can be used, eg.

the carbonyls and the metal alkyls as well as the chelates, association compounds and enol salts. Examples of other complexes include bis (triphenylphosphine) nickel dicarbonyl, tricyclopentadienyl trinickel dicarbonyl, tetrakis (triphenylphosphite) nickel and the corresponding complexes of the other components such as molybdenum hexacarbonyl and tungsten hexacarbonyl. Among the catalyst components mentioned are also complexes of these metals with organic promoters as ligands derived from the organic promoters mentioned below.

Particular preference is given to the elemental form, the halides (especially the iodides) and the organic salts, and especially those of the monocarboxylic acid which is to be prepared. It will be appreciated that the above listing of possible compounds and complexes is for illustrative purposes only and should not be taken as limiting.

The listed catalyst precursors may contain the impurities normally associated with commercial qualities and do not require further purification.

The organophosphorus promoter is preferably a phosphine of the formula PR2 wherein the three groups R may be the same or different and are alkyl, cycloalkyl, aryl, amide or halogen (with 3 to 20 carbon atoms in the alkyl and cycloalkyl groups and 6 to 18 carbon atoms in the aryl groups), e.g. hexamethylphosphoric triamide.

Representative of the phosphines are trimethylphosphine, tri-propylphosphine, tricyclohexylphosphine and triphenylphosphine. When the promoter is an organo nitrogen compound it is preferably a tertiary amine or a polyfunctional nitrogen containing compound such as an amide, a hydroxyamine, a ketoamine, a di-, tri- or other polyamide, or a nitrogen compound containing 8301891 •. . 6 two or more other functional groups. Representative of these organo nitrogen promoters are 2-hydroxypyridine, 2,6-diamino-pyridine, 2- and 8-hydroxyquinoline, 1-methylpyrrolidinone, 2-imidazolidone, Ν, Ν-dimethylacetamide, dicyclohexylacetamide, 5 dicyclohexylmethylamine, Ν, Ν-diet and imidazole.

Although the organic promoter is usually added separately to the catalyst system, it is also possible to introduce it as a complex with one of the constituent metals, such as bis (triphenylphosphine) nickel dicarbonyl and tetrakis (triphenylphosphite nickel). If a complex of an organic promoter with a metal is used, one can also add free organic promoter.

The amount of the catalyst components is immaterial and that is not a parameter of the process of the invention; this can vary over a wide range. As those skilled in the art know, so much catalyst is used that a reasonable and sufficient reaction speed is obtained. However, essentially any amount of catalyst 20 supports the intended reaction and can thus be considered catalytically active. In general, however, 1 mole of catalyst per 100 to 10,000 moles of alcohol will be used, preferably 1 mole per 100 to 5000 moles of alcohol and most preferably 1 mole per 300 to 1000 moles of alcohol.

The ratio of nickel to the other metal can also vary. Generally, there is 1 mol of nickel per 0.01 to 100 mol of the other metal, preferably that ratio is 1 mol per 0.1 to 20 mol, most preferably 1 mol per 1 to 10 mol.

30. The amount of organic promoter can also vary widely, but generally 1 mole per 0.1 to 10 mole of catalyst components is used.

The amount of iodide can also vary widely, but generally at least 10 moles per 300 moles of alcohol must be present. In general, per 100 moles of alcohol there are 10 to 50 moles, preferably 17 to 35 moles of iodide. Usually, no more than 200 moles of iodide per 100 moles of alcohol will be used. However, it should be recognized that the iodide does not necessarily have to be used as alkyl iodide, it can also be used as HJ or as inorganic iodide and even as elemental iodine.

A particularly favorable embodiment of the catalyst with molybdenum-nickel or tungsten-nickel as metals can be represented by the formula X: T: Z: Q, wherein X is molybdenum or tungsten and T nickel (both with valence 10 zero or as halide , oxide or carboxylate, carbonyl or hydride), ZEJ or alkyl iodide and Q is an organophosphorus or organo nitrogen compound in which phosphorus or nitrogen is trivalent. Preferred nitrogen and phosphorus compounds have already been indicated above, and most preferred is a phosphine of the formula PR2, also as defined above. The X to T molar ratio is between 0.1: 1 and 10: 1, the X + T to Q molar ratio is between 0.05: 1 and 20: 1, and the Z to X + T molar ratio is between 1: 1 and 1000: 1.

It will be appreciated that the reaction described above lends itself well to continuous operation in which starting materials water and catalyst are continuously fed into a proper reaction zone and the reaction mixture is continuously distilled so that the volatiles are separated therefrom, then a net yield of carboxylic acid is obtained and the other organic constituents are recycled, and in the liquid phase there is also left over catalyst which is also recycled.

The invention is now further illustrated by the following non-limiting examples.

Example I

The preparation was performed in an I liter autoclave with electrically heated jacket, a magnetically driven stirrer, gas and liquid feeds, and a drain tube near the vapor-liquid interface. The test was carried out at a temperature of 210 ° C and a carbon monoxide tension of 57 bar and a hydrogen tension of 8301891.8, 13.7 bar, so that the ratio E ^ / CO was 0.24: 1 . The CO and Efforts were maintained by continuously supplying those two gases in the required quantities.

720 g of the liquid feed was supplied per hour, which consisted of 25.2% by weight of methanol, 37% by weight of methyl iodide, 9.6% by weight of methyl acetate, 12.2% by weight from acetic acid and 4.6 wt.% of water, with 0.2 wt.% nickel (introduced as nickel iodide), 0.3 wt.% molybdenum (introduced as molybdenum carbonyl) and 5 wt.% triphenyphosphine.

After a steady state was set, the reaction was run continuously for about 16 hours. Gas chromatography of the effluent showed that 12.1 gmol acetic acid per hour per liter had formed from methanol.

Comparative Example A

The preparation of Example I was repeated in the same apparatus and under the same conditions, except that the partial CO tension was now 36 bar and the partial CO tension was now 28 bar, so that the H 2 / CO ratio was now 0.77: 1. The liquid feed had the same composition as in Example I and was supplied at a flow rate of 540 g / hour. Analysis of the effluent showed that the methanol formation was now only 6.8 gmoles per hour per liter.

Comparative example B

Again, the apparatus described in Example 1 was used under the same conditions, except that the CO voltage was now 84 bar and no hydrogen was added to the system, so that the voltage was zero. The liquid feed had the same composition as in Example I, and was supplied at a flow rate of 300 g / h. Analysis of the effluent showed that the acetic acid formation was now only 3.9 gmoles per hour per liter.

8301891

Claims (3)

1. Process for the preparation of acetic acid, wherein methanol is reacted with carbon monoxide in the presence of molybdenum-nickel or tungsten-nickel as catalyst, an iodide and an organophosphorus or organo-nitrogen compound as promoter, characterized in that the reaction it also happens in the presence of so much hydrogen that the ratio between the partial hydrogen voltage and the partial carbon monoxide voltage is between 0.05: 1 and 0.4: 1. A method according to claim 1, characterized in that the ratio between the partial hydrogen voltage and the partial carbon monoxide voltage is between 0.15: 1 and 0.30: 1. 3. Method mainly according to description and / or example. 8301891