This invention relates to the production of ethanol,

Ethanol is a major commercial bulk chemical. Although some ethanol for industrial purposes is produced by fermentation of sugar, starch and similar raw materials of biological origin, the feedstock predominantly used in most industrialised countries for production of ethanol is ethylene, which is catalytically hydrated by direct vapour phase reaction with water in the presence of a supported phosphoric acid catalyst.

Ethylene is a petroleum-based feedstock and so, with increasing depletion of the world's petroleum reserves there is interest in producing ethanol by processes that are ultimately based upon coal or natural gas as feedstock, rather than upon petroleum based feedstocks, such as ethylene.

Several processes have been described in the literature for conversion of carbonaceous feedstocks, such as coal and natural gas, to ethanol. The first stage in such procedures is usually the production of synthesis gas, i.e. a CO:H2 mixture of appropriate composition, followed by ^' methanol synthesis and then by further reactions such as reaction of methanol with further synthesis gas. Although methanol can be synthesiεed in high yield and with good selectivity from synthesis gas, the second stage of this route to ethanol has usually proceeded at- best at about 80% selectivity to "realisable ethanol ^H , based upon the methanol starting material (see, for example, British Petroleum's British Patent Specification No. 2053915A). Moreover the conversion of methanol to ethanol has required the use of high temperatures and high pressures and results in production of significant quantities of by-products of relatively low value. ^--gTREAT"?--

There is accordingly a need for a process capable of converting methanol to ethanol under conditions employing relatively low temperatures and pressures and at high selectivity. The present invention accordingly seeks to provide an improved process for production of ethanol from methanol, and hence ultimately from synthesis gas, the source of which can be any carbonaceous material, such as coal or natural gas, besides petroleum based feedstocks such as naphtha or gas oil. It further seeks to provide a route to ethanol from synthesis gas which can be carried out in high yield and at high conversion rates whilst using relatively low temperatures and pressures.

According to the present invention there is provided a process for the production of ethanol which comprises reacting methanol with carbon monoxide under carbonylation conditions in a carbonylation zone in the presence of a carbonylation catalyst to form acetic acid, converting resulting acetic acid to an acetate ester, contacting a vaporous mixture comprising resulting acetate ester and hydrogen in a hydrogenolysis zone with a hydrogenolysis catalyst comprising a reduced mixture of copper and zinc oxides at a temperature in the range of from about 75°C to about 300 ^C C and at a pressure in the range of from about 0-1 kg/cm ^** - absolute (about 9-8 kPa) up to about 100 kg/cm ^** - absolute (about 9813 kPa), and recovering from the hydrogenolysis zone a hydrogenolysis product comprising ---thanol.

The process of the invention may be operated batchwise or on a semi—continuous basis, but is preferably operated on a continuous basis.

The methanol used as starting material can be εynthesised directly in known manner from synthesis gas, desirably an approximately 2:1 E_2 : CO mixture, or a suitable H2:C0:C02 mixture, the source of which may be

OMP1

any suitable carbonaceous feedstock, e.g. coal, natural gas, naphtha, gas oil or other petroleum fraction. The synthesis gas as formed may have an H2 : CO molar ratio of, for example, from about 1:1 to about 3:1. This ratio may be adjusted In known manner to give an approximately 2:1 H£JCO molar ratio or the synthesis gas may be used as formed for synthesis of methanol after suitable purification.

The reactions involved In the process of the invention can be summarised thus: (a) carbonylation -

CH3OH + CO = CH3COOH (1);

(b) esterification, e.g. -

CE3COOH + ROH = CH3COOR + H ₂ 0 (2);

(c) hydrogenolysis - CH3COOR + 2H ₂ - CH3CH2OH + ROH (3).

As the synthesis of methanol from synthesis gas involves the reaction of one mole of carbon monoxide with 2 moles of hydrogen over a suitable catalyst: CO + 2H2 = CH3OH the overall series of reactions from synthesis gas to ethanol involves the use of 2 moles of CO and A moles of H2 : CO + 4H ₂ - CH3CH2OH + H 0.

In the above equations R is a onovalent organic radical, typically an alkyl or cycloalkyl radical containing, for example, from 1 to about 12 carbon atoms. In one preferred process R is ethyl so that ^* the acetate ester subjected to hydrogenolysis is ethyl acetate. Such a process has the advantage that ethanol is essentially the sole alcohol produced so that product recovery is greatly ii -implified. In cases where R is other than ethyl, for example where R is methyl, then the hydrogenolysis yields a mixture of ethanol and the alcohol of formula ROH, such as methanol, which can then be separated, e.g. by fractional distillation under normal, reduced, or elevated pressure to

enable recovery of product ethanol as well as of the alcohol of formula ROH.

As examples of other acetate esters that can be used in the process of the invention there can be mentioned, for example, methyl acetate, n- and iso-propyl acetates, n-, iso- and _t-butyl acetates, n_-amyl acetate, ii-hexyl acetate, 2—ethylhexyl acetate, cyclohexyl acetate, and the like.

In operation of the process it is preferred to recycle alcohol of formula ROH recovered from the hydrogenolysis step for esterification of further acetic acid. When R is ethyl, each mole of ethyl acetate that is hydrogenolysed yields essentially 2 moles of ethanol; part of the ethanol produced can be recycled for esterification of further acetic acid and production of further ethyl acetate, whilst the remainder of the ethanol produced by hydrogenolysis is passed forward for product recovery and possibly for further purification.

Carbonylation of methanol to form acetic acid is effected in the presence of an effective amount of an appropriate catalyst. Such catalysts are effective for catalysing the addition of carbon monoxide across the carbon-oxygen bond of methanol .to form acetic acid. Examples of suitable catalysts include those containing Group VIII metals, such as cobalt, nickel, rhodium, and the like. Preferred catalyst systems include those comprising a rhodium containing component and a promoter component, such as iodine, bromine or a compound of bromine or iodine. As rhodium component there may be mentioned rhodium metal (in finely divided form), simple rhodium salts, organorhodlum compounds, and coordination compounds of rhodium. Specific examples include:

Rh metal, RI-1CI3, RhBr3, Rh2θ3,

Rh(N0 ₃ )3.2H ₂ 0, Rh (CO)g, RhCl ₃ . 3H ₂ '0, RhBr3,

Rh ₂ (C0) I ₂ , RhBr[(C ₆ H5) ₃ P]3, RhI [ (C ₆ H ₅ ) ₃ P]3,

RhCl[(C ₆ H ₅ ) ₃ P]- ₃ , RhCl[(C ₆ H5)3P3 ₃ H ₂ , f (C ₆ H ₅ ) ₃ P] ₃ Rh(CO)H, K ₄ Rh ₂ I ₂ (SnI ₃ )4, Rh[(C ₆ H ₅ ) ₃ P] ₂ (CO)I, RhCl [(C ₆ H ₅ ) ₃ P] ₂ (CH ₃ I) ₂ ,

Rh(CO) [P(C ₆ H ₅ ) ₃ ] ₂ C1, Rht(n-C H ₉ } ₃ P] ₂ (CO)I,

RhCl(CO) [(C6H ₅ ^' ) ₃ As3 ₂ , and the like.

Promoters include compounds of the formula RX _n wherein X is halogen, n is 1 to 3, and R is alkyl, alkylene or aryl (e.g. CH3I, Cό^Br, CH ₃ CH ₂ I and

ICH ₂ CH ₂ I), as well as l ₂ , 13 ^" / Br , Br3~,

HBr, HI, RCOBr or RCOI, where R is alkyl or aryl (e.g.

CH3COI), and substituted and unsubstituted ammonium, phosphonium, or stibonium bromides or iodides (e.g. NH4I, NH4I3, PH4I3, PH ₄ Br ₃ , and (C6H ₅ ) ₃ Pl2).

• Typical carbonylation conditions include- use of partial pressures of carbon monoxide in the range of from about 1 psig (0.07 kg/cm ² gauge or 6.9 kPa gauge) to about 25,000 psig (1757.5 kg/cm ² gauge or 172.5 MPa gauge), preferably in the range of from about 10 pεig (0.7 kg/cm ² gauge or 69 kPa gauge) to about 1000 psig (70.3 kg/cm ² gauge or 6900 kPa gauge) and temperatures in the range of from about 50°C to about 300°C, preferably in the range of from about 100°C to about 240°C. When using such rhodium catalysts there is produced in the carbonylation zone acetic acid, and possibly also methyl acetate as by-product. Such by-product methyl acetate can be recycled to the carbonylation zone and admixed with feed methanol so as to form a feed mixture containing methanol and methyl acetate in a ratio of not more than about

2:1, e.g. a feed mixture having a methanol:methyl acetate ratio of from about 0.001:1 or less up to about 2:1. If insufficient methyl acetate is produced in the carbonylation zone to provide the desired methanol:methyl acetate ratio, then part of the acetic acid can be esterified to give methyl acetate, which is recycled whilst the remaining acetic aj _-_---

_S UBSTITUTESHE

Is passed on to form an acetate ester, such as ethyl acetate, for hydrogenolysis. Alternatively the crude acetic acid/methyl acetate product mixture can be passed forward to the es erification zone where the acetic acid is esterified to give, for example, ethyl acetate and the resulting methyl acetate/ethyl acetate mixture can then be hydrogenolysed to give a mixture of methanol and ethanol which is separated into methanol and ethanol streams, of which the methanol stream can be recycled to the carbonylation zone. In any event the conditions in the carbonylation zone are preferably selected to ensure at least about 90% conversion of methanol so as to maximise production of acetic acid. Carbonylation using the above—mentioned rhodium and promoter combination catalyst systems can be operated as a liquid phase process or as a vapour phase process. Further details can be obtained, for example, from British Patent Specification No. 1233121, the entire disclosure of which is herein incorporated by reference.

Other carbonylation catalysts include nickel based catalysts, such as those described in United States Patent Specifications Nos. 4134912 and 4356320, as well as in British Patent Specification No. 2089803A. Such nickel-based catalysts are used in the presence of an iodide and include, for example: 1. Catalysts consisting essentially of nickel or a nickel compound and a promoter which -is tin or a tin compound (see United States Patent Specification No. 4134912).

2. Catalysts consisting essentially of nickel or a nickel compound and a phosphlne promoter (see United States Patent Specification No. 4356320).

3. Catalysts comprising nickel or a nickel compound, a molybdenum or tungsten co—catalyst component, and a promoter comprising an organo—phosphorus compound or an organo—nitrogen compound wherein the phosphorus and nitrogen _ar e trivalent (see British Patent Specification No.

2089S03A) .

When using a nickel-based carbonylation catalyst, the more volatile components such as alkyl iodide, e.g. .T.ethyl iodide, unreacted methanol, and by-products such as dimethyl ether and methyl acetate in the carbonylation mixture can be readily removed, for example by distillation, for recycling, and the net yield of product is substantially exclusively the desired acetic acid. In the case of liquid-phase reaction, which is preferred, the organic compounds are easily separated from the metal-containing components, for example by distillation. The reaction is suitably carried out in a reaction zone to which the carbon monoxide, methanol, the iodide, nickel or a nickel compound, the molybdenum or tungsten co-catalyst component (if used) and the promoter are fed.

In use of the nickel-based carbonylation catalysts a wide range of temperatures, e.g. 25 to 350 ^C C are suitable but temperatures of 100 to 250°C are preferably employed and the more preferred temperatures generally lie in the range o 125 to 225°C. - Typical residence ^' times will generally fall i the range of 0.1 to 20 hours. The carbonylation reaction is carried out under super-atmospheric pressure but excessively high pressures, which require special high-pressure equipment, are not necessary. In general, the carbon monoxide partial pressure is preferably at least about 1.05 kg/cm ² (103.5 kPa) but less than 141 kg/cm ² (13800 kPa) , particularly about 2.1 kg/cm ² (207 kPa) to about 14.1 kg/cm ² (1380 kPa), although 'CO partial pressures of about 0.07 kg/cm ² (6.9 kPa) up to 703 kg/cm ² (69000 kPa) can also be employed. The total pressure is, of course, that which will provide the desired carbon monoxide partial pressure to sustain a commercially acceptable reaction rate and preferably it is that required to maintain the liquid phase. The final carbonylation reaction mixture will

O PI

normally contain volatile components such as hydrocarbyl iodide, unreacted methanol and may contain the corresponding methyl acetate and/or dimethyl ether along with the product acetic acid; these volatile components, after separation from the acetic acid, can be recycled to the reaction. At the end of the desired residence time the reaction mixture is separated into its several constituents, for example by distillation. Preferably, the reaction product is Introduced into a distillation zone which may be a fractional distillation column, or a series of columns, effective to separate the volatile components from the acetic acid product and to separate the acetic acid product from the less volatile catalyst and promoter components of the reaction mixture. The boiling points of the volatile components are sufficiently far apart that their separation by conventional distillation presents no particular problem. Likewise, the higher-boiling organic components can be readily distilled away from the metal catalyst component or components and any organic promoter which may be ^' in the form of a relatively non—volatile complex. The thus recovered metal catalyst component or components as well as promoter, including the iodide component, can then be combined with fresh amounts of methanol and carbon monoxide and reacted to produce additional quantities of acetic acid. Although not necessary, the carbonylation step can be carried out in the presence of a solvent or diluent. The presence of a higher-boiling solvent or diluent, preferably - he product acid Itself, i.e. acetic acid, or the corresponding ester, i.e. methyl acetate, will make it possible to employ lower total pressures. Alternatively, the solvent or diluent may be any organic solvent which is- inert in the environment of the process such as hydrocarbons, e.g octane, benzene, toluene, xylene and tetralin, or halogenated hydrocarbons such as the chlorobenzenes , e.g. trichlorobenzene, or carboxyllc acids, or esters such as

ethylene glycol ether monoacetateε , and the like. Mixtures of solvents can also be used, such as mixtures of methyl acetate and acetic acid.

In the carbonylation step the carbon monoxide may be employed In substantially pure form, as available commercially, but inert diluents such as carbon dioxide, nitrogen, methane, and noble gases can be present if desired. The presence of inert diluents does not affect the carbonylation reaction but their presence makes it necessary to increase the total pressure in order to maintain the desired CO partial pressure. The presence of minor amounts of water such as may be found in the commercial forms of the reactants Is, however, entirely acceptable. Hydrogen is not objectionable and even may tend to stabilise the nickel-based carbonylation catalysts. Indeed, in order to obtain low CO partial pressures the CO fed may be diluted with hydrogen or. any inert gas such as those mentioned above. Surprisingly, the presence- of hydrogen does not lead to the formation of re.duction products. The diluent gas, e.g. hydrogen, may generally be used in amounts up to about 95%, if desired.

The metal catalyst components can be employed in any convenient form, viz., in the zero valent state or In any higher valent form. For example, the nickel and the tin, molybdenum or tungsten can be the metals themselves In finely divided form, or a compound, both organic or Inorganic, which is effective to introduce the co-catalyst components into the reaction system. Thus, typical compounds include the carbonate, oxide, hydroxide, bromide, iodide, chloride, oxyhalide, hydride, lower alkoxide (methoxide), phenoxide or •-ϊn, Mo, W or Ni carboxylateε wberein the carboxylate ion is derived from an alkanoic acid of 1 to 20 carbon atoms such as acetates, butyrates, decanoates, laurates, benzoates, and the like. Similarly, complexes of any of the metal co-catalyst components can be employed, e.g. carbonyls and metal alkyls as well aε chelates, association compounds and enol salts.

O PI

Exampleε of other complexes Include bis-(trlphenylphoεphine) nickel dicarbonyl, trlcyclopentadienyl trinickel dicarbonyl, tetrakis (triphenylphosphite) nickel, and corresponding complexes of the other components, such as molybdenum hexacarbonyl and tungsten hexacarbonyl. Included among the catalyst components listed above are complexes of the metal co-catalyst components with organic promoter ligands derived from the organic promoters hereinafter described.

Particularly preferred are the elemental forms of the metals, their halides, especially Iodides, and their organic salts, e.g. acetates. It will be understood that the foregoing compounds and complexes are merely illustrative of suitable forms of the several co-catalyst components and are not intended to be limiting. , The tin promoter is preferably employed in elemental form or in the form of a halide, such as stannic iodide, εtannous iodide, stannic chloride and stannic bromide, or a hydrocarbyl tin compound such as tetraphenyl tin, tetra n.-butyl tin and dibutyl diphenyl tin, or an oxide such as stannous oxide and stannic oxide, or an organo oxide such as dimethyl tin oxide and diphenyl tin oxide, or a carboxylate such as stannous caproate and tri n-propyltin acetate, or an organo—halide such as dimethyl tin di-chloride and methyl tin trichloride. The most preferred tin compounds are the halides, the organo halides and the hydrocarbyl tins.

The specified metal catalyst components employed may contain impurities normally associated with the commercially available metal or metal compounds and need not be purified further. * ^■

The organo-phosphorus promoter is preferably a trivalent phosphorus compound, for example a phosphlne of the formula

OMPI

R wherein R , and R3 may be the same or different, and are alkyl, cycloalkyl, or aryl groups, preferably containing 1 to 20 carbon atoms in the case of alkyl and cycloalkyl groups and 6 to 18 carbon atoms in the case of aryl groups. Other organo-phosphorus promoters include cyclic amides, such as hexamethylphoεphoramide, and halophosphineε . Typical phosphlnes include trimethylphoεphine, tripropylphosphine, tricyclohexylphosphine and triphenylphosphine . Preferably the organo-nitrogen promoter is a tertiary amine or a polyfunctional nitrogen-containing compound, such as an amide, a hydroxy amine, a keto amine, a di-, tri and other polyaminε or a nitrogen-containing compound which comprises two or more other functional groups. Typical organo-nitrόgen promoters include 2-hydroxypyridine, 8-quInolinol, 1-methylpyrrolldone, 2-lmidazolidone, N,N-dimethylace-tamide, dicyclohexylacetamide, dicyclohexyl—methylamine, 2 ,6-diaminopyridine, 2-quinolinol, N,N-diethyltoluamide, and imidazole.

Although generally the organic promoter is added separately to the carbonylation catalyst system, it is also possible to add it as a complex with any of the co-catalyst metals, such as bis (triphenyl-phosphine) nickel dicarbonyl and tetrakis(triphenyl phosphite) nickel. Both free organic promoters and complexed promoters can also be used. When a complex of the organic promoter and the. co-catalyst metal is used, free organic promoter can also be added.

The amount of each metal catalyst component employed Is in no way critical and is that which will provide the desired suitable and reasonable reaction rate since reaction rate is influenced by the amount of catalyst. Typically, however, each catalyst component is employed in ^' the amount of 1 ol per 100 to 10,000 olε of methanol, preferably 1 mol per 100 to 5,000 mols of methanol and most

preferably 1 mol per 500 to 1,000 mols of methanol.

The ratio of nickel to the second metal co-catalyst component can vary. Typically, it is one mol of the nickel per 0.01 to 100 mols of the εecond metal co-catalyst component, preferably the nickel component iε used in the amount of 1 mol per 0.1 to 20 mols, moεt preferably 1 mol per 1 to 10 mols of the εecond metal co-catalyεt component.

The quantity of organic promoter can alεo vary widely but typically it is used in the amount of 1 mol per 0.1 to 10 mols of the nickel catalyst component, preferably 1 mol-per 0.5 to 5 mols, most preferably 1 mol per 1 to 5 mols of the nickel catalyst component.

The amount of iodide component may also vary widely but, in general, it should be present In an amount of at least 10 mols (expressed as I) per hundred mols of methanol. Typically, there are used 10.to 50 mols of the Iodide per 100 mols of methanol, preferably 17 to 35 moles per 100 moles. Ordinarily, more than 200 mols of Iodide per 100 mols of methanol are not used. It will be understood, however, that the iodide component does not have to be added to the system as a hydrocarbyl iodide but may be supplied as another organic iodide or as the hydroiodide or other inorganic iodide, e.g. a salt, such as the alkali metal or other metal salt, or even as elemental Iodine.

Preferably carbonylation is conducted In a continuous manner in which the reactants and catalyst are continuously supplied to the appropriate reaction zone and the reaction mixture continuously distilled to separate the volatile organic constituents and to provide a net product consisting essentially of acetic acid with the other organic components being recycled and, in a" liquid-phase reaction a residual catalyst contining fraction also being recycled.

It will also be apparent that the catalytic carbonylation reaction can be carried out in the vapour phase, if desired, by appropriate control of the total

fCfREX$- OMPI

pressure in relation to the temperature so that the reactan are in vapour form when in contact with the carbonylation catalyst. In the case of vapour-phase operation, if desire the carbonylation catalyst components may be supported i.e. they may be, dispersed on a carrier of conventional type suc as alumina, silica, silicon carbide, zirconia, carbon, bauxite, attapulgus clay, and the like. The catalyst components can be applied to the carriers in conventional manner, e.g. by impregnation of the carrier with a solution of the catalyst component. Concentrations upon the carrier may vary widely, e.g. 0.01 weight percent to 10 weight percent, or higher. Typical operating conditions for vapour-phase operation are a temperature of 100°C to 350°C, preferably 150°C to 275°C and most preferably 175°C to 255-C, a pressure of about 0.07 kg/cm ² absolute (6.9 kPa absolute) to about 350 kg/cm ² absolute (34500 kPa • absolute), preferably about 3.5 kg/cm ² absolute (345 kPa absolute) to about 105.5 kg/cm ² absolute (10350 kPa absolute), and most preferably about 10.5 kg/cm ² absolute (1035 kPa absolute) to about 35 kg/cm ² absolute (3450 kPa absolute), with gas hourly space velocities of 50 to 10,000 hr ^_1 , preferably 200 to 6,000 hr" ¹ and most preferably 500 to 4,000 hr ^-1 (STP). In the case of a supported catalyst, the iodide component is included with th reactants and not on the support.

As noted above a mixture of CO and H ₂ can be used in the carbonylation step. Hence it is possible to supply a synthesis gas, e.g. a 2:1 H ₂ :CO molar mixture to the carbonylation stage, and to recover the hydrogen subsequentl for use in the hydrogenolysis step. Alternatively, as described further below, it is possible to supply to the carbonylation zone a CO-rich gas exiting the hydrogenation zone or a mixture thereof with make up synthesis gas.

In the ester formation step of a particularly

preferred proceεs acetic acid produced In the carbonylation step iε eεterified with an alcohol, preferably ethanol, in the preεence of a εuitable esterification catalyst. Conveniently such ethanol comprises product ethanol recycled following recovery subsequent to the hydrogenolysis step.

Esterification can be carried out in the liquid phase or In the vapour phase. Preferably the esterification step is operated as a continuous process.

In liquid phase esterification an acidic esterification catalyst is preferably used. Examples of suitable acidic esterification catalysts include sulphuric acid, p-tolueneεulphonic acid, and acidic ion exchange resinε such as Amberlyεt 15. (The word "Amberlyst" is a trade mark). According to one preferred procedure acetic acid from the carbonylation zone is admixed with exceεε ethanol and with a catalytic amount, e.g. from about 0.1% to about 1% by weight of the reaction mixture, of sulphuric acid typically having a specific gravity of from about 1.53 to about 1.84 (about 50 to about 66 Be). After allowing the mixture to equilibriate for a suitable period, e.g. about 5 minutes to about 3 hours, it is preheated to about 80°C and fed to the upper end of a diεtillation column, from which an ethanol/ethyl acetate/water mixture is recovered overhead, whilεt sulphuric acid and water are removed as a bottom product. This diεtillation column iε heated by introduction of live steam at its bottom end. The overhead mixture is fed to a recovery column from which, at a top temperature of 70°C, a ternary azeotrope (83% ethyl acetate, 9% ethanol, and 8% water) is recovered overhead. The resulting distillate is cooled and the organic layer which separateε iε "redistilled to give a εmall amount of ternary azeotrope as overhead product, the bottom product comprising substantially pure ethyl acetate which is passed on to the hydrogenolysis step. Water and alcohol in excess of that required to form the

ternary azeotrope is returned from the bottom end of the recovery column to the first mentioned diεtillation column.

Vapour phase esterification of acetic acid with alcohols, such as ethanol, can be effected by passage of a vaporous mixture thereof over a suitable catalyst, such as silica gel, zirconium dioxide, activated charcoal, potassium hydrogen sulphate, phosphoric acid-treated coal, calcium phosphate, or an acidic ion exchange resin, such as Amberlyst 15.

According to an alternative procedure acetic acid is esterified by reaction with an olefin to yield the corresponding acetate ester. In such a step it iε preferred to use an acidic es erification catalyst, for example one of those listed above.

The hydrogenolysis of esters is discloεed In International Patent Publication No. W082/03854 which waε published after the priority date claimed in the present application. The entire disclosure of that International Patent Publication Is herein incorporated by reference.

In the hydrogenolysis step the vaporous mixture to be contacted with the catalyst containε, in addition to the acetate ester, hydrogen either alone or in admixture with other gases (desirably gases inert to the ester and the catalyst). Hence the gaseous mixtures containing hydrogen may include inert gases such as nitrogen, or carbon monoxide.

In view of the well known behaviour of carbon monoxide as a hydrogenation catalyst poison or inhibitor it is surprising that carbon monoxide does not appear to poison the copper-containing hydrogenolysis catalysts used in the present invention, even when present in gross amounts .of about 10% or more by volume.

Thus the make up gas supplied to the hydrogenolysis zone may contain at least about 1% CO, and more usually at least about 10% by volume of carbon monoxide. Suitably it

comprises a synthesiε gaε produced by partial oxidation of a suitable hydrocarbon feedstock, such as methane, natural gas, a petroleum gas, such as propane or butane, or a petroleum fraction, such as naphtha, a gaε oil, or a fuel oil, or by

5 catalytic εteam reforming of a suitable hydrocarbon feedstock such as methane, natural gas, propane, butane, or naphtha, or by reaction of εteam/02 mixtures with coal. It is preferred to use a synthesis gas that contains CO and H2 in a molar ratio in the range of from about 1:1.5 to about

10 1:2.5, especially about 1:2. If the synthesis gas as produced, either by partial oxidation or by steam reforming, has a C0:H2 molar ratio that differs greatly from this range, then the C0:H2 ratio is preferably adjusted by known methods, e.g. subjection to a water gas shift reaction

--5 followed by CO2 removal, to a value within ^* this range. A gas with an approximately 1:2 C0 H2 molar ratio is preferably used.

Prior to its introduction to the hydrogenolysis zone steps are desirably taken to ensure that the syntheεiε

20 g _as is substantially free from sulphurous impurities. When using steam reforming to generate the synthesis gas from a hydrocarbon feedstock containing sulphurous impurities it is best to desulphurise the feedstock prior to εteam reforming. Similarly when using partial oxidiation of gaseous or liquid

² 5 hydrocarbon feedstocks, such as methane or naphtha, containing sulphurous impurities, it will often be convenient to desulphurise the feedstock prior to oxidation thereof by contact with a suitable desulphurisation medium, such as activated alumina or zine oxide, although it is also posεible

³⁰ to effect desulphurisation after partial oxidation by contacting the synthesis gas with a bed of a desulphurisation medium such as zinc oxide, possibly after removing the bulk of the sulphur—containing compounds by any of the well known sulphur removal processes (e.g. Rectisol process,

^• '-' diethanolamine waεh proceεs etc).

OMPI .* . WIPO

The hydrogenolysis zone may take the form of one or more externally cooled tubular reactors. Alternatively the hydrogenolysis zone may comprise one or more adiabatic reactors arranged in series- The hydrogenolysis step is conducted at a temperature of between about 75 ^C C and about 300 ^C C; although in many caseε the temperature may lie in the range of from about 150°C to about 200°C, In most cases it typically is between about 180 ^C C and about 240 ^C C. The total pressure is between about 0.1 kg/cm2 absolute (about 9.8 kPa) and about 100 kg/c absolute (about 9813 kPa), preferably not more than about 50 kg/cm ² absolute (about 4906 kPa), and even more preferably between about 5 kg/cπ-2 absolute (about 491 kPa) and about 25 kg/cm ² absolute (about 2453 kPa). The mixture of CuO and ZnO, before reduction, preferably contains from about 5 to about 95 percent by weight, typically from about 10 to about 70 percent by weight, of CuO and from about 95 to about 5 percent-by weight, typically from abou 90 to about 30 percent by weight, of ZnO. Hence the mixture may contain, for example, from about 20 to about 40 percent by weight of CuO and from about 60 to about 80 percent by weight of ZnO. A preferred mixture, for example, comprises from about 30 to about 36 percent by weight of CuO and from about 62 to about 68 percent by weight of ZnO. Other particularly preferred mixtures comprise from about 65 to about 85 percent by weight of CuO and from about 35 to about 15 percent by weight of ZnO, for example mixtures comprising from about 68 to about 75 percent by weight of CuO and from about 32 to about 25 percent by weight of ZnO. The hydrogenolysis catalyst may contain minor amounts of other materials εuch as carbon, sodium, titanium, zirconium, manganese, silica, diatomaceous earth, kieselguhr, and aluminium oxide. Such other materials do not usually comprise more than about 20 percent by weight calculated (except in the case of carbon) as oxide. In the

case of sodium it iε best not to exceed about 0.5 percent by weight, calculated as oxide. Hence other preferred catalyεtε Include mixtures comprising from about 40 to about 50 weight percent each of CuO and ZnO and from 0 to about 20 weight percent of alumina. The catalyst is, however, preferably esεentially free from other metals, particularly from metals of Group Vlll of the Periodic Table, such as Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt, as well as from Group VIB metals, such as Cr, Mo, and W, from the metals Ag, Re, Au and Cd, and also from elements of atomic number 80 and above, e.g. Hg and Pb. By the term "essentially free", we mean that the catalyst contains not more than about 0.1 wt% (i.e. not more than about 1000 ppm), and preferably not more than about 250 ppm, of the element in question. The catalyst may be prepared by any of the methods known in the art of forming a composite of copper oxide and zinc oxide. The catalyst may be prepared by fixing the separate oxides, by coprecipitation of the oxalates, nitrates, carbonates, or acetates, followed by calcination. The coprecipitation method is preferred. Generally, the mixture of CuO and ZnO is reduced by hydrogen or carbon monoxide at a temperature in the range of between about 160°C and about 250 ^C C for several hours, preferably for 8 to 24 hours, prior to contact with the vaporous mixture containing acetate ester and hydrogen. If the catalyst is charged in a pre-reduced form the period required for reduction can be reduced accordingly.

The mixture of CuO and ZnO is reduced prior to its use as catalyst in the hydrogenolysiε step. Hydrogen or CO, or mixtures thereof, are generally mixed with a diluent gas such as εteam, nitrogen, or cbmbustion gas, to maintain the catalyst bed temperature and to carry away the heat of reduction..

Reduction of the mixture of CuO and ZnO is complete when no more hydrogen is being reacted as shown by analysis of the inlet and outlet hydrogen. Complete reduction of the

mixture occurs when the total amount of water produced In th reduction iε equal to the εtoichiometrlc value of water whic should be produced when a given amount of copper oxide is reduced to copper. This value is about 0.079 kg of water per kg of catalyst for a mixture containing 35 weight percen of CuO.

An inert carrier material may be Included in the hydrogenolysis catalyst composition. The catalyst is generally formed Into pellets, tablets, or any other suitabl shape prior to use, by conventional techniques.

It iε advantageous that the mixture of CuO and ZnO have an internal surface area of from about 25 to about 50 sq.m. per gram. The internal surface area may be determined by the well-known BET method.

The reaction product mixture from the hydrogenolysis step may be separated from any excesε hydroge by condensation and the excess hydrogen can be compressed an recycled-. This reaction product mixture comprises, "when the acetate ester is ethyl acetate, ethanol or, when another acetate ester iε hydrogenolysed, a mixture of ethanol and another alcohol, i.e. the alcohol derived from the alcohol moiety of the acetate ester; in addition the reaction product mixture may comprise, possibly a minor amount of unconverted acetate ester. This mixture is separated In any suitable manner, e.g. by distillation. The ethanol may be used aε recovered or it can be further purified in a conventional manner such as by fractional distillation. Preferably any unconverted acetate ester recovered is recycled to the hydrogenolysis stage. When ethyl acetate is used as the acetate ester, part of the ethanol recovered may be recycled to the esterification stage; when another acetate ester, such as methyl acetate is used, the other alcohol derived from the alcohol moiety of the acetate ester (e.g. methanol) may be recycled to the esterification stage.

In the hydrogenolysis step the partial pressure of

the acetate ester may vary within wide limits, e.g. from about 0.05 kg/cm ² (about 4.9 kPa) or less up to about 10 kg/cm ² (about 981 kPa) or more. Care must however be taken to ensure that at all times the temperature of the vaporous mixture in contact with the hydrogenolysis catalyst is above the dew point of the acetate ester under the prevailing pressure conditions.

The vaporous mixture preferably contains at least an amount of hydrogen corresponding to the stoichiometric quantity of hydrogen required for hydrogenolyεiε . Uεually an excess of hydrogen over the εtoichiometric quantity will be present. In this case the excess hydrogen remaining after product recovery can be recycled to the hydrbgenolysiε zone. As will be apparent from equation (3) above, 2 moles of hydrogen are required for hydrogenolyεis of the acetate ester.

The hydrogen: ester molar ratio within the vaporous ^" mixture may vary within wide limits, from about 2:1 to about 100:1 or more. ^' s mentioned above, hydrogenolysis can be carried out using a gaseous mixture containing both CO and H2. In this case the concentration of CO- in the gaε passing into the hydrogenolysiε zone may range from about 1% by volume up to about 90% by volume or more. Uεually i will be at leaεt about 10% by volume and will often lie in the range of from about 10% ^* by volume up to about 60% b volume, e.g. about 20% by volume to about 40% by volume. The make up gaε supplied to the hydrogenolysis zone may thus be a synthesis gas, preferably one with a C0:H2 molar ratio of about 1:2. Alternatively the make up gas..supplied to the hydrogenolysiε zone may be an off gas from the carbonylation zone or a mixture thereof with synthesis gas.

Hence, in a preferred process according to the invention, the carbonylation and hydrogenolysis stepε are carried out uεing a circulating gaε containing both carbon

-gTREAlr OMPI

monoxide and hydrogen and make up gas for the circulating gaε is supplied In the form of synthesis gas at a convenient point in the gas circuit. Thus the make up synthesis gas can be admixed with the off gaε from the carbonylation zone and the resulting mixture supplied to the hydrogenolysis zone. Alternatively make up synthesis gaε can be admixed with off gas from the hydrogenolysis zone and supplied to the carbonylation zone. Either or both of the carbonylation and hydrogenolysis zones may employ internal recirculation of gas. The level of inertε (e.g. CH4 , 2, A, etc) may be controlled by taking a purge stream from the circulating gas stream at some convenient point.

If a synthesis gas other than a 2:1 H2 :CO molar mixture is used, e.g. a 3:1 H2 : CO molar mixture synthesis g _a s resulting from εteam reforming of methane, then a pressure swing absorption unit can be Included in the gas circulation loop in order to recover the excess H2 for export beyond battery limits or for use as a fuel.

In order that the invention may be clearly understood and readily carried into effect some preferred forms of ethanol synthesis plant operating according to the process of the Invention will now be described, by way of example only, with reference to the accompanying diagrammatic drawings, Figures 1 to 3 of which are each a simplified flow sheet of the respective plant.

Referring to Figure 1 of the drawings, a synthesis gaε plant 1 of conventional design iε arranged to deliver a synthesis gas of suitable composition, desirably an approximately 2:1 H2 : CO molar mixture, by way of line 2 to _a methanol synthesis plant 3, also of conventional design.

Product methanol is passed by way of line 4 to carbonylation stage 5 in which it is reacted with carbon monoxide, supplied in the form of synthesis gas, by way of line 6. Carbonylation stage 5 may operate, for example, according to the teachings of the afore-mentioned British Patent

Specification No. 1233121 or according to the teachingε of United States Patent Specification No. 4134912 or 4356320 or British Patent Specification No. 2089803A. A gaseous stream comprising hydrogen and any unreacted carbon monoxide is paεεed by way of line 7 to hydrogenolysis zone 8, whilst product acetic acid (containing possibly minor amounts of methyl acetate) produced in carbonylation zone 5 iε paεεed to esterification zone 9 by way of line 10.

In esterification zone 9 acetic acid supplied in line 10 is esterified with ethanol recycled in line 11.

EεterifIcation can be effected in zone 9 aε a liquid phase process using as catalyst an esterification catalyst such as sulphuric acid or Amberlyst 15. Alternatively esterification can be carried out as a vapour phase process using one of the catalysts listed above as suitable for such vapour phase esterification. Water produced in esterification zone 9 is removed in line 18.

Hydrogenolysis zone 8 contains a charge of -a catalyst comprising a reduced mixture of CuO and ZnO, e.g. a catalyst of composition, before reduction, 30 to 36% by weight CuO and 62 to 68% by weight ZnO. This is supplied with a vaporous mixture comprising hydrogen supplied by way of line 7 and ethyl acetate supplied from esterification zone in line 12. Hydrogenolysis zone 8 may comprise a multi-tubular reactor or a single or multi-stage adiabatic reactor. The temperature in hydrogenolysis zone 8 may range from about 75°C to about 300°C and the pressure may range from- bout 0.1 kg/cm_2 absolute (about 6.9 kPa) up to about 50 kg/cm ² absolute (about 4906 kPa). Ethanol (together possibly with a minor amount of methanol and with unchanged ethyl and/or methyl acetate) is recovered from hydrogenolysis zone 8 in line 13 and is separated by fractional diεtillation in product recovery zone 14 to give ethanol, which is recovered in line 15, methanol, which iε recycled'to carbonylation zone 5 in line 16, and methyl and/or ethyl

OMPI

acetate which is or are recycled to hydrogenolysis zone 8 in line 17. Ethanol for esterification of acetic acid in - ^' esterification zone 9 is recycled from line 15 in line 11. In the plant of Figure 2 methanol Is supplied in line 101 from a conventional methanol synthesis plant (not shown) to a carbonylation zone 102, which is generally . similar to zone 5 of the plant of Figure 1. Zone 102 iε also supplied with a CO-rich gaε In line 103.

Acetic acid product from zone 102, together with any by-product methyl acetate, is passed by way of line 104 to esterification zone 105 to which is fed ethanol in line 106. Water produced aε a result of the esterification reaction iε recovered in line 107, whilεt ethyl acetate is passed in line 108 to ester hydrogenolysis zone 109, which is generally similar to zone 8 of Figure 1. Hydrogenolysiε zone 109 may contain, for example, a charge of a hydrogenolyεiε catalyst comprising about 70 to about 72% by weight copper oxide and about 18 to about 20% by. weight zinc oxide- on a suitable support. Hydrogen for the hydrogenolysiε step is provided in the form of a mixture of recycled gas from carbonylation zone 102 in line 110 and make up synthesis gas supplied in line 111. This make up synthesis gaε comprises an approximately 2:1 H2: CO molar mixture and may be generated for example by subjection of methane to partial oxidation:

CH4 + 0 ₂ = CO + 2H ₂ . Alternatively a hydrocarbon feedstock such as naphtha, can be F ibjected to catalytic steam reforming: -(CH2) _n - + nH2θ = ^* nCO + 2nH2- T e same synthesis gaε plant -can be used to supply the synthesis gas in line 111 as is used to supply εynthesis gas to the plant used to produce the methanol in line 101. Within zone 109 there may be internal recycle of gas. This mixture is fed to ester hydrogenatlon zone by means of compressor 112 and line 113. A purge gas stream is taken in

- JR ^E Alr OMPI Λr- _. IPO .

line .114 in order to control the level of inerts (e.g. N2, A, CH etc) In the circulating gas.

Ethanol product from the hydrogenolysis zone, together posεibly with εmall amounts of ethyl acetate and methyl acetate Is supplied by way of line 114 to a separation/purification zone 115. This typically comprises a fractional distillation arrangement. Ethanol product is recovered in line 116. Any methanol present is recovered in line 117. Ethanol required for esterification Is recycled in line 106 to esterification zone 105. Any methyl acetate can be recycled in line 110.

As described, the alcohol of formula ROH used In the plant of Figure 2 Is ethanol. If another alcohol,- e.g. methanol or jl-propanol, iε used as the alcohol of formula ROH, then such alcohol ROH will be recycled in line 106 and the ester in line 108 will be the ester, CH3COOR. Line 114 will then contain a mixture of ethanol and the alcohol ROH which is separated In zone 116.

It will be appreciated that as the drawings are diagrammatic many features which would be present in an operating plant have not been described. For example, there may be provision for internal recycle of CO-rich gas within carbonylation zone 102.

In the plant of Figure 3, methane is supplied by way of line 201 to a steam reforming plant 202 of conventional design, which is also supplied with steam in line 203. Approximately 50% of the resulting approximately 3:1 H2:CO molar mixture synthesis gas is fed by way of line 204 to a methanol synthesis plant 205, also of conventional design. Product methanol is fed in line 206 to a carbonylation/esterification plant 207 in which it is first carbonylated, by reaction with CO in a gaseous mixture containing both CO and H2 supplied in line 208 ^' , to give acetic acid and then esterified by reaction with methanol supplied in line 209 to form methyl acetate. Carbonylation

can be effected according to the teachings of British Patent Specification No. 1233121 or 2089803A or of United States Patent Specification No. 4134912 or 4356320. The resulting methyl acetate iε passed in line 210 to a hydrogenolyεiε zone 211. Water produced in the esterification reaction iε removed in line 212, whilst a CO-depleted gas iε passed in line 213 to a pressure swing absorption plant 214 which is used to control the C0:H2 ratio in the gaε supplied in line 208. A hydrogen stream is recovered from plant 214 in line 215. Off gas from methanol synthesis plant 205 iε introduced Into the circulating gaε by means of line 216. In hydrogenolysiε plant 211 methyl acetate is subjected to hydrogenolysiε, using as the source make up gas, the remainder of the synthesis gas from plant 202 which is supplied in line 217. The catalyst iε that used In the plants of Figures 1 and 2. Ethanol is recovered, after suitable separation e.g. by fractional distillation, in line 218 whilst methanol is recycled in line 209. The . off gas from hydrogenolysis plant 211 Is supplied to presεure swing absorption plant 214 in line 219.

The invention is further illustrated In the following Examples. Example 1

A. A solid supported catalyst containing a rhodium component dispersed upon an inert support is prepared in the following manner: 2.37 grams of rhodium chloride trihydrate having the formula, hCl3.3H2θ iε dissolved in 50 ml of water as solvent. The resulting solution is added to 30 of a support consisting of 12-30 mesh high surface area carbon.

The mixture Is dried at 25°C in air for 8 to 16 hours and then vacuum dried at 110°C for 8 to 16 hours. A batch reactor is charged with the following ingredients: 0.396 grams of a rhodium compound having the formula RI1CI3.3H2O, 28.8 grams of a promoter consisting of methyl iodide, CH3I, 196.9 grams of acetic acid as a

OMPI IPO .

solvent, and 79 gramε of methanol aε feedstock.

The reactor iε pressurised with carbon monoxide to a total preεsure of 70.3 kg/cm2 gauge (6900 kPa gauge) corresponding to a carbon monoxide partial presεure of about 56.25 kg/cm2 gauge (5520 kPa gauge) at the reaction temperature of 175 ^C C. The reaction iε carried out at a constant presεure to yield a solution containing the following distribution of products:

89.0 wt % Acetic Acid 3.6 wt % Methyl Iodide

8.4 wt % (Catalyst, etc.) The selectivity to the formation of the desired acetic acid product is greater than 95% at substantially 100% conversion of methanol. No substantial amounts of byproducts such as aldehydes, dimethyl ether,- higher boiling carboxylic acids, methane, or carbon dioxide are formed. The time required for 50% of the methanol to be converted to acetic . acid is 335 minutes.

B. 58g ethanol, 228g acetic aci ^' and 3g concentrated sulphuric acid are heated under reflux for 6 hours. The reaction mixture is then fractionated and the crude- ester fraction is then washed with a little water, saturated with salt, washed with saturated sodium bicarbonate solution, saturated with salt, and dried over anhydrous sodium and magnesium sulphate, prior to redistillation. Ethyl acetate b.p. 76-77°C is obtained.

C. Ethyl acetate iε pumped at a rate of 34.8 ml/hr to an electrically heated gas/liquid mixing device to which hydrogen was alεo εupplied at a controlled rate and preεsure. The resulting vaporous mixture Is passed through a lagged, electrically heated line to a pre—heating coil prior to passage through a tubular reactor packed with 146 ml of a powdered catalyst. Both the tubular reactor and the pre—heating coil are immersed in a molten salt bath which waε heated to 203°C. The vaporouε mixture exiting the reactor is

OMPI

' &>

paεsed through a water cooled condenεer and the resulting condensate is collected in a water-cooled knock out pot. The exit gas presεure iε controlled to 10.55 kg/cm2 absolute (1035 kPa). The non-condensed gases are then passed through a let-down valve, the gas flow rate being monitored downstream from this valve in a wet gaε meter. A gaε flow rate of 160.4 litres/hr (measured at atmospheric presεure) Is maintained throughout the experiment.

The liquid condenεate iε analyεed by gas chromatography uεing a 2 metre εtainlesε steel column (6 mm outεide diameter) packed with polyethylene glycol (nominal molecular weight 20,000) on Chromoέorb PAW, a helium gas flow rate of 30 ml/minute and a flame ioniεation detector. The instrument iε fitted with a chart recorder having a peak integrator. The condensate iε shown to contain both ethanol and -ethyl acetate corresponding to a 86.2% conversion. with essentially 100% selectivity to ethanol.

The catalyst used is charged to the reactor as a co-precipitated mixture of CuO and ZnO containing 33 3% by weight CuO and 65*3% by weight ^' ZnO having a particle size in the range of 1.2 mm to 2.4 mm and an internal surface area of about 45 εq. m. per gram. This is pre-reduced in the reactor using a 5 vol % H2 in N2 gas mixture at 200°C for 17 hours followed by pure hydrogen at 200°C for 8 hours, the gas flow .rate in each case being about 20 litres/hr (measured at atmospheric pressure using the wet gaε meter) and. the gas pressure being 10.55 kg/cm2 absolute (1035 kPa). After this pre—reduction stage the catalyst iε at all times maintained in a hydrogen-containing atmosphere. Example 2

A magnetically-s irred, glass-lined bomb iε charged with 20g methanol, 15g of methyl acetate, lOOg of methyl iodide, lOOg of biε-triphenylphoεphine nickel dicarbonyl plus 3g of molybdenum hexacarbonyl aε co-catalyst, and 3g of triphenylphoεphine , is swept out with argon and is presεured

OMPI

to 34.6 kg/cm ² absolute (3400 kPa) with carbon monoxide.

The vessel is heated to 185°C with stirring and the temperature is maintained at 185°C. After 1 hour reaction time, gas chromatographic analysis of the reaction effluent shows it to contain 5 mol % methyl acetate in excess of the charged amount and 87 mol % acetic acid. This represents a

92% conversion of methanol to the acetyl group.

B. Ethyl acetate is prepared by a similar method to that used in Part B of Example 1. C. Using a similar procedure to that of Part C of

Example 1, ethyl acetate is subjected to hydrogenolysis usin a 1:1 CO:H2 molar mixture in place of the hydrogen stream of Example 1. At an ethyl acetate flow rate of 292 ml/hr

(i.e. a liquid hourly space velocity of 2.0 hr ^~ l) and a bath temperature of 198°C, conversion of ethyl acetate to ethanol is 25.6%, the selectivity to ethanol being essentially 100%.

In a comparison run using a 1:1 N2 H2 molar mixture- in place of the 1:1 C0:H2 molar mixture, the observed conversion to ethanol is 46.2%.

Example 3

The general procedure of Step C of Example 2 is repeated using methyl acetate in place of ethyl acetate. using a 1:1 C0:H2 molar mixture at a catalyst temperature of 180°C and a liquid hourly space velocity of 1.0 hr ^-** -, the observed conversion of methyl acetate to a mixture of methanol and ethanol is 17.7%.

In a comparison run under otherwise identical conditions, the observed conversion of methyl acetate is 35.2% when a 1:1 N2 ^:H 2 molar mixture replaces the 1:1

CO:H2 molar mixture.

Example 4

The general procedure of Step C of Example 2 is repeated using a 20 ml sample of a pelleted catalyst (3mm x 3mm) comprising 71.5% by weight CuO and 18.5% ZnO on an iner

OMPI

εupport heated by means of a fluidlsed sand bath to 185°C. At an ethyl acetate flow rate of 27.2 l/hr and using a mixture of 62.4 litres/hr H2 and 31.3 litres/hr CO the conversion to ethanol Is 22.1% at essentially 100% selectivity. In a comparison run under the same conditions except that the gas mixture is 62.4 litres/hr H2 and 31.3 litres/hr N2 the conversion to ethanol is 56.0%, alεo at esεentially 100% selectivity.