Process for the production of alcohol from a carbonaceous feedstock

The present invention relates to a process for the production of ethanol from methanol. In particular the present invention relates to a process for the production of ethanol from a carbonaceous feedstock; wherein the carbonaceous feedstock is first converted to synthesis gas which is then converted to methanol, which is then (following several process steps) converted to alkyl ester(s) which is then hydrogenated to produce ethanol. In recent years increased use and demand for alcohols such as methanol, ethanol and higher alcohols has led to a greater interest in processes relating to alcohol production. The said alcohols may be produced by the fermentation of, for example, sugars and/or cellulosic materials.

Alternatively alcohols, such as ethanol, may be produced from synthesis gas. Synthesis gas refers to a combination of H ₂ and carbon oxides produced in a synthesis gas plant from a carbon source such as natural gas, petroleum liquids, biomass and other carbonaceous materials including coal, recycled plastics, municipal wastes, or any organic material. Thus, alcohol and alcohol derivatives may provide non-petroleum based routes for the production of valuable chemicals and fuels.

Generally, the production of alcohols, for example methanol, takes place via three process steps: synthesis gas preparation, methanol synthesis, and methanol purification. In the synthesis gas preparation step, an additional stage may be employed whereby the feedstock is treated, e.g. the feedstock is purified to remove sulphur and other potential catalyst poisons prior to being converted into synthesis gas. This treatment can also be conducted after synthesis gas preparation; for example, when coal or biomass is employed. The reaction to produce alcohol(s) from synthesis gas is generally exothermic. The formation of C ₂ and C ₂₊ alcohols is believed to proceed via the formation of methanol for modified methanol catalysts and cobalt molybdenum sulphide catalysts. However, the production of methanol is equilibrium-limited and thus requires high pressures in order to achieve viable yields. Hence, pressure can be used to increase the yield, as the reaction which produces methanol exhibits a decrease in volume, as disclosed in US 3326956.

A low-pressure, copper-based methanol synthesis catalyst is commercially available from suppliers such as BASF, Johnson Matthey, and Haldor-Topsoe. Methanol

yields from copper-based catalysts are generally over 99.5% of the converted CO+CO ₂ present. Water is a by-product of the conversion of CO ₂ to methanol and the conversion of CO synthesis gas to C ₂ and C ₂₊ oxygenates. In the presence of an active water-gas shift catalyst, such as a methanol catalyst or a cobalt molybdenum catalyst the water equilibrates with the CO to give CO ₂ and H ₂ . A paper entitled, "Selection of Technology for Large Methanol Plants," by Helge Holm-Larsen, presented at the 1994 World Methanol Conference, Nov. 30-Dec. 1, 1994, in Geneva, Switzerland, reviews the developments in methanol production and shows how further reduction in costs of methanol production will result in the construction of very large plants with capacities approaching 10,0001 per day. Other processes for the production of C ₂ and C ₂₊ alcohol(s), include the processes described hereinafter;

WO 8303409 describes a process whereby ethanol is produced by carbonylation of methanol by reaction with CO in the presence of a carbonylation catalyst to form ethanoic acid which is then converted to an ethanoate ester followed by hydrogenolysis of the ethanoate ester formed to give ethanol or a mixture of ethanol and another alcohol which can be separated by distillation. Carbonylation can be effected using a CO/H ₂ mixture and hydrogenolysis can similarly be conducted in the presence of CO, leading to the possibility of circulating gas between the carbonylation and hydrogenolysis zones with synthesis gas, preferably a 2:1 H ₂ :CO molar mixture being used as make up gas. US 4122110 relates to a process for manufacturing alcohols, particularly linear saturated primary alcohols, by reacting CO with H ₂ at a pressure between 2 and 25 MPa and a temperature between 150 and 400 ^° C, in the presence of a catalyst, characterized in that the catalyst contains at least 4 essential elements: (a) copper (b) cobalt (c) at least one element M selected from chromium, iron, vanadium and manganese, and (d) at least one alkali metal.

US 4831060 relates to the production of mixed alcohols from CO and H ₂ gases using a catalyst, with optionally a co-catalyst, wherein the catalyst metals are molybdenum, tungsten or rhenium, and the co-catalyst metals are cobalt, nickel or iron. The catalyst is promoted with a Fischer-Tropsch promoter like an alkali or alkaline earth series metal or a smaller amount of thorium and is further treated by sulphiding. The composition of the mixed alcohols fraction can be selected by selecting the extent of intimate contact among the catalytic components.

Journal of Catalysis, 1988, 114, 90-99 discloses a mechanism of ethanol formation from synthesis gas over CuO/ZnO/Al ₂ O ₃ . The formation of ethanol from CO and H ₂ over a CuO/ZnO methanol catalyst is studied in a fixed-bed microreactor by measuring the isotopic distribution of the carbon in the product ethanol when isotopically-enriched ¹³ C methanol was added to the feed.

As the importance of ethanol is ever increasing in today's world, so is the need and desire to produce ethanol from a carbonaceous feedstock with a higher carbon efficiency, a higher conversion and an improved productivity and selectivity. Hence, the present invention provides a process that allows one to produce ethanol from a carbonaceous feedstock, with an improved carbon efficiency, a higher selectivity and, in particular, with a more productive conversion to ethanol.

Figure 1 represents an embodiment of a process scheme according to the present invention, wherein the references correspond to those used in the present description and appending claims. Thus, the present invention relates to a process for the conversion of methanol to ethanol, characterised by the following steps:

1. introducing methanol into an etherification unit to produce dimethyl ether (DME),

2. introducing at least a part of the dimethyl ether from step 1, together with CO and optionally H ₂ , into a carbonylation unit, in the presence of an acidic homogeneous or heterogeneous carbonylation catalyst, to produce methyl ethanoate,

3. introducing at least a part of the methyl ethanoate from step 2, together with H ₂ , into a hydrogenation unit, to produce methanol and ethanol,

4. optionally recycling at least part of the methanol from step 3, into the etherification unit of step 1, and 5. recovering ethanol from step 3.

Furthermore, the present invention relates to a process for the conversion of a carbonaceous feedstock(s) into ethanol, wherein the carbonaceous feedstock is first converted into synthesis gas, which is subsequently converted into ethanol, characterised by the following steps: 1. introducing a carbonaceous feedstock, into a synthesis gas generation unit, in order to produce a mixture of carbon oxide(s) and H ₂ (synthesis gas), 2. introducing at least a part of the carbon oxide(s) and H ₂ , from step 1 , into a methanol

synthesis unit, to produce methanol,

3. introducing at least a part of the methanol from step 2, into an etherification unit to produce dimethyl ether,

4. introducing at least a part of the dimethyl ether from step 3, together with CO and - 5 optionally H ₂ , into a carbonylation unit, in the presence of a heterogeneous carbonylation catalyst, to produce methyl ethanoate,

5. introducing at least a part of the methyl ethanoate from step 4, together with H ₂ , into a hydrogenation unit, to produce methanol and ethanol,

6. optionally recycling at least part of the methanol, from step 5, into the etherification 10 unit of step 3, and

7. recovering the ethanol from step 5.

For the purposes of the present invention and appending claims the following terms are defined hereinafter:

- The 'dew point temperature' is a threshold temperature, for example, for a given 15 pure component or mixture of components, at a given pressure, if the system temperature is raised to above the dew point temperature, the mixture will exist as a dry gas. Likewise below the dew point temperature, the mixture will exist as a vapour containing some liquid.

- ^' 'Gas' and/or 'gas phase' are defined as a pure component, or mixture of 20 components, that are above the dew point temperature.

- 'Gas hourly space velocity' (GHSV) is defined as the volume of gas fed per unit volume of catalyst per hour, at standard temperature (0 °C) and pressure (0.101325 MPa).

- 'Liquid hourly space velocity' (LHSV) is defined as the volume of liquid fed per 25 unit volume of catalyst per hour.

According to one aspect of the present invention, the synthesis gas feedstock, a mixture of carbon oxide(s) and H ₂ , that is used to produce the methanol feed stream, is preferably produced from a carbonaceous feedstock.

The carbonaceous feedstock is preferably a material such as biomass, plastic, 30 naphtha, refinery bottoms, crude synthesis gas (from underground coal gasification or biomass gasification), smelter off gas, municipal waste, coal bed methane, coal, and/or natural gas, with coal and natural gas being the preferred sources. To one skilled in the art

a combination of sources can also be used, for example coal and natural gas to advantageously increase the H ₂ to carbon ratio.

Natural gas commonly contains a range of hydrocarbons (e.g. C ₁ -C ₃ alkanes), in which methane predominates. In addition to this, natural gas will usually contain nitrogen, CO ₂ and sulphur compounds. Preferably the nitrogen content of the feedstock is less than 40 mol %, more preferably less than 10 mol % and most preferably less than 2 mol %. Processes for producing synthesis gas, in a synthesis gas plant, are well known. Each method has its advantages and disadvantages, and the choice of using a particular reforming process over another is governed by economic and available feed stream considerations, as well as by the desire to obtain the optimum (H ₂ -CO ₂ )I(CCHCO ₂ ) molar ratio in the resulting synthesis gas that is suitable for further chemical processing. A discussion of the available synthesis gas production technologies is provided in both Hydrocarbon Processing, 1999, 78:4, 87-90, and 92-93 and Petrole et Techniques, 1998, 415, 86-93, and are both hereby incorporated by reference. It is also known that the synthesis gas may be obtained by catalytic partial oxidation of hydrocarbonaceous material in a microstructured reactor as exemplified in IMRET 3: Proceedings of the Third International Conference on Microreaction Technology, ed. W. Ehrfeld, Springer Verlag, 1999, pages 187-196. Alternatively, the synthesis gas may be obtained by short contact time catalytic partial oxidation of hydrocarbonaceous feedstocks as described in EP 0303438. The synthesis gas can also be obtained via a 'compact reformer' process as described in Hydrocarbon Engineering, 2000, 5:5, 67-69; Hydrocarbon Processing, 2000, 79:9, 34; Today's Refinery, 2000, 15:8, 9; WO 9902254; and WO 0023689.

Typically, for commercial synthesis gas production the pressure at which the synthesis gas is produced from a steam reformer ranges from approximately 0.1 to 10 MPa, preferably 2 to 3 MPa and the temperatures at which the synthesis gas exits the reformer ranges from approximately 700 to 1000 °C. Likewise, for commercial synthesis gas production the pressure at which the synthesis gas is produced from an auto-thermal reformer ranges from approximately 0.1 to 10 MPa, preferably 2 to 5 MPa and the temperatures at which the synthesis gas exits the reformer ranges from approximately 700 to 1300 °C. Where the high temperatures are necessary in order to produce a favourable equilibrium for synthesis gas production, and to avoid metallurgy problems associated with

carbon dusting. The synthesis gas contains a molar ratio of (H ₂ -CO ₂ ) :(CO+CO ₂ ) ranging from 0.8 to 3.0, which is dependent on the carbonaceous feedstock(s) and the method of reforming used. For example, when natural gas is used as the carbonaceous feedstock for steam reforming, the synthesis gas obtained usually has a maximum (H ₂ -CO ₂ ) :(CO+CO ₂ ) ratio of 3.0. However, when natural gas is used as the carbonaceous feedstock for auto- thermal reforming, the synthesis gas obtained usually has a (H ₂ -CO ₂ ) :(CO+CO ₂ ) ratio of 1.5.

According to a preferred embodiment of the present invention, the molar ratio, (H ₂ - CO ₂ ):(CO+CO ₂ ), of the synthesis gas stream exiting the synthesis gas generation unit(s) is greater than 1.6, more preferably greater than 1.8 and most preferably greater than 2.0. Preferably, the molar ratio, (H ₂ -CO ₂ ):(CO+CO ₂ ), of said synthesis gas stream exiting the synthesis gas generation unit(s) is less than 3.0, preferably less than 2.75, more preferably less than 2.4 and most preferably less than 2.2.

According to another embodiment of this invention when the carbonaceous feedstock used for synthesis gas generation is not an aliphatic hydrocarbon (e.g. coal, aromatic material, biomass) the molar ratio (H ₂ -CO ₂ ):(CO+Cθ ₂ ) of the exit synthesis gas is preferably adjusted to the target value by addition of H ₂ or removal of CO ₂ .

CO ₂ may be removed by the use of a simple, yet effective, separation method known to those skilled in the art, for example, a "membrane separation method". Such membrane technologies can be found in 'Purification and Recovery Options for Gasification' D. J. Kubek, E. Polla, F. P. Wilcher, UOP, 1996.

Alternatively, CO ₂ may be recovered and removed by any suitable method(s) known to those skilled in the art, for example, by reacting with amines; performing a methanol wash (i.e. the RECTISOL process) and/or by using hot potassium carbonate (e.g. the BENFIELD process).

According to a preferred embodiment of the present invention, the exit stream obtained from the synthesis gas generation unit (e.g. using a steam reformer), comprises essentially a mixture of carbon oxide(s) and H ₂ . It can also comprise water, nitrogen and traces of unconverted hydrocarbons (e.g. C ₁ -C ₃ alkanes). According to a preferred embodiment of the present invention, during synthesis gas generation, an additional stage may be employed whereby the carbonaceous feedstock is first purified to remove sulphur and other potential catalyst poisons (such as halides or

metals e.g. mercury) prior to being converted into synthesis gas; alternatively this treatment can also be performed after synthesis gas preparation for example, when coal or biomass are used.

According to an embodiment of another aspect of the present invention, at least part of the said syngas stream is then introduced into a methanol synthesis unit, in order to produce a stream comprising methanol. Preferably the molar ratio, (H2-CO2):(CO+CO2), of said synthesis gas stream fed into the methanol synthesis unit is greater than 1.6, more preferably greater than 1.8 and most preferably greater than 2.0. Preferably the molar ratio, (H2-CO2):(CO+CO2), of said synthesis gas stream fed into the methanol synthesis unit is less than 3.0, more preferably less than 2.5 and most preferably less than 2.2.

According to a preferred embodiment of the present invention, the methanol synthesis unit may be any reactor that is suitable for producing methanol, for example a fixed bed reactor, which can be run in adiabatic or isothermal mode e.g. a multi-tubular reactor; or a fluidised bed reactor. Preferably the methanol synthesis unit is operated at a temperature of more than

200 ⁰ C, preferably more than 220 ⁰ C and most preferably more than 240 ⁰ C; and less than 310 ⁰ C, preferably less than 300 ⁰ C and most preferably less than 290 ⁰ C. Preferably the methanol synthesis unit is operated at pressure of more than 2 MPa and preferably more than 5 MPa; and less than 10 MPa and preferably less than 9 MPa. In fact, since methanol synthesis is an exothermic reaction, the chosen temperature of operation is governed by a balance of promoting the forward reaction (i.e. by not adversely affecting the equilibrium) and aiding the rate of conversion (i.e. higher productivity).

The catalysts used for methanol synthesis can be divided into 2 groups: i. the high pressure zinc catalysts, composed of zinc oxide and a promoter; and ii. low pressure copper catalysts, composed of zinc oxide, copper oxide, and a promoter.

Hence, according to a preferred embodiment of the present invention, the preferred methanol synthesis catalyst is a mixture of copper, zinc oxide, and a promoter such as, chromia or alumina. Under the aforementioned operating conditions, these said mixtures can catalyse the production of methanol from CO and H ₂ with a high selectivity.

Additionally by-products such as methane, ethanol and other higher alcohols may also be produced during methanol synthesis. According to a preferred embodiment of this

aspect of the present invention, the stream exiting the methanol synthesis unit is subsequently purified to remove said by-products by any methods known to those skilled in the art.

According to the present invention, methanol is introduced into an etherification unit to produce dimethyl ether. Preferably, said methanol introduced into the etherification unit emanates from the aforementioned methanol synthesis unit; however, the methanol may also be produced by any other appropriate method known to those skilled in the art, e.g. via wood pyrolysis.

For example, dimethyl ether is commonly produced by the Lewis or Bronsted acid catalysed etherification of methanol; numerous catalysts for this reaction have been reported e.g Houben-Weyl, vol. VI/3, part 3, Georg Thieme Verlag, Stuttgart 1965, pp. 17, 18. Dr. Alexander Wacker, Gesellschaft fur elektrochemische Industrie GmbH, DE 680328, 1934 (P. Halbig, O. Moldenhauer). R. L. Brown, W. W. Odells, US 1873537, 1927. N. V. de Bataafsche Petroleum Maatschappij, FR 701335, 1930; GB 332756, 1929;GB 350010, 1931;GB 403402, 1932. Further examples of etherification catalysts include iron chloride, copper sulphate, copper chloride, manganese chloride, aluminium chloride, aluminium sulphate, chromium sulphate, alums, thorium compounds, aluminium oxide, titanium oxide, barium oxide, silica gel, and aluminium phosphate, acidic ionic liquids. The chosen etherification catalyst may be either homogeneous or heterogeneous. According to the present invention the preferred etherification catalysts are heterogeneous catalysts, such as aluminium oxides and aluminium silicate, which can be modified by doping. Corresponding etherification catalysts and process operating conditions that may be advantageously used according to the present invention are described in Mobil Oil Corporation, DE 2818831, 1978 (F. G. Dwyer, A. B. Schwartz); DE-OS 3201155, 1982 (W. K. Bell, C. Chang); Du Pont, EP-A 99676, 1983 (D. L. Brake); Mitsubishi Chemical Industries, EP-A 124078, 1984 (N. Murai, K. Nakamichi, M. Otake, T. Ushikubo).

Zeolites, strong acid ion exchange resins, supported heteropolyacids (such as silicotungstenic acid) and mixtures thereof, can also advantageously be used as etherification catalysts; wherein supported heteropolyacids catalysts are preferably used according to the present invention.

During the methanol etherification process to produce dimethyl ether, water is also

produced. It is preferred according to the present invention to proceed with the removal of said water prior to introduction of the dimethyl ether to the next stage of this invention.

According to the present invention, at least a part of the aforementioned dimethyl ether together with CO, and optionally H ₂ , are introduced into a carbonylation unit in the presence of an acidic homogeneous or heterogeneous carbonylation catalyst. The applicants have found a preferred embodiment, whereby it is especially advantageous (e.g. increased catalyst lifetime) to conduct the carbonylation process in the presence of hydrogen. Preferably at least part, most preferably all, of the said dimethyl ether, emanates from the aforementioned etherification unit. However, according to an alternative embodiment of the present invention, at least a part of the dimethyl ether introduced into the carbonylation unit may also originate from a direct synthesis gas to dimethyl ether process. The reaction to produce dimethyl ether is exothermic and equilibrium controlled, as is the methanol synthesis reaction; however the reaction to produce dimethyl ether is more thermodynamically favourable than the methanol synthesis, under methanol synthesis reaction conditions. This has lead to close-coupling of these reactions in a single reactor as a means of increasing the potential yield per pass. This can be achieved by either adding acidity to the methanol catalyst, or by mixing an acidic catalyst with a methanol catalyst. Such processes are often referred to as direct synthesis e.g. Snamprogetti, SpA., DE 2362944, 1973 (G. Giorgio); DE 2757788, 1977 (G. Manara, B. Notari, V. Fattore); DE 3220547, 1982 (G. Manara).

Suitable carbonylation catalyst(s) used in the present invention are acidic homogeneous or acidic heterogeneous catalysts. For example, the catalyst can be selected from homogeneous strong acid catalysts, e.g. phosphoric acid; and, heterogeneous analogues such as supported phosphoric acid and SAPO' s catalysts. Other heterogeneous catalyst(s) include: zeolites; strong acid ionic resins catalysts; and mixtures thereof.

According to an embodiment of the present invention the carbonylation process for producing methyl ethanoate comprises reacting DME with carbon monoxide and preferably hydrogen under substantially anhydrous conditions, in the presence of a catalyst comprising a zeolite having at least one 8-member ring channel, said'8-member ring channel being interconnected with a channel defined by a ring with greater than or equal to 8 members, said 8-member ring having a window size of at least 2.5 Angstroms x at least 3.6 Angstroms and at least one Brønsted acid site and wherein the zeolite has a silica:X ₂ θ3

ratio ofat least 5, wherein X is selected from aluminium, boron, iron, gallium and mixtures thereof, such as mordenite or ferrierite.

The feed to the carbonylation unit may comprise substantially pure carbon monoxide (CO), for example, carbon monoxide typically provided by suppliers of industrial gases, or it may contain impurities that do not interfere with the conversion of the DME to the target methyl ethanoate, such as nitrogen, helium, argon, methane and/or carbon dioxide. For example, the feed may comprise CO that is typically made commercially by removing hydrogen from synthesis gas via a cryogenic separation and/or use of a membrane. In fact, according to a preferred embodiment of the present invention, the carbon monoxide feed contains substantial amounts of hydrogen. For example, the feed may be what is commonly known as synthesis gas, and is preferably derived from the above synthesis gas generation unit.

As indicated, the carbonylation process takes place in the presence of both carbon monoxide and hydrogen; preferably, the ratio of carbon monoxide: hydrogen is in the range 1:3 to 15:1 on a molar basis, such as 1:1 to 10:1.

Zeolites, both natural and synthetic are microporous crystalline aluminosilicate materials having a definite crystalline structure as determined by X-ray diffraction. The chemical composition of zeolites can vary widely but they typically consist of SiO ₂ in which some of the Si atoms may be replaced by tetravalent atoms such as Ti or Ge, by trivalent atoms such as Al, B, Ga, Fe or by bivalent atoms such as Be, or by a combination thereof. A zeolite, is comprised of a system of channels which may be interconnected with other channel systems or cavities such as side-pockets or cages. The channel systems are uniform in size within a specific zeolite and may be three-dimensional but are not necessarily so and may be two-dimensional or one-dimensional. The channel systems of a zeolite are typically accessed via 12-member rings, 10-member rings or 8 member rings. The specific zeolite mentioned hereinabove contains at least one channel which is defined by an 8-member ring. Preferred zeolites are those which do not have side-pockets or cages within the zeolite structure. The Atlas of Zeolite Framework Types (C. Baerlocher, W. M. Meier, D. H. Olson, 5 ^th ed. Elsevier, Amsterdam, 2001) in conjunction with the web-based version (http://www.iza-structure.org/databasesλ is a compendium of topological and structural details about zeolite frameworks, including the types of ring structures present in

the zeolite and the dimensions of the channels defined by each ring type. For the purposes of the present invention, the term 'zeolite' also includes materials having a zeolite-type structure such as delaminated porous crystalline oxide materials such as ITQ-6 and pillared layered oxide materials such as ITQ-36. The process of the present invention preferably employs a zeolite having at least one channel defined by an 8-member ring of tetrahedrally co-ordinated atoms (tetrahedra) with a window size having a minimum dimension of 2.5 Angstroms x 3.6 Angstroms. The 8-member ring channel is interconnected with at least one channel defined by a ring with equal to or greater than 8 members, such as 10 and/or 12 members. The interconnected 8-, 10, and 12- member ring channels provide access to Brønsted acid sites contained in the 8- member ring channels to enable the carbonylation of the DME to proceed at acceptable rates.

The zeolite for use in the present invention may consist of interconnected channels defined solely by 8-member rings, such as zeolites of framework type CHA, for example, chabazite and framework type ITE, for example ITQ-3. Preferably, however, the zeolite has at least one channel formed by an 8-member ring and at least one interconnecting channel defined by a ring with greater than 8 members, such as a 10, and/or 12 member ring. Non-limiting examples of zeolites having 8- member ring channels and interconnecting larger ring channel systems include zeolites of framework type MOR, for example, mordenite, FER, such as ferrierite and ITQ6, OFF, for example, offretite, GME, for example Gmelinite, MFS, such as ZSM-57, EON such as ECR-I and ETR such as ECR-34. Preferably, the zeolites for use in the process of the present invention have at least one 8-member ring channel interconnected with at least one 12-member ring channel, such as those of framework type MOR, OFF and GME, for example, mordenite, offretite and gmelinite.

The aperture (pore width) of an 8-member ring channel of the zeolite has a minimum dimension of 2.5 x 3.6 Angstroms. Channel dimensions of zeolite framework types may be found, for example, in the Atlas of Zeolite Framework Types. In addition, M.D. Foster, I. Rivin, M.M.J. Treacy and O. Delgado Friedrichs in "A geometric solution to the largest-free-sphere problem in zeolite frameworks" Microporous and Mesoporous Materials 90 (2006) 32-38, have used Delaunay triangulation methods applied to known zeolite frameworks and have tabulated the largest free-sphere diameters for diffusion along

the three principal crystallographic directions for the 165 zeolite frameworks that are currently listed in the Atlas of Zeolite Framework Types. Ring window sizes may be modified by suitable atomic substitutions that change bond lengths and bond angles of the tetrahedrally co-ordinated atoms and the bridging oxygens.

A partial listing of zeolite framework types having at least one interconnected 8 member ring channel of minimum dimension of 2.5 x 3.6 Angstroms taken from The Atlas of Zeolite Framework Types is given below:

Zeolites are available from commercial sources. Alternatively they may be synthesized using known techniques. In general, synthetic zeolites are prepared from aqueous reaction mixtures comprising sources of appropriate oxides. Organic directing agents may also be included in the reaction mixture for the purpose of influencing the production of a zeolite having the desired structure. After the components of the reaction mixture are properly mixed with one another, the reaction mixture is subjected to appropriate crystallization conditions. After crystallization of the reaction mixture is complete, the crystalline product may be recovered from the remainder of the reaction mixture. Such recovery may involve filtering the crystals, washing with water followed by a calcination treatment at high temperature. The synthesis of zeolites is described in numerous references. For example, zeolite Y and its synthesis is described in US

3,130,007, zeolite ZSM-23 is described in US 4,076,842 and J.Phys. Chem. B, 109, 652- 661 (2005), Zones, S.I. Darton, R.J., Morris, R and Hwany, S-J; ECR-18 is described in Microporous Mesoporous Mat., 28, 233-239 (1999), Vaughan D.E.W. & Strohmaier, K.G.;

Theta-1 is described in Nature, 312, 533-534 (1984). Barri, S.A.I., Smith W.G., White, D and Young, D.; Mazzite is described in Microporous Mesoporous Mat., 63, 33-42 (2003), Martucci, A, Alberti, A, Guzmar-Castillo, M.D., Di Renzo, F and Fajula, F.; Zeolite L is described in Microporous Mesoporous Mat., 76, 81-99 (2004), Bhat, S.D., Niphadkair, P.S., Gaydharker, T.R., Awate, S.V., Belhekar, A.A. and Joshi, P.N and also in J. Ind. Eng. Chem. Vol. 10, No. 4 (2004), 636-644, Ko Y.S, Ahn W.S and offretite is described in Zeolites 255-264, Vol. 7, 1987 Howden M.G.

The zeolite catalyst for use in the process of the present invention is used in the acid form, generally referred to as the 'H' form of the zeolite, for example, H-mordenite, H-ferrierite. Other forms of the zeolite, such as the NH ₄ form can be converted to the H- form, for example, by calcining the NH ₄ form at elevated temperature. The acid form of a zeolite will possess Brønsted acid (H ⁺ ) sites which are distributed among the various channel systems in the zeolite. For example, H-mordenite has H ⁺ sites located in the 12 member ring channels and in the 8 member ring channels. The number or concentration of H ⁺ species residing in any particular channel system can be determined by known techniques such as infra-red and NMR spectroscopic techniques. Quantification of Brønsted acidity by FTIR and NMR spectroscopy is described, for example, in Makarova, M.A., Wilson, A.E., van Liemt, B.J., Mesters, C. de Winter, A.W., Williams, C. Journal of Catalysis 1997, 172, (1), 170. The two types of channels in H-Mordenite (defined by 12 member rings and 8 member rings) give rise to at least two bands associated with the hydroxyl region of H-mordenite, one corresponding to vibration into the larger pores and the other, at a lower frequency, vibrating into the smaller pores. It has been shown that there is a correlation between the number OfH ⁺ sites located in an 8-member ring channel and the carbonylation rate whereas no such correlation has been observed for 12-member ring channels. It has been found that carbonylation rates increase in parallel with the number of H ⁺ sites within 8 member ring channels. In contrast, no correlation is evident with the number of H ⁺ sites within 12 member ring channels. The number of H ⁺ sites within 8-member ring channels can be controlled by replacement of the H ⁺ with metal cations such as Na ⁺ or Co ²⁺ using known ion-exchange techniques. The chemical composition of a zeolite may be expressed as involving the molar relationship:

SiO ₂ : X ₂ O ₃

wherein X is a trivalent element, such as aluminium, boron, iron and/or gallium, preferably aluminium. The SiO ₂ : X ₂ O ₃ ratio of a given zeolite is often variable. For example, it is known that mordenite can be synthesized with SiO ₂ : Al ₂ O ₃ ratios of 6 to 90 or greater, zeolite Y, from about 1 to about 6, chabazite from about 2 to 2000 and gmelinite may be synthesised with SiO ₂ : Al ₂ O ₃ ratios of greater than 4. In general, the upper limit of the SiO ₂ : X ₂ O ₃ ratio is unbounded, for example, the zeolite ZSM-5. The specific zeolite mentioned hereinabove have a SiO ₂ : X ₂ O ₃ molar ratio of at least 5, preferably in the range 7 to 40, such as 10 to 30. Suitably, the SiO ₂ : X ₂ O ₃ molar ratio is less than or equal to 100. Particular SiO ₂ : X ₂ O ₃ ratios can be obtained for many zeolites by dealumination (where X is Al), by standard techniques using high temperature steam treatment or acid washing.

In one embodiment the carbonylation catalyst is composed of mordenite or ferrierite, or mixtures or combinations of the two, either per se (i.e., in the acid form, generally referred to as H-mordenite and H-ferrierite), or optionally ion-exchanged or loaded with one or more metals such as copper, nickel, iridium, rhodium, platinum, palladium, or cobalt. Mordenite catalysts may, in addition to silicon and aluminum atoms, contain further elements in the zeolite framework, particularly gallium and/or iron. Ferrierite catalysts may, in addition to silicon and aluminum atoms, contain further elements in the zeolite framework, particularly boron, gallium and/or iron. Framework modifier elements to both types of catalysts may be introduced to the framework by any conventional means. Where a framework modifier element is used in either a mordenite or ferrierite catalyst, the catalyst suitably has a ratio of silica to the oxide of the framework modifier element would be from about 10:1 to about 100:1. T-atom incorporation where T is B, Ga or Fe into zeolites of the ferrierite structure is disclosed in Melian-Cabrera et al., Catalysis Today 110 (2005) 255-263; Shawki et al., EP (Application) 234,766 (1987), Sulikowski et al., J Chem. Soc, Chem. Comm., 1289 (1989); Borade et al., J. Chetn. Soc, Chem.Comm., 2267 (1996); Jacob et al, Zeolites 430 (1993) Vol. 13. T-atom incorporation into zeolites of the mordenite structure where the T-atom is Ga or Fe is disclosed in Smith, WO 05/085162.

Mordenite (commonly available as Na-mordenite, Nt- ₄ -mordenite or H-mordenite) is a member of the aluminosilicate zeolite class of minerals. The formula of mordenite in its Na-form is usually given as Na(AlSi ₅ Oi ₂ ).3H ₂ O or (Na ₂₅ Ca ₅ K ₂ )Al ₂ Si ₁₀ O ₂₄ JH ₂ O. It is available from a number of commercial sources of such materials. H-Mordenite has

elliptical 6.5 x 7.0 A channels (12 member rings with window openings running in the crystallographical c-direction). It also has a system of smaller channels running perpendicular to the 12 member ring channels (and running in the 6-direction). These small channels consist of 3.4 x 4.8 A channels having 8 member ring windows of these dimensions. The mordenite structure also possesses a zig-zag Y-branching of the pore structure due to twisted 8 member rings (in the crystallographic c-direction). This results in a distorted window to each side of the Y-branching of 2.6 x 5.7 A. Ferrierite is another member of the aluminosilicate zeolite class of minerals, also available in the Na-, NH ₄ - and H- forms. In the Na-form its formula generally is given as Na _{0 8} K ₀₂ MgSi ₁₅ Al ₃ O ₃₆ .9H ₂ O or (Mg ₅ Na ₂₃ K ₂₅ Ca) _3- SMg[Al _5-7 Si _{27 S-31} O ₇₂ J-ISH ₂ O. It, too, is available from various commercial sources. Additional information on these and other materials can be found on the website of the International Zeolite Association, www.iza-online.org. H-Ferrierite has 4.2 x 5.4 10-member ring channels running in the crystallographic c-direction and 3.5 x 4.8 8-member ring channels running in the δ-direction. The reaction is to be conducted substantially in the absence of water when using the specific zeolite mentioned hereinabove, for example by drying the zeolite catalyst before beginning the operation, for example, by preheating to 400- 500°C.

According to an embodiment of the present invention, the carbonylation process is run at temperatures at or below about 600°C, that is, at temperatures of from about 200 to about 600 °C, preferably from about 250 to about 400 °C. One feature of the process is that, surprisingly, the carbonylation of dimethyl ether (DME) to methyl ethanoate using mordenite zeolite based catalysts and in the substantial absence of water can be performed with very high selectiyities.

According to an embodiment of the present invention total operating pressures in the carbonylation process are from about 1 MPa to about 20 MPa, preferably from about 1 MPa to about 15 MPa and most preferably from about 2.5 MPa to about 10 MPa; preferably with carbon monoxide pressures greater than 1 MPa and dimethyl ether pressures below 0.5 MPa

The zeolite carbonylation process is carried out under substantially anhydrous conditions, i.e. in the substantial absence of water. The carbonylation of dimethyl ether to methyl ethanoate does not generate water in situ. Water has been found to inhibit the carbonylation of dimethyl ether to form methyl ethanoate. This is in comparison to prior

art processes, in which dimethyl ether was a co-feed, and in which water was also fed to the reaction. Water is thus kept as low as feasible, in order to allow the desired reaction to proceed best. To accomplish this, the ether and CO reactants and the catalyst are preferably dried prior to introduction into the process. The carbonylation process may be run as either a continuous or a batch process, with continuous processes typically preferred. Essentially, the heterogeneous process is a gas-phase operation, with reactants being introduced in either liquid or gaseous phase, and products withdrawn as gases. As desired, the reaction products may subsequently be cooled and condensed. The heterogeneous catalyst may be used as convenient, in either a fixed bed or a fluidized bed. In operating the process, unreacted starting materials may be recovered and recycled to the carbonylation unit. The product methyl ethanoate is then fed to the hydrogenation unit; however, some methyl ethanoate may also be recovered and sold as such, or may be forwarded to other chemical process units as desired. Preferably, the entire reaction product may be sent to the hydrogenation unit for conversion of the methyl ethanoate to methanol and ethanol.

According to a preferred embodiment of the present invention, the methyl ethanoate stream exiting the carbonylation unit is then purified to remove ethanoic acid and water, before being introduced into the hydrogenation unit. After purification and before introduction into the hydrogenation unit, the alkyl ester stream contains preferably less than 5 wt% of ethanoic acid, more preferably less than 1 wt%, even more preferably less than 0.1 wt% and most preferably less than lOOppm by weight.

After purification and before introduction into the hydrogenation unit, the alkyl ester stream contains preferably less than 20 mol % of water, preferably less than 2 mol %, more preferably less than 0.2 mol % and most preferably the hydrogenation unit is operated in the absence of water.

According to an alternative embodiment of the present invention, water represents between 0.5 and 20 mol %, more preferably between 0.5 and 15 mol % and most preferably between 1 and 5 mol % of the total liquid feed (alkyl ester, alcohol and water) to the hydrogenation unit. These said purification treatments are especially advantageous when a copper- based catalyst is employed in the hydrogenation unit.

Furthermore, the methyl ethanoate feed is preferably vapourised prior to contacting

the hydrogenation catalyst by either, heating the methyl ethanoate in a separate vapouriser prior to having contact with the H ₂ or, by heating the methyl ethanoate within the H ₂ (e.g. either in a separate vessel or on a prebed to the hydrogenation unit). The feed mixture, including recycles, entering the hydrogenation unit (e.g. the H ₂ together with the methyl ethanoate and optionally ethyl ethanoate) is more than 10 ⁰ C, preferably more than 20 ⁰ C above its dew point temperature.

According to the present invention, methyl ethanoate is introduced into a hydrogenation unit together with H ₂ to produce a stream comprising ethanol and optionally methanol. In addition to the production of ethanol the hydrogenation process also produces other reaction products such as ethyl ethanoate (e.g. trace amounts of methane, ethane, diethyl ether, water and ethanal) and unreacted starting materials e.g. unreacted methyl and/or ethyl ethanoate (from recycle) and unreacted H ₂ .

When methyl ethanoate is used as a feed for the hydrogenation process, it also produces ethyl ethanoate by trans-esterifϊcation. The proportion of ethyl ethanoate present in the exit stream will be determined by the nature of the catalyst and the degree of conversion. The proportion of ethyl ethanoate may be further increased, if desired, by introducing an acidic function into the catalyst bed to promote in situ trans-esterification.

This trans-esterification is advantageous as it reduces the amount of unreacted methyl ethanoate passed through with methanol, as a mixture to other process stages such as the etherifϊcation unit and/or as a recycle to the hydrogenation unit. More preferably the quantities of these fluxes are minimized.

Preferably the hydrogenation unit is operated at high conversion of methyl and/or ethyl ethanoate to ethanol, such as more than 75%, preferably more than 90% and most preferably more than 95%. It is has also been found to be an advantage of this process, that the selectivity of the methyl and/or ethyl ethanoate hydrogenation to ethanol can be further increased at the expense of undesirable by-products, such as the aforementioned alkanes (e.g. ethane and methane).

Preferably, at least a part, preferably all, of the hydrogen fed into the hydrogenation unit emanates from the synthesis gas generation procedure (i.e. it is obtained during the aforementioned CO/H ₂ separation), where, if need be, the H ₂ content can be further increased by subjecting the said synthesis gas to a water-gas shift reaction and a

subsequent H ₂ separation.

Alternatively the H ₂ stream may originate from a variety of other chemical processes, including ethene crackers, styrene manufacture and catalytic reforming. However, it is known that the main commercial processes for purposeful generation of H ₂ are autothermal reforming, steam reforming and partial oxidation of hydrocarbon feedstocks such as natural gas, coal, coke, deasphalter bottoms, refinery residues and biomass. H ₂ may also be produced by electrolysis of water.

The overall choice of technology for producing H ₂ is generally determined by the following economic considerations and factors: i. feedstock cost ii. feedstock availability iii. capital cost iv. local energy and operating costs; and v. environmental considerations The catalyst(s) employed in the hydrogenation unit is selected from any of the following:

(i) a precious metal based catalyst, comprising of at least one noble metal from Group VIII of the periodic table (CAS version, for example iron, ruthenium, osmium, cobalt, rhodium, iridium, nickel, palladium, platinum) and at least one of the metals chosen from rhenium, tungsten and/or molybdenum; and optionally an additional metal, that is capable of alloying with said Group VIII noble metal; or

(ii) a copper-based catalyst (for example a copper chromite or a mixed copper metal oxide-based catalyst wherein the second metal can be copper, zinc, zirconium or manganese).

According to a preferred embodiment of the present invention, the catalyst(s) employed in the hydrogenation unit is a copper-based catalyst, most preferably comprising copper and zinc.

AU of the aforementioned catalysts may advantageously be supported on any suitable support known to those skilled in the art; non-limiting examples of such supports include carbon, silica, titania, clays, aluminas, zinc oxide, zirconia and mixed oxides. Preferably, the palladium-based catalyst is supported on carbon. Preferably, the copper-

based catalyst is supported on zinc oxide and preferably comprises between 20 and 40 wt % of copper.

According to a preferred embodiment of the present invention, the catalyst(s) employed is heterogeneous. The hydrogenation process may be operated in a gas phase, or a mixed gas/liquid phase regime. The mixed gas/liquid phase regime is where the reactant mixture, at the reactor conditions, is below the dew point temperature.

The hydrogenation can be conducted in batch or semi continuous or continuous mode. Continuous mode of operation is the most preferred. The hydrogenation reaction can be conducted in adiabatic or isothermal mode; where adiabatic mode of operation is preferred. Suitable reactors include single, or a plurality, of adiabatic bed reactors which can be used in series or parallel. For reactors utilised in series, heat exchangers and/or intercoolers and/or additional reactant and/or recycle of intermediates can be employed in between successive reactors to control the reaction temperature. The preferred adiabatic temperature rise is less than 50 °C, preferably less than 25 °C and most preferably less than 10 °C. The preferred use of adiabatic reactors is in series. The adiabatic reactors may be operated at different temperatures depending on composition of the individual reactor feeds.

The hydrogenation can also be conducted in multi-tubular reactors in which case a cooling/heating medium is circulated around the tubes to control the temperature. For exothermic reactions, as such, there will be a radial temperature gradient in the reactor, the preferred gradient is less than 50 ⁰ C preferably less than 25 °C most preferably less than 10 °C. The preferred flow regime in this type of reactor is turbulent rather than laminar, this corresponds to a Reynolds number greater than 2100 (where the velocity is approximated by velocity in an unpacked tube).

The hydrogenation reaction can also be conducted in other reactor types such as fluidised bed, spinning basket and buss loop, heat exchanger reactors. A mixed liquid/gas phase hydrogenation reaction can be conducted with co-flow or counterflow of the H ₂ and gas to the liquid (e.g. a bubble reactor). The preferred mode of operation of gas/liquid reactors is co-flow, also known as trickle bed operation; this can be conducted in at least one tubular and/or multi-tubular reactor in series. The hydrogenation reaction may change from a mixed gas/liquid phase to a fully gas phase reaction, as the reaction proceeds down

the reactor. The mixed phase hydrogenation can also be conducted in other types of reactors, or within a combination of different reactors, for example in a slurry or stirred tank reactor with, or without, external circulation and optionally operated as a cascade or stirred tanks, a loop reactor or a Sulzer mixer-reactor. The hydrogenation reactor(s) preferably operate at a temperature of more than 150

⁰ C, but less than 290 ⁰ C.

According to a preferred embodiment of the present invention the reaction temperature is more than 150 °C, preferably more than 170 °C and most preferably more than 190 ⁰ C; and less than 250 °C. The hydrogenation reaction may be operated at a pressure of more than 3 MPa, preferably at a pressure of more than 5 MPa; and at a pressure of less than 15 MPa, more preferably at a pressure less than 13 MPa and most preferably at a pressure less than 9

MPa.

According to an embodiment of the present invention, when the hydrogenation unit(s) is operated with a copper-based catalyst, the feed mixture introduced into the unit(s) is always above its dew point temperature.

The GHSV for continuous operation of the hydrogenation unit may be in the range

50 to 50,000 h ^"1 , preferably from 1,000 to 30,000 h ^"1 and most preferably from 2,000 to

9,000 If ¹ . . The ester liquid substrate introduced into the hydrogenation unit preferably has an

LHSV of less than 1O h ^"1 , more preferably of less than 5 h ^'1 and most preferably of less than 3 h ^'1 ; for example, a typical LHSV for normal operation is approximately I h ^'1 .

According to a preferred embodiment of the present invention, the molar ratio of H ₂ to [methyl ethanoate and ethyl ethanoate] that is introduced into the hydrogenation unit is greater than 2:1, preferably the molar ratio is greater than 4:1 and most preferably the molar ratio is greater than 5:1; and is less than 100:1, preferably less than 50:1 and most preferably less than 15:1.

According to a preferred embodiment of the present invention, the effluent exiting the hydrogenation unit(s), (or the last in a train of reactors, where the unit comprises more than one reactor), is subjected to a separation stage, e.g. cooling the effluent and then feeding the cooled effluent to a flash vessel (wherein the said effluent can be both cooled and/or flashed several times). After this treatment, a vapour stream containing the majority

of the hydrogen and inert gases (including light alkanes; nitrogen; CO; and CO ₂ ) in the effluent is recovered for recycle to the hydrogenation unit(s). The concentration of said inert gases can be controlled by diverting a proportion of the gas recycle stream to purge.

According to the present invention, the liquid stream exiting the hydrogenation unit is then subjected to a separation stage (e.g. distillation), whereby a fraction comprising alcohol(s) (i.e. methanol and ethanol) is separated and recovered and at least a part, preferably all, of the recovered methanol is used as the feed for the etherification unit. The ethanol is recovered as the desired product.

Additionally, another separation embodiment at the liquid exit of the hydrogenation unit consists of isolating methyl ethanoate/methanol mixture and/or an ethyl ethanoate/ethanol mixture and recycling said isolated compounds to the hydrogenation unit. Alternatively, the methyl ethanoate/methanol mixture can be passed to the etherification unit. As a further embodiment to the present invention, the methyl ethanoate can be hydrogenated by co feeding with stoichiometric amount of H ₂ to the methanol unit. It should be noted that whilst all of the aforementioned temperature and pressure operating conditions form preferred embodiments of the present invention, they are not, by any means, intended to be limiting, and the present invention hereby includes any other pressure and temperature operating conditions that achieve the same effect.