Methyl formate production

Skeletal Cu-Zn catalysts show great potential as alternatives to coprecipitated Cu0-Zn0-Al203 catalysts used commercially for low temperature methanol synthesis and water gas shift (WGS) reactions. They can also be used for other reactions such as steam reforming of methanol, methyl formate production by dehydrogenation of methanol, and hydrogenolysis of alkyl formates to produce alcohols. In all these reactions zinc oxide-promoted skeletal copper catalysts have been found to have high activity and selectivity. 

The catalysts or catalysts precursors employed in these studies were anionic group 6 carbonyl complexes ( ) or group 8 metal carbonyl clusters (37— 0) where reaction conditions were 500 psi (CO2/H2) and 125 C. For the group 6 metal catalysts, the turn-over numbers obtained for the methyl formate production were ca. 15 using methanol as solvent for a 24 hour period. The anionic metal carbonyls examined as catalysts precursors included HM2(CO)io , HC02M(C0)5, and CH3C02M(C0)s as their PPN salts (PPN = bis(triphenylphosphine)-iminium and M Cr or W). The proposed reaction pathway is depicted in Scheme 3. 

Fig. 2.—Expansion of copper lattice compared with methyl formate production (dotted), both as functions of catalyst composition in the decomposition of methanol over Zu-Cu catalysts."...

The conventional electrochemical reduction of carbon dioxide tends to give formic acid as the major product, which can be obtained with a 90% current efficiency using, for example, indium, tin, or mercury cathodes. Being able to convert CO2 initially to formates or formaldehyde is in itself significant. In our direct oxidation liquid feed fuel cell, varied oxygenates such as formaldehyde, formic acid and methyl formate, dimethoxymethane, trimethoxymethane, trioxane, and dimethyl carbonate are all useful fuels. At the same time, they can also be readily reduced further to methyl alcohol by varied chemical or enzymatic processes. 

The fermentative fixing of CO2 and water to acetic acid by a species of acetobacterium has been patented acetyl coen2yme A is the primary reduction product (62). Different species of clostridia have also been used. Pseudomonads (63) have been patented for the fermentation of certain compounds and their derivatives, eg, methyl formate. These methods have been reviewed (64). The manufacture of acetic acid from CO2 and its dewatering and refining to glacial acid has been discussed (65,66). 

Anhydrous, monomeric formaldehyde is not available commercially. The pure, dry gas is relatively stable at 80—100°C but slowly polymerizes at lower temperatures. Traces of polar impurities such as acids, alkahes, and water greatly accelerate the polymerization. When Hquid formaldehyde is warmed to room temperature in a sealed ampul, it polymerizes rapidly with evolution of heat (63 kj /mol or 15.05 kcal/mol). Uncatalyzed decomposition is very slow below 300°C extrapolation of kinetic data (32) to 400°C indicates that the rate of decomposition is ca 0.44%/min at 101 kPa (1 atm). The main products ate CO and H2. Metals such as platinum (33), copper (34), and chromia and alumina (35) also catalyze the formation of methanol, methyl formate, formic acid, carbon dioxide, and methane. Trace levels of formaldehyde found in urban atmospheres are readily photo-oxidized to carbon dioxide the half-life ranges from 35—50 minutes (36). 

Between 50 and 60% of the formaldehyde is formed by the exothermic reaction (eq. 23) and the remainder by endothermic reaction (eq. 24) with the net result of a reaction exotherm. Carbon monoxide and dioxide, methyl formate, and formic acid are by-products. In addition, there are also physical losses, hquid-phase reactions, and small quantities of methanol in the product, resulting in an overall plant yield of 86—90% (based on methanol). 

Formic acid is currently produced iadustriaHy by three main processes (/) acidolysis of formate salts, which are ia turn by-products of other processes (2) as a coproduct with acetic acid ia the Hquid-phase oxidation of hydrocarbons or (3) carbonylation of methanol to methyl formate, followed either by direct hydrolysis of the ester or by the iatermediacy of formamide. 

Coproductioa of ammonium sulfate is a disadvantage of the formamide route, and it has largely been supplanted by processes based on the direct hydrolysis of methyl formate. If the methanol is recycled to the carbonylation step the stoichiometry corresponds to the production of formic acid by hydration of carbon monoxide, a reaction which is too thermodynamicaHy unfavorable to be carried out directly on an iadustrial scale. 

Other potential processes for production of formic acid that have been patented but not yet commerciali2ed include Hquid-phase oxidation (31) of methanol to methyl formate, and hydrogenation of carbon dioxide (32). The catalytic dehydrogenation of methanol to methyl formate (33) has not yet been adapted for formic acid production. 

World installed capacity for formic acid is around 330,000 t/yr. Around 60% of the production is based on methyl formate. Of the remainder, about 60% comes from Hquid-phase oxidation and 40% from formate salt-based processes. The largest single producer is BASF, which operates a 100,000 t/yr plant at Ludwigshafen in Germany. The only significant U.S. producer of formic acid is Hoechst-Celanese, which operates a butane oxidation process. 

Even though form amide was synthesized as early as 1863 by W. A. Hoffmann from ethyl formate [109-94-4] and ammonia, it only became accessible on a large scale, and thus iadustrially important, after development of high pressure production technology. In the 1990s, form amide is mainly manufactured either by direct synthesis from carbon monoxide and ammonia, or more importandy ia a two-stage process by reaction of methyl formate (from carbon monoxide and methanol) with ammonia. 

However, BASF developed a two-step process (25). After methyl formate [107-31-3] became available in satisfactory yields at high pressure and low temperatures, its conversion to formamide by reaction with ammonia gave a product of improved quaUty and yield in comparison with the earlier direct synthesis. 

Two-Step Process. The significant advantage of the two-step process is that it only requkes commercial-grade methyl formate and ammonia. Thus the cmde product leaving the reactor comprises, in addition to excess starting materials, only low boiling substances, which are easily separated off by distillation. The formamide obtained is of sufficient purity to meet all quaUty requkements without recourse to the costiy overhead distillation that is necessary after the dkect synthesis from carbon monoxide and ammonia. 

The methanol carbonylation is performed ia the presence of a basic catalyst such as sodium methoxide and the product isolated by distillation. In one continuous commercial process (6) the methyl formate and dimethylamine react at 350 kPa (3.46 atm) and from 110 to 120°C to effect a conversion of about 90%. The reaction mixture is then fed to a reactor—stripper operating at about 275 kPa (2.7 atm), where the reaction is completed and DMF and methanol are separated from the lighter by-products. The cmde material is then purified ia a separate distillation column operating at atmospheric pressure. 

Another method of preparation involving methyl formate has been reported whereia the formate reacts with ammonia and methanol ia the presence of ammonium chloride at 255°C and 2.9 MPa (28.6 atm) (15). In this case, monomethylformamide is present ia considerable quantities as a by-product. 

The reaction is mn for several hours at temperatures typically below 100°C under a pressure of carbon monoxide to minimise formamide decomposition (73). Conversions of a-hydroxyisobutyramide are near 65% with selectivities to methyl a-hydroxyisobutyrate and formamide in excess of 99%. It is this step that is responsible for the elimination of the acid sludge stream characteristic of the conventional H2SO4—ACH processes. Because methyl formate, and not methanol, is used as the methylating agent, formamide is the co-product instead of ammonium sulfate. Formamide can be dehydrated to recover HCN for recycle to ACH generation. 

Fiaal purification of propylene oxide is accompHshed by a series of conventional and extractive distillations. Impurities ia the cmde product iaclude water, methyl formate, acetone, methanol, formaldehyde, acetaldehyde, propionaldehyde, and some heavier hydrocarbons. Conventional distillation ia one or two columns separates some of the lower boiling components overhead, while taking some of the higher boilers out the bottom of the column. The reduced level of impurities are then extractively distilled ia one or more columns to provide a purified propylene oxide product. The solvent used for extractive distillation is distilled ia a conventional column to remove the impurities and then recycled (155,156). A variety of extractive solvents have been demonstrated to be effective ia purifyiag propylene oxide, as shown ia Table 4. 

The solvent is then evaporated, and the unconverted sterol is recovered by precipitation from an appropriate solvent, eg, alcohol. The recovered sterol is reused in subsequent irradiations. The solvent is then evaporated to yield vitamin D resin. The resin is a pale yeUow-to-amber oil that flows freely when hot and becomes a brittie glass when cold the activity of commercial resin is 20 30 x 10 lU/g. The resin is formulated without further purification for use in animal feeds. Vitamin D can be crystallized to give the USP product from a mixture of hydrocarbon solvent and ahphatic nitrile, eg, benzene and acetonitrile, or from methyl formate (100,101). Chemical complexation has also been used for purification. 

Ethanol can also be obtained by the reaction of methanol with synthesis gas at 185°C and under pressure (6.9—20.7 MPa or 68—204 atm) in the presence of a cobalt octacarbonyl catalyst (177). However, although ethanol was the primary product, methyl formate, methyl, propyl and butyl acetates, propyl and butyl alcohols, and methane were all present in the product. 

Among many polar aprotic solvents, including ethers, BL, PC, and ethylene carbonate (EC), methyl formate (MF) seems to be the most reactive towards lithium. It is reduced to lithium formate as a major product which precipitates on the lithium surface and passivates it [24], The presence of trace amounts of the two expected contaminants, water and methanol, in MF solutions does not affect the surface chemistry. C02 in MF causes the formation of a passive film containing both lithium formate and lithium carbonate. 

Residues of PCBs in animal tissues include not only the original congeners themselves, but also hydroxy metabolites that bind to cellular proteins, for example, transthyretin (TTR Klasson-Wehler et al. 1992 Brouwer et al. 1990 Fans et al. 1993). Small residues are also found of methyl-sulfonyl metabolites of certain PCBs (Bakke et al. 1982, 1983). These appear to originate from the formation of glutathione conjugates of primary epoxide metabolites, thus providing further evidence of the existence of epoxide intermediates. Further biotransformation, including methylation, yields methyl-sulfonyl products that are relatively nonpolar and persistent. 

The importance of solvation on reaction surfaces is evident in striking medium dependence of reaction rates, particularly for polar reactions, and in variations of product distributions as for methyl formate discussed above and of relative reactivities (18,26). Thus, in order to obtain a molecular level understanding of the influence of solvation on the energetics and courses of reactions, we have carried out statistical mechanics simulations that have yielded free energy of activation profiles (30) for several organic reactions in solution (11.18.19.31. ... 

Yeom and Frei [96] showed that irradiation at 266 nm of TS-1 loaded with CO and CH3OH gas at 173 K gave methyl formate as the main product. The photoreaction was monitored in situ by FT-IR spectroscopy and was attributed to reduction of CO at LMCT-excited framework Ti centers (see Sect. 3.2) under concurrent oxidation of methanol. Infrared product analysis based on experiments with isotopically labeled molecules revealed that carbon monoxide is incorporated into the ester as a carbonyl moiety. The authors proposed that CO is photoreduced by transient Ti + to HCO radical in the primary redox step. This finding opens up the possibility for synthetic chemistry of carbon monoxide in transition metal materials by photoactivation of framework metal centers. 

Both oxidative and non-oxidative routes with similar share are followed, yielding hydrogen or water as additional products. As by-products, carbon dioxide and carbon monoxide, methyl formate and formic acid are generated. It is advised to quench the exit stream as formaldehyde decomposition can occur. 

Sodium hydrosulfite is produced through the Formate process where sodium formate solution, sodium hydroxide, and liquid sulfur dioxide reacted in the presence of a recycled stream of methanol solvent. Other products are sodium sulfite, sodium bicarbonate, and carbon monoxide. In the reactor, sodium hydrosulfite is precipitated to form a slurry of sodium hydrosulfite in the solution of methanol, methyl formate, and other coproducts. The mixture is sent to a pressurized filter system to recover sodium hydrosulfite crystals that are dried in a steam-heated rotary drier before being packaged. Heat supply in this process is highly monitored in order not to decompose sodium hydrosulfite to sulfite. Purging is periodically carried out on the recycle stream, particularly those involving methanol, to avoid excessive buildup of impurities. Also, vaporized methanol from the drying process and liquors from the filtration process are recycled to the solvent recovery system to improve the efficiency of the plant. 

Interestingly, one-electron oxidants partly mimic the effects of OH radicals in their oxidizing reactions with the thymine moiety of nucleosides and DNA. In fact, the main reaction of OH radicals with 1 is addition at C-5 that yields reducing radicals in about 60% yield [34, 38]. The yield of OH radical addition at C-6 is 35% for thymidine (1) whereas the yield of hydrogen abstraction on the methyl group that leads to the formation of 5-methyl-(2 -de-oxyuridylyl) radical (9) is a minor process (5%). Thus, the two major differences in terms of product analysis between the oxidation of dThd by one-electron oxidants and that by the OH radical are the distribution of thymidine 5-hydroxy-6-hydroperoxide diastereomers and the overall percentage of methyl oxidation products. 

In Fig. 2.13 the mass intensities for carbon dioxide (m/e = 44) and methyl formate (m/e = 60) during a potential scan are given. While the signal for C02 follows the current pattern in the whole potential range that for HCOOCH3 does not. This indicates the existence of parallel pathways. Methyl formate was also detected as an electrolysis product in long duration experiments [66],... 

The SNIFTIRS results presented here confirm the presence of formic acid and methyl formate as by-products of methanol oxidation. Other by-products such as formaldehyde could not be detected under our experimental conditions. In fact, formaldehyde hydrolyses (99.99%) in aqueous solutions to a gemdiol H2C(OH)2, and the typical aldehyde bands are, therefore, not expected. 

Some of these techniques are perhaps self-evident it is known, for example, that methanol can be oxidised not only to C02 but also to formaldehyde, formic acid and even (reportedly) methyl formate in strongly acidic aqueous solutions. Clearly any mechanism must account for the formation of all of these products. 

---