GB2531088A - Process for making esters - Google Patents

Abstract

A process for preparing an alkyl alkanoate, comprising contacting a an alkene, carbon monoxide (CO), an alcohol, a catalyst and a co-catalyst comprising a Bronsted acid, under alkoxycarbonylation conditions to produce an alkyl alkanoate, wherein the catalyst comprises a metal-organophosphorous ligand complex comprising a ligand of the following formula: wherein X1 - X6 are independently, H, R, Ar, substituted Ar, OR, OAr, CO2R, SiR3, SO3R, SO3H, or fluoro; where R is alkyl or substituted alkyl; where Ar is aryl; and where Y1 - Y8 are independently Ar or substituted Ar, with the proviso that the Y1 - Y8 groups that are bound to the same P atom optionally may be linked with a carbon to carbon bond, CH2, NH, NR, NAr or O.

Description

Intellectual Property Office Application No. GB1503206.3 RTTVI Date:30 November 2015 The following terms are registered trade marks and should be read as such wherever they occur in this document: Gilson (page 21) Intellectual Property Office is an operating name of the Patent Office www.gov.uk/ipo

PROCESS FOR MAKING ESTERS

BACKGROUND OF THE INVENTION

This invention relates to a process for the alkoxycarbonylation of olefins to S carboxylic acid esters.

The use of transition metal complexes as catalysts for the carbonylation of olefins in the presence of water or alcohols to generate carboxylic acids and esters, respectively, is well known. See, e.g., EP 106379 and US 4,245,115. However, only afew catalytic systems afford sufficient activity, selectivity and stability to be attractive for an industrial-scale process.

US 6,348,62] teaches the use of a Group 8 metal and a bulky bidentate phosphine containing tertiary carbon phosphorus substituents, such as bi s(di-t-butylphosphino)-o-xylene (dtbpx), to provide stable homogeneous catalysts that have high reaction rates and produce small amounts of by-products. The steric bulk of the phosphine substituent, the basicity of the bidentate phosphine ligand and the rigidity of the aromatic ligand backbone are taught to be important features of the catalyst. For example, when bis(di-t-butylphosphino)propane was used under identical conditions, a significant decrease in reaction rate and catalyst stability was noted. Likewise, when the t-butyl groups of dtbpx were replaced with less spatially demanding substituents (i.e., cyclohexyl, isopropyl or phenyl), the high reaction rates, high selectivity and turnover numbers were compromised.

The need to use highly basic, sterically cumbersome phosphine substituents to achieve high activity and selectivity in alkoxycarbonylations was confirmed in additional, subsequently published reports disclosing a second class of highly active bidentate ligands.

US 7,485,739 discloses that Pd complexes bearing bulky i,2-bis(di-i-butylphosphinomethyl) ferrocene and I,2-bis(di-adamantlyphosphinomethyl) ferrocene ligands exhibit comparable or improved reaction rates and lifetimes relative to dtbpx. US 5,679,831 discloses that], I -bis(diphenylphosphino)ferrocene, commonly referred to as dppf, also generates a highly active alkoxycarbonylation catalyst. However, it is reported in Organoinetaliics 2003, 21, 3637, that the product profile observed low molecular weight oxygenates and alternating oligomeric ketones in addition to the desired methyl propionate.

In J Organomel. Chern. 2006, 691, 921, high selectivity for alkoxycarbonylation was only observed when the more sterical I y demanding bi s(di-o-tolylphosphino)ferrocene and bis(diphenylphosphino)octamethylferrocene ligands were employed.

These contributions, as well as others, led to the development of general guidelines, which are disclosed in JbJL Organornet Chern. 2006, 18, 125, for the design of carbonylation catalysts that are tailored to give selectivity for high molecular weight alternating polyketones, low molecular weight oxygenates or alkoxycarbonylation products.

In the case of alkoxycarbonylation, the state of the art emphasizes the need for bisphosphine ligands that have wide bite angles and that contain rigid backbones and sterically demanding, highly basic substituents. In addition, such ligands are known to be sensitive to the presence of oxygen or air.

It would be desirable to have an alkoxycarbonylation catalyst that would not require bulky, highly basic phosphine substituents, and that would be less sensitive to the presence of oxygen.

SUMMARY OF THE INVENTION

The invention employs such a catalyst in a process comprising contacting an alkene, CO, an alkanol, a catalyst and a co-catalyst under alkoxycarbonylation reaction conditions to produce an alIcyl allcanoate, wherein the co-catalyst comprises a Bronsted acid, and the catalyst comprises a metal-organophosphorous ligand complex catalyst comprising a ligand of the formula: x2 yi yyX3y3 -P4' R. Yl I Yx, _kX 6 A6 J X4 xs where Xi -X6 are independently, H, R, Ar, substituted Ar, OR, OAr, CO2R, SiR3, SO3R, SO3H, or fluoro; where R is alkyl or substituted alkyl; where Ar is awl; and where Y1 -Y8 are independently Ar or substituted Ar, with the proviso that the Yi -Yg groups that are bound to the same P atom optionally may be linked with a carbon to carbon bond, CH2, NH, NR, NAr or 0.

Surprisingly, the tetrakis(phosphinomethyl)biphenyl ligands of this invention can be employed in the preparation of remarkably active and selective alkoxycarbonylation catalysts that employ less bulky, less basic phosphine substituents, relative to the

BRIEF DESCRIPTION OF THE DRAWINGS

Figure is a schematic of the process configuration employed in Example 1] Figure 2 is a schematic of the process configuration employed in Example t2.

DETAILED DESCRIPTION OF THE INVENTION

The disclosed process comprises contacting an alkene, CO, an alkanol, a catalyst and a co-catalyst under allcoxyoxycarbonylation reaction conditions to produce an alkyl alkanoate, wherein the co-catalyst comprises a Bronsted acid, and the catalyst comprises a metal-organophosphorous ligand complex catalyst comprising a ligand of the formula shown hereinabove.

iO All references to the Periodic Table of the Elements and the various groups therein are to the version published in the CRC Handbook of Chemistry and Physics, 72nd Ed.

(t991-1992) CRC Press, at page I-iO.

As used herein, "a," "an," "the," "at least one," and "one or more" are used interchangeably. The terms "comprises," "includes," and variations thereof do not have a limiting meaning where these terms appear in the description and claims. Thus, for example, an aqueous composition that includes particles of "a" hydrophobic polymer can be interpreted to mean that the composition includes particles of "one or more" hydrophobic polymers.

Also herein, the recitations of numerical ranges by endpoints include all numbers subsumed in that range (e.g., ito 5 includes i, 1,5, 2, 2.75, 3, 3.80, 4, 5, etc.). For the purposes of the invention, it is to be understood, consistent with what one of ordinary skill in the art would understand, that a numerical range is intended to include and support all possible subranges that are included in that range. For example, the range from I to 100 is intended to convey from 1.01 to 100, from Ito 99.99, from 1.01 to 99.99, from 40 to 60, from Ito 55, etc. Also herein, the recitations of numerical ranges and/or numerical values, including such recitations in the claims, can be read to include the term "about." In such instances the term "about" refers to numerical ranges and/or numerical values that are substantially the same as those recited herein.

The term "complex" as used herein means a coordination compound formed by the union of one or more electronically rich molecules or atoms with one or more electronically poor molecules or atoms. For example, the organophosphorous ligands employable herein may possess four phosphorus donor atoms, each having one available or unshared pair of electrons that are each capable of forming a coordinate bond independently or possibly in concert (e.g., via chelation) with the metal. Carbon monoxide, which is also properly classified as a ligand, can also be present and coordinated to the metal. The ultimate composition of the complex catalyst may also contain an additional ligand, e.g., hydrogen or an anion satisfying the coordination sites or nuclear charge of the metal. Illustrative additional ligands include, for example, halogen (Cl, Br, I), alkyl, aryl, substituted aryl, acyl, CF3, C2 F5, CN, (Z)2P0 and ZP(O)(OH)O (wherein each Z is the same or different and is a substituted or unsubstituted hydrocarbon radical, e.g., the alkyl or awl), acetate, acetylacetonate, 504, PF4, PF5, NO2, NO3, CH3, CH2=CHCH2, CH3CHCHCH2, CÔH5CN, Cl-I3CN, NH3, pyridine, (C21-15)3N, mono-olefins, diolefins and triolefins, tetrahydroffiran, and the like, It is to be understood that the complex species are preferably free of any additional organic ligand or anion that might poison the catalyst or have an undue adverse effect on catalyst performance. It is preferred in the metal-organophosphine ligand complex catalyzed alkoxycarbonylation reactions that the active catalysts be free of sulfur directly bonded to the metal, although such may not be absolutely necessary.

As nsed herein, the term "substituted" is contemplated to include all permissible substituents of organic compounds unless otherwise indicated. In a broad aspect, the permissible substituents include acyclic and cyclic, branched and unbranched, carbocyclic and heterocyclic, aromatic and non-aromatic substituents of organic compounds.

Illustrative substituents include, for example, alkyl, alkyloxy, aryl, aryloxy, hydroxyalkyl, aminoalkyl, or fluoroalkyl, in which the number of carbons can range from to 20 or more, preferably from ito 12, as well as silyl, sulfonyl, hydroxy, fluoro and amino. The permissible substituents can be one or more and the same or different for appropriate organic compounds. This invention is not intended to be limited in any manner by the permissible substituents of organic compounds.

Unless stated to the contrary, or implicit from the context, all parts and percentages are based on weight and all test methods are current as of the filing date of this application.

For purposes of United States patent practice, the contents of any referenced patent, patent application or publication are incorporated by reference in their entirety (or its equivalent US version is so incorporated by reference) especially with respect to the disclosure of definitions (to the extent not inconsistent with any definitions specifically provided in this

disclosure) and general knowledge in the art.

Carbon monoxide may be obtained from any suitable source, including petroleum cracking and refinery operations. Inert gases such as, for example, nitrogen, and the noble gases, such as argon, may be present during the reaction.

Similarly, the alkene may be obtained from any suitable source. The alkene can be a substituted or unsubstituted olefinic unsaturated reactant containing from 2 to 40, preferably 2 to 20, carbon atoms. Such olefinic unsaturated compounds can be terminally or internally unsaturated and be of straight-chain, branched chain or cyclic structures, as well as olefin mixtures, such as obtained from the oligomerization of ethylene, propene, butene, isobutene, etc. (such as so called dimeric, trimeric or tetrameric propylene and the like, as disclosed, for example, in US 4,518,809 and US 4,528,403), The alkene preferably comprises ethylene. The molar ratio of the alkene to carbon monoxide in the gaseous phase is advantageously from 0.5:1 to 50:1, preferably at least 0.8:1 to 5:1, more preferably from 0.9: t to 1,5:1 and most preferably in the range from 0,95:1 to t,05: t. In one embodiment of the invention, this ratio is about 1:1.

Suitable alkanols include C130 alkanols, optionally substituted with one or more substituents such as cyano, carbonyl, alkoxy or aryl groups. Examples of suitable alkanols include methanol, ethanol, propanol, 2-propanol, 2-butanol, and t-butyl alcohol, Methanol and ethanol are particularly useful. Methanol is the preferred alkanol. Mixtures of alkanols can be employed.

The catalyst is formed when a ligand structure of Formula I is combined with a suitable metal source and a Bronsted acid co-catalyst. x2

Y1xaY3 Zf kZt:__L% Y6 "6 X4 x5 Formula where Xi -X are independently, H, R, Ar, substituted Ar, OR, OAr, CO2R, SiR3, SO3R, 505H, or fluoro; where R is alkyl or substituted alkyl; where Ar is awl; and where Y1 -are independently Ar or substituted Ar, with the proviso that the Y1 -Y8 groups that are bound to the same P atom optionally may be linked with a carbon to carbon bond, C112, NH, NR, NAr or 0. Preferred alkyl groups have from I to 6 carbon atoms. Preferably, each X is independently H, OR, Ar or alkyl, more preferably H or alkyl, and most preferably H. Preferably, each Y is independently Ar, more preferably phenyl or alkyl-substituted phenyl.

Even more preferred Y groups include phenyl, para-tolyl, and 3,5-dimethyl phenyl. The catalyst and ligand are known to those skilled in the art and can be prepared according to known methods. See, for example, US 7,531,698. Mixtures of ligands may be employed in the metal-organophosphorous ligand complex catalyst and/or free ligand.

The metal source provides the catalytic metal to the catalyst. The catalytic metal can include Group 8, 9 and 10 metals selected from rhodium (Rh), cobalt (Co), iridium (Ir), ruthenium (Ru), iron (Fe), nickel (Ni), palladium (Pd), platinum (Pt), osmium (Os) and mixtures thereof, preferably palladium or rhodium, and most preferably palladium.

Examples of effective metal sources, using Pd as an exemplary metal, include Pd(dba)2 (where dba = dibenzylideneacetone), Pd(acac)2 (where acac = acetylacetonato), and bis-(acetonitrile) palladium dichloride. Additional suitable compounds of palladium include salts of palladium with, or compounds comprising weakly coordinated anions derived from the following: nitric acid; sulfuric acid; lower alkanoic (up to C12) acids such as acetic acid and propionic acid including halogenated carboxylic acids such as trifluoroacetic acid and trichloroacetic acid; sulfonic acids such as methanesulfonic acid, chlorosulfonic acid, fluorosulfonic acid, trifluoromethanesulfonic acid, benzenesulfonic acid, naphthalenesulfonic acid, toluenesulfonic acids, e.g., p-toluenesulfonic acid, t-butylsulfonic acid, and 2-hydroxypropanesulfonic acid; sulfonated ion exchange resins; perhalic acids such as perchloric acid; halogenated carboxylic acids such as trichloroacetic acid and trifluoroacetic acid; orthophosphoric acid; phosphonic acids such as benzenephosphonic acid; and acids derived from interactions between Lewis acids and Bronsted acids, Other sources that may provide suitable anions include the optionally halogenated tetraphenylborate derivatives, e.g., perfluorotetraphenyl borate, Additionally, zero-valent palladium complexes, particularly those with labile ligands, e.g., alkenes such as dibenzylideneacetone or styrene, may be used.

In general, the catalyst may be preformed or formed in situ and comprises a metal in complex combination with one or more organophosphorous ligarids as well as other ligands that may be present in the reaction mixture. The other ligands may include organic or inorganic anions added to the reaction mixture along with the metal source, electrically neutral ligands added to the reaction mixture along with the metal source, organic anions or neutral species produced in the reaction, the substrate olefin, the reactant alkariol, carbon monoxide and hydride. The ligand complex species may be present in mononuclear, dinuclear and/or higher nuclearity forms. The exact structure of the catalyst present during the reaction depends on the reaction conditions as well as the identity of the reactants and generally is not known.

The catalyst is preferably present as part of the liquid phase, which may be formed by one or more of the reactants and/or by the use of a suitable, optional, solvent.

A Bronsted acid is employed as a co-catalyst. Bronsted acids are well known to those skilled in the art. In one embodiment of the invention, Bronsted acids of varying strength, ranging from acetic acid to trifluoromethanesulfonic acid, can be employed to produce a highly efficient catalyst system. Examples of suitable Bronsted acids include: sulfuric acid; phosphoric acid; nitric acid; phosphoric acid esters such as binapthyl hydrogenphosphate and BIMOL hydrogenphosphate; carboxylic acids such trifluoroacetic acid, acetic acid and propionic acid; sulfonic acids such as trifluoromethanesulfonic acid, methanesulfonic acid, benzenesulfonic acid, toluene sulfonic acid and propanesulfonic acid; and sulfonimide acids such as bis(trifluoroniethane)sulfonimide and 4,4,5,5,6,6-hexafluoro- 1,3,2-dithiazinarie 1,i,3,3-tetraoxide. Mixtures of Bronsted acids may be employed. Acids possessing weakly coordinating or non-coordinating counter anions are preferred. The preferred acid and the preferred proton to metal catalyst ratio vary according to the structure of the ligand employed, metal source used and the identity of the nucleophile, The acid advantageously has a pKa measured in aqueous solution of less than 4, more preferably less than 1. The Bronsted acid is employed in an amount suitable to facilitate the reaction.

Advantageously, the molar ratio of Bronsted acid to catalytic metal may be from:1 to 1000:1, preferably from 2:1 to 500:1 and more preferably from 3:1 to 100:1. Preferably, the catalyst system is capable of achieving > 99 % selectivity for methyl propionate at rates of greater than 1 mol/liter/hr, A solvent is optionally employed. Suitable solvents that may be used in conjunction with the catalyst include one or more aprotic solvents, such as ethers, e.g., diethyl ether, dimethyl ether of diethylene glycol, anisole and diphenyl ether; aromatic compounds, e.g., benzene, toluene, ethyl benzene, o-xylene, m-xylene, p-xylene; alkanes, including halo variants of such compounds, e.g., hexane, heptane, 2,2,3 -trimethylpentane, methylene chloride and carbon tetrachloride; nitriles, eg,, benzonitrile and acetonitrile; esters, e.g., methyl benzoate, methyl acetate, methyl propionate and dimethyl phthalate; sulfones, e.g., diethyl sulfone and tetrahydrothiophene 1,1-dioxide; and carboxylic acids, e.g., propionic acid, Particularly suitable solvents are the reactants and products of the reaction. For example, in the methoxycarbonylation of ethylene with carbon monoxide in the presence of methanol to form methyl propionate, the preferred solvents are methyl propionate and methanol.

In one embodiment of the invention, the catalyst is employed in a process comprising: a) supplying a first feed stream comprising carbon monoxide and ethylene in the gas phase to a reactor; b) supplying a second feed stream comprising methanol to the reactor; c) reacting together the reactants of the first and second feed streams in the reactor in the presence of a alkoxycarbonylation catalyst, which catalyzes the reaction between methanol, carbon monoxide and ethylene, to form a product comprising methyl propionate; d) removing a product-containing stream from the reactor; e) separating a product-containing stream from a catalyst-containing stream by a separation process such as distillation; and f) optionally recycling the catalyst-containing stream to the reactor. In another embodiment of the invention, the catalyst is used in a process comprising: a) supplying a first feed stream comprising carbon monoxide and ethylene in the gas phase to a reactor; b) supplying a second feed stream comprising methanol to the reactor; c) reacting together the reactants of the first and second feed streams in the reactor in the presence of a alkoxycarbonylation catalyst, which catalyzes the reaction between methanol, carbon monoxide and ethylene, to form a product comprising methyl propionate; d) removing a product-containing stream from the reactor by distillation leaving a catalyst-containing liquid in the reactor.

The reaction conditions of the alkoxycarbonylation process may include any suitable alkoxycarbonylation conditions for producing esters. For instance, the total gas pressure of carbon monoxide and the alkene starting compound of the alkoxycarbonylation process may range from Ito 69,000 kPa (0,14 to 10,000 psig). In general, however, it is preferred that the process be operated at a total gas pressure of carbon monoxide and alkene starting compound of from 276 to 6,900 kPa (40 to 1,000 psig) and more preferably from 552 to 2,068 kPa (80 to 300 psig), The minimum total pressure is limited predominantly by the amount of reactants necessary to obtain a desired rate of reaction, More specifically, the carbon monoxide partial pressure of the allcoxycarbonylation process is preferably from I to 14,000 kPa, and more preferably from 21 to 5,000 kPa. In general, the molar ratio of gaseous ethylene:CO may range from 1:10 to 100:1 or higher, preferably from 1:10 to 10:1.

In general, the alkoxycarbonylation process may be conducted at any operable reaction temperature. Advantageously, the process is conducted at a reaction temperature from 0°C to 200°C, preferably from 50°C to 120°C, and more preferably from 70°C to 100°C.

The alkoxycarbonylation process may be carried out using one or more suitable reactors such as, for example, a continuous stirred tank reactor (CSTR). The reaction zone employed may be a single vessel or may comprise two or more discrete vessels. The separation zone employed may be a single vessel or may comprise two or more discrete vessels. The reaction zone(s) and separation zone(s) employed herein may exist in the same vessel or in different vessels. For example, reactive separation techniques such as reactive distillation, reactive membrane separation, and the like, may occur in the reaction zone(s).

The reaction is preferably operated as a continuous process, however batch operation is possible.

The reaction can be conducted in a single reaction zone or in a plurality of reaction zones, and in series or in parallel. The reaction may be implemented by the incremental addition of one of the starting materials to the others, The reaction can be implemented by the joint addition of all starting materials. When compl etc conversion is not desired or not obtainable, the starting materials can be separated from the product, for example by distillation, and the starting materials then recycled back into the reaction zone.

The alkoxycarbonylation process may be conducted in glass lined, stainless steel, HastalloyTM or similar type reaction equipment. The reaction zone may be fitted with one or more internal and/or external heat exchanger(s) in order to control undue temperature fluctuations, or to prevent any possible "runaway" reaction temperatures, The alkoxycarbonylation process of this invention may be conducted in one or more steps or stages. The exact number of reaction steps or stages will be governed by the best compromise between capital costs and achieving high catalyst selectivity, activity, lifetime and ease of operability, as well as the intrinsic reactivity of the starting materials in question and the stability of the starting materials and the desired reaction product to the reaction conditions.

The alkyl alkanoate, e,g,, methyl propionate, product may be obtained from the product stream by a separation method, such as distillation, via various recovery schemes known to those skilled in the art; see, e.g., US 6,476,255.

SPECIFIC EMBODIMENTS OF THE INVENTION

The following examples are given to illustrate the invention and should not be construed as limiting its scope.

Example t. Methoxycarbonylation using 2,2',6,6'-tetrakisdiphenylphosphinomethyl)- 1,1 -biphenyl (hereinafter Ligand fl, palladium dibenzylideneacetone (hereinafter Pd(dba)2) and benzenesulfonic acid.

Pd((dba)2) (0.0466 g, 0.08 10 mmol) and Ligand 1 (0.039t g, 0.04 13 mmol) are added to a mixture of methanol (20 niL, 0.49 mol) and p-dioxane (17 mL, 0.20 mol) in a mL volumetric flask to prepare a solution. The solution is diluted to the mark using methyl butyrate. A stirbar is added to the resulting solution, which is stirred for 2 days to form a precatalyst stock solution. An aliquot of the precatalyst stock solution (20 mL) is transferred to a 20 mL scintillation vial, Benzenesulfonic acid (IBSA) (0.0752 g, 0.48 mmol) is added to the vial to form a catalyst stock solution, and the catalyst stock solution is allowed to stir for 20 minutes. An aliquot of the catalyst stock solution (5 mL) is added to a parallel pressure reactor glass liner and loaded into a 20 niL reactor, The reactor is charged to 65 psig with 1:1 C21-L:CO, heated to 100°C and pressurized to 120 psig with t:1 C2H4:CO. After two hours, the reactor is cooled to 40°C and vented. By gas chromatographic GC) analysis, 0. t93 g of unreacted methanol is recovered, and 1.60 g methyl propionate (73.5 % yield) and 0.011 g propionic acid (0.58 % yield) are found. The turnover frequency (TOF) in moles methyl propionate per moles Pd per hour is 2.24 x iO' Y For the purposes of the invention, the term "yield" means the moles of product produced divided by the moles of methanol charged to the reactor.

Example 2. Methoxycarbonylation using Ligand I and Pd(dba)2 with different Bronsted acids.

The procedure of Example 1 is repeated, except that 0.47 mmol of the desired Bronsted acid is added to the vial in place of BSA. By GC analysis, the product mixture consists of methanol, methyl propionate, propionic acid and 3-pentanone. Table 1 summarizes the results for a variety of Bronsted acids, For the purposes of the invention, the term "selectivity" means the moles of a given product or by-product produced divided by the total moles of products and by-products produced.

Rate constants are obtained by fitting the gas uptake data in Microsoft Excel using the assumptions and rationale that follows. The kinetics are expected to be first order in active catalyst concentration. Under a constant gas headspace pressure, the kinetics can appear to be zeroth order with respect to methanol as observed from a constant gas uptake rate as the methanol is depleted. A slight deviation from zeroth order is seen in some experiments. This deviation is corrected for by fitting a first order decay, with rate constant kd, of the zeroth order rate constant, k. The overall reaction rate is expressed as: rate = lc*(etit) The data presented here include only the zeroth other rate constant, k, because the decay rate is very small and can be neglected over the time frame of the experiments. The ethylene partial pressure and the CO partial pressure are included in the observed /c, so this rate constant is valid only for the conditions of the experiment. Changes in liquid volume are neglected for the purposes of calculating rate constants.

For the purposes of the invention, the capitalized term "Rate Constant" means the rate constant calculated as described above as measured under the conditions of Example t.

In various embodiments of the invention, the Rate Constant is at least 350, at least 800, or at least 1100. -1]--a -a -t

Table 1: Summary of results for various Bronsted acids.

Total % % g 3-g g Methyl g Propionic Propionate Propionatc Acid Type piCa Pentanone Methanol Propionate Acid Yield Selectivity k, (mollhr/mol Pd) Hexafluoropropanedisulfonimide -11.6 0.0097 0,56 0.11 0.0030 6.1% 92% 1.0 x I 3 Bis(trifluoromethane)sulfonimide -10.4 0.010 0.55 0.14 0.0061 7.9% 93% 1.4 x Trifluoromethanesulfonic acid -5.5 0.013 0.59 0.45 0.0088 21% 97% 2.5 x i0 p-Nitrobenzcnesulfonic acid -3.8 0.0000 0.61 0.048 0.0000 2.6% 100% 2.6 x Benzenesulfonic acid -2.8 0.0000 0.24 1.07 0.0077 57% 100% 1.8 x i0 Methanesulfonic acid -I.9 0.0000 0.56 0.59 0,0050 28% 100% 7.9 x I 02 Propanesulfonic acid -1.5 0.0000 0.49 0.34 0.0032 18% 100% 4.0 x 102 Trifluoroacetic Acid -0,25 0.0000 0.74 0.064 0.0000 3% 100% 32 Binaphthyl hydrogenphosphate 1,1 0.0000 0.59 0.071 0.0000 3.8% 100% 29 Example 3. Effect of ligand equivalents on the methoxycarbonylation using Ligand 1, Pd(dba)2 and BSA, The procedure of Example I is repeated except that the amount of Ligand I used in the precatalyst stock solution is varied from 0.25 equivalents to 2.00 equivalents with respect to Pd. The results are shown in Table 2 below. The results show that the yield of methyl propionate increases as the ligand:Pd ratio increases. Ui

Table 2: Summary of results obtained using various ligand:Pd (L:Pd) ratios. 0/

mol Pd mol L L/Pd g g g Methyl g Total % Pentanonc Methanol Propionat2 Propionic Propionatc Propionate (mollhr/ni ol Acid Yield Selectivity Pd) 4,06E-06 I.OIE-06 0,25 0.0000 0.40 0.60 0.0038 32.0% 100.0% 7,7x 102 4.07E-06 2.06E-06 0.51 0.0000 0.24 1.07 0.0077 57.0% 100.0% 1.8 x io 4.03E-06 4.03E-06 1.00 0.0051 0.028 1.72 0.010 92.1% 99.7% 3.5 x I 3 4.02E-06 8.06E-06 2.01 0.0050 0.029 1.73 0.011 92.4% 99.7% 3.9 x I0 Example 4. Effect of ligand substituent on the rate of methoxycarbonylation.

The procedure of Example 1 is used, except as noted. A variety 2,2',6,6'-tetrakis((diphenylphosphino)methyl)-],]'-biphenyl ligands(Ligands I through 9, shown in Table 3) possessing phenyl substituents of varying steric and electronic characteristics are explored. In Table 3, in terms of Formula, each Xis H and the column labeled "Y" indicates that each Y substituent of the ligand has the stmcture shown. All ligands are used in a 0.5 mol ratio with respect to Pd(dba)2. Four acid co-catalysts (30 equivalents ofBSA, trifluoromethanesulfonic acid (l-lOTf), methanesulfonic acid (MSA) and trifluoroacetic acid (TFA)) are investigated independently with each ligand/Pd combination, and the optimal acid varies with the identity of the ligand. A summary of results is displayed in Table 3, showing the results only for the best co-catalyst for a given ligand.

Table 3: Effect of ligand substituent on selectivity and yield. % %

Acid Propionate Propionate Zero order rate constant k, Y Ligand Type Yield Selectivity mol/hr/mol Pd 1 BSA 67.2% 100% 2.6 x io 2 BSA 66.5% 99.3% 3.8 x i03 OH, i) 3 BSA 50.1% 100% 1.7 x io3 4 HOTf 36.1% 100% 1.1 x io HOTI 23.7% 100% 6.2 x 102 6 BSA 10.3% 100% 2.8 x 102 7 HOTf 4.1% 100% 79 OF, 8 MSA 2.4% 100% 1.1 x 102 9 HOTf 2.0% n.m. 64 OH3 n.m. = not measured -Is-As can be seen from Table 3, the selectivity to the desired product is very high for Ligarids 1-8, but the yield of product decreases with bulkier, more electron donating and more electron withdrawing ligand substitution, Ligands 1, 2 and 3 are the most active. The rate constants are also highest for Ligands 1 and 2.

Example 5: Effect of initial methanol concentration on reaction rate constant when Pd(dba)2 is the metal source.

A precatalyst stock solution is prepared as in Example I, adding the requisite amount of methanol to the scintillation vials. The results are shown in Table 4 below.

Table 4: Effect of initial methanol equivalents on reaction rate using Ligand L % Zero order rate Volume % Propionate constant k, Methanol Yield (mol/hr/mol Pd) 22% l.0x103 24% 6.8x102 27% 7.5x102 28% 7.5x102 Example 6: Effect of methanol and water on reaction rate when (Cl-I3CN)2PdCI2 is the metal source, The procedure of Example 5 is used. except the palladium source is (CH3CN)2PdC12, The results are shown in Table 5.

Table 5: Effect of water and methanol on reaction rate constant when (CH3CN)7PdC17 is the metal source, Volume, Volume Zero order rate constant k, % MeOH Acid % Water (mol/hr/mol Pd) BSA 0 3.2 x 102 BSA 2 1.5 x BSA 0 1.9x 10 BSA 2 2.9x 10 Example 7: Effect of ethylene and CO partial pressures on reaction rate.

The procedure of Example 1 is repeated, except that instead of using 1:1 C2H4:CO, the reactors are purged with C21-L:CO (5:1) or C21-L:CO (1:5) for approximately 30 seconds.

The reactors are charged to 100 psig C2H1:CO (5:1) or C21-L:CO (1:5), sealed, and then heated to 98°C. Once pressurized, reaction data collection is initiated (approximately 2 minutes after initial pressurization). After three hours, the reactor contents are cooled to 40°C and vented. The glass liners are removed from the reactors and samples are collected for INMR and OC analysis. When the partial pressure of ethylene is in excess of the partial pressure of CO, the reaction is finished in less than 2 minutes. When the partial pressure of CO is in excess of ethylene, the reaction finishes more slowly, i.e., in about one hour.

Example 8: Methoxycarbonylation using lOOmL Pan reactor as a semibatch reactor under constant pressure and head space composition with gas supplied to the reactor head space from a separate feed cylinder.

To a 1 00 ml nitrogen-purged Parr reactor equipped with heated mantel, magnetic stirrer, dip tube, gas feed cylinder, pressure regulator, and liquid charge cylinder is added a degassed solution of about 23 grams methyl propionate, and I gram of methanol, A 1:1 mixture of ethylene and CO in a gas uptake cylinder is prepared. The reactor solution is purged with the gas mix 3 times to 300 psig and vented. After the final purge, the pressure is brought to about 60 psig in the reactor using the mixed gas, then the contents of the Parr reactor are heated to 100°C. In a glove box, an air free 200 ppm palladium catalyst charge solution is prepared by adding 30mg Pd(AcAc)2, 46 mg Ligand 1 (0.5 ligand to Pd molar ratio), and 9 g methanol to a vial and then stirring for about 2 hours. Then, about 0.56gm toluenesuli'onic acid (acid to Pd molar ratio -3O) is added and the solution is stirred for 30 minutes. 5.46 grams of the catalyst solution is transferred air-free to the liquid charge cylinder on the Parr reactor. With the reactor contents at OO°C, the catalyst charge is transferred to the reactor using 150 psig mix gas pressure. The pressure in the reactor is equilibrated to 150 psig, then the gas supply is switched from the catalyst charge cylinder to the dip tube. The reaction progress is monitored by tracking the amount of CO and ethylene consumed from the mixed gas cylinder using a data logger. The methanol concentration in the reactor at various times is given in Table 6 below.

TaNe 6: Concentration of methanol over time.

Elapsed [MeOH] Time (hr) (v1oles/L) o s,ii 0.2 4.29 0.4 3,34 0.6 2.46 0.8 164 1.2 0.16 IA 0 1,6 0 Example 9: Batch methoxycarbonylation with different levels ofp-toluene sulfonic acid co-catalyst.

Example 8 is repeated except that different amounts ofp-toluene sulfonic acid are used. The rates of product formation at the difference acid levels are given in Table 7.

TaNe 7: Rates of product formation at different acid levels.

Acid:Pd ppm Pd Ligand:Pd Ethylenc:CO Pressure Temp Initial rate Mol ratio Mol ratio Mol ratio (psig) °C MolJhr/mol Pd 198 0.5 1:1 150 100 1.7 x io3 189 0.5 1:1 150 100 2.7 x io3 200 0.5 1:1 150 100 6.1 x io Example 10: Methoxycarbonylation using lOOmL Parr reactor as a batch reactor under constant volume, A reaction solution containing the desired amounts of methyl propionate and methanol is syringed into a clean, dry, nitrogen-inerted 100 ml stainless steel Parr reactor, If the reaction is run with an excess of ethylene, the reaction solution is purged with ethylene three times and the reactor pressure is brought to the desired pressure. If the reaction is run with an excess of carbon monoxide, the reaction solution is purged with carbon monoxide. If the reaction is run with a 1:1 ratio of CO:ethylene, the reaction solution is purged with a 1:1 ratio of CO:ethylene. While purging, the catalyst solution containing the ligand (2,2',ô,ô'-tetrakis((diphenylphosphino)methyl)-l.1 -biphenyl 0), palladium source (Pd(AcAc)2 or Pd(dba)2, and BSA is added to a purged catalyst addition reservoir. The catalyst reservoir is pressured with the remaining gas to a higher pressure than the reactor pressure, determined by the desired reactor composition. When all reaction components are added, the reactor is heated and agitated, and the reactor is brought to operating temperature. When the desired temperature is reached, the catalyst solution is added from the reservoir, bringing the total reactor pressure to the final reactor pressure.

The reactor headspace, containing the desired ratio of ethylene to CO. is blocked in. The data recorder is activated, and reaction proceeds, leaving the headspace pressurized with the component charged in excess. The rates of gas uptake from the reactions at various conditions are given in Table 8.

TaNe 8: Rates of gas uptake from reactions at various initial conditions.

Acid:Pd ppm Pd Ligand:Pd Ethylene CO C2:CO Temp Initial Mol ratio Mol ratio Pros. Pros. °C rate (psig) (psig) psig/min 100.5 0,5 210 210 1,0 100 4.62 54.8 0.5 550 ISO 3.67 100 9.18 96.3 0.5 250 450 0.56 100 2.54 Example 11: Continuous methoxycarbonylation using process in which reaction and catalyst separation take place in separate vessels.

The reaction is run continuously in a CSTR (II) configured as shown in Figure 1, with ethylene feed (12), CO feed (13) and methanol feed (14) . With the reactor in this configuration, methanol is fed to the bottom of the reactor at the desired rate. Ethylene and carbon monoxide enter the reactor through a sparger at a rate great enough to assure that the head space vent rate is sufficient to allow effective pressure control, The vent stream (15) passes through a condenser (16) to condense any product that comes overhead with the vapor. Catalyst, dissolved in methanol and methyl propionate, is fed into the reactor via the catalyst recycle stream (19). Ethylene, carbon monoxide and methanol react in the reactor to give methyl propionate, As material is added to the reactor, liquid in the reactor overflows through a stand pipe (20) and goes to the vaporizer (21), where it separates into a product stream (22) and a catalyst-containing stream (19). The product distills out of the vaporizer, and the catalyst returns to the reactor via a liquid catalyst recycle stream (19).

The desired amount of methanol, Pd(AcAc)2, and phosphine ligand are combined and allowed to dissolve over 2 hours and form a solution. The desired amount of benzenesulfonic acid or methanesulfonic acid is then added to the solution. The resulting solution is combined with methyl propionate and the mixture is transferred to the reactor air free. 50 psig of 1:1 ethylene CO pressure is applied to the reactor and the reactor contents are heated to 100°C. The methanol feed is set at 100 to 150 g/hr and the ethylene and carbon monoxide feeds are set at the desire rates. As material fills the vaporizer, recycling of the vaporizer tails to the reactor begins.

The reaction conditions are given in the Table 9.

Table 9: Reaction conditions.

MeOH feed rate 75 g/hr Ethylene feed rate 85 SLPH CO feed rate Si SLPH Reactor temperature 100°C Reactor pressure 150 psig Pd concentration 200 ppm Lig:Pd 0,55 BSA:Pd 20 SLPH = standard liters per hour.

Samples are taken periodically and analyzed by GC. The normalized concentrations of methanol and methyl propionate in the samples are shown in Table 10 below.

Table 10: The normalized concentrations of methanol and methyl propionate in the samples.

Elapsed time (hrs) [MeOH1 IMeP1 0.00 9.27% 90.73% 8,00 23.46% 76.54% 17,00 35.50% 64.50% 22.00 40.14% 5. 86% 25.00 44.99% 55.01% 41.00 65.12% 34.88% 48.00 71.16% 28.84% 72.00 88.00% 12.00% 113,00 96,75% 3.25% 120.00 9910% 0.90% Example 12: Continuous methoxycarbonylation using process in which reaction and catalyst separation take place in a single vessels.

For this example, the reaction is run continuously in a CSTR (I) configured as shown in Figure 2, with ethylene feed (2), CO feed (3) aid methanol feed (4). In this configuration, the product distills overhead from the reactor in the vapor phase. Ethylene and carbon monoxide are fed to the CSTR through a sparger at an operator-determined, desired feed rate, Methanol is fed to the reactor at a rate controlled by the control system to maintain the operator-detennined, desired reactor level. A differential pressure cell (not shown) determines the reactor level and the methanol feed rate is adjusted accordingly in order to maintain a constant liquid level in the reactor. The reactor head space contains methanol and methyl propionate at the vapor pressure of the materials at the reaction temperature. Unreacted ethylene and carbon monoxide exit the reactor through the head space via the reactor vent (5), carrying the methanol and methyl propionate out of the reactor. The vapor passes through a high efficiency condenser (6) that condenses the methanol and methyl propionate, allowing the bulk of the uncondensed excess ethylene and carbon monoxide to vent via line (7) to a hood. The condensed liquid stream (8), comprising product methyl propionate, unreacted methanol and a small amount of dissolved ethylene and carbon monoxide, go to the product can or drum, A solution of Pd(AcAc)2, ligand and methanol (degassed) and/or methyl propionate is stirred in a glove box for 2 hours. At that time, the co-catalyst, benzenesulfonic acid, is added to the solution md the solution is stirred for 30 minutes, The reactor is filled halfway with a 10-50 wt% solution of methanol in methyl propionate. The reactor is heated to the desired reaction temperature. The pre-made catalyst charge solution is staged air free into a pressure cylinder to feed the reactor through a Gilson 5SC piston. CO and ethylene gas feeds are started at about 50 SLPH each. The reactor pressure increases to the desired reaction pressure (300 psig). The catalyst charge is fed to the reactor at about 1.7-4 ml/min.

As the reaction begins to take up gas, the gas feeds are increased to maintain a blow off gas rate of about 100-400 SLPH. Once the desired amount of catalyst is added, the catalyst addition is stopped and the methanol feed to the reactor is started. The methanol feed is kept nnder manual control at a feed rate of at least 20g/hr methanol until the reaction stabilizes, The level is controlled where the level stabilizes, The material from the overhead collection tank is collected in 5-gallon plastic containers, Under the reaction conditions, which are shown in Table II, the methyl propionate:methanol wt, ratio in the product after three hours is 80:20. -2]-

JaNe 11: Reaction conditions for Example 12.

MeOH feed rate 75 glhr Ethylene feed rate 130 SLPI-I CO feed rate 130 SLPH Reactor temperature 100°C Reactor pressure 300 psig Pd concentration 100 ppm Lig:Pd 0.55 BSA:Pd 30 Example 13: Continuous methoxycarbonylation using process in which reaction and catalyst separation take place in a single vessel with ethylene:CO ratio greater than 1.0.

S Example 12 is repeated except that the conditions are as shown in Table 12. After three hours, the methyl propionate production rate is 2, 1 mol/liter/hr.

JaNe 12: Reaction conditions for Example 12.

MeOH feed rate 75 g/hr Ethylene feed rate 300 SLPH CO feed rate 150 SLPI-I Reactor temperature 100°C Reactor pressure 300 psig Pd concentration 100 ppm Lig:Pd (molar) 0,55 BSA:Pd (molar) 30 Example 14: Continuous methoxycarbonylation using process in which reaction and catalyst separation take place in a single vessel with ethylene:CO ratio less than 1.0.

Example 13 is repeated except that the conditions are as shown in Table 13. After three hours, the methyl propionate production rate is 2.1 mol/liter/hr.

Table 13: Reaction conditions for example 13.

MeOH feed rate 75 glhr Ethylene feed rate 150 SLPI-I CO feed rate 275 SLPH Reactor temperature 100°C Reactor pressure 300 psig Pd concentration 200 ppm Lig:Pd (molar) 0.55 BSA:Pd (molar) 30