FI113016B - High productivity shell impregnated catalyst for vinyl! acetate prodn. - Google Patents

Abstract

Catalyst for vinyl acetate prodn. from ethylene, acetic acid and O2-contg. gas has a productivity of above 661g/hr. per litre of catalyst at 150 deg.C and consists of (i) a support with particle dia. 3-7mm and pore vol. 0.2-1.5 ml/g; (ii) Pd amd Au distributed in the outermost 1.0mm layer of the support particles, with the Au:Pd wt. ratio being 0.60-1.25, rather than the value of below 0.5 as per US4048096 or the apparently non-critical value of ca. 0.05-1.5 as per US4087622; and (iii) 3.5-9.5 wt% of K acetate.

Description

11301 /

Catalysts and processes for the preparation of vinyl acetate

The invention relates to improved palladium / gold catalysts useful for enhancing the production of vinyl 5 acetate from ethylene, acetic acid and oxygen-containing gas.

The production of vinyl acetate by the reaction of ethylene, acetic acid and oxygen in a single gasase in the presence of a catalyst containing palladium, gold and an alkali metal acetate promoter is a known method. The catalyst components are usually supported by a porous support material, e.g., silica or alumina.

In early examples of these catalysts, both palladium and gold were more or less evenly distributed throughout the support (see, for example, US 3,725,680, US 3,743,607 and GB 1,333,449). Later, this was considered to be detrimental because it was found that the material in the inner region of the support did not promote the reaction because the reagents did not diffuse significantly into the support prior to the reaction. In other words, a significant amount of palladium and gold is not legal; '·· Never come into contact with reagents.

:, · \ Ί To overcome this problem, other inventors were invented; processes for the preparation of a catalyst for the purpose of such catalyst; Preparation of 25 tions in which the active components are concentrated on the outer shell of the support (shell impregnated catalysts). For example, GB 1,500,167 discloses catalysts in which at least 90% of palladium ... and gold is distributed in a portion of the support having a radius of less than 30% from the surface when it is · · · · · 1 GB 1 283 737 teaches that the degree of penetration into a porous carrier can be controlled by pretreating the porous carrier with an alkali solution, for example, sodium carbonate or sodium hydroxide.

• · · • · 1 · • »• · 2 113016

Another approach which has been found to produce particularly active catalysts is disclosed in US 4,048,096. In this patent, shell impregnated catalysts are prepared by a process comprising the steps of: (1) impregnating a carrier with an aqueous solution of water-soluble palladium and gold compounds; with a total volume of 95 to 100% of the absorption capacity of the catalyst support, 10 (2) of the water-insoluble palladium and gold compounds are precipitated on the support by washing the impregnated support with alkali metal silicate solution, the alkali metal silicate being hours, the pH of the solution is 6.5 to 9.5; (3) converting the water-soluble palladium and gold compounds to palladium and gold metals by treatment with a reducing agent; (4) washed with water; (5) contacting the catalyst with alkali metal acetate; and (6) drying the catalyst. By this method it is claimed that it is possible to prepare catalysts with specific activity : a viscosity of at least 83 g vinyl acetate of precious metal ·; · 25 grams per hour, measured at 150 ° C.

It is taught in column 5 of U.S. Patent 4,048,096 that the above process is advantageously used to prepare catalysts containing 1.65-3.3 g palladium and 0.75-1. , 5 g of gold per liter of finished »· M 30 catalyst In a typical silica carrier having a density of approximately 600 g / l, these ranges, expressed as a percentage by weight, range from 0.25 to 0.5% and from 0.12 to 0.22 % and thus *: ··: Au: Pd% by weight ratio of 0.16 to 0.75 The effective amount of alkali metal acetate * was taught to be 5 to 60 grams per liter, preferably 25 to 35 grams per liter, which corresponds to 0.75 to 9.2% by weight, preferably 3.8 to 5.4%, in the case of potassium acetate.

Six examples of the invention are disclosed in U.S. Patent 4,048,096. They all teach the use of catalysts having a gold: palladium to weight ratio of 0.42 to 0.45 and a gold content of 2.1 grams per liter or less. The most active catalyst is shown in Example 3 with a gold content of 2.1 grams per liter, a gold: palladium weight ratio of 0.42 and a productivity of 610 g vinyl acetate per hour catalyst at 150 ° C. Computer predictions of catalyst activity based on the results we obtained in our studies showed higher activity, 661 g.

It has now been found that shell impregnated catalysts having a productivity of more than 661 g vinyl acetate per hour per liter of catalyst at 150 ° C can be obtained by ensuring that the gold: palladium weight ratio is in the range of 0.60 to 1.25. In addition, it has been found that certain metal batches can further improve productivity by making sure that the catalyst has a potassium acetate content in the range of 3.5 to 9.5% by weight.

i '· This finding is quite unexpected and contrary to what is taught in U.S. Patent 4,048,096, which proposes that gold palladium below 0.5 should be used; ·· 25 weight ratios.

; * Shell-impregnated catalysts are also shown. · ·. in U.S. Pat. No. 4,087,622. Column 2, lines 47-50, teaches that the percentage of gold in palladium and ... the total weight of gold is preferably from about 5% to about 60% by weight. These numbers correspond to a wide range of gold: palladium weights in the range of 0.05 to 1.5, indicating that no attention was paid to the criticality of this variable. Further, the alkali metal acetate range is taught to have a wide concentration range (1 to 30% by weight), and the examples show only 'II! 35 catalysts with a potassium acetate content of 3%.

* · • # * 113016 4

Examples 4-1 to 4-7 show the effect of varying the gold content of the catalyst on a given palladium charge, but nothing suggests gold-palladium-ratio criticality over a wide palladium charge range.

The present invention provides a shell impregnated catalyst for use in the preparation of vinyl acetate from ethylene, acetic acid and oxygen gas having a catalyst productivity greater than 661 g vinyl acetate per hour per liter of catalyst at 150 ° C consisting essentially of: (1) a catalyst carrier; having a particle diameter of about 3 to 7 mm and a pore volume of 0.2 to 1.5 ml per gram (2) of palladium and gold distributed on the outermost 1.0 mm thick layer of catalyst support particles, 15 and (3) of about 3.5 to 9.5 wt. % potassium acetate with a gold: palladium weight ratio in the catalyst ranging from 0.60 to 1.25 and a palladium content greater than 3.9 grams per liter of catalyst.

The catalyst according to the invention has the additional advantage that they are highly selective in the production of vinyl acetate at a cost of by-products, for example carbon dioxide.

: * As far as the catalyst components are concerned, it is preferred that the catalyst support is either porous silica, alumina, silica / alumina or titanium oxide; : cid, the former being the most preferred. The carrier should have a surface area in the range of 100 to 800 m2 / g. With respect to the palladium content of the catalyst, it is suitably more than 2.5 grams per liter of catalyst to achieve the above-mentioned high productivity. Typical catalysts having '. *: The sodium content is about 0.5% by weight, preferably the palladium content would be greater than 3.0 g per liter of catalyst, more preferably more than 3.9 g per liter of catalyst, and most preferably 35.3, 9.6 to 6.1 g per liter of catalyst. On the other hand, the content of gold should suitably be more than 1.5 g per liter of catalyst and for catalysts having the sodium content indicated above, preferably more than 1.8 g per liter of catalyst, more preferably more than 2.3 g per liter of catalyst and most preferably 2.3 to 7.6 g per liter of catalyst. Finally, it is preferred that the weight ratio of gold to palladium is in the range 0.7 to 1.05, most preferably 0.85 to 1.05.

With respect to the potassium acetate content of the catalyst, it is suitably 5-9% by weight to achieve optimum activity.

In one embodiment of the present invention, it has been found that the level of undesirable ethyl acetate by-product can be substantially reduced by using a gold to palladium weight ratio greater than or equal to about 0.9.

The catalysts of the invention may conveniently be prepared by the method detailed in GB 1,559,540. In the first step of the process, the support is impregnated with a solution containing the required amount of palladium and gold in the form of soluble salts. Examples of salts are soluble halide derivatives. The impregnating solution is preferably an aqueous solution and the volume of solution used is such that it corresponds to a pore volume of the support of 95-100%, preferably 98-99%.

| After impregnation, the wet support is treated with an aqueous solution of 25 alkali metal salts, which include alkali metal silicates, carbonates and hydroxides.

The amount of alkali metal salt used is such that after contact of the support with impregnated: »t ·. with the carrier for 12-24 hours, the solution will suitably have a pH of 6.5 to 9.5, preferably 7.5 to 8, measured at 25 ° C. Preferably

t I

The t * 'alkali metal salts include sodium metal silicate, sodium carbonate, and sodium hydroxide.

·: **: During the above treatment, palladium and gold hydroxides are believed to precipitate or combine with the support. To convert these materials into the metal state * · · 113016 6 the impregnated support is treated with a reducing agent such as ethylene, hydrazine, formaldehyde or hydrogen. If hydrogen is used, it is usually necessary to heat the catalyst to 100-300 ° C to achieve complete reduction.

After completion of the above steps, the reduced catalyst is washed with water, impregnated with the required amount of alkali metal acetate and then dried.

In the preparation of the catalysts of the invention by the above process, they typically contain 0.5% by weight of sodium, which is largely derived from a precipitant. As described in detail in another separate patent application, it is desirable that the sodium content of these catalysts be lower than this number in order to achieve optimum catalyst activity. The sodium content of the catalyst may be reduced, for example, by washing with potassium acetate solution or by using reagents which do not contain sodium.

The preparation of vinyl acetate using the catalyst according to the invention is typically carried out by contacting ethylene, acetic acid and oxygen or air at a temperature of 100 to 200 ° C, preferably 100 to 200 ° C. , At 140-180 ° C and at a pressure ranging from 25 bar to 20 bar. The process is typically carried out heterogeneously with the reagents in the gas phase and; when the oxygen level is below the flammability limits. The reaction is usually carried out with excess ethylene, whereas the amount of acetic acid, in contrast, is determined by the dew point considerations.

After the reaction, the vinyl acetate is separated and purified by standard procedures.

• · * .1 ·: The invention will now be illustrated by the following examples.

1 · »113016 7

General procedure for the preparation of catalyst samples

Only deionized water was used in the process.

Step (1) - Impregnation of the carrier 15 g of a large surface spherical silica support (KA160-ex Sud Chemie) was added to 8.7 ml of an aqueous solution of Na2PdCl4 and HAuC14. The amounts of palladium and gold complexes used were such that the desired concentrations of palladium and gold on the support were achieved. The addition was done in one portion and the mixture was gently vortexed until the solution was completely absorbed. After impregnation, the impregnated support was allowed to stand for two hours at room temperature.

Step (2) - Precipitation

An aqueous solution of 18 ml of sodium metasilicate was rapidly added to the moist impregnated carrier. The concentration of a solution containing 18 ml of sodium metasilicate (SMS) was determined by the formula: 1.8 x [(1 mol SMS / mol Na2PdCl4) + (2 mol SMS / mol HAuC14) + (0.02 mmol SMS / g carrier)] . The mixture was vortexed occasionally over a 15 min period and allowed to stand without interruption overnight.

Step (3) - Reduction The aqueous phase on the black pellets was treated with 85% hydrazine hydrate solution. The amount of hydrazine hydrate used was determined using the formula; 22.5 x [(1 mol N2H4 / mol Na2PdCl4) + (1.5 mol N2H4 / mol: v, HAuCl4)]. The mixture was gently vortexed and then allowed to stand undisturbed overnight.

Step (4) - Washing

The aqueous phase containing a small amount of suspended black solids was decanted and the phases were washed four times with about 50 ml of water, decanting after each wash. The catalyst was transferred to a | * · »8

T130U

fitted with a stopcock and then washed with water at an approximate rate of one liter for 12 hours until a negative chloride test result with silver nitrate stock solution was obtained.

5 Steps (5) to (7) - Drying, Potassium Acetate Batching, and Final Drying

The catalyst was dried overnight in a convection oven at 60 ° C, cooled, then impregnated with a solution of the required amount of potassium acetate in 8.7 ml of 10 water. The mixture was vortexed until all of the liquid had been absorbed, then the catalyst was re-dried at 60 ° C with a stainless steel wire in a convection oven.

Catalyst Test Methods and Results 15 Tests were performed at 7.8 bar and 150 ° C for 2.5 g samples of 5-6 mm catalyst pellets diluted with 30 ml of 1 mm glass beads and charged with a stainless steel tube having an internal diameter of 10-11 mm. The catalyst was activated at 7.8 bar by heating at 160 ° C for 3 hours in a stream of nitrogen or helium and then at 150 ° C for 10 minutes in a stream of ethylene. The acetic acid vapors were then combined with ethylene and passed over the catalyst for at least one hour. A mixture which contains; containing 21% oxygen in helium was gradually added to the feed gas while maintaining a maximum catalyst bed temperature of 150 ° C. The catalyst hot spot was maintained at 150 ° C for 6 hours and then allowed to drop when the catalyst was deactivated. The final composition of the reagent mixture was ethylene: acetic acid: oxygen '. Helium = * · 30 53.1: 10, 4: 7.7: 28.6 and the total gas volume velocity * · · · * * per hour was 3850 h -1. The vapor phase of the product stream was analyzed every hour by on-line gas-liquid chromatography.

· p; Alternative Method for Preparing Catalyst Samples • * · i 35 The following is an alternative, improved method for preparing * ··· 'catalyst.

113016 9

Step (1) - Impregnation of the Carrier The same procedure as described above was used. Step (2) - Precipitation The same procedure as described above was used except that sodium hydroxide was used as the precipitant. The amount of sodium hydroxide used was determined using the formula: 1.8 x [(2 mol NaOH / mol Na2PdCl4) + (4 mol NaOH / mol HAUCl4) + (0.04 mmol NaOH / g carrier)].

Step (3) - Reduction The same procedure as above was used.

Step (4) - Washing

The product from step (3) was washed with water as described above. However, potassium acetate (5%) was used instead of water to wash the column.

Step (5) - Drying

The product of step (4) was dried overnight at 60 ° C on a stainless steel wire in a convection oven. Example 1 20 Catalysis test showing noble metal effect of other on activity and selectivity. The catalysts were prepared according to the above general procedure. Precious metal concentrations were varied to obtain • a statistically designed test series. You test- *: "! The compositions of the supported catalysts were determined by XRF analysis and are shown in Table 1. The potassium acetate content sought in each case was > 7.0% by weight, corresponding to 2.8% by weight of potassium. The resulting carrier was found to contain 0.4% by weight of potassium, giving a total potassium content of 3.2% by weight. Used catalyst »t • · ·; the weight in each test ranged from 0.5 to 2.5 g, * ·: / · allowing the activities and selectivities of the catalyst to be measured in the conversion range. The catalysts had a sodium content of about 0.5% by weight.

tt · 10 11301 (5 The test conditions used are given above. The total reagent flow was such that, in a 2.5 g catalyst assay, the GHSV (gas volume per hour) was 3850 h "1. The 5 tests performed with lower catalyst volumes maintained the total flow rate, activity, selectivity, and oxygen conversion measurements were made after 20 hours from the current and are shown in Table 2.

The variations in activity, seismic activity and conversion with the catalyst composition and weight are best suited to the following equations:

Equation 1. STY = ey where: y = 6.76 + 0.40 (Pd-0.82) + 0.13 (Au / Pd-0.81) - 0.57 (Au / Pd-0.81) ) 2 - 0.16 (Cat. Weight - 1.5) R 2 = 0.95 15 Equation 2. Selectivity = 100 - ez, where z = 1.41 - 0.29 (Au / Pd - 0.81) + 0.49 (Au / Pd-0.81) 2 - 0.39 (K-3.25) + 0.10 (Cat. Weight - 1.5) - 0.10 (Cat. 1.5 - 1.5) 2 R2 = 0.74 Equation 3. Oxygen Conversion: 20 = 24.5 + 10.0 (Pd-0.82) + 7.74 (Pd-0.82) (Cat. Weight - 1.5) + 1, 97 (Au / Pd-0.81) - 14.1 (Au / Pd-0.81) 2 * * '+ 11.6 (Cat. Weight - 1.5) R 2 = 0.96

The abbreviations used in the equations are Pd = palladium; , ·% By weight of catalyst in the catalyst; Au / Pd = gold: palladium-weight-G '*: 25 ratio; Cat. Weight = weight of catalyst, in grams. K = weight percent of potassium; STY = volume time yield in grams of vinyl acetate per liter of catalyst per hour. Correlation factors R indicate good fit of the results.

: V. To determine the effect of the catalyst composition at constant conversion, equation 3 was rearranged to reflect the weight of the catalyst as a function of palladium content, gold / palladium ratio and conversion. A representative oxygen conversion of 30% was placed in this function. The catalyst weight terms in equations 1 and 2 were then replaced by the weighting function 35 obtained by rearrangement of equation 3, whereby the resulting equations depicted the variation in catalyst activity and selectivity with metal concentration at constant conversion.

The effect of gold / palladium content on activity 5 is shown in Figure 1, while the effect on selectivity is shown in Figure 2. These curves show that improved catalyst activity can be obtained by increasing the catalyst palladium content and increasing the gold / palladium ratio to an optimum of 0.9 - 0.95. An increase in gold / palladium ratio greater than this reduces the activity of the catalyst.

Higher selectivity is also achieved by increasing the gold / palladium ratio. Finally, as can be seen in Figure 3, catalysts with a gold to paladium to-to-15 ratio of 0.9 or higher and little or no ethyl acetate by-product are obtained.

; - '· * t · * 1 »1» · • · »• · · e •» f I i2 11301 e

Table 1 1> Results used in the STY, selectivity, and conversion models as a function of palladium content and Au / Pd ratio

Pd Au Au / Pd K Cat. STY Select. 02 conversion wt% wt% wt% wt (measured after 20 h) 10 0.58 0.26 0.45 3.3 0.77 801 95.4 12.3 0j 58 0.26 0.45 3 , 3 2.49 602 94.5 30.0 0.88 0.42 0.48 3.3 2.50 702 94.9 36.3 0/91 0.34 0.37 3.2 2.50 654 93.8 33 , 9 0.91 0.34 0.37 3.2 2.50 624 94.2 32.2 0.91 0.34 0.37 3.2 0.76 835 95 «2 12.7 15 1.01 o'44 O , 44 3.1 2.50 727 94.7 35.9 0.91 0.34 0.37 3.2 1.25 747 94.7 19.0 l'06 1.21 1.14 3.1 0.85 993 96.2 16.0 0.58 0.83 1.43 3.2 2y50 592 95.4 28.6 θ ', 58 0.83 1.43 3.2 0.80 771 96.0 11.4 0.92 0.83 0, 90 3f1 2.46 760 95.4 35.1 1.00 0.87 0.87 3.1 1.55 939 95.3 28.0 f.01 0.44 0.44 3.1 0.80 991 95.1 15 ? 7 20 o'91 0.34 0.37 3.2 0.50 879 95.8 9.2 0.91 0.34 0.37 3.2 2.09 709 94.0 30.7 0 ^ 74 0.49 0166 3 ^ 3 2.50 693 94.8 35.7 0.70 0.29 0 j 41 3.4 2.50 644 95-2 32.2 0.88 0.42 0.48 3.3 2.50 691 94 , 9 36.3 0.87 1.16 1.33 3.2 0.85 916 96.0 14.5 0/58 0.26 0.45 3.3 2.48 555 95.5 25.3 l 9R 0.92 0 | 83 0 , 90 3.1 1.48 937 96.2 26.6 0.87 1.16 1.33 3.2 0.89 857 96.4 14.0 1 i

* I

• * 3.2 3.2 wt% K is approximately equal to 7.0 wt% KOAc * ♦ * i I t »# 113016 13

Example 2

Catalysis Assays showing the effect of potassium acetate content on catalyst activity and selectivity Catalysts containing varying amounts of potassium acetate were prepared by the general method described above as part of a statistically designed test series. The catalyst charge was synthesized at a concentration of 0.9% w / w palladium and a gold / palladium ratio of 0.44. Potassium-10 acetate concentrations of 0, 3, 5, 7, 9% by weight were impregnated into the same catalyst charge. Metal analyzes were performed using XRF and are shown in Table 2.

The catalysts were evaluated using the above test conditions. The activities (STY, g VA / h / l catalyst-15) and selectivity (% ethylene) after 20 hours from the stream are also shown in Table 2.

Activity variations best fit the equation: STY = 160.5 KoAc - 12.4 KoAc2 + 147 R2 = 0.998 20 where KoAc = actual KoAc by weight.

The square of the correlation factors (R2) shows that the results fit the equation well. The effect of potassium on activity (STY) is shown in Figure 4. The figure shows t. , · A clearly optimum range of potassium acetate content for better catalyst activity. Selectivity: I have also been found to increase potassium levels studied.

• · • ·} I i * · I>

»I

I t> · »<> •» * · »» * »I» 1; > I i f>

I I

) * »14 113016

Table 2

Catalyst activities and selectivities

5 Aiming for XRF-Ana- Really-STY SEL

lysis KOAc (at 20 h) KOAc% by weight KOAc by weight% by weight 10 0 0.48 0 148 84.3 3 1.77 3.24 539 93.2 5 2.48 5.02 631 94.2 7 3.31 7.10 677 94.0 9 4.06 8.99 585 94.3 15 # • · • ·> I t • • • • I · • ·

t 'I

• I · • 1 · • • • • • • • • • • • »»

«I I

Claims (16)

113016 Catalyst impregnated catalyst for use in the manufacture of vinyl acetate from ethylene, acetic acid and 5 oxygen-containing gas, characterized in that the catalyst has a productivity of more than 661 g vinyl acetate per hour at 150 ° C and consists essentially of: (1) a catalyst support the particle diameter is 10 about 3 to 7 mm and the pore volume is 0.2 to 1.5 ml per gram; (2) palladium and gold distributed over an outer 1.0 mm thick layer of catalyst support particles; and (3) about 3.5 to 9.5 weight percent potassium acetate, wherein the weight ratio of gold to palladium in the catalyst ranges from 0.60 to 1.25 and the palladium content is greater than 3.9 grams per liter of catalyst. Catalyst according to claim 1, characterized in that the gold: palladium weight ratio 20 is 0.7 to 1.05. Catalyst according to claim 2, characterized in that gold: palladium weight ratio. .0 is between 0.85 and 1.05. Any of the preceding claims; ; 25, characterized in that the potassium acetate content is 5 to 9% by weight. , ·; . Catalyst according to any one of the preceding claims, characterized in that . the di content is from 3.9 to 6.1 grams per liter of catalyst. 6. Shell impregnated catalyst for use in the preparation of vinyl acetate from ethylene, acetic acid and; ·,: An oxygen-containing gas, characterized in that it · · * consists essentially of: (1) a catalyst support having a particle diameter of about 3 to 7 mm and a pore volume of 0.2 to 1.5 ml per gram 113016 (2) palladium and gold distributed over an outer 1.0 mm thick layer of catalyst support particles; and (3) about 3.5 to 9.5 weight percent potassium acetate, wherein the weight ratio of gold to palladium in the catalyst is in the range of about 0.9 to 1.25 and the palladium content is greater than 3.9 grams per liter of catalyst. 7. Catalyst according to claim 6, characterized in that the potassium acetate content is 10 to 9% by weight. Catalyst according to claim 6 or 7, characterized in that the palladium content is from 3.9 to 6.1 grams per liter of catalyst. 9. A process for the preparation of vinyl acetate, characterized in that ethylene is reacted with acetic acid in the presence of an oxygen gas at 100 to 200 ° C and in the presence of a shell impregnated catalyst with a productivity greater than 661 g vinyl acetate per hour per 20 ° C: and consisting essentially of: (1) a catalyst support having a particle diameter of about 3 to 7 mm and a pore volume of 0.2 to 1.5 ml per gram; • (2) palladium and gold distributed over the outer • 1.0 mm thick layer of catalyst support particles; 25 and; j (3) about 3.5 to 9.5 weight percent potassium acetate, wherein the gold: palladium weight ratio in the catalyst is in the range 0.60 * 1.25 and the palladium content is greater than 3 , 9 grams ... per liter of catalyst. * · 10. A process according to claim 9, characterized in that the weight ratio of gold to palladium is 0.7 to 1.05. The method of claim 10, *. Known for being the catalyst for gold: Palladium 'll! The weight ratio of 35 is 0.85 to 1.05. • · «Il 113016 Process according to any one of claims 9 to 11, characterized in that the catalyst has a potassium acetate content of 5 to 9% by weight. Catalyst according to any one of claims 9 to 12, characterized in that the palladium content is from 3.9 to 6.1 grams per liter of catalyst. 14. A process for the preparation of vinyl acetate, wherein the ethylene is reacted with acetic acid in the presence of an oxygen gas at 100-200 ° C and in the presence of a shell impregnated catalyst consisting essentially of: (1) a catalyst support having a particle size of about 3 to 7 mm and a pore volume of 0.2 to 1.5 ml per gram (2) of palladium and gold distributed in the outermost layer of 1.0 mm thick catalyst support particles; and (3) about 3.5 to 9.5 weight percent potassium acetate, wherein the gold: palladium weight ratio in the catalyst is in the range of about 0.9 to 1.25 and the palladium content is greater than 3.9 grams per 20 liters of catalyst. Process according to claim 14, characterized in that the catalyst has a potassium acetate content of 5 to 9% by weight. Process according to claim 14 or 15, characterized in that the palladium content is from 3.9 to 6.1 grams per liter of catalyst. »> · ♦» »» • »» I I t t »k» I I • l k · f * · k ·> 11 »» • k k • * * k • kk * kk · t k k ·· 113016