Catalyst for the synthesis of alkyl mercaptans and process for producing it

Abstract

The present invention relates to a catalyst which comprises a support material and an oxidic composition containing at least one alkali metal and tungsten, a process for producing such catalysts and also a process for preparing alkyl mercaptans by reaction of alkanols with hydrogen sulphide in the presence of such a catalyst.

Claims

1. A process for producing a catalyst comprising a support material, wherein the support material has a particle size from 1-25 μm, and an oxidic composition comprising oxygen together with at least one alkali metal and tungsten, wherein the oxidic composition has a formula A.sub.xWO.sub.y, wherein A is at least one alkali metal, x is from 0.8 to 2 and represents the mole fraction of the alkali metal to tungsten in the composition, and y is from 3.4 to 4 and represents the mole fraction of oxygen in the composition, the process comprising: 1) mixing of the support material with an oxidic tungsten compound and at least one separate alkali metal compound to obtain a catalyst composition; and 2) shaping of the catalyst composition.

2. The process of claim 1, wherein the oxidic tungsten compound is selected from the group consisting of tungsten trioxide (WO.sub.3), tungstic acid (WO.sub.3.H.sub.2O), metatungstic acid, paratungstic acid, isopolytungstic acids, heteropolytungstic acids, ammonium salts thereof, hydrates thereof and mixtures thereof, ammonium orthotungstate, ammonium metatungstate and ammonium paratungstate.

3. The process of claim 1, wherein at least one alkali metal compound is a basic alkali metal compound.

4. The process of claim 1, wherein the support material is an oxidic inorganic support material.

5. The process of claim 1, wherein the oxidic tungsten compound and the separate alkali metal compound are added in succession to the support material.

6. The process of claim 1, wherein at least one organic and/or inorganic binder is added in one or both of 1) and 2).

7. The process of claim 1, wherein the shaping of the catalyst is carried out by extrusion or pressing.

8. The process of claim 3, wherein the basic alkali metal compound is selected from the group consisting of a hydroxide of an alkali metal and a carbonate of an alkali metal.

9. The process of claim 4, wherein the support material is selected from the group consisting of aluminium oxide, silicon dioxide, titanium dioxide, zirconium oxide, and amorphous aluminosilicates and mixtures thereof.

10. The process of claim 5, wherein the oxidic tungsten compound is added as a solid to the support material or to the mixture of support material and a least one alkali metal compound.

11. The process of claim 7, further comprising: (i) drying of the catalyst composition and/or of the shaped catalyst; and (ii) calcination of the catalyst composition and/or of the shaped catalyst.

12. A catalyst, comprising: a support material having a particle size from 1-25 μm; and an oxidic composition comprising oxygen together with at least one alkali metal and tungsten, wherein the oxidic composition has a formula A.sub.xWO.sub.y, wherein A is at least one alkali metal, x is from 0.8 to 2 and represents the mole fraction of the alkali metal to tungsten in the composition, and y is from 3.4 to 4 and represents the mole fraction of oxygen in the composition.

13. The catalyst of claim 12, wherein the catalyst is produced by a process comprising: mixing a solid tungsten compound with the support material.

14. The catalyst of claim 12, wherein the proportion of the oxidic composition composed of alkali metal and tungsten in the catalyst is greater than 15% by weight, based on the total weight of the catalyst.

15. The catalyst of claim 12, wherein the support material comprises at least one oxidic inorganic compound.

16. The catalyst of claim 12, which further comprises at least one organic and/or inorganic binder.

17. The catalyst of claim 12, wherein a standard deviation, stadev, of the normalized tungsten concentration, c.sub.norm (W), over the cross section of the catalyst is less than 20, determined by quantitative EDX analysis in accordance with ISO 22309 (2006) in square measurement spots, which have a side length of 100 μm in each case and whose midpoints lie on a straight line and are in each case 100 μm from the midpoint of the adjoining square, with the first and last midpoint of a square being in each case 100 μm from the edge of the catalyst cross section.

18. A catalyst comprising, based on a total weight of the catalyst: from 25% to 50% by weight of a support material having a particle size from 1-25 μm; and greater than 40% by weight of an oxidic composition comprising oxygen together with at least one alkali metal and tungsten.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows the conversion in the reaction of hydrogen sulphide and methanol to form methyl mercaptan achieved when using the catalysts according to the invention (Examples 3 to 7) at various temperatures under the conditions described in Example 8 compared to that achieved when using catalysts which have been produced by processes known from the prior art (Comparative Examples 1 and 2).

(2) FIG. 2 shows the selectivity in the reaction of hydrogen sulphide and methanol to form methyl mercaptan achieved when using the catalysts according to the invention (Examples 3 to 7) at various temperatures under the conditions described in Example 8 compared to that achieved when using catalysts which have been produced by processes known from the prior art (Comparative Examples 1 and 2).

(3) FIG. 3 shows the relationship between conversion and selectivity for the catalysts of Examples 1 to 7.

(4) FIG. 4 shows the relationship between conversion and selectivity for catalysts according to the invention having different loadings (Examples 9 to 13) in the reaction of hydrogen sulphide and methanol to form methyl mercaptan under the conditions described in Example 8 compared to a catalyst which has been produced by a process known from the prior art (Comparative Example 2).

(5) FIG. 5 shows the relationship between conversion and selectivity for various catalysts according to the invention having the same loading (Examples 11 and 14 to 17) in the case of which the catalyst composition was produced in different ways or shaping was carried out by different methods compared to a catalyst known from the prior art (Comparative Example 2) in the reaction of hydrogen sulphide and methanol to form methyl mercaptan.

(6) FIGS. 6 to 8 show the surface distribution of the elements caesium, tungsten and aluminium in two catalysts which have been produced by processes known from the prior art (FIG. 6: Example 1; FIG. 7: Example 2) compared to a catalyst according to the invention (FIG. 8: Example 6), determined by EDX mapping. As respective FIG. 6a, FIG. 7a, and FIG. 8a, a material contrast image of a catalyst particle in which the substantial homogeneity (FIG. 8a: Example 6) or inhomogeneity (FIG. 6a: Example 1) of the catalyst structure or the distribution of the elements therein can be seen is shown. The respective FIGS. 6 b, 6c, and 6d, FIGS. 7b, 7c, and 7d, and FIGS. 8b, 8c, and 8d show the concentration of the elements caesium (FIGS. 6 b, 7b, and 8b), tungsten (FIGS. 6c, 7 c, and 8c) and aluminium (FIGS. 6 d, 7 d, and 8d) in the catalyst particle. A scale which shows, to scale, the length of one millimeter of the original structure is also shown under the respective figure. The respective FIGS. 6e, 7e, and 8e show the distribution of the measurement squares for quantifying the homogeneity of the distribution of the elements caesium and tungsten over the catalyst cross section, as is described in detail in Example 18; the respective FIGS. 6 f, 7f, 8f show the graphical evaluation of these measurements.

EXAMPLES

Example 1 (Comparative Example)

(7) 200 g of spherical aluminium oxide having a particle diameter of from 2 to 5 mm (Spheralite 501 A from Axens having a specific surface area of 303 m.sup.2/g, a pore volume of 45 ml/100 g and a bulk density of 815 kg/m.sup.3) were impregnated in a three-stage impregnation with a total of 52.8% by weight of oxidic composition of the formula Cs.sub.1.44WO.sub.3.72 by means of vacuum impregnation. The following procedure was employed for this purpose: 103.3 g of tungstic acid were suspended in 206.5 g of 32% strength ammonia solution and dissolved by stirring for about 30 minutes. 126.9 g of a 70% strength solution of caesium hydroxide in water were added to the ammonia solution and the resultant solution was stirred for about 23-24 hours. The aluminium oxide was placed in a glass vessel which had been evacuated to 150 mbar. The impregnation solution was drawn in by opening a stopcock until there were about 4.5 cm of impregnation solution over the total aluminium oxide. After admission of air into the glass vessel, the support was incubated in the solution for about 15 minutes. The solution was subsequently drained off and the catalyst was predried for about one hour by passing through a stream of 200 standard I/h (volume flow in the standard state at 0° C. and 1.013 bar absolute in accordance with DIN 1343) of air, with adhering impregnation solution being flushed into the receiver.

(8) The catalyst was subsequently heated at a heating rate of 1° C./min to 120° C. under a stream of 60 m.sup.3/h of air and maintained at this temperature for three hours. The temperature was then increased at a heating rate of 5° C./min to 455° C. and the catalyst was calcined at this temperature for 3 hours.

(9) To carry out the second impregnation, an impregnation solution as described above for the first step was made up and applied in the same way by vacuum impregnation to the previously loaded catalyst obtained from the first impregnation. Predrying at room temperature followed by three-stage drying at 120° C. and subsequent calcination at 455° C. for 3 hours were carried out as described above.

(10) The third impregnation was carried out in the same way.

Example 2 (Comparative Example)

(11) Comparative Example 1 was repeated with a loading of 17.8% by weight of WO.sub.3 and 17.3% by weight of Cs.sub.2O on the aluminium oxide.

Example 3

(12) In this example according to the invention, 80 g of a pulverulent aluminium oxide having a particle diameter of from 7 to 15 μm (Spheralite 509A from Axens having a specific surface area of 335 m.sup.2/g, a pore volume of 56 ml/100 g and a bulk density of 840 kg/m.sup.3) were mixed in succession with solid tungstic acid and a calcium hydroxide solution.

(13) For this purpose, the procedure was as follows:

(14) In a glass beaker, 80 g of the aluminium oxide were mixed with 40.98 g of solid tungstic acid. 69.85 g of a 50% strength solution of caesium hydroxide in water and 5.33 g of a 6% strength aqueous methylhydroxyethylcellulose solution (Tylose MH 1000, ShinEtsu, Tokyo, Japan) were added to the pulverulent mixture and the mixture was kneaded with the aid of a spatula for 10 minutes until an extrudable composition was formed, i.e. the liquid had been taken up completely and a dough-like composition which was not sticky had been obtained (about 10 minutes). 5.33 g of petroleum (Merck, Darmstadt, Germany) were added and kneaded into the mixture. The mixture was dried and subsequently calcined in a muffle furnace by firstly heating it at a heating rate of 2° C./min to 120° C., maintaining it at this temperature for 3 hours, then heating it at a heating rate of 5° C./min to 455° C. and maintaining it at this temperature for 3 hours. The mixture was subsequently cooled to 20° C.

(15) After cooling, the granular material obtained was milled in a mortar. From 1 to 2 g of the catalyst powder obtained were subsequently pressed in a tabletting press at a pressure of 4 t for about 1 minute to give a pellet having a diameter of 20 mm.

(16) For subsequent use in a test reactor for preparing methyl mercaptan, the pellet was broken up into pieces having a maximum edge length of 5 mm.

Example 4

(17) Example 3 was repeated with a loading of 17.8% of WO.sub.3 and 17.3% by weight of Cs.sub.2O on the aluminium oxide.

Example 5

(18) Example 4 was repeated using aluminium oxide having a particle diameter of less than 250 μm obtained by milling of spherical aluminium oxide having a particle diameter of from 2 to 5 mm (Spheralite 501 A) instead of the pulverulent aluminium oxide having a particle diameter of from 7 to 15 μm (Spheralite 509 A).

Example 6: Production of the Catalyst Particles According to the Invention with Addition of Binders and Shaping by Extrusion

(19) 1.05 kg of Spheralite 509 A and 537.9 g tungstic acid were mixed in a laboratory batch kneader (Coperion LUK 2.5, Weinheim, Stuttgart, Germany) at 40 revolutions per minute of the kneading hook and 11 revolutions per minute of the discharge screw (backward-directed), with the barrel of the kneader being cooled to 10° C. by means of a cryostat. 740.5 g of a 70% strength by weight aqueous caesium hydroxide solution were subsequently added over a period of 1 minute with continual mixing, resulting in the temperature rising briefly from 30 to 40° C. 127.5 g of deionized water and then 175 g of a colloidal silica dispersion (Lithosol 1530, Zschimmer & Schwarz GmbH & Co. KG, Lahnstein, Germany) were added 10 minutes after addition was completed. The mixture obtained was mixed for a further 10 minutes before a mixture of 30 g of a high-polymer polysaccharide (Zusoplast PS 1, Zschimmer & Schwarz GmbH & Co. KG, Lahnstein, Germany) and 30 g of hydroxyethylcellulose (Tylose H 10000 P2 ShinEtsu, Tokyo, Japan) were added. The binders were allowed to swell for 120 minutes while kneading the composition continually. 15 g of a nonionic wax dispersion (Zusoplast WEB, Zschimmer & Schwarz GmbH & Co. KG, Lahnstein, Germany) were subsequently added. After a total kneading time of 190 minutes, extrusion was commenced, for which purpose the direction of rotation of the screw was changed over to transport at a constant speed of rotation of the kneader and the speed of rotation of the screw was increased to 13 revolutions per minute. An attachment having four horizontal holes each having a diameter of 3.2 mm was used as pressing tool. Two cutting wires which were cut horizontally were operated at 400 revolutions per minute in order to obtain extrudates having a length of about 3.2 mm. The die pressure was 12.7 bar. The cut extrudates were allowed to fall onto a drying belt and predried at 60° C. before being heated at a heating rate of 1° C./min to 120° C. in a muffle furnace and maintained at this temperature for 3 hours. To carry out calcination, the extrudates were directly afterwards heated at a heating rate of 5° C./min to 455° C. and maintained at this temperature for 3 hours.

Example 7

(20) Example 6 was repeated, this time with the addition of the caesium hydroxide solution being carried out before the addition of the solid tungstic acid.

1.1. Example 8: Use Example

(21) The catalysts produced in Examples 1 to 7 were examined in respect of their performance characteristics in the synthesis of methyl mercaptan from hydrogen sulphide and methanol.

(22) The reaction of hydrogen sulphide and methanol to form methyl mercaptan in the presence of the respective catalyst was carried out in a stainless steel tube having a diameter of 18 mm and a length of 500 mm. A catalyst bed which had a volume of 76 ml and was in each case fixed in the reaction tube by means of inert beds of glass spheres on both sides was used in each case. The reaction tube was heated via a double wall by means of a heat transfer fluid to the various reaction temperatures in the range from 300 to 360° C. indicated in Table 1 below.

(23) TABLE-US-00001 TABLE 1 Con- Selec- Loading T version tivity Sup- WO.sub.3 Cs.sub.2O Catalyst [° C.] [%] [%] port* [% by wt.] [% by wt.] Example 1 300 81.3 95.9 I (imp) 28.20 24.6 (comparative 320 87.4 95.3 example) 350 94.6 94.2 360 95.2 93.9 Example 2 300 80.6 96.6 I (imp) 17.80 17.3 (comparative 320 84.4 96.3 example) 350 93.6 95.4 360 95.5 95.0 Example 3 300 89.3 97.6 II (press 25.81 22.49 320 93.7 97.2 4) 340 97.1 96.7 350 98.2 96.5 Example 4 300 84.5 96.5 II (press 17.80 17.3 320 90.8 96.5 4) 340 95.7 95.8 350 97.4 95.3 Example 5 300 77.9 96.6 I (mill, 17.80 17.3 320 83.4 96.5 press 4) 340 88.7 96.1 350 91.1 95.7 Example 6 300 88.2 97.5 II (extr) 25.81 22.49 320 92.7 97.6 350 98.4 96.7 360 99.2 96.4 Example 7 300 86.6 98.3 II (extr) 25.81 22.49 320 91.8 98.2 350 97.7 97.4 360 98.6 97.1 *I: Spheralite 501A, particle size 2-5 mm; II: Spheralite 509A, particle size 7-15 μm; imp: impregnated; press 4: pressed into pellet shape at a pressure of 4 t; mill: milled (particle size after milling ≦250 μm); extr: extruded

(24) Further experimental conditions are indicated below:

(25) GHSV: 1300 h.sup.−1 (based on standard conditions at 0° C. and 1.013 bar in accordance with DIN 1343)

(26) LHSV: 0.4 h.sup.−1 (based on liquid methanol)

(27) Mass ratio of H.sub.2S/MeOH: 1.9

(28) Pressure: 9 bar

(29) The reaction mixture obtained, which comprised the products methyl mercaptan, dimethyl sulphide, dimethyl disulphide and dimethyl ether and also the unreacted starting materials methanol and hydrogen sulphide was analysed by on-line gas chromatography.

(30) The measurement results are shown in Table 1 and also FIGS. 1 to 3.

(31) It can be seen that a catalyst produced by the process of the invention and having a more finely divided support material displays, compared to a catalyst produced by impregnation of particles of support material having a size of 2 mm and more, a higher conversion at the same loading at a given temperature (Comparative Example 2 compared to Example 4 according to the invention) and thus a higher selectivity to methyl mercaptan at the same conversion.

(32) Furthermore, catalysts in the form of extrudates and pressed bodies which have a high loading (for example above 45% by weight) of the oxidic composition can be produced by the production process of the invention, enabling the conversion and the selectivity of the catalysed reactions to be increased further. This is not possible at a loading above 45% by weight in the case of a catalyst produced by the impregnation process (Comparative Example 1 and Example 3 according to the invention compared to Comparative Example 2 and Example 4 according to the invention.

(33) Addition of inorganic and/or organic binders, preferably inorganic binders, not only enables the mechanical strength of the catalyst according to the invention to be increased, but it is surprisingly also possible to observe a further increase in the selectivity and the conversion at a particular temperature (Example 6).

(34) It can also be seen that the order of addition of the alkali metal hydroxide and the oxidic tungsten compound also influence the conversion and the selectivity of the reaction at a particular temperature (Examples 6 and 7 according to the invention).

1.2. Examples 9 and 13: Variation of the Loading

(35) Example 5 was repeated with an increased loading of tungsten(VI) oxide and caesium hydroxide on the aluminium oxide. The respective loading of the catalyst and the conversion achieved therewith at a particular temperature and the selectivity achieved in a use example as per Example 8 are shown in Table 2 and also FIG. 4.

(36) TABLE-US-00002 TABLE 2 Con- Selec- Loading T version tivity Sup- WO.sub.3 Cs.sub.2O Catalyst [° C.] [%] [%] port* [% by wt.] [% by wt.] Example 9 300 81 97.23 I (mill, 20.5 19.9 320 86.05 97.05 press 4) 340 91.17 96.63 350 93.3 96.35 Example 10 300 83.05 97.2 I (mill, 21.8 21.2 320 89.13 97 press 4) 340 94 96.65 350 96.05 96.3 Example 11 300 83.3 97.7 I (mill, 23.1 22.5 320 88.47 97.43 press 4) 340 92.9 97.1 350 94.57 96.9 Example 12 300 90.15 97.25 I (mill, 24.5 23.8 320 94.93 97.1 press 4) 340 98.03 96.43 350 98.8 96.1 Example 13 300 83.3 97.4 I (mill, 25.8 25.1 320 88.8 96.95 press 4) 340 93.55 96.6 350 95.15 96.4 *I: Spheralite 501A, particle size 2-5 mm; press 4: pressed to pellet shape at a pressure of 4 bar; mill: milled (particle size after milling ≦250 μm)

(37) It can be seen that an increase in the loading has a positive influence on the conversion and the selectivity of the catalysed reaction at a given temperature and both conversion and selectivity at a given temperature and also the selectivity of the reaction of methanol and hydrogen sulphide to form methyl mercaptan in the presence of the catalyst according to the invention can be increased by increasing the loading.

1.3. Examples 14 to 17: Influence of the Particulate Support Material

(38) In Examples 14 to 17, various processes for producing the catalyst according to the invention were compared with one another at a loading of 23.1% by weight of WO.sub.3 and 22.5% by weight of Cs.sub.2O.

Example 14

(39) Example 11 was repeated with the same loading, but the pressure in the tabletting press was 15 t instead of 4 t.

Example 15

(40) Example 3 was repeated with a loading of 23.1% by weight of WO.sub.3 and 22.5% by weight of Cs.sub.2O.

Example 16

(41) Example 11 was repeated with the same loading, but the aluminium oxide support was not milled before mixing with the tungstic acid but instead was milled together with the tungstic acid in a ball mill (from Haldenwanger Berlin) using a Schwinherr Multifix drive for 2 hours.

Example 17

(42) Example 16 was repeated with the support material being milled with the tungstic acid for 65 hours instead of 2 hours in the ball mill.

(43) The conversion and selectivity of the reaction of methanol and hydrogen sulphide to form methyl mercaptan was determined for the catalysts according to the invention of Examples 14 to 17 as described in Use Example 8. The results are shown in Table 3 and FIG. 5.

(44) TABLE-US-00003 TABLE 3 Loading Con- Selec- WO.sub.3 Cs.sub.2O T version tivity Sup- [% by [% by Catalyst [° C.] [%] [%] port* wt.] wt.] Example 14 300 85.2 97.35 I (mill, press 23.1 22.5 320 91 97.1 15) 340 95.47 96.73 350 97.03 96.43 360 98.25 96 370 99.1 95.43 Example 15 300 88.45 97.3 II (press 4) 23.1 22.5 320 93.53 97.33 340 97 96.85 350 98.13 96.33 Example 16 300 88.1 97.65 I (ball mill 2, 23.1 22.5 320 92.75 97.6 press 4) 340 96.7 97.25 350 98.05 96.75 Example 17 300 87.23 97.93 I (ball mill 23.1 22.5 320 91.7 97.7 65, press 4) 340 96.05 97.35 350 97.3 96.95 360 98.5 96.65 *I: Spheralite 501A, particle size 2-5 mm; II: Spheralite 509A particle size 7-15 μm; imp: impregnated; press 4/15: pressed to pellet shape at a pressure of 4 or 15 t; ball mill 2/65: support material milled with tungstic acid for 2 or 65 h in a ball mill; mill: milled (particle size after milling ≦250 μm)

(45) Like the preceding examples, Examples 14 to 17 also show that the present invention makes it possible to provide catalysts by means of which a selectivity of above 95% can be obtained in the reaction of methanol and hydrogen sulphide to form methyl mercaptan even at a conversion of greater than 95%. Thus, for example, a selectivity of above 95% can still be achieved at a conversion of above 99% by means of the catalyst of Example 14 at a temperature of 370° C.

(46) A small particle size of the support material and intensive mixing of the support material, the oxidic tungsten compound and the alkali metal hydroxide, for example in a ball mill as per Example 16 or 17, have a positive effect on the conversion and the selectivity of the catalyst at a particular temperature with otherwise unchanged catalyst parameters (Examples 16 and 17 compared to Example 11).

1.4. Example 18: Energy-Dispersive X-Ray Spectroscopy

(47) The concentration distribution of the elements aluminium, caesium and tungsten on the surface of the catalysts as per Examples 1, 2 and 6 was made visible by means of energy-dispersive X-ray spectroscopy (EDX mapping) in accordance with ISO 22309 (2006) (FIGS. 6 to 8). A Jeol 7600F scanning electron microscope (SEM) (Jeol Ltd., Tokyo, Japan) together with an Oxford INCA Energy 400 energy dispersive X-Ray analysis (EDX) system (Oxford Instruments, Abingdon, Great Britain) was used for this purpose. The scanning electron micrographs and the EDX mappings of the selected elements were recorded at a primary electron beam energy of 20 keV. The cut, embedded and polished catalyst particles were coated with a 20 nm thick carbon coating to make them electronically conductive for the analytical electron beam.

(48) It is here surprisingly found that the elements caesium and tungsten are more homogeneously distributed in the catalyst when using the process of the invention in which the support material is preferably firstly mixed with compounds of the one element before the compound of the other element is added to this mixture than when catalyst particles are treated by the impregnation process known from the prior art with a solution containing both elements in the form of caesium tungstate.

(49) A material contrast image of a catalyst particle is shown as respective Figure a), in which the substantial homogeneity (FIG. 8a: Example 6) or inhomogeneity (FIG. 6a) of the catalyst structure or the distribution of the elements in the particle can be seen. The respective figures b)-d) show the concentration of the elements caesium (Fig. b), tungsten (Fig. c) and aluminium (Fig. d) in the catalyst particle.

(50) The lighter a place appears in these images, the denser the material and therefore the higher the concentration of the element concerned.

(51) It can clearly be seen that the catalyst according to the invention (FIGS. 8a to d) displays a clearly homogenous distribution of the elements over the entire catalyst, i.e. also on the surface thereof.

(52) To quantify this, the element distribution was determined along a straight line in the cross section of a catalyst particle by means of quantitative EDX measurements in about 40 square sections each having a side length of 100 μm. The midpoints of the squares were located on the straight line and are each 100 μm away from the midpoint of the respective adjoining square. The first and last midpoints of squares on this straight line are in each case about 100 μm away from the edge of the catalyst body. The distribution of these measurement squares over the catalyst cross section is shown in the respective figures e).

(53) The proportion of an element was determined from the intensities of the signals in the respective square based on the sum of the proportions of all elements in this square (hereinafter referred to as concentration), with the sum in each measurement spot being 100%. From the concentrations of caesium and tungsten determined in the respective squares, the average was formed over the cross section of the specimen, i.e. over the about 40 individual values. The ratio of the experimentally determined concentration of caesium or tungsten in the respective square to the average of this concentration over the cross section of the specimen multiplied by the factor 100 will be referred to as the normalized concentration c.sub.norm, from which the standard deviation stadev, the average deviation avdev, the largest and smallest measured value in each case (max or min) and the difference between maximum and minimum of the normalized concentrations Δ (max-min) over the cross section of the respective specimen were calculated by means of the table calculation Excel (Microsoft Office Excel 2003, Microsoft Corporation, Redmont, U.S.A.). The results are shown in Tables 4 (Example 1 which is not according to the invention), 5 (Example 2 which is not according to the invention) and 6 (Example 6 according to the invention).

(54) TABLE-US-00004 TABLE 4 Quantification of the normalized concentration of caesium and tungsten over the catalyst cross section in Example 1 c.sub.norm (Cs)/ MP c.sub.norm (Cs) [%] c.sub.norm (W) [%] c.sub.norm(W) 1 103.31 159.65 0.65 2 85.44 122.15 0.70 3 80.13 112.44 0.71 4 76.01 105.57 0.72 5 87.64 113.18 0.77 6 89.13 113.61 0.78 7 68.99 90.78 0.76 8 94.38 113.92 0.83 9 71.26 98.21 0.73 10 122.95 103.53 1.19 11 116.79 100.56 1.16 12 113.24 108.29 1.05 13 117.99 98.21 1.20 14 111.89 96.66 1.16 15 105.44 100.74 1.05 16 90.41 72.83 1.24 17 94.87 77.48 1.22 18 92.11 70.17 1.31 19 105.58 75.19 1.40 20 97.92 69.86 1.40 21 101.47 74.01 1.37 22 94.52 68.75 1.37 23 90.97 72.40 1.26 24 96.08 67.88 1.42 25 107.85 84.96 1.27 26 112.11 92.57 1.21 27 106.79 94.74 1.13 28 116.15 99.32 1.17 29 120.05 99.20 1.21 30 120.54 103.71 1.16 31 121.82 109.16 1.12 32 127.70 107.61 1.19 33 111.54 116.46 0.96 34 63.68 99.94 0.64 35 100.19 116.71 0.86 36 97.29 116.09 0.84 37 82.75 106.81 0.77 38 90.62 119.74 0.76 39 112.39 146.91 0.77 Stadev 15.97 20.51 0.25 Avdev 13.12 15.22 0.22 Max 127.70 159.65 1.42 Min 63.68 67.88 0.64 Δ (Max − Min) 64.03 91.77 0.78 MP stands for square measurement spots.

(55) TABLE-US-00005 TABLE 5 Quantification of the normalized concentration of caesium and tungsten over the catalyst cross section in Example 2 c.sub.norm (Cs)/ MP c.sub.norm (Cs) [%] c.sub.norm (W) [%] c.sub.norm(W) 1 166.40 213.91 0.78 2 147.14 187.15 0.79 3 128.88 153.28 0.84 4 91.08 101.77 0.90 5 99.89 94.92 1.05 6 98.05 97.83 1.00 7 98.05 91.19 1.08 8 97.76 93.27 1.05 9 91.65 89.32 1.03 10 94.85 88.36 1.07 11 95.70 82.69 1.16 12 93.29 84.35 1.11 13 91.87 88.49 1.04 14 96.41 81.24 1.19 15 95.13 85.25 1.12 16 97.83 85.11 1.15 17 95.56 85.25 1.12 18 91.51 85.59 1.07 19 97.48 78.33 1.24 20 92.58 82.20 1.13 21 92.01 88.91 1.03 22 92.43 83.52 1.11 23 89.38 82.48 1.08 24 87.11 86.77 1.00 25 91.44 82.69 1.11 26 93.50 79.51 1.18 27 92.36 80.96 1.14 28 93.93 85.73 1.10 29 94.49 88.01 1.07 30 98.90 88.56 1.12 31 98.26 90.43 1.09 32 95.35 86.35 1.10 33 95.63 91.88 1.04 34 93.00 91.88 1.01 35 91.79 95.34 0.96 36 95.28 97.07 0.98 37 95.63 102.74 0.93 38 86.32 103.77 0.83 39 95.13 106.12 0.90 40 141.32 170.14 0.83 41 125.61 167.66 0.75 Stadev 16.88 31.20 0.12 Avdev 10.21 19.83 0.09 Max 166.40 213.91 1.24 Min 86.32 78.33 0.75 Δ (Max − Min) 80.07 135.58 0.50

(56) TABLE-US-00006 TABLE 6 Quantification of the normalized concentration of caesium and tungsten over the catalyst cross section in Example 6 c.sub.norm (Cs)/ MP c.sub.norm (Cs) [%] c.sub.norm (W) [%] c.sub.norm (W) 1 112.60 107.38 1.05 2 110.31 99.52 1.11 3 102.72 97.71 1.05 4 103.75 97.24 1.07 5 112.52 97.71 1.15 6 106.75 108.11 0.99 7 109.67 110.64 0.99 8 95.85 99.31 0.97 9 102.48 112.97 0.91 10 103.19 105.73 0.98 11 97.82 101.85 0.96 12 105.72 99.26 1.07 13 100.43 99.57 1.01 14 99.01 99.78 0.99 15 99.80 101.49 0.98 16 107.70 101.74 1.06 17 96.56 92.59 1.04 18 105.56 105.16 1.00 19 108.73 108.42 1.00 20 102.56 87.05 1.18 21 99.56 100.14 0.99 22 92.37 91.35 1.01 23 104.06 109.45 0.95 24 111.10 106.61 1.04 25 91.18 94.55 0.96 26 93.24 90.73 1.03 27 85.34 86.85 0.98 28 73.88 89.28 0.83 29 77.04 91.35 0.84 30 100.35 95.23 1.05 31 103.98 104.80 0.99 32 88.50 97.81 0.90 33 92.61 92.33 1.00 34 96.32 102.68 0.94 35 84.86 85.40 0.99 36 87.79 92.85 0.95 37 104.77 102.78 1.02 38 106.91 107.38 1.00 39 109.99 108.57 1.01 40 112.44 116.64 0.96 Stadev 9.48 7.62 0.07 Avdev 7.41 6.13 0.05 Max 112.60 116.64 1.18 Min 73.88 85.40 0.83 Δ (Max − Min) 38.72 31.24 0.35

(57) It can clearly be seen that standard deviation and average deviation of both the caesium distribution and the tungsten distribution over the catalyst are significantly smaller in the catalyst according to the invention of Example 6. The difference between the maximum and minimum of the concentration determined in each case is also significantly smaller in the catalyst according to the invention. In addition, the ratio of the normalized caesium concentration to the normalized tungsten concentration c.sub.norm (Cs)/c.sub.norm (W) in the catalyst according to the invention is also significantly more uniform over the catalyst cross section, as indicated by the standard deviation and the average deviation of this value and also the difference between maximum and minimum value for the respective catalysts.

(58) These quantitative measurements confirm the significantly more homogeneous distribution of the elements caesium and tungsten over the entire catalyst.