CATALYST BED COMPRISING SILVER CATALYST BODIES AND PROCESS FOR THE OXIDATIVE DEHYDROGENATION OF OLEFINICALLY UNSATURATED ALCOHOLS

Abstract

The present invention relates to a catalyst bed comprising silver catalyst bodies and a reactor comprising such a catalyst bed. Further, the invention relates to the use of the catalyst bed and the reactor for gas phase reactions, in particular for the oxidative dehydrogenation of organic compounds under exothermic conditions. In a preferred embodiment, the present invention relates to the preparation of olefinically unsaturated carbonyl compounds from olefinically unsaturated alcohols by oxidative dehydrogenation utilizing a catalyst bed comprising metallic silver catalyst bodies.

Claims

1.-18. (canceled)

19. A process for the preparation of an olefinically unsaturated carbonyl compound in a tubular reactor comprising a plurality of reactor tubes, comprising reacting an olefinically unsaturated alcohol with oxygen in the presence of a catalyst bed, comprising full-metallic silver catalyst bodies, wherein the catalyst bed has a packing density of the full-metallic silver catalyst bodies in the range of 3.0 g/cm.sup.3 to 10.0 g/cm.sup.3.

20. The process according to claim 19, wherein the catalyst bed has a packing density of the full-metallic silver catalyst bodies in the range of 5.5 g/cm.sup.3 to 10.0 g/cm.sup.3.

21. The process according to claim 19, wherein the catalyst bed has a void space ratio in the range of 5% to 70%, based on the volume of the catalyst bed not occupied by the catalyst bodies per volume of the catalyst bed.

22. The process according to claim 19, wherein the full-metallic silver catalyst bodies have a mean particle size of 0.5 mm to 5.0 mm.

23. The process according to claim 19, wherein the full-metallic silver bodies have a cylindrical shape or spherical shape or sphere-like shape or combinations thereof.

24. The process according to claim 19, wherein the full-metallic silver bodies have a geometric surface area in the range of 100 mm.sup.2/g to 600 mm.sup.2/g.

25. The process according to claim 19, wherein the catalyst bed is located in a tube reactor.

26. The process according to claim 19, wherein the olefinically unsaturated carbonyl compound is an α,β- and/or β,γ-olefinically unsaturated aldehyde and the olefinically unsaturated alcohol is an α,β- and/or β,γ-olefinically unsaturated alcohol.

27. The process according to claim 19, wherein the unsaturated carbonyl compound is an olefinically unsaturated aldehyde, selected from a compound of formula (Ia), formula (Ib) and mixtures thereof ##STR00003## wherein R.sup.1, R.sup.2, R.sup.3 and R.sup.4 are, identical or different, selected from the group consisting of H, substituted or unsubstituted C.sub.1-C.sub.10-alkyl and substituted or unsubstituted C.sub.3-10-cycloalkyl; or R.sup.1 and R.sup.2 together with the carbon atoms to which they are bonded form a substituted or unsubstituted, 5- or 6-membered carbocyclic ring; or R.sup.2 and R.sup.4 together with the carbon atoms to which they are bonded form a substituted or unsubstituted, 5- or 6-membered carbocyclic ring; or R.sup.4 and R.sup.3 together with the carbon atoms to which they are bonded form a substituted or unsubstituted, 5- or 6-membered carbocyclic ring.

28. The process according to claim 27, wherein R.sup.1 is selected from H and C.sub.1-4-alkyl, R.sup.2 is selected from H and C.sub.1-4-alkyl, R.sup.3 is selected from H and C.sub.1-4-alkyl, R.sup.4 is selected from H and C.sub.1-4-alkyl.

29. A catalyst bed as defined in claim 19, wherein the catalyst bed has a packing density of the full-metallic silver catalyst bodies in the range of 5.5 g/cm.sup.3 to 10.0 g/cm.sup.3.

30. A catalyst bed according to claim 29, wherein the full-metallic silver bodies have a geometric surface area in the range of 100 mm.sup.2/g to 600 mm.sup.2/g.

31. The catalyst bed according to claim 29, wherein the catalyst bed is located in a tube reactor.

32. A reactor, comprising a plurality of reactor tubes containing a catalyst bed as defined in claim 29.

33. The reactor according to claim 32, wherein the catalyst beds have a radial thermal conductivity Ar in the range of 1.0 to 1.5 W/m/K.

34. The reactor according to claim 32, wherein the catalyst beds have a heat transfer value α.sub.w in the range of 1000 to 1550 W/m.sup.2/K.

35. A method for preparing olefinically unsaturated carbonyl compounds from olefinically unsaturated alcohols comprising oxidatively dehydrogenating over the catalyst bed according to claim 29.

36. The method according to claim 35 where the olefinically unsaturated carbonyl compounds are 3-methyl-3-buten-1-al (isoprenal) or 3-methylbut-2-enal (prenal), and where the olefinically unsaturated alcohols are 3 -methyl-3 -butene-1-ol (isoprenol) or 3 -methylbut-2-en-1-ol (prenol).

37. The process according to claim 19, wherein the catalyst bed has a void space ratio in the range of 10% to 50%, based on the volume of the catalyst bed not occupied by the catalyst bodies per volume of the catalyst bed.

38. The process according to claim 19, wherein the full-metallic silver catalyst bodies have a mean particle size of 1.0 mm to 4.0 mm.

Description

EXAMPLES

[0160] FIG. 1: Selectivity towards prenal and iso-prenal as a function of the iso-prenol conversion, for the catalysts described in the examples

[0161] FIG. 2: Heat profile of the catalyst bed of examples under operation

[0162] Analytics:

[0163] A) Method for determining the packing density:

[0164] A glass tube with an inner diameter of 13 mm is filled with a material of interest to a defined packing height. The mass of the packed material is divided by the inner volume of the tube corresponding to that packing height.

[0165] B) Method for determining the geometric surface area ranges in the case of sphere-like catalyst bodies:

[0166] The geometric surface area ranges of catalyst bodies are calculated by assuming ideal sphericity of the catalyst bodies and using minimum and maximum diameters of a corresponding sieve fraction. The specific density of silver is used to calculate mas-specific geometric surface areas expressed in mm.sup.2/g.

[0167] C) Method for determining the size distribution (sieve fraction):

[0168] The size distribution, expressed as sieve fraction, is measured using sieves with defined sieve sizes. For example: a material with a sieve fraction of 1 to 4 mm will pass through a sieve with a sieve size ≥4 mm and will be completely retained by a sieve with a sieve size of ≤1 mm.

[0169] D) Method for determining void fraction in the tubular reactor:

[0170] The void fraction is calculated starting from the density of the catalyst bed. The combined volume of catalyst particles in the catalyst bed is calculated using the intrinsic material density (specific density) of the material. In the case of silver we have used a value of 10.5 g/ml. The void fraction is then the ratio between the void volume (volume of the catalyst bed minus the calculated combined volume of all the catalyst bodies in the catalyst bed) and the volume of the catalyst bed.

[0171] E) Method for determining the weight of the catalyst bed

[0172] A glass tube with an inner diameter of 13 mm is filled with a material of interest to a defined packing height. The mass of the packed material is then measured.

[0173] Oxidative Dehydrogenation

[0174] A setup was used comprising a continuous alcohol evaporation chamber where the educt was evaporated and mixed with air, after which the gaseous reagent was directed to a quartz reactor. The reactor had an internal diameter of 13 mm and held the catalyst bed by a metal sieve. The reactor contained a central thermocouple placed inside a glass tube (OD 3 mm), which went through the length of the catalyst bed. The catalyst bed length was kept at 7 cm. The reactor was surrounded by a chamber which was heated by an electric heating coil. This chamber contained sand which could be fluidized by a nitrogen flow, which was used to control the temperature of the reactor. Initially, the reactor was heated by the sand bath to ignite the reaction. Once the reaction was started, the fluidized sand bath was used as cooling medium to remove heat from the reactor, originating from the highly exothermic oxidation of the alcohol. A water-cooled condensation chamber was placed immediately after the reactor where the unconverted reagent and condensable products were accumulated. This condensate was periodically analyzed by a gas chromatographer. The non-condensable products left the condensation chamber and are monitored with an on-line gas chromatographer.

Example 1 (according to the invention)

[0175] Fully metallic silver shot (1-3 mm, Sigma-Aldrich, ≥99.99%) was placed inside the above described reactor to obtain a catalyst bed length of 7 cm. 110 g/h of isoprenol was evaporated and mixed with 50 NL/h of air. This reagent stream was sent to the reactor which was heated at 360° C. After 3 hours of operation, the sandbath temperature was adjusted between 380 and 400° C. to obtain iso-prenol conversion levels between 45 and 60%. At an iso-prenol conversion of 50%, a prenal and iso-prenal selectivity of 91% was obtained. The results are depicted in table 1 below.

Example 2 (according to the invention)

[0176] Fully metallic silver cylinders (height=2.8 mm, diameter=2 mm, Sigma-Aldrich, 9.99%) were placed inside the above described reactor to obtain a catalyst bed length of 7 cm. The material was initially ordered as a longer rod which was cut to the defined length. 110 g/h of isoprenol was evaporated and mixed with 50 NL/h of air. This reagent stream was sent to the reactor which is heated at 360° C. After 3 hours of operation, the sandbath temperature was adjusted between 380 and 400° C. to obtain iso-prenol conversion levels between 45 and 60%. At an iso-prenol conversion of 50%, a prenal and iso-prenal selectivity of 92% was obtained. The results are depicted in table 1 below.

Example 3 (not inventive)

[0177] A “shell-catalyst”, as described in EP 263385 B1, comprising 5 wt.-% of silver coated on a spherical steatite carrier (1.8-2.2 mm), was placed inside the above described reactor to obtain a catalyst bed length of 7 cm. 110 g/h of iso-prenol was evaporated and mixed with 50 NL/h of air. This reagent stream was sent to the reactor which is heated at 360° C. After 3 hours of operation, the sandbath temperature was adjusted between 380 and 400° C. to obtain iso-prenol conversion levels between 45 and 60%. At an iso-prenol conversion of 50%, a prenal and iso-prenal selectivity of 87.5% was obtained. The results are depicted in table 1 below.

Example 4 (not inventive)

[0178] Fully metallic silver rings (height =3 mm, outer diameter =3 mm, inner diameter=2.5 mm, Sigma-Aldrich, ≥99.99%) were placed inside the reactor described above to obtain a catalyst bed length of 7 cm. The material was initially ordered as a longer tube which was cut to the defined length. 110 g/h of isoprenol was evaporated and mixed with 50 NL/h of air. This reagent stream was send to the reactor which was heated at 360° C. After 3 hours of operation, the sandbath temperature was adjusted between 380 and 400° C. to obtain iso-prenol conversion levels between 45 and 60%. At an iso-prenol conversion of 50%, a prenal and iso-prenal selectivity of 85% was obtained. The results are depicted in table 1 below.

Example 5 (not inventive)

[0179] Fully metallic silver crystals as described in EP 0 244 632 in two different sieving fractions (0-1 mm and 1-2 mm). Such silver crystals have a rather undefined, needle-like, appearance. This material leads to low packing densities (void fraction above 80%) and a broad spread in pressure drop over different tubes. The practical application of this material as catalyst is therefore not desired.

TABLE-US-00001 TABLE 1 Packing density and void fraction of selected materials Packing Size range density Void fraction Catalyst Example (mm) (g/mL) (%) Shell catalyst.sup.1 3 1.8-2.2 1.4 40.8 Silver crystals 5 0.2-1.0 2.1 80.1 Silver crystals 5 1.0-2.0 2.1 80.1 Silver rings.sup.1 4 3.0; 3.0; 2.5.sup.2 1.4 86.5 Silver cylinders.sup.1 2 2.0; 2.8.sup.3 6.1 41.9 Silver shot.sup.1 1 1.0-3.2 6.0 46.2 Silver shot 1 1.5-2.5 6.3 40.3 .sup.1Performance shown in performance examples .sup.2Outer diameter; height; inner diameter .sup.3Diameter; height

[0180] Discussion

[0181] Table 1 lists five different materials of which two have two different sieve size ranges. The ‘shell catalyst’ consists of supported silver on steatite spheres, as in EP 263385 B1. The ‘silver crystals’ are fully metallic particles, as in EP 244632 B1. A person skilled in the art generally refers to this material as electrolytic silver or cathode silver. This material leads to low packing density (<4 g/mL) and has the disadvantage to lead to unsatisfactory pressure drop differences between the individual tubes of a multitubular reactor. The silver rings of example 4 are fully metallic bodies which lead to a low packing density (g/mL). The performance examples demonstrate that, using these silver rings as catalyst, no improvement in selectivity is observed in comparison to the prior art. Silver cylinders and silver shot (mainly round silver bodies) are fully metallic bodies which lead to high packing densities (≥4 g/mL). The performance examples demonstrate that, using silver shot or silver cylinders, a significant improvement in selectivity is observed in comparison to the prior art.

[0182] The table 2 lists the value of the parameters Λ.sub.r, and α.sub.w under typical operation conditions over a shell-type catalyst bed as described in EP 263385 and for a catalyst bed according to the invention.

TABLE-US-00002 TABLE 2 Λ.sub.r in (W/m/K) α.sub.w in (W/m.sup.2/K) Coated - shell type - catalyst 0.561 530 Ag on steatite, 2 mm spheres Fully metallic silver bodies 1.34 1465 2 mm spheres according to the invention