WO1990008592A1

WO1990008592A1 - Catalyst

Info

Publication number: WO1990008592A1
Application number: PCT/GB1990/000130
Authority: WO
Inventors: George Edwin Harrison; Alan James Dennis; Geoffrey Webb
Original assignee: Davy Mckee (London) Limited
Priority date: 1989-01-30
Filing date: 1990-01-29
Publication date: 1990-08-09
Also published as: EP0455661A1; KR910700098A; CA2045664A1; GB8901977D0; CN1044601A; JPH04503027A

Abstract

A reduced supported Group VIII metal catalyst of enhanced activity is obtained by an ante-pre-reduction treatment in which a supported Group VIII metal catalyst precursor is soaked in a reducing atmosphere at temperatures below a pre-reduction temperature (typically about 140°C) at which appreciable pre-reduction of the catalyst can be detected. This catalyst may be, for example, a ruthenium on alumina catalyst and is characterised by a Group VIII metal surface area of at least about 45 m2/g of Group VIII metal present in the catalyst, as measured by adsorption of carbon monoxide thereon.

Description

CATALYST

This invention relates to catalysts. In particular it relates to a Group VIII metal catalyst.

According to the present invention there is provided a catalyst comprising a supported reduced Group VIII metal comprising a support and up to about 10 percent by weight of a Group VIII metal which is characterised by a Group VIII metal surface area of at least about 45 m²/g of Group VIII metal present in the catalyst as measured by adsorption of carbon monoxide thereon.

In measuring the Group VIII metal surface area we have assumed that the area occupied by an adsorbed carbon monoxide molecule is 16.8 x 10^~20 m² (16.8 A²).

In contrast to the catalysts of the present invention, which are characterised by a Group VIII metal surface area of at least about 45 m²/g of Group VIII metal present in the catalyst, we have found that a supported Group VIII metal catalyst prepared from the same catalyst precursor by a conventional pre-reduction technique has an exposed Group VIII metal surface area that is significantly lower than about 45 m²/g of Group VIII metal present in the catalyst.

The literature suggests that the catalytic activity of Group VIII metal hydrogenation catalysts is due to the reduced metal particles present following pre- reduction. Hence it follows that the activity of the catalyst will tend to bear a more or less direct relationship to the exposed surface area of reduced metal. Thus the larger is the surface area of exposed reduced metal, the greater will be the activity of the catalyst. In the limit the Group VIII metal is in the form of individual metal atoms on the support.

In preferred forms of catalyst in accordance with the invention substantially all of the Group VIII metal content thereof is present as particles of reduced metal. The catalyst precursors contain a Group VIII metal oxide or salt and a support. Suitable precursors can be obtained by impregnation of a support with a solution of a Group VIII metal salt, followed if desired by ignition. Coprecipitation can be used in suitable cases. Methods of making suitable catalyst precursors are well known to those skilled in the art of catalyst manufacture. Such catalyst precursors consist, it is believed, of crystallites of a reducible form of the Group VIII metal more or less uniformly distributed over the support. The size of these crystallites will depend upon the conditions prevailing during preparation of the catalyst precursor. The smaller that such crystallites are, the larger the surface area of metal in the reduced catalyst can be, it is postulated. Thus the method of manufacture of the catalyst precursor will have an influence on the final surface area of Group VIII metal in the reduced catalyst, as will also the method of catalyst reduction used.

The Group VIII metal may be any of those conventionally used in production of supported catalysts, especially those used for production of Group VIII metal hydrogenation catalysts. Examples include platinum, palladium, rhodium and ruthenium.

Any suitable support can be used, for example alumina, silica-alumina, thoria, silicon carbide, titania, chromia, zirconia, or carbon. The Group VIII metal content of the catalyst precursor will usually range from about 0.01% by weight up to about 10% by weight, typically up to about 5% by weight of the catalyst precursor, e.g. about 0.5% by weight.

The catalyst precursor is typically in the form of a powder having a particle size of not more than about 100 μ . Such a powder may be formed by conventional techniques into any conventional catalyst shape, such as cylindrical pellets, rings, saddles or the like using the usual binders, and die lubricants, so that the material can be used in fixed bed operations.

The invention also provides a process for producing a catalyst in which a supported Group VIII metal catalyst precursor is subjected to a pre-reduction treatmen by heating in a hydrogen containing atmosphere at a pre- reduction temperature at which appreciable pre-reduction of the catalyst can be detected characterised in that, prior to effecting said pre-reduction treatment, the catalyst precursor is subjected to an ante-pre-reduction treatment by soaking it under hydrogen starvation conditions in an atmosphere comprising a major amount of an inert gas and a minor amount only of hydrogen at a temperature below said pre-reduction temperature. Conveniently heating to said pre-reduction temperature from ambient temperature is effected in said atmosphere. Hydrogen starvation conditions are maintained throughout the ante-pre-reduction step of the process of the present invention, that is to say the catalyst precursor is always starved of hydrogen so that the rate of reduction of Group VIII metal oxide or salt is limited by the availability of hydrogen at the catalyst precursor surface. In this way the rate of reduction to Group VIII metal is conducted at a controlled rate. This procedure differs from procedures conventionally recommended by manufacturers of such catalyst precursors in that heating from ambient temperature up to about 140°C is effected throughout in the presence of a gas mixture containing only a minor amount of a reducing gas rather than in the conventional manner in a 100% hydrogen atmosphere. It is not known exactly what mechanism may be involved in production of an active Group VIII metal catalyst from the catalyst precursor but it would appear that the mechanism involves reduction of at least a portion of the Group VIII metal oxide or salt present on the support to the corresponding Group VIII metal. It would appear that, although no discernible reaction can be detected between the catalyst precursor and the hydrogen containing gas at temperatures below the pre- reduction temperature (which is typically about 140°C) , yet some miniscule amounts of Group VIII metal oxide or salt are in fact reduced whereby sub-microscopic nucleation of Group VIII metal atoms occurs in a manner analogous to the sub- microscopic nucleation of silver atoms that occurs upon exposure of a photographic film in a camera. Although the latent image cannot be detected visually in an exposed photographic film yet it can be rendered visible as a result of the conventional development process by exposure to a reducing agent. In an analogous fashion, it is postulated, the ante-pre—reduction step of the present invention produces a "latent image" consisting of numerous sub- microscopic nucleations of Group VIII metal atoms which can grow individually to form a large number of small particles of Group VIII metal, thus ensuring that the resulting reduced Group VIII metal catalyst has a correspondingly large exposed surface area of Group VIII metal. On the other hand, if conventional pre-reduction techniques are used, so that rapid pre-heating to temperatures of about 180°C and higher in the presence of a hydrogen-containing gas is used, or if the catalyst precursor is pre-heated to a temperature of at least about 180°C in an inert gas atmosphere prior to contact with a hydrogen-containing gas, the first relatively few nucleations of Group VIII metal atoms that form serve as a focus for subsequent reduction of Group VIII metal oxide or salt to metal, with the result that relatively large crystallites of Group VIII metal may form, thus resulting in a lower exposed surface area of Group VIII metal and in a lower catalyst activity in hydrogenation reactions.

By adopting a suitable temperature-time profile and monitoring the inlet and exit gas compositions to and from the pre-reduction zone, it can be ensured that any reactions involved in the ante-pre-reduction step or soaking step occur always at the lowest possible temperature and are permitted to occur as completely as possible before the temperature is again raised significantly. In addition any heat produced as a result of exothermic ante-pre-reduction reactions is removed by the hydrogen containing gas with minimum risk of thermal damage to the catalyst.

The process of the invention involves use of a gas soaking step. In this gas soaking step no liquid is present but the gas is allowed to permeate fully the catalyst precursor and to equilibriate therewith.

In the gas soaking step of a preferred process of the invention the Group VIII metal catalyst precursor is maintained in a hydrogen containing atmosphere at temperatures intermediate ambient temperature (e.g. about 15°C to about 25°C, typically about 20°C) and the pre- reduction temperature (which is typically about 180°C) . Gas soaking can be commenced at temperatures below ambient temperature, e.g. 0°C or below. In this gas soaking step the hydrogen containing atmosphere typically contains hydrogen in a minor amount only, typically not more than about 1% by volume and preferably 0.5% by volume or less, in addition to an inert gas, such as nitrogen, helium or argon. Although it is preferred to heat the catalyst precursor during the gas soaking step from ambient temperature to the pre-reduction temperature throughout in a hydrogen containing atmosphere, it is alternatively possible to commence heating in an inert gas atmosphere, and to introduce the hydrogen containing atmosphere at ambient temperature (if heating from below ambient temperature is taking place) or at a moderately elevated temperature (e.g. about 40°C to about 50°C) . It is, however, an essential feature of the process that, the nearer the temperature during the gas soaking step reaches the pre-reduction temperature, the more important it is that the catalyst precursor be always in contact with a hydrogen containing gas atmosphere but with the catalyst precursor still under hydrogen starvation conditions.

In the gas soaking step of a preferred process of the invention the catalyst precursor is heated under controlled conditions from ambient temperature (e.g. about 20°C) in a stream of a hydrogen containing gas at a suitable gaseous hourly space velocity, e.g. about 500 hr^-1 to about 6000 hr^-1. The hydrogen containing gas preferably comprises a mixture of a minor amount of hydrogen (typically less than about 1% by volume) and a major amount of one or more inert gases, such as nitrogen, argon, neon, methane, ethane, butane, or a mixture of two or more thereof. In a particularly preferred process the reducing gas is a mixture of a minor amount of hydrogen (preferably less than about 0.5% by volume of hydrogen, e.g. about 0.2% by volume) and a major amount of nitrogen, preferably substantially oxygen- free nitrogen.

The gas soaking step of the process of the invention may be operated at normal or reduced pressure but is preferably operated at an elevated pressure in the range of from about 1 bar to about 20 bar, preferably from about 2 bar to about 10 bar.

The partial pressure of the hydrogen need be no more than about 0.01 bar, and can be in the range of from about 0.0005 bar up to about 0.005 bar, during the gas soaking step.

In a particularly preferred process of the invention the catalyst precursor is heated from ambient temperature to about 140°C in a reducing atmosphere containing a minor amount of hydrogen. Preliminary heating of the catalyst precursor from ambient temperature to about 140°C is preferably effected at a controlled rate; typicall this preliminary heating step takes from about 12 hours to about 48 hours or more, e.g. about 24 hours. The temperature can be increased at a substantially linear rate during the soaking step or can be increased in an approximately stepwise fashion, in steps of, for example about 5°C to about 10°C, followed by periods during which the temperature is maintained substantially constant before the temperature is raised again. Over the range of from about 140°C to about 180°C heating may follow any temperature-time curve, provided that the rate of heating is such that at all times the catalyst precursor is maintained under reducing conditions with the inlet and exit gas compositions to the pre-reduction zone being substantially identical one to another. Preferably the temperature is increased in an approximately linear fashion from about 140°C to about 180°C. In one procedure heating is carried out in a series of steps, conveniently steps of approximately 10°C, and a careful check of the inlet and exit gas compositions to the pre-reduction zone is made before, during and after each heating step. Under typical operating conditions the rate of increase of temperature over the temperature range from about 140°C to about 180°C is from about l°C/hour up to about 15°C/hour, e.g. about 10°C/hour.

In this heating step from about 140°C to about 180°C the gas flow rate generally corresponds to a gaseous hourly space velocity (measured at 0°C and 1 bar) of from about 400 hr^-1 to about 6000 hr^"1 or more, e.g. about 3000 hr^"1.

The composition of the hydrogen containing gas is dependent upon the operating pressure; the higher the total pressure is, the lower is the maximum permitted hydrogen concentration. Conversely, the lower the total pressure is the higher can be the concentration of hydrogen in the reducing gas. Typically the H₂ concentration is from about 0.01% v/v up to about 1% v/v, e.g. about 0.2% v/v, under preferred operating conditions.

Once the catalyst precursor has reached the final temperature of about 180°C the hydrogen partial pressure is gradually increased. However, during this phase of catalyst activation the inlet and exit gas compositions to the pre- reduction zone should still be closely monitored so that the two compositions are substantially identical at all times. Further heating up to about 210°C or more can then be carried out if desired.

It is important to ensure that, when the catalyst precursor reaches the pre-reduction temperature, there should not be a substantial excess of hydrogen present so as to minimise any danger of damage to the catalyst resulting from a thermal runaway due to the exothermic catalyst pre- reduction step.

The pre-reduced catalyst produced in accordance with the teachings of the invention is sensitive to oxidation, probably due to some re-oxidation of Group VIII metal particles. Hence the pre-reduced catalyst is preferably maintained under an inert gas or hydrogen containing atmosphere after it has been prepared.

The catalysts of the invention can be used in a wide variety of reactions for hydrogenation of an unsaturated organic compound to produce at least one hydrogenation product thereof.

The invention is further illustrated in the following Examples. Examples 1 and 2

Two samples of PG 88/10 0.5% Ru on alumina catalyst precursor (obtainable from Davy McKee (London) Limited of Davy House, 68 Hammersmith Road, London, 14 8YW) in the form of 3.0 x 3.0 mm pellets were in each case loaded into a reaction chamber. The amount of catalyst precursor was between 2.5 g and 2.6 g in each case and was carefully weighed. Oxygen-free nitrogen was passed through the reaction chamber at an exit pressure of 4.45 bar and a flow rate of 200 litres/hour (measured at 0°C and 1 bar) at room temperature (20°C) for 30 minutes. Then hydrogen was admitted to the nitrogen flow to give a 0.2% hydrogen in nitrogen flow at the same total flow rate. The temperature of the reaction chamber was gradually raised at 5°C per hour to 140°C, the 0.2% hydrogen in nitrogen flow being maintained at the same flow rate. The inlet and outlet gases were continuously monitored by thermal conductivity. The hydrogen level in the gas flow was then gradually increased in small steps to 1% over approximately 24 hours. When no further uptake of hydrogen could be detected, the temperature was increased at 15°C per hour to 160°C and maintained at this temperature until the inlet gas composition was the same as the outlet gas composition, thus indicating that hydrogen uptake had ceased. The catalyst temperature was then increased over a period of 2 hours to 180°C. Next the hydrogen level was gradually increased to 100% with the catalyst temperature still at 180°C and the system conditions maintained for a period of 18 hours. When a 100% hydrogen gas flow had been established the temperature is increased to 200°C and maintained at this level for 1 hour before the catalyst is used.

After this reduction procedure, the catalyst was cooled in an oxygen free helium flow to 20°C, and the ruthenium surface area was determined by reaction of the reduced ruthenium surface with carbon monoxide. Assuming that one molecule of carbon monoxide occupies 16.8 x 10~²^ nr (16.8 A^Λ) of surface, the extent of reaction, and hence the number of carbon monoxide molecules adsorbed on the ruthenium surface, was determined by successive injections of carbon monoxide into the helium flow until no further reaction was detected. In this way an area of exposed metallic ruthenium in the reduced catalyst was calculated.

The following uptakes and the corresponding Ru - 10 -

metal surface areas were recorded:-

TABLE 1

Comparative Examples A and B

Comparative runs using the same catalyst precursor as was used in Examples 1 and 2 were carried out in the following manner.

The reaction chamber was charged in each case with a further 3ml sample of the catalyst precursor pellets. The catalyst precursor was heated to 210°C over a period of 1.5 hours in a 100% hydrogen stream at an exit pressure of 4.45 bar and a gaseous hourly space velocity of 3000 hr^-1 and maintained at this temperature for 30 minutes before characterisation. using the carbon monoxide reaction technique of Examples I and 2 the following results were obtained:-

TABLE 2

Comparison of these results shows that the gas soaking step of Examples 1 and 2 leads to a greatly increased ruthenium surface area compared with that obtained in the corresponding one of Comparative Examples A and B. Examples 3 to 6

In these Examples a sample of catalyst was in each case loaded into the reactor of Example 1 and subjected to the following steps:

1. 0 -free nitrogen at 4.45 bar was passed through the reactor at room temperature at a gaseous hourly space velocity of 1800 hr^"1; this gaseous hourly space velocity was maintained throughout the experiment.

2. 0.2% H₂ in N₂ was flowed over the catalyst at 20°C for 24 hours.

3. The temperature was raised at 5°C per minute to 140°C whilst monitoring the inlet and outlet gas compositions.

4. When the inlet and outlet gas compositions were equal the H₂ content was slowly raised to 1% over 24 hours.

5. The temperature was raised to 160°C at 0.25°C per minute and held at 160°C for 4 hours, after which the inlet and outlet compositions were equal.

6. The H₂ content was increased slowly to 5% over 4 hours and the gas flow was maintained at this concentration for a further 4 hours.

7. The temperature was raised to 180°C at 0.25°C per minute and maintained at this value for a further 4 hours.

8. The temperature was further increased to 200°C at 0.25°C per minute and then the H₂ content was gradually increased to 100% over the next 6 to 8 hours.

9. The 100% H₂ gas was maintained at 200°C for a minimum of a further 8 hours.

10. The system was flushed with 100% N₂ at 200°C and then cooled to 25°C under ₂ for subsequent adsorption studies.

The catalyst samples were then characterised by their ability to chemisorb carbon monoxide irreversibly in a pulsed flow dynamic mode, using the following experimental technique: 11. The system was flushed at 25°C with pure helium at a gaseous hourly space velocity of 1800 hr^-1.

12. A 500 μl pulse of 5% carbon monoxide in helium was injected into the flowing gas stream and the amount of carbon monoxide eluted from the reactor measured by a thermal conductivity detector. The amount of CO in the pulse adsorbed by the catalyst sample was then calculated as the difference between the area of the peak obtained from a standard 500μl 5% CO in the pulse and that of the eluted peak following adsorption.

13. Step 12 was repeated until it was evident that no further adsorption was occurring. From the results obtained the total amount of CO which was irreversily adsorbed was determined.

Since the dynamic pulsed flow mode of adsorption only gives the amount of irreversibly adsorbed carbon monoxide, it is important to determine the total amount of carbon monoxide adsorbed in order to determine the metal area of the catalyst. This was achieved using a radiotracer method to determine the ¹ C carbon monoxide adsorption isotherm under static conditions at 25°C, i.e. at the same temperature as was used in the dynamic pulsed flow experiments. The fraction of CO irreversibly adsorbed was determined by flushing the catalyst sample with a flow of helium and determining the decrease in surface count rate due to removal of the reversibly adsorbed species. With all samples examined the amount of irreversibly adsorbed CO represented 50 -2% of the total adsorption capacity of the catalyst.

The areas of exposed metal in the reduced catalyst were then determined from the absolute amounts of carbon monoxide irreversibly adsorbed using the dynamic pulsed flow procedure described above. In each case the amount of irreversibly adsorbed carbon monoxide was assumed to be 50% of the total adsorbed carbon monoxide capacity of the catalyst. Assuming that the carbon monoxide is adsorbed in a linear form, which is confirmed by infra red spectra, and that the area of the adsorbed carbon monoxide molecule is 16.8 x 10^"20 m² (16.8 A²), the total area of the metal was calculated.

Four portions of PG 88/10 0.5% Ru on alumina catalyst precursor taken from two different batches were used in Examples 3 to 6; these batches are identified as samples A and B. Two portions, those of Examples 3 and 5, were from samples A and B respectively in unreduced form; those of the other two Examples 4 and 6 were portions from samples A and B respectively which had been subjected to a conventional pre-reduction technique followed by exposure to air. The results obtained are set out in Table 3 below.

TABLE 3

It will be seen by comparison of the results for Examples 3 and 4, which are identical apart from the fact that sample A had been already pre-reduced, that the method by which the initial reduction step is performed has a critical influence on the exposed surface area of metal in the reduced catalyst. Thus conventional pre-reduction apparently leads to some aggregation of reducible ruthenium-containing crystallites in the catalyst precursor, whilst the method of the invention does not produce such aggregation. Similar results are observed from Examples 5 and 6, which are identical apart from the fact that the catalyst of Example 6 had been subjected to a conventional pre-reduction step. (Examples 3 and 5 demonstrate catalysts falling within the ambit of the invention but Examples 4 and 6 do not) . Comparative Examples C to G

In this Comparative Example further portions of catalyst samples A and B were subjected to pre-reduction by a conventional technique in the reactor of Example 1. This pre-reduction regime consisted of the following steps:-

1. Hydrogen at 4.45 bar was passed through the reactor at 25°C at a gaseous hourly space velocity of 1800 hr^"1.

2. The temperature was increased, while maintaining the hydrogen flow rate, at 10°C per minute to 200°C.

3. The catalyst was maintained for a minimum of 11 hours under these conditions.

4. The reactor was flushed with 100% N₂ at 200°C and cooled to 25°C whilst maintaining the N₂ flow.

5. The CO adsorption method described in Examples 3 to 6 was then used to determine the metal surface area.

The results obtained are set out in Table 4 below. TABLE 4

It will be noted that, in contrast to the results of Examples 3 to 6, the metal surface area measured following the conventional pre-reduction technique of these Conmparative Examples did not change significantly between the unreduced and pre-reduced samples, whereas using the method of the invention set out in Examples 3 to 6, the metal surface area from the unreduced catalyst precursor was approximately double that from the pre-reduced samples. Example 7

Following the general procedure of Examples 3 to 6 there are treated samples of the following catalysts with similarly good results:

0.1% Pt on alumina

0.1% Rh on alumina

0.1% Pd on alumina

0.1% Ru on carbon

0.1% Pt .on carbon

0.1% Rh on carbon

0.1% Pd on carbon.

Claims

1. A catalyst comprising a promoted reduced supported Group VIII metal comprising a support and up to about 10 percent by weight of a Group VIII metal which is characterised by a Group VIII metal surface area of at least about 45 m²/gram of Group VIII metal present in the catalyst, as measured by adsorption of carbon monoxide thereon.

2. A catalyst according to claim 1, in which substantially all of the Group VIII metal content thereof is present as particles of reduced Group VIII metal.

3. A catalyst according to claim 1 or claim 2, in which the Group VIII metal comprises ruthenium.

4. A catalyst according to any one of claims 1 to 4, in which the support is alumina.

5. A process for producing a catalyst in which a supported Group VIII metal catalyst precursor is subjected to a pre-reduction treatment by heating in a hydrogen containing atmosphere at a pre-reduction temperature at which appreciable pre-reduction of the catalyst can be detected characterised in that, prior to effecting said pre- reduction treatment, the catalyst precursor is subjected to an ante-pre-reduction treatment by soaking it under hydrogen starvation conditions in an atmosphere comprising a major amount of an inert gas and a minor amount only of hydrogen at a temperature below said pre-reduction temperature.

6. A process according to claim 5, in which heating to said pre-reduction temperature from ambient temperature is effected in a hydrogen containing atmosphere.

7. A process according to claim 5 or claim 6, in which the pre-reduction temperature is about 140°C.

8. A process according to any one of claims 5 to 7, in which the precursor is maintained in a hydrogen containing atmosphere at temperatures intermediate ambient temperature and the pre-reduction temperature.

9. A process according to any one of claims 5 to 8, in which in the catalyst precursor is heated under controlled conditions from ambient temperature in a stream of a hydrogen containing gas which comprises a mixture of a minor amount only of hydrogen and a major amount of one or more inert gases.

10. A process according to claim 9, in which the hydrogen containing gas is a substantially oxygen-free mixture of a minor amount of hydrogen and a major amount of nitrogen.

11. A process according to any one of claims 5 to 10, in which the soaking step is operated at a pressure in the range of from about 2 bar to about 10 bar.

12. A process according to any one of claims 5 to 11, in which the partial pressure of the hydrogen is in the range of from about 0.0005 bar up to about 0.005 bar during the soaking step.

13. A process according to any one of claims 5 to 12, in which the catalyst precursor is heated at a controlled rate from ambient temperature to about 140°C in an atmosphere containing a minor amount of hydrogen.

14. A process according to any one of claims 5 to 13, in which the catalyst precursor is heated over the range of from about 140°C to about 180°C so as to follow a temperature-time curve with a rate of heating such that at all times the catalyst precursor is maintained under reducing conditions with the inlet and exit gas compositions to the pre-reduction zone being substantially identical one to another.

15. A process according to any one of claims 5 to 14, in which in heating the catalyst precursor from about 120°C to about 170°C the gas flow rate corresponds to a gaseous hourly space velocity (measured at 0°C and 1 bar) of from about 400 hr^"1 to about 6000 hr^"1.

16. A process according to any one of claims 5 to 15, in which the catalyst precursor contains ruthenium.

17. A process according to any one of claims 5 to 16, in which the support is alumina.