NO139002B - PROCEDURE FOR CATALYTIC HYDROGEN SULFURING OF HYDROCARBON RESIDUES - Google Patents

Description

The present invention relates to a method for catalytic hydrogen desulphurisation of hydrocarbon residual oils with a total vanadium and nickel content above 120 ppmw, in which no continuous or periodic replenishment of the present catalyst in the desulphurisation reactor takes place during operation.

Hydrocarbon residues usually contain a significant amount of sulfur compounds. When these oils are used as fuel, the sulfur present in the sulfur compounds is converted into sulfur dioxide which is released into the atmosphere. In order to limit air pollution as much as possible when burning these oils, it is desirable that the sulfur content be reduced. This can be achieved by catalytic hydrogen desulphurisation of the oils. For this purpose, methods can be used in which continuous or periodic replenishment of the catalyst present in the desulphurization reactor takes place during operation, or methods in which such replenishment does not take place. For the sake of simplicity, the latter processes, of which the present one is an example, will be referred to as "catalytic hydrogen desulphurisation without catalyst post-filling".

The catalytic hydrogen desulphurisation of hydrocarbon residual oils entails certain problems which do not arise when this process is used for the hydrogen desulphurisation of hydrocarbon oil distillates.

These problems are due to the fact that most hydrocarbon residues, such as residues obtained from the distillation of crude oils under atmospheric or reduced pressure, contain high molecular non-distillable compounds such as asphaltenes and metal compounds, especially vanadium and nickel compounds, which metal compounds are to a significant extent bound to the asphaltenes. During the hydrogen desulphurisation, the asphaltenes and the vanadium and nickel compounds tend to deposit on the catalyst particles. A portion of the high molecular weight compounds that are deposited on the catalyst particles are converted into coke. As a result of the increasing concentration of vanadium, nickel and coke on the catalyst, very rapid deactivation of the catalyst occurs. Consequently, as the activity of the catalyst decreases, higher temperatures must be used to maintain the desired degree of desulphurisation. At higher temperatures, however, hydrocracking reactions begin to play a more important role and a fuel oil of varying quality is obtained.

In order to extend the lifetime of the catalyst, it has already been proposed that the asphaltenes should be removed from the starting material before it is desulfurized and that the separated asphaltenes should then be re-mixed with the desulphurised product. However, this method requires a further step, namely deasphaltenisation, and also causes other disadvantages. Therefore, a hydrogen desulphurisation method is preferred according to which the starting material as such, i.e. including the asphaltenes, is processed. However, this embodiment of the process requires catalysts with a better resistance to deactivation than those currently generally recommended for this purpose.

An investigation into catalytic hydrogen desulphurisation without catalyst replenishment of hydrocarbon residues with a total vanadium and nickel content above 120 ppmw has shown that optimal catalysts for this purpose, comprising one or more metals with hydrogenation activity on a support, must fulfill a number of requirements with regard to their particle diameter and porosity. These requirements for optimal catalysts partly depend on the hydrogen partial pressure at which the hydrogen desulphurisation is carried out. In this description, an optimal catalyst for dehydrogen desulphurisation of hydrocarbon residual oils without catalyst refill means a catalyst which has a sufficiently long lifetime, so that an acceptable amount of hydrocarbon residual oils can be desulphurised per kg of catalyst before the catalyst shows rapid deactivation, and which catalyst has a sufficiently high average activity so that the end product with the desired low sulfur content is obtained at an acceptable space velocity.

It has been found that for catalytic hydrogen desulfurization without catalyst replenishment of hydrocarbon residues with a total vanadium and nickel content above 120 ppmw, the requirements with regard to particle diameter and porosity that an optimal catalyst comprising one or more metals with hydrogenation activity on a support must meet are as follows. First of all, the catalyst particles must have such a specific mean pore diameter (p) and specific mean

0 9

particle diameter (d) that the quotient p/(d) ' meets the requirement 45 x 10~ x(<p>H2P/(d)°'<9> < 9 x 10~4 x(PH2)2, where PH2 is the applied hydrogen partial pressure (p in nm, d in mm, PH2 in bar). Furthermore, the catalyst must have a pore volume above 0.45 ml/g with at least 0.4 ml/g of the pore volume in the pores with a diameter of at least 0.7 x p, and a maximum 1.7 x p. Finally, the catalyst must have a sharp pore diameter distribution which is characterized by

(a) less than 20% of the pore volume in the pores with a diameter

less than 0.7 x p,

(b) less than 20% of the pore volume in the pores with a diameter

greater than 1.7 x p, and

(c) less than 10% of the pore volume in the pores with a diameter greater than 100 nm.

The way in which d is determined depends on the shape of the catalyst particles. If they are of such a shape that the particle diameter distribution of the catalyst can be determined by sieve analysis, d is determined as follows. After a complete sieve analysis of a representative catalyst sample has been carried out, d is read from a curve where for each successive sieve fraction the weight percentage has been set, based on the total weight of the catalyst sample, cumulatively as a function of the linear mean particle diameter of the relevant sieve fraction, where d is the particle diameter corresponding to 50% of the total weight. This method can be used to determine d for spherical and granular materials and materials with a similar shape such as extrudates and pellets with a length/diameter ratio between 0.9 and 1.1. The determination of d for extrudates and pellets with a length/diameter ratio less than 0.9 or greater than 1.1 and similar cylindrically shaped materials, for which the particle diameter distribution cannot be determined by means of a sieve analysis, is carried out with this as follows . After a complete length distribution analysis (in case the length/diameter ratio is less than 0.9) or after a complete diameter distribution analysis (in case the length/diameter ratio is greater than 1.1) of a representative catalyst sample has been performed, d is read from a curve where, for each successive length and diameter fraction, the weight percentage has been set based on the total weight of the catalyst sample, cumulatively as a function of the linear mean size of the relevant fraction, d being the value corresponding to 50% of the total weight .

After a determination of the complete pore diameter distribution of a catalyst sample has been carried out, p is read from a curve where, for the pore diameter range 0 - 100 nm, for each successive pore volume increase less than or equal to 10% of the pore volume, the quotient of the pore volume increase is set and the corresponding pore diameter interval cumulatively as a function of >the linear mean pore diameter over the relevant pore diameter interval, where p is the pore diameter corresponding to 50% of the td/tal quotient at 100 nm.

A determination of the complete pore diameter distribution of the catalyst can very conveniently be carried out by the nitrogen adsorption/desorption method (as described by E.V. Ballou and O.K. Doolen in Analytical Chemistry 32,; 532 (1960)) combined with the mercury permeation method (as described by H.L. Ritter and L.C. Drake 1 Industrial and Engineering Chemistry, Analytical Edition 17, 787 (1945)), where mercury pressures of 1 - 2000 bar are used. The pore diameter distribution of the catalyst in the pore diameter range below and including 7.5 nm is preferably calculated from the nitrogen desorption isotherm (where cylindrical pores are assumed) according to the method described by J. C. P. Broekhoff and J. H. de Boer in Journal of Catalysis 10, 377 (1968) and the pore diameter distribution of the catalyst in the pcrc diameter range above 7.5 nm is preferably calculated using the formula: r

The above-described catalysts for which the quotient p/(d)<0>,<9> fulfills the requirement 4.0 < p/(d)<0>,<9>4 20.0 have already been described as new compositions in British Patent Application No. 56803/71.

The present invention relates to a method for catalytic hydrogen desulphurisation without catalyst replenishment of hydrocarbon residues with a total vanadium and nickel content above 120 ppmw, which method is carried out at a hydrogen partial pressure (Ph2) of 50 "129 °9 137 "200 Dar' and a temperature of 300 - 475 ° C, a space velocity of 0.1 - 10 parts by weight of fresh starting material per volume fraction of catalyst per hour, and a hydrogen/starting material ratio of 150 - 2000 NI H2/kg starting material, over a per se known catalyst comprising 0.5

- 20 parts by weight of nickel and/or cobalt and 2.5 - 60 parts by weight of molybdenum and/or tungsten per 100 parts by weight of alumina or silica-alumina support, which catalyst has a total pore volume of more than 0.45 ml/g, less than 10% of the pore volume is present in pores with a diameter greater than 100 nm, a specific mean particle diameter (d) of 0.5 - 2.5 Eim and a specific mean pore diameter (p, expressed in nm), which method is characterized by the catalyst having such a pore diameter distribution that

(a) the quotient p/(d) 0 ' 9 meets the requirements

(b) at least 0.4 ml/g of the total pore volume is present in pores

with a diameter of at least 0.7 x p, and a maximum of 1.7 x p,

(c) less than 20% of the total pore volume is present in pores

with a diameter less than 0.7 x p, and

(d) less than 20% of the total pore volume is present in pores with a diameter greater than 1.7 x p.

Norwegian patent document 137 208 discloses a method for catalytic hydrogen desulphurisation of residual hydrocarbon oils which have a total vanadium and nickel content greater than 120 ppm on a weight basis using a catalyst containing one or more metals from group VIB, VIIB and/or VIII in the periodic table.

However, these catalysts do not meet the pore diameter distribution requirements set for the catalysts used according to the present method.

At standard hydrogen desulphurisation conditions, catalysts used according to the present method showed a better initial activity, and a better average activity than catalysts according to Norwegian patent 137 208 with practically the same P/(d)' but which, however, do not meet the present pore diameter distribution requirements.

The catalysts used are characterized by a given ratio between their mean pore diameter and mean particle diameter at a given hydrogen partial pressure and by a given pore diameter distribution within pore diameter ranges that depend on the mean pore diameter of the catalysts. It is important that the average pore diameter and average particle diameter used for characterizing the catalysts have been determined according to the methods previously indicated for the specific average pore diameter and the specific average particle diameter. If an average pore diameter or an average particle diameter determined by methods different from those described here for the specific average pore diameter and the specific average particle diameter (e.g.

a mean pore diameter calculated as four times the quotient of the pore volume and the surface area or a mean particle diameter calculated as the linear mean) completely different results can be obtained.

The found relationship between p, d and PH2 can serve three different purposes. Primarily, this relationship will make it possible to determine the range within which PH^ should be chosen in order to achieve good results with a catalyst with a given p and d.

Furthermore, the ratio can be used to determine the area within

i■-

which d for a catalyst material with a given p should be chosen to achieve good results at a given Ph2- Finally, this relationship makes it possible to determine the range within which p of a catalyst with a given d should be chosen to achieve good results at a given PH . The most suitable values for Pj^' ^ and p within the areas found are determined, among other things, by the composition of the hydrocarbon oil to be desulphurised.

The catalysts used in the method according to the invention preferably have a surface area of more than 50 m 2 /g and most preferably a surface area of more than 100 m <2>/g.

The catalysts used comprise, as mentioned, preferably 0.5 - 20 parts by weight and preferably 0.5 - 10 parts by weight of nickel and/or cobalt and 2.5 - 60 parts by weight, preferably 2.5 - 30 parts by weight of molybdenum and/or tungsten per 100 parts by weight carrier. The atomic ratio between nickel and/or cobalt on the one hand and molybdenum and/or tungsten on the other can vary within wide limits, but is preferably between 0.1 and 5. Examples of very suitable metal combinations for the present catalysts are nickel/tungsten, nickel/molybdenum, cobalt/molybdenum and nickel/cobalt/molybdenum. The metals can be present on the carrier in metallic form or in the form of their oxides or sulphides. Preferred

catalysts are those where the metals are present on the support in the form of their sulphides. Sulphiding of the catalysts can be carried out by any of the known methods for sulphiding catalysts. The sulphidation can e.g. is carried out by bringing the catalyst into contact with a sulphur-containing gas such as a mixture

of hydrogen bg hydrogen sulfide, a mixture of hydrogen and carbon disulfide, or a mixture of hydrogen and a mercaptan, such as butyl mercaptan. Sulfidation can also be carried out by bringing the catalyst into contact with hydrogen and with a sulphur-containing hydrocarbon oil, such as a sulphur-containing kerosene or gas oil. Very suitable carriers for the present catalysts are oxides, such as aluminum oxide and silicon oxide-aluminum oxide.

Production of the catalysts used according to the invention can be carried out by depositing the indicated metals on a carrier with such a pore diameter distribution and specific average pore diameter that after depositing the metals on this a catalyst is obtained which meets the requirements according to the invention, either as such or after the specific medium catalyst particle diameter has been increased or decreased.

The porosity of the finished catalyst depends to a certain extent on the amount of metal used. -In general, it can be said that starting from a support with a given porosity, a finished catalyst with lower porosity is obtained the greater the amount of metal used. This phenomenon only plays a minor role if relatively low amounts of metal are used, i.e. amounts of approx. 20 parts by weight of metal or less per 100 parts by weight carrier. This means that with relatively low amounts of metal, the porosity of the finished catalyst is mainly determined by the porosity of the carrier used and for the production of catalysts according to the invention with relatively low amounts of metal, carriers should be chosen with a porosity that deviates only slightly from the desired porosity of the finished catalyst. With higher amounts of metal, however, the influence of the amount of metal on the porosity will become more important and the use of a high amount of metal can serve as a means of producing catalysts according to the invention by starting from supports whose porosity is too high. The porosity of a carrier can also be influenced by means of a high temperature treatment either in the presence or in the absence of water vapour.

The porosity of a support is mainly determined by the way in which the support is produced. Catalyst supports of the metal oxide type are usually prepared by adding to one or more aqueous solutions of the salts of the metals in question one or more gelatinizing agents, which results in the metals being precipitated in the form of metal hydroxide gels, which are then formed and calcined. Before the metal hydroxide gels are formed, they are usually allowed to age for a certain time. During the production of the carrier, there are great opportunities to influence the porosity of the carrier. The porosity of the final carrier depends, among other factors, on the rate of addition of the gelatinizing agents, and on the temperature and pH used during the formation of the gel. The porosity of the finished carrier can also be influenced by the addition of certain chemicals to the gel, such as phosphorus and/or halogen compounds. If aging is used, the porosity of the finished carrier also depends on the aging time and the temperature and pH used during aging. In the production of mixed metal oxide supports, a further important point with regard to the porosity of the finished support is the way in which the metal hydroxide gels are precipitated: simultaneously or separately, e.g. one on top of the other. The porosity of the finished support further depends on the way in which the support particles are shaped, on the conditions used to facilitate the shaping, and the temperature used during the calcination.

During forming, the porosity of the carrier particles is influenced, e.g. by the type and amount of peptizing agents and binders that are usually added during this step of the carrier preparation, by the addition of certain chemicals and by the addition of small amounts of inert materials such as silicon oxide and/or zirconium oxide. If the carrier particles are formed by extrusion, the porosity of the finished carrier is influenced by the applied extrusion pressure. If use is made of the spray drying technique, the porosity of the support is influenced by the applied spray temperature and the applied spray pressure.

The used catalysts according to the invention can be produced according to any of the methods known for the production of catalysts with multi-component supports. It is not necessary for the catalytically active metals to be deposited on a finished carrier, they can also be incorporated into the carrier material during its production, e.g. before forming. Incorporation of the catalytically active metals into the support at an early stage of the catalyst preparation can also have a strong influence on the porosity of the finished catalyst. The used catalysts according to the invention are preferably produced by simple or multi-stage co-impregnation of a carrier with an aqueous solution comprising one or more nickel and/or cobalt compounds and one or more molybdenum and/or tungsten compounds, followed by drying and calcination. If the impregnation is carried out in several stages, the material can be dried and calcined if desired, between the successive impregnation stages. Drying and calcination are preferably carried out at temperatures between 50 and 150° C and between 150 and 550° C. Examples of suitable water-soluble compounds of nickel, cobalt molybdenum and tungsten which can be used in the production of the present catalysts are nitrates, chlorides, carbonates, formates and acetates of nickel and cobalt, ammonium molybdate and ammonium tungsten. To increase the solubility of these compounds and to stabilize the solutions, certain compounds such as ammonium hydroxide, monoethanolamine and sorbitol can be added to the solutions.

Aluminum oxides and silicon oxides-aluminum oxides are preferred as carriers for the catalysts used according to the invention. Very suitable carriers are aluminum oxide particles produced by spray-drying an aluminum oxide gel, followed by shaping the spray-dried microparticles into larger particles, e.g. by extrusion, and spherical aluminum oxide particles obtained by the well-known plje-drip method. The latter method involves the formation of an aluminum oxide hydrosol, which is combined with a suitable gelatinizing agent

after which the mixture is dispersed as droplets in an oil maintained at an elevated temperature. The droplets are allowed to remain in the oil until they are solidified into spherical hydrogel particles, which are then isolated, washed, dried and calcined. Very suitable silicon oxide-alumina carriers for the present catalysts are cogels of aluminura hydroxide gels on silicon oxide hydrogel. These cogels are preferably first prepared by precipitating a silicon oxide hydrogel from an aqueous, silica-containing solution by adding a mineral acid, after which a water-soluble aluminum salt is added to the mixture, and the aluminum hydroxide gel is then precipitated by adding an alkali-reactive compound. It is preferred that the produced silicon oxide hydrogel is allowed to age for a certain time at elevated temperature

temperature before cogel production continues. The aging conditions, especially the aging time and the aging temperature, have a strong influence on the porosity of the finished cogel. Production of catalysts used according to the invention based on the above-mentioned cogels can e.g. performed as follows. First of all, the cogel is formed, e.g. by extrusion, followed by drying and calcination of the hydrogel particles. The xerogel particles thus obtained are then neutralized with a nitrogen base and dried. Finally, the catalytically active metals are deposited on the carrier by impregnation of this with one or more aqueous solutions comprising salts of the metals in question, followed by drying and calcination of the material. The production of catalysts used according to the invention based on the above-mentioned cogels can also be carried out by the following simplified procedure. The catalytically active metals are incorporated into the cogel by mixing this with one or more aqueous solutions comprising salts of the metals in question, after which the material

shaped, e.g. by extrusion, dried and calcined.

The catalytic hydrogen desulphurisation of hydrocarbon residual oils without catalyst replenishment is preferably carried out by passing the hydrocarbon oil at elevated temperature and pressure and in the presence of hydrogen in an upstream, downstream or radial direction through one or more vertically arranged reactors comprising one or more stationary catalyst beds. The hydrocarbon oil to be desulphurised can be completely or partially saturated with hydrogen and in addition to the hydrocarbon phase and the catalyst phase, a hydrogen-containing gas phase can be present in the reactor. Furthermore, part of the liquid product, which may or may not contain dissolved hydrogen, hydrogen sulphide and/or hydrocarbon gases, can be recycled to the catalyst mixture.

The hydrogen desulphurisation can be carried out either in a single reactor or two or more reactors. As a rule, hydrogen desulphurisation reactors contain more than one catalyst bed. The catalysts used in the separate catalyst beds and/or the separate 1 reactors may differ from one another with respect to their p and/or d and/or chemical composition. If several reactors are used, it is possible to use all these reactors simultaneously when carrying out the desulphurisation reaction. It is also possible to use the reactors alternately for desulphurisation, in which case the desulphurisation is carried out in one or more reactors while the catalyst is replaced in the other reactors.

Hydrogen desulphurisation of hydrocarbon residual oils without catalyst filling can very suitably be carried out by passing the hydrocarbon oil together with hydrogen through a vertically arranged stationary catalyst bed in the upstream direction, where the applied liquid and/or gas velocity is such that an expansion of the catalyst bed takes place . Another advantageous embodiment of hydrogen desulphurisation without catalyst replenishment of hydrocarbon residues is one in which the hydrocarbon oil is passed together with hydrogen through a vertically arranged stationary bed in the upstream direction and in which the adiabatic rise in temperature resulting from the hydrogen desulphurisation reaction is kept below 20° C by recirculating a part of the desulphurised product to the catalyst layer and/or inject hydrogen at different places in the catalyst layer.

In catalytic hydrogen desulphurisation of hydrocarbon residues without catalyst refilling, catalyst particles with a specific average particle diameter of 0.5 - 2.5 mm are usually used. If the desulphurisation is carried out by passing the hydrocarbon oil to be desulphurised together with hydrogen through a vertically arranged stationary catalyst bed in the upstream or downstream direction, it usually makes use of catalyst particles with a specific mean particle diameter of 0.6 - 2.0 mm.

The found relationship between p and d and PH makes it possible to produce catalysts for hydrogen desulphurisation of hydrocarbon residues at a given Ph2' which show optimal performance. If a catalyst or catalyst support is available for which d is not optimal in relation to p at a given pH H it is possible to prepare from it an optimal catalyst or catalyst support by adjusting d to p. This can be done simply by increasing or decreasing the particle size of the catalyst or catalyst support (eg by binding the particles with or without the use of the binder or by grinding the particles).

When producing optimal desulphurisation catalysts by starting from a specific catalyst or catalyst carrier, the following problem can arise. The optimum d at a given PH^ corresponding to p in the material from which the catalyst must be produced can be so small that difficulties arise when such small catalyst particles are used for catalytic hydrogen desulphurisation. In this case, it is preferred that agglomerates are formed from the small particles, for which d is optimal in relation to p and Ph , which agglomerates have more than 10% of their pore volume in the pores with a diameter above 100 nm. It is preferred that agglomerates are formed from the small particles with more than 25% of their pore volume in the pores with a diameter above 100 nm. The use of these porous catalyst agglomerates in the hydrogen desulphurisation of hydrocarbon residual oils provides the same advantages as the use of the less optimal catalyst particles without the inherent disadvantages of using these small catalyst particles. The porous catalyst or catalyst carrier agglomerates can very suitably be produced from small optimal particles by binding them together with or without the use of a binder in the presence of a material which is incorporated into the agglomerates and then removed by evaporation, combustion, dissolution, leaching or otherwise, leaving sufficient pores with a diameter above 100 nm in the agglomerates. Suitable compounds for this purpose are cellulose-containing materials, polymers and compounds soluble in organic or inorganic solvents.

The reaction conditions used in the hydrogen desulphurisation process according to the invention can be varied widely. The hydrogen desulphurisation is preferably carried out at a temperature of 300 - 475° C, a hydrogen partial pressure of 50 - 12009 and 137 - 200 bar, a space velocity of 0.1 - 10 parts by weight of fresh feed per volume fraction of catalyst per hour and a hydrogen/feed ratio of 150 - 2000 NI H2/kg feed. Particularly preferred is a temperature of 3 50 -

445° C, a hydrogen partial pressure of 80 - 129 and 137 - 180 bar, a space velocity of 0.5 - 5 parts by weight fresh feed per volume fraction of catalyst per hour and a hydrogen/feed ratio of 250 - 1000 NI H2/kg feed.

The present invention is limited to the use of catalysts that meet certain requirements with regard to the quotient p/(d) 0 ' 9 and the pore diameter distribution, in hydrogen desulphurisation without catalyst replenishment of hydrocarbon residual oils with a vanadium and nickel content above 120 ppmw. Application of these catalysts by. hydrogen desulphurisation without catalyst replenishment of hydrocarbon residues with a total vanadium and nickel content of a maximum of 12 0 ppmw is described in the applicant's Norwegian patent application 3832/71

Examples of feeds that can be used in the hydrogen desulphurisation process according to the invention are crude oils and residues obtained from the distillation of crude oils at atmospheric pressure or reduced pressure. It is preferred that the feed to be desulphurised should contain less than 50 ppmw and in particular less than 25 ppmw alkali metal and/or alkaline earth metal. If the alkali metal and/or alkaline earth metal content in the feed is too high, this can be reduced, e.g. by desalination of the feed.

The invention is further illustrated by the following examples.

Example 1

Catalyst production

Catalyst A

A catalyst comprising 4.7 parts by weight of cobalt and 11.4 parts by weight of molybdenum per 100 parts by weight of aluminum oxide were prepared as follows: An aqueous solution of 1051 g of ammonium molybdate (the ammonium molybdate so.a used in the examples had a molybdenum content of 54.3% by weight) was mixed with an aqueous solution of 1163 g of cobalt nitrate 6 aq. After adding 350 ml of 25% ammonia, the mixture was diluted with water to a volume of 3800 ml. This mixture was used to impregnate 5000 g of 1.5 mm aluminum oxyextrudates. After 15 min, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst B

A catalyst comprising 4.7 parts by weight of cobalt and 11.4 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 35.1 g of ammonium molybdate was mixed with an aqueous solution of 38.9 g of cobalt nitrate 6 aq. After adding 10 ml of 25% ammonia, the mixture was diluted with water to a volume of 125 ml. This mixture was used to impregnate 166.9 g 1.5 ml aluminum oxyextrudates. After 15 min, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst C

A catalyst comprising 4.7 parts by weight of cobalt and 11.4 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 29.9 g of ammonium molybdate was mixed with an aqueous solution of 33.1 g of cobalt nitrate 6 aq. After adding 10 ml of 25% ammonia, the mixture was diluted with water to a volume of 108 ml. This mixture was used to impregnate 14 2.3

g 1.5 mm aluminum oxide extrudates. After 15 min, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalysts D and E

Two catalysts comprising 3.8 parts by weight of cobalt and 9.5 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows:

An aqueous solution of 876 g of ammonium molybdate and 1000 ml

30% H2°2 was mixed with an aqueous solution of 940 g cobalt nitrate 6 aq. After the mixture was diluted with water to a volume of 3450 ml, it was used to impregnate 5000 g of 0.8 mm aluminum oxide dextrodates. After 30 min, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours. Part of the catalyst D thus obtained was crushed to produce a catalyst E with a d of 0.2 mm.

Catalysts F and G

A quantity of 23.2 kg of an aqueous solution containing 5.256 kg of sodium silicate (SiC^ content: 26.5% by weight) was heated to 40° C. The pH of the solution was lowered from 11.1 to 6 by adding 2200 ml of 6 N HNO-j during 30 min with stirring. The silica gel obtained was aged for 24 hours at 40° C. An amount of 2448 g of an aqueous solution containing 1528 g of Al(NO-j)3.9 1^0 and at a temperature of 40° C was added to the mixture during 5 min while stirring. After a further 10 minutes of stirring, the pH of the mixture was increased to 4.8 by adding 25% ammonia. After 10 min, the pH of the mixture was further increased to 5.5 (total ammonia consumption was approx. 900 ml). The silica-alumina cogel was filtered off and washed with water until sodium-free. The gel was extruded into 1.5 mm extrudates. The extrudates were dried at 120°C and calcined at 500°C.

A quantity of 620 g of this silicon oxide aluminum oxide cogel was neutralized with 6.2 liters of 0.1 molar Nf^NO^ solution and 15 ml of 25% ammonia. The silica-alumina was filtered off, washed with water and dried at 120°C.

The above-mentioned silicon oxide-aluminum oxide (95% by weight dry matter) was used as a carrier for two catalysts comprising 2 parts by weight of nickel and 16 parts by weight of molybdenum per 100 parts by weight of silicon-aluminum oxide. The catalysts were prepared as follows: An aqueous solution of 37.7 g of nickel formate 2 aq was mixed with an aqueous solution of 176.4 g of ammonium molybdate. After adding 110 ml of monoethanolamine, the mixture was diluted with water to a volume of 700 ml. This mixture was used to impregnate 630 g of the above-mentioned silicon oxide-aluminium oxide (598.5 g dry material). After 15 min, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours. Part of the catalyst F thus obtained was crushed to produce a catalyst G with a d of 0.5 mm.

Catalyst H

An amount of 116.25 kg of an aqueous solution containing 2G.25 kg of sodium silicate (SiO2 content: 26.5% by weight) was heated to 40° C. The pH of the solution was lowered to 6 by the addition of 6N HNO3 during 30 min while stirring. The silicon oxide gel obtained was aged for 140 hours at 40° C. with stirring. A quantity of 30 liters of an aqueous solution containing 7.66 kg (Al(N03)3.9H20) and with a temperature of 40° C was added to the mixture during 5 min with stirring. After another 10 minutes of stirring, the pH of the mixture was increased to 4.8 by adding 25% ammonia. After 10 minutes the pH of the mixture was further increased to 5.5. The silicon oxide-aluminum oxide cogel was isolated by centrifugation and washed until it was sodium-free. This cogel (12.4 % dry material) was used as a carrier for a catalyst comprising 2 parts by weight of nickel and 16 parts by weight of molybdenum per 100 parts by weight of silicon oxide-aluminium oxide.

The catalyst was prepared as follows:

An amount of 1008 g of silicon oxide aluminum oxide cogel

(= 125 g dry material) was kneaded for 10 minutes. Then it was 7.87

g nickel formate 2 aq added and the material was again kneaded for 5 minutes. A solution of 36.83 g of ammonium molybdate in a small amount of water and H 2 O 2 (molar ratio t^C^^O 1 0.25) was added and the mixture was kneaded for 1 hour. The product was extruded into 1.2 mm extrudates. The extrudates were dried at 120°C for 18 hours and calcined at 500°C for 3 hours.

Catalyst I

A catalyst comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 11.9 g of molybdenum as ammonium molybdate and 7.1 g of H 2 O 2 (molar ratio H 2 O 2 /Mo = 0.5) was mixed with an aqueous solution containing 4.73 g of nickel as nickel nitrate. After the mixture was diluted with water to a volume of 110 ml, it was used to impregnate 110 g of aluminum oxide. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours. The aluminum oxide that was used as a carrier was obtained by extrusion of a spray-dried aluminum oxide. The aluminum oxyextrudates had the following properties: pore volume: 0.68 ml/g,

pore volume in pores with a diameter ^ 0.7 x p and ^!'7 x P: °/56 ml/g,

surface area: 250 m 2/g,

Catalyst J

A catalyst comprising 4.3 parts by weight of cobalt and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 38.6 g of ammonium molybdate and 28.4 ml of 30% H 2 O 9 was mixed with an aqueous solution of 30.3 g of cobalt nitrate 6 aq. After the mixture was diluted with water to a volume of 103 ml, it was used to impregnate 142.3 g of the same aluminum oxide extrudates as those used in the preparation of catalyst I. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C for 3 hours.

Catalyst K

A catalyst comprising 4.3 parts by weight of cobalt and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 60.22 g of ammonium molybdate and 19.3 g of 30% H 2 O 2 was mixed with an aqueous solution of 63.7 g of cobalt nitrate (Co content: 20.25% by weight) . After the mixture was diluted with water to a volume of 240 ml, it was used to impregnate 300 g of aluminum oxide extrudates obtained by extruding a spray-dried aluminum oxide. After 20 minutes, the impregnated material was dried at 120°C for 18 hours and calcined at 500°C

for 3 hours.

Catalyst L

A catalyst comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 30.1 g of ammonium molybdate and 7.5 ml of 30% H 2 O 2 was mixed with an aqueous solution of 31.9 g of nickel nitrate 6 aq. After the mixture was diluted with water to a volume of 200 ml, it was used to impregnate 150 g of spherical aluminum oxide particles obtained by the oil drip method. One. 15 minutes, the impregnated material was dried at 120°C for 18 hours and calcined at 500°C for 3 hours. The spherical aluminum oxide particles had the following properties:

pore volume: 0.80 ml/g

pprevolume in pores with a diameter > 0.7 xp and <: 1.7 xp: 0.58 ml/g surface area: 230 m <2>/g

Catalyst M

A catalyst comprising 4.3 parts by weight of cobalt and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 20.0 g of ammonium molybdate and 20 g of 30% H 2 O 2 were mixed with an aqueous solution of 21.2 g of cobalt nitrate 6 aq. After the mixture was diluted with water to i a volume of 110 ml was used to impregnate 100 g of them

in the same spherical aluminum oxide particles as those used in the production of catalyst L. After 15 minutes, the impregnated material was dried at 120°C for 18 hours and calcined at 500°C

for 3 hours.

Catalysts N and 0

Two catalysts comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows:

An aqueous solution of 57.2 g of ammonium molybdate was mixed

with an aqueous solution of 38.6 g of nickel formate 2 aq. After addition of 65 ml of monoethanolamine, the mixture was diluted with water to a volume of 280 ml. This mixture was divided into two equal portions and

each of these portions was used to impregnate 142.5 g of spherical aluminum oxide particles which had been steam treated at 475°C

and then calcined either at 500° C or at 700° C for 3 hours.

The spherical aluminum oxide particles were the same as those used in the production of catalysts L and M. In the production of catalyst N, the aluminum oxide particles that had been calcined at 700° C were used, and the aluminum oxide particles that had been calcined at 500° C were used as carrier for catalyst O. After 15 minutes, the impregnated materials were dried at 120° C. for 13 hours and calcined at 500° C. for 3 hours.

Catalyst P

A catalyst comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 12.0 g of ammonium molybdate was mixed with an aqueous solution of 8.1 g of nickel formate 2 aq. After addition of 13.5 ml of monoethanolamine, the mixture was diluted with water to a volume of 60 ml. This mixture was used to impregnate 60 g of the same steam-treated spherical aluminum oxide particles that were used in the production of catalyst N. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst Q

An amount of 11,600 g of an aqueous solution containing 2,628 g of sodium silicate (SiO 2 content: 26.5% by weight) was heated to 40° C. The pH of the solution was lowered from 11.6 to 6 by the addition of 1,160 ml of 6 N HN0 3 in the course of 30 minutes with stirring. Gelatin ring occurred at pH = 10.5. The resulting silica gel was aged for 140 hours at 40° C. An amount of 1224 g of an aqueous solution containing 780 g of Al(NO 3 ) 3 .9H 2 O was added to the mixture over 5 minutes with stirring. The pH of the mixture was increased to 4.8 by adding 4 35 ml of concentrated ammonia over 20 minutes. After 20 minutes, the pH of the mixture was further increased to 5.5 by adding 20 ml of concentrated ammonia. The silica-alumina cogel was isolated by centrifugation and washed six times, each time with 15 liters of water until sodium-free. The gel was extruded into 1.5 mm extrudates. The extrudates were dried at 120°C for 18 hours and calcined at 500°C for 3 hours.

A quantity of 377 g of this silica-alumina cogel was mixed with 3770 ml of 0.1 molar NH 4 NO 3 . The pH of this mixture was increased from 4.6 to 7 by the addition of 7.8 ml of concentrated ammonia. After 2 hours, the silicon oxide-aluminium oxide was filtered off, washed with 2 liters of water and dried at 100° C.

The above-mentioned silicon oxide-aluminum oxide (94.5% by weight dry material) was used as a carrier for a catalyst comprising 2 parts by weight of nickel and 16 parts by weight of molybdenum per 100 parts by weight silicon oxide-aluminium oxide. The catalyst was prepared as follows:

An aqueous solution of 23 g of nickel formate 2 aq was mixed

with an aqueous solution of 108 g of ammonium molybdate. After addition of 70 ml of monoethanolamine, the mixture was diluted with water to a volume of 500 ml. This mixture was used to impregnate 386 g of the above-mentioned silicon oxide-aluminium oxide (365 g dry material). After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst R

A catalyst comprising 1 part by weight of nickel and 8 parts by weight of molybdenum per 100 parts by weight of silica-alumina was prepared as follows: An aqueous solution of 11.6 g of nickel formate 2 aq was mixed with an aqueous solution of 54.4 g of ammonium mol<y>bdate. After addition of 45 ml of monoethanolamine, the mixture was diluted with water to a volume of 500 ml. This mixture was used to impregnate 388 g of the same silicon oxide-alumina oxide that was used in the preparation of catalyst Q. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst S

A catalyst comprising 4.3 parts by weight of cobalt and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution containing 15.25 g of molybdenum as ammonium molybdate and 0.5 mol of I^C^/ved Mo was mixed with an aqueous solution

ning containing 6.02 g of cobalt as cobalt nitrate. After mixing

was diluted: with water to a volume of 130 ml, it was used to impregnate 140 g of aluminum oxide. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalysts T and U

A quantity of 116.25 kg of an aqueous solution containing

26.25 kg of sodium silicate (SiO 2 content: 26.5% by weight) was heated to 40° C. The pH of the solution was lowered to 6 by adding 6 N NH 0 3 during 30 minutes after stirring. The silica gel obtained was aged for 140 hours at 40° C. with stirring. A quantity of 30 litres

of an aqueous solution containing 7.66 kg Al(N03)3.9 H2O and at a temperature of 40° C. was added to the mixture over 5 minutes with stirring. After further stirring for 10 minutes, the pH of the mixture was increased to 4.8 by adding 25% ammonia. After 10 minutes, the pH of the mixture was further increased to 5.5. The silicon oxide-aluminum oxide cogel was isolated by centrifugation and washed with water at 60°C until it was sodium-free.

A quantity of 4.6 kg (14% dry matter by weight) of this silica-alumina cogel was treated three times with 5 liters of 0.1 molar NH 4 NO 3 , washed with water and filtered off.

The above-mentioned silicon oxide-aluminum oxide (12.3% by weight dry material) was used as a carrier for two catalysts each comprising 1 part by weight of nickel and 8 parts by weight of molybdenum per 100 parts by weight of silicon oxide-aluminum oxide. The catalysts were prepared as follows:

An amount of 3661 g of silicon oxide aluminum oxide cogel

(= 450 g dry material) was kneaded for 10 minutes. Then 14.15 g of nickel formate 2 aq was added and the material was again kneaded for 5 minutes. An aqueous solution of 66.3 g of ammonium molybdate was added and the mixture was kneaded for 1 hour. From this product the catalysts T and U were obtained by extrusion into 1.3 and 1.6 mm extrudates. The extrudates were dried at 100° C and calcined at 500° C for 3 hours.

Catalyst V

A silica gel was prepared from a solution of 1500 g of sodium silicate (SiO2 content: 26.5% by weight) in 5000 ml of water by

lower the pH of the solution from 11.6 to 6.0 by adding 6 N HN03. The silica gel was aged for 24 hours at 100° C. A solution of 4 37.2 g of Al(NO 3 ) 3 .9H 2 O in 900 ml of water was added to the silica gel. By adding 300 ml of concentrated ammonia, the pH of the mixture was increased from 3.15 to 6. The silicon oxide-aluminium oxide cogel thus obtained was filtered off and washed with water until it was sodium-free.

The gel was dried at 120°C, calcined for 3 hours at 500°C and crushed into particles with a d of 0.7C mm.

The pH of a mixture of 99 g of the above silica-alumina particles and 1000 ml of 0.1 molar NH 4 NO 3 was increased from 4.5 to 7.0 by the addition of concentrated ammonia. The silicon oxide-aluminum oxide was filtered off and dried at 120° C. This silicon oxide-aluminum oxide (95.6% by weight dry material) was used as a carrier as a catalyst comprising 1 part by weight of nickel and 8 parts by weight of molybdenum per 100 parts by weight of silicon oxide-aluminum oxide. Kata-

the lysates were prepared as follows:

An aqueous solution of 13.7 g of ammonium molybdate and monoethanolamine was mixed with an aqueous solution of 2.95 g of nickel formate 2 aq. After the mixture was diluted with water to a volume of 175 ml, it was used to impregnate 98 g of the above-mentioned neutralized silicon oxide-aluminum oxide. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst W

A catalyst comprising 3.8 parts by weight of cobalt and 9.5 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 876 g of ammonium molybdate and 1000 ml of 30% H2O2 was mixed with an aqueous solution of 940 g of cobalt nitrate 6 aq. After the mixture was diluted with water to a volume of 3500 ml, it was used to impregnate 5000 g of 1.5 mm aluminum oxide dextrodates. After 30 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst X ■$'

A catalyst comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 2.01 kg of ammonium molybdate was mixed with an aqueous solution of 2.13 kg of nickel nitrate 6 aq. After adding 6 liters of 25% ammonia, the mixture was diluted with water to a volume of 11 liters. This mixture was used to impregnate 10 kg of aluminum oxide. After 15 minutes, the impregnated material was dried at 120° C. for 18 hours and calcined at 500° C. for 3 hours.

Catalyst Y

A catalyst comprising 4.3 parts by weight of nickel and 10.9 parts by weight of molybdenum per 100 parts by weight of aluminum oxide was prepared as follows: An aqueous solution of 44.13 g of ammonium molybdate containing monoethanolamine was mixed with an aqueous solution of 30.74 g of nickel formate 2 aq containing monoethanolamine. After the mixture was diluted with water to a volume of 190 ml, it was used to impregnate 220 g of aluminum oxide. After 15 minutes, the impregnated material was dried at 120°C for 18 hours and calcined at 500°C

for 3 hours.

Hydrodesulfurization experiments

A hydrocarbon residual oil with a total vanadium and nickel content of 245 ppmw, a C₁ asphaltene content of 7.2% by weight and a sulfur content of 2.1% by weight, which oil was obtained as a residue from the distillation at atmospheric pressure of a Caribbean crude oil, was catalytically hydrodesulfurized without catalyst replenishment and using catalysts A — Y. The oil together with hydrogen was passed through a cylindrical stationary catalyst bed in the downstream direction at a temperature of 420° C, a hydrogen partial pressure varying from 40 to 200 bar, an outlet gas velocity of 250 Nl/kg fresh feed and a space velocity of 4.35 kg oil/kg catalyst per hour. The catalysts were used in the form of their sulfides.

The performance of a catalyst in the hydrodesulfurization of hydrocarbon residual oils without catalyst replenishment can be described by means of the catalyst's lifetime and average activity (^i^g^)' which is defined as follows: The catalyst's lifetime (expressed in kg feed per kg of catalyst) is the maximum amount of residual oil which can be hydrodesulfurized over the catalyst before the catalyst shows a rapid deactivation.

The average activity (expressed in kg feed/kg catalyst, hour (wt% S) is the activity of the catalyst at the point where half of the catalyst's lifetime is reached.

The results of the hydrodesulfurization experiments together with the properties of the catalysts used are listed in Table I. The specific mean pore diameter (p) of the catalysts was calculated from a complete pore diameter distribution, which was determined by means of the nitrogen adsorption/desorption method in combination with the* mercury penetration method as previously described.

The criteria for a catalyst to be optimal under the conditions used in these hydrogen desulphurisation trials are: the lifetime of the catalyst must be at least 2500 kg feed/kg catalyst and the average activity must be at least 1.5 kg feed/kg catalyst

0 5

sator. hour (%w S) ' .

Experiments 1 - 9 and 11 - 24 (in which the catalysts have a lifetime greater than 2500 and a k^^g^ greater than 1.5) are hydrogen desulphurization experiments according to the present invention. In these experiments, the catalysts used fulfilled the requirements with regard to p/(d) 0 ' 9 and the pore diameter distribution of an optimal catalyst.

Experiments 25 - 35 in which the catalysts had a lifetime of less than 2500 and/or a ^j^gi of less than 1.5) are hydrogen desulphurization experiments outside the scope of the present invention. In these experiments, the catalysts used failed to fulfill at least one of the requirements regarding p/(d) 0 ' 9 and the pore diameter distribution of an optimal catalyst. Trial 10 lies outside the specified PH range.

2

In experiments 35 - 43, the catalysts used did not fulfill the requirement: 45 x 10~<5> x (P„ )2 < p/(d)<0>'<9> < 9 x 10"4(P„ )< 2>.

H2 H2

Catalyst W used in experiment 32 had less than

0.4 ml/g of the pore volume in pores with a diameter of at least 0.7 x p and a maximum of 1.7 x p and more than 20% of the pore volume in pores with a diameter greater than 1.7 x p.

Catalyst X used in experiments 33 and 34 had . less than 0.4 ml/g of the pore volume in pores with a diameter of at least 0.7 x p and a maximum of 1.7 x p, more than 20% of the pore volume in pores with a diameter greater than 1.7 x p and more than 10% of the pore volume in pores with a diameter greater than 100 nm.

Catalyst Y used in experiment 35 has less than 0.4 ml/g of the pore volume in pores with a diameter of at least 0.7 x p and a maximum of 1.7 x p, more than 20% of the pore volume in pores with a diameter less than 0, 7 x p and more than 10% of the pore volume in pores with a diameter greater than 100 nm.

The influence of on the catalyst's performance is clear when the following experiments are compared: Experiments 1 and 25 with catalyst A >

Experiments 4, 5 and 6 with catalyst C

Trials 6 and 2 7 with catalyst D

Experiments 8 and 28 with catalyst F

Trials-17 and 29 with catalyst N

Experiments 22 and 30 with catalyst S

Catalysts A, C, D, F, N and S which met the pore diameter distribution requirements of an optimal catalyst and which showed an optimal performance in the hydrodesulfurization of residual oil at a PH2 that meets the requirement

(trials 1, 4, 5, 6, 8, 17 and 22), is less suitable for this purpose if one does not meet the requirement

(experiments f25 - 30).

Example 2?

Using the nitrogen adsorption/desorption method combined with the ^mercury penetration method, a complete pore radius distribution of two Ni/Mo/Al2O3 catalysts (catalyst I and II) was determined. Tables II and III show the percentage of the pore volume that exists in pores with a given pore radius.

Of the catalysts I and II, the mean pore diameters were determined according to three different methods, each used in practice.

Method 1: Calculated using the formula

,.. 4 x pore volume .-3

pore diameter = £—=-r— x. 10

c surface ; -

Method 2: Readings from a curve set up using a complete pore radius distribution and where, for the pore diameter range from 0 to 100 nm, for each pore volume increase less than or equal to 10% of the pore volume, the quotient of the pore volume increase and the corresponding pore diameter interval is set as one

function of the linear mean pore diameter over the relevant pore diameter interval, the pore diameter being read at the point where the curve reaches a maximum.

Method 3: Reading from a curve set up using a complete pore radius distribution and where, for the pore diameter range from 0 to 100 nm, for each pore volume increase less than or equal to 10% of the pore volume, the quotient of the pore volume increase and the corresponding pore diameter interval is accumulated cumulatively as a function of the linear mean pore diameter over the relevant pore diameter interval, the pore diameter being read at the point corresponding to 50% of the total quotient at 100 nm.

Por catalyst I, the following mean pore diameter values are found by the different methods:

By method 1: 13.1 nm.

By method 2: Undeterminable as there was no distinct maximum on

Basket.

By method 3: 7.6 nm.

For catalyst II, the following mean pore diameter values were found by the different methods:

By method li 18.9 nm.

By method 2: 16.0 nm.

By method 3'in 13.6 nm..

As can be seen from this example, the different methods for determining the average pore diameter of the catalyst lead to widely varying results. Method 3 was used to determine the specific mean pore diameter (p) according to the invention.

Example 3

A complete sieve analysis was performed on a crushed Ni/Mo/Al 2 O 3 catalyst with a particle diameter between 0.115 and 1.10. The results of this sieve analysis are shown in table IV.

For this catalyst, the average particle diameter was determined according to different methods, both of which are used in practice.

Method 1: Calculated as a linear mean using the formula

dl + d2

particle diameter = ^ ' where d-^ and d2 represent the particle diameter of the largest and smallest particle.

Method 2: Reading from a curve set up by means of a complete sieve analysis and where for each successive sieve fraction a weight percentage has been set aside, based on the total weight of the catalyst sample, cumulatively as a function of the linear mean diameter of the relevant sieve fraction, the particle diameter is read at the point corresponding to 50% of the total weight.

For this catalyst, the following mean particle diameter values were found by the various methods.

For method 1: 0.61 mm.

For method 2: 0.225 mm.

It is clear from this example that the different methods for determining the average particle diameter of the catalyst lead to widely varying results. Method 2 was used to determine the specific mean particle diameter (d) according to the invention.

Claims (1)

Process for catalytic hydrogen desulphurisation without catalyst refilling of hydrocarbon oils with a total vanadium and nickel content above 120 ppmw, which process is carried out at a hydrogen partial pressure (Pu) of 50 - 129 and 137 - 200 bar, a temperature of 300 - 475°C, a 2-room rate of 0.1 - 10 parts by weight of fresh starting material per volume fraction of catalyst per ti-me, and a hydrogen/starting material ratio of 150 - 2000 Ni H2/kg starting material, over a per se known catalyst comprising 0.5 - 20 parts by weight nickel and/or cobalt and 2.5 - 60 parts by weight molybdenum and/or tungsten per 100 parts by weight of alumina or silica-alumina support, which catalyst has a total pore volume of more than 0.45 ml/g, less than 10% of the pore volume is present in pores with a diameter greater than 100 nm, a specific mean particle diameter (d) of 0.5 - 2.5 mm and a specific mean pore diameter (p, expressed in nm), characterized in that the catalyst has such a pore diameter distribution that (a) the quotient p/(d) 0 ' 9 meets the requirements (b) at least 0, 4 ml/g of the total pore volume is present in pores with a diameter of at least 0.7 x p, and a maximum of 1.7 x p, (c) less than 20% of the total pore volume is present in pores with a diameter less than 0 .7 x p, and (d) less than 20% of the total pore volume is present in pores with a diameter greater than 1.7 x p.