DE1039168B - Catalyst for reforming light hydrocarbons and for the production of aromatics - Google Patents

Description

GERMAN

Now *
B.

BL

REGISTRATION DATE: May 1, 1953

NOTICE
THE REGISTRATION
AND ISSUE OF THE
EDITORIAL: SEPTEMBER 18, 1958

The invention relates to catalysts containing alumina and platinum or another metal the platinum group included. The new catalysts are based on a special structure that is essentially determined by a special nature of the aluminum oxide from which the Catalyst! Is produced. These catalysts are used to reform light hydrocarbons, to increase their octane number or to use benzene and other selected aromatics, such as Toluene or xylene, to win, suitable.

With the catalytic «! Reforming are called the Starting product serving hydrocarbons at elevated temperatures and pressures, usually in The presence of a noticeable partial pressure of hydrogen treated. There are one or more chemical transformations of the hydrocarbon components of the starting material. With the high-quality In reforming processes, catalysts are used which initiate certain hydrocarbon reactions favor selectively, like the isomerization of normal paraffins to isoparaffins, the dehydrocyclization normal paraffins to naphthenes and aromatics, the dehydration of naphthenes to aromatics, the Isomerization of 5-ring naphthenes to 6-ring naphthenes, which then dehydrate to aromatics can be, as well as the hydrocracking of relatively long-chain paraffins to paraffins of shorter ones Chains. The selection of the catalyst is crucial and extremely difficult, since the equilibria and working conditions that favor the course of certain reactions, are often unfavorable for other reactions. For example, in isomerization, the initial rate becomes of the process favored by high temperatures, but with regard to the location of the In equilibrium, low temperatures are more favorable, while dehydration is more favorable due to high temperatures is favored. The presence of hydrogen favors hydrocracking and tends to occur to suppress dehydration under reforming conditions. The presence λόιι However, hydrogen is necessary to reduce the formation of coke, which poisons the catalyst, by covering the catalyst surface; Hydrogen hydrogenates the unsaturated components, which would otherwise form coke, it also leads to a hydro-splitting of the tarry precursors of the Coke at an early stage of its formation by condensation. Although the hydrocracking is the octane number if improved, it tends to lower the yield due to the formation of non-condensable hydrocarbon gases and contributes to coke formation by generating hydrocarbon fragments, Reforming catalyst

light hydrocarbons
and for the production of aromatics

Applicant:

Engelhard Industries, Inc.,
Newark, NJ (V. St. A.)

Representative: Dr.-Ing. W. Abitz, patent attorney,
Munich 27, Gaußstr. 6th

Claimed priority:
V. St. v. America May 15, 1952

John W. Teter, Chicago, IH.,

Carl D. Keith, Munster, Ind.,

and John L. Gring, Homewood, 111. (V. St. Α .;

have been named as inventors

which can coke. Hydrocracking is particularly undesirable when reforming selected hydrocarbons, such as a C ₆ or C ₇ fraction, which are used to produce benzene or toluene, etc. The invention is directed to catalysts which in the production of high octane motor gasoline both have a high level of activity and give a high yield of liquid products. They are used with particular advantage for the production of selected aromatic hydrocarbons with such yields and costs that enable an economically advantageous separation in commercially available degrees of purity.

An even more difficult problem of the technical implementation of catalytic reforming in the Petroleum industry is developing catalysts that have sufficient service life or resistance to aging have in order to achieve acceptable catalyst costs per space part of the charge. For this it is necessary that the activity of the platinum catalysts only slowly when they are used decreases, since these catalysts generally do not regenerate or revitalize with the simple means that would normally be available to an oil refinery, but that would anyway quite expensive in terms of material and manufacture

809 638 / 3S3

are. The invention aims at platinum catalysts with better aging resistance or life and thus brings an extraordinary value for the production of motor gasoline or aromatics by reforming technical progress.

The invention is based on the surprising finding that the aluminum oxide base of the reforming catalysts, containing alumina and a platinum group metal, a preeminent one Significance for the structural condition and the structure of the catalyst and that the thereby certain state of the microstructure and the structure of the catalyst for high catalytic activity and Selectivity and, in particular, high resistance to aging can be decisive.

Many hundreds of different catalysts of this type have been made and tested and evaluated under reforming conditions. It has been found that these catalysts are extremely sensitive to changes in their preparation and deviations in reforming conditions and charging. It has been found that catalysts of aluminum oxide and platinum or metals of the platinum group with a characteristic structure and a characteristic structure are formed when the platinum aluminum oxide hydrogels of varying composition are incorporated, and that unexpected advantages are achieved when in the hydrogel or the mixture of hydrate phases Alumina trihydrate predominates. The trihydrate phase contains the known hydrate forms gibbsite (7-Al ₂ Oj ₅ -SH ₂ O) and bayerite (Cx-Al ₂ O ₃ -SH ₂ O) (see Weiser, Inorganic Colloid Chemistry, 1935, Vol. 2, p. 90 and 92). It has also been found that a third and hitherto not yet described trihydrate form also appears to be present in the desired precursors or hydrate phases. Moreover, in the hydrate phase, the presence of a small amount of a monohydrate, either in the form of amorphous, gelatinous, hydrated aluminum oxide or in a form which, after drying, corresponds to boehmite (γ- Al ₂ O ₃ .H ₂ O), seems particularly advantageous for the creation of a valuable microstructure and an advantageous structure, measured by the properties of the calcined catalyst.

In the dried state, the precursors of the aluminum oxide base show characteristic crystalline or quasi-crystalline interferences in the X-ray structural analysis and, in contrast to the amorphous character and the small-pored structure of the aluminum oxide hydrates usually used for catalysts, have a large volume of accessible pores in the form of large pores. It is recommended that the aluminum oxide precursors are composed essentially of small crystallites, which are determined by X-ray analysis on samples which have been dried ^{at around HO 0 C.} Crystallites of about 1000 Å units or less are determined.

Starting from here, the invention aims at a reforming catalyst made of aluminum oxide and platinum or other metals of the platinum group and differ from the known Reforming catalysts based on aluminum oxide with a large-pore structure features a high surface area and consists essentially of a mixture of aluminum oxide modifications the y-series is composed. that from a mixture of aluminum oxide hydrate phases arose in which the trihydrates consist essentially predominantly of gibbsite and bayerite. The catalysts according to the invention can be carried out through the sequence of operations described below are produced: An aluminum oxide hydrogel is produced, which is in the form of a highly gelatinous precipitate is formed from a mixture of amorphous gelatinous oxide hydrates consists, which is usually boehmite or a

ίο amorphous looking monohydrate dries. One walks this gel into a mixture. which predominantly contains trihydrate phases of aluminum oxide, which are decisive for the structure and structure of the catalysts according to the invention.

The platinum metal or one containing the metal in question is now added to the mixture of the hydrate phases Component as well as any promoter, e.g. B. a fluoride. The slurry will dried and the dried product calcined at elevated temperature. Usually the crowd will tabletted or pressed into the desired shape before calcination.

The aluminum oxide hydrogel can be formed, for example, by precipitating the gelatinous hydrous oxide from the solution of a soluble aluminum salt such as aluminum chloride. Other soluble forms of aluminum can also be used, e.g. B. aluminum sulfate or sodium aluminate, although the subsequent washing out of the foreign ion in question, e.g. B. the sulfate ion, can sometimes be more difficult than the removal of the chlorine ion. Ammonium hydroxide is a useful means of precipitating the hydrogel from the salt solution. In the gelation, the p _H -control is important; it is expedient to maintain a p _H 8 to 10 If the p _H is too low, the conversion can be delayed. As a secondary treatment in gel formation, the foreign ions that are introduced into the mass, such as chlorine ions, are removed by washing out with water. For example, it is usually advisable to reduce the chlorine ion content to below 0.2% in order to avoid corrosion problems when using the catalyst.
The conversion of the hydrogel into the hydrate system desired as a precursor of the aluminum oxide base can be effected in various ways, e.g. For example, by aging which is held for this purpose several days at a p _H of _the 8 to 10 of the hydrogel, or z. B. by seeding the hydrogel with gibbsite crystals. The transition to the desired system, in which the crystalline trihydrate forms of aluminum oxide predominate, can be roughly checked by visual observation. The translucent hydrogel takes on a decidedly more white and opaque appearance. when the crystalline form grows and causes light scattering. However, it is desirable to monitor the transition by sampling. For this purpose, the sample taken is dried at about 110 ° C., for example, and the distribution of the hydrate phases is determined by X-ray analysis. The determination of the pore volume and the measurement of the surface, for example using the BET method, are also of value. In this way, the production can be standardized for a given starting material and a given process technology and facility and then, if it appears necessary, monitored by analytical random samples.

From the aluminum oxide hydrate system thus prepared, by mixing with platinum or

5 6

other platinum group metals originate from reforming hydrate system in which (prior to CaI activity, either alone or as a mixture) the aluminum oxide hydrates in the form of each other, catalysts of high activity and trihydrates predominate. Establish the percentage of AIu aging resistance. These metals miniumoxide in the form of trihydrates must include more than platinum itself, rhodium, palladium and 50% by weight, advantageously about 65 to 90 percent by weight. These are the platinum metals with cubic, weight percent of the total aluminum oxide surface-centered lattices in contrast to Ruthe systems. The trihydrates can be gibbsite, nium and osmium, which have a hexagonal lattice, bayerite and randomite. Randomite and appear to be of lesser value. For many, the trihydrate is of a crystalline form which is useful, especially when hydrocracking is related to its structure between gibbsite and bayerite. Increasing the octane number is desirable. It has been found that the addition of an acidic promoter, e.g. B. fluorine, catalysts prepared in oxydgel, which is the tri-form of a soluble fluoride. Contains other acidic prohydrate predominantly in the form of gibbsite. B. silica, zirconium oxide, boron oxide, have the best activity and resistance to aging. Molybdenum oxide, chromium oxide, chlorides, sulfates, phos- 15 It has also been found that it seems appropriate to phate and the like, at least about 5, probably 10 to The addition of the platinum and the promoter, e.g. B. 35 weight percent of the aluminum oxide monohydrate fluoride, can be done in any order. The ((AlO) OH), as it is in the original gelatinous fluoride, is in the form of a water-soluble compound oxide precipitation in the usual production, z. B. is contained as ammonium fluoride or hydrofluoric acid, maintain or add, substance, added, from which the foreign component is easily removed by evaporation during subsequent drying of reforming catalysts by precipitation and calcining. The platinum can be added to the solution of an aluminum salt prepared by adding an aqueous solution, it is in the form of an amorphous aluminum oxide of chloroplatinic acid, for example that from the gel. During drying, the gelatinous hydrate system obtained is added and the platinum an- serous aluminum oxide is converted into the monohydrate boehmite on the spot by an aqueous or a form which is precipitated by an X-ray solution of hydrogen sulfide. Another graphically proves to be amorphous. By converting the process of adding platinum, this system is obtained in the trihydrate phase, in which platinum sulfide sol of the desired concentration is obtained with a catalyst whose pore volume is mixed in the alumina hydrate system. This slurry, which is the same as the known reforming catalysts, is now dried, to be precise to a relatively large extent in the form of large, preferably fast. For example, the drying should be present in pores. For example, the known oxide catalysts do not last longer than about 24 hours and, in the case of an extremely high surface area, occur through pores in the size of a p _H of about 6 to 9. Characterized by observing certain precautionary measures, the drying can be accelerated by typical catalysts that are made from 100% boehmite. For example, there are essentially no pores, the example can be the mixture in a drum larger than 50 Å, and their pore distribution is the dryer to dry, or you can filter it beforehand, those of the silica-alumina catalysts to adjust the water content before drying very similar to a 40. On the other hand, catalysts made from alumina systems that contain high or spray drying percentages of trihydrates, to be reduced in a drum dryer, are of considerable importance. Licher part of the pore volume in the range of 100 The dried alumina catalyst mixture to 100E. For example, the spectrum of the pores can then be shown by tabletting or prepressing in 45 vohimens as it can be shaped by identifying the nitrogen tablets or pills. If the adsorption isotherm is determined analytically and if the catalyst is to be present in a finely distributed form, the high-performance catalysts can be ground according to the calcination. With the invention of at least about 0.2 to 5 ccm / g of your pore tableting, it is customary to add a compression lubricating volume, which is about half of the total pore medium, which preferably corresponds to an organic volume, is of an order of magnitude above nature and at the calcination with an acidic 100 Å (see Table I). With regard to the substance-containing gas can be burned out. The CaI distribution of the pore size below about 100 Å is achieved by considering the mass in units of what appears to be little differences which one would find oxygen-containing «! Gas burns off, e.g. B. can be brought in connection with catalytic properties by heating the catalyst for hours in a gas 55.

current at about 399 to 649 ° C. The calcination is The large pores can entin from the trihydrates an air-nitrogen mixture introduced, but in stand, because this is in a non-calcined state Air terminated. Before use, the catalyst will have a crystal size of about 1000 Å units. These reduced by having it in contact with flowing large pores will not be in the course of the calcinie-hydrogen several hours at a similarly high level, but actually already exist Temperature, for example 482 ° C and atmospheric in the non-cacinated alumina. These pressure, treated for 12 hours. large pores, on the other hand, do not exist - neither before nor The catalyst mixtures according to the invention after calcination - contained in the ammonium oxides about 0.1 to 1 percent by weight of platinum or products derived from boehmite or mono-other Platinum group metals or combinations 65 hydrate were formed. Boehmite is characterized the same. By a small size of the crystallites before and after As stated above, it was found that the calcination, on the order of 40 Å-one-die characterize new catalysts that units, and contains essentially no pores the aluminum oxide base from a mixture of over 50 Å. It seems, however, that the presence Modifications of the 7 series consists of a 70 a small amount of boehmite in the starting

I 039 168

mixture is advantageous, so that the large-pored volume apart from the normal fine-pored structure during the Calcinierang is maintained to the maximum extent. So contain those catalysts according to of the invention, which are characterized by larger pore volume, more in their starting mixture than 5 to 25% boehmite.

There was a connection between the aging resistance, which is achieved by a slow Loss of activity in the course of use can be expressed, and the distribution of the pore volume established. It has been found that the superior catalysts according to the invention, at those at least about 0.2 to 0.4 cc / g of their total pore volume in the size range of about 100 to 1000 Å under reforming conditions suffer a much slower loss of activity, both in the production of Motor gasoline as well as flavoring.

In addition to the pore size and the distribution of the pore volume, the nature of the aluminum oxide hydrate pliases used as starting material also has an influence in other ways on the final structure and the state of the catalyst, which seems to be important with regard to the performance of the catalyst. If the monohydrate has crystallite sizes of about 30 to 40 Å and a relatively large surface area of, for example, about 300 m ² / g (as determined by nitrogen adsorption), the crystallite sizes in the non-calcined trihydrate mixture appear for the most part between about 300 to 100OE and the surface areas at only about 150 to 200 m ² / g. After calcination, the crystallite size of the alumina appears to be predominantly in the range from about 35 to 65 Å and the surface area between about 350 and 550 m ² / g. This increase in the surface area of the trihydrate during calcination has a significant influence on the final state of distribution and the crystallite size of the platinum in the catalyst. Considered in context, the dry crystallite size of the trihydrate phases determined before calcination and the distribution of the pore volume after calcination can be an indicator of whether the catalyst base concerned can be used to obtain catalysts of high activity and stability.

It was also determined by X-ray that the platinum crystals in the catalysts according to the invention are small in size, often not appearing at all as defined lines when the platinum is in front the X-ray is not carefully isolated, reduced and concentrated from the wearer. Usually the sizes of the trihydrate-based platinum crystals are on the order of less than 50 Å, although they can be up to 150 Å. One surprising property of finely divided platinum is it to be considerably soluble in strong sulfuric acid. This fact suggests that the active platinum bound in some way rather than in metallic form.

The following examples are the processing conditions for making special catalysts and the determination of their physical and chemical properties and their behavior in the Use explained.

Example 1

11.35 kg aluminum chloride hexahydrate (Mallinckrodts Grade A.R.) were dissolved in 50 l of distilled water, using a wooden container, to prevent contamination. Ammonium hydroxide solution was used as the precipitant. This was obtained by mixing equal parts of the volume of deionized water and ammonium hydroxide with a density of 0.90. This mixture was placed in a container made of difficult-to-melt glass, from which the liquid could be added by siphoning off the aluminum chloride solution.

ίο While the aluminum chloride solution was being vigorously stirred, the solution of ammonium hydroxide was added at such a rate that the Pu of the solution was 8.0 to 8.1 after 27 to 30 minutes. The flow of the ammonium hydroxide solution was stopped at a pjj of about 4.7 because the mass thickened. With continued stirring, the mass first became liquid again, whereupon further ammonium hydroxide solution was added. Stirring was then continued for 30 minutes and the precipitate was separated from the mother liquor by means of frame filter presses. The filter cake from four batches was redispersed in each 26.8 1 deionized water and the Aluminiuimoxydhydrogel washed by repeated slurrying filtered and the filter cake again to give ceased the p _H at 8 before filtration. The aluminum oxide hydrate has now been aged for 14 days in order to convert it into a phase mixture which contains about 50% gibbsite (which contains some randomite), 25% bayerite and 25% boehmite. The material washed out 13 times was designated with the test number 400 E 9078. A slurry of 400E 9078 was made by mixing 35.13 kg of 400E 9078 and 19.41 kg of distilled water.

This slurry was stirred vigorously in a suitable mixer or homogenizer to produce a thorough distribution. With continued stirring, a solution of 45.1 g of ammonium fluoride in 1550 ecm of distilled water was added over the course of 5 minutes. Stirring was continued for 30 minutes, after which a platinum solution was added over the course of 5 minutes, which consisted of 251.5 ecm H ₂ PtCl ₆ (0.043 g Pt per ecm) plus 1150 ecm distilled water. After stirring for a further 5 to 10 minutes, hydrogen sulfide (2510 ecm water saturated with H., S) was slowly added. The slurry was stirred for a further 30 minutes, after which it was poured into bowls made of difficult-to-melt glass and dried in a circulating air dryer at 110 to 115 ° C.

The oven-dry product was ground and, after the addition of a tabletting lubricant, into tablets deformed with a diameter of 3.9 mm. The tablets were now calcined under controlled conditions.

The calcination was done initially in flowing nitrogen and was terminated in oxygen (6 hours. 482 ° C). The initial concentration of oxygen was 0.3 percent by volume; it was maintained until the lubricant was burned. The calcination was completed in 100% flowing oxygen and the catalyst is cooled in a stream of nitrogen. The finished catalyst was with the test number B-2 (400 E 9099).

Radiographically, the aluminum oxide was identified as a mixture of ^ 'modifications with a crystallite size of about 42 Å. The bulk density was 0.71 and the surface area, determined by means of nitrogen, was 411 m ² / g. The chemical analysis showed the following composition: 0.33% Pt, 0.62% F, 0.43% Cl, 0.10% C and 3.22% volatiles.

Example 2

There was prepared a catalyst in composition and manufacture to that according to Example 1 corresponded, but with the proviso that the Aluminiumoxydhydrogel aged for 15 days at room temperature and the p _H was adjusted to 8 before platinum and fluoride were added. The x-ray analysis of the aluminum oxide hydrate system showed: 25% gibbsite, 30% bayerite, 16% randomite and 20% boehmite + amorphous. The catalyst was designated with the test number 400 E 9027.

The aluminum oxide was identified radiographically as a mixture of / modifications with a crystal size of about 48 Å. The bulk density was 0.80 and the means. Surface area determined by nitrogen 463 m ² / g. The chemical analysis showed: 0.29% Pt, 0.72% F, 0.24% Cl and 2.57% volatiles.

Example 3

5.5 1 one molar aluminum chloride solution was filtered and placed in an earthenware vessel with a capacity of 30 1, which was equipped with a powerful air stirrer. With vigorous stirring, 5 kg of a slurry containing gibbsite seeds was added. This slurry contained a colloidal suspension of precipitates. and washed alumina hydrate corresponding to _{0.42% Al 2} O ₃ as gibbsite and 2.02% Al ₂ O _{3 as boehmite.}

The mixed AIuminiumchlorid-alumina hydrate slurry was stirred for 5 minutes, followed by slow 'zusetzte a achtmolare ammonium hydroxide solution, with continued vigorous stirring, was achieved to a p _H of 8.0. The ammonium hydroxide solution, which contained 1350 ecm of concentrated ammonium hydroxide ¹ , which was made up to 2.5 liters, was added at a relatively constant rate of about 75 to 80 ccm / min. The slurry thickened at a p _H of about 5 (determined by _H p paper), was discontinued followed by the addition of ammonium hydroxide until the slurry liquid again. When the precipitation was complete, the slurry was stirred for an additional 30 minutes and then the hydrogel was filtered from the mother liquor using a large cylindrical Buchner funnel.

The hydrogel filter cake was broken up into pieces of about 2.54 cm and then slurried in 20 liters of deionized water by stirring vigorously for about 1 hour. This slurry was then suction filtered using vacuum from the washing liquid, for about 5 minutes has been set, the p _H by the addition of ammonium hydroxide to 8 before this filtration. The hydrogel was washed repeatedly in this way until the alumina hydrate had a chlorine content of less than 0.2%. This required eight washing and nine filtering. The X-ray analysis of a dried sample of the hydrogel (400 E 9159) revealed a hydrate mixture which consisted of 63% gibbsite, 12% bayerite, 9% randomite and about 16% boehmite.

The X-ray crystal size of the trihydrates was about 500 Å.

2548 g of this hydrogel were mixed with 3 liters of distilled water in an earthenware container of one capacity of about 11 1 mixed and stirred vigorously for about 30 minutes until the gel is well distributed was. With vigorous stirring this slurry were 4.76 g of ammonium fluoride, which in 130 ecm distilled water were dissolved, slowly added. Stirring was continued for 30 minutes, whereupon one with constant vigorous stirring 26.5 ecm of chloroplatinic acid solution (0.043 g Pt / ccm), with Diluted 130 ecm of distilled water, added slowly. Others were stirred. 5 minutes and then continued with constant vigorous stirring slowly add 264 ecm of distilled water, which at about 26 ° C with

ίο hydrogen sulfide was saturated (100% H ₂ S excess over the amount calculated for the formation of Pt S _2). This slurry was colored light brown. After stirring for an additional 30 minutes, the slurry was poured into bowls of difficult-to-melt glass and dried overnight in an oven at 110 ° C. The dried catalyst was ground to such an extent that it passed through a sieve with a mesh size of 0.833 mm, mixed with 2% binder and tabletted in 4 mm tabs. The pelletized catalyst would be designated with the test number 400 E 9165. It was calcined for 6 hours at 482.degree. C. by the method of Example 1. The aluminum oxide was identified radiographically as a mixture of / modifications of a crystallite size of about 41 Å. The bulk density was 0.63 and the surface area, determined by means of nitrogen, was 440 m ² / g. The chemical analysis showed: 0.35% Pt, 0.62% F, Ο, 19ό / 8 Cl and 2.70% volatiles.

Example 4

The one used as a basis in this example

Aluminum hydroxide consisted of a mixture of the hydrogels (a) and (b) described below in a ratio of ₃ V (AE) and ^{_2/3 of} (b). The hydrogels were made as follows:

(a) To a vigorously stirred solution of 11.57 kg of AlCl ₃ -OH ₂ O in 51% distilled water, an ammonium hydroxide solution prepared from equal parts of water and ammonium hydroxide solution with a specific gravity of 0.90 was added. By "total of 19.71 of this 1: 1 mixture was brought -Ammoniumhydriixyd the p _H to 8.2: The addition was carried out in the course of 35 minutes: 'The Ammoniumhydroxydzusatz was 5 f wherein p _P" Kürz interrupted, so that the viscous liquid has become slurry was again .

After further stirring, the precipitate was separated off on a frame filter press. The filter cake however, was not too firm, requiring reslurry and re-filtration made, whereupon a firmer cake was won. This latter was in Crushed cubes of about 2.54 dm, placed in a vessel with distilled water and percolated washed. For this purpose, distilled water was added at a rate passed through the cake at a rate of about 76 l / hour. The washing was about 830 hours continued.

(b) A second batch was made using the same process steps as used above. The precipitation was carried out in two approaches. In each were 11.35 kg AlCl ₃ - dissolved 6H ₂ O and 19.7 to 20.0 of an aqueous 1 Ϊ: 1 NHjOH solution treated, ie brought to p _H 8.2 to 8.3. The crushed filter cakes were washed for 688 hours by percolation with 227 liters of water / hour and for 18.9 hours with 113.5 H ₂ O / hour.

809 638/383

X-ray examination of the above samples shows boehmite of small crystallites and small amounts of gibbsite. Therefore 18.88 kg of product were made (a) and 47.98 kg of product (b) and 50.58 kg of distilled water are thoroughly stirred together and referred to as sample no. 400 E 9086.

30minutiges by vigorous stirring 4240 g 400 E were 9086 (173 g Al _{2 O?} Dry weight) were dispersed in 1,350 cc of distilled water. Then an ammonium fluoride solution (3.41 g NH ₄ F and 135 ecm H ₂ O) was added. The addition took about 10 to 15 minutes and was temporarily discontinued when the consistency of the slurry became too thick.

After stirring for a further 30 minutes, a solution of chloroplatinic acid was added (0.613 g Pt, from 14.25 ecm H ₂ Pt Cl _e solution of 0.043 g Pt / ccm, diluted with 100 ecm distilled water). After stirring for 10 minutes, 142 cc of distilled water, which was _{saturated with H 2} S at room temperature, were slowly added. Stirring was continued for 30 minutes after which was transferred to the mass in shells made of hard glass and ¹ dried in a convection oven at 110 ⁰ C. The dried catalyst was tabletted and subjected to controlled calcination by treatment in oxygen at 482 ° C. for 6 hours. The product was designated with the test number 400 E 9113.

Example 5

Below are the results of the radiographic Methods (using a 90 ° X-ray chamber according to Moreloo) in application the aluminum oxide bases of the catalysts according to Examples 1 to 3 are given. In the examined The following aluminum oxide hydrates were found in aluminum oxide preparations:

(1) Boehmite, AlO-OH. This monohydiate exists from small crystallites of about 30 to 50 Å; the roentgenogram contains broad lines.

(2) Bayerite, Al (OH) ₃ . This trihydrate consists of large crystallites (300 to 1200 A) and gives sharp X-ray lines. It is metastable and eventually turns into gibbsite.

(3) Gibbsite, Al (OH) ₃ . This trihydrate also consists of large crystallites and gives sharp X-ray interference. It is the most stable hydrate of alumina.

(4) Randomit. This crystalline form is not reported in the literature. It is presumably a Trihydrate with an intermediate structure between bayerite and gibbsite.

Bayerite gives a line at 2.23 Ä, which can be easily interpreted. Bayerite also has a line closely adjacent at 4.73 a (18-70 ° 20, Cu radiation) a Gibbsite line at 4.85 Å (18 x 25 ° 20, Cu radiation). These three lines have the same intensity. The Gibbsit line at 4.85 Ä can be due to the adjacent Bavarian policy cannot be clearly identified. But the ratio of bayerite to gibbsite can be simple can be determined by taking the area of the Bayerit line of 2.23 Å with the combined areas the Bayerite line at 4.73 Å and the Gibbsite line

ίο 4.85 A compares.

The presence of Randomite, which has a line at about 4.70 Å, introduces further uncertainty a. The intensity of the line at 4.79 A is not known experimentally and so some assumption must be made be made. From the known structure of gibbsite and the assumed: structure of bayerite it can be assumed that the intensity of the 4.79 λ line is equal to the intensities of the aforementioned Bayerite and Gibbsite lines is. Under this assumption, the area of the Bayeritlinie is 2.23 Ä (determined planimetrically) divided by the combined areas of the bayerite maximum at 4.73 A plus des Randomite maximum at 4.79 Å plus the gibbsite maximum at 4.85 Å the ratio of bayerite to Gibbsite randomite. As a result, the Gibbsif concentrations given below include everything present Randomit.

The total area of the three lines (at 4.73, 4.79 and 4.85 Å) is a linear function of the boehmite concentration. So if the total area is zero, the boehmite concentration is 100 ^ / o, and if the Total area has a maximum value, the boehmite concentration is 0%. Because of this, the boehmite concentration of a hydrate whose roentgenogram has the largest total area for the three Lines resulted, set equal to zero. All other preparations were based on this.

Since a device for röntgenographi see structure determination always contains some elements, whose properties can undergo a small change from time to time is used to compensate of these variations used a quartz standard and the maximum intensity of a quartz line at 20 * 7 ° 20 determined simultaneously with the examination of each hydrate sample. The united area is then represented by the corrected corresponding factor.

It has been found that in the presence of significant amounts of trihydrates, the values for the various Components can be reproduced more precisely than + 5%. When the boehmite concentration increases, the specified values become more reliable and more reproducible.

The three samples examined had the following composition :

Basis for catalyst
No. area
4.7 to 4.9 Å Boehmite
»/ 0 Bayerite
• / 0 Gibbsite
(+ Randomit)
° / o 400 E 9027
B-2
400 E 9165 282 units
275 units
272 units 20th
25th
16 39
25th
12th 41
50
72

According to this list, the areas for the 4.7 and 4.9 Ä districts of the roentgenograms, i.e. H. the Areas under the trihydrate maxima, a very similar size. The values are in arbitrary Units given, a1> he is actually measured areas, i. i.e., they are uncorrected Values obtained using a quartz standard. The composition of the Samples vary somewhat in terms of distribution within the trihydrates present, but the total content of trihydrate is the same. So the 9027 foundation contains 16% Randomit, while the Contains relatively little basis for B-2. In the latter case, the bayerite maximum is at 4.13 Å clearly separated from the gibbsite maximum at 4.85 Å, which is somewhat unusual. The basis

for 9165 contains a high percentage of Gibbs-it. in the In contrast, the basis for the catalyst according to Example 4, No. 400 E 9113, was essentially made entirely of boehmite and contained only a small percentage of trihydrate.

Example 6

The following provides information on the determination of the surface area and pore volume of the catalysts given according to Example 1 to 4. These values are based on determinations of the substance adsorption-desorption isotherms using the usual volumetric methods. From these values one can the pore distribution according to the method of Barrett, Joyner and Halenda (JACS, 73, p. 373 [1951]) to calculate. Here one starts from the saturation value of the isotherms and takes volume increments according to certain incrementals of the pore radius and makes the necessary corrections. Pore volume and surface area for each increment will be summed up, and the calculation is terminated if the sum of the pore volume equals the (at

Saturation) is observed pore volume. The summation the surface agrees with the BET area (BET = Brunauer-Emmet registration method) within an average deviation of about 5%. The pore size at which the calculations broken off is recorded as the size of the smallest pores.

The pore spectrum can be calculated by cumulatively plotting the pore volume against the pore size applied and graphically differentiated. However, it is more common to keep the volume within certain arbitrary ranges of the pore size to be tabulated, what in the following table for a number of aluminum oxide bases as well as for the correspondingly untreated and aged Catalysts is done. Although the values are not given, the total pore volumes are correct, determined from the isotherm, with the values determined by helium-mercury displacement. So there are no very large pores left out in the isotherm method. The analytical Values are given in Table I.

Table I.

Identification of the sample

Area, mVg BET ¹ )

Pore distribution

size

the

smallest

Pores,

Total pore volume, ccm / g distribution of the pore volume, pore volume (ccm / g) ^s ) of the pore size range, in Ä

20 to
50

50 to

100

100 to 300

300 to 1000

1000 and more

400 E 9027

400 E 9027, 14% C
aged

400 E 9027, 0% C ² )
aged

"B-2"

B-2, 3.1% C

aged

B-2, 0 «/ · C η
aged

Aluminum oxide base for 400 U
9165, not calcined

400 E 9165

Boehmite base for
400E 9113, no
calcined

400 E 9113

463
107

228
409

168
199

163
446

352
362

419 107

216 399

183 188

160 420

367 366

8 16

12 9

15 17

11 8

13 12

0.617

0.505

0.617 0.659

0.603

0.626

0.571 0.760

0.361 0.390 0.195

0.022

0.080
0.177

0.047

0.041

0.052
0.190

0.166
0.140

0.055

0.067

0.115
0.086

0.127

0.147

0.094
0.092

0.190
0.235

0.046

0.056

0.061
0.053

0.066

0.080

0.045
0.038

0.003
0.005

0.216

0.195

0.188 0.213

0.177

0.177

0.106 0.202

0.002 0.006

0.098

0.154

0.164 0.105

0.166

0.161

0.192 0.205

0 0.004

0.007

0.011

0.009 0.025

0.020

0.020

0.082 0.033

^x ) Brunauer-Emmet registration method.

² ) C removed at 482 ° C using 2% oxygen-in-air.

⁸ ) "Pore size" means the radius of cylindrically shaped pores or the distance between parallel platelets; the term is used as long as the shapes of the pores are indefinite.

Example 7

Under comparable conditions, the following activity values were obtained for the catalysts according to Examples 1 to 4 in a test plant:

9027 catalyst
B-2 - 400 E -
9165 9113 Gas velocity, parts by weight (liquid) / hour /
Part by weight of catalyst
Pressure, at, 4th
50.6
5 4th
50.6
S. 4th
50.6
5 4th
50.6
5 Molar ratio of hydrogen to hydrocarbon.

15th 9027 catalyst
B-2 16 9113 * 90.2
75.4
86.9
84.3 88.5
79.6
83.3
2.6
84.5 - 400 E -
9165 89.2
78.3
86.8
83.6 Result at 466 ° C
Stabilized reformate, weight percent
Head proportion, weight percent
Octane number (research)
Result at 477 ° C
Stabilized reformate, weight percent
Head portion, weight percent
Octane number (research) 83.2
79.7
83.6
3.3
85.8

A mid-continent heavy gasoline with the following properties was used as feed:

Specific weight (15.6 ° C) 0.7612

Distillation, ° C

Beginning of boiling 114

10% 123

50 ° / »141

90 ° / »169

End of boiling 186

Paraffins 44.3%

Olefins 1.2 »/»

Naphthenes 43.8Vo

Aromatics 10.7%

Sulfur 0.02%

Octane number (Research) 39.0

Example 8

The effectiveness of the catalysts prepared according to the invention when applied to reforming can be seen when the catalysts are tested for their activity stability and selectivity in long-term aging tests. the Values of this example were obtained by placing the catalysts to be tested in a reaction tube made of stainless steel. The catalyst was either arranged as a fixed bed or with tabular aluminum oxide with a particle size of 2.362 to 1.194 mm clear mesh size dispersed. Before starting the hydrocarbon treatment, the reaction chamber is flushed with nitrogen, to remove air, followed by purging the catalyst bed with hydrogen at atmospheric pressure and the temperature gradually increases to operating level. After the reaction chamber has been brought to working temperature, d. H. to 482 ° C, the hydrogen purge is at a gas velocity of 1700 parts by volume of hydrogen / hour / part by volume of catalyst continued for 12 hours.

At the end of this reduction period, the entire system is put under hydrogen pressure and the temperature set. Now pass the hydrocarbon charge by opening the appropriate one Valve on. The mixture of feed and recycle gas flows through the catalyst bed. The reaction product is fed into a receiver, in which the condensable liquid is removed. the The gas phase is fed back into the catalyst bed. Excess circulating gas is drawn off from the system. The liquid product is periodically poured into a evacuated bomb withdrawn and stabilized by taking it out of this in a at a temperature of 16 ° C held glass container peeled off. The octane number (research neat) of the stabilized liquid is determined by the micro method. The pure production of hydrogen is made from the volume of the withdrawn circulating gas surplus is calculated. In the aging tests specified below the following products were used as feed:

Mid-Continent Commercial Gasoline

California Ce heavy fuel

California Ce Commercial Gasoline

Specific weight (15.6 ° C)

Distillation, ° C

Start of boiling

10%

50%

90%

End of boiling

Paraffins

Olefins

Naphthenes

Aromatics

sulfur

Octane number (research)

0.7608

0.7559

0.7571

114 72 118 123 73 120 141 74 122 169 76 124 186 91 130 44.3% 9.5% 36.2% 1.2% -% -% 43.8% 87.8% 59.0% 10.7% 2.7% 4.8% 0.02% 0.004% 0.010% 39.0 87.8 58.4

The catalyst no. 400 described in example 2 if one uses the mid-continent starting material (experiment E 9027 was used according to procedure 9-38) given above. During the trial period of checked. The working conditions of the steadily conducted 2500 hours was a pressure of 53 at, a water process were adjusted so that a reformate material-hydrocarbon ratio of 6: 1 and one of the octane number 80 (Research neat) was obtained. 70 feeding speed of 4.0 parts by weight (liquid

17 18

sig) / hour / part by weight of catalyst. with an average octane number of around 82

The initial temperature was 477 ° C, which resulted in and it did not seem profitable to

Was lowered to 471 ° C for 200 hours. After about continue with further temperature increases

1100 hours of operating time was added to the temperature. The yields of liquid product remained

again increased to 477 ° C., by the desired 5 during the entire experiment at or above

Maintain octane levels. After about 90 percent by volume.

The temperature was increased to 482 ° C. for 1750 hours. In contrast to the stability as it was in the case of the above, and the long aging tests described in the last 400 hours of the test, were carried out at 488 ° C. In the entire duration, an experiment had to be carried out in which the octane number of the reformate was always very close to 80 using the mid-continent heavy gasoline, although during the last one of the catalyst according to Example 4, 400 E 9113 to several 100 hours of the experiment gradually Böhmitgrundlage, a feed drop began at 482 ⁰ C, 35 at from about 80 to 75, wherein the initial velocity of 5 parts by weight (liquid) / hour / yield of liquid volume of about 90 % fraud. The part by weight of catalyst and a pure hydrogen-hydrogen generation remained approximately in the first 15 hydrocarbon ratio of 13: 1 processed 1800 hours of the experiment at 44.6 to 48.2 m ³ and an octane level of 85 hydrogen / m ³ charge (liquid) was maintained . Later (9 to 42), part of the experiment was stopped at about 550 hours, a gradual drop began because the octane numbers fell and also the approx. 17.8 m ³ hydrogen / m ³ charge (liquid) hydrogen free production decreased sharply,
a. 20 In an attempt with an octane level of

Similarly, an aging test (13-12) 95 was carried out at a temperature of 502 ° C (496 to 499 ° C, using the catalyst according to in the first 20 hours), 35at and a feed example 1, B-2 , used to have a mid-continent rate of 2.9 parts by weight / hour / Ge heavy gasoline at 482 ° C, 35 atm, one part by weight of catalyst and a hydrogen-carbon rate of 5 parts by weight (liquid) / hour / hydrogen -Ratio of 10: 1, the catalyst was part by weight of catalyst and a hydrogen- according to Example 1, B-2, on the octane level 95 hydrocarbon ratio of 13: 1 to be processed or above about 300 hours of use and exceeded. This experiment ran for 700 hours at a half octane level of 90 over 460 hours octane level of 85 (Research neat), with a yield of liquid product of about 81 yield of liquid product more than 90 volume - 30 to 82 volume percent. Percent deteriorated at 460 hours. The hydrogen-free generation was the catalyst as a result of an incident in the approximately 98 to 107 m ^{3 of} hydrogen / m ³ charge (liquid operation.

sig). After about 700 hours, the temperature was, in contrast, in the same comparison

Increased by 5.6 ° C, octane number and gas purity, which shows the effectiveness of the catalytic converter

to keep generation at the same level, and the 35 according to Example 4 on boehmite basis, No. 400 E 9113,

continued searching for another 300 hours. After about checked, the octane number 90 already after 200 operating

1000 hours, the temperature was reached again by hours and the attempt after 223 hours

5.6 ° C increased and the experiment continued. After canceled because the octane number was constantly down to about

about 1250 hours the temperature was again decreased by 87 and the hydrogen free generation of

2.8 ^r C increased, then successively to 499 and 504 and 40 about 143 to about 98 m ³ hydrogen / m ³ feed

after about 1550 hours to 510 ° C. At this height (liquid) sank.

it was held until the end of the experiment, which was canceled after 1846 hours because the catalyst described in game 1, B-2, had an octane rating of about 82 to about 72 to test. A long-term aging test (13-15) was carried out under The pure gas production in the last 100 hours was about 41 m ^{3 of} hydrogen / m ^{3 of} beet as charge using a California Cg heavy gasoline section of the test. This attempt sent (liquid). was at 488 ° C, 14 at ,, a feed speed

By way of comparison (experiment 14-12), the average of 1.5 parts by weight (liquid) / hour / product

Example 3 described catalyst based on gibbsite, part by weight catalyst and a hydrogen-carbon

400 E 9165, at the point at which the above hydrogen ratio of 10: 1 was carried out. the

such was terminated with the catalyst B-2, the yield of liquid product was about 80 vol.

from the mid-continent heavy gasoline a reformate lum percent and the hydrogen production about 357

from the octane number of 83 to 85 with more than 90Vo to 374 m ³ hydrogen / m ³ charge (liquid). the

lum percent yield of liquid components. This benzene yield in the liquid product was about 75 per cent.

In place, the run with catalyst 9165 was about 4% toluene yield. After 1185 hours it was

by 8.3 ° C lower temperature, the yield had increased the temperature to 493 ° C and after

by 2% higher and the octane number of the product by 1653 hours to 499 ° C. At this point, the

three units higher. After 1520 hours, the benzene yield became about 60% and the toluene yield

Temperature increased by 5.6 ° C to 493 ° C, by about 3 ° / o, and from about 1920 to 2002 there were 1 ben-

keep the octane level at 83 to 84. Formed after 60 zol / kg catalyst. After 2300 hours it was

about 2265 hours, the temperature in turn decreased the benzene yield to about 50 percent by weight.

increased by 5.6 ° C to 499 ° C, causing the octane to drop. The experiment was a.b-

number rose from 81 to 82 to almost 83 to 84. After broken. At this point the yield was up

In 2540 hours the temperature was dropped to 504 ° C about 40% and the recovery of liquid

increases as the octane numbers fell to 80 and the 65 product increased to a total of about 90%. It was

Hydrogen-free generation by around 17.8 to 98.2 m ³ which produces around 2087 1 benzene / kg catalyst.

Hydrogen / m ³ feed (liquid) decreased; Similarly, the effectiveness of the

The experiment was ended after 2707 hours because the same catalyst was tested by using the 75% -

last temperature increase from 499 to 504 ° C only head portion of the California-Cg-heavy gasoline-concentrate

a further life of 150 to 160 hours 70 occurred on xylene (12-17) processed. At 477 ° C,

21 at and a feed rate of 3 parts by weight (liquid) / hour / part by weight catalyst and a hydrogen-hydrocarbon ratio of 10: 1, the octane number of the product was 95 to 97 and the yields of liquid product in the order of 90 to 91 percent by volume . After 455 hours the feed was switched from a 75% top cut to an 80% top cut while all other operating conditions remained unchanged. The octane numbers remained at 95 to 96. After 682 hours the feed was switched to a 65% overhead cut. The octane ratings of the product were roughly one number lower than those obtained when processing the 75% and 80% head parts. After 2000 hours of operation (yields and octane numbers had remained constant during this time) the feed was switched to a C ₇ heavy gasoline cut from the same crude oil in order to recover toluene. After 500 hours, the toluene yield was still about 55 percent by weight of a liquid product totaling 90 percent by volume. At this point the feed was switched back to the 75% overhead portion of the C ₈ fraction to determine whether catalyst deactivation had occurred. After this original charge had run through for 24 hours, the octane numbers averaging 97.4 (Research), the experiment was terminated after 2508 hours because it was evident that catalyst deactivation had occurred only to a very small extent, if at all.

Example 9

35

A. Description of the alumina hydrate base The base for this catalyst, 400 E 9757, was an alumina hydrate, 516-27, which was 71 days old. The composition of the hydrate (determined by X-ray) was: 29% gibbsite, 6% bayerite, 22% randomite, 42% boehmite and traces amorphous components. The crystallite size of the trihydrate phases seemed low.

The 516-27 base was prepared by deionizing a solution of NH ₄ OH in water (1: 1) to a solution of vigorously stirred AlCl ₃ -OH ₂ O (0.454 kg Al Cl ₃ · 6H ₂ O / 2 1 water) zusetzte, was until the p _H 8.0. The hydrate was filtered off from the mother liquor and washed out in filter presses to a chlorine content of 0.27%. The mass was re-slurried using about 25 liters of distilled water / kg Al ₂ O ₃ and the pH of the slurries adjusted to 8.0 (first re-slurry), 9.0, 8.5 and 8.5 (fourth re-slurry) . The washed hydrate was aged in the form of a filter cake.

B. Impregnation of the Akiminiumoxydhydratgrundlage

21 Aluminiumoxyd.hydrat- (containing 123 g A1 ₂ O _{3/1) (516-27)} slurry was introduced into an earthenware vessel of 11 1 capacity and vigorously stirred for 30 minutes to effect a thorough distribution. With continued stirring, a platinum ^{solution consisting of 34.3 cm 3} H ₂ PtCl ₆ solution (0.043 g Pt / cm ³ ) in 170 cm ^{3 of} distilled water was added over the course of 5 minutes. After stirring for a further 10 minutes, hydrogen sulfide (343 cm ^{3 of} _{distilled water, saturated with H 2} S at 25.6 ° C.) was slowly added. The slurry was stirred for an additional 30 minutes before drying. The resulting slurry was very thin and light brown in color.

C. Drying, tableting and calcining

The slurry obtained according to the above procedure was put into a bowl made of difficult to melt Poured glass and dried in an aminco oven (circulating air dryer) at 110 ° C. After a few hours it was notes that the drying did not take place uniformly and that a hard, rubber-like film was formed on the surface of the material to be dried would have. This film could be redispersed in the not yet dry portion of the slurry. Therefore the slurry (solids content probably 40 to 50%) became in about 30 seconds homogenized in a mixer. The drying was then completed at 110 ° C.

The dry catalyst was ground to a size of 0.833 mm clear mesh size, with 2% Binder mixed and shaped into 4 mm tablets. The binder was burned out at 482 ° C for what 5 parts of air and 300 parts of nitrogen atmosphere were initially used. The oxygen content was then slowly increased and the catalyst finally calcined in air at 482 ° C. for 6 hours.

The calcined catalyst was designated with the test number 400 F 9757.

Example 10

The values for the activity of an unused catalyst, 400 E 9757, are given below. The test conditions were 35 at, the feed rate 4.4 parts by weight (as liquid) / Hour / part by weight catalyst, the hydrogen to hydrocarbon mole ratio was 5, and as a feed served a heavy fuel with a boiling range of about 118 to 204 ° C, which contains 56% paraffin, Contained less than 1% olefins, 24% naphthenes and 19% aromatics.

Stabilized reformate,
Weight percent ....

Head portion, weight percent

Reaction temperature, ° C
466 I 477 I 488

86.8

82.8

1.3

79.0

Octane number (Research) 79.9 86.4 91.1

In a series of aging tests at 496 ° C, 35 at, a hydrogen-hydrocarbon ratio of 10: 1 and a feed rate of 3 parts by weight (as liquid) / hour / part by weight of catalyst, the activity of the unused catalyst decreased from an initial octane number of about 97 (CFRR) drops to around 90 in 500 hours. After regeneration with dilute oxygen at 427 ° C, the same catalyst worked for about 800 hours before the octane number dropped to 90. After a second regeneration, the catalyst again showed a decrease in activity which was somewhat better than that of the untreated catalyst, calculated after a further 375 hours. The yield of liquid product was about 80% (15.6 ° C.) and the hydrogen production was 160 to 178 m ³ hydrogen / m ³ feed (liquid) as long as the catalyst was switched to reaction.

In the examples given above, the results are given that were obtained with aluminum oxide-platinum

Catalysts and fluoride-activated alumina-platinum catalysts were obtained because these catalysts appear to have the highest selectivity and activity stability in common reforming processes. Using the same production technique, however, rhodium, palladium or iridium or mixed catalysts of these platinum metals can also be produced, which also have better properties based on the structure of the aluminum oxide base achieved according to the invention. The metal ίο is incorporated into the aluminum oxide starting mixture by adding the aqueous solution of a water-soluble salt, e.g. B. rhodium or iridium chloride or mixtures of such salts are added to the hydrate slurry and the metal is then _{precipitated with H 2} S in situ. As in the above examples, a promoter can also be added here.

Claims (5)

Claims: 20 1. Catalyst for reforming light hydrocarbons and for producing aromatics, which essentially consists of y-modifications of aluminum oxide and metal of the platinum group is composed, characterized in that the aluminum oxide base is a large-pored Structure and a large surface area and the dehydration product of aluminum oxide hydrate phases, in which trihydrates predominate is. 2. Catalyst according to claim 1, characterized in that the metal of the platinum group in finely divided state, in particular in a particle size of predominantly less than 50 Å present is. 3. Catalyst according to claim 1 or 2, characterized in that the mixture of aluminum oxide hydrate phases contains about 5 to 35 weight percent monohydrate. 4. Catalyst according to one of claims 1 to 3, characterized in that it is used as a metal the platinum group contains platinum. 5. Catalyst according to one of claims 1 to 4, characterized in that it contains a promoter. Considered publications:
U.S. Patent No. 2,479,109. © «09 6Ϊ8 / 3Ϊ3 9.5 &