Synthetic Silica Alumina Catalysts

Acid-treated clays were the first catalysts used in catalytic cracking processes, but have been replaced by synthetic amorphous silica-alumina, which is more active and stable. Incorporating zeolites (crystalline alumina-silica) with the silica/alumina catalyst improves selectivity towards aromatics. These catalysts have both Fewis and Bronsted acid sites that promote carbonium ion formation. An important structural feature of zeolites is the presence of holes in the crystal lattice, which are formed by the silica-alumina tetrahedra. Each tetrahedron is made of four oxygen anions with either an aluminum or a silicon cation in the center. Each oxygen anion with a -2 oxidation state is shared between either two silicon, two aluminum, or an aluminum and a silicon cation. 

Accordingly, work has been done on series of n-paraffins,. isoparaffins, naphthenes, aromatics, and naphthene-aromatics which have been chosen as representative of the major components of petroleum. In addition, olefins, cyclo-olefins, and aromatic olefins have been studied as a means of depicting the important secondary reactions of the copious amounts of unsaturates produced in the majority of catalytic cracking reactions. A silica-zirconia-alumina catalyst was used principally it resembles closely in cracking properties typical commercial synthetic silica-alumina catalysts. 

The inherent variability of the raw mineral, particularly with respect to minor constituents which in certain cases were known to have major effects on the cracking reaction, led to the development by the Houdry Process Corp. of a synthetic silica-alumina catalyst of controlled chemical composition and more stable catalytic properties. Full scale manufacture of synthetic catalyst was started in 1939. 

Most of the commercial Fluid plants have found it economical to continue operations on the silica-alumina catalyst which was used exclusively during the war to maximize raw material production for aviation gasoline and synthetic rubber. The lower cost of the natural clays, however, coupled with the different product distribution, has influenced a number of plants to switch to the use of this material. As the choice of catalyst is... 

Double-bond isomerization was once used in the multistep synthesis of isoprene developed by Goodyear.266-268 2-Methyl-1-pentene produced by the dimerization of propylene was isomerized to 2-methyl-2-pentene over a silica-alumina catalyst at 100°C. The product was cracked to isoprene and methane. Because of the lower cost of isoprene isolated from naphtha or gas oil-cracking streams, synthetic isoprene processes presently are not practiced commercially. 

Rates of model reactions are more commonly used to determine relative rather than absolute surface acidities and a variety of acid-catalyzed reactions have been used for this purpose (1-3). Xylene isomerization is a particularly well-substantiated model reaction, thanks to work by Ward and Hansford (43). They demonstrated that the conversion of o-xylene to p- and /n-xylenes over a series of synthetic silica-alumina catalysts increases as the alumina content is increased from 1 to 7%. The number of strong Brdnsted acids in each member of the catalyst series was measured by means of infrared spectroscopy. Since conversion of o-xylene was found to be a straight-line function of the number of Br0nsted acids (see Fig. 9), rate of xylene isomerization appears to be a valid index of the amount of surface acidity for this catalyst series. This correlation also indicates that the acid strengths of these silica-alumina preparations are roughly equivalent. 

Fig. 9. Conversion of o-xylene as a function of Br0nsted acidity for synthetic silica-alumina catalysts (43).

Since silica-alumina contains Br nsted as well as Lewis acid sites, a clear correlation between rates of a heterogeneously catalyzed reaction and surface acidity as measured by pyridine adsorption is only possible if a distinction between PyH+ and PyL is made. This is possible by infrared spectroscopy as shown in this section. Thus, Ward and Hansford (226) found a good linear correlation between the percent conversion of o-xylene and the Br nsted acidity of synthetic silica-alumina catalysts. This correlation is shown in Fig. 4, where the Br nsted acidity is expressed as peak height of the band at 1545 cm-1 per unity of catalyst weight. 

In 1940, Houdry Process Corporation initiated commercial manufacture of a synthetic silica-alumina catalyst at Paulsboro, New Jersey (133). The synthetic catalyst is produced in pellet form (51,265) and contains 12 to 13% alumina (221,276). It has the advantages of controlled chemical composition, higher purity, and greater heat stability, but is more expensive than the activated-clay catalyst. 

In 1944, Socony-Vacuum Oil Company started manufacture of synthetic silica-alumina catalyst in the form of beads (262). This catalyst was reported to contain about 10% alumina. The bead catalyst gives about the same product distribution as the pelleted synthetic catalyst and was developed primarily to achieve greater physical strength for use in the TCC process. The bead catalyst has also been used in Houdry fixed-bed units (51,171). Subsequently, a harder bead catalyst was developed for use in the air-lift units. The improved bead catalyst consists of approximately 15% alumina and 85% silica and contains 0.003% chromium to minimize afterburning by suppressing formation of carbon monoxide during regeneration (333). 

Commercial production of synthetic silica-alumina catalysts for use in fluid cracking was initiated in 1942. The synthetic catalysts were first manufactured in ground form, but means were later developed for production in MS (micro-spheroidal) form. First shipments of the MS catalyst were made in 1946. The synthetic catalysts contain 10 to 25% alumina. Synthetic silica-magnesia catalyst has also been used commercially in fluid-catalyst units (19,100). Magnesia content is 25 to 35% as MgO (276). 

Fluid grades of synthetic silica-alumina catalyst are manufactured by the American Cyanamid Company, The Davison Chemical Corporation, Morton Salt Company, and National Aluminate Corporation. At first the catalyst was dried and ground to produce the desired range of particle sizes. It was later found that by using spray driers micro-spheroidal particles of the desired size distribution could be produced directly without any grinding (7,10,145). The particle-size distribution can be altered, within limits, by changing the spray-drier conditions (145). 

An exact comparison of activity ratings by the various methods is not available in the absence of data on identical samples of catalyst. However, the relative magnitudes of the numbers reported as activity rating by various methods are illustrated in the following tabulation of typical values for fresh synthetic silica-alumina catalysts. 

Fm. 47. Effect of iron on selectivity of synthetic silica-alumina catalyst. [Mills, Ind. Eng. Chem. 42, 182 (1950). Reprinted by permission.]... 

Cracking catalysts are highly porous materials with large internal surface areas. Thus, for example, a fresh synthetic silica-alumina catalyst typically has a pore volume of about 0.5 cc./g. and a specific surface of the order of 500 m.Vg., equivalent to about 56 acres (almost 0.1 square mile) per pound. Compared with the internal pore surface, the external surface of the discrete particles of catalyst used in commercial plants is insignificant. This is illustrated by the following tabulation, which shows the external surface areas of spherical particles of the diameters employed commercially. A particle density of 1.0 g./cc. was assumed, about equal to the observed particle density for fresh synthetic silica-alumina. 

Based on experiments with pure hydrocarbons and synthetic silica-alumina catalyst, it has been estimated that the cracking-rate constant at 932°F. should decrease by a factor of to when particle diameter is increased from about 0.5 mm. to 4 mm. (74). The influence of particle size on effective activity is especially pronounced at very high cracking temperatures (49). This behavior is in line with predictions because, with increasing temperature, reaction rate on the catalyst surface increases more rapidly than the rate of diffusion of reactants into the pores. Cracking of unsymmetrical diarylethanes is an exceptional case in which the reaction appears to depend entirely upon the number of collisions of the hydrocarbon with the external area of the catalyst particles (208). 

The ratio of CO2 to CO with Filtrol SR catalyst has been reported to be 1.2-1.3, closely resembling that obtained with synthetic silica-alumina (326). The ratio tends to increase with use for silica-alumina catalyst but not with silica-magnesia (355). The increase with silica-alumina is presumably due, at least in part, to accumulation of metal contaminants that promote complete combustion to CO2 total iron pick-up during commercial use was reported to be much less in the case of silica-magnesia than in a companion commercial run on silica-alumina (355). Intentional addition of a small amount of chromium to TCC bead catalyst is practiced commercially for the specific purpose of insuring complete combustion to CO2 and thereby avoiding afterburning (333). 

A common feature of these catalysts is their acidic nature (i.e., they all act as solid phase acids in the hot gas oil vapor stream). Synthetic silica/alumina catalyst composites, for example, have an acidity of 0.25 mEq/g distributed over an active surface area of some 500m /g). This acidity is the key feature that distinguishes catalytic cracking from straight thermal cracking. 

Commercially, pyridine is prepared by the gas phase, high-temperature reaction of crotonaldehyde, formaldehyde, steam, air, and ammonia over a silica-alumina catalyst in 60-70% yield. However, in the laboratory, the challenge is in the preparation of substituted pyridine derivatives in a process that allows one to control regioselectivity and chemoselectivity in the most efficient manner. In this regard the utility of palladium-catalyzed cross-coupling reactions has enabled synthetic chemists by providing the ability to construct highly diversified pyridine derivatives in an efficient fashion [2]. 

Pyridine was first isolated, like pyrrole, from bone pyrolysates the name is constrncted from the Greek for fire, pyr , and the suffix idine , which was at the time being used for all aromatic bases - phenetidine, toluidine, etc. Pyridine and its simple alkyl derivatives were for a long time produced by isolation from coal tar, in which they occur in quantity. In recent years this source has been displaced by synthetic processes pyridine itself, for example, can be produced on a commercial scale in 60-70% yields by the gas-phase high-temperatnre interaction of crotonaldehyde, formaldehyde, steam, air and ammonia over a silica-alumina catalyst. Processes for the manufacture of alkyl-pyridines involve reaction of acetylenes and nitriles over a cobalt catalyst. 

The preparation of synthetic silica-alumina catalysts is a relatively simple one, involving the coprecipitation or cogelation of the two hydrous oxides from mixed solutions of sodium silicate and aluminum sulfate. Depending on how the solutions are mixed and on the pH and concentration of the resulting mixture, the combined hydrous oxides will be formed as a coprecipitate, which separates from a greater part of the aqueous phase, or as a true hydrogel, which embraces the entire solution volume. 

There are many other methods of preparing active synthetic silica-alumina catalysts. A fair catalyst can be made by impregnating dried silica gel with an aluminum compound which is easily converted to the oxide by calcination, e.g., A1(NC>3)3. A preferred impregnation technique is to soak a sodium-free silica hydrogel in a solution of an aluminum salt and to follow this with an aqueous ammonia treatment to precipitate the hydrous alumina on the silica (Thomas, 16 Ryland and Tamele, 17). It should be noted that silica hydrogel can easily be freed of sodium ions by water washing, since it is not a zeolite. Exceptionally pure silica-alumina composites can also be prepared by the hydrolysis of mixtures of ethyl orthosilicate and aluminum alkoxides (Thomas, 18). 

The synthetic silica-alumina catalysts, TCC Beads, Aerocat, and Diakel, are composed of pores appreciably larger but with average pore radius values almost exclusively in the small pore range of 15 to 25 A. (Ries, Johnson, Melik, and Kreger, 48 Shull, Elkin, and Roess, 57). The complete absence of large pores is indicated for the TCC Beads and the Aerocat Microspheres. In the case of silica-alumina catalysts, the hysteresis loops have considerable breadth and area in contrast to the silica-magnesia isotherms. 

The adsorption-desorption isotherms for synthetic silica-alumina catalysts are somewhat similar to those of silica-magnesia in that there is very little adsorption in the higher relative pressure region. However, the average pore radius of the silica-alumina preparations, generally in... 

---