CA1158631A - Stabilized reforming catalyst - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
Thermal Stability of a halide-promoted, supported noble metal reforming catalyst is improved by employing a modified alumina support whose alumina precursor comprises at least about 75 weight percent boehmite, having an average crystalline diameter no greater than about 60 Angstroms.
The thermal stability of the support is further improved by the inclusion of a minor amount of silica.

Description

1 1~8~3~

BACKGROUND OF TEIE Ie~VENTION
Catalytic reforming of naphtha fractions has lonq been regarded as an attractive means for providing gasoline blending components having high octane numbers. The demand for such blending components has become critical as the use of organomet-allic octane appreciators, such as lead aklyls, has diminished in response to enviro~mental constraints.
Catalytic reforming generally involves a complex series of hydrocarbon reactions, employing a hydrogenation-dehydrogena-tion catalyst, wherein a substantially paraffinic and/or naphthe-nic naphtha fraction of petroleum variously undergoes dehydrocy-clization, dehydrogenation, isomerization and hydrocracking to provide a mixture substantially comprising aromatic, olefinic, naphthenic and isoparaffinic hydrocarbons. Such mixtures possess sultably high octane numbers and generally excellent blending characteristics. The reforming reactions are, on balance, endo-thermic and are generally conducted with serial flow through a plurality of reactors at elevated temperatures in the presence of hydrogen, with provision for heating between reactor stages.
Noble metal catalysts generally are effective in the promotion of the spectrum of chemical conversions characteristic istic of reforming. Such metals are quite effective at rela-tively low concentrations, for example, 1 weight percent or less, when extended upon a support material. Suitable support ; - materials must possess sufficient surf~ce area to provide an adequate base for dispersion of the noble metal. Further, the support must possess pores having dimensions large enough to accommodate the chemical reactants to be acted upon. In general, an adequate surface area for most reactions will be in excess of at least about 1~0 square meters per gram. For most reforming operations, a generally preferred noble metal is platinum and a preferred support material is alumina. Platinum is often used ' ~.
,~ --1-- ~

together with a second metal. Alumina is often modified to increase acidity of the support, as by contacting with a halide-affording material.
Catalysts comprising platinum, such as platinum dispersed on an alumina support, are generally employed in the reforming of naphthas because of their overall excellent per-formance, despite high cost, and high selectivity toward the production of aroma~ic hydrocarbons boiling in the gasoline range. Maintenance of platinum catalyst activity and selectiv-ity can be improved by the use of a second metal. A preferredsecond metal is rhenium, whose use is particularly described in United States Patents numbered 3,415,737, 3,496,096 and 4,176,088.
Since most of the desirable chemical reactions, including hydroisomerization and dehydrocyclization, require acidic conditions, it is necessary to provide an effective level of acidity in the catalyst. This is generally accomplished by the introduction of a halogen, usually chloride, which is held on the surface of the catalyst support material together with the catalytic metals. The halogen material is replenished throughout the catalytic cycle by addition, usually in the form of an organic halide, to the hydrocarbon feedstock. Control of the halide concentration on the catalyst by control of the water-halide ratio in the feed is well known.
One method for increasing the activity of the reforming catalyst, particularly in terminal reactors, comprises significantly increasing the chloride content in the catalyst, for example, to a level substantially above 1.0 weight percent. However, this approach is often not feasible because loss of catalyst surface 3~ area with continuing on-stream time leads to loss of chloride from the catalyst, thus requiring an excessive make-up rate for chloride and causing excessive downstream corrosion.

i ~5~63~

Although much attention has been given to this problem, there is a continuin~ nee~ for an impro~ed support material, possessing s~itably hi~h and stable surface area and pore size characteristics. A desirably improved support will exhibit a substantially diminished rate of change in desirable surface parameters, such as surface area and pore size, upon continued e~posure to high tempe~atures. Such increased stability of the support material will effectively reduce the tendency toward a gradual loss of halide promoter.
SUMMARY OF THE INVENTION
This invention relates to a reforming catalyst, having improved catalytic activity and thermal stability, comprising a noble metal, such as platinum, on a support, such as a modified alumina. The catalyst may preferably contain a second metal, such as rhenium, as well as platinum. The alumina support pre-cursor is selected to comprise at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms. Its thermal stability is greatly improved by the incorporation of a minor amount of silica, as by addition of an orthosilicate ester, alkali metal silicate or a low-sodium colloidal silica dispersion, preferably to a mull of the boehmite-containing alumina before extrusion and subsequent calcination of the extruded support material.
This invention further relates to a process for reforming petroleum naphtha, employing a catalyst as described hereinabove. Acidity of the catalyst is maintained by addition of a chloride-affording substance such as an alkyl chloride or polychloride. It has been surprisingly observed that a desired chloride level can be more readily maintained with the catalyst of this invention. For example, a lower chloride make-up rate is required under typical reforming conditions. Additionally, it has been surprisingly found that higher chloride levels can I ~ 5~3 ~

be achieved and maintained, such that higher activity levels can be achieved and maintained.
DESCRIPTION OF THE INVENTION
This invention relates to a reforming catalyst comprising a noble metal on a modified alumina support, having a surface area typically within the range from about 100 to about 3~00 m.2/g., wherein the modi~ied alumina support precursor comprises at least abo~t 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and from about 0.25 to about 4.0 weight percent, preferably from about 0.5 to about 2.0 weight percent, of silica, and to a re-forming process employing said catalyst.
The reforming catalyst of this invention preferably includes platinum as the noble metal and may optionally include a second metal, preferably rhenium. In a particularly preferred embodiment, the alumina support material is derived from an alumina precursor comprising at least about 80 weight percent boehmite and c~ntains from about 1.0 to about 2.0 weight percent silica. The minor quantity of silica associated with the support remarkably improves the stability of both surface area and pore diameter dimensions, under reforming conditions, of the boehmite-rich material. Longer catalyst cycle length and ulti-mate catalyst life are both realized with the catalyst of this invention. Additionally, the chloride deposited on the catalyst during reforming operations is more securely held so that the - activity-promoting effect of the halide is realized to a higher degree and with a significantly lessened make-up requirement.
One particularly effective technique for incorporating the silica component into the alumina support material comprises incorpora-3~ tion of a silica forming compound selected from the group com-prising orthosilicate ester, alkali metal silicate and low-sodium colloidal silica, into the alumina mull prior to extrusion and 3 ~

subsequent calcination.
The reforming cata~yst of this invention comprises from about 0.3 to about 1.5 weight p~rcent of a noble metal, such as platinum, together with from about 0.3 to about 2.0 weight percent chloride maintained upon the surface of the alu-mina support. When a second metal, such as rhenium, is also present, such second metal should be present in an amount to provide from about 0.3 to about 1.5 weight percent thereof dis-tributed upon the alumina support material.
The reforming catalyst of this invention desirably comprises from about 0.3 to about 1.0 weight percent of a noble metal such as platinum, preferably about 0.6 weight percent of platinum. Additionally, in bimetallic embodiments of the cata-lyst, from about 0.3 to about 1.0 weight percent of a second metal is desirably employed, preferably about 0.85 weight percent of rhenium. For effective promotion of the reforming reactions, there is incorporated from about 0.3 to about 2.0 weight percent of a halide, preferably chloride, into the support material, as by addition in the form of an alkyl halide to the naphtha feedstock. When preferred amounts of platinum and rhenium are present in the catalyst, it has been found that the presence of about 1.4 weight percent chloride provides a particularly desir-able promoting effect.
The catalysts of this invention surprisingly maintain the preferred high chloride levels to a degree permitting a much lower chloride make-up rate. This ability to hold chloride also minimizes the stripping effect of moisture on catalyst halide concentration, as well as the downstream corrosion problems attributable thereto. Accordingly, the catalysts of this inven-tion permit the ready maintenance of a particular chloride level,with less make-up required, or the achievement of higher chloride levels, or activity levels, than generally realized heretofore.

i ~5863~

This improvement is accomplished without significant loss of yield.
With the catalyst of this invention, having improved thermall surface-area ~tabili~y characteristics under reforming conditions, the initial high activity and selectivity, in the processing of naphtha feedstoc~s to high-octane number blending components for gasolines, are surprisingly maintained over a substantial period of time, when measured either as cycle length or as the number of process cycles achieved prior to such a loss of catalyst activity as to dictate catalyst replacement. Such catalyst qualities significantly increase the operating effi-ciency of a naphtha reforming process and the economic production of critical blending stocks.
In seeking means for improving the stability of reforming catalyst supports, it has been found that an alumina precursor in the form of boehmite, a crystalline alumina mono-hydrate, affords surprisingly greater thermal stability than other crystalline forms, such as the tri-hydrates (bayerite, nordstrandite or gibbsite) or than amorphous alumina.
Suitable catalyst support precursor materials include any crystalline alumina having the requisite content of boehmite.
One suitable source for boehmite-rich alumina is the product obtained from hydrolysis of the aluminum alko~ide catalyst employed in olefin oligomerization and in alcohol production.
Another means for achieving substantially increased boehmite content in an alumina comprises hydrothermal aging of various alumina precursors. Preferred boehmite-rich alumina precursors contain at least about 80 weight percent boehmite, the remainder customarily being amorphous alumina, and are further character-ized in having an average crystallite diameter of up to about 60 Angstroms, preferably at least about 30 Angstroms and no more than about 60 Angstroms, and more preferably within the range ~ ~5~3~

from about 30 to about 50 Angstroms. It has been observed that catalysts employing boeh~ite having larger crystallite dimensions exhibit a lesser degree of surface-area stability, comparable to that observed in catalysts derived from crystalline alumina tri-hydrates or amorphous alumina.
In preparation of the support material for convention-al impregnation with platinum and/or other metals, the alumina precursor is customarily formed into a mull or paste with water, extruded and calcined under selected conditions. Such opera-tions may ~e repeated if desired. It has been found that therequisite amount of the silica precursor selected from the group comprising orthosilicate ester, alkali-metal silicate and low-colloidal silica, may be conveniently added to the catalyst of this invention either before extrusion or after calcination.
It is preferred to add the silica precursor to the alumina mull prior to extrusion. Alternatively, the silica precursor may be added to the finished catalyst prior to a final calcination.
Although any orthosilicate ester may be suitably employed, the preferred orthosilicate ester is ethyl orthosili-cate. A preferred silica precursor is alkali-metal silicate.
It is preferred to add the alkali-metal silicate to the alumina mull, followed by washing to remove the alkali metal following extrusion and a preliminary calcination. Alternatively, the alkali-metal silicate may be added to the finished catalyst, with the alkali metal being washed out prior to a final calcina-tion.
Another preferred silica precursor is low-sodium colloidal silica. It is preferred to add the colloidal silica dispersion to the alumina mull prior to extrusion. Alterna-tively, the silica dispersion may be added to the finishedcatalyst prior to a final calcination. In any event, the weight ratio of silica to sodium oxide should be greater than 100/1, 3 ~5~3~

and preferably greater than 300/1. If necessary, this may be achieved by water extraction of the calcined support mater-al or of the ~inished catalyst.
The catalyst of this invention is conveniently employed under generally conventional reforming conditions. Typically, a virgin naphtha, boiling generally in the range from about 200 to about 450F, is processed at reaction temperature within the range from about 850 to about 1050F, in the presence of hydro-gen at process pressures ranging from about 100 to abo~t 700 psig. A series of reactors is customarily employed with provi-sion for reheating between stages. Process hydrogen is recycled in part and the liquid product can be fractionated to provide the desired reformate blending stock.
The following procedures and tests are exemplary, without limitation, of the catalysts of this invention.
EXAMPLE I
PART A
An alumina support material, comprising 52 weight percent boehmite and having an average crystallite diameter of 34 Angstroms, as a powder, was blended with water to provide a thick paste. Mixing was further carried out in a muller with the addition of water as required to provide an extrudable mixture.
After mulling for 30 minutes, the mixture was extruded through a die plate having l/6-inch diameter holes. After drying 12 hours in a forced air drying oven at 125C, the extrudate was broken into small particles and calcined for 2 hours at 300F
furnace temperature. Temperature was then raised to 950F, at a rate of 300F/hour, and maintained for 3 hours. During calcina-tion, dry air was passed through the oven at 1000 VHSV.

The calcined extrudate particles were impregnatedwith a solution (0.75 ml./g.) having a pH = 7, containing I ~L5~31 chloroplatinic acid, perrhenic acid and ammonium chloride in amounts calculated to provide 0.6 weight percent platinum, 0.85 weight percen-~ rhen~um~ and }.5 w~ight percen~ chloride in the finished catalyst. The volume of solution was selected to be slightly greater than the pore volume of the extrudate particles.
A final calcination was carried out as before.
PART A' The procedure of Part A was repeated to provide a finished catalyst containing 0.35 weight percent platinum and 10 0.35 weight percent rhenium.
PART B
An alumina support material, comprising 82 weight percent boehmite and having an average crystallite diameter of 45 Angstroms as a powder, was blended with sufficient dilute (2 volume percent? nitric acid to partially peptize the alumina and with sufficient water to provide a thick paste. Mixing was further carried out in a muller with the addition of water as required to provide an extrudable mixture. After mulling for 30 minutes, the mixture was extruded through a die plate having 20 1/6-inch diameter holes. After drying 12 hours in a forced air drying oven at 125C, the extrudate was broken into small parti-cles and calcined for 2 hours at 300 furnace temperature.
Temperature was then raised to 950F, at a rate of 300F/hour, and maintained for 3 hours. During calcination, dry air was passed through the oven at 1000 VHSV.
PART C
196 g. of a catalyst prepared as in Part A was impregnated under vacuum with 151 ml. of a solution comprising denatured ethanol containing tetraethyl orthosilicate in an 30 amount calculated to provide a finished catalyst containing 0.25 weight percent silica. After 5 hours of equibration, the product was placed in a drying oven overnight. Calcination was carried _g_ 1 15~3 11 out by passing 3001 ml/hr of 1% 2 in N2 while increasing the temperature to 950F and holding for ~5 hours, then switch.ing to 5% 2 in N2 for 0.5 hours, then to 100% air for 3 hours, then 3 hours i~ 100~ oxygen before cooling.
PART D
The procedure of Part C was repeated to provide a finished catalyst containing 0.5 weight percent silica.
PART E
A portion of the calcined extrudate of Part B was impregnated with platinum, rhenium and chloride in the same manner as the catalyst of Part B to provide a duplieate product.
152 g. of this calcined catalyst was vacuum impregnated with a solution of 2.6 g. of tetraethyl orthosilicate diluted to 130 ml.
with denatured ethanol. After 2 hours of equilibration, the catalyst was dried in an oven for 1 hour at 125C. Calcination was carried out under carefully controlled flow conditions to oxidize the organic matter in 1% O2/99%N2 while increasing the temperature to 950F over a period of 4 hours, then 1 hour at 950F in 5% 02/95% N2, then 2 hours in air at 950F, and 20 finally 3 hours in 100~ 2 at 950F.
PART F
The high-boehmite precursor of Part B was mulled as in Part B, except that an amount of tetraethyl orthosilieate sufficient to provide 1.0 weight pereent siliea in the finished eatalyst was dissolved in 3/1 ethanol/water and added to the dilute nitric acid solution used for mulling. The extrusion, calcination, i~pregnation and recalcination steps were carried out as in Part B.
PART G
The high-boehmite precursor of Part B was mulled as in Part F to provide a finished catalyst containing 2.0 weight percent silica.

1 ~5~3~.

The relative effects of surface area-stabilizing additives were determined by holding the above catalyst samples at 1100F in flowing air, containing 1.4-1~5 psia water to accel-erate loss of surface area. Samples were withdrawn at various times over a run length of up to 550 hours. Data were analyzed to provide surface area data for an included run length of 400 hours and for an ex~rapolated run length of 104 hours. The standard deviation of the average area is approximately 2 m.2/g.
Data are presented in Table I.
Reforming tests were conducted by holding the above catalyst samples at 950F while processing a standard naphtha feedstock at 4 weight hourly space velocity, under a pressure of 300 psig, and at a hydrogen rate of 3 moles/mole naphtha feed.
The catalyst was first reduced in flowing hydrogen gas for 16 hours at 1 atmosphere pressure and was presulfided to provide one atom of sulfur per atom o. rhenium. Chloride was added to the feed to maintain 1.5 weight percent chloride on the catalyst.
A sample of reformate was taken when conditions had stabilized and the clear research octane number was determined on the 20 debutanized product. Data are presented in Table I.
TABLE
STABILITY AND REFORMING TESTS
Part Catalyst Composition Surface Area (m.~/g.) Octane No.

Precursor Silica Boehmite, wt % wt % 400 hrs __ hrs RONC
A 52 0.0 -- - 101.9 A' 52 0.0 140 113 102.0 B 82 0.0 157 127 C 52 0.25148 119 102.8 D 52 0.5 157 131 E 82 0.5 166 142 F 82 1.0 175 151 99.8 G 82 2.0 185 159 1 158~3~

Other comparable tes~s, not otherwise described herein, have shown that large quantities of silic~ ~ca~ 4 weight percent) have a stabilizing effect on surface area, but reduce reforming activity. Similarly, phosphorus compounds, alkaline earth oxides, and some rare earth oxides significantly diminish reforming activity.
Based on correlati~ns with commercial testing, a difference of 10 m.2/g corresponds to the difference between 5.5 and 7 cycles, or about 25~ greater stability.
The octane numbers observed for the reforming tests described above were substantially the same, indicating that substantially identical activities were obtained in each instance.
EXAMPLE II
PART A
An alumina support material, comprising 82 weight percent boehmite and having an average crystallite diameter of 45 Angstroms as a powder, was blended with sufficient dilute (2 volume percent) nitric acid to partially peptize the alumina and with sufficient water to provide a thick paste. Mixing was further carried out in a muller with the addition of water as required to provide an extrudable mi~ture. After mulling for 30 minutes, the mixture was extruded through a die plate having 1/6-inch diameter holes. After drying 12 hours in a forced air drying oven at 125C, the extrudate was broken into small particles and calcined for 2 hours at 300F furnace temperature.
Temperature was then raised to 950F, at a rate of 300F/hour, and maintained for 3 hours. During calcination, dry air was passed through the oven at 1000 VHSV.
The calcined extrudate particles were impregnated with a solution (0.75 ml./g.) containing chloroplatinic acid, perrhenic acid, and hydrochloric acid in amounts calculated to '.:

3 ~ 5863 ~

provide 0.6 weight percent platinum, 0.85 weight percent rhenium, and 1.5 weight percent chloride iD ~he finished catalyst. The volume of solution was selected to be slightly greater than the pore volume of the extrudate particles. A final calcination was carried out as before.
PART B
The high-boehmite alumina support material of Part A
was mulled as in Part A with the addition of a filtered dilute aqueous solution of commercially available water glass (sodium silicate) in an amount sufficient to provide 1.0 weight percent silica in the calcined product. Additional dilute nitric acid was added to the mull in an amount calculated to substantially neutralize the alkalinity of the sodium silicate; i.e., the weight ratio of silica to sodium oxide should be greater than 100/1 and preferably greater than 300/1. After extrusion, drying and calcination, the support material was extracted with water to substantially remove the sodium ions therefrom. The washed support material was then calcined under the same conditions employed prior to the washing step.
The relative effects of surface area-stabilizing additives were determined by holding the above catalyst samples at 100F in flowing air, containing 1.4-1.5 psia water to accel-erate loss of surface area. Samples were withdrawn at various times over a run length of up to 550 hours. Data were analyzed to provide surface area data for an included run length of 400 hours and for an extrapolated run length of 104 hours. The standard deviation of the average area is approximately 2 m.2/g.
Data are presented in Table I.

~ ~58B~

TABLE I
STA~ILITY TESTS
Part Catalyst C~mpositio~ Surface Area (m.2/g.) .
Precursor Si~ica Boehmite, wt % wt % 400 hrs 104 hrs _ A g2 0.0 147 127 B 82 1.0 179 147 Other comparable tests, not otherwise described herein, have shown that large quantities of silica (ca. 4 weight percent) have a stabilizing effect on surface area, but reduce reforming activity. Similarly, phosphorus compounds, alkaline earth oxides, and some rare earth oxides significan~ly diminish reforming activity.
Based on correlations with commercial testing, a difference of 20 m.2/g., as between Parts A and B in Table I, above, corresponds to the difference between 5.5 and 8.5 cycles, or about 50% greater stability.
EXAMPLE III
PART A
An alumina support material, comprising 52 weight percent boehmite and having an average crystallite diameter of 34 Angstroms, as a powder, was blended with water to provide a thick paste. Mixing was further carried out in a muller with the addition of water as required to provide an extrudable mixture. After mulling for 30 minutes, the mixture was extruded through a die plate having l/6-inch diameter holes. After drying 12 hours in a forced air drying oven at 125C, the extrudate was broken into small particles and calcined for 2 hours at 300F
furnace temperature. ~emperature was then raised to 950F, at a rate of 300F/hour, and maintained for 3 hours. During calcina-tion, dry air was passed through the oven at 1000 VHSV.
The calcined extrudate particles were impregnated with a solution (0.75 ml./g.) having a pH = 7, containing 1 ~5~3~

chloroplatinic acid, perrhenic acid and ammonium chloride in amounts calculated to provi~e 0.6 weight percent platinum, 0.85 weight percent rhenium, and 1.5 weight percent chloride in the finished catalyst. The volume of solution was selected to be slightly greater tha~ the pore volume of the extrudate parti-cles. A final calcination was carried out as before.
PART A' The procedure of Part A was repeated to provide a finished catalyst containing 0.35 weight percent platinum and 0.35 weight percent rhenium.
PART B
An alumina support material, comprising 82 weight percent boehmite and having an average crystallite diameter of 45 Angstroms as a powder, was blended with sufficient dilute (2 volume percent) nitric acid to partially peptize the alumina and with sufficient water to provide a thick paste. Mixing was further carried out in a muller with the addition of water as required to provide an extrudable mixture. After mulling for 30 minutes, the mixture was extruded through a die plate having 1/6-inch diameter holes. After drying 12 hours in a forced air drying oven at 125C, the extrudate was broken into small parti-cles and clacined for 2 hours at 300F furnace temperature.
Temperature was the~ raised to 950F, at a rate of 300F/hour, and maintained for 3 hours. During calcination, dry air was passed through the oven at 1000 VHSV.
The calcined extrudate particles were impregnated with a solution (0.75 ml./g.) containing chloroplatinic acid, perrhenic acid, and hydrochloric acid in amounts calculated to provide 0.6 weight percent platinum, 0.85 weight percent rhenium, and 1.4 weight percent chloride in the finished catalyst. The volume of solution was selected to be slightly greater than the pore volume of the extrudate particles. A

3 ~

final calcination was carried out as before.
PART C
The high-boehmite alumina support material of PART
B was mulled ~s in PART B with the addition of a commercialy available low-sod}~m colloidal silica dispersion in an amount sufficient to provide 1.0 weight percent silica in the calcined product. After mulling,, the support material was prepared and converted into a catalyst as in PART B.
PART D
The procedure of PART C was repeated with the addition of low-sodium colloidal silica dispersion in an amount sufficient to provide 2.0 weight percent silica in the calcined product.
PART E
The procedure of PART A was repeated with the addition, during mulling, of low-sodium colloidal silica disper-sion in an amount sufficient to provide 2.0 weight percent silica in the ca~cined product.
The relative effects of surface area-stabilizing additives were determined by holding the above catalyst samples at 1100F in flowing air, containing 1.4-1.5 psia water to accelerate loss of surface area. Samples were withdrawn at various times over a run length of up to 550 hours. Data were analyzed to provide surface area data for an included run length of 400 hours and for an extrapolated run length of 104 hours.
The standard deviation of the average area is approximately

2 m.2/g. Data are presented in Table I.
Reforming tests were conducted by holding the above catalyst samples at 950F while processing a standard naphtha feedstock at 4 weight hourly space velocity, under a pressure of 300 psig, and at a hydrogen rate of 3 moles/mole naphtha feed. The catalyst was first reduced in flowing hydrogen gas ` I ~S8~3~

for 16 hours at 1 atmosphere pressure and was presulfided to provide one atom of sul~ur per atom of rhenium. Chloride was added to the feed to maintain 1.5 weight percent chloride on the catalyst. A sample of reformate was taken when conditions had stabilized and the clear research octane number was determined on the debutanized product. Data are presented in Table I.
TABLE I
STABILITY AND REFORMING TESTS
-10 Part Catalyst Composition Surface Area (m.~ 9.) Octane No.

Precursor Silica Boehmite, wt ~ wt % 400 hrs _4 hrs RONC
A 52 0.0 - - 101.9 A' 52 0.0 140 113 B 82 0.0 157 127 102.0 C 82 1.0 166 142 102.1 D 82 2.0 179 160 E 52 2.0 172 151 Other comparable tests, not otherwise described herein, have shown that large quantities of silica (ca. 4 weight percent) have a stabilizing effect on surface area, but reduce reforming activity. Similarly, phosphorus compounds, alkaline earth oxides, and some rare earth oxides significantly diminish reforming activity.
Based on correlations with commercial testing, a difference of 10 m.2/g. corresponds to the difference between 5.5 and 7 cycles, or about 25% greater stability.

The octane numbers observed for the reforming tests described above were substa~tially the same, indicating that substantially identical activities were obtained in each instance.

Claims (21)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A reforming catalyst comprising a noble metal on a modified alumina support derived from an alumina precursor, wherein the modified alumina support precursor comprises at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and from about 0.25 to about 5.0 a weight percent silica.

2. The catalyst of claim 1 wherein the silica is supplied as a silica precursor selected from the group compri-sing orthosilicate ester, alkali metal silicate and low-sodium colloidal silica.

3. The catalyst of claim 1 wherein the modified alumina support precursor comprises at least about 80 weight percent boehmite.

4. The catalyst of claim 1 wherein the modified alumina support comprises from about 0.5 to about 2.0 weight percent silica.

5. The catalyst of claim 1 wherein the noble metal is platinum.

6. The catalyst of claim 4, additionally comprising rhenium.

7. A reforming catalyst, comprising from about 0.3 to about 1.5 weight percent platinum and from about 0.3 to about 2.0 weight percent chloride, on a modified alumina support, derived from an alumina precursor, wherein the modified alumina support precursor comprises at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and from about 0.25 to about 5.0 weight percent silica, the silica being added to the boehmite-containing alumina as an aqueous colloidal dispersion.

8. The catalyst of claim 7 wherein the aqueous colloi-dal dispersion is added to the alumina prior to extrusion and calcination of the modified alumina support.

9. A reforming catalyst, comprising from about 0.3 to about 1.5 weight percent platinum and from about 0.3 to about 2.0 weight percent chloride, on a modified alumina support, derived from an alumina precursor, wherein the modified alumina support precursor comprises at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and from about 0.25 to about 4.0 weight percent silica, the silica being added to the boehmite-containing alumina as an orthosilicate ester.

10. The catalyst of claim 9 wherein the orthosilicate ester is added to the alumina prior to extrusion and calcination of the modified alumina support.

11. The catalyst of claim 9 wherein the orthosilicate ester is added to the finished catalyst prior to a final calcination.

12. The catalyst of claim 9 wherein the orthosilicate ester is ethyl orthosilicate.

13. A reforming catalyst, comprising from about 0.3 to about 1.5 weight percent platinum and from about 0.3 to about 2.0 weight percent chloride, on a modified alumina support, derived from an alumina precursor, wherein the modified alumina support precursor comprises at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and from about 0.25 to about 5.0 weight percent silica, the silica being added to the boehmite-containing alumina as an aqueous solution of an alkali-metal silicate, followed by washing to remove the alkali metal.

14. The catalyst of claim 13 wherein the alkali-metal silicate is added to the alumina prior to extrusion and calcina-tion of the modified alumina support.

15. The catalyst of claim 13 wherein the alkali-metal silicate is sodium silicate.

16. The process for reforminq a naphtha fraction which comprises contacting the naphtha fraction under reforming conditions and in the presence of hydrogen with a reforming catalyst comprising a modified alumina support, derived from an alumina precursor, having distributed thereon from about 0.3 to about 1.0 weight percent platinum, from about 0.3 to about 1.0 weight percent rhenium, and from about 0.5 to about 2.0 weight percent chloride, wherein the modified alumina support precursor comprises at least about 75 weight percent boehmite, having an average crystallite diameter no greater than about 60 Angstroms, and the modified alumina support additionally comprises from about 0.25 to about 5.0 weight percent silica.

17. The process of claim 16 wherein the modified alumina support comprises at least about 80 weight percent boehmite.

18. The process of claim 16 wherein the modified alumina support comprises from about 0.5 to about 2.0 weight percent silica.

19. The process of claim 16 wherein the reforming catalyst comprises about 0.6 weight percent platinum, about 0.85 weight percent rhenium, and about 1.4 weight percent chloride.

20. The process of claim 16 wherein the average crystallite diameter of the boehmite component is within the range from about 30 to about 60 Angstroms.

21. The process of claim 16 wherein the silica is derived from a silica purcursor selected from the group compri-sing orthosilicate ester, alkali metal silicate and low-sodium colloidal silica.