US2573149A - Three-stage catalytic process for the reforming of gasoline - Google Patents

Description

L. S. KASSEL oct. 30, 1951 THREESTAGE CATALYTIC PROCESS FOR THE REFORMING OF' GASOLINE Filed Nov. 20, 1948 NNUWAM' Patented Oct. 30, 1951 THREE-STAGE CATALYTIC PROCESS FOR THE REFORMING OF GASOLINE Louis S. Kassel, Riverside, Ill., assignor to Universal Oil Products Company, Chicago, Ill., a

corporation of Delaware Application November 20, 1948, Serial No. 61,155

Claims. (Cl. 196-49) This invention relates to a method of conducting endothermic catalytic reactions of organic compounds, particularly substantially olefin-free hydrocarbon fractions containing naphthenes and boiling approximatelywithin the gasoline range. It is more specically concerned with a particular method of reforming straightrun gasolines and naphthas in the presence of hydrogen and a platinum-containing catalyst.

In a number of processes for the conversion of organic compounds, it has been noticed that the initial stages of the reaction are highly endothermic whereas the later stages are less endothermic and may approach thermal neutrality or may even be mildly exothermic. This is particularly true in the reforming of straight-run gasolines and naphthas in the presence of hydrogen and platinum-alumina catalysts. In reforming processes of this type, the octane number increase appears to vbe the result of two separate and independent reactions, namely, aromatization and hydrocracking. Aromatization is a highly endothermic and reversible reaction, and as such has a high reaction rate temperature coeicient. Hydrocracking, on the other hand, is a nonreversible reaction which is mildly exothermic. Its reaction rate is less affected by change in reaction temperature. Accordingly, the balance between the two reactions is very markedly inuenced by the temperature at which A the reactions are carried out, high temperatures tending to favor aromatization and low temperatures tending to favor hydrocracking because of the unfavorable eifect of the low temperature on the naphthene-aromatic equilibrium as well as the effect of temperature on the reaction rates.

The yield-octane vincrease relationship resulting from aromatization is approximately linear over a wide range of octane increases. This is due to the loss of a portion of the mass of the charge stock to hydrogen, and in part to a density increase inherent in changing naphthenes to aromatics. Hydrocracking, on the other hand, involves a combination of cracking, and hydrogenation of the fragments thereby produced. In the initial stages, this results in the formation of lower molecular weight gasoline hydrocarbons which are of lower density than the charge.

Thus the initial octane increases due to hydro` cracking often are obtained at a volumetric recovery or yield greater than 100%. As the reaction proceeds, however, there is an increase in the tendency to produce gaseous hydrocarbons, which results in a loss in yield without any appreciable increase in octane number. Hydrocracking, therefore, is an efcient reforming process if not carried to extremes, but becomes very inefficient when an attempt is made to secure high octane products principally by this means. In general, to obtain the optimum yield at a given product octane number, it is necessary to control the balance between the hydrocracking and aromatization reactions to limit the hydrocracking to the degree at which it is eiiicient and to obtain the remainder of the octane number increase by aromatization. However, at high octane levels, the maximum yields are obtained with some stocks by securing the maximum extent of aromatization and limiting the hydrocracking to that needed to meet volatility requirements.

Since the same types of hydrocarbons which are susceptible to aromatization are also susceptible` to hydrocracking, it is important in either case that the aromatization reaction be carried out first, since if it is not, some of theseV particular hydrocarbons will be hydrocracked in the early stages, and it is impossible to obtain a maximum of aromatization or maximum yield for a given octane number.

When adiabatic reaction chambers are used with intermediate heating of the charge it has been found that the aromatization reaction takes place very rapidly in the forepart of the ilrst reactor, but that due to the high endothermicity of this reaction the temperature is lowered to a y point at which equilibrium prevents formation of more aromatics and hence tends to be relatively more favorable to hydrocracking. Subsequent heating again tends to induce aromatization. However, the hydrocracking which has taken place in the cold section of the rst reactor has destroyed some material which might potentially be aromatized, with the final result that the potentialities for the production of high octane numbers are less for a reactor system of this type than would have been the case if the temperature had not been permitted to drop.

Recently a more desirable type of reactor system has been proposed, namely, one in which heat is supplied during the initial stages of the reaction at a rate to substantially retard or prevent the temperature drop due to the aromatizaftion reaction, thus'permitting said reaction to proceed substantially to equilibrium or to the degree desired.- Subsequently, the hydrocarbons which have not been aromatized may be hydrocracked by passing the reaction mixture from said rst stages of the reaction zone into an adiabatic reaction zone.V With this arrangement,

heat is supplied to the catalyst bed during the initial stages of the reaction and the balance of the reaction is allowed to proceed in the presence of catalyst under substantially adiabatic conditions. Ordinarily, the aromatization reaction is not carried to equilibrium in order that a small quantity'of` naphthenes, of theorder of 15.-20% of the amount originallypresent, may be present in the charge to the adiabatic reaction zone. In such cases, the endothermic heat of dehydrogenation of naphthenes to aromatics approximately offsets the exothermic heaty of hydrocracking. The result is that the overall reaction in the adiabatic section is close to thermalA neutrality.

By using an operation of this type, itis necessary to supply heat only to the initial stages of the reaction and, therefore, a great deal of the expense associated with completely isothermal reaction systems is thereby avoided. The heated reaction zone may advantageously contain a catalyst that possesses primarily aromatization activity and relatively less hydrocracking activity; whereas the catalyst in the adiabaticA reaction Zone may possessv relatively more hydrocracking activity.

High heat input rates. are needed during the initiall stages ofy the reaction in order to. olset the high degree of endothermicity of the reaction;` It hasy been found thatl these high heat rates can most conveniently and economically be achieved by means of radiant heating. One such method ofheatingv comprises placing a minor proportion of the catalyst, say about 20%, inl a plurality oftubesv arranged vertically around the inner periphery of the walls of. a substantially cylindrical verticalheating chamber and subjecting said tubes to radiant heat from a radiant name and'` thev resulting hot products of: combustion; directed inl a central longitudinal` path throughsaid chamber, while passingl the charging stock through said tubes. Again, the feedx may be initially contacted with a; minor amountof catalyst disposed in vertical tubes,- arranged ina` row and iianked' by parallel radiating walls, oneorbothl ofl which` is heated by a name and the resulting. hot products of combustion; The4v flow of reactants throughthe tubes? and the direction of propagation of` the flame and hotl products of combustion usually are in the same direction, which direction ordinarily is downwardly. The reason for this is that the initial stages of4 this reaction are more highly endothermic than subsequentstages and, consequently, a greater` heat: input is` needed during said initial stages of the reaction in order to maintain thereaction temperature substantially constant. Thisl isi accomplished by having those sections of thee catalyst where the initial reaction takes place nearest the. flame, the flame being at a higher temperature than thev combustion products; and`v consequently de'- livering or radiating a greater amount of heat.

`As the reaction proceeds the endothermicity befractions thereof in. the` presence of; hydrogen and platinum-alumina` catalyst of a type that willwbe more fullyi described hereinafter, the

yprocessl isv substantially nonregenerative and may be operated continuously:- without'y regeneration for weeks and even months at; av time.

This is one of the principal reasons that radiant heating of the first part of the catalyst bed is attractive. There is, however, one serious potential hazard in radiant heating. If a portion of the catalyst starts to become disproportionately inactivated because of the deposition thereon` of hydrocarbonaceousmaterials or because of poisoning due to contaminants in the charging stock, a critical situation is presented. The portion of the catalyst that has lost more activity will not promote the endothermic reaction to` thev same extent as the remainder of the catalyst, and, since it is subjected to a large amount of radiant heat, the temperature thereof and of the reactants flowing therethrough will rise rapidly. This increase of itself will speed up the rate of carbon deposition in that section andV in the sections immediately following. In addition, a certain amount of thermal cracking will be induced, and the olefinic hydrocarbons thus produced will tend to cause'Y further contamination of the catalyst in that and subsequent sections. Thus it can be seen that-it the section of the catalyst near the upper end ofthe tubes becomes deactivatedby carbonaceous deposition or by poisoning, the effect is not only cumulative, but deactivation tends to spread through the bed atI anl increasing rate; For this reason it is essential that undesirable constituents inthe charging stock that promote deactivationv and poisoning beY removed. therefrom before the charging stock is passed into the radiantly heated section of. the plant. My invention relates to an endothermic conversion process of the type described wherein these harmful andl undesirable constituents of the charging stock are effectively removed before saidl charging stock isv contacted with the catailyst inthe radiant-heated reactor.,

In one embodiment my inventionv relates' toa process which comprises effecting the endo:- thermic conversion of an organic reactant by .passing said reactant at endothermic reaction conditions and in series flow throughsolid' catalyst contained in a first substantiallyI adiabatic reaction zone, a. reaction Zone to` which' heat is supplied', anda second substantially adiabatic reaction zone.

In a more specific embodiment, my invention relates to a reforming process which comprises passing hydrogen and a substantially olefinfree hydrocarbon reactant containingv naphthenes and boiling approximately withinv the gasoline range at a conversion temperature within the range of from about 750 F. to about 1000 F. through an adiabatic reaction Zone containing a minor proportion of' acatalyst comprising platinum and alumina, then in parallel flow through a plurality of heat absorbingl tubes containing a minor proportion of catalyst comprising platinum and aluminasaid tubes being in substantially uniform radiative relation to a radiant flame and the resulting hot products 0f combustion, and finally throughV a major proportion of catalyst comprising platinum and alumina maintained at substantially adiabatic conditions.

In one of its primary aspects, my inventionA is concerned with a method for prolonging the life of catalyst that is used inthe endothermic reforming of straight-run gasolinesY or fractions thereof, and wherein atleast a portion ofthe endothermic heat of reforming is supplied from a source of radiant heat. For-example; if the catalyst is disposedin, tub esthat: areinsuhstantially uniform radiative relation to a radiant flame and the resulting hot products of combustion, the apparatus and processing conditions are designed to attain relatively high rates of heatv input to the catalyst and the charging stock ow therethrough, If a portion of the catalyst in one of the tubes becomes inactivated due to the presence of carbon-forming constituents or poisons in the charging stock, the amount of reaction in that section decreases and a greater proportion of the heat input is utilized in raising the temperature of the charging stock. As heretothis necessary purification of the feed by subjecting it under adiabatic conditions to the action of catalyst that reacts with and removes catalyst- Y inactivating constituents therefrom. Only a relatively minor amount of catalyst need be used in this treating step since the capacity of the catalyst to react with poisons of this type is much greater than the amount of poison that will cause severe deactivation of the catalyst. lOrdinarily I use the same type of catalyst in the pretreater that I employ in the main conversion zones of the unit.

The hydrocarbon stocks that may -be converted in accordance with my process comprise nonoleflnic hydrocarbon fractions containing saturated hydrocarbons, particluarly naphthenes. By the term nonoleiinic I mean substantially olefin-free, i. e., a few per cent of olens can be present in the charge in some types of operation. Suitable stocks include narrow boiling fractions rich in naphthenes as well as substantially pure naphthenes such as cyclohexane and methylcyclohexane. Preferred stocks are those consisting essentially of naphthenes and paraiiins, although relatively minor amounts of aromatics also may be present. The naphthenes are dehydrogenated to aromatics and the paraflins are hydrocracked to lower boiling parains. This preferred class includes straight-run gasolines, natural gasolines, and the like. The gasoline may be a full boiling range gasoline having an initial boiling point within the range of from about 50 to about 100 F. and an end boiling point within the range of from about 325 to about 425 l.y or it may be a selected fraction thereof which usually will be a higher boiling fraction commonly referred to as naphtha, and generally having an initial boiling point of from about 125 to about 250 F. and an end boiling point within the range of about 350 F. to about 425 F. The expression straight-run gasoline fraction as used herein is intended to include both naphthas and full boiling gasolines.

The catalysts comprising platinum and alumina that are preferred for use in my process may contain substantial amounts of platinum, but, for economic as well as for product yield and quality reasons, the platinum content usually will be Within the range of from about 0.05% to about 5.0%. A particularly effective catalyst of this type contains relatively minor amounts, usually less than about 3% on a dry alumina basis,'of a halogen, especially chlorine or fluorine.- l One method of preparing the catalyst comprises 'wadding a suitable alkaline reagent such as ammonium hydroxide or carbonate to a salt of aluminum,

suchv as aluminum chloride, aluminum sulfate, aluminum nitrate, and the like, in an amount sufficient to form aluminum hydroxides, which upon drying, can be converted to alumina. The halogen may be added to the resultant slurry in the form of an acid such as hydrogen fluoride or hydrogen chloride, or as a volatile salt such as ammonium fluoride or ammonium chloride. The amount of combined halogen in the finished catalyst usually is maintained within the range of from about 0.1%' to about 8% by weight of the alumina on a dry basis. The uoride ion appears to be somewhat more active than other members of the halide group and, therefore, may be present vin a lower concentration, within the range of from about 0.1% to about 3% by weight of the alumina on a dry basis. The amount of chloride ion incorporated into the catalyst composite will be within the range of .from about 0.2% to 8%,

. preferably from about 0.5 to about 5 by weight of the alumina on a dry basis.

- A satisfactory method of adding platinum to thealumina-halogen composite comprises preparing a colloidal suspension of platinic sulde by introducing hydrogen sulfide into an aqueous solution of chloroplatinic acid until said solution reaches a constant color, which usually is a dark brown. The resultant colloidal suspension of platinic sulfide is commingled with the aluminum hydroxide slurry at room temperature followed by stirring to obtain intimate mixing. The resulting material is then dried at a temperature of from about 200" to about 400 F. for a period of from about 4 to about 24 hours or more to form a cake. The resulting material may then be converted into pills or other shaped particles. Thereafter the catalysts may be subjected to a high temperature calcination or reduction treatment prior to use. It is to be understood that the foregoing method of preparing satisfactory platinum-alumina catalysts is merely illustrative and is not to be taken in a limitative sense inasmuch as various other methods may be employed to produce satisfactory catalysts of this type.

The use of the term catalyst comprising platinum and alumina in the specification and appended claims is intended to include platinumalumina composites of the type described above including those containing minor amounts of a halogen. The exact manner in which the halogen or halide ion is present in the catalyst is not known although it is believed to be present -in the form of a chemical combination or loose complex with the alumina and/ or platinum components. Because the exact chemical constitution of such halogen-containing catalysts is not known, I sometimes refer to them as "catalysts comprising platinum, alumina, and a halogen. It is known, however, that the presence of a small amount of a halogen in the catalyst enhances the hydrocracking activity thereof; for platinum-alumina composites that are substantially halogen-free possess very little ability to promote hydrocracking.

Other platinum-containing catalysts that may be used in my process, although not necessarily with equivalent results, include platinum on charcoal, platinum on silica, platinum on asbestos, and platinum on bases or carriers that possess cracking activity such as silica-alumina composites. The corresponding palladium catalysts occasionally may be used with advantage in my process.y

Although the Vabove description has been directedvprimarily.k to catalytic composites conazzame 7i taining platinum,y my invention is adapted to processes employing other catalysts, for example, catalytic composites comprising acompound of the metals of the left-hand column of groups V and VI of the periodic table and' in particular, the oxides of chromium, molybdenum, tungsten or vanadium, either alone or in admixture with one another and a suitable refractory supporting material such as alumina, magnesia, silica, or

mixtures thereof. The particular processing conditions of temperature and pressure, etc. employed in any specic operation will, of course, vary somewhat dependingl upon the catalyst'used in the operation.

Hydrocarbon reforming operations carried out in accordance with my process in the presence of catalyst comprising platinum and alumina ordinarily will be conducted at temperatures of from about 750 F. to about 1000a F. At temperatures inthe vicinity of 750 F. and lower, the aromaticnaphthene equilibrium is, unfavorable, the reaction rates are quitev low, and very loW space velocities must be employedto obtain appreciable conversion. At temperatures in excess of about 1000 F. an appreciable amount of thermal reaction takes place accompanied by a poorer liquid' recovery and. more rapid catalyst deactivation.

The pressure at which my process will be conducted when employing platinum-alumina catalyst for the reforming off hydrocarbons usuallyv will be within the range of from about 50 to about 1200 pounds per'square inch; aweight hourly space velocity, which is definedA as the Weight of hydrocarbon charge per hour per weight of catalyst in. the reaction zone, that usually-will be within the range of from about' 0.2 to about 40; and the amount of hydrogen charged along with the hydrocarbons usually will'be from aboutY 0.5 to about'A 15 mols per mol of hydrocarbon.

Additional features and advantages of my invention will be apparent from thefollowing description of the attached drawing whichv illustrates a particular method for conducting a hydrocarbon reforming operation in accordance with the present invention.

Referring` to the drawing, a, W sulfur, fullboiling range straight-run gasoline which contains naphthenes and4 parains is charged through line I and;y is picked up by pump 2, passed through line B'containing valvefd, and is joined` by a` stream of: recycle hydrogen produced as hereinafterv described; The combined stream of straight-rungasoline andv hydrogen is passed through heater 5 wherein itis heated to a. temperature of. about 850F. The effluent from the heater is withdrawn through line 6 and charged to adiabatic reactor 1. This reactor contains a bed of platinum-alumina-uorine catalyst in the form of Ik x 1/8 pellets. The amount of catalyst in this reactor represents about 5% of the total amount in the system. Ordinarily, the quantity of catalyst in this treating zone will befrom about 1% to about 15-20% ofthe total catalyst inventory. Asthe charging stock passes through the catalyst a substantial portion ofthe poisons and catalyst-inactivating constituents are removed therefrom and adsorbed or deposited upon the catalyst. Only a small amount of conversion takes place because the space velocity in this reactor is comparatively high and because the endothermicity of theY reaction that does take place lowers the temperature toa point atwhich. the-reaction rate is relativelyv slow. If

the. vreactor were isothermal, or if a substantial portion of the endothermic heat of reaction were supplied from `an external source, a considerably greater degree of conversion would be obtained therein. Such a high degree of conversion is not necessary in order to remove catalyst contaminants from the charging stock and the major extent of conversion would of itself tend to deactivate the catalyst by causing the decomposition of carbonaceous deposit. This would shorten the life of the catalyst, thereby causing more frequent replacement of the catalyst. This larger catalyst consumption would to a large extent defeat the object of the present invention, namely,

.- the increasing of catalyst life with the concomitant decrease in catalyst consumption.

The effluent from reactor I is Withdrawn through line 8 containing valve 9 and is charged into header IU of heated reactor II. Reactor l-I' comprises a furnace of the down-draft type and is provided with a burner at the top thereof. Ver-- tically disposed tubes are arranged adjacent the wall of the furnace so as to be heated predominantly by radiant heat. A suitable gaseous or liquid fuel is charged to heater II through line I2 containing'valve I3 and the products of combustion are removed via line I4. The tubes are filled withl/'l x 1/8 pellets of platinum-aluminafluorine catalyst. The amount of catalyst in the tubes represents about 20% of the total amount in the plant. The tubes are disposed onV a circumference of a circle, the center of which substantially corresponds to the center of the combustion chamber. The flame and the products of combustion flow in a central longitudinal unobstructed path through the chamber substantially out of direct contact with said tubes.

I prefer to have the reactants flow in the same direction as Athe ame and the hot products of combustion. This is for the reason' that the initial stages of the reaction are more highly endothermic than subsequent stages and, consequently, a greater heat input is needed during the first stages of the reaction in order to maintain the reaction temperature substantially constant. This is accomplished by having those sections of the catalyst wherein the initial reaction takes place nearest the flame, the flame being at a higher temperature than the combustion product and consequently delivering or radiating a greater amount ofheat. As the reaction proceeds the endothermicity becomes less and the amount of heat needed to maintain the reaction temperature substantialy constant is accordingly reduced. The hot combustion products, which are cooler than the flame, are adequate for this purpose. Since it is better engineering practice tol pass the hydrocarbon reactants and the hydrogen downwardly through the catalyst bed in order to avoid lifting and bumping of the bed, I prefer to pass both the reactants and the hot combustion products in a downward direction through heater' I I.

Although I have described one methodl of supplying heat to this section of the unit, it is to be understood that any suitable method of supplying-heat may be employed. For example, a somewhat diiferent heated reactor may be used from the one shown in the drawing, said reactor comprising a number of vertical tubes iilledwith catalyst and flanked by parallel radiating walls, one or both of which is heated by a flame and hot products of combustion.

TheA effluent from the tubes in reactor II is passedthrough header I5V and line Iii-containing valve-I1y into adiabatic reactor I8'. The reactor contains platinum-alumina-fluorine catalyst and the reaction is completed as the material from line I 6 passes through the catalyst bed. The amount f catalystin this reactor represents about 75% of the total amount of catalyst in the plant. Reactor I 8 is insulated and is substantially adiabatic. The overall reaction that takes place in this reactor is substantially thermally neutral. However, it may be somewhat endothermic or even somewhat exothermic depending upon the degree of conversion obtained in the preceding stages, upon the operating conditions maintained, and the like.

The efliuent from reactor I3 passes through line I9 containing valve 20 and through the cooler 2I into receiver 22 wherein a separation is effected between the gas and liquid. The gas, which comprises about 90% hydrogen, the remainder being light hydrocarbons, is withdrawn from receiver 22 through line 23 containing valve 24. It is picked up by compressor 25 and returned to heater via line 26 containing valve 2l and line 3. Make-up hydrogen may be added to or excess hydrogen may be withdrawn from the system through line 28 containing valve 29. The liquid hydrocarbon layer in receiver 22 is withdrawn through line 25 containing valve 2'! and is sent to suitable fractionation and storage equipment.

It is to be understood that the process that has just been described is to be taken in an illustrative and not a limitative sense for the reason that a number of variations may be made in the process without departing from the spirit of the invention. For example, the adiabatic portion of the reaction taking place after heated reactor I I may take place in a series of beds instead of in a single bed as shown in the drawing. In a process in which a single endothermic reaction takes place and there is no compensating exothermic reaction, such as in the dehydrogenation of methylcyclohexane to toluene, all of the reaction may advantageously take place in the heated zone, i. e., no adiabatic reactor need follow thereafter. In such cases, of course, it is just as important if not more so that the catalyst-inactivating co'nstituents of the charging stock be removed before the charging stock is passed to the heated zone.

In some cases it may be more convenient to use a separate initial adiabatic reaction zone for each tube in the heated zone. This is readily accomplished for example, by moving the burners of reactor II lower down, so that the upper part of the tubes are not exposed to radiant heat. The tubes are filled with catalyst to a level in the unheated portion. The upper section of each tube is then an adiabatic reaction zone, and the lower section a radiant heated zone.

Again it may be desirable to heat or cool the material flowing from one reactor to another.

The method that I have described is capable of producing a high yield of high octane number reformate at an increased catalyst life. It makes practicable the use of a radiant heated reaction zone for endothermic catalytic reactions by removing from the charging stock, catalyst-inactivating components that would seriously interfere with the operation.

I claim as my invention:

1. A process for reforming saturated hydrocarbons boiling in the gasoline range which comprises passing the hydrocarbons serially through three successive bodies of solid reforming catalyst maintained at from about 750 F. to about 1000 F., maintaining the first catalyst body of the series under substantially adiabatic conditions and therein removing deleterious carbonforming constituents from the hydrocarbons without effecting more than a small amount of reforming in said rst body, subjecting the hydrocarbons to endothermic aromatization in the second catalyst body of the series and radiantly heating said second body to supply at least a portion of the endothermic heat of the aromatization reaction, and subjecting the hydrocarbons to hydrocracking in the third catalyst body of the series under substantially adiabatic conditions.

2. The process of claim 1 further characterized in that said bodies comprise reforming catalyst of substantially the same composition.

3. The process of claim 1 further characterized in that each of said bodies comprises a platinum-containing catalyst.

4. The process of claim 1 further characterized in that each of said rst and second bodies contains only a minor proportion of the total reforming catalyst employed in the process and said third body contains the major proportion of the total reforming catalyst.

5. The process of claim 4 further characterized in that each of said bodies comprises a platinum-containing catalyst.

LOUIS S. KASSEL.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 1,894,140 Wietzel et al. Jan. 10, 1933 2,322,622 Fischer et al. June 22, 1943 2,348,557 Mattox May 9, 1944 2,349,812 Day et al. May 30, 1944 2,378,531 Becker June 19, 1945 2,423,328 Layng July 1, 1947 2,439,934 Johnson et al Apr. 20, 1948 2,479,110 Haensel Aug. 16, 1949

Claims (1)

1. A PROCESS FOR REFORMING SATURATED HYDROCARBONS BOILING IN THE GASOLINE RANGE WHICH COMPRISES PASSING THE HYDROCARBONS SERIALLY THROUGH THREE SUCCESSIVE BODIES OF SOLID REFORMING CATALYST MAINTAINED AT FROM ABOUT 750* F. TO ABOUT 1000* F., MAINTAINING THE FIRST CATALYST BODY THE SERIES UNDER SUBSTANTIALLY ADIABATIC CONDITIONS AND THEREIN REMOVING DELETERIOUS CARBONFORMING CONSTITUENTS FROM THE HYDROCARBONS WITHOUT EFFECTING MORE THAN A SMALL AMOUNT OF REFORMING IN SAID FIRST BODY, SUBJECTING THE HYDROCARBONS TO ENDOTHERMIC AROMATIZATION IN THE

Images