OA11949A - Improved fischer-tropsch activity for non-promotedcobalt-on-alumina catalysts. - Google Patents

Description

119 4 9

IMPROVED FISCHER-TROPSCH ACTIVITY FOR “NON-PROMOTED”COBALT-ON-ALUMINA CATALYSTS

BACKGROUND OF THE INVENTION 1. Technical Field : 5 The présent invention relates to Systems and processes for conducting hydrocarbon synthesis and to cobalt-on-alumina catalysts employed in such processes. 2. Background:

In Fischer-Tropsch processes, a synthesis gas (“syngas”) comprising carbonoxide(s) and hydrogen is reacted in the presence of a Fischer-Tropsch catalyst to produceliquid hydrocarbons. Certain advanced cobalt catalysts hâve proven to be very effectivefor Fischer-Tropsch synthesis. However, for these catalysts, extensive promotion withnoble and/or near noble metals has been required in order to enhance the reducibility ofthe cobalt to an extent sufficient to achieve acceptable Fischer-Tropsch conversionactivities. Due in significant part to the cost of obtaining and adding such promoters, 15 these cobalt catalysts hâve typically been quite expensive. Thus, a need presently existsfor a means of significantly reducing the cost of cobalt catalysts useful for Fischer-Tropsch synthesis while maintaining activity levels which are at least comparable tothose heretofore obtained by promoting such catalysts with noble metals.

The “syngas” employed in Fischer-Tropsch processes can be produced, for 20 example, during coal gasification. Processes are also well known for obtaining syngasffom other hydrocarbons, including natural gas. U.S. Patent No. 4,423,265 to Chu et al.notes that the major processes for producing syngas dépend either upon (a) the partialcombustion of the hydrocarbon fuel with an oxygen-containing gas, (b) the reaction of ahydrocarbon fuel with steam, or (c) a combination of these two reactions. U.S. Patent 1 119 4 9

No. 5,324,335 to Benham et al. explains the two primary methods (i.e., steam reformingand partial oxidation) for producing syngas from methane. The Encyclopedia ofChemical Technology, Second Edition, Volume 10, pages 3553-433 (1966), IntersciencePublishers, New York, N. Y. and Third Edition, Volume 11, pages 410-446 (1980), John 5 Wiley and Sons, New York, N. Y. is said by Chu et al. to contain an excellent summary ofgas manufacture, including the manufacture of synthesis gas.

It has long been recognized that syngas can be converted to liquid hydrocarbonsby the catalytic hydrogénation of carbon monoxide. The general chemistry of theFischer-Tropsch synthesis process is as follows:

W (1) CO+ 2¾ (-CH2-)+H2O (2) 2CO+H2 (-CH2-) + CO2

The types and amounts of reaction products, i.e., the lengths of carbon chains, obtainedvia Fischer-Tropsch synthesis can vary depending upon process kinetics and choice ofcatalyst. ^5 Many attempts at providing effective catalysts for selectively converting syngas to liquid hydrocarbons hâve been disclosed. U.S. Patent No. 5,248,701 to Soled et al.,présents an over-view of relevant prior art. The two most popular types of catalystsheretofore used in Fischer-Tropsch synthesis hâve been iron-based catalysts and cobalt-based catalysts. U.S. patent No. 5,324,335 to Benham et al. discusses the fact that iron- v based catalysts, due to their high water gas shift activity, favor the overall reaction shownin (2) above, while cobalt-based catalysts tend to favor reaction scheme (1).

The current practice is to support the catalytic components on porous, inorganicreffactory oxides. Particularly preferred supports hâve included silica, alumina, silica-alumina, and titania. In addition, other reffactory oxides from Groups ΙΠ, IV, V, VI and ^5 VIII hâve been used as catalyst supports.

As mentioned above, the prevailing practice has been to also add promoters to thesupported catalysts. Promoters hâve typically included noble metals, such as ruthénium,and near noble metals. Promoters are known to increase the activity of the catalyst,sometimes rendering the catalyst three to four times as active as its unpromotedcounterpart. Unfortunately, effective promoter materials are typically quite costly both toobtain and to add to the catalyst. 2 119 4 9

Contemporary cobalt catalysts are typically prepared by impregnating the supportwith the catalytic material. As described in U.S. Patent No. 5,252,613 to Chang et al., atypical catalyst préparation may involve imprégnation, by incipient wetness or otherknown techniques, of, for example, a cobalt nitrate sait onto a titania, silica or aluminasupport, optionally followed or preceded by imprégnation with a promoter material.Excess liquid is then removed and the catalyst precursor is dried. Following drying, or asa continuation thereof, the catalyst is calcined to convert the sait or compound to itscorresponding oxide(s). The oxide is then reduced by treatment with hydrogen, or ahydrogen-containing gas, for a period of time sufficient to substantially reduce the oxideto the elemental or catalytic form of the métal. U.S. Patent No. 5,498,638 to Long pointstoU.S. Patent Nos. 4,673,993, 4,717,702, 4,477,595, 4,663,305, 4,822,824, 5,036,032,5,140,050, and 5,292,705 as disclosing well known catalyst préparation techniques.

Fischer-Tropsch synthesis has heretofore been primarily conducted in fixed bedreactors, gas-solid reactors, and gas-entrained fluidized bed reactors, fixed bed reactorsbeing the most utilized. U.S. Patent No. 4,670,472 to Dyer et al. provides a bibliographyof several référencés describing these Systems.

Recently, however, considérable efforts hâve been directed toward conductingFischer-Tropsch synthesis in three-phase (i.e., solid, liquid, and gas/vapor) reactors. Onesuch System is the slurry bubble column reactor (SBCR). In a SBCR, catalyst particlesare slurried in liquid hydrocarbons within a reactor chamber, typically a tall column.Syngas is then introduced at the bottom of the column through a distributor plate, whichproduces small gas bubbles. The gas bubbles migrate up and through the column,causing a bénéficiai turbulence, while reacting in the presence of the catalyst to produceliquid and gaseous hydrocarbon products. Gaseous products are captured at the top of theSBCR, while liquid products are recovered through a filter which séparâtes the liquidhydrocarbons ffomthe catalyst fines. U.S. Patent Nos. 4,684,756, 4,788,222, 5,157,054,5,348,982, and 5,527,473 référencé this type of System and provide citations to pertinentpatent and literature art.

It is recognized that conducting Fischer-Tropsch synthesis using a SBCR Systemcould provide significant advantages over the reaction Systems commonly employed 3 119 4 9 heretofore. As noted by Rice et al. in U.S. Patent No. 4,788,222, the potential benefits ofa slurry process over a fixed bed process include better control of the exothermic heatproduced by the Fischer-Tropsch reactions as well as better maintenance of catalystactivity by allowing continuons recycling, recovery and rejuvenation procedures to be 5 implemented. U.S. Patent Nos. 5,157,054, 5,348,982, and 5,527,473 also discuss advantages of the SBCR process. However, the slurry bubble column process has beenexpensive to operate, owing in part to the significant catalyst costs required.

SUMMARY OF THE INVENTION

The présent invention provides “nonpromoted” cobalt-on-alumina catalysts 10 unexpectedly and surprisingly having conversion activities at least comparable to those ofthe best promoted formulations. The inventive catalysts also exhibit superior productselectivity characteristics and are particularly effective for use in SBCR processes andother three-phase reaction Systems. This remarkable discovery significantly decreasesthe cost of the Fischer-Tropsch conversion process, as the more expensive promoters 15 need not be utilized to achieve acceptable résulte.

In one aspect, the présent invention provides a cobalt catalyst for hydrocarbon synthesis. The cobalt catalyst comprises cobalt supported on a -alumina support. Thecatalyst is not promoted with any noble metals and is not promoted with any near noblemetals. However, the -alumina support includes a dopant in an amount effective for 2© increasing the activity of the catalyst for hydrocarbon synthesis. The dopant is preferablya titanium dopant.

In another aspect, the présent invention provides a process for hydrocarbonsynthesis comprising the step of reacting a synthesis gas in the presence of a cobaltcatalyst. The cobalt catalyst comprises cobalt supported on a -alumina support. The 25 cobalt catalyst is not promoted with any noble metals and is not promoted with any nearnoble metals. However, the -alumina support includes a dopant in an amount effectivefor increasing the activity of the cobalt catalyst for hydrocarbon synthesis. The dopant ispreferably a titanium dopant.

In yet another aspect, the présent invention provides a cobalt catalyst for 30 hydrocarbon synthesis, wherein the cobalt catalyst comprises cobalt supported on a -alumina support. The cobalt catalyst is not promoted with any noble metals and is not 4 11949 promoted with any near noble metals. However, the cobalt catalyst has been reduced inthe presence of hydrogen at a water vapor partial pressure effective to increase theactivity of the cobalt catalyst for hydrocarbon synthesis. The water vapor partial pressureis preferably in the range of from 0 to about 0.1 atmosphères

In yet another aspect, the présent invention provides a process for hydrocarbonsynthesis comprising the steps of : (a) reducing a cobalt catalyst in the presence ofhydrogen and at a water vapor partial pressure effective to increase the activity of thecobalt catalyst for hydrocarbon synthesis and (b) reacting a synthesis gas in the presenceof the cobalt catalyst. The cobalt catalyst is not promoted with any noble metals and isnot promoted with any near noble metals.

In yet another aspect, the présent invention provides a method of improving theactivity of a cobalt catalyst for hydrocarbon synthesis, wherein the cobalt catalyst has analumina support. The cobalt catalyst is not promoted with any noble metals and is notpromoted with any near noble metals. However, the alumina support includes a titaniumdopant in an amount, expressed as elemental titanium, of at least 500 ppm by weight ofthe total weight of the alumina support.

Further objects, features and advantages of the présent invention will be apparentupon examining the accompanying drawings and upon reading the following descriptionof the preferred embodiments.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 provides a graph comparing the effect of titanium dopant concentrationson the activities of ruthenium-promoted catalysts and “nonpromoted” catalysts forFischer-Tropsch synthesis processes conducted in an SBCR. Each test involved 15-25grams of catalyst which was sieved to 400-150 mesh, calcined, and thenreduced/activated outside of the SBCR System. Each Fischer-Tropsch reaction test wasconducted at 450 psig and 230 C using a synthesis gas flow rate of 15 liters per minute.The synthesis gas was diluted with 60% nitrogen and had a H2:CO ratio of 2.

Figure 2 provides a graph comparing the effects of titanium dopant concentrationson the activities of ruthenium-promoted catalysts and “nonpromoted” catalysts forFischer-Tropsch synthesis conducted in a fixed bed reactor. In each case, the Fischer- 5 119 4 9

Tropsch reaction was conducted at a pressure of 1 atmosphère, a température of 220 C,and a H2/CO ratio of 2.

DETA1LED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Catalvst Compositions

The présent invention provides supported cobalt catalysts which are well suitedfor use in Fischer-Tropsch synthesis. These catalysts are particularly well suited for usein slurry bubble column reactor processes. Examples of preferred, general catalystcompositions provided by the présent invention include, but are not limited to: (a) cobalt,without any noble métal or near noble métal promoter, preferably supported on a doped -alumina, and (b) cobalt promoted with one or more selectivity promoters (preferably analkali promoter and/or a rare earth oxide such as lanthana), but without a noble métal ornear noble métal promoter, and preferably supported on a doped -alumina.

Preferred catalyst compositions comprise (per 100 parts by weight of support):ffom about 10 to about 65 pbw cobalt; from about 0.1 to about 8 pbw potassium (whenprésent); and from about 0.5 to about 8 pbw lanthana (when présent). The catalysts willmost preferably comprise (per 100 parts by weight of support): from about 17 to about 45pbw (more preferably from about 20 to about 40 pbw, and most preferably about 30 pbw)cobalt; from about 0.2 to about 1.0 pbw potassium (when présent); and/or from about 0.9to about 2.1 pbw lanthana (when présent).

The Catalyst Support

The support employed in the inventive catalyst will preferably be -alumina. Wehâve determined that, for cobalt catalysts used in both fixed bed and a slurry bubblecolumn reactor Systems, the particular support employed plays a major rôle in influencingoverall hydrocarbon production rate (i.e., catalyst activity) with little or no effect onproduct selectivity. Catalyst activities generally rank in the following order: AI2O3 >S1O2 » T1O2. The source of the alumina and the pretreatment procedures used also playmajor rôles in determining the performance of the resulting, cobalt-based, Fischer-Tropsch catalysts.

Titania-supported cobalt catalysts, with or without promoters, were found to hâvepoor Fischer-Tropsch synthesis properties in both the fixed bed and SBCR Systems. 6 119 4 9

Compared to -alumina and silica, titania supports hâve much lower surface areas andpore volumes. Thus, they do not readily retain high cobalt loadings.

Although silica supports hâve relatively high surface areas, silica-supported cobaltcatalysts also provided low Fischer-Tropsch synthesis performance. Silica-supported 5 cobalt catalysts are unstable in reaction conditions, such as those usually encountered inFischer-Tropsch reaction Systems, where a significant amount of water is présent. Theformation of cobalt-silica compounds under these conditions is believed to cause thislower performance. To prevent or at least slow down silicate formation, the silica surfacemust typically be coated with oxide promoters, such as ZrC>2, prior to cobalt 10 imprégnation.

Characteristics and Préparation of Preferred Alumina SupportsThe catalyst support employed in the présent invention is preferably a -alumina support having: a low level of impurities, especially sulfur (preferably less than 100 ppmsulfur); a spheroidal shape; an average particle size in the range of from about 10 to about 15 150 pm (most preferably from about 20 to about 80 microns); a BET surface area, after calcination, in the range of from about 200 to about 260 m2/g; and a porosity in the rangeof from about 0.4 to about 1.0 cm3/g.

The alumina support is preferably produced from relatively high purity, syntheticboehmite. As discussed hereinbelow, the boehmite can be formed from aluminumalkoxide of the type obtained as a byproduct in the manufacture of synthetic fattyalcohols. Altematively, suitable, high-purity boehmite materials can be formed fromaluminum alkoxides produced by alcohol/aluminum métal reaction processes.

The aluminum alkoxide is preferably hydrolyzed to produce high purity,synthetic, monohydrate alumina. Next, this material is preferably spray-dried to yield 25 highly porous, spherical boehmite particles of relatively high surface area. The particulate boehmite material is preferably then sieved to remove fines and large particlesso that a desired particle size range is obtained (most preferably from about 20 to about80 microns). The sieved material is calcined to convert the boehmite particles to a -alumina support material having the desired surface area and porosity. The boehmite 30 material will preferably be calcined at a température of at least 350 C (more preferably from about 400 C to about 700 C and most preferably about 500 C) for a period of from 7 119 4 9 about 3 to about 24 hours (more preferably from about 5 to about 16 hours and mostpreferably about 10 hours). The desired calcination température is preferably reached byslowly heating the System at a rate of about 0.5-2.0 C/minute.

Examples of commercially-supplied boehmite materials suitable for forming thepreferred -alumina supports include, but are in no way limited to, the CAT AP AL andPURAL aluminas supplied by Condea/Vista. As discussed below, commercial materialsof this type are particularly effective when intentionally produced to hâve certain targetedtitanium “impurity” levels. Product quality reports for the CATAPAL aluminas indicatethat these products, as presently produced and sold, can hâve titania impurity levelsvarying ail the way up to 3000 ppm of elemental titanium by weight. The PURALproducts, on the other hand, typically hâve varying titanium impurity levels of up toabout 600 ppm.

Doping of -Alumina Supports

As shown hereinbelow, we hâve discovered that the presence of a controlledamount of dopant (preferably a titanium dopant) in the -alumina support unexpectedlyand surprisingly improves significantly the activities of “nonpromoted”, cobalt-on-alumina Fischer-Tropsch catalysts. As used herein, the term “nonpromoted” meanssimply that the catalyst is not promoted with any noble or near noble metals. The termdoes not exclude other types of promoters (e.g., potassium and/or lanthana). The phrase“near noble métal,” as used herein, encompasses rhénium and, although not practical foruse as a promoter, also encompasses technetium.

The titanium dopant should be présent in the -alumina support in an amount,expressed as elemental titanium, of at least 500 (preferably of at least 800) parts permillion (ppm) by weight. The dopant will more preferably be présent in the support in anamount, expressed as elemental titanium, in the range of fforn about 800 ppm to about2000 ppm by weight and will most preferably be présent in an amount in the range offrom about 1000 to about 2000 ppm. The titanium dopant can be added at substantiallyany time but will most preferably be added prior to crystallization of the boehmite.

As is well known in the art, one method of producing synthetic boehmitematerials utilizes aluminum alkoxides recovered as byproducts of certain processes (e.g.,the Ziegler Process) employed for manufacturing synthetic fatty alcohols. The Ziegler 8 119 4 9

Process typically comprises the steps of: (1) reacting high-purity alumina powder withethylene and hydrogen to produce aluminum triethyl; (2) polymerizing ethylene bycontacting it with the aluminum triethyl, thus forming aluminum alkyls; (3) oxidizing thealuminum alkyls with air to produce aluminum alkoxides; and (4) hydrolyzing the 5 aluminum alkoxides to produce alcohols and an alumina byproduct. The oxidation stepof the Ziegler process is typically catalyzed by an organic titanium compound which isitself converted to titanium alkoxide. The titanium alkoxide remains with, and is co-hydrolyzed with, the aluminum alkoxide, thus producing an alumina byproduct which isincidentally “doped” with a small amount of titania. 10 Another process for forming synthetic boehmite utilizes aluminum alkoxide produced by reacting an alcohol with a highly pure aluminum powder. The aluminumalkoxide is hydrolyzed to produce an alcohol, which is recycled for use in the alkoxideformation step, and alumina. Because this process does not involve an oxidation step, thealumina product typically does not contain titanium. However, for purposes of the 15 présent invention, any desired amount of titanium dopant can be included in the aluminaproduct by, for example, adding a titanium alkoxide to, and co-hydrolyzing the titaniumalkoxide with, the aluminum alkoxide. If desired, the same process can be used to addother dopants such as, for example, silica, lanthanum, barium, etc.

Heretofore, support manufacturers and catalyst users hâve simply considered 20 titania, if présent in the alumina support, to be a harmless impurity. Of the commercialsynthetic boehmite products presently available in the market, some are produced by theZiegler process, others are produced by the above-described aluminum alkoxidehydrolysis process, and still others are produced by a combination of these processeswherein the resulting products or product precursors are blended together. Such products 25 are sold and used interchangeably, without regard to the amount, if any, of titania présent.

Thus, the amount of titanium présent in commercial -alumina supports can vary ffom 0 ppm to as high as 3000 ppm titanium by weight or more. Titanium concentrationscan also vary significantly between different batches of the same commercial product.

Asdepicted in Figure 1, titania has a significant detrimental effect on the 3® activities of ruthenium-promoted, cobalt-on-alumina catalysts employed in slurry bubble column reactors. Figure 1 shows the activities (g-HC/kg-cat/hr) of three ruthenium- 119 4 9 promoted catalysts (catalysts 1, 2, and 3) which were produced and tested as describedhereinbelow in Example 1. Catalysts 1, 2, and 3 were identical in ail respects except that:catalyst 3 was formed on a -alumina support found to hâve a titania concentration,expressed as titanium, of about 7 ppm by weight; catalyst 2 was formed on a -aluminasupport found to hâve a titanium concentration of about 500 ppm; and catalyst 1 wasformed on a -alumina support found to hâve a titanium concentration of about 1000 ppm.As the amount of titania in the support increased, the activities of the catalysts, whichbegan at about 1400 for catalyst 3, declined to about 1322 for catalyst 2, and to about1195 for catalyst 1. Thus, any preference in the art as to the presence of titanium wouldheretofore hâve been that no titanium be included in the -alumina support.

We hâve discovered, however, that, in contrast to the detrimental effect oftitanium on the activities of “noble metal-promoted” catalysts employed in slurry bubblecolumn reactors, the activities of “nonpromoted”, cobalt-on-alumina catalysts, in ailFischer-Tropsch reaction Systems, are unexpectedly and surprisingly improved whencontrolled amounts of dopant are présent in the catalyst supports. The inventive, “non-promoted” catalysts hâve activities at least approaching those of catalysts promoted withnoble metals. Moreover, because they need not be promoted with noble metals, theinventive catalysts cost much less to produce. Thus, our invention significantly reducesthe cost of Fischer-Tropsch processes, particularly those processes conducted in slurrybubble column and other three-phase reaction Systems wherein catalyst attrition lossesare high.

Catalyst Préparation

Although other methods of préparation can be used, the catalytic components ofthe inventive catalysts are preferably added to the support by totally aqueousimprégnation using appropriate aqueous solution compositions and volumes to achieveincipient wetness of the support material with the desired component loading(s).Promoted catalysts are most preferably prepared by totally aqueous co-impregnation.Examples of typical promoters include, but are not limited to: métal oxides, such asoxides of Zr, La, K, and other oxides of éléments from Groups IA, IIA, IVA, VA, andVIA. 10 119 4 9

The totally aqueous imprégnation of cobalt onto the support, with or without oneor more desired promoters, is preferably accomplished by the steps of: (a) calcining thealumina support in the manner described above; (b) impregnating the support with anaqueous solution of cobalt nitrate, using a sufïïcient quantity of the solution to achieve 5 incipient wetness with the desired loading of cobalt; (c) drying the resulting catalystprecursor for about 5-24 hours at approximately 80-130°C, with moderate mixing, toremove solvent water and obtain a dried catalyst; and (d) calcining the dried catalyst inair or nitrogen by slowly raising the température of the System at a rate of about 0.5-2.0°Cper minute to approximately 250-400°C and then holding for at least 2 hours to obtain the 1θ oxide form of the catalyst. Multiple impregnation/coimpregnation steps (b) can be usedwhen higher cobalt loadings are desired.

Alkali (e.g., potassium) and/or rare earth oxide (e.g., lanthana)-promoted catalystscan be prepared, for example, by dissolving potassium nitrate [KNO3], lanthana nitrate[La(NÜ3)3'6H2O], and/or other precursors in the same solution which contains the 15 cobalt precursor. Alkali promoters, particularly potassium, can significantly improve product selectivity and reduce methane production. Moreover, when employed in properamounts, the alkali promoters do not substantially reduce catalyst activity. The additionof a lanthana (I^zCb) promoter can enhance the attrition résistance of the catalyst. Theimproved attrition résistance provided by the addition of LaîOa is not detrimental to 30 Fischer-Tropsch activity, or to Fischer-Tropsch selectivity. Preferred alkali and lanthanaconcentration ranges are provided hereinabove.

Catalyst Activation

To provide optimum performance, it is presently preferred that the catalyst beactivated by reducing the catalyst in a hydrogen-containing gas by slowly increasing the 25 température of the catalyst, preferably at a rate of about 0.5-2.0°C/minute, toapproximately 250-400°C (preferably about 350°C) and holding at the desiredtempérature for at least 2 hours. After réduction, the catalyst is preferably cooled inflowing nitrogen.

The reducing gas preferably comprises from about 1% to 100% by volume of 3° hydrogen, with the remainder (if any) being an inert gas, typically nitrogen. The reducing gas is preferably delivered at a rate of about 2-4 (preferably about 3) liters per 11 119 4 9 hour per gram of catalyst. The réduction procedure is preferably conducted in a fluidizedbed reactor vessel. The réduction procedure is most preferably conducted at conditions(i.e., température, flow rate, hydrogen concentration, etc.) effective to ensure that a verylow water vapor partial pressure is maintained during the procedure. 3 As shown hereinbelow, this activation procedure unexpectedly enhances the activity of the inventive “non-promoted” cobalt catalysts.

The Fischer-Tropsch Reaction Process

The catalysts prepared and activated in accordance with the présent invention canbe employed in generally any Fischer-Tropsch synthesis process. Where applicable (e.g., 1° for SB CR Systems or continuous stirred tank reactor (CSTR) Systems), the catalyst willpreferably be slurried in a Fischer-Tropsch wax or in a synthetic fluid (e.g., a C30 to C50range isoparaffm polyalphaolefin such as that available from Chevron under the nameSYNFLUID) having properties similar to those of Fischer-Tropsch wax. The catalystslurry will preferably hâve a catalyst concentration in the range of from about 5% to 15 about 40% by weight based on the total weight of the slurry.

The synthesis gas feed used in the reaction process will preferably hâve a CO.H2 volume ratio of from about 0.5 to about 3.0 and will preferably hâve an inert gas (i.e.,nitrogen, argon, or other inert gas) concentration in the range of from 0 to about 60% byvolume based on the total volume of the feed. The inert gas is preferably nitrogen. 20 Prior to initiating the reaction process, the activated catalyst will most preferably be maintained in an inert atmosphère. Before adding the catalyst thereto, the slurry fluidwill preferably be purged with nitrogen or other inert gas to remove any dissolvedoxygen. The slurry composition will also preferably be transferred to the reaction Systemunder an inert atmosphère. ^5 A particularly preferred SBCR reaction procedure comprises the steps of: (a) filling the SBCR, under an inert atmosphère, with the activated catalyst slurry; (b)heating and pressurizing the SBCR, under an inert atmosphère, to the desiredpretreatment conditions (preferably a température in the range of from about 220°C toabout 250°C and a pressure in the range of from about 50 to about 500 psig) ; (c)replacing the inert gas with hydrogen and holding the System at these conditions for fromabout 2 to about 20 hours; (d) purging the System with inert gas and lowering the reaction 12 1 7 9 4 9

System température, if necessary, to a point at least about 10°C below the desired reactiontempérature; (e) carefully replacing the inert gas with the desired synthesis gas; and (f)heating and pressurizing the reaction System, as necessary, to a desired operatingtempérature, preferably in the range of from about 190°C to about 300°C, and a desired 5 operating pressure, preferably in the range of from about 50 to about 900 psig.

Cobalt Catalvsts Without Noble Métal Promotion

It as long been believed that noble or near noble métal promotion is necessary toprovide cobalt catalysts which are truly viable for commercial Fischer-Tropschapplications. Heretofore, the accepted view in the art has been that a nonpromoted W CO/AI2O3 catalyst will only be from 50% to less than 25% as active as an otherwiseidentical catalyst promoted with one or more noble metals. U.S. Patent No. 5,157,054and other references discuss the necessity of using ruthénium or other promoters to obtainacceptable performance. However, the présent invention unexpectedly and surprisinglyprovides “nonpromoted”, cobalt-on-alumina catalysts having activities at least 15 approaching these of cobalt catalysts promoted with noble metals. As will be apparent tothose skilled in the art, eliminating the use of noble métal promoters without significantlydecreasing catalyst performance greatly enhances the cost effectiveness of the Fischer-Tropsch process, particulary in reaction Systems characterized by higher catalyst attritionlosses. £.0 As depicted in Figs. 1 and 2, the tests described hereinbelow amazingly show that the nonpromoted, cobalt-on-alumina catalysts produced and activated in accordance withthe présent invention perform at levels comparable to highly désirable, ruthenium-promoted catalysts. These results were unexpectedly obtained primarily through the useof doped -alumina supports. However, The catalyst activation procedure used, 25 particularly the maintenance of a very low water vapor partial pressure during the réduction process, also surprisingly enhanced the activity of the nonpromoted, cobalt-on-alumina catalysts.

The amount of dopant employed will preferably be an amount effect to provide acatalyst activity which is at least 60% (preferably at least 70%, more preferably at least 2>Û 80%, and most preferably at least 90%) of that of an otherwise identical catalyst promoted with ruthénium in a ruthénium to cobalt weight ratio of 1:40. The amount of 13 11949 dopant required to obtain a desired activity level for any given application can be readilydetermined.

As indicated above, the dopant is preferably employed in an amount of at least500 ppm (more preferably from about 800 ppm to about 2000 ppm), based on the total 5 weight of the -alumina support, and is most preferably added prior to the crystallizationof the boehmite precursor.

Low Water Vapor Partial Pressure

As to the bénéficiai results obtained by maintaining a very low water vapor partialpressure during the réduction process, it is believed that the presence of water vapor in 1® the activation System promûtes the formation of certain cobalt-alumina compounds whichare very difficult or impossible to reduce. Température programmed réduction studiesconducted by Applicants indicate that, in contrast to nonpromoted catalysts, the presenceof noble métal promoters such as ruthénium appear to enhance the réduction process in amanner which may counteract the deleterious effects of water vapor. 15 When activating a nonpromoted cobalt catalyst in accordance with the présent invention, the partial pressure of water vapor in the activation System will preferably bemaintained at or below a level effective for increasing the activity of the catalyst. Thewater vapor partial pressures effective for providing such results can readily bedetermined for any nonpromoted, cobalt-on-alumina catalyst. Although the values 20 necessary to obtain the desired results may vary depending upon the spécifie catalystselected, it is presently preferred that the partial pressure of water vapor in the activationSystem be maintained below 0.1 atmosphères.

EXAMPLES

In the foliowing examples 1-4, certain cobalt-on-alumina catalysts were preparedand then tested in various Fischer-Tropsch reaction Systems. Before testing, each catalyst 25 was reduced in pure hydrogen by slowly increasing the température of the catalyst, at arate of about 1.0 C per minute, to about 350 C and holding at this température for 10hours. The hydrogen was delivered at a rate of about 3 liters per hour per gram ofcatalyst. After réduction, the catalyst was cooled in flowing nitrogen.

For slurry bubble column reactor (SBCR) tests conducted in examples 1-4, the 3® réduction procedure was conducted in a fluidized bed reactor. After cooling to ambient 14 11949 température, the catalyst was weighed, slurried in S YNFLUID, and then transferred tothe SBCR under an inert atmosphère. The SBCR tests were then conducted at 230 C and450 psig using 900 sl/hr of syngas and from 15 to 25 grams of reduced catalyst. Thesyngas contained 60% nitrogen and had a H2/CO ratio of 2. In each case, the SBCRresults cited are those obtained after 24 hours on-stream.

For fixed bed micro-reactor (FBR) tests, the catalyst was reduced in-situ using thesame procedure just described. Prior to the introduction of syngas, the reduced catalystwas cooled to about 10-15 C below the reaction température. The FBR tests were thenconducted under differential conditions (i.e. low conversion) at atmospheric pressure and220 C using from 0.5 to 1.0 grams of catalyst and a H2/CO ratio of 2. In each case, theFBR results cited are those obtained after 24 hours on-stream. EXAMPLE 1

Effect of Titania Impurities on the Activities of Ru-PromotedCatalysts in a Slurry Bubble Column Reactor (SBCR)

The following ruthenium-promoted, cobalt-on-alumina catalysts were identicallyprepared and had identical loadings of cobalt and ruthénium. The catalysts differed onlywith respect to the amounts of titanium "impurity" contained in the -alumina supports.The aluminas were ail manufactured by Condea/Vista. CATALYST 1: (Ru-promoted cobalt catalyst on CATAPAL B alumina with 20 wt% cobalt and 0.5 wt% ruthénium)

Préparation Procedure: CATAPAL B alumina, supplied by Condea/Vista in the boehmite form, wascalcined at 500 C for 10 hrs to couvert it to -alumina. It was then preserved to 400-170mesh (i.e., a particle size range of from more than 38 microns to less than 88 microns)and impregnated using an amount of an aqueous solution of cobalt nitrate [Co(NO3)26H2O] and ruthénium (III) nitrosylnitrate [Ru(NO)(NO3)3 xH^O] appropriateto achieve incipient wetness (ca. 1.2 ml/g) with the desired loadings of Co. and Ru. Thecatalyst precursor was then dried in air at 115 C for 5 hours and calcined in air at 300 Cfor 2 hours (with a heating rate of ca. 1 C/min to 300 C). Réduction Procedure before Reaction: 15 1 7 9 4 9

The catalyst was reduced in 3000 cc/g/hr of pure hydrogen by heating at 1 C/minto 350 C and holding for 10 hrs.

Each of the following catalysts 2 and 3 were prepared and reduced in the samemanner as catalyst 1. The spécifie supports employed in catalysts 2 and 3 were as 5 follows: CATALYST 2: PURAL SB support supplied by Condea/Vista. CATALYST 3: PURAL SB1 support supplied by Condea/Vista.

The particular CAT AP AL B support material employed in catalyst 1 was determined to contain an amount of titania "impurity" of about 1000 ppm by weightj ü (expressed as ppm by weight of titanium) which was incidentally added as part of the

Ziegler Process prior to the crystallization of the boehmite. In contrast, the particularPURAL SB support material employed in catalyst 2 had been formed by a blendingprocess and was found to contain about 500 ppm of titanium. The PURAL SB1 supportemployed in catalyst 3 was identical to the PURAL SB support except that efforts were

15 made to prevent the addition of titanium. An elemental analysis showed that the PURAL SB 1 support contained only 7 ppm of titanium. The crystallite characteristics of theCAT AP AL B, PURAL SB and PURAL SB1 supports were substantially identical.

Figure 1 shows the activities (expressed in g-HC/kg-cat/hr) exhibited by catalysts 1-3 in an SBCR at the end of 24 hours on-stream. A comparison of catalysts 1-3 SC illustrâtes the detrimental effect of titania on the activities of ruthenium-promoted, cobalt-on-alumina catalysts. As the amount of titania in the support increased, activity declinedsignificantly. Catalyst 3 provided an activity of about 1400 and had selectivities (%C) of 80.5 for Cî+ and 8.4 for CH4. Catalyst 2 provided an activity of about 1322 and hadselectivities for Cs+ and CH4 of 81.9 and 8.5, respectively. Catalyst 1 provided an 25 activity of about 1195 and had selectivities of 82.2 (Cj+) and 8.2 (CH4). EXAMPLE 2

Effect of Titania Doping on the Activities of Non-PromotedCatalysts in a Slurry Bubble Column Reactor

The following catalysts 4-6 were respectively identical to catalysts 1-3, except 30 that catalysts 4-6 did not include any promoters. 16 119 4 9 CATALYST 4 (Non-promoted, alumina supported catalyst with 20 wt% Cobalt)Préparation Procedure: CAT AP AL B alumina, supplied by Condea/Vista in the boehmite form,was calcined at 500 C for 10 hrs to convert it to -alumina. It was then presieved 5 to 400-170 mesh (i.e, a particle size range of from more than 38 microns to lessthan 88 microns) and impregnated using an amount of an aqueous solution of Conitrate [Co(NO3)2'6HO] appropriate to achieve incipient wetness (ca. 1.2 ml/g)with the desired loading of Co. The catalyst precursor was then dried in air at 115C and calcined in air at 300 C for 2 hours (heating rate of ca. 1 C/min to 300 C). 10 Réduction Procedure before Reaction:

The catalyst was reduce in a pure hydrogen flow of 3000 cc/g/hr by heating at 1C/min to 350 C and holding for 10 hours.

Each of the following catalysts 5 and 6 were prepared and reduced in the samemanner as catalyst 4. The spécifie supports employed in catalysts 5 and 6 were as 15 follows: CATALYST 5: PURAL SB as described above. CATALYST 6: PURAL SB1 as described above.

Catalysts 4-6 were tested in a slurry bubble column reactor. Their activities(expressed in g-HC/kg-cat/hr) after 24 hours-on-stream are also shown in Figure 1. In 20 stark contrast to the results obtain with ruthenium-promoted catalysts 1-3, the presence oftitania in the -alumina support unexpected and surprisingly improved significantly theactivities of the non-promoted catalysts. Catalyst 6 (7 ppm Ti) provided an activity ofabout 606 and had selectivities (%C) of 85.6 (Cs+) and 5.2 (CH4). Catalyst 5 (500 ppmTi) provided an activity of about 775 and had selectivities of 84.0 (C5') and 6.2 (CH4). 25 Catalyst 4 (1000 ppm Ti) provided an activity of about 1032 and had selectivities of 86.5(Cî+) and 6.0 (CH4). Thus, in this SBCR test, the activity of non-promoted catalyst 4 was86% of that provided ruthenium-promoted catalyst 1. Further, the selectivities providedby catalyst 4 were significantly superior to those of the ruthenium-promoted catalysts. 17 119 4 9

Example 3

Effect of Titania Doping on the Activities of Non-PromotedCatalysts in a Fixed Bed Reactor

Ruthenium-promoted catalysts 1 and 3 and non-promoted catalysts 4 and 65 were also tested in a fixed bed micro-reactor (FBR) under the conditions described above, (atmospheric pressure and a température of 220 C). Figure 2 illustrâtes again thesignificant, unexpected, and surprising effect of titania doping on the activities of thenon-promoted cobalt catalysts. While the activities exhibited by the two ruthenium-promoted catalysts remained relatively constant (220 and 200 g-HC/kg-cat/h for Catalysts '(û 1 and 3, respectively) the activity of the titanium-doped (1000 ppm), non-promoted catalyst 4 was three times higher than that of non-doped (7 ppm), non-promoted catalyst 6. Moreover, whereas the activity of non-promoted catalyst 6 (7 ppm Ti) was only about25% of that of promoted catalyst 3, non-promoted catalyst 4 (1000 ppm Ti) amazinglyprovided an activity level which was about 75% of that of catalyst 3. 15 Example 4

Effect of Titania Doping on the Performances of Ru-Promoted and Non-Promoted

Catalysts in a Continuously Stirred Tank Reactor (CSTR)

In order to ascertain whether the SBCR results shown in Figure 1 represented theintrinsic activities of the non-promoted, cobalt-on-alumina catalysts, ruthenium-promoted 20 Catalyst 1 and non-promoted Catalyst 4 were tested in a continuously stirred tank reactor(CSTR). As discussed above, both catalysts were supported on CATAPAL B aluminacontaining about 1000 ppm by weight of titanium. In a CSTR, mass transfer limitationsare substantially negligible such that true intrinsic kinetics can be measured. Theperformance of the two cobalt catalysts in the CSTR was evaluated under reaction 25 conditions substantially similar to those employed in the SBCR. Amazingly, in accordance with the unexpected and surprising results obtained in the SBCR and FBRtests, the activities exhibited by promoted catalyst 1 and nonpromoted catalyst 4 in theCSTR were, within experimental errer, substantially the same. At 240 C and 450 psig,ruthenium-promoted Catalyst 1 and non-promoted Catalyst 4 exhibited activities of 1390 30 and 1330 (g-HC/g-cat/h), respectively. 18 119 4 9

Example 5

Température Programmed Réduction Studies of the Efïects of Titania Doping

It is well known that noble métal promoters enhance the reducibility of cobalt. 9 The effects of titania on the reducibilities of non-promoted catalysts 4 and 6 were determined using température programmed réduction (TPR). The résulte were comparedto TPR results obtained for ruthenium-promoted catalysts 1 and 3.

In each case, the calcined catalyst was initially dried ,under an argon flow, at 120C for 30 minutes to remove water. In the TPR tests, a 5 % Ffe/Ar gas mixture was used as 10 the reducing gas. The reducing gas flow rate was 30 cm3/min. During the réduction test,the catalyst was heated to 900 C at a rate of 5 C/min. The effluent gas was deliveredthrough a cooling trap (less than 50 C) to condense and collect water generated by theréduction process. The amount of Hî consumed by the catalyst was monitored, using athermal conductivity detector (TCD), and recorded as a fonction of température. From 15 this data (and assuming the oxide to be in the form of CO3O4), the total reducibility of each catalyst, up to a température of 900 C, was determined and expressed as the percentof cobalt completely reduced. The percent reducibilities of catalysts 1,3,4, and 6 at 900C are shown in Table 1. Table 1 also provides the low and high réduction températurepeaks exhibited by each catalyst. 20 Table 1

Température Programmed Réduction Results

Catalyst Low Température Peak(C) High Température Peak(C) Reducibility 1 (%) Catalyst 1 253 483 98 Catalyst 3 241 471 99 Catalyst 4 238/322 587 99 Catalyst 6 328 584 90 19 119 4 9

As expected, both of the ruthenium-promoted catalysts were almost totallyreduced (98+%) under the conditions used in the TPR experiments. In addition, theyreduced at lower températures than their non-promoted analogs. However, a comparisonof the results obtained for the non-promoted catalysts shows, again, that titania dopingprovided a significant, bénéficiai effect. Like the ruthenium-promoted catalysts, the non-promoted catalyst having a doped support was completely (99%) reduced. However, thenon-promoted catalyst having a no dopant was only 90% reduced.

It is believed that certain, hard-to-reduce, cobalt-alumina compounds can formduring réduction, especially when, as is typically the case, the réduction System has arelatively high water vapor partial pressure. Noble métal promoters, which allow cobaltréduction at lower température, either help to prevent the formation of these compoundsand/or enhance their reducibility. Our fmdings suggest that, for a non-promoted cobalt-on-alumina catalyst, the presence in the support of a controlled amount of dopant helpsprevent the formation of cobalt-alumina compounds, thus improving the overallreducibility of the catalyst. This would explain the significant improvement in Fischer-Tropsch activity observed for nonpromoted, cobalt-on alumina catalysts having dopedsupports.

Example 6

To test the effects of water vapor partial pressure during the réduction process,two different batches of the same nonpromoted, 20 wt% cobalt-on-alumina catalysts (i.e.,a 50 g (lab size) batch and a 220 g batch) were reduced as set forth below and then testedin an SBCR.

Test 1: The 50 g (lab size) batch was reduced at 350 C for 18 hours with 100% H2 at a flow rate above 3 1/hr per gram of catalyst. The SBCR test was started with acharge of 15 g of the reduced catalyst. The reaction conditions were as follows: atempérature of 220 C, a pressure of450 psig, a total gas flow rate of900 sl/hr with 60%N2, and a H2//CO ratio of 2. The CO conversion was 13.6%, the hydrocarbonproductivity (activity) was 0.66 g HC/g cat./hr, and the CH4 selectivity was 3.2%.

Test 2: The 220 g batch of the same catalyst was reduced at the same conditionsexcept that a hydrogen flow rate of 1.81/hr per gram of catalyst was used. The SBCR testwas started with a 15 g charge of catalyst and was run under the same reaction conditions 20 119 4 9 as in Test 1. The CO conversion was only 3.4%, with a hydrocarbon productivity(activity) of just 0.17 g HC/g cat./hr. A re-reduction of the catalyst employed in Test 2, using a smaller batch, did notproduce better résulte. The catalyst was irreversibly damaged during the first réduction. 5 Additional large batch réductions (ca. 0.6 -1 kg) produced similarly inactive or very lowactivity catalysts.

These results suggest that a higher water vapor partial pressure présent during theréduction of the larger batches had a négative effect on the reducibility of the catalyst.

Example 7 1U Following the unsuccessful, attempted réduction of large batches of the non- promoted CO/AI2O3 catalyst, and in view of the belief that the presence of a relativelyhigh water vapor partial pressure during the initial stages of the réduction process wasresponsible for their low reducibility, it was suggested that higher réduction températuresmight provide improved reducibility. Hence, two new lab size batches (ca. 50 g) of the 15 same unpromoted CO/AI2O3 catalyst tested in Example 6 were reduced at 410 C and 380C, respectively. In order to test the effect of high water vapor partial pressure, thehydrogen stream used for each réduction was saturated with water vapor by passing itthrough a saturator, at room température, before use in the réduction System.

The SBCR tests were again carried out at the same reaction conditions asdescribed in Example 6. In spite of the different réduction températures used, each batchhad a CO conversion activity of below 1%. The results showed that the highconcentration of water in the hydrogen stream had a drastic effect on catalyst reducibilityand that, under such conditions, the use of an increased réduction température did notimprove the reducibility of the catalyst. The high water vapor partial pressure is believed 25 to hâve caused the formation of cobalt-alumina compounds which were very difficult orimpossible to reduce. The use of a higher réduction température seemed to actually promote the production of such compounds rather than enhance the réduction process. * * * * *

Thus, the présent invention is well adapted to carry out the objects and attain theends and advantages mentioned above, as well as those inhérent therein. While theinvention has been described with a certain degree of particularity, it is manifest that 21 many changes may be made without departing from the spirit and scope of thisdisclosure. It is understood that the invention is not limited to the embodiments set forthherein for purposes of exemplification. 22

Claims (41)

119 4 9 WHATIS CLAIMEDIS:

1. A cobalt catalyst for Fischer-Tropsch hydrocarbon synthesis, said cobaltcatalyst comprising cobalt supported on a -alumina support wherein: said cobalt catalyst is not promoted with any noble metals deposited on 5 said -alumina support, is not promoted with rhénium or technetium deposited on said -alumina support, and is not promoted with titaniumdeposited on said -alumina support; said -alumina support has an internai structure comprising -alumina anda controlled amount of a titanium dopant; and 10 said controlled amount of said titanium dopant in said -alumina support is an amount effective for increasing the activity of said cobalt catalyst forsaid Fischer-Tropsch hydrocarbon synthesis.

2. The cobalt catalyst of claim 1 wherein said controlled amount of saidtitanium dopant is an amount effective to render said cobalt catalyst at least 60% as active 19 as a promoted catalyst which is identical to said cobalt catalyst except that said promotedcatalyst is promoted with ruthénium in a ruthénium to cobalt weight ratio of 1:40.

3. The cobalt catalyst of claim 2 wherein said controlled amount of saidtitanium dopant is an amount effective to render said cobalt catalyst at least 70% as activeas said promoted catalyst.

4. The cobalt catalyst of claim 2 wherein said controlled amount of saidtitanium dopant is an amount effective to render said cobalt catalyst at least 80% as activeas said promoted catalyst.

5. The cobalt catalyst of claim 1 wherein said controlled amount of saidtitanium dopant, expressed as elemental titanium, is at least 500 ppm by weight of the 25 total weight of said -alumina support. 23 119 4 9

6. The cobalt catalyst of claim 1 wherein said controlled amount of saidtitanium dopant, expressed as elemental titanium, is in the range of from about 800 toabout 2000 ppm by weight of the total weight of said -alumina support.

7. The cobalt catalyst of claim 1 wherein said controlled amount of said5 titanium dopant, expressed as elemental titanium, is about 1000 ppm by weight of the total weight of said -alumina support.

8. The cobalt catalyst of claim 1 wherein said cobalt catalyst is an activatedcatalyst which has been reduced in the presence of hydrogen and at a water vapor partialpressure effective to increase said activity of said cobalt catalyst.

9. The cobalt catalyst of claim 8 wherein said water vapor partial pressure is in the range of from 0 to about 0.1 atmosphères.

10. The cobalt catalyst of claim 1 wherein said cobalt catalyst is promotedwith at least one of potassium and lanthana.

11. The cobalt catalyst of claim 1 wherein: 1? said -alumina support is produced from synthetic boehmite and said titanium dopant is added to said -alumina support prior to the crystallization of said synthetic boehmite.

12. A process for hydrocarbon synthesis comprising the step of reacting asynthesis gas in the presence of a cobalt catalyst wherein: 20 said cobalt catalyst comprises cobalt supported on a -alumina support; said cobalt catalyst is not promoted with any noble metals deposited onsaid -alumina support, is not promoted with rhénium or technetiumdeposited on said -alumina support, and is not promoted with titaniumdeposited on said -alumina support; 24 11949 said -alumina support has an internai structure comprising -alumina anda controlled amount of a titanium dopant; and said controlled amount of said titanium dopant in said -alumina supportis an amount effective for increasing the activity of said cobalt catalyst for $ said Fischer-Tropsch hydrocarbon synthesis.

13. The process of claim 12 wherein said amount of said controlled amount ofsaid titanium dopant is an amount effective to render said cobalt catalyst at least 60% asactive, for said hydrocarbon synthesis, as a promoted catalyst which is identical to saidcobalt catalyst except that said promoted catalyst is promoted with ruthénium in 10 ruthénium to cobalt weight ratio of 1:40.

14. The process of claim 13 wherein said amount of said controlled amount ofsaid titanium dopant is an amount effective to render said cobalt catalyst at least 70% asactive as said promoted catalyst.

15. The process of claim 13 wherein said controlled amount of said titanium15 dopant is an amount effective to render said cobalt catalyst at least 80% as active as said promoted catalyst.

16. The process of claim 12 wherein said controlled amount of said titaniumdopant, expressed as elemental titanium, is at least 500 ppm by weight of the total weightof said -alumina support.

17. The process of claim 12 wherein said controlled amount of said titanium dopant, expressed as elemental titanium, is in the range of ffom about 800 to about 2000ppm by weight of the total weight of said -alumina support .

18. The process of claim 12 wherein said controlled amount of said titaniumdopant, expressed as elemental titanium, is about 1000 ppm by weight of the total weightof said -alumina support. 25 1 194 9

19. The process of claim 12 further comprising the step, prior to said step ofreacting, of activating said cobalt catalyst by reducing said cobalt catalyst in the presenceof hydrogen and at a water vapor partial pressure effective to increase said activity of saidcobalt catalyst.

20. The process of claim 19 wherein said water vapor partial pressure is in the range of ffom 0 to about 0.1 atmosphères.

21. The process of claim 12 wherein said step of reacting is conducted in aslurry bubble column reactor.

22. The process of claim 12 wherein said hydrocarbon synthesis is a Fischer-10 Tropsch synthesis process.

23. The process of claim 12 wherein: said -alumina support is produced ffom synthetic boehmite; and said titanium dopant is added to said -alumina support prior to the crystallization of said boehmite. 15 24. A cobalt catalyst for hydrocarbon synthesis comprising cobalt supported on a -alumina support, wherein: said cobalt catalyst is not promoted with any noble metals deposited onsaid -alumina support, is not promoted with rhénium or technetiumdeposited on said -alumina support, and is not promoted with titaniumdeposited on said -alumina support; said -alumina support has an internai structure comprising -alumina anda controlled amount of a titanium dopant; and said cobalt catalyst has been reduced in the presence of hydrogen at a water vaporpartial pressure effective to increase the activity of said cobalt catalyst for said 25 hydrocarbon synthesis. 26 119 4 9

25. The cobalt catalyst of claim 24 wherein said water vapor partial pressure isin the range of from 0 to about 0.1 atmosphères.

26. The cobalt catalyst of claim 24 wherein said -alumina support includessaid controlled amount of said titanium dopant in at least 500 ppm by weight, expressedas elemental titanium, of the total weight of said -alumina support.

27. The cobalt catalyst of claim 26 wherein said controlled amount of saidtitanium dopant, expressed as elemental titanium, is in the range from about 800 to about2000 ppm by weight of the total weight of said -alumina support.

28. The cobalt catalyst of claim 26 wherein: said -alumina support is produced from synthetic boehmite; and said titanium dopant is added to said -alumina support prior to the crystallization of said synthetic boehmite.

29. A process for hydrocarbon synthesis comprising the steps of: (a) reducing a cobalt catalyst in the presence of hydrogen and at a water vaporpartial pressure effective to increase the activity of said cobalt catalyst for saidhydrocarbon synthesis, said cobalt catalyst comprising cobalt supported on a -aluminasupport, said -alumina support having an internai structure comprising -alumina and acontrolled amount of a titanium dopant; and (b) reacting a synthesis gas in the presence of said cobalt catalyst, whereinsaid cobalt catalyst is not promoted with any noble metals deposited on said -aluminasupport, is not promoted with rhénium or technetium deposited on said -aluminasupport, and is not promoted with titanium deposited on said -alumina support;

30. The process of claim 29 wherein:step (a) is conducted in a fixed bed vessel; and step (b) is conducted in a slurry bubble column reactor. 27 119 4 9

31. The process of claim 29 wherein said water vapor partial pressure is in therange of from 0 to about 0.1 atmosphères.

32. The process of claim 29 wherein said controlled amount of a titaniumdopant is at least 500 ppm by weight, expressed as elemental titanium, of the total 5 weight of said -alumina support.

33. The process of claim 32 wherein said controlled amount of said titaniumdopant, expressed as elemental titanium, is in the range of from about 800 to about 2000ppm by weight of the total weight of said -alumina support.

34. The process of claim 29 wherein said hydrocarbon synthesis is a Fischer-10 Tropsch synthesis process.

35. A method of improving the activity of a cobalt catalyst for hydrocarbonsynthesis, wherein said cobalt catalyst has a -alumina support and said -alumina supporthas an internai structure comprising -alumina and a controlled amount of a titaniumdopant, said method comprising the step of including in said support an amount, 15 expressed as elemental titanium, of at least 500 ppm by weight of the total weight of saidalumina support.

36. The method of claim 35 wherein said amount, expressed as elementaltitanium, of said titanium dopant is in the range of from about 800 to about 2000 ppm byweight of the total weight of said alumina support.

37. The method of claim 35 further comprising the step of reducing said cobalt catalyst in hydrogen at a water vapor partial pressure in the range of from 0 toabout 0.1 atmosphères.

38. The cobalt catalyst of claim 1 wherein: 28 119 4 9 said -alumina support is formed from aluminum alkoxide which ishydrolyzed to produce an alumina product and said titanium dopant is added to said -alumina support by cohydrolyzingtitanium alkoxide with said aluminum alkoxide in an amount effective to 5 yield said controlled amount of said titanium dopant in said -alumina support.

39. The cobalt catalyst of claim 1 wherein said cobalt is deposited on said -alumina support by totally aqueous imprégnation using cobalt nitrate.

40. A cobalt catalyst for Fischer-Tropsch hydrocarbon synthesis, said cobalt 10 catalyst comprising cobalt supported on a -alumina support wherein: said cobalt catalyst is not promoted with any noble metals deposited onsaid -alumina support, is not promoted with rhénium or technetiumdeposited on said -alumina support, and is not promoted with titaniumdeposited on said -alumina support; 15 said -alumina support has an internai structure comprising -alumina and a controlled amount of a titanium dopant; and said controlled amount of said titanium dopant, expressed as elementaltitanium, is at least 500 ppm by weight of the total weight of saidalumina support.

41. The cobalt catalyst of claim 40 wherein said controlled amount of said titanium dopant, expressed as elemental titanium, is in the range of from about 800 toabout 2000 ppm by weight of the total weight of said -alumina support.

42. The cobalt catalyst of claim 40 wherein said controlled amount of saidtitanium dopant, expressed as elemental titanium, is about 1000 ppm by weight of the 25 total weight of said -alumina support. 29