WO2004096437A1

WO2004096437A1 - Low coke formation catalysts and process for reforming and synthesis gas production

Info

Publication number: WO2004096437A1
Application number: PCT/US2003/010068
Authority: WO
Inventors: Gregory J. Lewis; John E. Bauer
Original assignee: Uop Llc
Priority date: 2003-04-02
Filing date: 2003-04-02
Publication date: 2004-11-11
Also published as: AU2003222169A1; CN1764501A; MXPA05010550A; CA2520221A1

Abstract

Applicants have developed a novel catalyst composition comprising a crystalline metal oxide having the empirical formula: AvBt+wNixD(Gu-)yOz, where A is an alkali metal (e.g. Na), B is a basic metal (e.g. Ca), D is a framework component (e.g. P), and G is an anionic species (e.g. OH-). Nickel may be present in the framework of the crystalline metal oxide, dispersed thereon, or both. Preferably, the metal oxide component has an apatite or hydroxyapatite crystal structure. These crystalline metal oxide components have been found to have improved performance in partial oxidation and light hydrocarbon (e.g. methane) reforming to produce synthesis gas. A new process for synthesizing these metal oxides is also disclosed.

Description

" LOW COKE FORMATION CATALYSTS AND PROCESS FOR REFORMING AND SYNTHESIS GAS PRODUCTION "

BACKGROUND OF THE INVENTION

The present invention relates to a novel catalyst composition, a method of making the composition, and a hydrocarbon conversion process using the composition comprising a phosphate- or vanadate-based crystalline catalyst with nickel incorporated into the crystalline metal oxide framework, deposited thereon, or both. The catalyst is used to produce synthesis gas from light hydrocarbons (e.g. methane).

Useful gases, known as synthesis gases or syngases, are produced from the coversion of methane and light hydrocarbons to a mixture containing CO and H₂.

Conventional syngas-generating processes include the gas phase partial oxidation process (US-A-5,292,246), the autothermal reforming process (US-A-5,492,649), and various other processes involving CO₂ or steam reforming. The choice of a particular route depends primarily on the desired product composition, as determined by its end use. Syngas is typically used to produce methanol, ammonia, or heavier hydrocarbon fuels through Fisher-Tropsch technology.

The partial oxidation of hydrocarbons can generally take several pathways, depending on the relative proportions of the reacted hydrocarbons and oxygen as well as the conditions used. In the case of methane the following reactions are possible:

2CH₄ + 2O₂ -» 2CO + 2H₂ + 2H₂O

(-64 kcal/g mol CH₄)

2CH₄ + 1.5O₂ -» 2CO + 3H₂ + H₂O (-34.9 kcal/g mol CH₄) or , 2CH₄ + O₂ -> 2CO + 4H₂

(-5.7 kcal/g mol CH₄) The last reaction is the most desirable in terms of both the quality of the syngas produced and the minimization of liberated heat to protect the apparatus and catalyst bed from thermal damage. This pathway also reduces formation of steam with a corresponding increased yield of hydrogen and carbon monoxide. When combined with the reforming reaction, this reaction product provides a high quality syngas. Therefore, maximizing the carbon to oxygen molar ratio (C:O ratio) in the feedstock gas mixture usually optimizes the partial oxidation. Unfortunately, greater coke formation usually offsets the improvements in product quality at high C:O ratios by limiting the partial oxidation catalyst life. As a result, providing reasonable catalyst stability normally constrains C:O ratios to below their ideal for product quality purposes, i.e. downstream operational costs, such as those associated with recycling unconverted syngas, are increased.

An autothermal reforming process mixes and reacts the methane and oxygen- containing feeds in a diffusion flame. The oxidized effluent typically passes into a steam reforming zone where it contacts a conventional steam reforming catalyst. The catalyst may comprise a simple fixed bed or an impregnated component in a monolith carrier or ceramic foam. The high temperature in the catalytic reforming zone places great demands on reforming catalyst's activity and stability over many years of use. Again catalyst coking problems in autothermal reforming normally require operation at sub- optimal conditions for feed stock composition and product quality. Specifically, to suppress coking the level of steam injection must exceed that dictated by concerns about optimizing product quality and minimizing utility costs. In controlling catalyst coke formation, the molar C:H₂O ratio for steam reforming is analogous to the C:O ratio in partial oxidation. Higher values generally mean a greater coking tendency.

According to the autothermal steam reforming process of US-A-5,492,649, the production of high amounts of carbon or soot in the diffusion flame oxidation step is avoided by mixing the methane gas with the oxidizer gas while swirling the latter at the injection nozzle to provide a large number of mixing points in the diffusion flame. However, this process still causes partial oxidation in the diffusion flame, resulting in over-oxidation and an excessively high temperature effluent. The heat generated can damage the steam reforming catalyst as well as the face of the injector. Published Canadian Patent Application No. 2,153,304 teaches ameliorating soot by reducing the steam to carbon molar feed ratio, combined with increasing the steam reforming temperature to between 1100-1300°C, and/or introducing the gaseous hydrocarbon feed in increments. The use of noble metals (e.g. Pt) in a catalyst limit coke formation under a wide range of conditions. The prohibitive cost of noble metal-containing catalysts in industrial applications usually leads to the alternative use of metal component selected from uranium, Group VII metals, and Group VIII metals in steam reforming often in combination with other metals such as lanthanum and cerium. Generally, thermally stable inorganic refractory oxides support the metals. Preferred catalyst metals are the Group VIII metals. Nickel-containing materials are particularly useful, especially nickel aluminate materials, nickel oxide and nickel on supports such as alpha-alumina.

Particularly useful support materials include alpha-alumina, aluminosilicates, cement, magnesia, and fused tabular alumina. Preferred catalyst supports may be Group

II metal oxides, rare earth oxides, modified alpha-aluminas, alpha-alumina-containing oxides, hexa-aluminates, calcium aluminate, or magnesium-alumina spinel. In some cases a binder may stabilize the catalyst, for example, calcium aluminum oxide. Preferably the catalyst maintains a very low level of silicon dioxide, e.g. less than 0.3 wt- % to avoid volatilization and fouling of downstream equipment. The shape of the catalyst carrier particles may vary and includes saddles, stars, beads, spoked wheels and Raschig rings 16 mm in diameter and height, and having a single 6-8 mm hole in the middle.

US-A-3,595,808 discloses catalysts comprising calcium nickel phosphates for dehydrogenation applications where a crystalline hydroxyapatite phase results from the preparation of such catalysts. In this hydroxyapatite phase a nickel replaces 1 of every 6 to 12 calcium atoms. Brown & Constantz, Hydroxyapatite and Related Materials, CRC Press, Inc. (1994) describes apatite structures, having the general chemical formula A₁₀(BO₄)₆X₂ where A=Ca, Sr, Ba, Pb, Cd and other rare earth elements, BO₄=PO₄ ^3", VO₄ ^3" , SiO₄ ⁴-, AsO₄ ³\ CO₃ ²-, and X is Off, CT, F, CO₃ ²\ Boechat, Eon, Rossi, Perez, and Gil, Phys. Chem. Chem. Phys., 2000, 2, 4225-4230 describe the preparation and structural characterization of calcium-phosphate and vanadate solid solutions and the incorporation of Sr into the composition to yield Ca₁₀._xSr_x(PO₄)₆._s(VO₄)_s(OH)₂ apatite structures. Sugiyama, Minami, Higaki, Hayahi, and Moffat, Ind. Eng. Chem. Res. 1997, 36, 328-334 reports the conversion of methane to carbon monoxide in the presence of stoichiometric strontium hydroxyapatite.

Applicants unexpectedly found that the use of a catalyst comprising a crystalline metal oxide component having a basic metal (e.g. Sr) and a structural component (e.g. PO₄ or VO₄), and optionally an allcali metal within the crystalline framework significantly mitigates the problem of coke formation in conventional oxidation and reforming. Active nickel is also included in the framework and/or dispersed on the crystalline metal oxide component. The reduced coke formation in hydrocarbon oxidation and reforming operations using the catalyst of the present invention overcomes many of the above noted problems relating to operational constraints and the associated cost burdens.

SUMMARY OF THE INVENTION

One embodiment is a catalyst composition of the present invention for the production of synthesis gas from light hydrocarbons comprising a crystalline metal oxide component having a chemical composition on an anhydrous basis expressed by an empirical formula of:

A_v(B^t+)_wNi_xD(G^u-)_yO_z

where A is an alkali metal selected from the group consisting of Li⁺, Na⁺, K⁺, Rb⁺, Cs⁺, and mixtures thereof, "v" is the mole ratio of A to D and varies from 0 to 2, B is a basic metal, "w" is the mole ratio of B to D and varies from 1 to 3, "t" is the weighted average valence of B and varies from 2 to 3, "x" is the mole ratio of Ni to D and varies from 0 to 0.5, D is a framework component selected from the group consisting of P⁺⁵, V⁺⁵, and mixtures thereof, and G is an anionic species selected from the group consisting of Off, Cl^", F^", CO₃ ^2", and mixtures thereof, "u" is the average valence of G and varies from 1 to

2, "y" is the mole ratio of G to D and varies from 0 to 2, and "z" is the mole ratio of O to D and has a value determined by the equation:

z = y₂ (v + t^«w + 2^«x + 5 - u^«y),

and when B is Ca, "v" is not 0, and when "x" is 0, the catalyst composition further comprises a nickel component dispersed on the crystalline metal oxide component.

In a preferred embodiment the crystalline metal component has the hydroxyapatite crystal structure.

In another embodiment, the present invention is a process for preparing the catalyst composition of the present invention by the process comprising: a) reacting a mixture containing reactive sources of B basic metal, optionally Ni, D framework component, and optionally A alkali metal, at a pH from 8 to 14 and a temperature and time sufficient to form the crystalline metal oxide component, the mixture having a composition expressed by:

h A₂O : j BO_(/2 : k NiO : D₂O₅ : l N : m H₂O

where N is a mineralizer, "h" varies from 0 to 10, "j" varies from 0.10 to 6.0, "k" varies from 0 to 1.0, "1" varies from 0 to 20, and "m" varies from 40 to 500,

b) contacting, when "k" is 0, the crystalline metal oxide component with an aqueous solution of a nickel salt selected from the group consisting of nickel nitrate, nickel chloride, nickel bromide, nickel acetate, and mixtures thereof, and

c) calcining the crystalline metal oxide component of step (a) or (b) at a temperature from 600°C to 1000°C for a period from 1 to 10 hours to yield the catalyst.

Another embodiment of the present invention is a process for producing synthesis gas comprising reacting a light hydrocarbon and an oxidant at reaction conditions in the presence of the catalyst composition of this invention.

BRIEF DESCRIPTION OF THE DRAWING

The Figure compares spent catalyst coke levels for catalyst of the present invention to conventional catalysts.

DETAILED DESCRIPTION OF THE INVENTION

The catalyst composition of the present invention includes the components as previously described. Typically metal B has a rather large cationic radius, generally from 0.85 A to 1.40 A, and a cationic charge of +2 or +3. These metals are broadly found in the alkaline earth series, rare earth series, and in selected cases in both the transition metal series, such as Cd²⁺and Hg²⁺, and the main group metals such as Pb²⁺. Preferably, B is selected from the group consisting of Ca²⁺, Sr²*, Cd²⁺, Pb²⁺, Ba²⁺, La³⁺, Eu³⁺, Gd³⁺, Pr³*, Nd³⁺, Sm³⁺, Y³⁺, Yb³⁺, and mixtures thereof. Catalysts give exceptional results when B is a mixture of Sr and another metal selected from this group of basic metals. Preferably B comprises a mixture of Sr and a second metal selected from the group consisting of Ca²⁺, Ba²⁺, La³⁺, Eu³⁺, Gd³⁺, Pr³*, Nd³⁺, Sm³⁺, Y³⁺, Yb³⁺, and mixtures thereof.

When B is one metal, its weighted average valence of B is that of the metal. When more than one basic metal is used, the total amount of

and the weighted average valance "t" is defined by

w

The weighted average valence of the anionic species ("u") is measured in a similar manner and depends on the amount of CO₃ ^2" present in the composition, relative to the other possibilities for G, that have a oxidation state of -1. Thus, if the total amount of G, or "y" is defined by y = p + q, where "p" is the mole ratio of CO₃ ^2", and "q" is the mole ratio of species having a valence of -1, then the average valence is

-2p-q u = — - — - p + q

The crystalline metal oxide component of the catalyst is prepared by dissolving sources of Ni (if incorporated into the framework) and the basic metals in a first solution. Sources of the framework component and optionally the alkali metal are dissolved in a second solution. Thoroughly mixing the two solutions with any optional mineralizer forms the reaction mixture. The pH of the reaction mixture is generally from 8.0 to 14.0 and is controlled by the amount of alkali hydroxide or mineralizer added to the reaction mixture. After thorough homogenization, the reaction mixture is digested from 2 hours to 7 days at temperatures from 20°C to 200°C, preferably using hydrothermal digestion at a temperature of 50°C to 150°C for a period of 12 to 96 hours and more preferably from 75°C to 125°C for a period from 24 to 48 hours. After digestion, the product is isolated by filtration or centrifugation, washed with de-ionized water and dried at room temperature or in an oven up to 200°C. The dried product is then calcined at a temperature from 600°C to 1000°C for 1 to 10 hours under a flow of air. Optional screening or adjusting of sample size can follow drying. Generally, after the initial hydrothermal synthesis, the crystalline metal oxide component exhibits the hydroxyapatite structure as determined by x-ray diffraction. Not usually but on occasion, the product of the hydrothermal synthesis may also include some Ni(OH)₂. Calcination of the isolated hydroxyapatites from the hydrothermal syntheses can result in multiple phases forming in the final catalyst, depending on calcination conditions. Along with the hydroxyapatite some NiO can form outside of the framework or another phase such as Sr₃(PO₄)₂ will appear. In certain instances a new structure can form after calcination (e.g. Example 8 for SrCaP-1). Hence, the crystalline metal component may contain multiple crystalline phases.

In forming the reaction mixtures above, sources of the alkali metal may include without limitation hydroxides, acetates, halides, or carbonates of sodium, potassium, lithium, rubidium, or cesium. Sources of the basic metals may include nitrates, chlorides, and acetates of calcium, strontium, lead, cadmium, barium, lanthanum, neodymium, gadolinium, europium, yttrium, ytterbium, and samarium. Sources of nickel include nickel nitrate, nickel chloride, and nickel acetate. Mineralizing agents such as hydroxide, fluoride, chloride, or carbonate may be introduced for example as the alkali salt, such as sodium fluoride or sodium carbonate, organoammonium salts such as tetramethylammonium hydroxide or tetraethylammonium fluoride or in the case of the halides as acids such as HF or HC1. The framework species P and V are introduced to the reaction mixture via sources such as phosphoric acid,

Na₃VO₄, KVO₃, and V₂O₅. Overall, the reaction mixture has a composition expressed by: h A₂O : j BO_Λ : kNiO : D₂O₅ : I N : m H₂O

where N is a mineralizer, "h" varies from 0 to 10, "j" varies from 0.10 to 6.0, "k" varies from 0 to 1.0, "1" varies from 0 to 20, and "m" varies from 40 to 500. When "k" is 0, nickel is dispersed onto the crystalline metal oxide component (which contains no nickel within its crystalline framework) prior to calcination. The structure of the crystalline metal oxide component of this invention was determined by x-ray analysis. The x-ray patterns presented in the following examples were obtained using standard x-ray powder diffraction techniques. The radiation source was a high-intensity, x-ray tube operated at 45 kV and 35 ma. The diffraction pattern from the copper K-alpha radiation was obtained by appropriate computer based techniques. Flat compressed powder samples were continuously scanned at 2° (2Θ) per minute from 2° to 70° (2Θ). friterplanar spacings (d) in Angstrom units were obtained from the position of the diffraction peaks expressed as θ where θ is the Bragg angle as observed from digitized data. Intensities were determined from the integrated area of diffraction peaks after subtracting background, "I₀" being the intensity of the strongest line or peak, and "I" being the intensity of each of the other peaks.

Alternatively, the x-ray patterns were obtained from the copper K-alpha radiation by use of computer based techniques using a Siemens D-500 x-ray powder diffractometer, Siemens Type K-805 x-ray sources, available from Siemens Corporation, Cherry Hill, N.J., with appropriate computer interface.

The determination of 2Θ is subject to both human and mechanical error, which in combination can impose an uncertainty of ±0.4° on each reported value of 2Θ. This uncertainty is also manifested in the reported values of the d-spacings, which are calculated from the 2Θ values. This imprecision is general throughout the art and is not sufficient to preclude the differentiation of the present crystalline materials from oilier compositions. In some of the x-ray patterns reported, the relative intensities of the d-spacings are indicated by the notations vs, s, m, and w which represent very strong, strong, medium, and weak, respectively. In tenns of 100 x I/I₀, the above designations are defined as

w = 0-15; m = 15-60; s = 60-80 and vs = 80-100.

In certain instances the purity of a synthesized product may be assessed with reference to its x-ray powder diffraction pattern. Thus, for example, if a sample is stated to be pure, it is intended only that the x-ray pattern of the sample is free of lines attributable to crystalline impurities, not that there are no amorphous materials present.

To allow for ready reference, the different structure types in the following examples have been given arbitrary numbers such as CaSrNiP-1. Thus CaSrNiP-1 and KCaSrNiP-1 have the same structure, i.e., structure type 1. The elements in the name are indicative of the composition. One particular structure is the hydroxyapatite structure, designated -HA. Both the apatite and hydroxyapatite structures are preferred for the crystalline metal oxide component used in the catalyst of the present invention. These structures are further defined, with reference to atomic positions, by Kikuchi, Masanori; Yamazaki,

Atsushi; Otsuka, Ryohei; Akao, Masaru; Aoki, Hideki. Crystal structure of Sr- substituted hydroxyapatite synthesized by hydrothermal method. J. Solid State Chem. (1994), 113(2), 373-8.

Depending on the specific intended application for the catalyst of the present invention, incorporation of nickel within the framework of the crystalline metal oxide component may provide sufficient activity without further treatment. Synthesis of the crystalline metal oxide without framework nickel, requires the addition of a nickel component (e.g. Ni or NiO) for catalytic activity. It may also be desirable to disperse a nickel component onto the surface of the crystalline metal oxide that also has nickel incorporated into its framework. Any dispersal of the nickel component onto the crystalline metal oxide occurs prior to the calcination step.

The nickel component may be deposited, if desired, onto the crystalline metal oxide in any suitable manner that results in a uniform dispersion. Usually the nickel component is deposited by contacting (e.g. impregnating) the crystalline metal oxide component with an aqueous solution of a nickel salt selected from the group consisting of nickel nitrate, nickel chloride, nickel bromide, nickel acetate, etc. Contacting may be effected by conventional means such as dipping, spraying, etc. at contacting conditions (e.g. time, temperature, and solution concentration) as required to achieve a desired loading of dispersed nickel component. A preferred method of nickel component deposition uses a steam-jacketed rotary dryer to achieve evaporative impregnation. The crystalline metal oxide component particles are immersed in an impregnating solution of any of the previously mentioned salt solutions, where the slurry of metal oxide component and nickel solution are contained in the dryer. Rotation of the dryer tumbles the crystalline metal oxide component therein and the application of steam to the dryer jacket expedites evaporation of the impregnation solution in contact with the tumbling crystalline metal oxide. After completing impregnation, the resulting nickel containing crystalline metal oxide particles are dried at a temperature from 20°C to 250°C for 1 to 48 hours. The crystalline metal oxide, now having a nickel component dispersed thereon, is then calcined according to the technique outlined above. The calcination generally converts the nickel component, despite deposition form, to substantially nickel oxide.

In addition to the catalytic component, other metals, such as catalyst promoter metals, may be deposited simultaneously or sequentially onto the crystalline metal oxide component by substantially the same procedure as described above for contacting the crystalline metal oxide component with a nickel solution. Promoter metals for enhancing activity and/or selectivity include lanthanum and cerium.

The catalyst of the present invention finds use in previously described syngas processes. These processes comprise contacting a light hydrocarbon and an oxidant under effective conditions and in the presence of the crystalline metal oxide catalyst. Light hydrocarbons are preferably the C_j- paraffins, namely methane, ethane, propane, and butane. Natural gas, an impure mixture of methane and other components, is also a preferred feed. Olefins and other unsaturates may be used but tend to form polymers and other undesirable side products. Preferred oxidants are oxygen, when used for partial oxidation, and steam, when applied to steam reforming. The reaction mixture may incorporate pure oxygen, but more commonly incorporates air or enriched air. The use carbon dioxide as an oxidant is known in CO₂ reforming.

The crystalline metal oxide catalyst of the present invention improves processes for partial oxidation, steam reforming, autothermal reforming, and CO₂ reforming. These processes use known conditions that generally range from a temperature of 600°C to 1200°C, an absolute pressure of 5 to 60 atmospheres, and a gas hourly space velocity (GHSV) of 500 to 500,000 hr^"1. GHSV is the volumetric hourly gas flow rate of the all feed components divided by the catalyst bed volume.

A tendency for reduced coke make distinguishes the catalyst of the present invention when applied to any of the above processes. It is believed that the incorporation of nickel into the framework structure of the crystalline metal oxide provides a degree of dispersion previously unknown in the art. Conventional surface- impregnated catalytic metals, even when well dispersed initially, are prone to migration and agglomeration under the severe conditions of reforming. The resulting agglomerated metal sites can initiate the growth of both filamentous and layered carbon. In contrast, by attempting to incorporate nickel uniformly into the framework of the catalyst of the present invention during the initial synthesis, the high initial dispersion and strong interactions are believed to prevent the migration and massive agglomeration of nickel that are characteristic of conventional catalysts. Overall, the catalysts of the present invention substantially diminish the sites for the nucleation of carbon growth.

The catalyst of the present invention relaxes the requirement for excess oxygen or steam injection to suppress coking in partial oxidation and reforming reactions. Excess oxygen or steam means the amount beyond the stoichiometric requirements for the reaction or beyond that necessary for optimal syngas composition such as diluent steam added in partial oxidation to inhibit the formation of catalyst coke.

For partial oxidation, catalyst coking at similar conditions relates closely to the atomic carbon to oxygen atom ratio (C:O ratio) in the reaction mixture. Conventional catalysts typically show a substantial rate of coking at a C:O ratio of greater than 0.5 (i.e. a CH₄:O₂ ratio of greater than 1). In contrast, catalysts of the present invention produce minimal coking in partial oxidation processes at preferred C:O ratios that range from 0.5 to 2.0. While the catalyst of the present invention provides a reduced coking tendency in dry partial oxidation processes, the ability to lower the injection rate of steam as a coke inl ibitor benefits steam reforming as well. The measure for catalyst coking tendency in steam reforming is generally the C:H₂O ratio, or carbon to steam molar ratio. When using the catalyst for steam reforming, the carbon to steam molar ratio can equal or exceed 2.0. The advantages may be manifested in a longer catalyst life and/or a reduction in utility costs (e.g. compression), heating duties, and raw material consumption as a result of operating under higher severity (e.g., higher C:O ratios). In following examples illustrate the some catalyst species of the present invention. EXAMPLES EXAMPLE 1

Preparation of Na-Ni-Sr-P-O

A 159 gram portion of 85% phosphoric acid (H₃PO₄) was added to 200 grams of de-ionized water in a glass beaker. Similarly, 220.8 grams of NaOH was dissolved in 300 grams of de-ionized water. Both solutions were placed in an ice bath. A third solution was prepared by dissolving 131.4 grams of Sr(NO₃)₂ and 20.0 grams of Ni(NO₃)₂»6 H₂O in 331.4 grams of de-ionized water. The chilled NaOH solution was slowly added to the

H₃PO₄ solution with stirring and the resulting mixture was transferred to a 2-liter plastic bealcer and placed under a high speed stirrer. Adding the Sr/Ni solution to the phosphate solution with vigorous mixing formed a gelatinous precipitate. The reaction mixture was homogenized for an hour, transferred to two 1 -liter teflon bottles and digested therein at

100°C for 4 days at autogenous pressure. The solid products were isolated by filtration, washed thoroughly with de-ionized water, dried at room temperature and then calcined at 800°C for 1.5 hours.

Elemental analysis showed an empirical formula Na_o ^Ni_o.πSr_{! 55}PO₄₂₉₅ for the catalyst while x-ray diffraction showed the catalyst to have the hydroxyappatite structure with NiO barely detectable. Characteristic lines in the x-ray diffraction pattern for the catalyst are given in Table 1. This material is designated SrP-HA.

Table 1

EXAMPLE 2

Preparation of Na-Ni-Ca-P-O

A 56.25 gram portion of 85% phosphoric acid (H₃PO₄) was added to 150 grams of de-ionized water in a glass bealcer. Similarly, 156 grams of NaOH was dissolved in 180 grams of de-ionized water. Both solutions were placed in an ice bath. A solution was prepared by dissolving 103.5 grams of Ca(NO₃)2»4 H₂O and 14.0 grams of Ni(NO₃)2«6 H₂O in 110 grams of de-ionized water. The chilled NaOH solution was slowly added to the H₃PO₄ solution with stirring. The resulting mixture was transferred to a 2-liter plastic bealcer and placed under a high speed stirrer. Addition of the Ca/Ni solution to the phosphate solution with vigorous mixing formed a gelatinous precipitate. The reaction mixture was homogenized for an hour, transferred to two 1 -liter teflon bottles and digested at 100°C for 4 days at autogenous pressure. The solid products were isolated by filtration, washed thoroughly with de-ionized water, dried at room temperature, and calcined at 800°C for 1.5 hours.

Elemental analysis showed an empirical formula on an anhydrous basis of Na_αι5Ni_{< 7}Ca_1.60PO₄._34S for the catalyst while x-ray diffraction pattern showed the catalyst to have the hydroxyappatite structure with a minor NiO impurity. Table 2 gives characteristic lines in the x-ray diffraction pattern for the catalyst.

Table 2

EXAMPLE 3

Preparation of Ni impregnated Na-Ca-P-O

A 627 gram portion of de-ionized water was weighed. Some of this water diluted 160 grams of 85% phosphoric acid (H₃PO₄) and some dissolved 223 grams of NaOH. These solutions were chilled in an ice bath. The final portion of the water dissolved

164.36 grams of Ca(NO₃)₂ ^»4 H₂O. The chilled sodium hydroxide solution was stirred into the chilled phosphoric acid solution and the mixture was placed under a high-speed stirrer. Addition of the calcium nitrate solution to the sodium phosphate solution with vigorous stirring formed a gel in the process. After the addition, the reaction mixture was further homogenized for 20 minutes. The reaction mixture was then transferred to a teflon bottle and digested at 100°C for 72 hours under autogenous pressure. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature. X-ray diffraction showed the hydroxyapatite structure for the calcium phosphate solid. A solution was prepared by dissolving 20.32 grams of Ni(NO₃)₂*6H₂O in 200 grams of H₂O. A 74 gram portion of the calcium phosphate solid was added to the Ni solution and slurried using a stirrer. The slurry was added to a steam-jacketed rotary evaporator and rolled in the evaporator with steam applied after 15 minutes of rolling until the solid was dry. The resulting Ni-impregnated solid was dried in an oven at 150°C overnight and then calcined at 800°C for 1.5 hours.

Elemental analysis showed the solid to have the empirical formula on an anhydrous basis of Na₀_₁₈Nio.ι₆Ca ₅₅PO_4>13. Characteristic lines in the x-ray diffraction pattern for the catalyst are shown in Table 3.

Table 3

EXAMPLE 4

Preparation of Ni impregnated Na-Sr-P-O

200 grams de-ionized water diluted a 155.6 gram portion of 85% phosphoric acid (H₃PO₄). 250 grams de-ionized water dissolved 216 grams NaOH to prepare a sodium hydroxide solution. An ice bath chilled both solutions before mixing them to make a sodium phosphate solution, which was placed in a teflon beaker under a high-speed stirrer. 279 grams of de-ionized water dissolved 142.85 grams of Sr(NO₃)₂ into a separate solution that was added to the sodium phosphate solution with vigorous stirring to forming a gel. Addition of 250 grams of de-ionized water reduced the thickness of the mixture which was homogenized for 1 hour, transferred to 2 1 -liter teflon bottles and digested at 100°C for 2 days at autogenous pressure. The solid products were isolated by filtration, washed with de-ionized water and dried at room temperature. X-ray diffraction showed the strontium phosphate product to have the hydroxyapatite structure.

Dissolving 19.98 grams of Ni(NO₃)₂ ^»6H₂O in 200 grams of water prepared a nickel nitrate solution to produce a slurry by stirring a 111 gram portion of the strontium phosphate solid into the solution. Rolling the slurry in a steam-jacketed rotary evaporator evaporator and steam application after 15 minutes of rolling continued until the solid was dry. The resulting Ni-impregnated solid was dried in an oven at 150°C overnight and then calcined at 800°C for 1.5 hours. Elemental analysis showed the solid to have the empirical formula on an anhydrous basis of Na₀₁₇Ni₀.ι₇Srι ₅₄PO₄₂₉₅. Analysis by x-ray diffraction showed the catalyst to consist of Hydroxyapatite and NiO components. Characteristic lines in the x- ray diffraction pattern for the catalyst are given in Table 4.

Table 4

EXAMPLE 5

Preparation of Cs-Na-Sr-Ca-P-O Crystalline Metal Oxide

148 g de-ionized water dissolved 21.44 g Na₂HPO₄*7H₂O to form a solution that received 31.93 g CsOH (50%). The mixture was placed under a heidolph mixer.

Separately, 25.0 g de-ionized water dissolved 7.5 g Sr(NO₃)₂, 8.37 g Ca(NO₃)₂-4H₂O, and 2.29 g Ni(NO₃)₂«6H₂O to form metal nitrate solution was vigorously mixed with the phosphate solution. The resulting mixture thickened, but became less viscous with further stirring. The homogenization continued for 2 hrs. The reaction mixture was placed in a teflon bottle and digested at 100°C at autogenous pressure for 48 hrs. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature. The dried solid was calcined for 5 hr at 800°C before testing.

Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Cs₀₀₀₂Na₀₁₆₄Ni₀₀₇₆Sr₀₇₅₅Ca₀₇₅₇PO_4.253. X-ray diffraction of the catalyst showed a mixture of hydroxyapatite, designated CaSrP-HA, a minor new material component designated SrCaP-1, presented in Example 8, and a minor NiO component. Table 5 gives characteristic lines in the x-ray diffraction pattern for the crystalline metal oxide.

Table 5

EXAMPLE 6

Preparation of Na-Ni-La-Sr-P-O

300 g of de-ionized water diluted 46.12 g H₃PO₄ (98%) and 300 g of de-ionized water separately dissolved 65.95 g NaOH. Ice chilled both solutions for an hour and which were then carefully mixed. Separately, 300 g of H₂O dissolved 7.50 g of

Ni(NO₃)2^»6H₂O, 31.8 g of Sr(NO₃)₂, and 21.6 g La(NO₃)₃ to form a metal nitrate solution. The phosphate solution received an additional 300 g de-ionized water under a high-speed mixer. The nitrate solution was stirred into the phosphate solution, homogenized for an hour, placed in a teflon bottle, and digested at 100°C at autogenous pressure for 48 hr. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature, and calcined for 5 hr at 800°C before testing.

Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Na₀₁₉₉Ni₀₀₆₄La₀₃₃₃Sr₀₉₉₅PO₄₁₆₂. Analysis of the catalyst by x-ray diffraction showed the major component to have the hydroxyapatite structure, designated LaSr-HA, along with Sr₃(PO₄)₂. NiO was barely detectable in the diffraction pattern.

Characteristic lines in the x-ray diffraction pattern for the catalyst are given in Table 6.

Table 6

EXAMPLE 7

Preparation of Na-Sr-Ni-P-O

220g de-ionized water diluted, a phosphoric acid (85%), 57.65g. Similarly, 220g de-ionized water dissolved 82.50g NaOH. Both of these solutions were chilled for an hour in ice. Dissolving 14.54 g Ni(NO₃)»6H₂O and 14.3 g Sr(NO₃)₂ in 100 g de-ionized water prepared a metal nitrate solution. The phosphate solution was carefully added to the NaOH solution under a high-speed mixer. This was followed by the addition of the metal nitrate solution and further homogenization of the mixture for an hour before its placement in a teflon bottle and its digestion at 100°C for 48 hr at autogenous pressure. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature. The dried solid was calcined for 5 hr at 800°C before testing.

Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Na₀₁₀₅Sr_{ι 52}Ni₀₃₅PO₄₄₂₃. X-ray diffraction showed the major catalyst component to have the hydroxyapatite structure. Sr₃(PO₄)₂ and NiO were also observed. Table 7 shows characteristic lines in the x-ray diffraction pattern for the catalyst.

Table 7

EXAMPLE 8

Preparation of Na-Ni-Ca-Sr-P-O

602g de-ionized water dissolved 13.74g of sodium hydroxide pellets into a solution that received 88.67g of

Separately, lOO.Og de-ionized water dissolved Sr(NO₃)₂, 35.0g, Ca(NO₃)₂-4H₂O, 39.06g, and Ni(NO₃)₂*6H₂O, 10.69g to provide a metal nitrate solution was mixed vigorously into the phosphate solution. Homogenization of the reaction mixture for an hour preceded its placed in a teflon bottle and digestion at 95 °C for 48 hr at autogenous pressure. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature. The dried solid was calcined for 5 hr at 800°C before testing.

Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Na₀₁₂Ni₀₁₄Ca₀₇₁Sr₀₇₁PO_{4 !2}. The x-ray diffraction data indicated the catalyst to contain a new structure designated here as SrCaP-1, which indexed on a hexagonal unit cell with lattice parameters a = 10.614 A and c = 19.147 A. Characteristic lines in the x-ray diffraction pattern for the catalyst are given in table 8. A small amount of NiO was also detectable in the pattern.

Table 8

EXAMPLE 9

Preparation of Na-Ni-Ca-P-O 140g of de-ionized water diluted phosphoric acid (85%), 230.6g, and 1860g of de-ionized water dissolved 330.0 g of sodium hydroxide pellets. An ice bath chilled both solutions before their careful combination and addition of another 1500 g de-ionized water with vigorous mixing. The resulting slurry was transferred to a 5-liter flask and fully dissolved by heating to 35°C. Separately, 500g of de-ionized water dissolved 236.15g of Ca(NO₃)₂»4H₂O and Ni(NO₃)₂ ^»6H₂O, 32.31 g. The metal nitrate solution was added to the dissolved slurry over a period of about 2 minutes. The flask was fitted with a temperature controller, condenser, and a mechanical stirrer. The reaction mixture was heated at 100°C for 60 hr. The solid products were isolated by filtration, washed with de- ionized water, dried at room temperature, and calcined for 5 hr at 800°C before testing. Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Na_0.π2Ni₀.π_sCa_{! 64}PO₄₃₇. X-ray diffraction showed the catalyst to have the hydroxyapatite structure, designated CaP-HA. A minor NiO component was also observed. Table 9 gives characteristic lines in the x-ray diffraction pattern for the catalyst.

Table 9

EXAMPLE 10

Preparation of Na-Ni-Sr-P-O lOOOg of de-ionized water dissolved 330.0g of NaOH pellets. An ice bath chilled the solution that entered a 5 liter flask. The flask received 400g of phosphoric acid (85%), 230.6g, diluted with 400.0 g de-ionized water by stirring, and another lOOOg of water. Heating to 50°C dissolved the resulting suspension. A metal nitrate solution, prepared by dissolving Sr(NO₃)₂, 211.63g, and 32.31g Ni(NO₃)₂ ^«6 H₂O in 600g de- ionized water, was stirred into to the flask. The flask was equipped with a condenser and a temperature controller. With vigorous stirring, the reaction mixture was digested for 60 hours at 100°C. The solid products were isolated by filtration, washed with de-ionized water, and dried at room temperature. The dried solid was calcined for 5 hr at 800°C before testing.

Elemental analysis on the calcined solid yielded an empirical formula on an anhydrous basis of Na₀._09gNi₀_₁₆₉Sr₁__slgPO₄.₂₃6. X-ray diffraction showed the catalyst to consist of a single phase designated SrNiP-HA, having the hydroxyapatite structure. Table 10 gives characteristic lines in the x-ray diffraction pattern for the catalyst.

Table 10

EXAMPLES 11-17

Catalyst Coking Studies

Testing in a methane partial oxidation environment showed the relative coking tendencies of the catalyst compositions of the present invention described in Examples 1, 2, 4, 6, and 8. Their coke formation rate, conversion, and selectivity were compared to conventional partial oxidation catalysts comprising nickel on an alumina support (Example 16) and further comprising a magnesium oxide promoter (Example 17). The testing procedure loaded a 2.4g catalyst sample, screened to a 40-60 mesh particle size range, into a quartz reactor. The reactor was heated to 800°C under a nitrogen purge at atmospheric pressure. At 800°C, the nitrogen purge was stopped and a combined stream of methane (Matheson, 99.99% pure) and air (Matheson) entered the catalyst bed. Based on the flow rates of these feed components the carbon to oxygen atom ratio (equivalent to ^lA of the carbon/O₂ ratio) ranged from 1.04 to 1.09 and the gas hourly space velocity (GHSV) ranged from 17,000 to 23,000 hr^"1. Each test continued under constant conditions for 48 hours. During this time, the average methane conversion (except in Example 15) varied from 86-89%, with a molar selectivity to CO and H₂ partial oxidation products in excess of 97%. In Example 15, the methane conversion was 75% with molar selectivities to CO and H₂ of 87% and 93%, respectively. On-line using Gas Chromatography analyzed the gaseous products exiting the reactor. Table 11 summarizes C/O ration and the analyzed carbon content of the spent catalyst after each test.

Table 11

This study shows that catalyst compositions within the scope of the present invention (Examples 11-15) perform equivalently to conventional partial oxidation catalysts in conversion and molar selectivity. However, these catalysts had surprisingly reduced rates of coke formation as clearly supported by the accompanying Figure. The exceptional characteristics of catalysts of the present invention allow, in commercial operation, for extended operation or operation under more severe conditions (e.g. at an increased C/O atom ratio) that provide a higher quality syngas product.

Claims

CLAIMS:

1. A catalyst composition for the production of synthesis gas from light hydrocarbons, the catalyst composition comprising a crystalline metal oxide component having a chemical composition on an anhydrous basis expressed by an empirical formula of:

A_v(B^t+)_wNi_xD(G^u")_yO_z

where A is an alkali metal selected from the group consisting of Li⁺, Na⁺, K⁺, Rb⁺, Cs⁺, and mixtures thereof, "v" is the mole ratio of A to D and varies from 0 to 2, B is a basic metal, "w" is the mole ratio of B to D and varies from 1 to 3, "t" is the weighted average valence of B and varies from 2 to 3, "x" is the mole ratio of

Ni to D and varies from 0 to 0.5, D is a framework component selected from the group consisting of P⁺⁵, V*^"5, and mixtures thereof, and G is an anionic species selected from the group consisting of Off, Cl^", F^", CO₃ ^2", and mixtures thereof, "u" is the average valence of G and varies from 1 to 2, "y" is the mole ratio of G to D and varies from 0 to 2, "z" is the mole ratio of O to D and has a value determined by the equation:

z = ^lA (v + t»w + 2»x + 5 - u*y),

2. The catalyst composition of claim 1 where B is selected from the group consisting of Ca²⁺, Sr²*, Pb ⁺, Cd²⁺, Ba²⁺, La³⁺, Eu³⁺, Gd³⁺, Pr³*, Nd³⁺, Sm³⁺, Y³⁺, Yb³⁺, and mixtures thereof.

3. The catalyst composition of claim 1 where the crystalline metal oxide component has the apatite or hydroxyapatite crystal structure.

4. The catalyst composition of claim 1 further comprising, when "x" is greater than 0, a nickel component dispersed on the crystalline metal oxide component.

5. A process for preparing the catalyst composition of any of claims 1-4 comprising:

a) reacting a mixture containing reactive sources of B, optionally Ni, D, and optionally A, at a pH from 8 to 14 and a temperature and time sufficient to form the crystalline metal oxide component, the mixture having a composition expressed by:

h A₂O : j BO,_/2 : k NiO : D₂O₅ : I N : m H₂O

6. The process of claim 5 further comprising carrying out step (b) when "k" is greater than 0.

7. The process of claim 5 where the D framework component is phosphorous and the reactive source is selected from the group consisting of orthophosphoric acid, pyrophosphoric acid, alkali phosphates, sodium metaphosphate and mixtures thereof.

8. A process for producing synthesis gas comprising reacting a light hydrocarbon and an oxidant at reaction conditions in the presence of a catalyst composition according to any of claims 1-4.

. The process of claim 8 where the reaction is selected from the group consisting of partial oxidation, steam reforming, autothermal reforming, and CO₂ reforming.

10. The process of claim 10 where the oxidant comprises oxygen, air, enriched air, steam, carbon dioxide, and mixtures thereof; where the oxidant comprises oxygen, air, or oxygen-enriched air and the reaction conditions include a carbon to oxygen atom ratio from 0.5 to 2.0 and; or where oxidant comprises steam and the reaction conditions include a carbon to steam molar ratio of less than 2.