Auto thermal reforming (ATR) catalytic structures

Abstract

An autothermal reforming catalytic structure for generating hydrogen gas from liquid hydrocarbons, steam and an oxygen source. The autothermal reforming catalytic structure includes a support structure and nanosized mixed metal oxide particles dispersed homogenously throughout the support structure.

Claims

1. A method for formulating an autothermal reforming (ATR) catalytic structure consisting essentially of: obtaining a basic solution having a basic pH of between 10 and 12, the basic solution comprising an alkaline metal hydroxide, an alkaline metal carbonate, and water; obtaining an acidic solution having an acidic pH of between 4 and 6; mixing the basic solution with the acidic solution to create a sol-gel having layered double hydroxide (LDH) precursors through co-precipitation of the cations with the basic solution, the LDH precursors having layered crystal structures, the layered crystal structures consisting essentially of 20% by weight nickel oxide, 5% by weight magnesium oxide, and 75% by weight aluminum oxide; heating the sol-gel at a decomposition temperature of 500° C. for a time period of 8 hours such that the LDH precursors are at least partially decomposed to form a calcined material; and conducting a metal reducing step on the calcined material for collapsing the layered crystal structures within the LDH precursors using a gas mixture having molar hydrogen in a range of from about 5% to about 20% and nitrogen at a temperature in a range of from about 450° C. to about 600° C. to form an ATR catalytic structure, where the ATR catalytic structure has nanosized mixed metal oxide particles dispersed throughout having diameters in a range of from about 40 nm to about 300 nm.

2. The method as claimed in claim 1, wherein the nanosized mixed metal oxide particles have a surface area between 100 and 300 m.sup.2/g.

3. The method as claimed in claim 1, wherein the nanosized mixed metal oxide particles are homogeneously distributed throughout the ATR catalytic structure.

4. The method as claimed in claim 1, wherein the nanosized mixed metal oxide particles are operable to reform hydrocarbons for hydrogen production at temperatures less than about 800° C.

5. The method as claimed in claim 1, wherein the basic solution further comprises a mixture of NaOH and Na.sub.2CO.sub.3, such that the basic solution has a pH of about 12.

6. The method as claimed in claim 1, wherein the acidic solution has a total cationic concentration of 1.5 M.

7. The method as claimed in claim 1, wherein the cations have an aluminum concentration of between about 20 and 35 mol %.

8. The method as claimed in claim 1, wherein the ATR catalytic structure is operable to produce a hydrogen product stream from a feed stream having liquid hydrocarbons through the use of an ATR reaction when incorporated into a porous tubular support.

9. The method of claim 1, where the nanosized mixed metal oxide particles includes both α- and γ-alumina.

10. A method of formulating an autothermal reforming (ATR) catalyst consisting essentially of: a. preparing a basic solution having a pH of between 10 and 12; b. preparing an acidic solution having a pH of between 4 and 6; c. mixing the acidic solution and the basic solution together to form a sol-gel; d. aging the sol-gel to form a formed solid; e. washing and filtering the formed solid with water until a generally neutral pH is reached; f. drying the formed solid for a predetermined period of time to form a dry solid, the dry solid consisting essentially of 20% by weight nickel oxide, 5% by weight magnesium oxide, 75% by weight aluminum oxide; g. calcining the dry solid at a temperature of 650° C. for a time period of 8 hours to form a calcined material; and h. reducing the metal in the calcined material by contacting the calcined material with hydrogen at a temperature in a range of from about 450° C. to about 600° C. to form the ATR catalytic structure, the ATR catalytic structure having nanosized mixed oxide particles having diameters in a range of from about 40 nm to about 300 nm dispersed throughout.

11. The method as claimed in claim 10, wherein the basic solution of step a is prepared by combining NaOH and Na.sub.2CO.sub.3 such that the basic solution has a pH of about 12.

12. The method as claimed in claim 10, wherein the ATR catalytic structure is incorporated into a porous tubular support and is operable to produce a hydrogen product stream from a liquid hydrocarbon feed stream through the use of an ATR reaction.

13. The method as claimed in claim 12, wherein the hydrogen product stream is substantially free from carbon monoxide.

14. The method as claimed in claim 12, wherein the hydrogen product stream comprises no more than 0.5% by volume carbon monoxide.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) These and other features, aspects, and advantages of the present invention will become better understood with regard to the following description, claims, and accompanying drawings. It is to be noted, however, that the drawings illustrate only several embodiments of the invention and are therefore not to be considered limiting of the invention's scope as it can admit to other equally effective embodiments.

(2) FIGS. 1-3 include graphical representations of experimental data collected in accordance with embodiments of the present invention.

(3) FIG. 4 is an embodiment of the present invention.

(4) FIG. 5 is cross sectional view of an embodiment of the present invention.

(5) FIG. 6 is an axial view of an embodiment of the present invention.

(6) FIGS. 7-27 include graphical representations of experimental data collected in accordance with embodiments of the present invention.

DETAILED DESCRIPTION

(7) While the invention will be described in connection with several embodiments, it will be understood that it is not intended to limit the invention to those embodiments. On the contrary, it is intended to cover all the alternatives, modifications and equivalence as may be included within the spirit and scope of the invention defined by the appended claims.

(8) Nano sized catalysts provide advantages over their full size counterparts, particularly in autothermal and WGS reactions, in that they increase reaction activity, improve selectivity, improve reactivity towards hydrogen production, and minimize the undesirable methanation reaction.

(9) Carbon formation (as coke), which is a primary disadvantage of a reforming process, is a kinetic issue. As such, coke production depends on the relative reaction rates of possible carbon species reaction alternatives. The reaction mechanism is largely dependent on the hydrocarbon type, operating conditions, and catalyst characteristics. For example, catalyst characteristics can influence the reaction mechanism by increasing water adsorption-dissociation rate on the catalyst and gasification rate with respect to the C—C scission.

(10) Catalyst characteristics are generally determined by their physical-chemistry, composition, structural, and textural properties, such as: active area, metal particle size, metal dispersion, and reducibility. These properties depend on metal-support interaction, and they could be established on different stages of catalyst synthesis. For example, varying the material composition of a precursor, the preparation method and/or the heat treatments (calcination or reduction), can provide desired characteristics of the catalyst for optimum performance.

(11) The present invention provides novel mixed metal oxides that are obtained by thermal decomposition of layered double hydroxides (LDHs) and offers the opportunity to control a catalyst's active site nature and its environment, as well as catalyst texture and stability. LDHs are a unique class of layered materials having positively charged layers and charge balancing anions located in the interlayer region. Typically, LDHs can be synthesized by the co-precipitation of metallic salts with a concentrated alkaline solution. An alternative method for the preparation of LDHs is through the sol gel method. Thermal treatments of the mixed metal oxides resulting from the LDHs prepared by the sol gel method can lead to materials that demonstrate synergetic effects between the elements in the mixed metal oxide structure, and after appropriate activation treatment, give rise to well dispersed metal particles like supported metal catalysts, with the possibility of controlling metal-support interaction during the synthesis stage.

(12) In one embodiment of the present invention, a new catalytic structure that can be used for reforming reactions is provided. In one inventive method of preparing said catalytic structure, several different types of LDHs can be prepared using a sol-gel precipitation method. In one embodiment, two aqueous solutions are prepared; one acidic and one basic. In certain embodiments, the sol is base catalyzed. The acidic solution preferably contains one or metal salts, such as magnesium salts, and specifically nitrates and salts of the same for nickel and aluminum using a desired Al/(Ni+Mg+Al) ratio. The pH of the acidic solution is preferably within the range of about 4 to about 6. In one embodiment, the total cationic concentration of the metal salts in the acidic solution is about 1.5 M. The basic solution can be obtained by mixing suitable amounts of sodium hydroxide and sodium carbonate in order to maintain a ratio of carbonate ions to nickel, magnesium and aluminum ions of around 0.7 and a pH for synthesis of approximately 12.

(13) After the solutions are prepared, they can then be added into a large mixing device and put into mechanical agitation for an appropriate period of time to form a sol-gel. In one embodiment, the appropriate amount of time is determined by measuring the concentration of metal oxide. In one embodiment, 95% concentration of metal oxide is acceptable. At concentrations at or above 95%, a preferred amount of reactant has been consumed, and agitation can be stopped. In one embodiment, ICP-AES analysis can be used to verify the metal oxide concentration. In one embodiment, this mixing step can be for up to about 5 hours. Following the mixing step, the obtained sol-gel is left to age at a temperature of between about 50 and 75° C., alternatively at a temperature of about 60° C. for an appropriate amount of time, which can be up to about 6 hours, alternatively up to about 10 hours, alternatively up to about 15 hours, or greater than 15 hours. During aging, at least a portion of liquid within the sol-gel evaporates. In one embodiment, rapid aging is not preferred as it can detrimentally impact gel characteristics. The resulting solid can then be filtered and washed with distillated water until the wash water has a neutral pH of about 7. Water washing is desirable to remove all undesired and converted reactants. Those of ordinary skill in the art will understand neutral pH to including a pH within the range of about 6.5 to 7.5. Next, the resulting washed solid LDH material can be dried at about 100° C. for an appropriate amount of time, which can be up to about 5 hours, alternatively up to about 10 hours, alternatively up to about 15 hours. In one embodiment, the LDH material can then be calcined in air at temperatures up to 550° C., alternatively at a temperature of between about 400° C. and 600° C., alternatively at a temperature between about 400° C. and 500° C., alternatively at a temperature between about 500° C. and 600° C., for a period of time, which can be up to about 5 hours, alternatively up to about 8 hours, alternatively up to about 12 hours. The resulting solid product is a catalyst precursor of a mixed metal oxide that includes Ni/Mg/Al. Supported metal catalysts suitable for autothermal reforming reaction of liquid hydrocarbons into hydrogen can be obtained from these precursors after a metal reducing step by contacting with a gas mixture that includes hydrogen and can also include nitrogen at a temperature of about 500° C., alternatively between about 400° C. and 550° C., for a period of time, which is up to about 5 hours. In one embodiment, the gas mixture contains about 5% molar hydrogen, alternatively between about 1 and 10% molar hydrogen, alternatively between about 5 and 10%, alternatively between about 5 and 15%, alternatively between about 10 and 20% molar hydrogen.

(14) During the metal oxide reduction step, at temperatures greater than about 300° C., the LDL structure will collapse, and in the presence of hydrogen, the metal oxide can be reduced to its elemental state. The collapsed structure was confirmed using XRD (and is shown in FIGS. 1-3). In general, the collapsed structure has an increased density, relative to the starting material, and a porosity that is sufficient for diffusion of reactants into the catalyst to undergo the desired reactions. In certain embodiments, the porosity of the collapsed material can remain consistent with the starting material, having a porosity that is in the range of about 65-70%, which in turn can facilitate the desired reforming reactions.

(15) Experimental Design for Catalytic Structure:

(16) The following synthesis parameters were investigated to obtain catalyst structures in accordance with embodiments of the present invention: Al/Mg mole ratio between 0.2-5 in the form of mixed metal oxides Calcination temperature: 500° C.-600° C. Reduction temperature: 500° C. Concentration of Hydrogen: 5 to 20% by mole Nickel incorporation method: Co-precipitation simultaneously with magnesium and aluminum or impregnation on Al—Mg support.

(17) In order to study the synthesis parameter effects on catalytic characteristics of these materials, different series of samples were prepared. In each different sample, only one parameter was changed. The baseline material, identified as CP10C, had the following parameters: Nickel content: 10 wt % Cation ratio [Al/(Al+Mg+Ni)]: 0.20 Calcination temperature: 550° C. Reduction temperature: 500° C. Nickel incorporation method: co-precipitation with magnesium and aluminum.

(18) To determine the Al/Mg ratio effect on catalytic behavior, several catalysts were prepared wherein the Al/Mg ratio was varied between 0.2 and about 5.

(19) To determine the calcination temperature effect on catalytic behavior, several mixed oxides catalysts were prepared wherein the calcination temperatures were varied between about 400° C. and about 800° C.

(20) To determine reduction temperature effect on catalytic behavior, several catalysts were prepared wherein the reduction temperatures was varied between about 450° C. and about 700° C.

(21) To determine the effect of the nickel incorporation method on catalytic behavior, several catalysts were prepared by co-precipitation and by impregnation. Catalysts marked with “LO” were prepared by impregnation, and catalysts marked with a “CP” were prepared by co-precipitation.

(22) The results and characteristics for each of the materials were measured using several different techniques in order to correlate catalytic activity with corresponding structural properties. Catalyst support surface area was measured using the BET technique employing nitrogen physisorption at the temperature of liquid nitrogen in a Quantachrome Autosorb-1C instrument. The percentage of metal loading was measured by Inductively-Coupled Plasma Atomic Emission Spectrometry (ICPAES). X-ray Diffraction (XRD) was carried out using Rigaku Miniflex diffractometer employing a Cu—Kα radiation source (30 KV/15 mA). The average particle size for different phases present on each state of the sample (calcined, reduced or used) was estimated by the Scherre equation, which is reproduced below:

(23) $\tau =\frac{K\lambda }{\beta \cos \vartheta }$
where K is the shape factor, λ is the x-ray wavelength, β is the line broadening at half the maximum intensity (i.e., full width at half maximum) in radians, θ is the Bragg angle; and τ is the mean size of the ordered (crystalline) domains.

(24) Nickel surface area can be measured using well established advanced analytical techniques, such as hydrogen chemisorptions complemented by temperature programmed reduction (TPR) where it is possible to estimate total metal surface area, size of metal particle and metal dispersion.

(25) Temperature programmed reduction (TPR) was carried out using a 5% H.sub.2/N.sub.2 mixed gas flowed at 50 mL/min and used to study the reducibility of the calcined samples by means of hydrogen consumption. From TPR curves, the temperature at which a maximum curve appears (nickel reduction temperature) can be used to determine the most efficient reduction conditions. The degree of reduction can be estimated by comparing TPR curves corresponding to calcined and reduced samples. Coke formation on samples during activity test can be determined by analysis of the temperature-programmed oxidation (TPO). Autothermal reforming reaction study was carried out in a down low fixed-bed reactor catalytic system.

(26) Approximately 1 gram of the catalyst was used having particles the size of about 30-50 mesh without dilution. The catalyst tested was pre-reduced in an H.sub.2 gas stream (5%) at a temperature of about 500° C. for about 5 hours. Two thermocouples controlled the operation temperature, wherein a first thermocouple was positioned in the oven and the other thermocouple was positioned at the center of the catalyst bed. All reactants were introduced from the top of the reactor. Air was used as a source of oxygen. Two HPLC pumps fed water and hydrocarbon, which were mixed and evaporated before being introduced into the reaction region. The pressure of the reaction was maintained by a backpressure regulator connected with a precision gauge to read the pressures. Time zero was measured at the point when the reactor contents were heated to the chosen reaction temperature, which usually took about 45 minutes. Gas samples were removed through the gas sampling system throughout the length of the test once the reaction temperature was reached. Effluent from the reactor was cooled in a double pipe condenser to condense the condensable vapors. The gas and liquid samples were analyzed by gas chromatography.

(27) The autothermal reforming reaction was investigated at different reaction conditions as follows: Hydrocarbon feed: n-octane, Reaction Temperature: 500° C., 550° C. and 600° C. Total Pressure: 3 bars Weight Hourly Space Velocity (WHSV): 5000 h.sup.−1 Oxygen/Carbon (molar ration): 0.5 Steam/Carbon (molar ratio): 3

(28) The general reaction procedure included packing the catalysts into the reactor, starting the reactor temperature controller and then, following catalyst reduction, setting the desired temperature condition. Approximately, fifteen minutes before the test began, steam was passed through the catalyst bed. Following this, the air flow meter with the water and hydrocarbon pumps were turned on all together at the established conditions.

(29) Each experiment ran for approximately eight hours. The first two hours were conducted at temperature of about 500° C., and then increased to a temperature of about 600° C. for a further two hours, and again increased to a temperature of about 700° C. for two hours and finally returning to a temperature of about 500° C. for the final two hours. Every thirty minutes, a reformate gas sample was taken for analysis to determine the molar concentrations of H.sub.2, O.sub.2, CH.sub.4, CO and CO.sub.2. Volumetric flow (ml/min) and density (g/ml) of the liquid reformate products were also tested, and were passed through a Gas Chromatography-Flame Ionization Detector (GC-FID) assay.

(30) Two Shimadzu GC-17A Gas Chromatography (GC) units were used to estimate the composition of the product gases and liquid reformate collected from the ATR reaction. The GC units were equipped with thermal conductivity (TCD) and flame ionization (FID) detectors. Specifically, the GC-TCD was used to evaluate the presence of H.sub.2, O.sub.2, N.sub.2, CH.sub.4, CO and CO.sub.2 using a ⅛ inch 6 ft Carbosphere 80/100 packed column. The light and heavy liquid hydrocarbons were analyzed using a GC-FID coupled with a 100 m×0.25 mm ID BPI-PONA capillary column. Liquid product identification was also carried out in a Shimadzu GCMS-QP5050A mass spectrometer equipped with a DB-5 column.

(31) For the GC-TCD method, the injector was held at 100° C. and detector temperature was 150° C. The oven temperature was initially maintained at a temperature of 40° C. for 5 minutes, and then increased to a temperature of 120° C. at a rate of 5° C./min. In the case of the GC-FID, the injector was held at 280° C. and detector temperature was 320° C. The oven temperature was initially maintained at a temperature of 60° C. for 5 minutes, and then increased to 280° C. at a rate of 10° C./min.

(32) Referring to FIG. 1, x-ray diffraction spectra for 4 samples prior to calcination are provided. Samples LO1 and LO2 were prepared by impregnation, and samples CP1 and CP2 were prepared by co-precipitation. The samples had compositions of about 20.5% by weight NiO, 5.5% by weight MgO, and 75% by weight Al.sub.2O.sub.3. The samples were all air-dried at 110° C. for approximately 5 to 15 hours. As shown in the x-ray diffraction spectra stacked plot shown in FIG. 1, the individual spectra for each sample are relatively similar, thus demonstrating that the synthetic methods utilizing impregnation and co-precipitation, whether at a pH of 10 or 12, produces material having like crystallographic structure. Additionally, FIG. 1 also displays the LDH structure of different aluminum/magnesium crystallite sizes, along with the crystallinity at different catalyst synthesis conditions.

(33) FIG. 3 shows x-ray diffraction spectra for four calcined catalyst samples having a composition of about 20.5% by weight NiO, about 5.5% by weight MgO, and about 75% by weight Al.sub.2O.sub.3, wherein the samples were calcined at a temperature of about 650° C. for a period of about 8 hours. The materials correspond to the calcined versions of the materials shown in FIG. 1. The spectra show that samples LO2C and CP2C both have peaks at about 38, 46 and 66, which correspond to mixed metal oxide (as opposed to LDH structures). Sample CP2C also includes a slight peak at about 60, corresponding to metal oxide. In contrast, sample LO1C has peaks at about 38, 42, 46, 63 and 66, and sample CP1C has peaks at about 38, 42, 62, and 66, although the peak at 66 is of a lower intensity.

(34) Based on the experimental results, which are shown in FIGS. 1-3, Applicants surprisingly discovered the following:

(35) (1) Catalysts obtained by calcination/reduction of LDH materials proved to be suitable catalyst materials for the autothermal reforming of liquid hydrocarbons for hydrogen production. These catalyst materials exhibit highly dispersed metallic crystallites or grains that are stable inside a matrix, and advantageously have a high surface area.

(36) (2) The catalyst materials obtained in accordance with various embodiments of the present invention can be used in catalytic membranes to convert liquid hydrocarbons, steam, and air into hydrogen and carbon dioxide. The catalytic activities test produced nearly total liquid hydrocarbon conversion with large levels of hydrogen produced without any carbon monoxide production, thus indicating an excellent balance of ATR and WGS combined reactions. Additionally, the WGS reaction was favored over the methanation reaction, resulting in a reduced amount of methane gas being produced. Moreover, coke formation was also reduced, which improves overall catalytic activity, thus leading to longer run times.

(37) (3) Nickel incorporated into the catalyst materials by co-precipitation results in more highly and finely dispersed nickel particles than the impregnation method.

(38) (4) Synthesis parameters have an important influence on metal-support interaction. When stronger metal-support interactions are favorable, the catalyst structure is improved. Additionally, catalyst performance is strongly linked to the nickel particle size, as well as the selected support material and its properties.

(39) As noted previously, prior art catalytic membranes were used to produce and subsequently remove hydrogen through hydrogen permeable metal membranes, such as palladium or palladium alloys. These membranes, in particular palladium or palladium alloys, however, are expensive, sensitive to sulfur compounds, and difficult to sinter with or co-sinter onto a catalyst layer. Additionally, such devices typically produced hydrogen only by the WGS reaction.

(40) Some embodiments of the current invention integrate catalysts useful for both the ATR reaction and the WGS reaction, as well as providing a hydrogen permeable membrane that does not include palladium or a palladium alloys, which reduces costs while increasing the overall H.sub.2 yield of the process. In one embodiment, the ATR structure described herein can be incorporated into a catalytic membrane reactor (CMR) assembly, which is operable to perform the ATR reaction, the WGS reaction, and remove hydrogen.

(41) In one embodiment of the invention, a method for the production of hydrogen by oil reforming processes reduces the overall cost of producing high quality hydrogen from liquid oil, as compared to the prior art processes. In one embodiment, the method can include a gasification, steam reforming, partial oxidation, autothermal reforming, or like step, depending on the characteristic of the oil feed processed to obtain synthesis gas. This synthesis gas requires subsequent cleaning as it is shifted to produce additional hydrogen gas. In one embodiment, one or more of these steps can be combined to improve efficiency. In certain embodiments of the present invention, additional novel technologies, such as membrane separation and catalytic reactors, have been developed that can help to solve these needs. Additionally, some embodiments of the present invention are also operable to remove unwanted products, such as CO.sub.2, to thermally provide one stream that can be used in a secondary process, such as in enhanced oil recovery from depleted oil reservoirs.

(42) In one embodiment, methods for producing hydrogen from alternate hydrocarbon sources of oil (such as, gasoline, kerosene, diesel, petroleum coke, heavy residues, and the like) involves the step of first reacting oil with oxygen and/or steam to produce a gas mixture that includes carbon monoxide, carbon dioxide and hydrogen. The carbon monoxide can then react with steam to produce additional hydrogen and CO.sub.2. Finally, the hydrogen and CO.sub.2 can be separated, either by removing the CO.sub.2 from the mixture or by removing hydrogen from the mixture. The removal of at least a portion of the hydrogen gas will advantageously shift the reaction equilibrium toward the product side, allowing a lowering of reaction temperature and use of a decreased amount of steam. An optional cleaning step can also be employed, but CO contamination can be controlled to make such contamination negligible. For example, concentrations of CO as low as 0.001% can be obtained, making the hydrogen substantially free of carbon monoxide. In certain embodiments, all of the hydrogen produced can be removed.

(43) The combination of reaction and separation processes, as described herein for the CMR assembly built in accordance with embodiments of the present invention, offers higher conversion of the reforming reaction at lower temperatures due to the step of removing hydrogen gas from the steam reforming and WGS equilibrium reactions. Thus depending on the feed oil composition used, a membrane reactor as part of an engineering process, can allow one step reforming and/or partial oxidation with WGS reaction and parallel hydrogen separation. Unlike conventional prior art processes, the CMR assembly equipped with the catalysts described herein, benefits from high pressure operation due to the increased hydrogen partial pressure differences across the membrane, which act as the driving force for hydrogen permeation. For example, the pressure on the permeate side can be atmospheric or under vacuum. When a higher pressure on the retentate side is applied, the pressure difference across the membrane acts as a driving force for hydrogen to permeate through the membrane. The higher the pressure difference, the higher the amount of hydrogen permeating through the membrane. Those of ordinary skill in the art will recognize that the mechanical property of the membrane used will create a practical limit as to the pressure that can be applied on the retentate side.

(44) Another embodiment of the present invention discloses a method of manufacturing a catalytic coated silica membrane for the conversion of liquid petroleum hydrocarbon fuels into high purity hydrogen. The embodiment includes the step of providing a membrane tube, which includes an outer surface covered with an active silica layer that is highly permeable to hydrogen. A mixed metal oxide catalyst, which can include at least one metal selected from the group consisting of rhodium, platinum, nickel, ruthenium, palladium, rhenium, iridium, and combinations thereof, can be deposited within the pores of the alumina framework of the membrane tube. During use, air, steam, and liquid hydrocarbons are transported through the membrane and are in intimate contact with the metal sites. Following activation of the catalyst membrane, the feed components react and form hydrogen through a combination of ATR and WGS reactions. High purity hydrogen can be produced by separation of the hydrogen gas from the product mixture through the hydrogen permeable membrane deposited on the CMR assembly outer surface. Certain embodiments of the present invention bring a number of benefits to hydrogen production from liquid petroleum hydrocarbon fuels. These improvements can include high conversion of the liquid hydrocarbons, high molar yield of the hydrogen produced, low molar yield of the residual methane and low catalyst deactivation as a result of the shifting of reaction equilibrium to favor the forward reaction to nearly 100% at a much lower operating temperature range between about 500° C.-550° C. Typical hydrogen purity in this one-step conversion of liquid petroleum hydrocarbon to hydrogen can range from about 96 to 99% molar concentration, alternatively at least about 97% molar concentration, alternatively at least about 98% molar concentration, or alternatively between about 97 and 99% molar concentration.

(45) In one embodiment, the CMR assembly in accordance with an embodiment of the present invention can be selectively permeable to hydrogen and produce a hydrogen-rich permeate product stream on the permeate side of the membrane and a carbon dioxide rich product retentate. The CMR assembly can be used to produce a hydrogen-rich permeate product stream that is greater than about 99% by volume hydrogen. The CMR assembly can be a composite ceramic material having an outer hydrogen transport and separation layer, which in one embodiment is a metal doped silica. The CMR assembly can be also composed of one or more inner catalytic layers. The ATR and WGS reactions occur on the inner metal layers of the CMR assembly and the produced hydrogen can be transported and removed through the outer metal-doped silica layer.

(46) In one embodiment, the metal catalysts are capable of catalyzing the conversion of hydrocarbon fuels to hydrogen and carbon oxides (CO and CO.sub.2). In one embodiment, the catalyst can include one or more of the following metals: nickel, ruthenium, platinum, palladium, rhodium, rhenium, and/or iridium.

(47) The CMR assembly can include a stainless steel vessel that surrounds the tubular support and is positioned between a pair of high temperature material shells. In one embodiment, the high temperature material shells can help to provide sealing, manifolding, expansion support, separated regions for the catalytic reactions, delivery of pressurized feedstock, support of the membrane, and removal of product gases.

(48) In accordance with one embodiment of the present invention, the stainless steel vessel was constructed of 316 grade stainless steel. In an exemplary embodiment, a cylindrical α-alumina tube having an outer diameter of about 10 mm and an inner diameter of about 8 mm with an average pore size of 1.3 micro meters and an approximate porosity of 0.55 was used as ceramic tube support for the membrane. For the preparation of the active catalytic sites and the membrane separation pores, the following procedure was adopted:

(49) First, the α-alumina ceramic tube was soaked in a boehmite sol-gel, having a concentration in the range of about 5 to 10% by weight boehmite. After removing the excess boehmite by drip drying, the ceramic tube was dried by natural convection at room temperature for several hours and then at about 110° C. for up to about 10 hours. It was followed with calcination at a temperature of about 500° C. that resulted in the bimodal distribution for the original α-alumina structure into alpha and gamma phases. During the calcination step, the nitrates and other gaseous oxides were evaporated and converted to the metal oxide form of the active metals and support structure (e.g., NiO, MgO, and Al.sub.2O.sub.3). For the materials used, the preferred calcination temperature was between 500° C. and 800° C. At temperatures above 800° C., it was discovered that the metal oxide structure changed to a dense, low surface area material, which greatly reduced its catalytic activity. To introduce the catalyst metal active sites to the tubular structure, the ceramic tube was soaked in a solution containing salts of metal species, such as nickel, ruthenium, platinum, palladium, rhodium, rhenium, iridium, and combinations thereof, at the appropriate concentration. Drying and calcination steps were subsequently performed, resulting in a uniform metal distribution within the core of the ceramic tubular structure. The final step was to dope the outer surface of the metal loaded catalytic ceramic tube with a silica colloidal sol-gel to produce an even thickness and uniform pore size hydrogen permeable layer. In one embodiment, this process can be repeated several times to remove large pores that might result in pinholes in the final membrane. In one embodiment, the silica colloidal sols were coated on the ceramic tube by a hot coating method in which the tube was first heated to a temperature of up to around 180° C. and then subsequently and quickly contacted with a wet cloth containing the sols. This hot coating method helped make the sol dry more quickly, preventing the sol from penetrating deep into the pores. The catalytic membrane can be cut in pieces and then characterized by Scanning Electron Microscopy (SEM), nitrogen adsorption, porosimetry and hydrogen adsorption.

(50) In order to test the hydrothermal stability of the silica membrane, a stability experiment was carried out at about 500° C. using several different molar ratios of steam and liquid hydrocarbon. The permeability-selectivity of the ceramic tube to hydrogen was measured at timed intervals. In general, a very slow drop in the permeation characteristics due to humidity was observed. Typically, at high temperature, silica interacts with steam causing densification of the material. Within the temperature ranges used in the experiment (e.g., between about 500° C. and about 600° C.), the membrane did not show any substantial change in its permeation properties. In addition, the influence of the molar ratio of water (steam) to hydrocarbon on liquid hydrocarbon conversion within the temperature range of 500° C. to 550° C. showed that the conversion was enhanced at high steam to hydrocarbon molar ratios, such as a steam to hydrocarbon molar ratio of greater than about 3:1, or alternatively between about 2:1 and about 3:1. The hydrocarbon feed reacted very effectively with steam and air through the ATR reaction steps. A fraction of the CO produced during the ATR reaction subsequently reacted with steam in the WGS reaction, leading to increased hydrogen yield. In-situ removal of hydrogen by the membrane further enhanced conversions of both of these reactions. In addition, enhancement in the rate of CO removal by the WGS reaction can reduce coke deposition caused by the so-called Boudouard reaction.

(51) In FIG. 4, CMR assembly 10 includes: stainless steel vessel 12, shells 14, ceramic tube 20, feed stream inlet 30, retentate output 40, sweep stream inlet 50, and permeate output 60. The feed stream (not shown), which includes liquid hydrocarbons, steam, and the oxygen source, enters CMR assembly 10 thru feed stream inlet 30. The feed stream then moves into ceramic tube 20 where it undergoes both ATR and WGS reactions, thereby forming hydrogen and carbon dioxide. The hydrogen that permeates thru the permeable membrane of ceramic tube 20 is removed by a sweep gas, such as nitrogen, argon, steam, and combinations thereof, that enters thru sweep stream input 50 and exits permeate output 60.

(52) FIG. 5 shows an embodiment in which ceramic tube 20 has ATR layer 22 and WGS layer 24 disposed within the tube in two separate layers. Those of ordinary skill in the art will recognize that the ATR catalysts and WGS catalysts can also be disposed within the same layer. In the embodiment shown in FIG. 2, the feed stream enters ceramic tube 20 where it reacts with ATR catalysts disposed in ATR layer 22. This reaction converts the hydrocarbon, oxygen, and steam in the feed stream into hydrogen gas and carbon monoxide gas. The carbon monoxide can react with additional steam and the WGS catalysts disposed in WGS layer 24 to form carbon dioxide and additional hydrogen gas. The hydrogen gas then travels radially outward through the pores of γ-alumina layer 26 and then through membrane 28. The carbon dioxide gas cannot permeate membrane 28, and exits ceramic tube 20 on the retentate side of membrane 28. Unconverted hydrocarbons and carbon dioxide exit ceramic tube 20 through retentate output 40 of FIG. 4.

(53) FIG. 6 is an axial view of ceramic tube 20 as shown in FIG. 5 along line 6-6.

(54) Experimental Design for CMR Assembly:

(55) Commercial alumina tubes (obtained from Noritake, Japan) of 100 mm length and 10 mm ID were used for the CMR assembly. The tubes include α-alumina having a porosity of about 30% and average pre size of about 0.5-1 μm. A γ-alumina layer having a pore size of about 4 nm was disposed on the outer surface of the tubes. A metal doped sol-gel was prepared by mixing 120 g of tetraethyl orthosilicate (TEOS) in 600 g ethanol to form a solution. An acid solution was then synthesized by dissolving 14.84 g cobalt nitrate hexahydrate (Co(No.sub.3).sub.2.6H.sub.2O) in 51.77 g of 30% aqueous H.sub.2O.sub.2. Subsequently, both solutions were vigorously mixed together by stirring for about 3 hours in ice-cooled bath. The tubes were then externally coated with a stable Si—Co—O solution using a controlled immersion time of 1 minute and withdrawn speed of 2 cm/min. Sintering was then carried out at about 600° C. for about 4 hours at a heating rate and a cooling rate of 0.7° C./min.

(56) The internal area of the alumina tube was wetted with a sol-gel that includes a boehmite solution having a concentration in the range of 5-10 wt % of aluminum. Excess solution was wiped off, and the tube was dried for several hours at a temperature of about 110° C., and then fired at a temperature of about 500° C. for about 8 hours, which resulted in an inlet tube wall of γ- and α-alumina having a bimodal distribution. Subsequently, this internal layer of the tube was soaked with chloroplatinic acid and rhodium chloride solutions (0.5 wt % platinum and 2.5 wt % rhodium, respectively), followed with drying, calcination, and reduction under hydrogen to finely disperse the active metal particles over the γ- and α-alumina pores.

(57) In one embodiment for the preparation of the chloroplatinic acid and rhodium chloride solutions, the following procedure was used: TEOS was mixed together with the metal solutions, water and ethanol according to the following:

(58) TEOS: 0.9 parts

(59) Pt—Rh: 0.5 to 2.5 parts

(60) Water: 4 parts

(61) HNO.sub.3: 0.01 parts

(62) Ethanol: 10 parts

(63) Afterwards, the resulting solution was hydrolyzed for about 12 hours at room temperature. After adding an appropriate amount of a 0.1 M solution of nitric acid, the solution was left at room temperature for about 5 hours to convert into a colloidal sol solution. It was then used to coat the internal wall of the membrane tube.

(64) A silica metal doped hydrogen separation layer having a thickness of about 1 μm was formed on the external wall of the membrane tube. An internal, highly dispersed Rh—Pt catalytic layer, having a metal dispersion of about 65%, was impregnated on the bimodal layer (which includes α- and γ-alumina) to act as a catalytic layer for the simultaneous ATR and WGS reactions to occur in the internal wall of the membrane tube.

(65) Experimental Results of CMR Assembly:

(66) A set of initial experiments were conducted using the CMR assembly described above and shown in FIGS. 4-6, and a commercially available autothermal catalyst (FCR-71D autothermal catalyst provided by Sud Chemie). One gram of the commercially available catalyst was used under similar experimental conditions tested on a fixed bed reactor (FBR) during catalyst selection: Temperature of 500° C.; Pressure of 3 bars; oxygen:carbon ratio of 1:2; steam:carbon ratio of 3:1; and a hydrocarbon feed of commercial gasoline. FIG. 7 shows the total gas product distribution at the end of both the retentate and the permeate side of the membrane. Hydrogen concentration on the permeate side was observed to be as high as 92% by volume. A 77 wt % recovery of hydrogen was obtained, with the remainder of the product stream being a mixture of carbon monoxide, carbon dioxide, methane and unconverted oxygen. A total conversion of about 98% of commercial gasoline was achieved.

(67) The hydrogen yield produced using a fixed bed reactor system and when using a CMR assembly was compared and can be seen in FIG. 8. As shown in FIG. 8, embodiments of the present invention utilizing the CRM assembly provide a distinct advantage over fixed bed reactors due to shifting the conditions of the reaction equilibrium.

(68) In FIGS. 9 and 10, the temperature was increased to 550° C. Notably, the CMR reactor had a higher selectivity of hydrogen (about 15% more) as compared to the fixed bed operation. Additionally, as shown by comparing FIG. 9 and FIG. 10, the CMR operation reduced the production of methane by approximately 40%, as compared with the fixed bed reactor.

(69) FIG. 11 shows comparisons of the average concentrations of hydrogen, carbon monoxide and methane in the product streams for fixed bed operation and CMR operation as function of temperature (500° C. and 550° C.). Comparing both fixed bed reactor and CMR operations, an increased amount of hydrogen is produced at an operating temperature of 500° C. than is produced at an operating temperature of 550° C. Additionally, at increased temperatures, the reduction in methane produced is very significant. On the retentate side of the reaction, analysis of the CO demonstrates nearly the same values for both operations.

(70) The strength of the silica membrane and the hydrogen separation through the retentate was tested for both the CMR and FBR reactors at 600° C., using the same reaction parameters as provided above with respect to FIGS. 7-9. The experiment was conducted for about 6.5 hours, however, after completion, leaks were observed through the membrane and through the graphite sealing. FIG. 12 represents a comparison of the results on hydrogen selectivity are compared as function of temperature and type of reactor used. As shown in FIG. 12, hydrogen is produced in amounts greater than the thermodynamic equilibrium conditions at temperatures of 500° C. and 550° C. Additionally, both CMR and fixed bed reactors show decreased hydrogen selectivity with increased temperature.

(71) FIGS. 13 and 14 compare the selectivity of two reactors to methane and carbon monoxide. When using CMR operation, the methane yield is significantly reduced compared with the values for FBR. This behavior can be attributed to the fact that in using the CMR device, the thermal cracking reaction of the fuel and the Boudouard reaction that causes methanation, are inhibited because hydrogen is continually being removed through the silica membrane, which in turn causes more carbon monoxide to be consumed. On the other hand, the carbon monoxide yield between the two operation modes remains practically unchanged. As such, CMR operations increase yields of hydrogen, reduce yields of methane, and have a negligible effect on carbon monoxide production.

(72) Table I presents the comparison of gasoline conversion, total hydrogen produced and the percentage of hydrogen recovered through the membrane as function of the reaction temperature. Increased reaction temperature favors the hydrogen recovery from the membrane, although the total hydrogen yield decreased.

(73) TABLE-US-00001 TABLE I Composition Production for CMR as a Function of Temperature Temperature Gasoline Hydrogen % Hydrogen (Celsius) Conversion (%) produced (mol %) Recovered 500 89 72 90 550 94 64 94

(74) FIG. 15 shows the levels of catalyst deactivation for the two types of reactors. As shown in FIG. 15, carbon deposition on the catalyst surface is lower for CMR operation as compared with FBR operations. Additionally, carbon deposition on the catalyst surface is nearly independent of temperature for CMR operation, contrasting with FBR operation where carbon deposition increased by nearly 100% as the temperature was increased from about 500° C. to about 600° C.

(75) Coke deposit selectivity can be seen in FIG. 16, which shows the temperature programmed oxidation (TPO) analysis of the coked catalysts using CMR and FBR operations. The results indicate that coke deposition from catalysts used in CMR operation mainly covers the support sites of the catalysts, thereby leaving the metal active sites clean and free from deactivation. FBR, on the other hand, shows deposition on support sites and also on metal sites and in the vicinity of metal-support sites. This coke deposition can break the reaction chain to produce hydrogen as reflected by the high methane and light unsaturated hydrocarbons that were found during FBR operations.

(76) FIGS. 17 and 18 show the gas selectivity for permeate and retentate for the membrane sample CMR-1. In this case, hydrogen purity in the permeate was greater than about 95 mol %. Relatively low amounts of CO and methane were produced, which is an indication that the methanation reaction is suppressed and the WGS reaction is enhanced.

(77) FIGS. 19 and 20 show the gas selectivity for permeate and retentate for membrane sample CMR-2. CMR-2 is a tube having similar physical characteristics to that of CMR-1, and it was used to show reproducibility. Similar to CMR-1, CMR-2 displayed a high hydrogen purity in permeate of around 90 mol %. CMR-2 also showed good membrane stability and very low production of CO and methane, which is an indication that the methanation reaction is suppressed and the WGS reaction is enhanced.

(78) FIGS. 21 and 22 show the gas selectivity for permeate and retentate for membrane sample CMR-3. CMR-3 is a tube having similar physical characteristics to that of CMR-1 and CMR-2; however, it was from a different supplier, and was used to show reproducibility. Similar to CMR-1 and CMR-2, use of CMR-3 resulted in a high hydrogen purity in permeate of around 90 mol %. CMR-3 also showed good membrane stability and very low production of CO and methane, which is an indication that the methanation reaction is suppressed and WGS reaction is enhanced.

(79) FIGS. 23-25 show the hydrogen purity concentration on permeate and retentate for each of the 3 membranes; CMR-1, CMR-2 and CMR-3. FIGS. 23-25, show that CMR-1 has the largest differences in hydrogen purity between the permeate and the retentate, indicating that CMR-1 is the membrane with the best hydrogen recovery rates (i.e., about 82%), which is shown in FIGS. 26 and 27.

(80) There are various alternative uses for the CMR assembly as constructed. For example the membrane system can be used as a hydrogen extractor for removing hydrogen from refinery gas streams such as the catalytic cracker off gas, hydrodesulfurization gas, etc. Other uses can also include hydrogenation, dehydrogenation, fuel cells, and the like.

(81) By integrating a hydrogen perm-selective silica layer with a reforming catalytic layer, an efficient and compact CMR assembly for reforming liquid hydrocarbon fuels into hydrogen has been shown. The system shows increased performance in the reforming of liquid hydrocarbons, at comparatively lower operating temperatures and lower steam to carbon molar ratios, than is typically the case for conventional fixed-bed reforming processes. The process efficiency, to a large extent, depends on various process parameters. Under optimized conditions, a nearly 30% improvement from the equilibrium conversion levels was achieved as a result of continuous hydrogen removal from the product stream through the CMR assembly that employs a hydrogen permeable silica membrane integrated with the catalyst layer. The system offers a promising alternative to conventional hydrogen permeable membrane reactors that are typically packed with either reforming or WGS catalysts.

(82) Use of terms such as “first” or “second” are strictly for labeling purposes and do not connote order of operations.

(83) While the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the foregoing description. Accordingly, it is intended to embrace all such alternatives, modifications, and variations as fall within the spirit and broad scope of the appended claims. The present invention may suitably comprise, consist or consist essentially of the elements disclosed and may be practiced in the absence of an element not disclosed.