IRON OR COBALT SELECTIVE TOWARDS HIGHER HYDROCARBONS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit from earlier filed U.S. Utility Patent Application No. 13/072,683, filed on March 26, 2011, the entire disclosure of which is incorporated by reference herein for all purposes.

BACKGROUND

Field of the Invention

[0002] The present teachings are directed towards a method for producing hydrocarbons by a Fischer-Tropsch mechanism and the catalysts for the method. The hydrocarbon producing method includes providing a catalyst of a manganese oxide-based octahedral molecular sieve nanofibers with at least one of iron, cobalt, nickel, copper, manganese, vanadium, zinc, and mixtures thereof, and further containing an alkali metal. The formation of iron carbides and cobalt carbides by exposing the catalyst to conditions sufficient to form those carbides is also taught. After the catalyst has been appropriately treated, a carbon source and a hydrogen source are provided and contacted with the catalyst to thereby form a hydrocarbon containing product. The characteristics of the hydrocarbon products can be controlled by the formulation and treatment of the catalyst.

Discussion of the Related Art

[0003] Production of clean fuel and fine chemicals via Fischer-Tropsch ("FT") synthesis has attracted interest in both academia and industry. Catalysts (Co, Fe, Ru, and Ni) for the conversion of CO2 supported on inert silica, alumina, zeolites, and carbon nanotubes have been developed for FT synthesis. The hydrocarbon products of these catalysts are mainly paraffins. Additional manganese oxides have been used to increase the selectivity towards long chain olefins, for instance, alkenes, but these manganese oxides decreased overall activity due to their enrichment on the catalyst surfaces. [0004] K-OMS-2 (KMn ³⁺ Mn ⁴⁺ ₇ 0i ₆ ) is composed of 2 x 2 edge-shared Mn0 ₆ ^" octahedral chains, which are corner shared to form one-dimensional tunnels (4.6 x 4.6 A), with K ⁺ ions located in the tunnels. The unique expandable structure and components suggest that K-OMS-2 nanofibers can be suitable templates for engineering design of FT catalysts. The high surface area, 60-150 m ² /g OMS-2 nanofibers can be supports for incipient wetness impregnation ("IWI") of Co, Fe, and Cu-based catalysts. Additionally, the mixed valence K-OMS-2 structure can be selective oxidation catalysts.

SUMMARY

[0005] The present disclosure teaches a catalyst based on manganese oxide-based octahedral molecular sieve nanofibers, (also known as synthetic cryptomelane), an active catalyst metal of at least one of iron, cobalt, nickel, copper, manganese, vanadium, zinc, and mixtures thereof, and an alkali metal typically present as a promoter. In some embodiments of the catalyst, the active catalyst metals are present as phases of iron carbide or cobalt carbide.

[0006] A method for producing hydrocarbons by an FT mechanism from CO and/or C0 ₂ hydrogenation is also disclosed herein. This method can include providing a catalyst composed of a manganese oxide-based octahedral molecular sieve nanofiber, at least one of iron, cobalt, nickel, copper, manganese, vanadium, zinc, and mixtures thereof, and an alkali metal. The method also features the formation of iron carbides and cobalt carbides by exposing the catalyst to conditions sufficient to form those carbides. After the catalyst has been appropriately treated, a carbon source and a hydrogen source are provided and contacted with the catalyst to thereby form the desired hydrocarbon containing product.

[0007] The presently disclosed OMS-2 supported Co and Fe catalysts can be utilized for the highly selective and efficient production of jet fuel, alkenes, C ₂ -C6 carboxylic acids, a-hydroxylic acids, and their derivatives via C0 ₂ and CO hydrogenation under different conditions. Neither of the OMS-2 supported Fe, Co, and Co/Cu catalysts formulations for FT synthesis nor the selective production method of the high value hydrocarbons by adjusting the oxidation ability of K-OMS-2 supports have been previously disclosed. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate preferred embodiments of the invention and together with the detailed description serve to explain the principles of the invention. In the drawings:

[0009] Figs. 1A - ID illustrate the FESEM micrographs;

[0010] Fig. 2 illustrates XRD patterns;

[0011] Figs. 3A and 3B illustrate TEM and EDS spectra;

[0012] Fig. 4 illustrates in situ XRD patterns;

[0013] Fig. 5 illustrates in situ XRD patterns;

[0014] Fig. 6 illustrates CO/C02 conversion activity;

[0015] Fig. 7 illustrates distribution of hydrocarbon by carbon number;

[0016] Fig. 8 illustrates TPR-MS spectra;

[0017] Fig. 9 illustrates FT-IR spectra;

[0018] Fig. 10 illustrates in situ XRD patterns;

[0019] Fig. 11 illustrates in situ XRD patterns;

[0020] Fig. 12 illustrates GC-MS spectra;

[0021] Fig. 13 illustrates distribution of hydrocarbon products;

[0022] Fig. 14 illustrates distribution of hydrocarbon products, and

[0023] Fig. 15 illustrates in situ FT-IR spectra.

DETAILED DESCRIPTION

[0024] Disclosed in the present application is a catalyst for producing hydrocarbons by means of the FT reaction mechanism, the catalyst can be composed of manganese oxide -based octahedral molecular sieve nanofibers, at least one of iron, cobalt, nickel, copper, manganese, vanadium, zinc, and mixtures thereof, and an alkali metal. Generally, iron-containing catalysts will have iron present in one or more iron carbide phases, and cobalt-containing catalysts will have cobalt present in one of more cobalt carbide phases. Carbide phases include, for example, Fe ₂ C, Fe ₂ . ₅ C, Fe ₃ C, Co ₂ C, and Co ₃ C. The preferred carbides for each metal are the Fe ₃ C and the Co ₃ C phases, respectively.

[0025] In some embodiments, the alkali metal can be potassium, and it can be present in an amount ranging from about 0.5 wt. % to about 7.5 wt. %, or an amount ranging an amount ranging from about 4.0 wt. % to about 6.0 wt. %. Preferably, the potassium is present as potassium hydrogen carbonate.

[0026] In those embodiments where the catalyst contains a mixture of cobalt and copper, the copper can be present at a Cu:Co molar ratio ranging from about 0.01 : 1 to about 0.5 : 1.

[0027] In those iron-containing embodiments, the catalyst can have iron present at a Fe:Mn molar ratio of no greater than about 1 : 5.

[0028] The alkali metal should be present, in the presently disclosed catalyst, at an alkali metakMn molar ratio of no greater than about 1 : 8.

[0029] The presently disclosed catalyst can include manganese present not only in the manganese oxide-based octahedral molecular sieve support material but also as manganese added via any suitable impregnation or deposition method to the surface of the OMS-2 support to provide a catalytic functionality, either as a catalyst or as a catalyst promoter or modifier.

[0030] The catalysts according to the present disclosure can have the active catalyst species present on the surface of the molecular sieve support material or the catalyst species can be incorporated into the lattice framework of the molecular sieve support material, and in some cases the catalyst species can be present in both locations. Known impregnation and/or deposition methods, such as IWI or CVD, can be utilized to achieve the desired placement of the catalyst species. In some cases of the present catalyst, the catalyst species can be initially impregnated or deposited on the surface of the molecular sieve, and can then migrate to positions within the lattice itself under reaction conditions. Furthermore, the molecular sieve material can be synthesized with the catalyst species directly incorporated into the lattice framework during the synthesis.

[0031] The present disclosure also teaches a method for producing hydrocarbons comprising providing a catalyst containing a manganese oxide-based octahedral molecular sieve, at least one of iron, cobalt, nickel, copper, manganese, vanadium, zinc, and mixtures thereof, and an alkali metal. The catalyst is exposed to conditions sufficient to form carbides of either iron or cobalt, and then contacted with the provided carbon and hydrogen sources to produce a hydrocarbon.

[0032] The carbon source can include carbon monoxide and carbon dioxide, while the hydrogen source can include hydrogen and water.

[0033] The catalyst heat treatments can include heating to a temperature of at least 280 °C under an atmosphere, ranging in pressure from about 2 atmospheres to about 20 atmospheres, comprised of hydrogen and either one of carbon monoxide or carbon dioxide. In some embodiments, the catalyst can be heated to a temperature of at least about 320 °C and no greater than about 450 °C under a similar atmosphere.

[0034] In those embodiments including a catalyst with a mixture of cobalt and copper, preferably with copper is present in an amount ranging an amount ranging from about 0.5 wt. % to about 5.0 wt. %, particularly preferred is a copper loading of about 2.4 wt. %, the hydrocarbon species produced under CO hydrogenation conditions include alpha-hydroxy acids, such as, 4-hydroxybutanoic acid, 5-hydroxypentanoic acid, and also heterocyclic compounds, such as, tetrahydro-2H-pyran-2-one, and tetrahydro-4-methyl-2H-pyran-2-one.

[0035] In the present application, oxidative OMS-2 supported Fe and Co catalysts have been disclosed for CO and C0 ₂ hydrogenation. The unique structure and compositions of OMS-2 can provide three active components (such as, K, Mn, and Fe or Co) as Fischer-Tropsch catalysts. After reduction, HRTEM-EDS and FESEM images show that these components form nanostructural catalysts: a thin layer of Fe or Co on the manganese oxide support. K ₂ 0 is gradually released to the interfaces of Fe or Co and manganese oxide supports. In situ XRD studies have shown a phase transformation of OMS-2, Fe or Co, and K ₂ 0 to MnO, Fe ₃ C or Co ₃ C, and KHC0 ₃ , respectively. 9-Fe ₃ C, Co ₃ C, and KHC0 ₃ are formed under an elevated temperature and high-pressure CO/H ₂ mixture.

[0036] One preferred embodiment of the presently disclosed catalyst is composed of Fe ₃ C, KHC0 ₃ , and MnO and has demonstrated high CO conversion activity (87%) and C0 ₂ conversion (45%) with selectivity of up to 75%> toward C ₂₊ carbon chains. Liquid products like jet fuel (C ₈ -C ₁₅ ), C ₂ -C ₆ carboxylic acids, and a-hydroxy acids can be selectively produced on the oxidative manganese oxide supports. The present catalysts have also shown good stability during a 144 h time -on- stream tests.

[0037] In the present disclosure, in situ X-ray diffraction (XRD) under similar conditions of the FT synthesis has been used to identify the phase transformations under dynamic reaction conditions. High-resolution transmission electron microscopy (HRTEM), field emission electron microscopy (FESEM), and energy dispersive spectroscopy (EDS) show the catalyst nanostructure, components, and their interfaces. Combined with gas chromatography (GC), mass spectroscopy (MS), and nuclear magnetic resonance (NMR) techniques, these studies have led to the presently disclosed OMS-2 supported catalysts.

[0038] As used herein, "CAT A" refers to a Co/Cu catalyst deposited on a K-OMS-2 support; "CAT B" refers to a Co catalyst deposited on a K-OMS-2 support; "CAT C" refers to an Fe catalyst with 5.4 wt. % K present on a K-OMS-2 support, and "CAT D" refers to an Fe catalyst with 0.7 wt. % K present on a K-OMS-2 support.

[0039] When Fe-K-OMS-2 catalysts are reduced in H ₂ or CO, Fe or Fe ₃ C nanoparticles formed on the MnO _x (mainly MnO) surface (see Fig. 3). K ₂ 0 promoters also gradually migrated to the MnO _x surface, and then formed KHC0 ₃ in CO/H ₂ atmosphere.

[0040] Similar processes are involved in the reduced Co ₃ 0 ₄ (CuO)/K-OMS-2 catalysts. Hep Co and K ₂ 0 were formed on the MnO _x surface. Hep Co can convert into fee Co at high temperatures (300-450°C) due to a small energy gap (0.03 eV), and both hep Co and fee Co form Co ₃ C in CO/H ₂ . Co nanoparticles are easier to oxidize by air or H ₂ 0 than Fe nanoparticles, and therefore, these reduced FT catalysts should be kept in an oxygen free dry environment.

[0041] The molar ratios of Fe/Mn and Co/Mn in the OMS-2 supported Fe and Co catalysts can be increased by adding more Fe(N0 ₃ ) ₃ or Co(N0 ₃ ) ₂ in the IWI process. In pure OMS-2, the maximum molar ratio of Fe/Mn in the catalysts is 1/5 because an excess of iron will lead to secondary phases in the framework doped Fe-K-OMS-2. The maximum molar ratio of K/Mn (1/8) in the OMS-2 support is determined by the formula ΚΜη ₈ 0ι ₆ (K-OMS-2). Potassium oxide (K ₂ 0) is formed during the reduction of K-OMS-2. It is noted that pure K ₂ 0 decomposes at 350 °C, and while other metal oxides can stabilize K ₂ 0 in the catalysts, some K ₂ 0 can be lost during calcination of the OMS-2 supported catalysts at elevated temperatures (>450 °C). As-synthesized hydrothermal Fe-K-OMS-2 contained 5.4% of K after reduction in H ₂ /He, higher more than the HT Fe-K-OMS-2 (0.7% K) prepared by a high temperature ("HT") method set forth herein.

[0042] In situ XRD patterns (Figs. 4, 5, 10, 11) show that the formation of carbides and KHCO3 is determined by the reaction temperature. Apparent XRD peaks of carbides (Co ₂ C, C03C, Fe ₃ C, and Fe ₂ .5C) and KHCO3 are present at 220°C, and the corresponding catalysts have low activity in C0 ₂ and CO hydrogenation (Fig. 6A). Co catalysts have higher activity than Fe catalysts at low temperature (220°C). The intensity of carbide peaks keeps increasing until 320°C, indicating the formation of more carbides. The conversion of CO and C0 ₂ also kept increasing with temperature. The catalytic activity of the Fe catalysts is greater than the CoCu catalysts at 280°C. Fig. 6B shows that the CO and C0 ₂ hydrogenation reactions experienced an induction stage and reached a maximum conversion due to the gradual formation of active carbide phases. The H ₂ /He reduced OMS-2 supported Fe catalysts (CAT C) have very limited activity in the beginning of C0 ₂ hydrogenation (Fig. 6A).

[0043] Formation of carbides in the presently disclosed catalysts is desirable to achieve a catalyst active for FT synthesis. Under C0 ₂ hydrogenation conditions, the extra step of the reverse water-gas shift ("RWGS") reaction (see Eq. 1) forms CO which then forms the carbides. The formation of carbides is determined by the concentrations of adsorbed [H] _a d and [CO] _a d on the catalyst surfaces, which are determined, in turn, by the partial pressures of H ₂ and CO and the nature of the catalysts. Fewer Fe ₂ .sC (or Co ₂ C and Co ₃ C) carbides are formed in the C0 ₂ /¾ mixtures, while more Fe ₃ C (Co ₃ C) carbides are formed in the CO/H ₂ mixtures at 320°C. C0 ₂ adsorption on these catalysts is stronger than CO adsorption on the bulk catalysts as determined by in situ FT-IR (Fig. 15).

[0044] In situ XRD patterns (Figs. 4h, 5i, 10 and 11 24 h) show four catalysts with various Co and Fe carbides and they are listed in Table 2. These active catalysts are stable after 24 h reaction. Figs. S5-S6 indicate that the Fe and Co carbides and KHC0 ₃ do not change in C0 ₂ hydrogenation. Therefore, the type and amount of bulk iron (or cobalt) carbides can be controlled in C0 ₂ hydrogenation.

[0045] Table 2 shows that the H ₂ /CO reduced Co and Fe catalysts have 3 and 2.5 times conversion of C0 ₂ than the H ₂ /C0 ₂ reduced Co and Fe catalysts, respectively. These carbide catalysts have shown high selectivity (up to 75%) towards C ₂₊ carbon chains with a high C0 ₂ conversion rate (45%). Fe ₃ C and Co ₃ C are apparently the most active phases among the listed species and Fe ₃ C carbide-containing catalysts are very efficient catalysts towards long chain hydrocarbons. H ₂ /C0 ₂ reduced Co and Fe catalysts can still contain some metallic particles, and these metallic catalysts may account for the high selectivity to CO and CH ₄ .

[0046] Manganese has a wide range of possible oxidation states (2 ⁺ , 3 ⁺ , and 4 ⁺ ) determined by the reduction temperature in the OMS-2 support. TPR-MS spectra (Fig. 8) of the K-OMS-2 support show that the manganese in OMS-2 changed its valences at three temperatures of about 250, 310, and 415 °C due to loss of lattice oxygen during CO/He reduction. The corresponding valences of manganese in OMS-2 catalysts changed in the order of Mn ⁴⁺ — » Mn ³⁺ — »Mn ²⁺ . The oxidation activity also decreased in this order.

[0047] In some embodiments of the present teachings, the catalyst reduction temperature can be varied to change the species of hydrocarbon products. As set forth in Fig. 13, the reduction temperature was increased from 350 to 450°C, which results in Mn0 ₂ in the K-OMS-2 support being fully reduced to less oxidative MnO, and longer chain hydrocarbons are produced in this reaction sequence. These oxidative MnO _x supports are capable of selectively producing light alkenes (C ₂ -C ₆ ) from C0 ₂ (or CO) hydrogenation using OMS-2 supported Fe and Co catalysts.

[0048] According to present theory, the following mechanisms may proceed under C0 ₂ hydrogenation conditions. Adsorbed C0 ₂ are transformed into CO on the supported Co and Fe catalysts via the RWGS reaction (Eq. 1). CH ₂ * radicals are generated from the hydrogenation of CO on the supported Co and Fe catalysts (Eq. 2). One pathway involves carbon chain growth on the metallic phases and metal carbides. Another pathway of carbon chain growth is that CH ₂ * reacts with intermediate KHC0 ₂ (formic acid potassium) and the carbon chain continues to grow with the insertion of more CH ₂ * (Eqs. 5-6). KHC0 ₂ species were formed on the supported Co and Fe catalyst surface via Eq. 4. It is believed that under higher (450 °C) reduction, the intermediate KHC0 ₂ is further reduced. Due to existing Mn ³⁺ and Mn ⁴⁺ ions in low temperature (350 °C) reduced Co catalysts, CH ₂ * radicals are more likely to follow the second pathway to form

carboxylic acids. Moreover, the products of the first pathway may be oxidized by Mn ^x+ (x=3-4) to carboxylic acids, and the chance of growth of long hydrocarbon chains via the first pathway is low. The liquid products are 100% C ₂ -C6 carboxylic acids (Fig. 13 A).

C0 ₂ (g) + H ₂ (g) <-> CO (g) + H ₂ 0 (g)

CO (g) + 2 H ₂ (g) <-> CH ₂ * (ad) + H ₂ 0 (g)

K ₂ 0 (s) + 2 C0 ₂ (g) +H ₂ 0 (g) <-> 2 KHCO3 (s)

KHCO ₃ (s) + H ₂ (g) <-> KHC0 ₂ (s) + H ₂ 0 (g)

KHC0 ₂ (s) + CH ₂ *(ad) <-> CH3COOK (s)

CH ₃ COOK (s) + n CH ₂ * (ad) <-> CH ₃ (CH ₂ ) _n COOK (s)

R-K (s) + H ₂ 0 (g) + C0 ₂ (g) <-> RH (g) + KHCO3 (s)

(R= CH3COO ^" , CH ₃ (CH ₂ ) _n COO ^" , n=l to 4)

[0049] In the present disclosure, in situ XRD patterns show the high-pressure induced growth of single KHCO3 crystals from potassium oxide under FT synthesis conditions. It has been reported by others that addition of potassium, such as K ₂ C03, K ₂ C ₂ 0 ₄ , and KOH, promoted long carbon chain growth and reduced the formation of methane in the FT synthesis. In the method of the present disclosure, single KHCO3 crystals form from potassium oxide in high-pressure CO/H ₂ mixtures. KHCO3 crystals do not form at low pressure (1 atm), and form less in high-pressure C0 ₂ /H ₂ mixtures. This suggests that added potassium may also convert to KHCO3 crystals under the FT reaction conditions. Small KHCO3 crystals (<80 nm) show good promotion effect, and large KHCO3 crystals (294 nm) may actually decrease the catalytic activity of the catalysts (Fig. 6B). Tables 1 and 3 show that the OMS-2 supported Fe catalysts containing 5.4% K and carbide, Fe3C, significantly decreased CH ₄ formation from 52% to 15%. The KHCO3 promoters may decrease the CH ₄ formation activity due to competing adsorption of CO and ¾. According to present theory, and in contrast to previous potassium promotion studies, the K-OMS-2 support may gradually release K ⁺ ions to the catalyst surface to compensate for the loss of K ⁺ at the catalyst surface. This controlled release of K ⁺ prevents overly high enrichment of KHCO ₃ which can, in turn, hinder the adsorption of CO and ¾.

[0050] For some embodiments of the present teachings, Cu dopants are added to the OMS-2 supported CoCu catalysts to shift the products from carboxylic acids to a-hydroxy acids. Ethanoic acid and propanoic acid were produced on the OMS-2 supported Co catalysts. According to present theory, 4-hydroxybutanoic acid can be formed from CH ₃ OH and propanoic acid via the aldol condensation (Eq. 8):

Tetrahydro-2H-pyran-2-one can be formed from 5-hydroxy-pentanoic acid via the intermolecular condensation (Eq. 9). 5-hydroxy-pentanoic acid and tetrahydro-4-methyl-2H-pyran-2-one can be formed via a similar aldol condensation and intermolecular condensation.

[0051] According to the present disclosure, a-hydroxy acids were formed along with C ₂ -C ₃ carboxylic acids when 2.4 wt. % of Cu was doped into the OMS-2 supported Co catalysts (CAT A) (see Fig. 14). The liquid product molar ratio of acetic acid, propanoic acid, 4-hydroxybutanoic acid, 5-hydroxy-pentanoic acid, tetrahydro-2H-pyran-2-one, and tetrahydro-4-methyl- 2H-pyran-2-one, is 63: 18:4:8:5:2. [0052] In some embodiments of the present disclosure, potassium promotion, with for instance KHCO ₃ , enhances the activity and selectivity of the OMS-2 supported catalysts towards highly desirable long carbon chain hydrocarbons in both C0 ₂ and CO hydrogenation processes. Catalysts reduced at temperatures of 400-450°C produced long carbon chain hydrocarbons as discussed above. Compared with the OMS-2 supported Co catalysts reduced at the same temperatures, the Fe catalysts show much higher activity, and longer carbon chain hydrocarbons and less oxygenates were produced. The increased performance of the Fe catalyst is presently believed to be largely dependent on the enhanced RWGS activity of the Fe catalysts over the Co catalysts. Table 4 shows that the CO hydrogenation products of the presently disclosed OMS-2 Fe catalyst contains higher amounts of alkenes (2 times) and aldehydes (3 times), and fewer aromatics (0.5 time) than C0/AI2O3 catalysts.

[0053] In the presently disclosed method, the reduction temperature of the catalysts impacts the selectivity of the products (see Fig. 13). K-OMS-2 supported Co catalysts (CAT B) reduced at a low temperature of 350°C underwent CO ₂ hydrogenation, the liquid products were ethanoic acid (53 molar%), propanoic acid (27%), butanoic acid (14%), and pentanoic acid (6%>). The gas products were CH ₄ (38%), CO (19%), C ₂ H ₆ (16%), and C ₃ -C7 olefins (21%). Aldehydes and other hydrocarbons were not detected by GC-MS in the liquid products. When CAT C was reduced at the high temperature of 450°C, the liquid products of CO ₂ hydrogenation are jet fuel (C ₈ -C ₂ o), and 67%o of the jet fuel is alkenes.

[0054] Typical Fischer-Tropsch products using C0/AI ₂ O ₃ catalysts are mainly paraffins and few olefins, including small fractions of linear and branched alcohols, ketones, carboxylic acids, and linear aldehydes. According to the presently taught method, the products show a different distribution of aldehydes, aromatics, alkenes, and alkanes from typical FT products. As shown in Table 4, for CAT C, the molar ratio of aldehydes/aromatics/alkenes for CO hydrogenation is 1 : 2: 45, compared with 1.0: 1.5: 3.5 for CO ₂ hydrogenation. The same ratio for the C0/AI ₂ O ₃ catalysts in CO hydrogenation is 1 : 13: 67. Note that CO ₂ hydrogenation using the C0/AI ₂ O ₃ catalysts did not produce these liquid products. The low concentrations of aromatics (2%) and aldehydes (1%) present in the CO hydrogenation products of the OMS-2 supported Fe catalysts (CAT C), while the C0 ₂ hydrogenation with the same catalysts produced significant amounts of aldehydes (12%) and aromatics (18%>) is attributed to the strong C0 ₂ adsorption which hinders H ₂ adsorption on these catalysts as described herein.

[0055] Fig. 7 shows that the gaseous products of CO and CO ₂ hydrogenation using the OMS-2 supported Fe and Co/Cu catalysts are mainly C ₂ -C7 light hydrocarbons. GC data shows that the ratio of an alkene (e.g C ₃ H ₆ ) to an alkane (C ₃ H ₈ ) is up to 10: 1. More desirable C ₂ -C ₆ alkenes are selectively produced. This carbon chain growth toward long chains (C ₂₊ ) on the supported Fe catalysts is clearly shown in Figs. 7 A and B in both CO and CO ₂ hydrogenation. In contrast, the long carbon chain growth on supported Co/Cu catalysts is more difficult than on supported Fe catalysts. Jet fuel (C ₇ -C ₁₅ ) has been produced with high selectivity using the OMS-2 supported Fe catalysts in CO and CO ₂ hydrogenation, respectively.

[0056] Typically light products (C ₂ -C ₇ ) are the main products of the OMS-2 supported catalysts with solid wax products are less than 2% of the weight of the total products. However, under specific conditions of a high concentration (> 20 wt. %) of cobalt or iron on manganese oxide supports and a low gas hourly space velocity, the chance of growth of long carbon chains increases. The carbon number can reach 45, and most products are straight chain hydrocarbons including olefins and paraffins (see Fig. 12).

EXAMPLES

[0057] Catalysts Preparation and Characterization

[0058] The synthesis of OMS-2 supports has been described elsewhere; see U.S. Patents Nos. 5,702,673 and 7,767,770, the disclosures of which are incorporated by reference herein in their entireties for all purposes.

[0059] The OMS-2 supported cobalt catalysts were synthesized by an IWI method. Cobalt (II) nitrate (Co/Mn: 0.2, all in molar ratio) and copper nitrate (Cu/Co: 0.1 molar ratio) were dissolved in deionized (DI) water, and then OMS-2 was added into the cobalt nitrate solution under agitation for 2 h. The cobalt loading can range from about 10 to about 25 wt. %. The impregnated samples were dried at 80°C in flowing air for 4 h, then at 120°C for 2 h, and finally calcined in flowing air (200 cm ³ /min) at 450°C for 6-12 h with a ramp rate of 2°C/min using a tube furnace. OMS-2 supported CoCu catalysts (CAT A) and OMS-2 supported Co catalysts (CAT B) were synthesized following the above procedures except that no copper nitrate was added during the CAT B preparation.

[0060] In some instances, the Co/Cu containing catalyst can be produced by CVD coating of a Co K-OMS-2 catalyst with Cu(acac) ₂ utilized as the copper precursor in the CVD reactor.

[0061] Framework doping of Fe ³⁺ ions in OMS-2 (Fe-K-OMS-2) occurred by conversion from Fe-K-birnessite. MnCl ₂ .4H ₂ 0 (0.2 mol) was dissolved in 150 mL of DI water. FeCl ₃ was added into the solution for an Fe / Mn atomic ratio of 1/5 at room temperature under agitation. Oxygen (500 cm ³ /min) was bubbled into the solution. NaOH solution (100 mL of 5 M) was added dropwise into the solution. After stirring for 2 h, the product was filtered, washed, and then transferred to a 250 mL KC1 solution (1 M) for ion-exchange at room temperature (RT) under stirring for 12 h. The Fe, K-birnessite was then washed with DI water.

[0062] Two Fe-K-OMS-2 samples with 5.4% and 0.7% K, respectively, were synthesized using a hydrothermal method and a high temperature calcination method, respectively. For the hydrothermal synthesis, 70 mL DI water, 0.15 mol K ₂ S0 ₄ , and 0.2 mol Fe-K-birnessite were mixed in an autoclave. The autoclave was sealed in an acid digest bomb and placed in an oven at 180 °C for 24 h. As-synthesized Fe-K-OMS-2 products (CAT C) were washed and dried at 120 °C for 12 h.

[0063] For high temperature ("HT") calcination syntheses, Fe-K-birnessite (0.2 mol) was calcined stepwise in oxygen (200 cm ³ /min) at 120, 200, 350, 500, and 625 °C for 2 h in a tube furnace. As-synthesized HT Fe-K-OMS-2 (CAT D) was obtained.

[0064] Powder XRD was performed with a Scintag XDS 2000 X-ray diffractometer equipped with a Cu Ka X-ray source (λ = 1.54 A). Field emission scanning electron microscopy (FESEM) was performed using a Zeiss DSM 982 Gemini FESEM instrument with a Schottky emitter.

Transmission electron microscopy (TEM) images were obtained using a JEOL 2010 FasTEM at accelerating voltages of 200 kV with an energy dispersive spectroscopy (EDS) system. A

Micromeritics ASAP 2010 instrument was used to measure the surface area using nitrogen adsorption at -196°C. Temperature programmed reduction- mass spectrometry (TPR-MS) analysis was performed with a designed setup and an MKS-UT1 PPT quadrupole mass spectrometer.

[0065] Hydrogenation of CO and C0 ₂

[0066] Hydrogenations of C0 ₂ or CO were performed in a packed bed reactor at temperatures of 120-320°C and a pressure of 13.6 atm. Pre-mixed C0 ₂ /H ₂ /Ar (Gas 1, molar ratio: 3:6: 1) and CO/H ₂ /Ar (Gas 2, 3:6: 1) gas mixtures (Airgas) were used for FT syntheses. Ar was utilized as an internal standard for GC analysis. 0.75 g catalysts were loaded into a packed stainless tube reactor (ID 0.9 cm). The catalysts were reduced in 10% CO (or H ₂ )/He at a pressure of 1 atm and temperatures of 350-450°C with a ramp rate of 2°C/min for 2 h. The reactor was then cooled below 100°C, and the feed gases (30-200 cm ³ /min) were switched to Gas 1 or Gas 2. A tube furnace was heated at a ramp rate of 5°C/min to the reaction temperature (200-360°C).

[0067] In Situ XRD Measurements

[0068] The in situ XRD studies were performed under similar conditions of hydrogenation of CO and C0 ₂ on a XTRA X-ray diffractometer (Cu K _a radiation) equipped with an Anton Parr XRK 900 heater chamber (max. pressure, 10 atm). The same pre-mixed gases for the FT synthesis were used for in situ XRD. The heating rate was 2°C/min; the total pressure was 1 atm for the H ₂ /He reduction and 8.5 atm for the hydrogenation of CO and C0 ₂ ; and the gas flow rate was 40 cm ³ /min. In situ XRD patterns were obtained by step scanning over the angle range 2O=5~80° at an increment of 1.2 min and at an accumulation time of 0.5 s at each point. To calculate the average crystallite size, LaB ₆ was used as an external standard for the instrumental full width at half maximum (FWHM) calibration. XRD of LaB ₆ was operated under identical conditions as that of the OMS-2 supported Fe and Co catalysts.

[0069] Product Analyses

[0070] Gaseous products were analyzed using an online GC (SRI 86 IOC) equipped with two detectors in series (a flame ion detector for hydrocarbons (>1 ppm) and a thermal conductivity detector for CO, CH ₄ , C0 ₂ , and Ar (>300 ppm)) and two columns (molecular sieve 13X and HayeSep D). The initial oven temperature was held at 10°C for 8 min by adding dry ice in the column chamber, and was then raised to 200°C at a ramp rate of 10°C/min and heated at 200°C for 5-20 min. A standard deviation of the GC analysis was about 3-5%.

[0071] Liquid products collected from the condensing tank and solid wax collected at the filter were analyzed by GC-MS. The detection limit is about 1-5 ug/L. The standard deviation of GC-MS was about 2-3%. A Hewlett-Packard GC (HP5890 series II) equipped with a

mass-selective detector (MSD, HP5971 Series) and a DB-5HT capillary polar column (30 m x 0.25 mm). Solid wax products (light yellow) were dissolved in CS ₂ (HPLC, 99.8%>) for GC-MS analyses. Liquid products were diluted in acetonitrile (HPLC, 99%) and ^g/mL CeH ₅ F was added as an internal standard for GC-MS analyses. The temperatures of the injector and GC-MSD interfaces were 270 and 280°C, respectively. For wax (liquid) analyses, the initial oven temperature was held at 50°C for 5 min and then raised to 350°C (320°C) at a rate of 5°C /min. The dwell time was 5 min. Liquid products were also analyzed by 1H NMR (Bruker DRX-400 400MHz) using deuterated dimethyl sulfoxide (DMSO-d6) as an internal standard.

[0072] Catalyst Structure, Composition, and Morphology

[0073] K-OMS-2 nanofibers with a diameter of 20-25 nm and a length of 0.5-5 μιη have a surface area of 65 m ² /g (Fig. 1). Synthetic K-OMS-2 (KMn ₈ 0i ₆ ) has up to 4 % (atomic) of K ⁺ ions in its tunnel structure. Fig. 2 shows the XRD patterns of the K-OMS-2 support and three OMS-2 supported Fe and Co catalysts according to the present disclosure. The labeled peaks show that K-OMS-2 is a pure tetragonal phase (JCPDS No. 29-1020) with the cryptomelane-type structure. Framework doped Fe-K-OMS-2 shows an identical structure to that of K-OMS-2.

Hydrothermal Fe-K-OMS-2 shows low intensity peaks indicating smaller particles than HT

Fe-K-OMS-2. Some unconverted Fe-birnessite existed in CAT C. Its K ⁺ concentration was 5.4%, which is higher than the 4% of K-OMS-2.

[0074] After the IWI of Co(N0 ₃ ) ₂ on K-OMS-2 supports, the XRD patterns of CAT B confirmed that a-Co(N0 ₃ ) ₂ .6H ₂ 0 (JCPDS No. 25-1219) was coated on the K-OMS-2 supports. Co(N0 ₃ ) ₂ was decomposed to Co ₃ 0 ₄ on K-OMS-2 supports after calcination at 450°C. Cu(N0 ₃ ) ₂ was added to Co(N0 ₃ ) ₂ in the IWI process for the preparation of CAT A. Cu(N0 ₃ ) ₂ was

decomposed to CuO after calcination. [0075] Fe-K-OMS-2, Co ₃ 0 ₄ /K-OMS-2, and Co ₃ 0 ₄ /CuO/K-OMS-2 catalyst precursors are not active and must be reduced before FT synthesis. After reduction in 20% H ₂ /He (balance), Co ₃ 0 ₄ was reduced to hexagonal close packed (hep) and face-center cubic (fee) Co nanoparticles; CuO was reduced to fee Cu; and K-OMS-2 was reduced to cubic MnO (JCPDS No. 7-230) (Fig. 4a). The crystallite size of Co nanoparticles along the [101] direction, calculated from the Sherrer Equation, is 42 nm. Typical elemental maps of four reduced catalysts (CAT A, B, C, and D) showed uniform elemental distribution for each Co, Mn, K, and Cu element.

[0076] Fig. 3 shows the TEM image and EDX spectra of the reduced OMS-2 supported Co and Fe catalysts. The TEM image (Fig. 3 A) of CAT B shows the cubic spinel Co ₃ 0 ₄ (JCPDS No.

42-1467) phase and the hexagonal close-packed (hep) Co phase present in the CAT B catalyst. The TEM image (Fig. 3B) of CAT D shows that the distance between the lattice fringes is 2.0 A, which is close to the -spacing (2.03 A) of the (110) planes of a-Fe in the XRD data.

[0077] In Situ XRD Analyses

[0078] Fig. 4a shows that the in situ XRD pattern of the reduced CAT C has a small peak at a 2Θ of 44.6°, which is assigned to a-Fe (body-centered cubic (bec), JCPDS No. 6-696) and other five major peaks (2Θ: 34.9, 40.6, 58.7, 70.2, and 73.6°) were assigned to cubic MnO. When the temperature was increased to 220°C in CO hydrogenation, two broad peaks at 2Θ of 31.2 and 51.6° appeared, indicating KHC0 ₃ (Kalicinite, JCPDS No. 12-292) was formed. With the temperature increased to 320°C for 15 minutes, the Fe ₃ C (Cohenite, JCPDS No. 35-772) phase was observed at 2Θ of 37.8°, 41.4°, and 45.3°. A small Fe ₂ C peak was also seen at a 2Θ of 28.2°. The particle sizes of Fe ₃ C, KHC0 ₃ , and MnO crystals exposed to reactants at different reaction times were calibrated from the Sherrer Equation. After a 24 h reaction, the Fe ₃ C crystals along the [031] direction grew from 8 nm at 220°C to 60 nm at 320°C; KHC0 ₃ crystals along the [-311] direction grew from 5 nm at 120°C to 81 nm at 320°C (see Table 1); the MnO crystallite size decreased from 22 nm at room temperature to 13 nm at 320°C due to the etching of acidic products and KHC0 ₃ . FT-IR data (Fig. 17) of the post-reaction Fe and Co catalysts further confirmed the formation of KHC0 ₃ . The FESEM images (Figs. 1C and D) show the morphology of post-reaction catalysts of CAT C. The Fe ₃ C phase is clearly shown in the black deposit and Fe-K-OMS-2 nanofibers were broken into smaller fibers and particles likely due to etching of acidic products and diffusion of K ⁺ ions.

[0079] The in situ XRD pattern (Fig. 5a) of the reduced CAT A shows hep cobalt (JCPDS No. 5-727) peaks at 2Θ of 41.4, 47.3, and 75.4°, fee cobalt (JCPDS No. 15-806) peaks at 2Θ of 44.2 and 51.6°, and cubic copper peaks (JCPDS No. 4-836) at a 2Θ of 43.1°. The formation of orthorhombic Co ₃ C (JCPDS No. 26-450) was observed in XRD patterns. After the temperature was increased to 320°C for 15 mins, the Co ₃ C peaks were clearly seen at 2Θ of 42.6, 56.6, 71.6 and 76.3°. After a 24 h reaction, the crystallite size of Co ₃ C showed an increase from 23 nm to 39 nm along [103] direction; the crystallite size of hep cobalt decreased from 42 nm to 9 nm; the crystallite size of copper decreased from 86 nm to 75 nm; and the crystallite size of MnO decreased from 40 nm at room temperature to 29 nm at 320°C. KHC0 ₃ crystals along the [-311] direction grew from 2 nm at 120°C to 294 nm at 320°C after a 24 h reaction (Table 1). The increased size of KHC0 ₃ crystals is believed to be due to less water formation for Co catalysts than for Fe catalysts. Fig. IB shows the morphologies of different phases in CAT A. Dark carbon species are shown on the catalyst surface.

[0080] The in situ XRD patterns (Fig. 10) of CAT C in C0 ₂ hydrogenation show that a-Fe gradually transformed to x-Fe ₂ . ₅ C (JCPDS No. 36-1248) with the temperature increasing from room temperature to 320°C. A small decrease in the crystallite size of the MnO support along the [111] direction was found. The in situ XRD patterns (Fig. 11) of C0 ₂ hydrogenation using the OMS-2 supported Co and Cu catalysts show that the crystallite size of hep Co increased from 5 nm at room temperature to 18 nm at 320°C, indicating the sintering of nanoparticles. The crystallite sizes of Cu and MnO have slightly decreased. Compared to large KHC0 ₃ crystals (75-300 nm) formation in CO hydrogenation (Table 1), only small KHC0 ₃ crystals (6-12 nm) were formed in C0 ₂ hydrogenation.

[0081] All publications, articles, papers, patents, patent publications, and other references cited herein are hereby incorporated by reference herein in their entireties for all purposes.

[0082] Although the foregoing description is directed to the preferred embodiments of the present teachings, it is noted that other variations and modifications will be apparent to those skilled in the art, and which may be made without departing from the spirit or scope of the present teachings.

[0083] The examples are presented herein to provide a more complete understanding of the present teachings. The specific techniques, conditions, materials, and reported data set forth to illustrate the principles of the present teachings are exemplary and should not be construed as limiting the scope of the present teachings.

[0084] The foregoing detailed description of the various embodiments of the present teachings has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the present teachings to the precise embodiments disclosed. Many modifications and variations will be apparent to practitioners skilled in this art. The embodiments were chosen and described in order to best explain the principles of the present teachings and their practical application, thereby enabling others skilled in the art to understand the present teachings for various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the present teachings be defined by the following claims and their equivalents.

Table 1. The crystallite sizes of KHCO ₃ with increasing time and temperatures in CO hydrogenation using the OMS-2 supported Fe and CoCu catalysts.

Note: The unit of the crystallite size is nm. NA represents "below the detection limit".

Table 2. Catalytic activity and product distribution of the OMS-2 supported catalysts in C0 ₂ hydrogenation

Note: The reaction conditions are the same as Fig. 6B, and the reaction time conversion and selectivity of the OMS-2 supported Fe catalysts with different K

Note: The conditions are the same as Table 2.

Table 4. Comparisons of aldehydes, aromatics, and alkenes distributions in liquid products of CAT C and C0/AI2O3 reference catalysts

Note: These data are 1H NMR results.