The present disclosure relates to a versatile and flexible, environmentally friendly and economically viable process for converting sour natural gas to sweet natural gas, green hydrogen and carbon disulfide.
Natural gas is classified according to its CO2 and H2S content: sweet natural gas that contains <2% CO2 and <4 ppm H2S used without further treatment and sour natural gas that does not meet the sweet natural gas criteria (Richard W. Baker and Kaaeid Lokhandwala, “Natural Gas Processing with Membranes: An Overview”, Ind. Eng. Chem. Res. 2008, 47, 2109-2121). There are various levels of sour natural gas, as listed in Table A. The very sour natural gas accounts for a relatively high percentage of the total natural gas available. One of the grand challenges in utilization of this large resource is to remove H2S from such highly sour natural gas streams to enable their applications in the energy and chemical sectors. Therefore, there is a high incentive to develop a novel process that removes H2S and CO2 and produces sweet natural gas, hydrogen and carbon disulfide.
Available world sour gas reserves (% of total gas) based on the levels of CO2 and H2S concentration (mol %). Adopted from William Taifan and Jonas Baltrusaitis, “Minireview: direct catalytic conversion of sour natural gas (CH4+H2S+CO2) components to high value chemicals and fuels”, Catal. Sci. Technol., 2017, 7, 2919
The commercial technology currently employed to convert sour to sweet natural gas consists mainly of the acid gas removal unit to selectively strip off H2S and CO2 (amine-based absorption is frequently used) and sulfur recovery unit (SRU) where H2S in the H2S-rich stream reacts by the well-known Claus process to produce sulfur as depicted in
U.S. Pat. No. 10,759,722 relates to hydrogen sulfide methane reformation, showing that sour natural gas can be upgraded in a DHA reactor (dehydroaromatization) to produce liquid aromatic hydrocarbons and CS2. Exemplified feed compositions have relatively low H2S content.
Although hydrogen sulfide methane reformation offers a promising route to produce clean hydrogen while simultaneously removing hydrogen sulfide and eliminating the need for Claus unit, very little has been published on this topic (A. L. Martínez-Salazar, J. A. Melo-Banda, M. A. Coronel-García, Pedro M. García-Vite, Iris Martínez-Salazar, and J. M. Domínguez-Esquivel, “Technoeconomic analysis of hydrogen production via hydrogen sulfide methane reformation”, International Journal of Hydrogen Energy 44 (24), 2019, 12296-12302) with no real leads to commercial processes. Furthermore, a very high H2S/CH4 ratio at feed (>8) is usually used to avoid potential carbon deposits (Cunping Huang and T. Ali, “Liquid hydrogen production via hydrogen sulfide methane reformation”, Journal of power sources 175(1), 2008, 464-472), which will cause catalyst deactivation.
We have now developed an elegant process design that reacts methane with H2S in sour or ultra-sour natural gas (possibly in the presence of CO2) to produce hydrogen, CS2 and sweet natural gas. That is, the feed to the reformer reactor consists of CH4, H2S and possibly CO2; with the aid of selected catalysts capable of advancing both reactions (H2S reforming and dry reforming), and under a set of suitable conditions, sour natural gas with high H2S levels can be converted to sweet natural gas, hydrogen and carbon disulfide, as shown by reactions R1 and R2 below, achieving not less than 25% conversion, e.g., ≥30%, and even ≥40% H2S conversion (40-50%). As to the products obtained, the hydrogen can be further reacted with CO2 to produce green liquid fuels and chemicals (Herskowitz, Mordechay and Hos, Tomy, “Novel, highly efficient eco-friendly processes for converting CO2 or CO-rich streams to liquid fuels and chemicals”, U.S. Pat. No. 10,865,107 (2020)). CS2 is a valuable material that is more desirable than elemental sulfur as a feedstock to produce chemicals.
Hydrogen sulfide methane reformation is shown by reaction R1:
CH4+2H2S→CS2+4H2 ΔH298K=233 KJ/mole R1.
Since CO2 may be a component of the feed, dry reforming may take place in parallel:
CH4+CO2→2CO+2H2 ΔH298K=247 KJ/mole R2.
and reverse water gas shift (RWGS) in series:
H2+CO2→CO+H2O ΔH298K=41 KJ/mole R3.
Since all reactions are reversible and endothermic, the product composition at thermodynamic equilibrium varies significantly over range of temperatures, pressure and feed composition. The equilibrium plot for reforming hydrogen sulfide and carbon dioxide with methane by Gibbs free energy minimization simulation indicates that temperature and pressure affect the equilibrium significantly, as shown in
Accordingly, one aspect of the invention is a process for preparing hydrogen by catalytic conversion of sour natural gas, comprising feeding sour natural gas mixed with one or more H2S recycled streams (which may contain also CH4, H2 and CO2), and optionally fresh CO2 to a reformer reactor packed with a catalyst, for example, a catalyst activated in-situ by sulfidation (selected catalysts are described in detail below).
Feed streams, with compositions suitable to afford hydrogen in industrially acceptable quantities, comprise from 50 to 90 vol % methane, not less than 10 vol % H2S, e.g., for example from 10 to 40 vol % H2S, and 0 to 40 vol % CO2. The process is well suited to convert feed streams with >12 vol % H2S, e.g., >15 vol % H2S, for example, from 15 to 35 vol % H2S.
The catalytic conversion of sour natural gas takes place over the catalyst in the reformer reactor under the following conditions: temperature in the range from 800 to 950° C., e.g., up to 900° C.; WHSVH2S in the range of 0.5-5 h−1 at a total pressure of 1-3 atm.
The effluent from the reactor is passed through a separation system consisting of several units; unreacted feed material, namely, H2S-containing streams, are collected at several points and are recycled to the reformer reactor, whereas CO2 can be produced downstream and may be either directed to the reformer reactor, or even better, used as a feed component in a plant where CO2 and hydrogen-the key product of the process-are converted into liquid hydrocarbons, e.g., as described in U.S. Pat. No. 10,865,107.
Separation of unreacted H2S from the effluent of the reformer reactor, and separation of the effluent into a liquid stream consisting of the CS2 by-product and the (CH4+H2)-containing gas product stream, includes two major steps, i.e., A) membrane separation followed by B) condensation and gas-liquid separation. Reversal of steps is also acceptable, that is, first condensation and gas-liquid separation, followed by membrane separation. The order of steps, either A→B or B→A, affects the management of process streams and recycle structure. But basically, whilst H2S-rich streams generated by separation steps A→B or B→A are returned to the reforming reaction, H2S-lean streams are jointly treated to further minimize H2S level, then recover the products H2 and CH4. Thus, the invention provides a process comprising:
H2S-lean streams generated by the separation methods (A→B or B→A) are jointly treated to recover the products H2 and CH4; the treatment includes removal of residual acidic gases (H2S, CO2; for example, by absorption) to afford an essentially H2S-free gas stream (e.g., not more than 4 ppm H2S), recycling of the acidic gasses to the reforming reaction; optionally reduction of CO level by mixing the essentially H2S-free gas stream with steam under conditions advancing water gas shift reaction; and ultimately, separation of H2 and CH4 from one another by membrane separation.
The description that follows focuses on the A→B order of steps.
One preferred variant of the invention, using an efficient recycle design of the H2S and CO2 streams, is a process for converting sour natural gas to sweet natural gas and producing hydrogen and carbon disulfide by H2S reforming of methane to hydrogen and carbon disulfide, comprising:
More specifically, the process of the invention comprises feeding sour natural gas and four recycle streams (possibly mixed with fresh CO2 as needed) to a preheater then to a reformer packed with a solid catalyst. The projected H2S conversion is at least 25%, e.g., from 35 to 50%, for example, about 40%. The effluent is cooled and compressed and fed to a two-stage membrane unit to separate a stream containing concentrated H2S that is recycled to the reformer; the first retentate stream is cooled in a series of condensers to separate the by-product CS2 in liquid form. One stream of non-condensable component of said retentate stream is fed together with the second retentate stream to an H2S capture unit consisting of an absorption system or a membrane or a combination of the two; the condensable stream is fed to a separator at atmospheric pressure to separate liquid CS2 and second non-condensable stream that is fed back to the reformer; the H2S-rich stream from the H2S capture unit is fed back to the reformer and the hydrogen-rich stream from the unit is fed into one, or two Water Gas Shift (WGS) reactors in series, together with steam to form CO2 and hydrogen.
The WGS reaction takes place in one or more adiabatic fixed-bed reactors packed with a suitable WGS catalyst, for example, Cu/ZnO based low temperature water gas shift catalyst. The effluent is cooled by a condenser, and water is separated by a gas-liquid separator. The conditions in WGS reactor include WHSVco of not less than 1 h−1, preferably not less than 2 h−1. The reaction is carried out at a temperature in the range from 180 to 230 oC at a pressure of not less than 15 atmospheres, e.g., from 20 to 30.
The WGS reactor effluent stream is fed into a membrane that separates the sweet natural gas stream (containing <2% CO2) from the hydrogen stream (containing CO2 and methane); part of the sweet natural gas is combusted with oxygen to supply the heat for the reformer producing CO2 that can optionally be recycled back to the reformer; the remaining sweet natural gas stream is fed into WGS reactor to reduce CO concentration below 4 ppm; the obtained hydrogen stream can be converted in the Blechner Center process (Herskowitz, Mordechay and Hos, Tomy, “Novel, highly efficient eco-friendly processes for converting CO2 or CO-rich streams to liquid fuels and chemicals”, U.S. Pat. No. 10,865,107 (2020)) to produce green liquid fuels and chemicals. CO2 generated in the process and separated from the very sour natural gas is reacted with hydrogen to produce green fuels and chemicals to minimize the carbon footprint of the process.
Turning now to
Part of the sweet natural gas stream (26) is combusted (8) with oxygen [27] to supply the heat for the reformer and produce CO2 that is recycled [28] to the reformer (1) or mixed [30] with stream [19]. The rich-hydrogen stream is used to produce higher hydrocarbons according to the process developed in the Blechner Center described in U.S. Pat. No. 10,865,107. The remaining sweet natural gas stream [21] flows to the WGS reactor, to react with steam [22], producing stream [23] that exits the WGS reactor, and led to a gas-liquid separator, to reduce CO concentration below 4 ppm in the treated natural gas stream [25]. Unreacted water [24] is removed.
By “catalyst activated in-situ by sulfidation” we mean a catalyst which undergoes activation in the reformer reactor; when 0.5-2.0 g of the catalyst are treated at a temperature in the range from 500 to 550° C., e.g., ˜530° C. and atmospheric pressure in a 50-150 mL min−1 flow of 20 vol % H2S/N2 gas for 1 hour, the catalyst transforms into active phase(s) useful in catalyzing the conversion of sour natural gas. Examples of useful catalysts (in their form prior to activation) include:
The invention contemplates the use of catalysts composed of metals (such as
molybdenum) dispersed on the surface of spinel compounds of the formulas (A12+)·(B13+β1B2+β2)2O4, (A12+α1A22+α2)·(B13+β1B23+β2)2O4, (A12+α1A22+α2A32+α3)·(B13+β1B23+β2)2O4, (A12+α1A22+α2A32+α3)·(B13+β1B23+β2B33+β3)2O4 and (A12+α1A22+α2A32+α3A42+α4)·(B13+β1B23+β2B33+β3)2O4. For example, the divalent ion(s) Ai2+ is selected from the group consisting of Ni2+, Co2+, Cu2+ and Zn2+ and the trivalent ion Bj3+ is selected from the group consisting of Fe3+, Cr3+ and Al3+.
For example, Ai2+ is Ni2+ and B13+ is Fe3+. The spinel compounds include: (Ni2+)·(Fe3+β1B23+β2)2O4, in which B23+ is selected from Cr3+ and Al3+, with 0.1≤βj≤0.9; e.g., 0.25≤βj≤0.75;
The spinel of Formula 1 exhibits surface area from, e.g., 10 to 50 m2/g, for example, from 10 to 20 m2/g, measured by the BET method.
Some of the spinel compounds are novel and form a separate aspect of the invention. X-ray diffraction analysis (using CuKα radiation) of samples of a few selected spinel compounds, e.g., Ni(Fe0.5Cr0.5)2O4, (Ni0.5Cu0.5)(Fe0.5Cr0.5)2O4 (Ni0.5Co0.5)(Fe0.5Cr0.5)2O4 and (Ni0.5Zn0.5) (Fe0.5Cr0.5)2O4 shows that the materials consist of a single spinel phase.
Table 1 lists the diffraction angles (2θ) and the [hkl] planes to which they refer.
The spinel of Formula 1 exhibits catalytic activity by itself. However, it was found that the performance of Ni(Co,Zn,Cu)—Fe(Cr,Al)—O spinel catalyst in direct catalytic conversion of sour natural gas is significantly improved after deposition on its surface nanocrystals of MoO3 oxide phase thus forming a catalyst by Formula 2:
xMoO3/(A12+)·(B13+β1B23+β2)2O4;
xMoO3/(A12+α1A22+α2)·(B13+β1B23+β2)2O4;
xMoO3/(A12+α1A22+α2A32+α3)·(B13+β1B23+β2B33+β3)2O4; and
xMoO
3/(A12+α1A22+α2A32+α3A42+α4)·(B13+β1B23+β2B33+β3)2O4.
The spinel of Formula 1 is prepared by the sol-gel method, assisted by a complexing agent. For example, the method of synthesis of the compound (A12+)19 (B13+β11B23+β2)2O4, such as Ni2+(Fe3+β1Cr3+β2)2O4, comprises dissolution in water of the A12+, B13+ and B23+ (e.g., Ni2+, Fe3+ and Cr3+) salt precursors, e.g., nitrate salts, to form Ni2+, Fe3+and Cr3+solution (this could also be done by combining separate aqueous solutions of the individual salts), adding a complexing agent, such as citric acid, to the mixed salt solution, heating the mixture to 60-90° C. until the gel is formed. The spinel powder is recovered by drying the material overnight at ˜110° C., followed by two calcination cycles, first in air at ˜200° C. (e.g., at 10° C./min) for 1-3 h and then at 500-700° C. (5° C./min) for 4 h.
The catalyst of Formula 2 is prepared by loading the catalytically active metal component, e.g., molybdenum, onto the spinel surface by impregnation. That is, the spinel powder is impregnated with an aqueous solution of Mo precursor salt (suitable Mo sources were described above), followed by drying (e., first in air then in the oven) and calcination (e.g., >500° C.). Illustrative conditions are provided in the experimental section below.
Another aspect of the invention is an apparatus suitable for converting sour natural gas to sweet natural gas, hydrogen and carbon disulfide, comprising:
Conventional wide-angle XRD patterns were measured with a Panalytical Empyrean Powder Diffractometer equipped with position sensitive detector X'Celerator fitted with a graphite monochromator, at 40 kV and 30 mA and analyzed using software developed by Crystal Logic. The phase identification was performed by using an SBDE ZDS computer search/match program coupled with the International Center for Diffraction Data (ICDD). BET was measured by NOVA 3200e Quantachrome adsorption analyzer.
Mo/γ-Al2O3 catalyst was prepared by incipient wetness impregnation. 8 g of γ-Al2O3 support (NORTON, SA6175, ⅛″ extrudates, 230-290 m2/g) were calcined at 500° C. for 2 h prior to impregnation. After calcination the support was held under vacuum for 1 h and further impregnated with solution of 2.454 g (NH4)6Mo7O24·4H2O in 4.25 g H2O and 1.5 ml NH4OH 25%. The material was dried at room temperature for 24 h. Next, the catalyst was dried at 120° C. for 12 h and calcined at 550° C. for 4 h (3° C./min).
K-Mo/γ-Al2O3 catalyst was prepared by incipient wetness impregnation in two steps. At first step Mo/γ-Al2O3 material was prepared according to previous synthesis (Preparation 1). After that, the 10.67 g of obtained material was held under vacuum for 1 h and further impregnated with solution of 0.988 g K2CO3 in 5.85 g H2O. The catalyst was dried at 120° C. overnight and calcined at 500° C. for 2 h (5° C./min). BET surface area was 166 m2/g.
Ni(Fe0.5Cr0.5)2O4 material was synthesized by the sol-gel method. The metal salt precursors were dissolved separately in 10 ml H2O each: 8.591 g Cr (NO3)3·9H20; 6.570 g Ni(NO3)2·6H2O; 9.128 g Fe(NO3)3·9H2O. Once fully dissolved, the metal precursors were mixed and 32.566 g of citric acid (complexant) was added to the solution (the ratio of moles complexant to total moles of metal ions was 2.5). Then the mixed solution was heated to 80° C. on a hot plate until the gel was formed. The material was dried overnight at 110° C., calcined in air at 200° C. (10° C./min) for 2 h and then at 700° C. (5° C./min) for 4 h. X-ray powder diffraction is shown in
4.61 g of Ni(Fe0.5Cr0.5)2O4 material was held under vacuum for 1 h and impregnated with solution of 0.298 g (NH4)6Mo7O24·4H2O in 3.24 g H2O and 0.18 ml NH4OH 25%. The material was dried at room temperature for 24 h. After that the catalyst was dried at 120° C. for 12 h and calcined at 550° C. for 4 h (3° C./min).
(Ni0.5Cu0.5)(Fe0.5Cr0.5)2O4 material was synthesized by the sol-gel method. The metal salt precursors were dissolved separately in 10 ml H2O each: 8.591 g Cr(NO3)3·90H2O; 3.286 g Ni(NO3)2·6H2O; 9.128 g Fe(NO3)3·9H2O; 2.730 g Cu(NO3)2·3H2O. Once fully dissolved, the metal precursors were mixed and 32.566 g of citric acid (complexant) was added to the solution (the ratio of moles complexant to total moles of metal ions was 2.5). Then the mixed solution was heated to 80° C. on a hot plate until the gel was formed. The material was dried overnight at 110° C., calcined in air at 200° C. (10° C./min) for 2 h and then at 700° C. (5° C./min) for 4 h. X-ray powder diffraction is shown in
4.66 g of (Ni0.5Cu0.5)(Fe0.5Cr0.5)2O4 material was held under vacuum for 1 h and impregnated with solution of 0.286 g (NH4)6Mo7O24·4H2O in 2.88 g H2O and 0.18 ml NH4OH 25%. The material was dried at room temperature for 24 h. After that the catalyst was dried at 120° C. for 12 h and calcined at 550° C. for 4 h (3° C./min).
(Ni0.5Co0.5)(Fe0.5Cr0.5)2O4 material was synthesized by the sol-gel method. The metal salt precursors were dissolved separately in 10 ml H2O each: 8.591 g Cr(NO3)3·9H2O; 3.286 g Ni(NO3)2·6H2O; 9.128 g Fe(NO3)3·9H2O; 3.288 g Co(NO3)2·6H2O. Once fully dissolved, the metal precursors were mixed and 32.566 g of citric acid (complexant) was added to the solution (the ratio of moles complexant to total moles of metal ions was 2.5). Then the mixed solution was heated to 80° C. on a hot plate until the gel was formed. The material was dried overnight at 110° C., calcined in air at 200° C. (10° C./min) for 2 h and then at 700° C. (5° C./min) for 4 h. X-ray powder diffraction is shown in
3.58 g of (Ni0.5Co0.5)(Fe0.5Cr0.5)2O4 material was held under vacuum for 1 h and impregnated with solution of 0.219 g (NH4)6Mo7O24·4H2O in 1.58 g H20 and 0.11 ml NH4OH 25%. The material was dried at room temperature for 24 h. After that the catalyst was dried at 120° C. for 12 h and calcined at 550° C. for 4 h (3° C./min).
(Ni0.5Zn0.5)(Fe0.5Cr0.5)2O4 material was synthesized by the sol-gel method. The metal salt precursors were dissolved separately in 10 ml H2O each: 8.591 g Cr(NO3)3·9H2O; 3.286 g Ni(NO3)2·6H2O; 9.128 g Fe(NO3)3·9H2O; 2.480 g Zn(CH3COO)2·2H2O. Once fully dissolved, the metal precursors were mixed and 32.566 g of citric acid (complexant) was added to the solution (the ratio of moles complexant to total moles of metal ions was 2.5). Then the mixed solution was heated to 80° C. on a hot plate until the gel was formed. The material was dried overnight at 110° C., calcined in air at 200° C. (10° C./min) for 2 h and then at 700° C. (5° C./min) for 4 h. X-ray powder diffraction is shown in
4.79 g of (Ni0.5Zn0.5)(Fe0.5Cr0.5)2O4 material was held under vacuum for 1 h and impregnated with solution of 0.293 g (NH4)6Mo7O24·4H2O in 4.23 g H2O and 0.23 ml NH4OH 25%. The material was dried at room temperature for 24 h. After that the catalyst was dried at 120° C. for 12 h and calcined at 550° C. for 4 h (3° C./min).
A schematic description of the experimental set-up used for running the H2S reforming of methane is depicted in
Methane was contacted with H2S and CO2 by passing a mixture of CH4, H2S and CO2 streams (indicated by numerals [101], [102] and [103], respectively) at a molar ratio H2S/CH4 and CO2/CH4 of 0.6 and 0.3, respectively, through a tubular reactor (11) (11 mm ID, 600 mm long) made of alumina, packed with 1.5 gram of the catalyst powder of Preparation 1 (Mo/γ-Al2O3) or 0.75 gram of the catalyst powder of Preparation 2 (K-Mo/γ-Al2O3) and 4 gram of quartz powder and heated up to 900° C. at a total pressure of 1 atm (Examples 1 and 2, respectively). All gaseous reactants are fed via line [106] to the reactor (11).
The reaction products are cooled down to 150° C. with the aid of an electric heater (12) to separate and capture sulfur residues, which may form in the H2S decomposition reaction. With the aid of a cooler (13), the gaseous products [108] were cooled down to 5° C. The gaseous products [109] flow in line [110] to GC analyzer or to an absorption column (14) containing paraffinic solvent to absorb most of the CS2 produced in the reformer (11). The effluent gas stream [112] containing H2S, CS2, CH4, CO2, CO and H2 is fed into two scrubber vessels in series (15) containing 2 L of sodium hydroxide solution, to remove H2S, CO2 and CS2 effectively.
The exhaust components [113] flowing in line [114] were analyzed in online Agilent 7890A Series Gas Chromatograph (GC) equipped with 7 columns and 5 automatic valves using helium as a carrier gas. The flow rate was measured by Alicat mass flow meter (FI).
In the tables below, the capital letters X and S stand for conversion and selectivity, respectively. The selectivity towards H2S reforming reaction was calculated as SH2S_reforming=[(XH2S(reforming)*λ/2)/XCH4], where λ is the H2S/CH4 molar ratio at feed.
The reaction of H2S (H2S reforming) and CO2 (dry reforming) with CH4 in reactor (11) to produce H2, CO and CS2 was run under specific conditions at close to equilibrium conversions: Temperature of 900° C., total pressure of 1 atm, H2S/CH4=0.6 mol/mol and CO2/CH4=0.3 mol/mol. Results are shown in Table 2.
This experiment was conducted in an experimental unit with a similar design as in Examples 1-2, schematically described in
CO2 [103] and H2S [102] streams were contacted with CH4 [101] to form a feed mixture [106] containing H2S/CH4 and CO2/CH4 in a molar ratio of 0.4-0.6 and 0.08-0.30, respectively. CO2 was partially reacting with CH4 and H2 in the dry reforming and RWGS reactions, respectively.
The reaction was run under the following specific conditions:
This experiment was conducted in an experimental unit with a similar design as in Examples 1-2 schematically described in
H2S [102], CO2 [103] and H2 [104] streams were contacted with CH4 [101] to evaluate the performance of the projected gas composition with hydrogen at feed mixture [106].
The reaction was run under the following specific conditions:
This experiment was conducted in an experimental unit with a similar design as in Examples 1-2 schematically described in
H2S [102] stream was contacted with CH4 [101] to form a feed mixture [106] containing H2S/CH4 in a molar ratio of 0.6, without CO2 in feed.
The reaction was run under the following specific conditions:
While the present disclosure has been illustrated and described with respect to a particular embodiment thereof, it should be appreciated by those of ordinary skill in the art that various modifications to this disclosure may be made without departing from the spirit and scope of the present disclosure.
This application is a National Stage application of International Patent Application No. PCT/IL2023/050185, filed on Feb. 23, 2023, which claims priority to U.S. Provisional Patent Application No. 63/314,488, filed on Feb. 28, 2022, each of which is hereby incorporated by reference in its entirety.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/IL2023/050185 | 2/23/2023 | WO |
Number | Date | Country | |
---|---|---|---|
63314488 | Feb 2022 | US |