Improved bilayer membrane for membrane reactor

Abstract

The invention relates to an improved bilayer membrane for a membrane reactor. According to one aspect, a hydrogen permeable membrane device is provided that includes a porous support layer, a hydrogen selective Pd-Ag membrane, and a porous protective layer. According to one aspect, it is possible to provide a hydrogen permeable membrane having high permeability and permeability selectivity as well as wear resistance to particles.

Description

[0001] This application is a divisional application of the Chinese Patent Application No. 201780085814.1, filed on December 7, 2017, with the title of "Improved Dual Layer Membrane for Membrane Reactor". TECHNICAL FIELD

[0002] The present invention relates generally to the production of hydrogen. More specifically, the present invention relates to a dual layer membrane having a Pd thin film covered by a nano-porous ceramic protective layer. BACKGROUND

[0003] It is well known that Pd-based thin films are used as hydrogen separation devices with high purity hydrogen as the product. Over the past few decades, a lot of research work has been spent to produce membranes with high hydrogen flux, large permeation selectivity and low price. Currently, the production of ultra-pure hydrogen for small scale applications is being investigated by steam reforming of methane (or other feedstocks such as ethanol and methanol) followed by separation of hydrogen using Pd-based membranes (or other technologies). Integrating the membrane into the reactor to produce hydrogen and separate it in situ, enables a reduction in the number of process units required. In addition, this process integration results in a number of other positive aspects, such as an increase in the conversion of the feedstock to hydrogen due to the continuous removal of one of the products, causing a shift in the equilibrium according to the Le Châtelier principle.

[0004] Two types of membrane reactor concepts have been mainly investigated: packed bed and fluidized bed. The performance of packed bed membrane reactors is severely impaired by hot (cold) spots due to low heat transfer rates and concentration polarization, especially for membranes with high permeation selectivity, such as Pd supported membranes. In fluidized bed membrane reactors (FBMR), the intense movement and circulation of the catalyst within the reactor increases the mass and heat transfer rates. The industrial development of membrane assisted fluidized bed membrane reactors is affected by the hydrogen production capacity and the amount of membrane required, which depends on the thickness of the selective layer. Self-supported membranes require a large amount of membrane, while thin film supported membranes appear to be the most attractive option. The thinner the layer, the greater the permeated H2 flux, and therefore, the smaller the membrane area required (thus having a double benefit in terms of Pd cost). However, thin layers (<4-5 pm) are not usually completely dense and there is usually some small porosity through which other gases can pass, thus reducing the purity of the permeated hydrogen. Even if the membrane presents a high permeation selectivity, once it is integrated into a fluidized bed membrane reactor, it must resist the erosion by the fluidized catalysts wash.

[0005] Palladium-based membranes are well known for their use in hydrogen separation devices that produce high-purity hydrogen as a product. Over the past few decades, significant research has been conducted to develop membranes with high hydrogen flux, high permeation selectivity, and low cost. Currently, ultrapure hydrogen production for small-scale applications is being investigated through steam reforming of methane (or other feedstocks such as ethanol and methanol) followed by hydrogen separation using Pd-based membranes (or other technologies). Integrating membranes into the reactor to generate and separate hydrogen in situ reduces the number of process units required. Furthermore, this process integration yields several other positive aspects, particularly the increased conversion of feedstock to hydrogen due to the continuous removal of one product, thus shifting the equilibrium according to Le Châtelier's principle.

[0006] Two main types of membrane reactor concepts were investigated: packed bed and fluidized bed. The performance of packed bed membrane reactors is severely compromised by hot (cold) points due to low heat transfer rates and concentration polarization, particularly for membranes with high permeation selectivity, such as Pd-supported membranes. In fluidized bed membrane reactors (FBMRs), the vigorous movement and circulation of the catalyst within the reactor enhances both mass and heat transfer rates. The industrial development of membrane-assisted fluidized bed membrane reactors is influenced by hydrogen production capacity and the required number of membranes, which depends on the thickness of the selective layer. Self-supported membranes require a large number of membranes, making thin-film supported membranes appear to be the most attractive option. The thinner the layer, the greater the permeated H2 flux, and therefore the smaller the required membrane area (thus offering a double benefit from the Pd cost perspective). However, thin layers (<4-5 μm) are typically not completely dense and often contain pores through which other gases can pass, reducing the purity of the permeated hydrogen. Even if the membrane exhibits high permeation selectivity, once integrated into a fluidized bed membrane reactor, it must resist the erosion caused by the fluidizing catalyst.

[0007] Compared to ceramic-supported membranes, metal-supported membranes offer significantly higher mechanical strength and are relatively easy to integrate into reactors. However, their long-term use in fluidized bed reactors at high temperatures (>400°C) is limited because defects may occur due to atomic movement within the selective layer or due to wear of the fluidizing particles.

[0008] What is needed is a hydrogen permeation membrane with high permeability and permeation selectivity, as well as resistance to particle abrasion, to be integrated into a fluidized bed membrane reactor. Summary of the Invention

[0009] To address the needs of the art, a hydrogen permeation membrane device is provided, the device comprising a hydrogen-selective Pd thin film or Pd alloy thin film and a porous protective layer deposited on a selective thin film comprising: zirconium oxide, yttrium-stabilized zirconium oxide (YSZ), γ / Al2O3 and / or YSZ-γ / Al2O3, silicon dioxide, titanium dioxide, magnesium oxide, cerium dioxide, nitrides and silicon carbide.

[0010] According to one aspect of the invention, the pore size of the porous ceramic protective layer ranges from 1 nm to 400 nm, or from 1 nm to 50 nm, or from 2 nm to 20 nm.

[0011] In another aspect, the present invention also includes a porous support layer. Here, the porous support layer comprises a metal, ceramic, or a combination of metal and ceramic. Furthermore, the ceramic porous support layer comprises a material including oxides, nitrides, and / or carbides. According to this embodiment, the porous support layer comprises a pore size ranging from 2 nm to 50,000 nm.

[0012] In another aspect of the invention, the thickness of the hydrogen-selective Pd layer or Pd alloy layer ranges from 0.5 μm to 150 μm.

[0013] According to another aspect of the invention, the thickness of the porous ceramic protective layer ranges from 0.5 μm to 150 μm.

[0014] In one aspect of the invention, the porous ceramic layer becomes catalytic by introducing a catalyst into the pores, the catalyst being, for example, ZnO, Pt, Ni, Rh, Ru, Ag, Pd or a mixture of metals, such as Pt-Rh, Pd-Ag, Pt-Pd.

[0015] In another aspect, the invention includes a housing encapsulating a hydrogen permeation membrane, wherein the housing comprises a retention end and a permeation end. According to one aspect of this embodiment, the housing comprises a material including stainless steel, carbides, ceramics, and / or Hastelloy corrosion-resistant nickel-based alloys. Here, the housing is disposed within a hydrogen production reaction vessel, wherein the generated hydrogen gas flows from the retention end of the housing to the permeation end of the housing. In another aspect, the housing may be tubular, microchannel, planar, and / or parallel plate in shape. Attached Figure Description

[0016] Figures 1A-1B (1A) A PID for osmosis measurement apparatus and (1B) a reactor vessel with a double membrane are shown according to an embodiment of the present invention.

[0017] Figures 2A-2EThe images show (2A) a bilayer system having a selective (Pd-based) membrane and a porous protective layer according to an embodiment of the present invention, (2B) further comprising a porous inorganic support under the selective membrane, (2C) a protective layer consisting of two sublayers, and SEM cross-sectional images of the bilayer membranes DS-2 (2D) and DS-3 (2E).

[0018] Figure 3 The diagram illustrates the hydrogen flux as a function of hydrogen pressure difference, measured at 300°C to 500°C after activation in air at 400°C, according to an embodiment of the invention.

[0019] Figure 4 The diagram shows the Arenius diagram based on H2 permeability after activation in air at 400°C according to an embodiment of the present invention.

[0020] Figure 5 The H2 / N2 permeation selectivity and H2 permeability of the DS-2 membrane in a temperature range of 300-500°C and a pressure difference of 1 bar, according to an embodiment of the present invention, are shown.

[0021] Figure 6 Table 2, “Permeability of Pd-based support membranes described in the literature,” is shown according to an embodiment of the present invention.

[0022] Figure 7 The hydrogen and nitrogen permeability of membrane DS-2 and the desired permeation selectivity are shown during long-term testing. Glass beads were absent for the first 250 hours. The remaining tests were conducted in the presence of particles in a bubbling fluidized state. According to an embodiment of the invention, the temperature, pressure, and U / Umf profiles during the tests are shown in the figure above.

[0023] Figures 8A-8D The driving forces of the hydrogen transmembrane flux of DS-2 at 400 °C relative to the calculated n value (0.783) of the binary mixture are shown in the case of H2 volume purity of 50 to 100%: (8A) H2-N2, (8B) H2-N2 using glass beads. 2、 (8C) H2-CO2 and (8D) H2-CO using glass beads 2。 Total feed flow rate according to an embodiment of the present invention: 10 L / min -1 (minute -1 ).

[0024] Figures 9A-9CThe H2 transmembrane flux of the binary mixture DS-2 at 400°C is shown under different total flow rates with and without glass beads: (9A) an H2-N2 mixture having 50 vol% H2, (9B) an H2-N2 mixture having 90 vol% H2, and (9C) an H2-CO2 mixture having 50 vol% and 90 vol% H2, according to embodiments of the present invention.

[0025] Figure 10 The long-term test DS-1 film with 4 wt% silver at the selective layer is shown. Rh supported on the promoted alumina particles was introduced after 400 hours. Temperature and pressure profiles are shown above. According to the invention, U / Umf was kept constant during test (3).

[0026] Figure 11 Long-term testing of the DS-3 film at 550°C is shown, with 4% by weight silver at the selective layer. According to the invention, Rh is introduced at the start of the test, supported on promoted alumina particles.

[0027] Figure 12 The present invention demonstrates the hydrogen permeation selectivity and ideal permeation selectivity of M33 and M15-DS under a pressure difference of 4 bar in the presence of an Rh-based catalyst under bubbling fluidized conditions.

[0028] Figure 13 The normalized H2 flux of the conventional PdAg membrane and two bilayer membranes at 400°C is shown. According to the invention, the particle sizes used for each test are summarized in Table 4. Detailed Implementation

[0029] This article provides hydrogen permeation membranes with high permeability and selectivity, as well as resistance to particle abrasion, for integration into fluidized bed membrane reactors. According to one embodiment of the invention, the membrane is produced by depositing a protective layer on top of a very thin selective Pd substrate. In another embodiment, a Pd-Ag layer is prepared by simultaneous ELP (electro-free plating), and the protective layer is prepared by dip coating. For embodiments with a metal porous support, a ceramic interdiffusion barrier layer is deposited by dip coating prior to depositing the hydrogen selective layer.

[0030] According to one embodiment, the present invention includes a hydrogen permeation membrane device having a porous substrate, a palladium core layer deposited on the porous substrate, a Pd-Ag layer deposited on the palladium core layer, and a mesoporous ceramic layer deposited on the Pd-Ag layer, wherein the palladium core of the palladium core layer is deposited in the pores of the porous substrate.

[0031] The permeation properties of the membrane were investigated using single-gas and binary-gas mixture tests. According to the invention, the ceramic-supported bilayer membrane exhibits high hydrogen permeability and H2 / N2 permeation selectivity. Furthermore, during long-term testing, the membrane demonstrated tolerance to two types of particles (glass beads and Rh on promoted alumina) under fluidized conditions. Additionally, interaction with a TiO2-based catalyst was demonstrated under fluidized conditions, compared to conventional membranes without a protective layer.

[0032] The conventional metal-supported membrane and the bilayer (DS) membrane of this invention have been tested under fluidized conditions, and their performance has been compared to study the long-term stability of the membranes. It was found that in the temperature range of 400-500°C, the initial ideal H2 / N2 permeation selectivity of the conventional supported membrane in the presence of a fluidized catalyst decreased significantly, while the DS membrane of this invention maintained a much higher selectivity than under the same conditions.

[0033] Now, regarding the preparation of the DS membrane, according to one embodiment, a porous tubular substrate (10 / 4 mm, outer diameter (od) / inner diameter (id)) made of α-Al₂O₃ with an asymmetric geometry is provided by Rauschert Kloster Veilsdorf and used as a porous carrier with a top layer having a pore size of 100 nm. Following the previously described method, a glass sealant (ASF-1761, Asahi Glass Co.) is used to connect the porous substrate and the dense ceramic tube.

[0034] As previously described, prior to the co-deposition of the Pd-Ag layer, the porous matrix was activated with palladium nuclei via the following process: the support was immersed in a chloroform solution of palladium acetate, followed by reduction with hydrazine. The tubular support containing palladium seeds was then immersed in a plating bath to co-deposit the Pd-Ag layer. The composition of the bath used in this example is summarized in Table 1. Co-deposition was performed at 64°C for 1 hour. Forty-five minutes after deposition, a vacuum was applied from the inside of the tube to seal the pores of the support.

[0035] Table 1. Chemical composition of the electroless plating bath:

[0036]

[0037] Following the deposition of the Pd-Ag layer, the sample was washed with distilled water, dried overnight, and annealed at 550 °C for 4 hours in a reducing atmosphere (10 vol% H2 / 90 vol% N2). Nitrogen was fed during the heating (3 °C / min) and cooling steps. Finally, a thin (<1 μm) mesoporous ceramic layer (50% YSZ / 50% γ-Al2O3) was deposited on top of the Pd-Ag layer using a dip-coating technique.

[0038] In another embodiment, a metal-supported film is prepared by coating a ceramic diffusion barrier layer onto a porous carrier of Hastelloy corrosion-resistant nickel-based alloy X (0.2 μm medium grade). A hydrogen-selective layer (4-5 µm) comprising Pd and Ag is deposited via electroless plating. A porous ceramic layer (protective layer) is deposited on the DS film by dip-coating followed by calcination at 550 °C.

[0039] The physicochemical characterization of the DS films was then examined, with cross-sections analyzed using scanning electron microscopy (FEI Quanta 250 FEG). At least four measurements were performed for each film to determine the layer thickness. The Pd-Ag composition in the selective layer was calculated by measuring the concentrations of silver and palladium in the plating bath before and after layer deposition using ICP-OES (VarianVista MPX inductively coupled plasma optical emission spectrometer). The pore size and surface area of ​​the protective layer were characterized using BET.

[0040] For permeation measurements, one end of the ceramic-supported membrane is connected to a dense metal tube, and the other end is sealed with a metal tap using graphite ferules. A torque wrench is used to tighten the connector onto the membrane at 7 Nm. The total length of the membrane with the connector is 150 mm. Once the seal is tightened, a leak test is performed by feeding helium gas from inside the membrane while immersing it in ethanol. No air bubbles were detected from the seal, indicating that it was properly tightened. Finally, the membrane is dried to remove the ethanol and integrated into the reactor. The permeation apparatus (single gas test and gas mixture) for gas permeation measurements is shown. Figure 1A The shell-and-tube module is made of stainless steel with an inner diameter of 42.7 mm. The gas pressure on the shell side is controlled by a back pressure controller connected to the retention side. The permeate side of the membrane is maintained at atmospheric pressure. No purge gas was applied during the experiment; the hydrogen flux was measured using an automatic mass flow meter (Defender 220), and nitrogen flux was measured using an automatic soap membrane flow meter (Horiba, model VP1) with a flow rate > 0.2 mL / min and another soap bubble flow meter with a flow rate < 0.2 mL / min. Single-gas tests were performed with nitrogen and hydrogen at temperatures of 300–500 °C. Mixed-gas permeate tests using a mixture of H₂-N₂ and H₂-CO₂ were performed at 400 °C. The invention can also be configured to separate / purify hydrogen from the gas mixture.

[0041] Figure 1B This diagram shows a schematic cross-sectional view of the reactor vessel of the present invention. A feed gas mixture is distributed along the reactor vessel, and the gas mixture reacts with the hydrogen-producing catalyst bed. Pure hydrogen gas passes through a hydrogen-selective composite membrane and is collected.

[0042] Metal-supported membranes were simultaneously integrated in a single reactor to evaluate and compare the performance of two membranes under the same conditions.

[0043] Figures 2A-2E Some exemplary embodiments of the present invention are shown, wherein Figure 2A In one aspect, a bilayer system comprising a selective membrane and a porous protective layer is shown, in which the protective layer is deposited on defects in the selective membrane to improve the membrane's selectivity. Here, the selective membrane is a dense (e.g., palladium or its alloys, perovskite) or microporous (e.g., carbon) selective membrane. According to another embodiment, the selective membrane is any inorganic hydrogen-selective membrane material. In yet another embodiment, the membrane is as follows: Figure 2A Self-supporting as shown, or as Figure 2B The support shown is on a porous inorganic carrier. Figure 2B The porous inorganic support is shown. Furthermore, the porous protective layer consists of a single material or a mixture of materials, such as metal oxides (e.g., alumina, ZrO2, YSZ, silicon dioxide, titanium dioxide, cerium dioxide, magnesium oxide), nitrides, and / or silicon carbide.

[0044] In other embodiments of the invention, the protective layer consists of one or more layers. Figure 2C The protective layer consists of two sublayers. In one embodiment, the thickness of the protective layer is from 0.2 μm to 6 μm, more preferably from 500 nm to 2 μm, or from 0.5 μm to 150 μm. The porous protective layer can be coated onto the film using various techniques, including but not limited to sol-gel, dip coating, wet powder spraying, atomic plasma spraying, spraying, sputtering (PVD-magnetron sputtering), or chemical vapor deposition (CVD).

[0045] The thickness of the selective layer and the protective layer are determined by DS-2 ( Figure 2D ) membrane and DS-3 ( Figure 2E The cross-sectional SEM images of the membranes were determined. The thicknesses of the selective layers in DS-2 and DS3 were measured to be ~1 μm (±0.26) and ~1.8 (±0.23) µm, respectively. The thickness of the selective layer of the protective layer in DS-2 was measured to be ~0.5 µm (±0.06), while in the case of DS-3, the protective layer thickness was slightly thicker, at ~0.67 µm (±0.10). EDS plots of the cross-sections of the DS-3 membranes for palladium, silver, aluminum, and oxygen were obtained. These plots confirmed that the Pd-Ag selective layer was disposed between the alumina support and the ceramic protective layer. The surface area of ​​the protective layer (300 m²) was measured. 2 g -1 The pore size (1-5 nm) and pore size were characterized by BET.

[0046] The DS membrane can have any shape or size, but is preferably tubular or planar. Furthermore, a protective layer is constructed to protect against mechanical damage, for example, under fluidized bed conditions.

[0047] In another embodiment, the invention is configured to produce hydrogen by protecting a catalyst bed (fluidized or packed) containing a suitable catalyst in a reactor. Here, conventional embodiments include introducing a mixture of carbonaceous gases (methanol, ethanol, methane, or propane) and a mixture of vapors (and / or oxygen) into a reactor vessel containing the catalyst. Furthermore, conventional embodiments include separating hydrogen from the mixture using a hydrogen-selective membrane.

[0048] In another embodiment, the protective layer is configured to prevent chemical reactions between the catalyst and the selective membrane in a reactor involved in hydrogen production. For example, a catalyst containing ZnO or TiO2 can react with a Pd membrane, reducing hydrogen permeation performance.

[0049] In other embodiments, the protective layer is configured to eliminate or reduce carbon formation on the selective membrane surface, for example, by depositing carbon in reactions containing organic gases (e.g., propane dehydrogenation using a Pd membrane). The protective layer is further configured to reduce hydrogen permeation or act as a barrier against the reaction of H₂S with the hydrogen-selective membrane, which can damage the membrane. The protective layer can be a layer of material that reacts with H₂S, preventing the gas from reaching the selective membrane. On the other hand, the protective layer is functionalized to function as a catalytic membrane reactor. For example, hydrogen is generated by introducing a catalyst into the pores, and then the generated hydrogen is passed through the selective membrane. Furthermore, the protective layer is functionalized by introducing an active compound into the pores, which can interact with substances that may damage the selective membrane.

[0050] Now let's examine the membrane permeation properties in an empty reactor. The equation describing the hydrogen flux through the Pd-based membrane, written using Siefer's law, is as follows:

[0051]

[0052] Where J is the hydrogen flux, Pe is the permeability, t is the thickness of the selective layer, and p H2 These are the partial pressures of hydrogen on the permeation side and the retention side, respectively, and n is the pressure exponent.

[0053] exist Figure 3 The figure shows the hydrogen flux through the DS-membrane at different temperatures (300-500℃) and different transmembrane pressures. It was found that the hydrogen flux (J) and [other parameters] were [higher / lower] relative to the pressure exponent n of 0.783 (after activation in air at 400℃ for 2 minutes). The best fit. For Pd-based films, the pressure exponent should be equal to 0.5 if the rate-limiting step diffuses through most of the palladium. Since the measured n value deviates from 0.5, this may indicate the effect of viscous flow and / or Knudsen diffusion in the porous support (100 nm pore size) within the mesoporous protective layer (1-5 nm pore size) on the overall hydrogen permeation rate; if both were rate-limiting, the n value would be consistent. Furthermore, the rate-limiting step in the film is a surface reaction, not bulk diffusion. Therefore, the n value is greater than 0.5.

[0054] Flux measured in the temperature range of 300–500 °C is used to estimate the activation energy of hydrogen permeation through the membrane. According to the Arrhenius relation provided below, permeability depends on temperature.

[0055]

[0056] Where Q is the membrane permeability, Q o Here, Ea is the activation energy, R is the universal gas constant, and T is the absolute temperature. One of the assumptions of this equation is that the pressure exponent n does not change with temperature.

[0057] By plotting the natural logarithm of the calculated permeability as a function of 1 / RT (see...) Figure 4 The activation energy was determined to be 5.11 kJ / mol. -1 The activation energy is a lumped value, which includes permeability through the mesoporous protective layer, the dense metal layer, and the support. This value is similar to that recorded by Melendez et al.

[10] , who recorded activation energies of 5.47 to 7.81 kJ mol for films with similar silver content (5.7 to 9 wt%) and the same selective layer thickness (0.46 to 1.29 μm). -1 Many parameters can affect the activation energy, such as the thickness of the selective layer, the grain microstructure of the layer, and the preparation method.

[0058] Hydrogen and nitrogen fluxes measured during single-gas tests were used to calculate the membrane's ideal H2 / N2 permeation selectivity. Figure 5 The data shows hydrogen permeability and H2 / N2 permeability selectivity at different temperatures (300-500℃) and pressure differences of 1 bar. This is likely due to permeation through a mesoporous protective layer lacking a Pd-Ag layer. Nitrogen leakage at 300℃ and 500℃ under a pressure difference of 1 bar was 2.10 g / L. -10 mol m -2 s -1 Pa -1 and 1.7∙10 -10 mol m -2 s -1 Pa -1This demonstrates the Knudsen osmosis mechanism.

[0059] exist Figure 6 Table 2 lists the permeation properties of the Pd-based supported membranes, which are thinner than 3 μm. It is recorded that a 1.4 µm thick Pd... 77 Ag 23 The layer is supported in a microchannel reactor, and the hydrogen flux is 22.5 × 10⁻⁶ at 300 °C and a pressure difference of 1 bar. -6 mol m -2 s -1 Pa -1 At the start of the test, the selectivity of the membrane was 5700. However, after 7 days of testing, this value dropped to 390. Regarding tubular Pd-based supported membranes, ~1.9–3.9 μm thick Pd-based membranes prepared by PVD-MS and transferred to PSS tubes are also documented. 77 Ag 23 The maximum hydrogen flux of the membrane. At 400°C and a pressure difference of 1 bar, the H2 flux is 14.9 × 10⁻⁶. -6 mol m -2 s -1 Pa -1 The selectivity is 7700. This is for a very thin (0.49 μm) Pd substrate supported on a porous ceramic substrate. 91 Ag9 membranes recorded similar fluxes. However, this high flux was achieved at the cost of low permeability selectivity (48). Membranes of different thicknesses were tested, and high H2 permeability at 400 °C was measured to be 9.2·10. -6 mol m -2 s -1 Pa -1 The optimal selectivity (3500) for films with relatively large thicknesses was determined. The optimal combination of flux and selectivity at 400 °C was recorded, with an H2 flux of 5.1 × 10⁻⁶ for a 2.5 μm thick Pd₈₀Ag₂₀ film under a pressure difference of 1 bar. -6 mol m -2 s -1 Pa -1 Furthermore, the selectivity exceeds 10,000. According to the present invention, a membrane (DS-2 containing 5 wt% Ag) with a selectivity layer of ~1 μm was prepared, which exhibits a similar flux (4.6·10⁻⁶). -6 mol m -2 s -1 Pa -1 The permeability and selectivity are ~26000 H2 / N2. The results described in this study are among the best recorded for a combination of high permeability and selectivity.

[0060] Once the membrane properties (DS-2) have been studied using single-gas and binary gas mixtures in the absence of glass beads, such as... Figure 7 As shown, during the first 250 hours (with a gray background), the reactor was cooled to room temperature to introduce glass beads with a particle size of 250-350 μm. The catalyst volume was sufficient to completely cover the membrane. Subsequently, the reactor was reheated to 350 °C, and the particles were fluidized (U / U... mf = 2) 100 hours. Then, increase the fluidization rate to U / U mf = 3, and the temperature was increased to 400°C (transmembrane pressure of 1 bar). The same binary gas mixture test was performed before the repeated introduction of particles to evaluate the effect of the presence of fluidized particles on bed-membrane mass transfer limitation. Total flow rate was increased from 10 L / min. -1 Change to 20L per minute -1 Therefore, the membrane surface suffers more wear due to increased fluidization rate and particle mixing. The H2 / N2 permeation selectivity was approximately 25,000, but after the introduction of particles, the selectivity began to decrease until the end of the long-term test (selectivity decreased to 5,500 after 975 hours). During the fluidization test with glass beads, the filter on the downstream side remained clogged twice. Lowering the temperature to 300°C allowed for filter cleaning (see...). Figure 7 (Temperature and pressure peaks in the sample). A gray powder was found and characterized by XRD. The obtained diffraction pattern corresponds to glass beads (mainly composed of SiO2). This indicates that particle breakage during testing was likely due to erosion of the reactor wall and membrane surface. Once the long-term test was completed and the module cooled, the protective layer was observed to be removed from the membrane surface. As selectivity decreased, a leak test was performed to locate the leak. As described above, the membrane was pressurized to 1 bar from the inside with helium while being immersed in ethanol. The membrane was slowly impregnated to assess which part of the membrane caused the gas leak. It was found that the seals (top and bottom seals) caused the leak, not the membrane surface. Therefore, even with the removal of the protective layer, the selectivity layer remained intact. It can be expected that the protective layer would be removed due to the collision of hard glass particles with the membrane surface. To evaluate this, further tests were performed using actual catalyst particles that are harder than glass.

[0061] For binary mixtures, membrane DS-2 (with 5% silver by weight) was tested using a binary mixture of H2-N2 and H2-CO2, with hydrogen feed content varying from 50% to 100% by volume. Permeation tests were conducted at 400°C with the total feed flow rate fixed at 10 L / min. -1 These experiments were repeated in the presence of fluidized glass beads. In this case, different feed flow rates (10, 15, and 20 L / min) were investigated. -1 In all cases, the osmotic side remains at atmospheric pressure.

[0062] Figure 8A and Figure 8B The results show that even with only 10% by volume N2 or CO2 mixed with H2, the hydrogen flux decreases significantly in the presence of N2 or CO2. The reduction in H2 flux was very similar for all the mixtures studied, therefore no CO2 interaction with the membrane (and no reverse water-gas shift) was observed. These results indicate the formation of a concentration-transfer boundary layer in the gas phase, which reduces the H2 concentration near the membrane surface. For membranes with high H2 permeability, this phenomenon, known as concentration polarization, becomes significant, even when the H2 volume percentage on the stagnant side remains high.

[0063] Once the glass beads are introduced into the reactor ( Figure 8B and Figure 8D The same binary gas mixture tests were then performed. The results showed that the H2 permeate flux was slightly higher in the fluidized bed test compared to the empty tube experiment. This demonstrates that the fluidized bed increases gas mixing due to particle motion, thereby reducing the degree of external mass transfer constraint (i.e., the thickness of the concentration distribution near the membrane surface is reduced). However, for each mixture, this improvement was only observed at low transmembrane pressures. For example, for an H2-N2 mixture with 50 vol% H2, the H2 flux was higher in the presence of glass beads when the partial pressure of hydrogen was less than 1.4 bar. Figure 9A For a feed with 90% H2 by volume, this value increases to 1.7 bar. Figure 9B Above this pressure, the hydrogen flux in the fluidized bed is lower than that measured in the empty tube. This may be related to another phenomenon. At high fluxes, particles are dragged toward the membrane surface and form dense regions close to the membrane, which causes additional mass transfer resistance, thus reducing the flux.

[0064] The total feed rate also affects the hydrogen flux, such as Figures 9A-9C As shown, in the presence of glass beads, the total locked feed rate increases, and the permeate H2 flux also increases. However, hydrogen recovery on the permeate side decreases, and the membrane surface area relative to the feed rate decreases, thus reducing the membrane's effect on hydrogen recovery. The same trend was observed for H2-CO2 mixtures (see [reference]). Figure 8C ).

[0065] Now let's examine Rh-promoted alumina and ceramic-supported DS membranes. Another membrane (DS-1) with 4 wt% silver in the selective layer was tested for a longer time under fluidized conditions using a catalyst supplied by Johnson Matthey®. The catalyst consisted of modified alumina particles (100-300 μm size) loaded with 0.5 wt% rhodium. Unlike glass particles, the promoted alumina-supported Rh particles are non-spherical and harder. Therefore, more severe wear on the membrane surface and a faster decrease in selectivity are expected compared to glass bead experiments. The nitrogen permeation at 400 °C was 4.15∙10⁻⁶ before integrating the membrane into the reactor. -10 molm -2 s -1 Pa -1 Furthermore, defects on the membrane surface have been identified during the seal leak test. During the first 350 hours, this value increased (see...). Figures 8A-8C Hydrogen and nitrogen permeation tests were performed until permeability reached a constant (350 hours). During this period, single-gas and binary gas mixture tests were conducted (the results of which are not included in this paper). At this point, the ideal permeation selectivity of the membrane at 300°C and 500°C at a pressure difference of 1 bar was ~2800 and ~3700, respectively, and N2 permeability increased up to ~1.4∙10⁻⁶. -9 mol m -2 s -1 Pa -1 The reactor was then cooled to 200°C, and catalyst was introduced to completely submerge the membrane. At 400°C and a pressure difference of 1 bar, the bed was in a free bubbling state (U / U...). mf = 3) Downflow. H2 and N2 permeability were monitored for over 1500 hours. During this period, the ideal permeability selectivity remained at approximately 2000, and the H2 flux remained at 3.35 × 10⁻⁶. -6 mol m -2 s -1 Pa -1 Then, the temperature was raised to 450°C and maintained for 7 days. Finally, the temperature was raised to 500°C. From Figure 10 It can be clearly seen that nitrogen leakage begins to increase sharply when the temperature rises to 450℃. At the end of the experiment, the nitrogen leakage at 500℃ was 3.10. -8 mol m -2 s -1 Pa -1 .

[0066] The membrane surface and seals were investigated in more detail at room temperature. Following this long-term test, the protective layer was observed to remain on the membrane surface, and surface abrasion was not the primary cause of the previous removal of this layer. Helium leakage was tested by immersing the membrane in ethanol and introducing a gas pressure of 1 atm (atmospheres). Leakage was observed at both seals, but a larger amount was observed at the top seal. To quantify the contribution of the membrane surface to the total nitrogen leakage, both seals were covered with an airtight resin. Nitrogen leakage at the membrane surface was calculated to be only 8% of the total leakage. Therefore, it can be concluded that the membrane performance was only affected by 8% after long-term testing (~2500 hours), and the main leakage was found at the seals.

[0067] The third membrane (DS-3) with 4% silver by weight was tested under fluidized conditions at 550 °C and a pressure differential of 1 bar for ~700 hours, using the same catalyst as the previous membrane DS-1 (alumina-based catalyst, see [link]). Figure 11 Hydrogen permeability of DS-3 at 550℃ (1.3·10) -6 mol m -2 s -1 Pa -1 It is lower than the hydrogen permeability of DS-2 at 500℃ (3.2·10). -6 mol m -2 s -1 Pa -1 This may be partly due to the fact that the DS-3 selective layer (~1.86 μm) is thicker than the DS-2 selective layer (~1 μm), but the difference in the thickness of the protective layer (~0.5 μm for DS-2 and ~0.67 μm for DS-3) may also have played a role.

[0068] like Figures 9A-9C As shown, the ideal H2 / N2 permeability of DS-3 is not as high as that of DS-2. This is not only related to the lower hydrogen permeability of DS-3, but also to the large defects found on the membrane surface that lead to nitrogen leakage. N2 leakage increases extremely rapidly in the first 250 hours. The experiment was stopped when the ideal permeability selectivity decreased to approximately 50. As with the DS-1 membrane, the leakage contribution at room temperature was quantified by pressurizing the inside with helium before and after covering the seal with hermetically sealed resin. For DS-3, which had defects from the start of the test, the membrane's total contribution to leakage was 20%. Furthermore, investigations should be conducted to improve the sealing of the supported membrane on the ceramic carrier.

[0069] Conventional metal-supported membranes and bilayer (DS) membranes have been tested under fluidized conditions to compare their performance in FBMR during long-term testing.

[0070] First, the membrane was tested in an empty reactor with pure gases (H2 and N2) in a temperature range of 300–500 °C to obtain key membrane permeability characteristics, such as hydrogen permeability, ideal H2 / N2 selective permeability, activation energy, and n-value (summarized in Table 3). The membranes were observed to exhibit very similar hydrogen permeability (1.34–1.55 · 10⁻⁶) at 500 °C and a pressure difference of 4 bar. -6 molm -2 s -1 Pa -1 Under the same conditions, the permeability selectivity of the metal-supported membrane (M33) is higher than 90,000, while for the bilayer membrane, the permeability selectivity is almost infinite. The activation energies of the two membranes were found to be very similar and below 10 kJ mol⁻¹. -1 With the addition of a thicker protective layer, the pressure exponent (n-value) increased from 0.62 to 0.71, which may be related to viscous flow and Knudsen diffusion in the porous protective layer. Since both mechanisms exhibit consistent pressure exponents, the addition of a porous layer may lead to an increase in the n-value, as observed in this study.

[0071] Table 3. Key parameters of conventional metal-supported membranes (M33) and bilayer membranes (M15-DS) in an empty reactor.

[0072]

[0073] Once the catalyst was integrated into the reactor, the ideal permeation selectivity of the M33 membrane decreased to ~14000 during the first hour. When operating at 400°C and 500°C, this value further decreased to ~1000 after 615 hours, due to nitrogen leakage from 8.76 × 10⁻⁶ at a pressure differential of 4 bar. -11 Increased to 1.4·10 -9 mol m -2 s -1 Pa -1 Caused by ( Figure 12 On the other hand, the DS-membrane (M15-DS) did not exhibit any nitrogen leakage under the same conditions, demonstrating better abrasion resistance than conventional metal-supported membranes. During testing in a bubbling fluidized state, the equipment was cooled twice, but the leakage of both membranes did not increase due to the temperature change. Figure 12 ).

[0074] These results indicate that bilayer membranes exhibit better performance under fluidized conditions. Future research will focus on testing the performance of bilayer membranes at higher temperatures.

[0075] Adding a porous protective layer to metal-supported Pd-based membranes increases their performance stability under fluidized conditions, paving the way for their application in fluidized bed membrane reactors. Indeed, membranes without the protective layer were observed to significantly increase nitrogen leakage under fluidized conditions, while bilayer membranes remained stable for over 615 hours at 400–500 °C and a pressure differential of 4 bar.

[0076] Regarding titanium dioxide-based catalysts, it has been reported that conventional Pd-based membranes (E54) supported by ceramics interact with TiO2-based particles in the presence of these particles. Hydrogen flux decreases when the particles are fluidized in the membrane reactor, as... Figure 13 As shown. Post-morten analysis of the membrane revealed the presence of Ti on the membrane surface, indicating a strong chemical interaction. The particle size used in this test ranged from 76 μm to 106 μm. The same test was performed using a ceramic-supported bilayer (DS-44) with a larger particle size distribution (100–305 μm). Hydrogen flux did not decrease over more than 140 hours (see [reference]). Figure 13 Furthermore, in FBMRs above 10000, the H2 / N2 permeation selectivity of DS-44 remained nearly equal. To determine whether smaller particles could interact with the membrane, fine particles (<100 μm) were incorporated into the reactor. For this purpose, the reactor was cooled to room temperature in nitrogen. Nitrogen leakage was observed from 3.7·10 at 400 °C. -10 mol m -2 s -1 Pa -1 The temperature rose to 2.4·10 at 20℃ -9 mol m -2 s -1 Pa -1 This increase in N2 leakage is likely related to the sealing section and was investigated after testing with small particles. The membrane was exposed to hydrogen once the reactor was heated to 400°C. The measured hydrogen flux was greater than in previous experiments without fine particles. Furthermore, the H2 flux did not decay during the tests, so no interaction between the particles and the membrane was observed. Descriptions of the various membranes and particle sizes used in each experiment are summarized in Table 4. It can be observed that for E54 and for DS-44 with particles between 100–305 μm, the hydrogen permeability measured in the empty reactor and in FBMR was lower at the beginning. However, after 50 hours, DS-44 almost recovered to its initial value (in the empty reactor), while E54 continued to decrease. For DS-44 with <305 μm particles, H2 permeability increased slightly in the first hour of FBMR and remained constant for more than 110 hours. Selectivity was lower than in previous tests due to nitrogen leakage. Leakage through the sealing section was found to account for approximately 90% of the total leakage.

[0077] Table 4. Main parameters of the films tested in FBMR with TiO2-based particles

[0078]

[0079] The present invention is now described with reference to several exemplary embodiments, which should be considered illustrative rather than restrictive. Therefore, various variations of the specific embodiments of the invention can be derived by those skilled in the art from the description herein. All such variations should be considered to be within the spirit and scope of the invention and are covered by, for example, the following technical solutions and their equivalents.

[0080] This application also includes, for example, the following technical solutions:

[0081] 1. A hydrogen permeation membrane device, comprising:

[0082] a. Hydrogen-selective Pd thin films or Pd alloy thin films; and

[0083] b. A porous protective layer comprising a material selected from the group consisting of: zirconium oxide, yttrium-stabilized zirconium oxide (YSZ), γ / Al2O3, YSZ-γ / Al2O3, silicon dioxide, titanium dioxide, magnesium oxide, cerium dioxide, nitrides, and silicon carbide.

[0084] 2. The hydrogen permeation membrane device as described in claim 1, wherein the pore size of the porous protective layer is in the range of 1 nm to 400 nm, or 1 nm to 50 nm, or 2 nm to 20 nm.

[0085] 3. The hydrogen permeation membrane device as described in item 1, wherein the hydrogen permeation membrane further comprises a porous support layer.

[0086] 4. The hydrogen permeation membrane device as described in item 3, wherein the porous support layer comprises: metal, ceramic, or metal and ceramic.

[0087] 5. The hydrogen permeation membrane device of claim 4, wherein the ceramic porous support layer comprises a material selected from the group consisting of oxides, nitrides, and carbides.

[0088] 6. The hydrogen permeation membrane device as described in item 3, wherein the pore size of the porous support layer ranges from 2 nm to 50,000 nm.

[0089] 7. The hydrogen permeation membrane device as described in claim 1, wherein the thickness of the hydrogen-selective Pd layer or the Pd alloy layer ranges from 0.5 μm to 150 μm.

[0090] 8. The hydrogen permeation membrane device as described in claim 1, wherein the thickness of the porous protective layer ranges from 0.5 μm to 150 μm.

[0091] 9. The hydrogen permeation membrane device of claim 1, wherein the porous protective layer further comprises a catalyst in the pores, wherein the catalyst is selected from the group consisting of Pt, Ni, Rh, Ru, Ag and Pd.

[0092] 10. The hydrogen permeation membrane device of claim 1, further comprising a housing encapsulating the hydrogen permeation membrane, wherein the housing includes a retention end and a permeation end.

[0093] 11. The hydrogen permeation membrane device of claim 10, wherein the housing comprises a material selected from the group consisting of stainless steel, carbides, ceramics, and Hastelloy corrosion-resistant nickel-based alloys.

[0094] 12. The hydrogen permeation membrane apparatus of claim 10, wherein the housing is disposed in a hydrogen production reaction vessel, wherein the generated hydrogen flows from the stagnation end of the housing to the permeation end of the housing.

[0095] 13. The hydrogen permeation membrane device of claim 10, wherein the shape of the housing is selected from the group consisting of: tubular, microchannel, planar, and parallel plate.

Claims

1. A hydrogen permeation membrane comprising: a porous support layer comprising a metal and / or a ceramic; a hydrogen selective Pd-Ag membrane having a thickness in the range of 0.5 pm to 150 pm arranged on the porous support layer; and a porous protective layer comprising YSZ-γ / Al2O3 having a thickness in the range of 0.5 pm to 2 pm arranged on the hydrogen selective Pd-Ag membrane, wherein the porous protective layer has a pore size in the range of 1 nm to 50 nm, and wherein the porous protective layer comprises a catalyst in the pores.

2. The hydrogen permeation membrane device of claim 1, wherein the porous protective layer of the hydrogen permeation membrane has a pore size in the range of 2 nm to 20 nm.

3. The hydrogen permeation membrane device of claim 1, wherein the porous support layer of the hydrogen permeation membrane comprises a material selected from the group consisting of oxides, nitrides, and carbides.

4. The hydrogen permeation membrane device of claim 1, wherein the porous support layer of the hydrogen permeation membrane has a pore size in the range of 2 nm to 50,000 nm.

5. The hydrogen permeation membrane device of claim 1, wherein the catalyst is selected from the group consisting of ZnO, Pt, Ni, Rh, Ru, Ag, Pd, Pt-Rh, Pd-Ag, and Pt-Pd.

6. A fluidized bed membrane reactor comprising the hydrogen permeation membrane device according to any one of the preceding claims.

Images