A kind of MOF separation membrane, preparation method and its application in organic azeotropic system separation

Abstract

The application discloses a MOF separation membrane, a preparation method and application of the MOF separation membrane in separation of an organic azeotropic system, and in view of the problem of lattice defects of a MOF-801 membrane prepared by an in-situ solvothermal method reducing the separation performance of the membrane, a strategy of adjusting a metal source type and post-synthesis repair is proposed to inhibit the lattice defects in the MOF-801 membrane, precise regulation of effective screening pores in the MOF-801 membrane is realized, intrinsic sub-nanometer mass transfer lattice pores in the membrane are constructed, and based on the size screening mechanism, the separation performance of the MOF membrane in a methanol / dimethyl carbonate, methanol / methyl acetate, methanol / methyl tert-butyl ether or other azeotropic system is significantly improved.

Description

Technical Field

[0001] This invention relates to a MOF separation membrane, its preparation method, and its application in the separation of organic azeotropic systems, belonging to the field of membrane separation technology. Background Technology

[0002] In 2012, Behrens' group first synthesized a novel zirconium-based MOF, Zr-fumarate, via hydrothermal synthesis. In 2014, Yaghi's group officially named it MOF-801. MOF-801 is a PN3... - Structure, Zr 4+ It has a tetragonal antiprism coordination geometry and contains 8 oxygen atoms. Figure 1 Its three-dimensional framework is constructed from ZrO2 secondary structural units. Each ZrO2 secondary structural unit consists of six crystallographically equivalent Zr... 4+ The space group is composed of two sub-structural units, each coordinated with 12 organic ligands (fumaric acid). Its three-dimensional framework consists of a fumarate salt and two sub-structural units, forming porous tetrahedral and octahedral cages. This space group has two independent tetrahedral cavities with different dimensions, having diameters of [missing information]. and The diameter of the octahedral cavity is Its minimum window size Just between methanol With most ester organic compounds (such as dimethyl carbonate): The kinetic diameters of MOF-801 are between those of Zr, and furthermore, based on the theory of hard and soft acids and bases, MOF-801 is composed of Zr. 4+ It binds to carboxylic acid ligands via a hard acid-hard base relationship, resulting in extremely strong coordination bonds and thus excellent water stability, thermal stability, and chemical stability. Therefore, MOF-801 holds promise for achieving efficient sieving of liquid mixtures.

[0003] Using MOF-801 as the separation membrane research object, MOF-801 membranes were constructed by in-situ synthesis (see schematic diagram of the membrane). Figure 2 As shown in the figure, the separation performance of the prepared MOF membrane for organic azeotropic systems was found to be positively correlated with the kinetic diameter of the organic molecules, indicating that size sieving is dominant. Furthermore, the MOF-801 membrane exhibited limited separation performance for systems with kinetic diameters larger than its intrinsic lattice pore size, suggesting the presence of lattice defects within the membrane. Summary of the Invention

[0004] The purpose of this invention is to provide a MOF separation membrane, a preparation method thereof, and its application in the separation of organic azeotropic systems. By controlling the metal source and using a post-synthetic repair strategy, the crystal defects of the membrane are reduced, giving the membrane excellent separation performance, thermal stability, and chemical stability.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A MOF separation membrane includes a support and a membrane layer, wherein the support is a porous material;

[0007] The membrane is composed of MOF-801, with a thickness of 1-5 μm, a pore size of 0.5-0.6 nm, and a specific surface area of ​​400-500 m². 2 / g; characterized in that the lattice defects of MOF-801 in the film layer are less than 15%.

[0008] Preferably, the support is made of α-Al2O3, and the support has a porous structure with a pore size of 100-300 nm and a porosity of 20%-45%.

[0009] The preparation method of the above-mentioned MOF separation membrane includes the following steps:

[0010] Casting membrane: The support is fixed in the reaction vessel, casting solution is poured in to form a membrane, the sample is cooled and solvent exchanged after completion, and then dried to obtain MOF separation membrane;

[0011] The casting solution is a mixed solution containing ZrCl4, fumaric acid, an organic solvent, and formic acid. The molar ratio of ZrCl4 to fumaric acid in the casting solution is 1:0.8 to 1.3, and the volume ratio of the organic solvent to formic acid is 100:20 to 50.

[0012] Preferably, the process also includes a post-repair step, in which after obtaining the separation membrane, the membrane is fixed in the reaction vessel, a repair solution is poured in and repaired, the sample is cooled and solvent exchanged, and then dried to obtain the post-repaired MOF separation membrane.

[0013] The repair solution is a mixed solution of an organic solvent containing fumaric acid and formic acid; the volume ratio between the organic solvent and formic acid is 100:20 to 50.

[0014] Preferably, the thickness of the support is 1-5 mm and the diameter is 20-40 mm.

[0015] Preferably, in the casting step, the film preparation specifically involves drying at 65-100℃ for 8-15 hours.

[0016] Preferably, in the post-repair step, the repair specifically involves drying at 70-90℃ for 9-14 hours.

[0017] Preferably, the lattice defects of MOF-801 in the post-repaired MOF separation membrane layer are less than 6%.

[0018] The above-mentioned MOF separation membrane is used in the separation of organic azeotropic systems.

[0019] Preferably, the organic azeotropic system refers to azeotropic systems such as alcohol / dimethyl carbonate, methanol / methyl acetate, and methanol / methyl tert-butyl ether.

[0020] The beneficial effects of this invention are as follows:

[0021] To address the issue of lattice defects in MOF-801 membranes prepared by in-situ solvothermal methods reducing membrane separation performance, a strategy of modulating the metal source type and post-synthesis repair was proposed to suppress lattice defects within the MOF-801 membrane. This enabled precise control of the effective sieving pores within the MOF-801 membrane, thereby constructing intrinsic sub-nanometer mass transfer lattice pores within the membrane. Based on the size sieving mechanism, this significantly improved the separation performance of the MOF membrane in azeotropic systems such as methanol / dimethyl carbonate, methanol / methyl acetate, and methanol / methyl tert-butyl ether. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the MOF-801 structure;

[0023] Figure 2 Schematic diagram of MOF-801 membrane;

[0024] Figure 3 This is a schematic diagram of a pervaporation membrane device;

[0025] Figure 4 Here are SEM images of the α-Al2O3 support, where (A) is a SEM surface photograph of the α-Al2O3 support; and (B) is a SEM cross-sectional photograph of the α-Al2O3 support.

[0026] Figure 5 SEM images of MOF-801 membranes at different film-forming temperatures;

[0027] Figure 6 SEM images of MOF-801 membranes at 80℃ and different film-forming times;

[0028] Figure 7 The following are the XRD patterns, Fourier transform infrared spectra, TG spectra, and BET spectra of MOF-801 films prepared on α-Al2O3 supports using in-situ synthesis: (A) XRD pattern; (B) Fourier transform infrared spectra; (C) TG spectra; (D) BET spectra.

[0029] Figure 8 Figure 1 shows the effect of membrane preparation temperature and time on the separation performance of MOF-801 membrane in the pervaporation separation of methanol / DMC.

[0030] Figure 9 The graph shows the pervaporation separation performance of the MOF-801 membrane for three organic-organic systems.

[0031] Figure 10For MOF-801 membrane gas separation performance;

[0032] Figure 11 SEM images of MOF-801 membranes with different film-forming times using zirconium chloride as the metal source;

[0033] Figure 12 The graph shows the effect of membrane preparation time using zirconium chloride as a metal source on the pervaporation separation performance of MOF-801 membrane for methanol / DMC.

[0034] Figure 13 The pervaporation performance of the MOF-801 membrane using zirconium chloride as the metal source for three organic-organic systems is shown in the figure.

[0035] Figure 14 SEM images of MOF-801 membranes at different repair times;

[0036] Figure 15 Figure 1 shows the methanol / dimethyl carbonate pervaporation separation performance of MOF-801 membranes at different repair times.

[0037] Figure 16 The diagram shows the pervaporation separation performance of three organic-organic systems in the post-repaired MOF-801 membrane.

[0038] Figure 17 The following diagrams show the (A) NMR defect concentration, (B) PXRD patterns, (C) N2 adsorption capacity, and (D) specific surface area and pore size of three different MOF-801 membranes. Detailed Implementation

[0039] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0040] Comparative Example 1

[0041] Preparation of alumina support

[0042] A certain amount of α-Al₂O₃ powder was extruded into discs under 20 MPa. The pore size of the prepared α-Al₂O₃ support was approximately 200 nm, and the porosity was 35%. The sintered α-Al₂O₃ had a thickness of 2 mm and a diameter of 28 mm. Before growing the MOF-801 film, the alumina surface was first polished with 800-grit sandpaper, then with 2000-grit sandpaper, rinsed with deionized water, and dried at 120°C for 24 hours.

[0043] Preparation of MOF-801 membrane

[0044] First, a self-made α-Al₂O₃ support was fixed in the liner of the reactor. An equimolar ratio of zirconium oxychloride octahydrate and fumaric acid was fully dissolved in a DMF / formic acid mixed solvent to prepare a synthesis solution. Then, the fully dissolved reaction casting solution was slowly poured into the reactor liner. The reactor liner was sealed and placed in a metal reactor, which was then placed in an oven for the coordination reaction. After the reaction was complete, the reactor was cooled to room temperature, the sample was removed and solvent exchanged, and finally, the membrane was dried in a vacuum drying oven at 150°C for 12 hours.

[0045] Main analysis and testing methods

[0046] Basic characterization of samples

[0047] The thermal stability of the MOF-801 film was studied by thermogravimetric analysis (TGA, NETZSCH, STA, 449F3) under a nitrogen atmosphere at a heating rate of 10 °C / min. The structural characteristics of the MOF-801 film were characterized by X-ray diffraction (XRD, Rigaku, Miniflex 600, Japan) at a scan rate of 10 °C / min. The cross-sectional and surface morphology of the MOF-801 film were observed by field emission scanning electron microscopy (FESEM, Hitachi Limited, S-4800, Japan). The sample was also sputtered with gold under vacuum using an ion sputtering system (Hitachi, MC1000, Japan) for 120 seconds. The chemical properties of the MOF-801 film were analyzed by Fourier transform infrared spectroscopy (FTIR, AVATAR-FT-IR-360, Thermoncolet, USA).

[0048] Gas separation test

[0049] In laboratory single-component gas tests, the constant pressure-variable volume method was used to evaluate the membrane's separation performance. The single-component gas permeation test was conducted at 0.2 MPa and 25 °C. When the system reached steady state, the gas permeation rate P was the average of three or more test results, calculated using the following formula:

[0050]

[0051] Where P represents the gas permeation rate (1 GPU = 10^6). -6 cm 3 (STP)cm -2 s -1 cmHg -1Patm represents atmospheric pressure (atm), Δp and T represent the transmembrane pressure difference (atm) and test temperature (°C), respectively, A is the effective area of ​​the membrane, and dV / dt corresponds to the volumetric displacement rate in the soap bubble flowmeter. The ideal selectivity of H2 / CO2 is calculated from the ratio of the permeation rates of the pure gases, and can be expressed as follows:

[0052]

[0053] Pervaporation test

[0054] The prepared MOF-801 membrane was loaded into a self-made assembly. Figure 3 This is a schematic diagram of a pervaporation apparatus. The feed liquid circulates between the feed tank and the membrane module at a certain flow rate via a circulating pump, and the feed liquid is at the test temperature. Its concentration is generally expressed as the mass ratio of small molecule components (e.g., 10 wt% methanol). The pervaporated vapor is condensed using liquid nitrogen and collected through a cold trap. After liquefaction, the concentrations on the feed side and the permeate side are determined using gas chromatography (GC-2014, Shimadzu, Japan). The pervaporation performance of the membrane is evaluated based on the total flux J and the separation factor β. The feed liquids selected are methanol / DMC solution, methanol / MTBE solution, and methanol / MAC solution with a methanol content of 10 wt%. Total flux J (kg·m³) -2 ·h -1 The following calculations were performed using the sample mass collected in the cold trap and the testing time:

[0055]

[0056] Where W represents the total mass of the collected samples (kg), and t and A represent the test time (h) and effective membrane area (m²), respectively. 2 ).

[0057] The separation factor is calculated using the following formula:

[0058]

[0059] Where X and Y represent the mass fraction of the component on the permeation side and the feed side, respectively.

[0060] membrane micromorphology

[0061] Field emission scanning electron microscope (SEM) image of a self-made α-Al2O3 support. Figure 4 As shown in the electron microscope images of the surface and cross-section of the support, the α-Al2O3 ceramic sheet has a smooth surface, which is suitable for film growth, and its pore size is approximately 200 nm.

[0062] To investigate the optimal preparation conditions for MOF-801 membranes, the effects of different temperatures and membrane preparation times on membrane formation were examined. Figure 5 The images show electron microscope (EM) images of MOF-801 films prepared at different temperatures for 6 hours. The images reveal that at 60℃ and 70℃, continuous films cannot be formed; the film layer is barely visible in the cross-section. This is because the temperature is too low, resulting in insufficient energy supply, preventing MOF grains from nucleating on the α-Al₂O₃ support and thus hindering the formation of a dense film. At 80℃ and 90℃, continuous and dense films can be formed, with a cross-sectional thickness of approximately 1.5 μm. However, cracks are clearly visible on the surface of the film prepared at 90℃. We believe this is due to the mismatch in thermal expansion coefficients between the α-Al₂O₃ support and the MOF-801 film. At 100℃ and 110℃, due to the excessively high temperature, MOF-801 grains preferentially nucleate in the mother liquor rather than on the support surface, resulting in only particle accumulation on the α-Al₂O₃ support. Therefore, we selected 80℃ as the optimal temperature.

[0063] Subsequently, the effect of synthesis time on the preparation of MOF-801 membranes was investigated at a synthesis temperature of 80℃. Figure 6 As shown, when the synthesis time is less than 6 hours, a large number of MOF particles accumulate on the support surface, failing to form an intergrowth film. This is because the reaction time is short, resulting in small MOF particles that are difficult to intergrow and form a continuous film. When the synthesis time is extended to 6 hours, the α-Al₂O₃ support surface is completely covered by MOF particles, and the crystals exhibit good intergrowth morphology. The film thickness is approximately 1.5 μm. Further extending the synthesis time did not significantly alter the surface morphology or thickness of the prepared film; therefore, the optimal film deposition time is 6 hours.

[0064] Physicochemical property characterization of membranes

[0065] The structural properties of MOF-801 films prepared on α-Al2O3 supports were studied using XRD. Figure 7 (Region A). The prepared MOF-801 film exhibits characteristic peaks of MOF-801 crystals around 10°, which are consistent with the standard spectrum of MOF-801. The characteristic peaks of the α-Al₂O₃ support appear around 25° and 35°, and the intensity of the MOF-801 characteristic peaks is weaker, which may be due to the thinness of the prepared MOF film. Subsequently, we used FTIR to study the chemical properties of the MOF-801 film, such as... Figure 7 As shown in region B, at 3217cm -1 The characteristic peak generated at 1655 cm⁻¹ is from the -OH bonds of the Zr cluster. -1 The peak at 1580 cm⁻¹ represents the -C=O bond in the solvent DMF. -1 1398cm -1 And 1207cm -1and 1103cm -1 The peaks represent the -C=O-OH and -C=O bonds generated by the functional groups of the organic ligand fumaric acid.

[0066] The thermal stability of MOF-801 film was studied by TGA. Figure 7 Region C). The results show that the initial slight weight loss of the MOF-801 sample before 100℃ is due to the removal of water absorbed by MOF-801. A second weight loss occurs between 100℃ and 260℃, due to the removal of guest molecules (such as methanol, N,N-dimethylformamide, etc.) within the pores of MOF-801. With further temperature increases, between 260℃ and 500℃, carboxylate groups decompose successively, leading to the final decomposition of MOF-801. These results indicate that the structural collapse temperature of MOF-801 is approximately 450℃, and TGA results show that MOF-801 exhibits superior thermal stability compared to other MOF materials (approximately 350℃). N2 adsorption measurements at 77K show ( Figure 7 MOF-801 (region D) exhibits type I adsorption behavior, and the calculated BET surface area of ​​MOF-801 is 483.83 m². 2 / g. The above results demonstrate the successful preparation of the MOF-801 membrane.

[0067] Performance testing of organic azeotropic systems

[0068] The separation performance of the MOF-801 membrane for organic azeotropes was investigated by pervaporation experiments on a methanol / dimethyl carbonate mixture (10 / 90 wt%) at 50 °C. First, the effect of membrane fabrication temperature on the pervaporation separation performance of the MOF-801 membrane was examined, such as... Figure 8 As shown in region A, a continuous MOF film did not form on the surface of the porous support when the synthesis temperature was below 80℃ and above 90℃. Figure 5 The prepared MOF-801 membrane has a permeation flux as high as 30 kg·m³. -2 ·h -1 The methanol content on the permeate side was similar to that of the feedstock, and no separation performance was observed. At a synthesis temperature of 80℃, the cross-grown MOF membrane layer formed, significantly improving the methanol / dimethyl carbonate separation performance (flux: 7.2 kg·m³). -2 ·h -1 Separation factor: 135, feed temperature: 50℃, feed concentration: 10wt% methanol). We then investigated the effect of membrane synthesis time on the separation performance of methanol / dimethyl carbonate. Figure 8 (Region B). As the synthesis time gradually increased, the permeation flux showed a decreasing trend, while the methanol / dimethyl carbonate separation factor gradually increased. When the synthesis time was 2–4 h, the membrane permeation flux reached 150 kg·m. -2·h -1 The separation factor of methanol / dimethyl carbonate is only about 3. This is because the continuous membrane layer does not completely cover the surface of the porous support, resulting in a large number of macropore defects (such as...). Figure 6 (As shown). Extending the synthesis time to 6 hours reduced the methanol / dimethyl carbonate permeation flux to 7.5 kg·m³. -2 ·h -1 The separation factor increased to 130, and after further extending the synthesis time, the separation factor and permeation flux tended to stabilize. This indicates that when the synthesis time is greater than 6 hours, the MOF-801 membrane with a molecular sieving effect mass transfer channel has been formed, which can also be verified by SEM characterization. Figure 6 ).

[0069] To further investigate the molecular sieving capability of the MOF-801 membrane, we also examined the separation performance of the prepared MOF membrane for two other organic azeotropic systems with different kinetic diameters: methanol / methyl tert-butyl ether. and methanol / methyl acetate like Figure 9 As shown, for the methanol / methyl acetate system, the MOF-801 membrane exhibits a high permeation flux of 8.9 kg·m³. -2 ·h -1 The separation factor was as low as 72, and for the methanol / dimethyl carbonate system, the permeation flux decreased to 7.8 kg·m³. -2 ·h -1 The separation factor improved. This may be due to the larger kinetic diameter of dimethyl carbonate, which hinders methanol diffusion to some extent. Interestingly, further increasing the kinetic diameter of organic molecules to... At this point, the separation factor of the MOF membrane increased to 200, exhibiting certain separation performance. This is because methanol molecules have a relatively small kinetic diameter. It can pass through the MOF membrane pores quickly, while ester or ether molecules have larger dynamic diameters and encounter greater mass transfer resistance when passing through the MOF membrane pores, thus achieving effective sieving of both, indicating that the size sieving mechanism is dominant.

[0070] According to literature reports, the sieve aperture diameter of MOF-801 is approximately... It lies precisely within the kinetic diameter of methanol. With dimethyl carbonate Methyl tert-butyl ether Based on the kinetic diameter and size sieving, theoretically, MOF-801 membranes can achieve highly efficient sieving of methanol / dimethyl carbonate and methanol / methyl tert-butyl ether. However, the separation factors of the MOF-801 membranes we prepared via in-situ hydrothermal treatment for these two systems were only 72–130. We hypothesize that the prepared MOF-801 membranes may contain lattice defects (cluster or ligand deficiencies), increasing their pore size and thus reducing their molecular sieving capacity. To verify our hypothesis, we evaluated the membrane regularity using H2 / CO2 separation performance. Three MOF-801 membranes (M1, M2, and M3) were selected respectively, as follows: Figure 10 As shown, the membrane prepared under optimal conditions exhibits a high H2 permeability coefficient (8500 GPU) and an H2 / CO2 selectivity of around 5, which is close to its Knudsen diffusion selectivity. Since the cell size of MOF-801 is 2 nm, the minimum pore size corresponding to Knudsen diffusion when gas molecules move within the membrane pores is 2 nm. Therefore, the gas separation performance confirms the presence of lattice defects in the prepared MOF-801 membrane.

[0071] Example 1

[0072] MOF-801 membrane prepared by controlling the metal source

[0073] In this embodiment, we selected ZrCl4 as the metal source. The specific film preparation process is as follows: A self-made α-Al2O3 support was fixed on the inner lining of the reactor. Different masses of ZrCl4 and fumaric acid were dissolved in a mixed solvent of DMF / formic acid (200 mL / 70 mL) to prepare the casting solution. Then, a certain amount of the synthesis solution was slowly poured into the reactor lining, and the lining was placed inside the metal reactor. The reactor was then placed in an oven for the coordination reaction. After the reaction was completed and cooled to room temperature, the sample was removed, subjected to solvent exchange, and finally dried at 150°C for 12 hours under vacuum.

[0074] Example 2

[0075] Post-repair strategy for preparing MOF-801 membrane

[0076] A MOF-801 membrane prepared using ZrCl4 as the metal source was immobilized at the bottom of the reaction vessel. Fumaric acid ligands were dissolved in a mixed solvent of DMF / formic acid (200 mL / 70 mL) to prepare a retrieval solution. The retrieval solution was then slowly poured into the reaction vessel, which was sealed and placed in an oven for a certain reaction time. After the reaction, the mixture was cooled to room temperature, the sample was removed for solvent exchange, and finally dried in a vacuum drying oven at 150°C for 12 hours.

[0077] NMR method for calculating defect concentration

[0078] During MOF crystal synthesis, modifiers (mainly referring to unit acids) can coordinate with metal clusters, leading to ligand loss defects. Therefore, the content of the modifier can be used to quantify the concentration of lattice defects. Based on... 1 Based on HNMR data, the molar ratio of modifier to ligand can be calculated using the following formula:

[0079]

[0080] Where N HMod and N Hlinker These refer to the NMR signals in the modulator and ligand, respectively. 1 The number of H protons. For example, for the ligand fumarate, Nbdc = 4. Mod.1H Int. and linker 1H Int. refer to the intensity of the NMR signal of the modulator and ligand, respectively.

[0081] Therefore, the lattice defect concentration (Cdef.) obtained from NMR data can be calculated using the following formula:

[0082]

[0083] In Comparative Example 1, zirconium oxychloride was used as the metal source for preparing the MOF-801 membrane. Due to its low Zr-O bond energy, it reacts readily and rapidly, leading to excessively fast nucleation of MOF-801 crystals and the generation of numerous lattice defects during this process. To reduce the lattice defect density within the membrane, zirconium chloride, with its higher bond energy, was chosen as the metal source. The Zr-Cl bond energy is higher than that of the Zr-O bond, thus slowing down the reaction rate. With a slower nucleation rate, the prepared MOF-801 membrane exhibits fewer defects and is expected to demonstrate higher molecular sieving performance. The effect of synthesis time on the preparation of the membrane using ZrCl4 as the metal source was systematically investigated at 80℃. Figure 11 As shown, the time required to synthesize a continuous and dense membrane using zirconium chloride as the metal source was 12 h, which is longer than the time (6 h) when using ZrOCl2 as the metal source. This is because in the MOF synthesis process, ZrCl4 needs to be hydrolyzed to ZrOCl2 first, then form metal clusters, and finally coordinate with organic ligands. Therefore, ZrCl4 has a lower reaction rate, which may result in a lower lattice defect concentration, thus improving the separation performance of the organic azeotropic system. Further extending the membrane deposition time did not significantly change the membrane morphology.

[0084] The MOF-801 membrane, prepared using zirconium chloride as the metal source, underwent pervaporation testing. The pervaporation performance of methanol / dimethyl carbonate was also tested at a temperature of 50°C and a methanol / dimethyl carbonate mass ratio of 1 / 9. Figure 12As shown, due to the use of zirconium chloride as the metal source, the nucleation rate is slowed down, the defects in the membrane are reduced, and the separation factor is increased to about 250, while the corresponding flux decreases to 5 kg·m. -2 ·h -1 The effect of membrane preparation time on performance was correlated with electron microscopy results. The membrane performance reached its optimal level at a preparation time of 12 hours, and subsequent extensions of the preparation time yielded almost no change in performance. Both pervaporation performance tests and electron micrographs demonstrated that the optimal conditions for preparing MOF-801 membranes using zirconium chloride as the metal source were 80°C and 12 hours.

[0085] Subsequently, the pervaporation separation performance of the MOF-801 membrane prepared using zirconium chloride as the metal source was tested for three different organic-organic systems. The results showed that, due to the reduction of defects, the separation factor of all three systems was improved compared to the MOF-801 membrane using zirconium oxychloride, while the flux decreased accordingly. Figure 13 As shown, for methanol / methyl tert-butyl ether, the separation factor reached 300, and the flux was 7 kg·m. -2 ·h -1 The methanol / dimethyl carbonate system has a flux of approximately 250 and a throughput of 5.5 kg·m³. -2 ·h -1 After switching to a metal source ligand, the nucleation rate of MOF-801 crystals slowed down, resulting in fewer defects throughout the film formation process compared to when zirconium oxychloride was used as the metal source ligand. This led to a significant improvement in the separation performance of the MOF-801 membrane using zirconium chloride as the metal source for three different organic azeotropic systems. From a performance perspective, the improvement in membrane separation performance due to the regulation of defects in the MOF-801 membrane is more clearly observed.

[0086] Although the MOF-801 membrane prepared using zirconium chloride as the metal source has significantly fewer defects compared to the MOF-801 membrane prepared using zirconium oxychloride as the metal source, and also has a substantial improvement in the pervaporation separation performance of organic azeotropic systems, the in-situ synthesis method itself is prone to producing defects in the prepared membrane. This method involves simultaneous crystal nucleation and growth, inevitably generating a certain amount of defects during film formation. Therefore, we further modulate the MOF-801 membrane through post-repair techniques to reduce its defects and achieve its optimal separation performance for pervaporation separation of organic-organic mixtures.

[0087] Post-synthesis repair

[0088] MOF-801 membranes prepared using zirconium chloride as the metal source have fewer defects compared to those prepared using zirconium oxychloride as the metal source. However, defects are still unavoidable during the membrane preparation process. Therefore, we perform post-synthesis repair on the membrane to further reduce its defects.

[0089] The specific method involves immersing the MOF-801 membrane, prepared using zirconium chloride as the metal source, in an organic ligand solution, and then placing the membrane in an 80°C oven for repair. Figure 14 As can be observed, the repaired membrane surface is smoother, but its thickness remains unchanged. Furthermore, the membrane thickness does not change significantly with prolonged repair time. However, this is not directly observable from SEM images, so further characterization tests and pervaporation performance tests were conducted.

[0090] The repaired MOF-801 membrane was then subjected to pervaporation tests. The methanol / dimethyl carbonate pervaporation performance test was also conducted at a test temperature of 50℃ and a methanol / dimethyl carbonate mass ratio of 1 / 9. (Performance graph...) Figure 15 At 6 hours of repair, the membrane's separation performance was not yet optimal; its separation factor was only slightly improved compared to the unrepaired MOF-801 membrane prepared using zirconium chloride as the metal source. However, when the repair time was extended to 12 hours, the membrane's separation factor reached its optimal level, reaching approximately 350 for the methanol / dimethyl carbonate system, with a flux of 4 kg·m³. -2 ·h -1 (Feed temperature: 50℃, feed concentration: 10wt% methanol).

[0091] Flux of MOF-801 membrane (methanol / DMC system) prepared using zirconium oxychloride: 7.2 kg·m -2 ·h -1 Separation factor: 135, feed temperature: 50℃, feed concentration: 10wt% methanol) and unrepaired MOF-801 membrane prepared using zirconium chloride (methanol / DMC system, flux: 5.5 kg·m) -2 ·h -1 Compared to the previous method (separation factor: 250, feed temperature: 50℃, feed concentration: 10wt% methanol), we used a step-by-step strategy to gradually reduce membrane defects, thereby significantly improving the separation factor while the flux decreased accordingly. When the repair time was further extended, the membrane performance did not change significantly, indicating that a repair time of 12 hours was optimal. Once a certain limit was reached, extending the repair time no longer effectively repaired the defects.

[0092] For the MOF-801 membrane with a repair time of 12 hours, we conducted pervaporation performance tests on three organic-organic systems, such as... Figure 16 For methanol / methyl tert-butyl ether, the separation factor reached 520, and the flux was 3.8 kg·m. -2 ·h -1 The methanol / dimethyl carbonate system has a flux of approximately 350 and a throughput of 3.7 kg·m³.-2 ·h -1 The separation factor of the methanol / methyl acetate system reached approximately 220, with a flux of approximately 3.8 kg·m³. -2 ·h -1 .

[0093] Defect characterization

[0094] First, we used NMR to characterize the lattice defects (ligand loss) of three MOF-801 films. The concentration of lattice defects could be obtained based on the peak area ratio of formic acid to fumaric acid ligand. Figure 17 As shown in region A, the MOF-801 film prepared using ZrOCl2 as the metal source exhibits the highest lattice defect concentration, reaching 19%. The lattice defect concentration of the film prepared by changing the metal source to ZrCl4 decreased to 13.5%. This is because during the film preparation process, ZrCl4 needs to undergo hydrolysis to form ZrOCl2, and then form metal clusters to coordinate with organic ligands. The reaction rate is significantly reduced, thus lowering the defect concentration. After soaking in the ligand solution, the lattice defect concentration further decreased (5%). Lattice defects in MOF films include ligand deficiency defects and cluster deficiency defects. We also used PXRD characterization to study the cluster deficiency defects in MOF films. The changes in the intensity of the characteristic peaks around 5°, which represent cluster deficiency defects, can be observed to reflect the changes in cluster deficiency defects within the MOF film. Figure 17 As shown in region B, with the implementation of metal source regulation and post-repair strategies, cluster deficiency defects gradually decrease, which is consistent with the variation pattern of ligand deficiency defects. Furthermore, the reduction in lattice defects leads to the MOF film pore size gradually approaching the intrinsic pore size of the MOF material, resulting in a decrease in surface area and adsorption sites. Therefore, the reduction in lattice defects can also be confirmed by gas adsorption and changes in specific surface area and pore size. Figure 17 The results above (C region, D region) confirm that our proposed metal source regulation and post-repair strategy can effectively reduce the lattice defect concentration, thereby regulating the pore structure of the MOF film and achieving efficient molecular sieving.

[0095] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A MOF separation membrane, comprising a support and a membrane layer, wherein the support is a porous material; The membrane is composed of MOF-801, with a thickness of 1-5 μm, a pore size of 0.5-0.6 nm, and a specific surface area of ​​400-500 m². 2 / g; characterized in that, The lattice defects of MOF-801 in the film are less than 15%; The preparation method of MOF separation membrane includes the following steps: Casting membrane: The support is fixed in the reaction vessel, casting solution is poured in to form a membrane, the sample is cooled and solvent exchanged after completion, and then dried to obtain MOF separation membrane; The casting solution is a mixed solution containing ZrCl4, fumaric acid, an organic solvent, and formic acid. The molar ratio of ZrCl4 to fumaric acid in the casting solution is 1:0.8~1.3, and the volume ratio of the organic solvent to formic acid is 100:20~50. The preparation method also includes a post-repair step, which involves fixing the membrane in a reaction vessel after obtaining the separation membrane, pouring in a repair solution for repair, cooling and solvent exchange of the sample after the repair is completed, and then drying to obtain the post-repaired MOF separation membrane. The repair solution is a mixed solution of an organic solvent containing fumaric acid and formic acid; the volume ratio between the organic solvent and formic acid is 100:20~50.

2. The MOF separation membrane according to claim 1, characterized in that, The support is made of α-Al2O3 and has a porous structure with a pore size of 100-300 nm and a porosity of 20%-45%.

3. The MOF separation membrane according to claim 1, characterized in that, The thickness of the support is 1-5mm and the diameter is 20-40mm.

4. The MOF separation membrane according to claim 1, characterized in that, In the casting process, the film preparation specifically involves drying at 65-100℃ for 8-15 hours.

5. The MOF separation membrane according to claim 1, characterized in that, In the post-repair step, the specific repair process involves drying at 70-90℃ for 9-14 hours.

6. The MOF separation membrane according to claim 1, characterized in that, The lattice defects of MOF-801 in the post-repaired MOF separation membrane layer are below 6%.

7. The application of the MOF separation membrane according to any one of claims 1-6 in the separation of organic azeotropic systems.

8. The application according to claim 7, characterized in that, Organic azeotropic systems refer to alcohol / dimethyl carbonate, methanol / methyl acetate, and methanol / methyl tert-butyl ether azeotropic systems.

Images