Branched Metal-Organic Framework Nanoparticles in Mixed Matrix Membranes and Related Methods

Abstract

Branched metal-organic framework nanoparticles in mixed matrix films and related methods are provided. [Solution] Compositions, devices, and methods of use of mixed matrix membranes comprising metal-organic frameworks for gas separation are generally described. In some embodiments, branched nanoparticles fabricated at least in part from metal-organic frameworks are described. In some embodiments, the morphology and size of the branched nanoparticles are controlled by the presence of chemical modifiers during synthesis. In some embodiments, the branched nanoparticles are evenly distributed in the mixed matrix membrane. In some embodiments, the mixed matrix membrane is configured to separate one or more gases from a gas mixture. In some embodiments, the branched nanoparticles contribute, at least in part, to increasing the permeability, selectivity, and / or plasticization resistance of the mixed matrix membrane.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit under 35 U.S.C. § 119(e) of U.S. Provisional Patent Application No. 62 / 683,516, filed June 11, 2018, and entitled "Branched Metal-Organic Framework Nanoparticles in Mixed Matrix Membranes and Related Methods," which is incorporated by reference in its entirety for all purposes.

[0002] Technical Field The synthesis of nanoparticles comprising metal-organic frameworks and the fabrication of membranes for gas separation comprising said nanoparticles are generally described. [Background technology]

[0003] background The chemical and petrochemical industries consume approximately 30% of global energy and are responsible for 16% of direct CO2 emissions. Approximately half of the energy consumption comes from separation processes, most notably thermal separations such as distillation. In the area of ​​CO2 separation required for carbon capture applications and natural gas purification, amine absorption is the go-to technology. While this process is efficient, it requires high-energy thermal regeneration cycles and the use of toxic amines. With improved material design, membranes can significantly reduce the energy consumption and costs associated with absorption processes and distillation, thereby providing an energy-efficient modular alternative to traditional unit operations.

[0004] To date, all commercially available gas separation membranes are fabricated using polymeric materials, and these gas separation membranes are typically fabricated over the large areas (1,000–500,000 m) required for large-scale separations. 2) and can be formed into thin selective layers (approximately 100 nm). In contrast to these attractive processing advantages, polymeric membranes have several drawbacks. In terms of efficiency and productivity, traditional polymeric materials are often limited by the trade-off between permeability and selectivity. Furthermore, polymeric membranes are often vulnerable to a deleterious process known as plasticization, which is described as the swelling of linear, chain-entangled polymers in the presence of highly adsorptive molecular diluents, disrupting chain packing. Plasticization often increases gas permeability but significantly reduces selectivity. Carbon dioxide is one of the most difficult and commonly observed substances to manipulate in this manner. These drawbacks necessitate the development of novel membrane materials to improve transport properties and reduce plasticization in practical separation processes.

[0005] Mixed-matrix membranes, in which inorganic fillers are dispersed within polymers, have attracted considerable attention to form composite membranes with relatively high mechanical integrity and superior separation performance while still using practical, process-friendly polymer-based systems. Inorganic materials such as various metal oxides, zeolites, and carbons have been incorporated into polymers to achieve a molecular sieve effect, thereby defining the pore dimensions and size of the resulting structures. However, these inorganic materials lack the functionality of organic frameworks, resulting in poor compatibility between the dispersed phase and the polymer. This compatibility issue leads to particle aggregation in the nascent composite membranes and a lack of selectivity in the dispersion pathways. Metal-organic frameworks (MOFs) are an attractive platform of crystalline materials with metal ions or metal clusters bridged by organic ligands, offering a variety of inorganic topologies, exceptionally high porosity and internal surface areas, tunable pore sizes, and chemical properties based on the selection of MOF building blocks. MOFs also offer tunable pore systems due to the organic composition of part of the structure, allowing for the integration of the pores into the polymer matrix. These features may facilitate the incorporation of MOFs into polymers and may be preferable to other inorganic materials needed for efficient gas separation.

[0006] HKUST-1 (also known as Cu(BTC)) is one of the MOFs being investigated for mixed matrix membrane applications. 2+ Constructed from paddlewheel units and benzene-1,3,5-tricarboxylic acid (trimesic acid), it exhibits a twisted cubic boron trioxide topology with two types of pores. These two types of pores have a main pore channel with a diameter of 9 Å, which is surrounded by a tetrahedral pocket with a diameter of 5 Å. One particularly interesting aspect is that HKUST-1 possesses coordinatively unsaturated (open) metal sites that are weakly bound to solvent molecules by coordination during synthesis but can be exposed by solvent exchange and / or heating (so-called activation) processes. Once the weakly bound solvent molecules are removed, the exposed coordination vacancies of the copper metal sites are Lewis acidic and can strongly interact with Lewis basic polar gas molecules, thereby enhancing gas adsorption. For HKUST-1, the activation energy barrier for nucleation is approximately 71.6 kJ mol -1 and the activation energy barrier for growth is approximately 63.8 kJ mol -1 Nucleation is therefore typically much slower than growth, resulting in the formation of micron-sized particles with a broader particle size distribution than other MOFs, which energetically favor nucleation over growth, and thus facilitate the formation of uniformly structured nanoparticles. When typical micron-sized HKUST-1 particles are incorporated into a polymer matrix, they can undergo phase separation, leading to the aggregation of the MOF into a "sleeve-in-cage" morphology, which creates nonselective interfacial voids between the bulk filler and the polymer, resulting in reduced gas selectivity. Therefore, despite HKUST-1's advantageous properties for gas separation, forming defect-free HKUST-1 mixed matrix membranes can be challenging. Summary of the Invention [Means for solving the problem]

[0007] Abstract Compositions, devices, and methods of use of mixed matrix membranes comprising metal-organic frameworks for gas separation are generally described. In some embodiments, branched nanoparticles fabricated at least in part from metal-organic frameworks are described. In some embodiments, the morphology and size of the branched nanoparticles are controlled by the presence of chemical modifiers during synthesis. In some embodiments, the branched nanoparticles are evenly distributed throughout the mixed matrix membrane. In some embodiments, the mixed matrix membrane is configured to separate one or more gases from a gas mixture. In some embodiments, the branched nanoparticles contribute, at least in part, to increasing the permeability, selectivity, and / or plasticization resistance of the mixed matrix membrane. The subject matter of the present invention includes, in some cases, interrelated products, alternative solutions to a particular problem, and / or multiple different uses of one or more systems and / or articles.

[0008] In one aspect, a device is described. In some embodiments, the device comprises a mixed matrix membrane. In some embodiments, the mixed matrix membrane comprises a matrix comprising a polymer and particles, the particles comprising a metal-organic framework comprising a metal ion and a multidentate ligand. In some embodiments, the particles are uniformly distributed throughout the matrix, and convective transport cannot be detected by permeation measurements.

[0009] In one aspect, a composition is described. In some embodiments, the composition comprises branched nanoparticles comprising a metal-organic framework, the metal-organic framework comprising a metal ion and a multidentate ligand. In some embodiments, the branched nanoparticles have a hydrodynamic diameter of 2 μm or less. The composition has an average diameter, an aspect ratio of at least 5, and branches that are 200 nm or less in diameter. In some embodiments, the composition is constructed and arranged as a gas separation membrane.

[0010] In another aspect, a method is described. In some embodiments, the method includes combining a metal salt comprising a metal ion, a multidentate ligand, and a chemical modulator in a liquid to form branched nanoparticles, wherein the branched nanoparticles comprise a metal-organic framework, and the metal-organic framework comprises a metal ion and a multidentate ligand. In some embodiments, the metal salt, the multidentate ligand, and / or the chemical modulator are selected such that the presence of at least 0.1 equivalents of the chemical modulator relative to the concentration of the multidentate ligand increases the aspect ratio of the branched nanoparticles by at least three times relative to particles formed under the same conditions in the absence of the chemical modulator. In embodiments, the method further includes combining the branched nanoparticles with a polymer to form a mixed matrix membrane comprising a polymer network.

[0011] Other advantages and novel features of the present invention will become apparent from the following detailed description of various non-limiting embodiments of the invention when considered in conjunction with the accompanying drawings. In cases where the present specification and a document incorporated by reference include conflicting and / or inconsistent disclosure, the present specification shall control. [Brief explanation of the drawings]

[0012] Non-limiting embodiments of the present invention are described by way of example with reference to the accompanying drawings, which are schematic and are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown is typically designated by a single numeral. For clarity, not every component in every figure is labeled, nor are every component in each illustrated embodiment of the present invention labeled, unless the label is necessary for those skilled in the art to understand the invention. A description of the figures is provided below.

[0013] [Figure 1A-1B] 1A-1B show transmission electron microscopy (TEM) images of (FIG. 1A) branched HKUST-1 particles and (FIG. 1B) bulk HKUST-1 particles according to certain embodiments.

[0014] [Figure 2A] FIG. 2A shows a schematic diagram of an exemplary branched nanoparticle according to certain embodiments.

[0015] [Figure 2B] FIG. 2B shows a schematic diagram of an exemplary branched nanoparticle, according to certain embodiments.

[0016] [Figure 3A] FIG. 3A shows X-ray diffraction (XRD) patterns of branched and bulk HKUST-1 particles according to certain embodiments.

[0017] [Figure 3B] FIG. 3B shows a number-average dynamic light scattering (DLS) plot of bulk and branched HKUST-1 particles according to certain embodiments.

[0018] [Figure 4A-4B] 4A-4B show cross-sectional SEM images of mixed matrix membranes containing bulk and branched nanoparticle HKUST-1 particles according to certain embodiments.

[0019] [Figures 5A-5E] 5A-5E show FIB-SEM images of 30 wt% branched HKUST-1 / 6FDA-DAM MMM at different magnifications according to certain embodiments (FIGS. 5A-5C) and reconstructed 3D images from FIB-SEM tomograms (FIGS. 5D-5E).

[0020] [Figures 6A-6D] 6A-6D show low magnification TEM images of a series of HKUST-1 particles formed using various sodium acetate concentrations according to certain embodiments.

[0021] [Figure 7A-7C] 7A-7C include FIB-SEM images of bulk HKUST-1 / 6FDA-DAM mixed matrix membranes according to certain embodiments.

[0022] [Figures 8A-8C] 8A-8C include FIB-SEM images of branched HKUST-1 / 6FDA-DAM MMMs using various HKUST-1 molecules according to certain embodiments.

[0023] [Figure 9A-9B] 9A-9B are charts showing FT-IR spectra of pure 6FDA-DAM polymer films and bulk HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings, according to certain embodiments.

[0024] [Figures 10A-10B] 10A-10B are charts showing FT-IR spectra of pure 6FDA-DAM polymer films and branched HKUST-1 / 6FDA-DAM MMMs with various HKUST-1 weight loadings, according to certain embodiments.

[0025] [Figures 11A-11D] 11A-11D are charts illustrating certain gas permeabilities as a function of supply pressure according to certain embodiments.

[0026] [Figures 12A-12D] 12A-12D are charts illustrating certain gas permeabilities as a function of gas supply pressure according to certain embodiments.

[0027] [Figures 13A-13B] 13A-13B are charts showing the gas separation performance of various membranes according to certain embodiments.

[0028] [Figures 14A-14B] 14A-14B are charts showing the gas separation performance of two sets of pure 6FDA-DAM polymer thin films and bulk and branched HKUST-1 / 6FDA-DAM MMMs with various HKUST-1 weight loadings according to certain embodiments.

[0029] [Figures 15A-15D] 15A-15D include charts showing normalized CO 2 permeability as a function of CO 2 gas supply pressure for various membranes according to certain embodiments.

[0030] [Figures 16A-16B] 16A-16B include FIB-SEM images of a rhombic dodecahedral ZIF-8 / 6FDA-DAM mixed matrix membrane (FIG. 16A) and a branched ZIF-8 / 6FDA-DAM mixed matrix membrane according to certain embodiments.

[0031] [Figure 17A] 17A-17D are charts showing the gas separation performance of mixed matrix membranes comprising rhombic dodecahedral ZIF-8 nanoparticles (RDZ) and branched ZIF-8 nanoparticles (BZ) according to certain embodiments. [Figure 17B] 17A-17D are charts showing the gas separation performance of mixed matrix membranes comprising rhombic dodecahedral ZIF-8 nanoparticles (RDZ) and branched ZIF-8 nanoparticles (BZ) according to certain embodiments. [Figure 17C] 17A-17D are charts showing the gas separation performance of mixed matrix membranes comprising rhombic dodecahedral ZIF-8 nanoparticles (RDZ) and branched ZIF-8 nanoparticles (BZ) according to certain embodiments. [Figure 17D] 17A-17D are charts showing the gas separation performance of mixed matrix membranes comprising rhombic dodecahedral ZIF-8 nanoparticles (RDZ) and branched ZIF-8 nanoparticles (BZ) according to certain embodiments.

[0032] [Figures 18A-18B] 18A-18B are TEM images of ZIF-8 particles formed in the presence of pyridine (FIG. 18A) and triethylamine (FIG. 18B) according to certain embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0033] Detailed Description Compositions, devices, and methods related to particle synthesis and the incorporation of such particles into mixed matrix membranes for use in gas separation applications are generally described. In some embodiments, a synthesis scheme for forming particles comprising metal-organic frameworks (MOFs) is provided. In some embodiments, the particles are branched nanoparticles. Branched nanoparticles, according to certain embodiments, have a relatively high aspect ratio and a relatively small hydrodynamic radius. In some embodiments, branched nanoparticles have an interconnected structure. In some embodiments, branched nanoparticles have a multidimensional structure. For example, according to certain embodiments, branched nanoparticles have relatively narrow branches connected end-to-end, and these branches are not collinear. The branching dimensions and morphology of branched nanoparticles can contribute to some of their desirable properties, such as limiting aggregation or agglomeration of suspensions in solution or forming multiple contact points in multidimensional structures. Branched nanoparticles can include carboxylate-based MOFs. For example, branched nanoparticles can include MOF HKUST-1, according to certain embodiments. In some embodiments, branched nanoparticles include imidazolate-based MOFs. For example, in some embodiments, the branched MOF nanoparticles comprise MOF ZIF-8.

[0034] In some embodiments, branched nanoparticles can be formed by combining a metal salt, a multidentate ligand, and a chemical regulator in a liquid. For example, according to certain embodiments, a metal salt containing a metal ion (e.g., copper), a multidentate ligand containing at least two carboxylate groups (e.g., trimesic acid), and a chemical regulator containing a carboxylate group (e.g., sodium acetate) are combined in a liquid (e.g., 100% methanol) at room temperature and 1 atm to form branched nanoparticles containing MOFs (e.g., HKUST-1). As another example, in some embodiments, a metal salt containing a metal ion (e.g., zinc), a multidentate ligand containing imidazolate (e.g., 2-methyl-imidazolate), and a chemical regulator containing amines (e.g., triethylamine) are combined in a liquid (e.g., methanol) at room temperature and 1 atm to form branched nanoparticles containing MOFs (e.g., ZIF-8). The presence of a chemical modulator and the conditions under which the metal salt and multidentate ligand are combined can, in part, result in particles with a high aspect ratio, branched morphology, and / or a relatively small hydrodynamic radius. For example, in some embodiments, particles formed under the same conditions as branched particles but in the absence of a chemical modulator or in a different liquid may lack the properties of branched nanoparticles. In other words, for example, the presence of a certain amount of chemical modulator (e.g., 0.1 equivalents relative to the concentration of the multidentate ligand) can increase the aspect ratio of the resulting particles by about three times compared to particles formed under the same conditions. In some embodiments, the metal salt, multidentate ligand, and / or chemical modulator are selected to result in particles with a high aspect ratio, branched morphology, and / or a relatively small hydrodynamic radius.

[0035] The particles may alternatively be incorporated into mixed matrix membranes. Mixed matrix membranes may, in some embodiments, include a matrix comprising a polymer. Throughout this disclosure, the term "polymer matrix" refers to a matrix comprising a polymer, where it should be understood that the matrix is ​​limited to a polymer or that the matrix includes polymers and other materials. Mixed matrix membranes may, in some cases, be used to separate gases. In some embodiments, particles comprising MOFs may be dispersed within the polymer rather than forming voids or undergoing phase separation, as typically observed in mixed matrix membranes comprising MOFs. The particles are uniformly distributed throughout the matrix. In some embodiments, the uniform distribution of particles in the polymer matrix of the mixed matrix membrane is limited to the extent that convective transport of gases through the mixed matrix membrane is observed. In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising MOFs (e.g., HKUST-1). In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising MOFs, where the MOFs are zeolite-like imidazolate structures (e.g., ZIFs, such as ZIF-8). The branched nanoparticles, according to certain embodiments, can form at least a partially percolated network within the mixed matrix membrane. Such a percolated network can improve gas permeation. Such a percolated network can also improve selectivity, particularly under mixed gas conditions where competitive adsorption can affect selectivity. For example, a mixed membrane network comprising a percolated network made from branched nanoparticles of HKUST-1 can exhibit improved selectivity relative to a mixed membrane network comprising HKUST-1 particles that do not form a percolated network. As another example, mixed matrix membrane networks containing percolating networks made from branched ZIF-8 nanoparticles may exhibit improved selectivity relative to mixed membrane networks containing ZIF-8 particles that do not form percolating networks. In some embodiments, the branched nanoparticles may form multiple contact points with the polymer within the polymer matrix. These multiple contact points can improve the durability and structural integrity of the mixed matrix membrane and limit the extent to which the mixed matrix membrane undergoes plasticization during the adsorption of certain gases. For example, in some embodiments, the plasticization pressure point for a gas (e.g., CO2) of a mixed matrix membrane containing branched nanoparticles is relatively high compared to the plasticization pressure point of a membrane made from the same polymer but lacking branched nanoparticles.

[0036] Certain embodiments relate to compositions. In some embodiments, the compositions comprise branched nanoparticles. In some embodiments, the branched nanoparticles comprise metal-organic frameworks (MOFs). For example, Figure 1A shows a transmission electron microscopy image (TEM) of branched nanoparticles comprising MOFs, according to certain embodiments.

[0037] In some embodiments, the MOF comprises a metal ion. The metal ion can generally be any metal ion capable of binding to a ligand. For example, the metal ion can be zinc, aluminum, or magnesium, according to certain embodiments. In some embodiments, the metal ion is zinc. In some embodiments, the metal ion is a transition metal ion. Non-limiting examples of transition metal ions that can be incorporated into the MOF include, according to certain embodiments, iron, cobalt, nickel, manganese, zirconium, or chromium. In some embodiments, the metal ion is copper.

[0038] In some embodiments, the MOF comprises a multidentate ligand. A multidentate ligand generally comprises at least two moieties (e.g., metal ions) capable of binding to a Lewis acid. In some embodiments, the multidentate ligand is an organic molecule. The multidentate ligand may be capable of binding to at least two, at least three, or at least four metals. In some embodiments, the multidentate ligand comprises at least two carboxylate groups. Non-limiting examples of some such groups include benzene-1,3,5-tricarboxylate, benzene-1,4-dicarboxylate, naphthalene-2,6-dicarboxylate, 1,3,5-tris(carboxyphenyl)benzene, 2-terephthalate, 2,5-dioxide-1,4-benzenedicarboxylate, and 5,5'-(9,10-anthracenediyl)di-isophthalate. In other embodiments, the multidentate ligand does not comprise multiple carboxylate groups. For example, in some embodiments, the multidentate ligand comprises an imidazolate. For example, the multidentate ligand may be an imidazolate according to certain embodiments. In some embodiments, the multidentate ligand is 2-methyl-imidazolate.

[0039] The branched nanoparticles can include various classes of MOFs. In some embodiments, The branched nanoparticles comprise carboxylate-based MOFs. Carboxylate-based MOFs generally comprise multidentate ligands containing at least two carboxylate groups. Non-limiting examples of carboxylate MOFs that the branched nanoparticles can comprise include UiO-66, MIL-53, MIL-96, MIL-100, MIL-101, MIL-125, and MOF-74. In some embodiments, the branched nanoparticles comprise the carboxylate-based MOF HKUST-1. The names of the MOFs listed herein are generally known to those skilled in the art of metal-organic frameworks. Other classes of MOFs are also suitable, including zeolitic imidazolate frameworks (ZIFs). The zeolitic imidazolate frameworks can comprise any of a variety of metal ions. For example, in some embodiments, the zeolitic imidazolate frameworks comprise iron, cobalt, copper, or zinc ions, or a combination thereof. In some embodiments, the zeolite-like imidazolate structure comprises zinc ions. For example, in some embodiments, the branched MOF nanoparticles comprise ZIF-8. Modified MOFs, which may include multidentate ligands derivatized by the addition of various functional groups (e.g., hydroxyl groups, alkyl groups, amino groups, halo groups, thio groups, nitro groups, etc.), as well as MOFs in which at least a portion of the metal is replaced with a different metal (e.g., zinc is replaced with magnesium), can also be used in the branched nanoparticles.

[0040] As mentioned above, in some embodiments, the composition comprises branched nanoparticles. Nanoparticles generally refer to particles having at least one cross-sectional dimension of 500 nanometers or less, as measured through the geometric center of the structure. Generally, a nanoparticle is branched if it comprises at least two branches, each of which has a relatively high aspect ratio (e.g., a relatively high length-to-width ratio) connected end-to-end, and at least two of the branches are not collinear. In some embodiments, the length-to-width ratio of the relatively high aspect ratio branches is 5 or greater, 10 or greater, and / or up to 15, 20, or greater. For example, FIG. 2A shows a branched nanoparticle 100 comprising branch 1, branch 2, and branch 3 according to certain embodiments. Branch 1, branch 2, and branch 3 are connected end-to-end and are not all collinear. As another non-limiting example, FIG. 2B shows a branched nanoparticle 200 comprising branch 1, branch 2, branch 3, and branch 4 according to certain embodiments.

[0041] In some embodiments, branched nanoparticles include branches that are nonplanar with respect to one another. For example, referring back to FIG. 2A, according to certain embodiments, branch 1, branch 2, and branch 3 are nonplanar. In some embodiments, branched nanoparticles may be considered interconnected. In some embodiments, interconnected branches can result in a denser MOF framework structure than relatively straight, one-dimensional rods, making devices easier to fabricate. One reason for this is that, according to certain embodiments, there is little need to align interconnected branched particles due to the fact that their branches are relatively randomly oriented, making it less likely that interconnected branched nanoparticles will arbitrarily rotate and disrupt interparticle contact with other interconnected branched nanoparticles. According to certain embodiments, such connections can occur when interconnected branched nanoparticles connect end-to-end with other interconnected branched nanoparticles, thereby forming multidimensional MOF structures. Such structures, according to certain embodiments, can be formed in a continuous phase, such as a polymer matrix or solution. Furthermore, branched nanoparticles, in some embodiments, have closely spaced branch points (i.e., relatively small angles between branches). Such closely spaced branching points allow for dense contact of the MOF-polymers in the mixed matrix membrane due to the relatively small interstitial spaces between the branches.

[0042] In some embodiments, branched nanoparticles have a relatively high aspect ratio. A high aspect ratio can, in some embodiments, help the branched particles form connected structures with long channels. The aspect ratio can be measured by dividing the hydrodynamic radius of the branched nanoparticle (as measured by dynamic light scattering) by the diameter of the branch in the branched particle with the largest diameter (as measured by TEM). In some embodiments, the branched nanoparticles , and have an aspect ratio of at least 5. In some embodiments, the branched nanoparticles have an aspect ratio of at least 8, at least 10, or more. In some embodiments, the aspect ratio is 20 or less, 15 or less, or less. Combinations of the above ranges are possible. For example, in some embodiments, the branched nanoparticles have an aspect ratio in the range of 5 to 20.

[0043] In some embodiments, branched nanoparticles have a relatively small hydrodynamic diameter. Hydrodynamic diameter refers to the diameter of a hypothetical hard sphere that diffuses in the same manner as the particle being measured. As described below, a relatively small hydrodynamic radius is, in part, responsible for preventing the branched nanoparticles from aggregating or clumping in suspension and allowing them to be well-dispersed and uniformly distributed in membrane structures. As described above, the hydrodynamic radius of a particle can be measured using dynamic light scattering (DLS). More specifically, the hydrodynamic radius can be measured by suspending branched nanoparticles in a methanol (MeOH) solution, sonicating to uniformly disperse the branched nanoparticles, and then performing DLS measurements at 35°C using a Zetasizer Nano S90 (Marvern). In some embodiments, the branched nanoparticles have a hydrodynamic diameter of 2 μm or less. In some embodiments, the branched nanoparticles have a hydrodynamic diameter of 1 μm or less, 500 nm or less, 400 nm or less, 300 nm or less, 200 nm or less, 150 nm or less, 100 nm or less, or less. In some embodiments, the branched nanoparticles have a hydrodynamic diameter of at least 20 nm, at least 30 nm, or more. Combinations of the above ranges are possible. For example, in some embodiments, the branched nanoparticles have a hydrodynamic diameter of at least 20 nm and less than 2 μm.

[0044] As mentioned above, branched nanoparticles have relatively small branch diameters. Such diameters can be measured by TEM. The small branch diameters may contribute, in part, to the formation of multidimensional structures with good transmission properties. In some embodiments, branched nanoparticles have branch diameters of 200 nm or less. In some embodiments, branch diameters are 100 nm or less, 50 nm or less, 30 nm or less, 20 nm or less, 15 nm or less, 10 nm or less, or less. In some embodiments, branch diameters are at least 2 nm, at least 5 nm, or more. Combinations of the above ranges are possible. For example, in some embodiments, branched nanoparticles have branch diameters of at least 2 nm and less than 200 nm. In some embodiments, at least 50% (e.g., at least 60%, at least 75%, at least 90%, or at least 95%) of branched nanoparticles have branch diameters of 200 nm or less. In some embodiments, at least 50% (e.g., at least 60%, at least 75%, at least 90%, or at least 95%) of the branched nanoparticles have branch diameters of 100 nm or less, 50 nm or less, 30 nm or less, 20 nm or less, 15 nm or less, 10 nm or less, or less. In some embodiments, at least 50% (e.g., at least 60%, at least 75%, at least 90%, or at least 95%) of the branched nanoparticles have branch diameters of at least 2 nm, at least 5 nm, or more. Combinations of the above ranges are possible. For example, in some embodiments, at least 50% (e.g., at least 60%, at least 75%, at least 90%, or at least 95%) of the branched nanoparticles have branch diameters of at least 2 nm and less than 200 nm. The percentage of branched nanoparticles having branches with diameters in the above ranges can be determined using TEM.

[0045] Particles comprising MOFs can be formed, in some embodiments, by combining precursors in solution. In some embodiments, particles comprising MOFs are formed by combining a metal salt, a multidentate ligand, and a chemical modifier in a liquid. For example, For example, according to certain embodiments, copper nitrate is combined with benzene-1,3,5-carboxylic acid (trimesic acid) and sodium acetate in methanol to form branched nanoparticles comprising HKUST-1. As another example, in some embodiments, zinc nitrate hexahydrate (Zn(NO3)2·6H2O), (or any of Zn(acac)2, Zn(ClO4)2, ZnSO4, Zn(OAc)2), ZnCl2, ZnBr2, ZnI2, or combinations thereof), 2-methyl-imidazole, and triethylamine in methanol to form branched nanoparticles comprising ZIF-8.

[0046] In some embodiments, the presence of a certain amount of a chemical modulator can affect the size and / or morphology of the resulting MOF particles. A chemical modulator generally refers to an additive to a reaction mixture that, when present during a chemical reaction to form particles, alters the size and / or morphology. In some embodiments, a chemical modulator can compete with a multidentate ligand to bind to the metal ion of a metal salt, thereby slowing the rate of particle growth. In some embodiments, a chemical modulator can promote particle nucleation, thereby increasing the number of individual particles formed during a reaction. In some embodiments, a chemical modulator can introduce defects into the crystalline structure of a metal-organic framework, thereby affecting the surface energy of the crystal faces and the resulting morphology. In some embodiments, a chemical modulator can affect the proton activity in solution (i.e., affect the acid-base chemistry in the reaction mixture).

[0047] In some embodiments, the chemical regulator is a salt comprising a cation and an anion. In some embodiments, the anion has at least one carboxylate group. For example, in some embodiments, the anion is acetate. For example, in some embodiments, the chemical regulator is sodium acetate. In some, but not all, embodiments, the chemical regulator is sodium formate or sodium propionate.

[0048] In some embodiments, the chemical modulator comprises an amine. In some embodiments, the chemical modulator is an amine. The amine of the chemical modulator can be any of a variety of suitable amines. For example, in some embodiments, the amine is a primary amine. In some embodiments, the amine is a secondary amine. In some embodiments, the amine is a tertiary amine. The amine can comprise any of a variety of suitable substituents. For example, in some embodiments, the chemical modulator comprises an amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 Independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. As an example, in some embodiments, the chemical modulator is or includes triethylamine. Other non-limiting examples of chemical modulators include trimethylamine, diethylamine, tri-n-butylamine, and n-butylamine.

[0049] As previously mentioned, in some instances, the chemical modulator may affect the acid-base chemistry of the reaction mixture. In some such cases, the chemical modulator may affect the acid-base chemistry of the reaction mixture within a certain range of pK a For example, in some embodiments, the pK of the chemical modulator that undergoes a chemical reaction in the liquid a is the pK of the multidentate ligand moiety that binds to the metal ion in the metal-organic framework. aFor example, in some embodiments where the multidentate ligand contains at least two carboxylate groups (and in the resulting MOF the carboxylate groups bind to the metal ion) and the solution in which synthesis occurs is methanol, the pK of the chemical modulator is a is higher than the carboxylate group of the multidentate ligand in methanol. As another example, in some embodiments where the multidentate ligand comprises an imidazolate group and the solution in which synthesis occurs is methanol, the pK of the chemical modulator is a is the pK of the imidazolate ligand in methanol a In some embodiments, the pKa of the chemical modulator is relatively high. For example, in some embodiments, the pKa of the chemical modulator (e.g., acetate, amine, etc.) in methanol is a is 3.0 or higher In some embodiments, the pK in methanol of the chemical modulator is greater than 4.0, greater than 5.0, greater than 6.0, greater than 7.0, greater than 7.5, greater than 8.0, greater than 8.0, greater than 9.0, or greater. a is 15.0 or less, 12.0 or less, 11.5 or less, 11.0 or less, 10.5 or less, 10.0 or less, or less. Combinations of these ranges are possible. For example, in some embodiments, the pK in methanol of the chemical modulator is a is greater than or equal to 3.0 and less than or equal to 15.0, greater than or equal to 5.0 and less than or equal to 12.0, or greater than or equal to 8.0 and less than or equal to 11.5.

[0050] In some embodiments, the metal salts, multidentate ligands, and / or chemical modulators that are combined in a liquid to form branched nanoparticles comprising MOFs are selected to impart certain properties to the resulting branched nanoparticles. In some embodiments, the presence of at least 0.1 equivalents of chemical modulator relative to the concentration of multidentate ligand increases the aspect ratio of the branched nanoparticles by at least three-fold relative to particles formed under the same conditions in the absence of the chemical modulator. In some embodiments, the metal salts, multidentate ligands, and / or chemical modulators are selected such that the presence of at least 0.1, at least 0.15, at least 0.2, at least 0.3, at least 0.5, at least 1.0, at least 2.0, at least 5.0, or more equivalents of chemical modulator relative to the concentration of multidentate ligand increases the aspect ratio of the branched nanoparticles by at least three-fold, at least four-fold, at least five-fold, at least ten-fold, or more relative to particles formed under the same conditions in the absence of the chemical modulator. Those skilled in the art having the benefit of this disclosure will understand how to select metal salts, multidentate ligands, and / or chemical modulators to achieve the above effects. For example, in some embodiments, metal salts and multidentate ligands are selected to form a desired MOF, and the chemical modulator is selected based on the relative pK of the chemical modulator compared to the binding moiety of the multidentate ligand. a or the ability to coordinate with the metal ion of the metal salt.Reactions carried out under the same conditions are generally considered to be reactions carried out in the same liquid (e.g., the same solvent mixture) at the same temperature and pressure for the same time using the same concentrations of reagents and the same mixing conditions (e.g., the same volume of reaction mixture and the same stirring speed), unless otherwise specified.For example, according to a specific embodiment, when copper nitrate trihydrate is mixed with trimesic acid in methanol at 23°C in the presence of 0.1 equivalents of sodium acetate relative to the concentration of trimesic acid, the aspect ratio of the resulting branched nanoparticles of HKUST-1 is at least three times higher than the aspect ratio of HKUST-1 particles formed by mixing the same concentrations of copper nitrate and trimesic acid in methanol at 23°C in the absence of 0.1 equivalents of sodium acetate.

[0051] In some embodiments, the liquid into which the metal salt, multidentate ligand, and chemical modifier are mixed contains a relatively high volume percent of methanol. This high volume percent of methanol may contribute, at least in part, to the surprising size and morphology of the MOF-containing branched nanoparticles described herein. In some embodiments, the liquid is at least 99% methanol, or 100% methanol. It should also be understood that in some embodiments, other solvent mixtures are used. For example, in some embodiments, the liquid from which the branched nanoparticles are formed is a solvent mixture containing less than 100% (e.g., less than 100% by volume), less than 99%, or less methanol. In some embodiments, the liquid is a solvent mixture that is methanol-free. While not wishing to be bound by any particular theory, differences in the miscibility of the protonated and deprotonated forms of components of the reaction mixture (such as the multidentate ligand) may contribute, at least in part, to the formation of the branched MOF nanoparticles.

[0052] In some embodiments, the precursors are combined under relatively mild conditions. For example, in some embodiments, the liquids are combined at temperatures of 15° C. or greater and 60° C. or less. The mixture is mixed for at least 10 minutes at a temperature below 100°C. Such mild conditions may allow for easy production and processing of branched nanoparticles, including MOFs.

[0053] In some embodiments, the branched nanoparticles are non-agglomerated. That is, when suspended in a continuous phase or polymer matrix, such as a liquid suspension, the branched nanoparticles do not aggregate, according to certain embodiments. In some embodiments, the branched nanoparticles do not agglomerate or aggregate in suspension in a continuous phase for at least one month. For example, in some embodiments, suspensions of HKUST-1 branched nanoparticles do not aggregate, clump, or settle for at least one month. As described below, the ability of branched nanoparticles to avoid aggregation or agglomeration in polymer suspensions may contribute, without being bound by theory, to the uniform distribution of particles in mixed matrix films.

[0054] In some embodiments, the branched MOF nanoparticles comprise a compound associated with the MOF. In some such embodiments, a chemical modulator used during the synthesis of the branched nanoparticles is associated with the MOF after the branched nanoparticles are formed. The compound (e.g., chemical modulator) may be associated with the MOF in any of a variety of ways, including, but not limited to, adsorption (e.g., physisorption, chemisorption), or coordination with a portion of the MOF. For example, in some embodiments, the compound (e.g., chemical modulator) is a Lewis base that coordinates with a Lewis acid of the MOF (e.g., a metal ion of the MOF). In some embodiments, the branched nanoparticles comprise any of the chemical modulators described above, or a combination thereof. In some embodiments, the branched nanoparticles comprise a carboxylate group associated with the metal-organic framework (e.g., an acetate group associated with the MOF). In some embodiments, the branched nanoparticles comprise an amine. In some embodiments, the amine is associated with the MOF of the branched nanoparticle (e.g., coordinates to a metal ion of the MOF). The amine can be any of the amines described above. For example, in some embodiments, the branched nanoparticles comprise an amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 The branched nanoparticles are independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. In some embodiments, the branched nanoparticles comprise triethylamine. The presence of the compound associated with the branched nanoparticles can be determined using analytical techniques such as, for example, X-ray photoelectron spectroscopy (XPS) or thermogravimetric analysis-mass spectrometry (TGA-MS).

[0055] In some embodiments, a device is described. In some embodiments, the device comprises a mixed matrix membrane. In some embodiments, the mixed membrane matrix comprises a matrix containing a polymer ("polymer matrix"). A variety of polymers can be used for the mixed matrix membrane. Non-limiting examples of suitable polymers include polyimide, polycarbonate, polysulfone, cellulose triacetate, polyphenylene oxide, or derivatives or combinations thereof. In some embodiments, the polymer is a polymer with inherent porosity. In some embodiments, the polymer has the following structure: [ka] The polyimide is 6FDA-DAM having the following structure.

[0056] As previously mentioned, a matrix comprising a polymer ("polymer matrix") refers to a matrix that is limited to polymers or a matrix that includes polymers and other materials. Other materials that may be present in the polymer matrix in some embodiments include gold. Examples of polymers include, but are not limited to, group ions, residual chemical modulator molecules, additives (such as plasticizers, zeolites, carbon, and silica particles, or combinations thereof). In some embodiments, the polymer is present in the matrix in an amount of 50 weight percent (wt%) or more, 75 wt% or more, 90 wt% or more, 95 wt% or more, 99 wt% or more, or more, based on the weight of the matrix. In some embodiments, essentially all of the matrix is ​​composed of polymer. In some embodiments, all of the matrix is ​​composed of polymer.

[0057] In some embodiments, the mixed matrix membrane comprises particles comprising a metal-organic framework. In some embodiments, the particles are present in the mixed matrix membrane at a certain weight percentage. In some embodiments, the weight percentage of the particles in the mixed matrix membrane is at least 10%, at least 20%, at least 30%, or more. In some embodiments, the weight percentage of the particles in the mixed matrix membrane is up to 50%. In some embodiments, the weight percentage of the particles in the mixed matrix membrane is up to 40% or less. Combinations of the above ranges are possible. For example, in some embodiments, the weight percentage of the particles in the mixed matrix membrane is at least 5% and up to 50%. In some embodiments, the mixed matrix membrane comprises particles comprising HKUST-1. In some embodiments, the mixed matrix membrane comprises at least 90% HKUST-1 particles. In some embodiments, the mixed matrix membrane comprises particles comprising ZIF-8. In some embodiments, the mixed matrix membrane comprises at least 90% ZIF-8 particles. In some embodiments, the particles of the mixed matrix membrane are branched MOF nanoparticles described herein (e.g., branched HKUST-1 nanoparticles, branched ZIF-8 nanoparticles).

[0058] As previously mentioned, mixed matrix membranes, according to certain embodiments, can separate a portion of a first gas from a gas mixture containing the first gas and a second gas. Several such separations can occur when a gas mixture passes through the mixed matrix membrane. For example, according to certain embodiments, the mixed matrix membrane can be configured to separate CO from CH or CO from N. In some embodiments, the mixed matrix membrane can perform separations involving hydrocarbons. In some embodiments, the mixed matrix membrane can separate olefins from paraffins. For example, in some embodiments, the mixed matrix membrane can separate ethylene from ethane or propylene from propane.

[0059] In general, it is beneficial for mixed matrix membranes to have relatively high selectivity and relatively high permeability or permeation during gas separation processes. The permeability of a gas through a membrane generally relates to the flow of the gas through the membrane divided by the thickness of the membrane. The selectivity of a first gas over a second gas in a gas mixture generally refers to the ratio of the permeability of the first gas to the permeability of the second gas. A common problem with mixed matrix membranes containing MOF particles is either low permeability or low selectivity. Such problems are often due to phase separation between the MOF particles and the polymer network or polymer matrix. An example of phase separation between a polymer and MOF particles other than the branched nanoparticles described herein is shown in the TEM of Figure 4B (described in Example 1 below), where the polymer phase forms an upper layer and the MOF particle phase forms a lower layer. Such phase separation can also result in voids between the MOF and the polymer matrix. Examples of such voids are shown in Figures 7A-7C (described in Example 1 below). Phase separation and the presence of voids in mixed matrix membranes reduces selectivity.

[0060] In some embodiments, the particles comprising the metal-organic framework are uniformly distributed in the mixed matrix film. In some such embodiments, the upper layer of the mixed matrix film and the lower layer of the mixed matrix film are uniformly distributed in the mixed matrix film. The bottom layer of the mixed matrix membrane has an identical, uniform distribution of particles and polymer. Figure 4B shows an example of such a uniform distribution according to certain embodiments. This can be determined by using ATR-FTIR techniques (the use of which is described in the Examples below) to compare spectra acquired in the top and bottom layers of the mixed matrix membrane (or vice versa). If the ATR-FTIR spectra show no resolvable differences, the particles can be considered to be uniformly distributed.

[0061] In some embodiments, the mixed matrix membrane has a relatively small amount of voids. The voids in the mixed matrix membrane tend to allow gas to permeate by convection. Convective transport is observed when the gas permeability increases with increasing pressure of the unplasticized gas. This tendency is characteristic of pressure-dependent pore flow due to interfacial defects in the polymer-metal-organic framework that have effective dimensions larger than the effective dimensions of the pores in the metal-organic framework. In some embodiments, convective transport cannot be detected by permeation measurements.

[0062] In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising a metal-organic framework. For example, in some embodiments, the particles of the mixed matrix membrane are any branched nanoparticles comprising a metal-organic framework as described herein. In some embodiments, at least 50 weight percent (wt%), at least 75 wt%, at least 90 wt%, at least 95 wt%, at least 99 wt%, or all of the metal-organic framework particles in the mixed matrix membrane are branched nanoparticles as described herein, based on the weight of the particles in the membrane. In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising carboxylate-based branched nanoparticles. In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising HKUST-1. In some embodiments, the mixed matrix membrane comprises branched nanoparticles comprising ZIF-8.

[0063] The presence of metal-organic framework particles having the branched architecture described herein at certain weight percentages can improve the gas separation performance of mixed metal membranes, according to certain embodiments (see below and Example 1 below). In some embodiments, the weight percentage of branched nanoparticles in the mixed matrix membrane is at least 5%. In some embodiments, the weight percentage of branched nanoparticles in the mixed matrix membrane is at least 10%, at least 20%, at least 30%, or more. In some embodiments, the weight percentage of branched nanoparticles in the mixed matrix membrane is up to 50%. In some embodiments, the weight percentage of branched nanoparticles in the mixed matrix membrane is up to 40% or less. Combinations of the above ranges are possible. For example, in some embodiments, the weight percentage of branched nanoparticles in the mixed matrix membrane is at least 5% and up to 50%.

[0064] In some embodiments, the branched nanoparticles form an at least partially leachable network. In some embodiments, the mixed matrix membrane comprises an at least partially leachable network. The at least partially leachable network can help improve the performance of the mixed matrix membrane during gas separation. Figures 5A-5E show examples of such leachable networks according to certain embodiments.

[0065] In some embodiments, mixed matrix membranes are resistant to swelling or plasticization. Plasticization refers to the swelling of the polymer network or polymer matrix upon adsorption of a particular gas. Membrane plasticization can be monitored by monitoring the permeability of gas through the membrane as a function of gas pressure. At low pressures, gas permeability decreases with increasing pressure. When plasticization occurs, a certain pressure point, referred to herein as the plasticization pressure point, is reached where gas permeability begins to increase with increasing pressure due to membrane swelling. Such swelling generally reduces selectivity. In some embodiments, mixed matrix membranes are resistant to swelling. The plasticization pressure point of a matrix membrane for a gas is at least 1.5 times that of a substantially identical membrane lacking said particles. In some embodiments, the plasticization pressure point of a mixed matrix membrane for a gas (e.g., CO2) is at least 1.5 times, at least 2.0 times, at least 5.0 times, at least 10 times, at least 20 times, and / or up to 50 times that of a substantially identical membrane lacking particles. In some embodiments, no plasticization pressure point is observed even with gases at pressures above 750 psi. In some embodiments, the gas is CO2. In some embodiments, the gas is a hydrocarbon. Plasticization pressure testing involves testing multiple pressure points, first below and then above the observed plasticization pressure, with a time lag of at least 6 × θ (where θ = 1) at each pressure. 2 / (6 * In this equation, l is equal to the thickness of the mixed matrix membrane, and D is the experimentally determined effective diffusion coefficient of CO2 in the membrane.

[0066] The following examples are intended to illustrate certain embodiments of the present invention, but do not exemplify the full scope of the invention. [Example]

[0067] Example 1 Synthesis of HKUST-1 particles HKUST-1 particles with various structures and sizes were synthesized using methanol (MeOH) solvent and a carboxylate chemical modifier (sodium acetate) at room temperature for 1 hour. Figure 1B shows a TEM image of typical HKUST-1 particles synthesized without chemical modifiers. These particles are micron-sized (1-2 μm cross-section) and octahedral in shape, forming large agglomerates. On the other hand, the introduction of a modifier (0.3 molar equivalents relative to trimesic acid) during HKUST-1 synthesis resulted in branched structures, as shown in Figure 1A. Figure 1A shows that these HKUST-1 particles exhibited end-to-end distances of approximately 100-300 nm in length and width, with branches of approximately 10-30 nm, corresponding to an aspect ratio of approximately 5-20. Branched HKUST-1 particles were dispersed from the dried particle solution onto a TEM copper grid. Several meshes on the TEM grid showed well-dispersed branched HKUST-1 nanoparticles without any significant agglomerations. Furthermore, the TEM images shown in Figures 6A–6D demonstrate the effect of systematically varying the chemical modulator concentration (0, 0.1, 0.3, and 0.5 molar equivalents relative to trimesic acid, which correspond to Figures 6A, 6B, 6C, and 6D, respectively) on the size and structure of HKUST-1 particles. As the concentration of the chemical modulator increased, the size of HKUST-1 particles decreased until reaching a minimum size at 0.3–0.5 molar equivalents relative to trimesic acid. This size also corresponds to a relatively high aspect ratio for the branched structure. To evaluate the effect of size and structure on HKUST-1 particles, micron-sized HKUST-1 particles synthesized without sodium acetate were selected as a bulk HKUST-1 reference sample. Figure 1B shows a TEM image of bulk HKUST-1 particles. Figure 1A shows a TEM image of HKUST-1 nanoparticles synthesized with 0.3 equivalents of sodium acetate relative to trimesic acid as a branched HKUST-1 sample. Figure 3A shows the XRD patterns for bulk and branched HKUST-1 particles. Both bulk and branched HKUST-1 particles showed the same crystalline phase of HKUST-1, but the peak width for branched HKUST-1 nanoparticles was broader, indicating a smaller crystallite size.Applying the Scherrer equation to the main Bragg peak of the (222) plate, the crystallite size of the branched HKUST-1 nanoparticles was estimated to be approximately 16 nm. This feature size corresponds to the width of the branched structure. XRD results indicated that increasing the sodium acetate concentration from 0 to 0.1 to 0.3 to 0.5 molar equivalents during HKUST-1 synthesis decreased the branch size of the crystallites, as suggested by the broadening of the peak width, while still maintaining the HKUST-1 crystalline structure. Figure 3B shows the particle distribution of bulk and branched HKUST-1 particles measured by DLS using a diluted suspension in MeOH as the solvent. Bulk HKUST-1 particles were observed by TEM. While the DLS measurements closely matched the 1-2 μm particle size range predicted, the HKUST-1 nanoparticles were much smaller, with diameters ranging from approximately 100 to 300 nm. It should be noted that the particle size distribution calculated by DLS measurements was based on the hydrodynamic radius of a sphere using the Stokes-Einstein equation, which means that the DLS instrument cannot account for the particle aspect ratio. Therefore, the DLS instrument assumed that the longer length of the particle was the radius of the hydrodynamic sphere that formed the skeleton. Therefore, the particle diameters estimated by DLS, ranging from 100 to 300 nm, strictly applied to the hydrodynamic radius of the interconnected branched structures. DLS measurements showed that varying the sodium acetate concentration during HKUST-1 synthesis altered the particle size distribution of HKUST-1 particles. Further addition of sodium acetate reduced the particle size distribution until the branched structures had the smallest particle size (100-200 nm). Overall, the branched HKUST-1 nanoparticles formed with sodium acetate as a modifier had branch widths of 10–30 nm and hydrodynamic diameters of 100–300 nm, as confirmed by TEM, XRD, and DLS analyses. These results suggest that the aspect ratios of these multidimensional MOFs are approximately 5–20.

[0068] Further characterization studies were performed on a series of HKUST-1 particles to investigate the effect of chemical modifier concentration on the synthesis and chemical structure. Fourier transform infrared (FT-IR) spectra were acquired for the independent reagents (i.e., copper nitrate trihydrate and trimesic acid), the chemical modifier (sodium acetate), and the product (branched HKUST-1 nanoparticles). The absorption bands of branched HKUST-1 nanoparticles were observed at 1649, 1448, and 1373 cm. -1 This indicates the coordination between the carboxylate group and the copper metal site (COO-Cu), suggesting that the HKUST-1 structure was successfully formed. Unreacted trimesic acid was thoroughly washed away by multiple washing steps, at least three times with fresh MeOH solvent and then at least three times with THF solvent. Furthermore, approximately 490 cm -1 The absorption band at approximately 1550 cm indicated a strong bond between the copper metal ion and the organic ligand via the bridging carboxylate groups. To demonstrate the effect of chemical modifiers on HKUST-1 synthesis, FT-IR spectra were acquired for a series of HKUST-1 particles synthesized using various sodium acetate concentrations. The FT-IR spectra of HKUST-1 synthesized using 0, 0.1, 0.3, and 0.5 molar equivalents of sodium acetate were all similar in appearance and all exhibited the characteristic HKUST-1 peaks described above. However, the peak at approximately 1550 cm -1 The intensity of the peak at 1550 cm gradually increased as the equivalent amount of sodium acetate increased. -1 The peak intensity at 177 cm increased. This change suggested that the acetate concentration from the sodium acetate modifier in the structure increased, creating defects in the crystal lattice of HKUST-1. These defects may be due to the substitution of a monodentate ligand (acetate ion) for the tridentate ligand (trimesic acid) characteristic of the HKUST-1 structure. As the sodium acetate concentration increased during HKUST-1 synthesis, more defects due to sodium acetate substitution in HKUST-1 were observed. Nevertheless, the structure of HKUST-1 was maintained. Raman spectra were also acquired for a series of HKUST-1 particles formed using various sodium acetate concentrations, revealing the existence of all chemical bonds (Cu-Cu (177 cm)) in the HKUST-1 particles.-1 ), CH (745 and 829 cm -1 ), C=C (1005 and 1614 cm -1 ), and C-O2 (1462 and 1542 cm -1 ) vibrational forms) were located at the same position regardless of the sodium acetate concentration. However, when the Raman spectrum was expanded, the C-O2 functional group peak in the branched HKUST-1 nanoparticles was located at 1462 cm -1 From 1456cm -1 A small peak shift to 0.025 was detected. This shift suggested that as the sodium acetate concentration during HKUST-1 synthesis increased, the position of the C-O2 functional group peak of the HKUST-1 particles approached the previously mentioned position of pure sodium acetate, thereby suggesting that only trimesic acid in the branched HKUST-1 structure was replaced by sodium acetate. Of particular interest, visual inspection of a series of HKUST-1 particles formed using various sodium acetate concentrations showed that they exhibited slightly different colors. The color difference was due to the Cu 2+ This indicates differences in particle size around the metal center and / or differences in the surrounding coordination environment. The branched HKUST-1 nanoparticles have a lower refractive index than the bulk HKUST-1 particles, resulting in a lower light reflectance. As the refractive index decreases with increasing particle size, these bulk particles appear dark blue. Furthermore, the branched HKUST-1 nanoparticles contain defects, which can cause Cu 2+ The coordination environment around the metal center can also affect the color of the sample. Diffuse reflectance UV-Vis spectra were acquired for a series of HKUST-1 particles formed with various sodium acetate concentrations (0, 0.1, 0.3, and 0.5 molar equivalents of HKUST-1) when prepared using a powder pellet accessory. Overall, the diffuse reflectance UV-Vis absorption of HKUST-1 particles has two characteristic ranges: from the oxygen in the tricarboxylate group to the Cu 2+ Ligand-to-metal charge transfer (LMCT) to metal ions (wavelengths less than 500 nm) and Cu 2+dd transitions around the metal center (wavelengths above 500 nm). Compared to typical bulk HKUST-1 particles, branched HKUST-1 nanoparticles exhibited an increased peak shift in the wavelength range below 500 nm and a decreased peak shift in the wavelength range above 500 nm. Both peak shifts indicate defects in the HKUST-1 structure, such as substituted acetates and loss of linkers.

[0069] Table 1 shows the results of N adsorption / desorption isotherms for a series of HKUST-1 particles formed with various sodium acetate concentrations. The BET surface area of ​​the bulk HKUST-1 sample is approximately 1140 m 2 When sodium acetate was introduced in the synthesis, the BET surface area of ​​the HKUST-1 particles was 1140 m 2 / g to 572m 2 / g. This decrease in the BET surface area of ​​the HKUST-1 particles is due to a decrease in the micropore volume, indicating that certain defects formed may reduce the exposed surface area of ​​the branched HKUST-1 nanoparticles compared to their bulk counterparts. These N adsorption isotherms also correlate with the decrease in crystalline intensity determined by XRD, indicating that the bulk HKUST-1 particles have higher crystalline intensity than the branched HKUST-1 nanoparticles.

[0070] [Table 1]

[0071] Thermogravimetric analysis (TGA) curves of bulk and branched HKUST-1 particles were determined by two different methods to observe the thermal behavior of as-prepared and activated samples. First, as-prepared bulk and branched HKUST-1 powder samples were loaded into a TGA pan and tested under air at a heating rate of 20 °C / min from room temperature to 800 °C. The branched HKUST-1 nanoparticles lost 20 wt% weight before decomposition, compared to a 30 wt% weight loss for the bulk sample. The difference in solvent weight loss was attributed to reduced solvent capacity due to defects in the branched HKUST-1 nanoparticles. For the activation and decomposition procedure, the bulk and branched HKUST-1 samples were activated at 150 °C for 2 h under a nitrogen atmosphere. After cooling to room temperature, a second TGA was performed in which the samples were gradually heated to 800 °C at the same rate as before. After activation, no significant weight change was observed up to 300 °C, indicating complete solvent removal during the activation process. At 300 °C, both bulk and branched HKUST-1 particles decomposed and converted to copper oxide. It should be noted that the final masses of bulk and branched HKUST-1 particles after annealing were used to back-calculate the actual weight loading of HKUST-1 particles in the mixed matrix films.

[0072] Preparation of mixed matrix membranes The dispersion stability of MOF particles in solution is a key factor determining whether they can be incorporated into a polymer matrix in a uniformly distributed state. Therefore, prior to fabricating mixed-matrix membranes, we performed dispersion tests on a series of HKUST-1 suspensions in MeOH (containing 0, 0.1, 0.3, and 0.5 molar equivalents of sodium acetate). These results demonstrated that all HKUST-1 suspensions showed good dispersion immediately after vigorous stirring and sonication. However, when the suspensions were left on the benchtop for several hours, bulk HKUST-1 began to settle to the bottom of the glass vial. After one day, all bulk HKUST-1 particles settled to the bottom of the glass vial, while all branched HKUST-1 suspensions maintained a uniform distribution. Particularly interestingly, while HKUST-1 suspensions require sufficient dispersion for up to 12 hours to incorporate uniformly dispersed MOFs into mixed-matrix membranes, branched HKUST-1 suspensions maintained a uniform dispersion for more than two weeks. Bulk and branched HKUST-1 mixed matrix membranes with various HKUST-1 weight loadings (10, 20, and 30 wt%) were fabricated using a solvent evaporation method, allowing evaporation for approximately 6 hours to ensure uniform film thickness and homogeneous filler distribution in the composite. For the fabrication of the mixed matrix membranes, 6FDA-DAM polyimide was used as the matrix. Detailed characterization results (i.e., purity and molecular weight) of the polymer are shown in Table 2. Nuclear magnetic resonance (NMR) spectra confirmed the assignment and purity of the 6FDA-DAM polyimide matrix. The actual HKUST-1 weight loadings in the mixed matrix membranes derived from the TGA curves were all close to the target weight loadings, but it should be noted that the uncertainty for independently prepared samples from different batches was ±1%. To demonstrate the dispersion and network formation of bulk and branched HKUST-1 in the 6FDA-DAM polymer, Figure 4 shows cross-sectional images of bulk (Figure 4A) and branched (Figure 4B) HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings. The mixed matrix membranes were fractured into two pieces using liquid nitrogen, and the cross sections were observed using FE-SEM.Figure 4A shows a cross-sectional FE-SEM image of a bulk HKUST-1 / 6FDA-DAM mixed matrix membrane with a 20 wt% bulk HKUST-1 loading. Membranes with increasing loadings of bulk HKUST-1 particles, from 10 to 30 wt%, were fabricated and observed using FE-SEM. The bulk HKUST-1 / 6FDA-DAM mixed matrix membrane exhibited an undesirable phase-separated structure into a MOF-rich phase and a polymer-rich phase, which was attributed to rapid particle sedimentation of bulk HKUST-1 during solvent evaporation. The thickness of the MOF-rich layer in the mixed matrix membrane increased with increasing bulk HKUST-1 weight loading. The thickness of the MOF-rich layer in the 30 wt% bulk HKUST-1 / 6FDA-DAM mixed matrix membrane was 20–25 μm, which was nearly half of the total membrane thickness (40–50 μm). Figure 4B shows an FE-SEM image of a branched HKUST-1 / 6FDA-DAM mixed matrix membrane with a 20 wt% branched HKUST-1 particle loading. Membranes with 10-30 wt% branched HKUST-1 loading were fabricated and observed using FE-SEM. Compared to the bulk HKUST-1 / 6FDA-DAM mixed matrix membrane, the branched HKUST-1 nanoparticles were uniformly distributed in the 6FDA-DAM polymer matrix, resulting in a uniform morphology throughout the thin film. No significant agglomeration of branched HKUST-1 particles was observed, even within the 30 wt% branched HKUST-1 sample. Furthermore, because the particle size and dispersity differ between bulk and branched HKUST-1, the cast HKUST-1 / 6FDA-DAM mixed matrix membranes exhibited different appearances. The bulk HKUST-1 / 6FDA-DAM mixed matrix membrane was opaque due to the high reflectivity caused by the large HKUST-1 particle agglomerations, whereas the branched HKUST-1 nanoparticles were opaque due to the high reflectivity caused by the large HKUST-1 particle agglomerations. The T-1 / 6FDA-DAM mixed matrix membrane was transparent due to the effective dispersion of small branched HKUST-1 nanoparticles. Images of the HKUST-1 crystal structure can be clearly observed behind the transparent branched HKUST-1 / 6FDA-DAM mixed matrix membrane, while the aforementioned images can be observed through the opaque bulk HKUST-1 / 6FDA-DAM mixed matrix membrane.

[0073] Table 2

[0074] Obtaining clear cross-sectional images using fractured mixed-matrix membranes was difficult. These cross-sectional SEM images, even at high resolution, did not clearly reveal the localized dispersion of HKUST-1 particles or the interfacial voids between the HKUST-1 particles and the 6FDA-DAM polymer. To address these issues, we performed FIB-SEM imaging of bulk and branched HKUST-1 / 6FDA-DAM mixed-matrix membranes for accurate cross-sectional analysis. Ga-ion milling was used for the FIB-SEM technique, which produced significantly smoother cross-sections than liquid nitrogen fracturing. Images of branched HKUST-1 samples were obtained by milling the top of the mixed-matrix membrane. However, because the bulk HKUST-1 particles settled to the bottom of the mixed-matrix membrane, the membrane was inverted for FIB-SEM milling to analyze only the MOF-rich phase. Figure 7 shows FIB-SEM images of bulk HKUST-1 / 6FDA-DAM mixed-matrix membranes. Various HKUST-1 weight loadings were used, all of which showed results consistent with those shown in Figures 7A-7C. In the FIB-SEM images, the bright areas represent HKUST-1 particles, whereas the dark areas represent the 6FDA-DAM polymer. Consistent with the previous FE-SEM images, the thickness of the MOF-rich layer increased with increasing HKUST-1 weight loading. Particularly interestingly, there was a clear and visible interfacial void between the bulk HKUST-1 particles and the 6FDA-DAM polymer, consistent with the permeation results shown below. Furthermore, the curtain effect highlighted by the lines in Figure 19 is due to a density gradient associated with inefficient material packing, thereby providing further evidence of the large voids. Figure 8 shows FIB-SEM images of branched HKUST-1 / 6FDA-DAM mixed matrix membranes with a 30 wt% branched HKUST-1 loading at various magnifications (Figures 8A-8C). Remarkably, in contrast to Figure 7, the branched HKUST-1 / 6FDA-DAM mixed matrix membrane in Figure 20 had branched HKUST-1 nanoparticles uniformly dispersed in the polymer matrix, and no interfacial voids were observed between the particles and the polymer.Particularly interestingly, Figures 5A-5C show that the 30 wt% branched HKUST-1 / 6FDA-DAM mixed matrix membrane retains the percolated network characteristic of the cast particle morphology. The branched HKUST-1 nanoparticles were well distributed throughout the thin film, demonstrating particle connectivity between free surfaces that benefits from the high aspect ratio of the branched structure. To more clearly assess the connectivity of the branched HKUST-1 nanoparticles in the mixed matrix membrane, FIB-SEM cross-sectional tomography was obtained for a thin film section measuring 11 μm in length, 29 μm in width, and 4 μm in depth. Figures 5D-5E show reconstructed 3D images of the 30 wt% branched HKUST-1 / 6FDA-DAM mixed matrix membrane using FIB-SEM tomography. A highly interconnected nanostructured composite thin film was formed from the branched HKUST-1 nanoparticles.

[0075] 9A-9B show the ATR spectra of pure 6FDA-DAM thin films and bulk HKUST-1 / 6FDA-DAM mixed matrix films with various HKUST-1 weight loadings. Figure 9A shows the FT-IR spectra using the ATR accessory. It is noteworthy that because the penetration depth of the ATR accessory is approximately 1–2 μm, it was possible to determine whether thin films with a thickness of approximately 50 μm had a uniform MOF distribution at both interfaces. Although the mixed matrix membranes had various bulk HKUST-1 weight loadings, the FT-IR spectra of the upper side of the mixed matrix membrane (Figure 9A) showed the same characteristic functional group absorption bands of 6FDA-DAM, without any notable bulk HKUST-1 peaks. On the other hand, as shown in Figure 9B, the FT-IR spectra of the lower side of the mixed matrix membrane showed the same peak at 1645 cm regardless of the bulk HKUST-1 weight loading. -1 and 490 cm -1The features showed almost identical HKUST-1 peak intensities at 1645 cm , indicating that most of the bulk HKUST-1 particles had settled on the lower side of the mixed matrix membrane. Figures 10A-10B show the FT-IR spectra of pure 6FDA-DAM thin films and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings, using the same ATR accessory. Unlike the FT-IR spectrum of the bulk HKUST-1 / 6FDA-DAM mixed matrix membrane, the FT-IR spectrum of the branched HKUST-1 / 6FDA-DAM mixed matrix membrane showed a peak intensity at 1645 cm for both the upper and lower sides of the mixed matrix membrane as the branched HKUST-1 weight loading increased. -1 and 490 cm -1 The results showed a gradual increase in the intensity of the branched HKUST-1 peak characteristic of the mixed matrix membrane, suggesting a uniform distribution of branched HKUST-1 nanoparticles in the 6FDA-DAM polymer matrix. These results further quantified the conclusions from cross-sectional FE-SEM and FIB-SEM images that the bulk HKUST-1 particles settled at the bottom of the mixed matrix membrane, whereas the branched HKUST-1 nanoparticles were well dispersed throughout the mixed matrix membrane. X-ray diffraction (XRD) patterns were obtained for pure 6FDA-DAM polymer thin films and bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings. In general, the higher the HKUST-1 weight loading, the stronger the crystalline peaks, regardless of the HKUST-1 morphology. At similar weight loadings, the bulk HKUST-1 peaks were more intense than branched HKUST-1. These results indicated that crystallinity was maintained in the thin film preparations and that the bulk MOF structure was inherently more crystalline than the branched structure.

[0076] Permeation Test Before determining the gas permeation properties of the samples, we investigated the effect of activation temperature by using a heating jacket covering the permeation cell while the mixed-matrix membrane was held under vacuum and heated to activate the HKUST-1 particles. H2 and N2 gases were used as probe gases for the 20 wt% branched HKUST-1 / 6FDA-DAM mixed-matrix membrane at 15, 30, and 45 psi. These results indicated that the H2 and N2 permeabilities decreased with slight temperature increases, likely due to increased polymer-MOF interfacial interactions or slight physical aging effects in the 6FDA-DAM polyimide phase, but also indicated that the samples were thermally stable under these conditions. Furthermore, nearly identical XRD measurements on the 30 wt% branched HKUST-1 / 6FDA-DAM mixed-matrix membrane obtained before and after activation using the heating jacket indicated that high HKUST-1 crystallinity was maintained during activation. Therefore, we concluded that the high-temperature activation process using the heating jacket did not significantly affect the gas transport properties and crystalline nature of the HKUST-1 particles. Furthermore, TGA analysis confirmed that the high temperature activation process aided in the complete removal of the solvent in the mixed matrix membranes.

[0077] Gas permeation tests were performed on pure 6FDA-DAM polymer thin films and bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings. Prior to testing, all membrane samples were annealed at 150 °C for 2 h using a heating jacket to activate the HKUST-1 particles in the mixed matrix membranes. Figures 11A-11D show the gas permeation tests for the pure 6FDA-DAM polymer membrane and the bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various weight loadings (10, 20). The transport properties of H2, CH4, N2, O2, and CO2 gases as a function of gas supply pressure at 35 °C are shown for bulk HKUST-1 / 6FDA-DAM mixed matrix membranes using 30% (30 wt%) and 30% (30 wt%). As the gas supply pressure increased, the light gases H2, CH4, N2, and O2 showed a slight decrease in permeability, while CO2, which has a higher characteristic gas-phase activity, initially showed a decrease in permeability associated with dual-mode and Langmuir adsorption effects, followed by an increase in permeability at higher pressures due to plasticization. The addition of bulk HKUST-1 particles into the 6FDA-DAM polymer matrix significantly increased gas permeability and reduced gas selectivity for the gas pairs of interest (i.e., CO2 / CH4 and CO2 / N2). For example, a 30 wt% bulk HKUST-1 / 6FDA-DAM mixed matrix membrane exhibited a CO2 gas permeability of approximately 2500 Barrer at 15 psi, which was 2.5 times that of pure 6FDA-DAM (i.e., 1000 Barrer), but the CO2 / CH4 selectivity decreased slightly from 19 to 15. Figures 12A-12D show the permeabilities of H2, CH4, N2, O2, and CO2 as a function of gas supply pressure at 35 °C for pure 6FDA-DAM polymer thin films and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various weight loadings (10, 20, and 30 wt%). As a function of increasing HKUST-1 weight loading, the branched HKUST-1 / 6FDA-DAM mixed matrix membranes showed a significant concomitant increase in gas permeability similar to that of the bulk HKUST-1 / 6FDA-DAM mixed matrix membrane. In comparison, a 30 wt% branched HKUST-1 / 6FDA-DAM mixed matrix membrane exhibited a CO2 permeability of approximately 2500 Barrer at 15 psi, which was similar to that of a 30 wt% bulk HKUST-1 / 6FDA-DAM mixed matrix membrane. However, the branched HKUST-1 / 6FDA-DAM mixed matrix membrane exhibited a smaller loss of selectivity for CO2 / CH4 and CO2 / N2 separations compared to the selectivity loss of a pure 6FDA-DAM polymer thin film.To investigate gas transport trends as a function of HKUST-1 particle type and weight loading, Figures 13A-13B show the transport properties at 35 °C and 15 psi for pure 6FDA-DAM polymer thin films and bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings (10, 20, and 30 wt%) for CO2 / CH4 and CO2 / N2 separations at the upper bound. As the weight loading of bulk and branched HKUST-1 in the mixed matrix membranes increased, CO2 gas permeability increased similarly. However, the branched HKUST-1 / 6FDA-DAM mixed matrix membranes experienced a more limited loss of selectivity, shifting their gas transport properties closer to the upper bound than those of the bulk mixed matrix membranes. The bulk HKUST-1 / 6FDA-DAM mixed matrix membranes experienced a loss of selectivity parallel to the upper bound. The theoretical gas transport properties of the mixed matrix membrane with the lowest HKUST-1 loading (i.e., 10 wt%) and the pure 6FDA-DAM polymer thin film could not be predicted by the Maxwellian model. The Maxwellian model assumes two assumptions: 1) the dispersed (i.e., filler) particles must be uniformly dispersed in the continuous (i.e., polymer) matrix, and 2) the dispersed (i.e., filler) particles must be spherical to apply a shape factor (n) of 1 / 3. However, branched HKUST-1 nanoparticles were not spherical, and bulk HKUST-1 particles were not uniformly distributed in the 6FDA-DAM polymer matrix. As a result, the addition of 10 wt% branched HKUST-1 nanoparticles (or even 10 wt% bulk HKUST-1 particles) significantly increased gas transport performance, far exceeding the theoretically predicted gas transport properties. In particular, the high aspect ratio of branched HKUST-1 nanoparticles (5–20) leads to a shape factor of less than 0.1 and a low percolation threshold of less than 10 vol%, but fitting the low shape factor to the Maxwell model still did not provide a reasonable fit when compared with the transport properties of pure HKUST-1 reported in the literature.It is possible that the highly unusual structures and defects of these novel branched particles may have more beneficial intrinsic transport properties that were not modulated in HKUST-1, but the permeability of pure branched MOFs (i.e., without polymer support) remains to be experimentally determined. This represents a significant and challenging task. Figures 14A-B show the trends in gas transport properties predicted by the Maxwellian model with a shape factor of 1 / 3 for HKUST-1 weight loadings from 0 to 100 wt % using pure 6FDA-DAM polymer thin film data and HKUST-1 polycrystalline membranes on porous supports derived from literature data based on interpolation. Figures 14A-B also include two sets of gas transport data points for pure 6FDA-DAM polymer thin films and bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings (10, 20, and 30 wt %), along with Maxwellian model predictions. Notably, the branched HKUST-1 / 6FDA-DAM mixed matrix membranes exhibited significantly improved gas transport properties compared to the theoretically predicted gas transport trends using HKUST-1 polycrystalline membranes. These results may suggest a non-standard transport mechanism in these composite systems, possibly related to the interconnected multidimensional branched structure. An additional feature of branched multidimensional MOFs is their exceptional resistance to plasticization, which is conferred by their polymeric support. To quantify this effect, Figures 15A-B show the normalized CO2 permeability of pure 6FDA-DAM polymer thin films, as well as bulk and branched HKUST-1 / 6FDA-DAM mixed matrix membranes with various HKUST-1 weight loadings. Adding bulk HKUST-1 to the 6FDA-DAM polymer slightly shifted the CO2 plasticization pressure point, suggesting a slight improvement in plasticization resistance. On the other hand, adding branched HKUST-1 to 6FDA-DAM significantly shifted the CO2 plasticization pressure point, indicating a substantial improvement in plasticization resistance. Particularly interestingly, the 30 wt% branched HKUST-1 / 6FDA-DAM mixed matrix membrane showed no observable plasticization pressure point even up to 750 psi, indicating that this membrane had high CO2 resistance. To further investigate the CO2 plasticization behavior, the difference in normalized CO2 permeability between the hysteresis curve and the initial permeation isotherm at the same CO2 pressure point was calculated.Within the scope of these calculations, the bulk HKUST-1 / 6FDA-DAM mixed matrix membranes showed no significant change in normalized CO2 permeability as a function of increasing MOF loading, whereas the branched HKUST-1 / 6FDA-DAM mixed matrix membranes showed a significant decrease in normalized CO2 permeability as a function of increasing MOF loading. These results were consistent with a molecular-level interpretation of decreased polymer chain mobility resulting from increased surface contact between the polymer chains and branched HKUST-1 nanoparticles, leading to increased plasticization resistance.

[0078] Materials and methods and characterization The materials used for the following examples are as follows: Copper(II) nitrate trihydrate (Cu(NO3)2·3H2O), trimesic acid (C9H6O6, 95%), sodium acetate (NaAc, C2H3NaO2, ≥ 99.0%), methanol (MeOH, CHO ≥ 99.9%), tetrahydrofuran (THF, C4H8O, ≥ 99.0%), and N-methylpyrrolidone (NMP, C5H9NO anhydride, 99.5%) were purchased from Sigma-Aldrich and used as received. Monomeric 4,4'-(hexafluoroisopropylidene)diphthalic anhydride (6FDA, C 19 H6F6O6 >99% and 2,4-diaminomesitylene (DAM, (CH3)3C6H(NH2)2 >99% were purchased from Akron Polymer Systems. 6FDA was dried overnight in vacuum at 160 °C to cyclize diacid contaminants before use, and DAM was dried overnight in vacuum at 70 °C before use. Triethylamine (Et3N, CH6H 15 N 99%) and o-dichlorobenzene (o-DCB >98%) were used as received from Fisher. Ethanol (EtOH, C2H6O 99.5%) was used as received from VWR. Gas cylinders of H2 (HY UHP300, 100%), CH4 (ME UHP300, 100%), N2 (NI UHP300, 100%), O2 (OX UHP300, 100%), and CO2 (CD UP300, 99.999%) were purchased from Airgas.

[0079] Synthesis of HKUST-1 particles For branched HKUST-1 nanoparticles, copper precursor solutions and organic ligand solutions were prepared separately. Copper nitrate trihydrate (0.9 g, 3.7 mmol) and trimesic acid (0.43 g, 2 mmol) were dissolved in separate MeOH solutions (100 mL) along with corresponding concentrations of sodium acetate (0.2, 0.6, and 1 mmol) (0.1, 0.3, and 0.5 molar equivalents relative to trimesic acid). The two separate solutions were immediately mixed, and the mixture was vigorously stirred (approximately 800 rpm) at room temperature for 1 h. After centrifugation of the final solution, at least three washing steps with fresh MeOH (100 mL) were performed. In these washing steps, the sample was exposed to MeOH at room temperature for several hours to remove unreacted monomers and chemical modifiers. The HKUST-1 suspension in MeOH was stored until use for characterization and the formation of mixed matrix films. For bulk HKUST-1 particle synthesis, the same procedure as above was used, except that sodium acetate was not added.

[0080] Synthesis of 6FDA-DAM polymer 6FDA-DAM polyimide was synthesized via the ester acid route to complete imidization. A three-neck 250 mL round-bottom flask and all additional glassware were dried overnight at 200 °C before use. The glassware was removed from the oven and cooled under a nitrogen stream, and the monomer was cooled under vacuum in a desiccator. 6FDA (17.7695 g, 40 mmol) was transferred to a three-neck flask containing 125 mL of EtOH. The reaction flask was equipped with a mechanical stirrer, a nitrogen inlet, and a Dean-Stark trap fitted with a condenser. The Dean-Stark trap was charged with EtOH, and EtN (10 mL) was added to the reaction flask. The reaction was heated to reflux under a nitrogen stream and stirred for 1 h. The Dean-Stark trap was drained, and excess EtN and EtOH were distilled off until a viscous ester acid solution remained. DAM (6.0088 g, 40 mmol) was added along with NMP (128 mL) and o-DCB (32 mL). After completion of distillation of EtOH and EtN, the Dean-Stark trap was charged with o-DCB and the reaction was heated to 175–180 °C. The reaction was heated and stirred under nitrogen for 48 h to yield a fully cyclized polyimide. The viscous polymer solution was precipitated dropwise into stirred MeOH to yield white polymer beads. The polymer beads were blended and subjected to Soxhlet extraction with MeOH overnight to remove additional NMP. The polymer was dried to a constant weight of 21.4 g (Mw = 149 kDa, D = 4.1, 96% yield).

[0081] Preparation of mixed matrix membranes The HKUST-1 particles were not dried before mixing with the polymer to prevent agglomeration. The solvent containing the HKUST-1 suspension was exchanged from MeOH to THF by repeated centrifugation and washing processes. For this step, the sample was centrifuged in a centrifuge tube, the supernatant was removed, and fresh THF (100 mL) was used to top the sample and exchange with MeOH at room temperature for several hours. At least three centrifugation and washing steps were used to complete the solvent exchange. To load the desired HKUST-1 weight into the mixed matrix membrane, the concentration of HKUST-1 in the stock solution was first determined by removing a small amount (0.5 mL) of the sonicated HKUST-1 suspension in THF. The volume was measured immediately after removal, and the solution was then completely evaporated in a vacuum oven. When the HKUST-1 particles turned purple due to adsorption of atmospheric gases and vapors, the vial containing the dried HKUST-1 particles was reweighed, allowing the approximate concentration of the HKUST-1 solution to be determined. Using the approximate concentration of a known HKUST-1 solution, mixed matrix membranes were prepared as follows. 6FDA-DAM polymer was dissolved in THF solvent and stirred using a stir bar for 3-4 hours to obtain a homogeneous solution. The total mass of the casting solution for membrane fabrication was kept constant at 0.15 g. Therefore, 0.15 g of 6FDA-DAM polymer was used for the pure polymer thin film. For the mixed matrix membranes, 0.015, 0.03, and 0.045 g of HKUST-1 particles were added to 0.135, 0.12, and 0.105 g of 6FDA-DAM to obtain 10, 20, and 300 μg concentrations, respectively, in the mixed matrix membranes. The HKUST-1 solution was sonicated using a horn (Qsonica, Q500) for 1 minute to further disperse the HKUST-1 particles, and then mixed with the 6FDA-DAM polymer solution. The mixed solution was stirred for 3–4 hours to homogeneously disperse the HKUST-1 particles in the 6FDA-DAM polymer. The mixed solution was sonicated using a horn for 1 minute, poured into a glass Petri dish, and covered with an aluminum foil cap with a small hole to allow the THF solvent to slowly evaporate. The resulting thin film was cast in an oven at 50°C overnight, resulting in a film thickness of 40–50 μm. The free-standing thin film was peeled off from the glass Petri dish and dried in a vacuum oven for 24 hours to remove the residual THF solution.

[0082] Pure gas permeation measurements Pure gas permeation measurements were performed using an automated, custom-built, constant-volume, variable-pressure pure gas permeation system (Maxwell Robotics). A stock disc of brass shim with an outer diameter of 47 mm and an internal concentric hole was specially machined for the membrane support. The membrane was cut to an area larger than the hole in the brass disc, and the membrane thickness was measured using a micrometer before coupon fabrication. The membrane was placed over the hole in the support disc and sealed to the exposed small area of ​​the membrane with an impermeable epoxy adhesive (Devcon, 5 Minute Epoxy, 14250). The active area of ​​the membrane was determined five times using a scanner and Image J software (NIH), and the average active area was used in the permeation calculations. The membrane was mounted in a stainless steel filter holder for permeation experiments. Prior to gas permeation testing, all membranes were activated in situ under dynamic vacuum at 150°C for 2 hours using a custom-built heating jacket (HTS / Amptek) in the permeation cell. After activation, the permeation cell was allowed to cool naturally overnight to allow the temperature inside the permeation cell to return to room temperature. To precisely control the test temperature, the permeation equipment was immersed in a constant-temperature water bath heated to 35 °C using an immersion circulator (ThermoFisher Scientific, SC150). A leak test was performed under static vacuum by closing all valves in the permeation system. This test was performed before the gas permeation test. Gas permeability was calculated by subtracting the leak rate from the measured gas transport rate. Note that for all gas permeation tests, the leak rate was less than 1% of the gas transport rate. Pure H2, CH4, N2, O2, and CO2 gases were continuously supplied for multi-pressure testing at 35 °C. The first two supply pressures considered were 15 psi and 75 psi. Next, the pressure was systematically increased in 75 psi increments up to 900 psi for all gases except CO2, which was tested up to 750 psi due to its saturated vapor pressure. To investigate the hysteresis effect after pressurization, the CO2 gas supply pressure was gradually decreased in units compatible with our initial pressurization step. After the depressurization step, the sample was held under dynamic vacuum for 2 hours. The same pressurization protocol as the first one was then repeated.Each gas was tested for permeation at different pressure hold times: H2 for 18 minutes, CH4 for 30 minutes, N2 for 30 minutes, O2 for 24 minutes, and CO2 for 3.6 minutes. To reliably assess steady-state permeation, a time lag of at least 6x was used to calculate the gas transport rate at each pressure point. The time lag is defined as the distance from the pressure (y-axis) to the intersection of the line obtained by extrapolating the time plot to the linear portion of the steady-state permeation pressure increase with the time axis (x-axis). Pure gas permeability (P) was calculated. i ) is the unit of Barrer (1 Barrer = 10 -10 (cm 3 (STP)·cm) / (cm 2 Calculated in s cmHg):

number

number

number

number

number

[0083] [Table 3]

[0084] [Table 4]

[0085] [Table 5]

[0086] The Maxwell model (4) was used to predict the theoretical gas transport properties of mixed matrix membranes.

number

[0087] At high feed pressures, it was necessary to consider non-ideal gas phase behavior, especially for gases such as CO. Therefore, fugacity (f) was calculated to more accurately assess permeability. Fugacity-based permeability was calculated as follows:

number

[0088] The fugitive power was estimated at 35 °C using a virial equation including the second and third virial coefficients. Gas-phase activity (f / f sat ) was calculated by dividing the estimated fugacity by the estimated fugacity at the theoretical saturation pressure point for CO2 at 35°C using data tables from NIST. Since CO2 is a supercritical gas at this temperature, T -1 Note that the extrapolation was done by extrapolating the saturated fugacity through the critical point when plotting ln of the fugacity as a function of .

[0089] Characterization Nuclear magnetic resonance (NMR) spectroscopy was performed using an Ascend™ 500 spectrometer (Bruker) equipped with a TopSpin™ 3.2. 1H-NMR spectra were acquired at 500 MHz. All spectra were obtained from 0.7 mL of 1–5% (w / v) DMSO-d6 solutions. Gel permeation chromatography (GPC) measurements were performed by Intertek (Allentown, PA) using a Waters 2695 chromatographic system, THF mobile phase, and refractive index detection. Solutions were analyzed using a polystyrene / divinylbenzene column set from Agilent Technologies. Sample solutions were prepared as 2 mg / mL THF solutions and filtered using a 0.45 μm nylon filter. Samples were manually shaken until completely dissolved. Molecular weights were measured using online light scattering and intrinsic viscosity signals. Narrow-range polystyrene standards ranging from 6 M to 162 Daltons were used for calibration. Fourier transform infrared (FT-IR) spectra were obtained using an Alpha II FT-IR spectrometer (Bruker) equipped with an attenuated total reflectance (ATR) accessory at a resolution of 4 cm with 32 scans. -1The data were obtained using a TA instrument (TGA500) at a heating rate of 20 °C / min for the MOF powder and MOF mixed-matrix membrane. To determine the mass of the remaining metal oxide after complete burn-off of the MOFs, the MOF powder was first activated in a TGA furnace at 150 °C for 2 h (the same conditions used for activation in gas permeation measurements) under nitrogen atmosphere. The furnace was then cooled to room temperature, and once the temperature stabilized, a ramp was run to 700 °C under air at a heating rate of 20 °C / min. The final product was pure metal oxide due to oxidation. To determine the actual MOF weight loading in the mixed-matrix membrane, the same protocol was followed for the MOF mixed-matrix membrane. Once the polymer matrix completely decomposed under air at 700 °C, leaving only the metal oxide, the ratio of final weight (mixed-matrix membrane) / final weight (MOF powder) was determined and is reported herein as a percentage of weight loading. The glass transition temperature was determined from the third trace of a repeated heating protocol performed between room temperature and 420 °C at a heating rate of 20 °C / min using a TA Instruments differential scanning calorimeter (DSC250). The number-average particle size distribution was obtained from dynamic light scattering (DLS) analysis. Dilute MOF suspensions in MeOH were sonicated to homogeneously disperse the MOF particles, followed by DLS measurements at 35 °C using a Zetasizer Nano S90 (Marvern). The diffuse reflectance of the MOF powder was measured using a UV-Vis spectrophotometer (Perkin Elmer Lambda 1050). For UV-Vis sample preparation, the MOF powder was fully packed into a circular powder pellet cell designed for measuring solid samples. Raman spectra were obtained using a Renishaw Invia Reflex Micro Raman spectroscopy system using a charge-coupled device (CCD) camera equipped with a 532 nm laser and a 1200 l / mm grating for the MOF powder on the substrate. Powder X-ray diffraction (XRD) measurements were obtained using a Bruker D8 Discover diffractometer equipped with a copper tube (l = 0.15418 nm) and a VANTEC-500 2D detector. Data were recorded over the 5–40° range. Nitrogen physisorption measurements were performed on a Micromeritics ASAP 2460 at −196 °C.Prior to measurement, all samples were analyzed by Micromeritics. The samples were degassed overnight under vacuum at 150 °C in a Smart VacPrep. The apparent surface area was determined using the Brunauer-Emmett-Teller (BET) method in the P / P range of 0.02–0.10. The micropore volume (V) was calculated using the t-plot method. mic ) was estimated. Transmission electron microscope (TEM) images were obtained using an FEI Tecnai (G2 Spirit TWIN) operating at 120 kV. A droplet of a diluted MOF solution with a concentration of less than 0.5% in MeOH was cast onto a copper TEM grid, and the TEM grid was allowed to air dry for several hours to completely evaporate the MeOH solvent. Cross-sectional field emission scanning electron microscope (FE-SEM) images of the thin films were collected using an FE-SEM instrument (Zeiss Merlin). The thin films were cryo-fractured into two pieces in liquid nitrogen to observe the cross-sections. The cross-sections of the thin films were sputtered with gold (Au) to create a thin conductive layer (3–5 nm) for charge dissipation on the non-conductive thin film. To observe smoother cross-sections, focused ion beam scanning electron microscope (FIB-SEM) imaging was performed using an FEI He microscope with gallium (Ga)-ion milling. This was performed using a lios Nanolab 600 Dual Beam System. The thin film surface was similarly coated with gold (Au) using a sputtering system to create a thin conductive layer of the same thickness. A rectangular platinum (Pt) thin layer was coated using an electron beam position. After tilting the sample at a 52° angle, Ga-ion milling was performed at a regular cross-section below this Pt layer at an appropriate beam current, such as 6.5 nA. Further Ga-ion milling was performed at a lower current using the cleaning cross-section mode to create a smooth cross-section before imaging. Note that for consistent FIB-SEM imaging, all thin films were investigated using the same protocol and conditions, regardless of the morphology of HKUST-1. For reconstruction of FIB-SEM tomographic images, a protective Pt / C mixture (approximately 250 nm) was deposited using an electron beam on the area to be sectioned and on an adjacent square area for alignment reference. A platinum layer (approximately 1 μm) was then deposited over these same areas using a Ga ion beam, engraving an X-shape within the alignment square. The initial trench was milled using an accelerating voltage of 30 kV and a current of 2.5 nA. A final milling step at a lower current (30 kV, 80 pA) was used to smooth the cross-section and remove material damaged during the initial trenching. Micrograph sections with a lateral spacing of 5 nm were then acquired using a software routine provided by FEI. The SEM was operated in backscattering mode (3 kV, 0.4 nA) using a through-the-lens detection system to increase the contrast between materials with different atomic weights for taking the final micrographs. The Avizo software package (FEI) was used for data processing, during which images were aligned, cropped, and stacked. An adaptive histogram equalization routine was used to remove artifacts from depth shadowing, and a frequency-domain FFT filter was used to remove curtain artifacts. Adaptive 3D histogram and edge detection routines were combined with smart thresholding techniques to find edges with large intensity changes, and then noise reduction routines were used to eliminate single pixel regions.Different intensity regimes were assigned to the polymer material, copper material, or void regions (bubbles), and then 3D reconstruction files were created for the copper material mesh. Open-source software (Blender) was used to create the final image of the 3D tomographic reconstruction for the copper mesh.

[0090] Example 2 This example describes the synthesis and characterization of branched nanoparticles of the metal-organic framework ZIF-8. In this example, triethylamine was used as a chemical modifier during the synthesis of ZIF-8 branched nanoparticles. Triethylamine has a higher pKa value than sodium acetate and is more soluble in methanol. The metal precursor was prepared by dissolving 0.5 g of zinc nitrate hexahydrate (Zn(NO3)2 6H2O, 1.68 mmol) in 20 mL of methanol (0.99 mmol), and the ligand precursor was prepared by dissolving 0.5 g of 2-methylimidazole (C4H6N2, 6.10 mmol). The ligand solution was poured into the stirred metal solution at room temperature and atmospheric pressure. Immediately after mixing, 0.2 mL of pure triethylamine (1.43 mmol) was added dropwise. The reaction solution was stirred at 600 rpm for 1 h to obtain multidimensional ZIF-8 nanoparticles. The particles were purified by washing with methanol and centrifuging the suspension at 11,000 rpm for 90 minutes. Between each washing and centrifugation step, the particles were resuspended in 40 mL of fresh methanol and sonicated by indirect sonication for 30 minutes, direct sonication for 90 seconds, and vortexing at 2000 rpm for more than 2 hours. Three washing steps were used for each sample. The multidimensionally branched ZIF-8 nanoparticles possessed previously unrecognized surface terminations, which resulted in different dispersion behavior in the solvents. The branched ZIF-8 nanoparticles were suspended in various solvents in different glass vials. The solvents used were dichloromethane, chloroform (solubility parameter 9.2 (cal / cm)), and HCl (solubility parameter 9.2 (cal / cm). 3 ) 1 / 2 ), tetrahydrofuran (solubility parameter 9.3 (cal / cm 3 ) 1 / 2 ), dimethylformamide (solubility parameter 12.1 (cal / cm 3 ) 1 / 2), and methanol (solubility parameter 14.5 (cal / cm 3 ) 1 / 2 ) was considered. Although it was difficult to form stable ZIF-8 suspensions in most solvents tested, a uniform dispersion could be formed in dimethylformamide (DMF), which is also a good solvent for use in making casting solutions for certain polymers (such as certain polyimide polymers).

[0091] Example 3 This example describes the fabrication and characterization of mixed matrix membranes containing branched nanoparticles of the metal-organic framework ZIF-8.

[0092] Mixed-matrix films containing branched ZIF-8 nanoparticles were fabricated using a 6FDA-DAM polymer matrix. The branched multidimensional ZIF-8 nanoparticles described in Example 2 (hereafter referred to as branched ZIF-8 or BZ) exhibited small hierarchical particle sizes (approximately 60 nm) and uniform distribution within the 6FDA-DAM polymer matrix. As a result, the BZ / 6FDA-DAM MMM was transparent. On the other hand, the rhombic dodecahedral ZIF-8 particles (called RDZ) used as a control sample had particle sizes ranging from 100 to 500 nm and exhibited agglomeration when mixed with the 6FDA-DAM polymer matrix to form the RDZ / 6FDA-DAM MMM. This agglomeration of RDZ particles reduced the transparency of the thin film, and the RDZ-6FDA-DAM sample was opaque due to its higher reflective index. Additionally, a film containing commercially available ZIF-8 particles (called CZ) (called CZ / 6FDA-DAM MMM) was fabricated. The CZ / 6FDA-DAM MMM resulted in significant aggregation of the MOF particles, which resulted in a noticeable pattern on the thin film. Therefore, the BZ and RDZ mixed-matrix membrane samples were used to fabricate mixed-matrix membranes for accurately comparing the effect of MOF particle structure on transport properties.

[0093] Cross-sectional SEM images of BZ / 6FDA-DAM MMMs with various BZ loadings (10, 20, 30, and 40 wt%) were acquired. The thin films were fractured into two pieces in liquid nitrogen. Cross-sectional SEM images of the 10 wt% BZ / 6FDA-DAM MMM showed a very rough, uneven, torn cross-section due to enhanced matrix elongation during thin film fracture in liquid nitrogen. However, as the BZ MOF loading increased, the cross-section became smoother due to immediate fracture without significant matrix elongation. Higher magnification cross-sectional SEMs were acquired, demonstrating that FE-SEM images of BZ / 6FDA-DAM MMMs exhibited uniform MOF distribution regardless of MOF loading.

[0094] To further investigate the dispersion of branched ZIF-8 particles within the membrane and whether these particles form a percolated network, 30 wt. % RDZ / 6FDA-DAM MMM and 30 wt. % BZ / 6FDA-DAM MMM were imaged by FIB-SEM. Figure 16A shows a FIB-SEM image showing that 30 wt. % RDZ / 6FDA-DAM MMM contained nearly spherical ZIF-8 nanoparticles (100–500 nm) in the 6FDA-DAM polymer matrix. At this weight loading, ZIF-8 (0.97 g cm) -3 ) and 6FDA-DAM (1.3 g cm -3 Considering the density of the RDZ nanoparticles, the volume fraction of the RDZ nanoparticle loading was determined to be 36%. Nevertheless, the 30 wt% RDZ / 6FDA-DAM MMM did not show clear evidence of a perfused network in the FIB-SEM images. It is possible that the polymer favorably interacts with the RDZ ZIF-8 particles to provide a coating layer on the MOF during casting, thereby preventing interparticle contact. While such a feature is useful for preventing defects, it also fails to overcome the Maxwellian model's ability to predict the transport properties of the MMM. On the other hand, Figure 16B shows FIB-SEM images showing that the 30 wt% BZ / 6FDA-DAM MMM exhibited a perfused network that maintained a uniform distribution throughout the thin film. Such a perfused network may be due to the characteristic interconnected structure of branched ZIF-8 nanoparticles and may enhance gas transport properties.

[0095] To investigate the transport performance, permeation tests of neat 6FDA-DAM polymer film, RDZ / 6FDA-DAM MMM, and BZ / 6FDA-DAM MMM were conducted at 35 °C and 15 psi for the separation of light gases (H2 separation from CH4, N2, and CO2, as well as O2 / N2 separation). The MMMs used various MOF loadings (10, 20, 30, and 40 wt%). The results of the permeation tests are shown in FIGS. 17A - D. As the MOF loading increased, an increase in selectivity was observed for BZ / 6FDA-DAM MMM, while RDZ / 6FDA-DAM MMM showed nearly constant selectivity. This surprising result indicates that the branched ZIF-8 sample exhibits molecular sieving behavior for gases smaller than those traditionally considered to be separated using this material (such as propylene / propane separation). The kinetic diameters of the gases were considered to increase in the following order: H2 (2.89 Å) < CO2 (3.30 Å) < O2 (3.46 Å) < N2 (3.64 Å) < CH4 (3.80 Å). Based on the findings shown in FIG. 17D, a molecular sieving window for O2 / N2 separation is available in the BZ / 6FDA-DAM membrane, suggesting that the BZ sample has effective pore openings between the sizes of O2 and N2. In particular, mixed matrix membranes containing branched ZIF-8 nanoparticles showed selectivity for light gases greater than 15 (such as H2 / N2) and even greater than 20. Without wishing to be bound by any particular theory, it is thought that the chemical regulator (triethylamine) is incorporated into the crystal structure of the ZIF-8 MOF of the BZ particles, thereby tethering to the organic polydentate ligand (2-methylimidazole) or inhibiting the gate-opening effect (mobility), resulting in smaller effective pore openings.

[0096] Example 4 This example describes attempts to synthesize branched ZIF-8 nanoparticles in methanol using various different potent chemical regulators. The different chemical regulators tested were pyridine (pK a = 5.44 (in methanol), pK a = 5.22 (in H2O)), aniline (pKa = 6.05 (in methanol), pK a = 4.6 (in H2O), p-phenylenediamine (pPDA, pK a = 6.2 (in H2O), 4-ethoxyaniline (p-phenetidine, pK a = 6.92 (in methanol), pK a = 5.32 (in HO), N,N-dimethylpropargylamine (pK a = 7.97 (in methanol), pK a = 8.04 (in H2O), sodium formate (pK a = 8.87 (in methanol), sodium acetate (pK a = 9.63 (in methanol), sodium propionate (pK a = 9.71 (in methanol), trimethylamine (TMA, pK a = 9.8 (in methanol), pK a = 9.74 (in H2O), triethylamine (pK a = 10.78 (in methanol), pK a = 10.67 (in H2O), diethylamine (pK a = 11.2 (in methanol), pK a = 10.72 (in HO), tri-n-butylamine (TBA, pK a = 10.89 (in HO), and n-butylamine (pK a = 11.48 (in methanol), pK a = 10.6 (in HO). ZIF-8 nanoparticles were synthesized by the procedure described in Example 2 using identical molar amounts of modulators of different structures: a ligand source (0.5 g of CHN dissolved in 20 mL of methanol) was poured into a metal source (0.5 g of Zn(NO)HO dissolved in 20 mL of methanol). 1.43 mmol of each modulator was added dropwise (alternatively, if the modulator was solid, it was dissolved in the ligand solution before mixing), stirred under ambient conditions for 1 hour, and then washed three times with fresh methanol as described in Example 2.

[0097] ZIF-8 nanoparticles synthesized using pyridine, aniline, pPDA, sodium formate, sodium acetate, and sodium propionate were found to be unbranched, instead possessing the RDZ morphology described in Example 2 above. Branched ZIF-8 nanoparticles were found to form when any one of the following was used as a chemical modifier: trimethylamine, triethylamine, diethylamine, tri-butylamine, and n-butylamine. Figure 18 shows images of ZIF-8 nanoparticles formed in the presence of aniline (Figure 18A, unbranched) and triethylamine (Figure 18B, branched). ZIF- pK of the basic nitrogen of the 2-methylimidazolate polydentate ligand in 8 MOFs a (pK a pK (approximately 7.97) or higher a It was observed that branched ZIF-8 nanoparticles were formed with each chemical modulator having a pKa of 0.01g or higher of the multidentate ligand moiety that binds to the metal ion in the metal-organic framework, although not wishing to be bound by any particular theory. a This shows that the use of chemical modulators having the formula (I) can contribute to the formation of branched morphologies of branched MOF nanoparticles.

[0098] X-ray photoelectron spectroscopy was also performed on RDZ and BZ ZIF-8 nanoparticles (BZ particles were synthesized using triethylamine as a chemical modifier). The XPS data were used to determine the mole percentages of elements in the two types of nanoparticles, as summarized in Table 6.

[0099] [Table 6]

[0100] As can be seen in Table 6, there is a significant difference in composition between the RDZ ZIF-8 nanoparticles and the BZ ZIF-8 nanoparticles. The coordination of the triethylamine chemical modulator to the zinc ions in the BZ nanoparticles can explain the difference in composition.

[0101] While several embodiments of the present invention have been described and illustrated herein, those skilled in the art will readily envision various other means and / or structures for performing the functions and / or obtaining the results and / or one or more advantages described herein, and each such variation and / or modification is deemed to be within the scope of the present invention. More generally, those skilled in the art will readily recognize that all parameters, dimensions, materials, and configurations described herein are intended to be exemplary, and that the actual parameters, dimensions, materials, and / or configurations will depend on the specific application(s) for which the teachings of the present invention are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, numerous equivalents to the specific embodiments of the invention described herein. Accordingly, the foregoing embodiments are presented by way of example only, and it should be understood that, within the scope of the appended claims and their equivalents, the invention may be practiced otherwise than as specifically described and claimed. The present invention relates to each individual feature, system, article, material, and / or method described herein. Furthermore, any combination of two or more such features, systems, articles, materials, and / or methods, if such features, systems, articles, materials, and / or methods are not mutually inconsistent, is included within the scope of the present invention.

[0102] The indefinite articles "a" and "an," as used in this specification and claims, unless clearly indicated otherwise, should be understood to mean "at least one."

[0103] The phrase "and / or," as used herein and in the claims, should be understood to mean "either or both" of the elements connected by the phrase (i.e., the elements may exist together or separately). Unless expressly indicated otherwise, other elements may be present as needed, whether in association with the elements specifically identified by the "and / or" clause. Thus, non-limiting examples are provided. As a general example, when "A and / or B" is used in conjunction with open-ended language such as "comprising," it can refer, in one embodiment, to A without B (optionally including elements other than B); in another embodiment, to B without A (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements), etc.

[0104] As used herein and in the claims, "or" should be understood to have the same meaning as "and / or" as defined above. For example, when separating items in a list, "or" or "and / or" shall be construed as inclusive (i.e., including at least one, but also more than one, of a number or list of elements and, where appropriate, additional unlisted items). Clearly exclusive terms such as "only one" or "exactly one," or, when used in the claims, "consisting of," shall refer to the inclusion of exactly one element of a number or list of elements. In general, the term "or" as used herein shall be construed to indicate exclusive alternatives only (i.e., "one or the other, but not both") when preceded by terms of exclusivity such as "either," "one of," "only one of," or "exactly one of." "Consisting essentially of," when used in the claims, shall have its ordinary meaning as used in the field of patent law.

[0105] As used herein and in the claims, the phrase "at least one," when referring to a list of one or more elements, should be understood to mean at least one element selected from any one or more elements in the list of elements, but not necessarily including at least one of each and every element specifically listed in the list of elements, nor excluding any combination of elements in the list of elements. This definition also allows that there may optionally be elements other than those specifically identified in the list of elements to which the phrase "at least one" refers, whether related or not to the specifically identified elements. Thus, as a non-limiting example, "at least one of A and B" (or, equivalently, "at least one of A or B" or, equivalently, "at least one of A and / or B") can refer, in one embodiment, to at least one A (optionally including more than one A) with no B present (and optionally including elements other than B); in another embodiment, to at least one B (optionally including more than one B) with no A present (and optionally including elements other than A); in yet another embodiment, to at least one A (optionally including more than one A) and at least one B (optionally including more than one B) (and optionally including other elements); etc.

[0106] In the claims and the above specification, all transitional phrases such as "comprising," "including," "carrying," "having," "containing," "involving," and "holding" shall be understood to be open-ended (i.e., meaning "including, but not limited to"). Only the transitional phrases "consisting of" and "consisting essentially of" shall be closed or semi-closed transitional phrases, respectively, as set forth in Section 2111.03 of the United States Patent Office Manual of Patent Examining Procedures. For example, the present invention provides the following items. (Item 1) A device, a matrix comprising a polymer; and A particle comprising a metal-organic framework, the metal-organic framework comprising metal ions and multidentate Particles containing ligands a mixed matrix membrane comprising A device wherein the particles are uniformly distributed throughout the matrix and no convective transport is detectable by transmission measurements. (Item 2) Item 10. The device of item 1, wherein the particles are branched nanoparticles, the branched nanoparticles having a hydrodynamic diameter of 2 μm or less, an aspect ratio of at least 5, and branches with diameters of 200 nm or less. (Item 3) 3. The device of any one of items 1 to 2, wherein the polymer is present in the matrix in an amount of 50% or more by weight of the matrix. (Item 4) 4. The device according to any one of items 1 to 3, wherein essentially all of the matrix consists of the polymer. (Item 5) 5. The device according to any one of items 1 to 4, wherein the multidentate ligand comprises at least two carboxylate groups. (Item 6) 6. The device according to any one of items 1 to 5, wherein the metal ions are transition metal ions. (Item 7) 7. The device according to any one of items 1 to 6, wherein the metal ion is copper. (Item 8) 8. The device according to any one of items 1 to 7, wherein the metal-organic framework is HKUST-1. (Item 9) 8. The device of any one of items 1 to 4 and 6 to 7, wherein the multidentate ligand comprises an imidazolate. (Item 10) Item 11. The device according to any one of items 1 to 6 and 9, wherein the metal ion is zinc. 11. The device according to any one of items 1 to 7 and 9 to 10, wherein the metal-organic framework is a zeolite-like imidazolate framework. (Item 12) 12. The device of any one of items 1 to 7 and 9 to 11, wherein the metal-organic framework is ZIF-8. (Item 13) 13. The device of any one of items 1 to 12, wherein the particles form a percolated network in the mixed matrix membrane. (Item 14) 14. The device according to any one of items 1 to 13, wherein the device is capable of separating a portion of the first gas from a gas mixture comprising a first gas and a second gas. (Item 15) 15. The device of any one of items 1 to 14, wherein the plasticization pressure point for a gas of the mixed matrix membrane is at least 1.5 times the plasticization pressure point of a substantially identical membrane lacking the particles. (Item 16) Item 16. The device of item 15, wherein the gas comprises CO2, ethylene, ethane, propane, and / or propylene. (Item 17) 17. The device of any one of items 1 to 16, wherein the polymer comprises a polyimide. (Item 18) The polyimide has the following structure: [ka] Item 18. The device according to item 17, comprising: (Item 19) 19. The device of any one of items 1 to 18, wherein the branched nanoparticles comprise an amine associated with the metal-organic framework. (Item 20) The branched nanoparticles comprise an amine associated with the metal-organic framework, the amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 20. The device of any one of items 1 to 19, wherein the alkyl groups are independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. (Item 21) 21. The device of any one of items 1 to 20, wherein the branched nanoparticles comprise triethylamine associated with the metal organic framework. (Item 22) 22. A method comprising contacting the device according to any one of items 1 to 21 with a gas mixture comprising a first gas and a second gas such that a portion of the first gas is separated from the gas mixture. (Item 23) 1. A composition comprising: Branched nanoparticles comprising a metal-organic framework, the metal-organic framework comprising a metal ion and a multidentate ligand. Including, wherein the branched nanoparticles have a hydrodynamic diameter of 2 μm or less, an aspect ratio of at least 5, and branches with diameters of 200 nm or less; A composition wherein said composition is constructed and arranged as a gas separation membrane. (Item 24) 24. The composition of claim 23, wherein the polydentate ligand comprises at least two carboxylate groups. (Item 25) 25. The composition according to any one of items 23 to 24, wherein the metal ion is a transition metal ion. (Item 26) 26. The composition according to any one of items 23 to 25, wherein the metal ion is copper. (Item 27) 27. The composition according to any one of items 23 to 26, wherein the metal-organic framework is HKUST-1. (Item 28) 27. The composition according to any one of items 23 and 25-26, wherein the polydentate ligand comprises an imidazolate. (Item 29) 29. The composition according to any one of items 23 to 25 and 28, wherein the metal ion is zinc. (Item 30) 20. The composition according to any one of items 23 to 26 and 28 to 29, wherein the metal-organic framework is a zeolite-like imidazolate framework. (Item 31) 31. The composition according to any one of items 23 to 26 and 28 to 30, wherein the metal-organic framework is ZIF-8. (Item 32) 32. The composition according to any one of items 23 to 31, wherein the branched nanoparticles are non-aggregated. (Item 33) 33. The composition according to any one of items 23 to 32, wherein the branched nanoparticles form an at least partially permeable network. (Item 34) 34. The composition of any one of items 23 to 33, wherein the branched nanoparticles comprise an amine associated with the metal-organic framework. (Item 35) The branched nanoparticles comprise an amine associated with the metal-organic framework, the amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 35. The composition of any one of items 23 to 34, wherein the alkyl groups are independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. (Item 36) 36. The composition of any one of items 23 to 35, wherein the branched nanoparticles comprise trimethylamine associated with the metal organic framework. (Item 37) 37. The composition according to any one of items 23 to 36, wherein the gas separation membrane is capable of separating a portion of the first gas from a gas mixture comprising a first gas and a second gas. (Item 38) 38. The composition of any one of items 23 to 37, wherein the plasticization pressure point for a gas of the gas separation membrane is at least 1.5 times the plasticization pressure point of a substantially identical membrane lacking the particles. (Item 39) 39. The composition of any one of items 23 to 38, wherein the gas comprises CO2, ethylene, ethane, propane, and / or propylene. (Item 40) 40. A method comprising contacting the gas separation membrane according to any one of items 23 to 39 with a gas mixture containing a first gas and a second gas such that a portion of the first gas is separated from the gas mixture. (Item 41) 1. A method comprising: a step of combining, in a liquid, a metal salt comprising a metal ion, a multidentate ligand, and a chemical modulator to form branched nanoparticles, wherein the branched nanoparticles comprise a metal-organic framework, and the metal-organic framework comprises the metal ion and the multidentate ligand; Including, wherein the metal salt, the multidentate ligand, and / or the chemical modulator are selected such that, when present in an amount of at least 0.1 equivalents relative to the concentration of the multidentate ligand, the aspect ratio of the branched nanoparticles increases by at least 3-fold relative to particles formed under the same conditions in the absence of the chemical modulator; The method further comprises a combining step, wherein the branched nanoparticles are combined with a polymer to form a mixed matrix membrane comprising a polymer network. (Item 42) Item 43. The method of item 41, wherein the branched nanoparticles have a hydrodynamic diameter of 2 μm or less, an aspect ratio of at least 5, and branches with diameters of 200 nm or less. 43. The method according to any one of items 41 to 42, wherein the chemical regulator is a salt comprising a cation and an anion, and the anion has at least one carboxylate group. (Item 44) 44. The method of claim 43, wherein the anion is acetate. (Item 45) 43. The method of any one of items 41 to 42, wherein the chemical modulator comprises an amine. (Item 46) The chemical modulator comprises an amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 46. ​​The method of any one of items 41-42 and 45, wherein the alkyl groups are independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. (Item 47) 46. ​​The method of any one of items 41 to 42 and 45, wherein the chemical regulator comprises triethylamine. (Item 48) pK of the chemical regulator in the liquid a is the pK of the multidentate ligand moiety that binds to the metal ion in the metal-organic framework a 48. The method according to any one of items 41 to 47, wherein the method further comprises: (Item 49) 49. The method according to any one of items 41 to 48, wherein the pKa of the chemical regulator in a liquid is 3.0 or greater. (Item 50) 50. The method according to any one of items 41 to 49, wherein the pKa of the chemical regulator in a liquid is 15.0 or less. (Item 51) 50. The method of any one of items 41 to 49, wherein the chemical modulator is associated with the branched MOF nanoparticles after formation of the branched nanoparticles. (Item 52) 52. The method of any one of items 41 to 51, wherein the branched nanoparticles comprise an amine associated with the metal-organic framework. (Item 53) The branched nanoparticles comprise an amine associated with the metal-organic framework, the amine having the formula NR3, where each R is hydrogen, optionally substituted C 1~8 53. The method of any one of items 41 to 52, wherein the alkyl groups are independently selected from branched and unbranched alkyl, optionally substituted cycloalkyl, and optionally substituted aryl. (Item 54) 54. The method of any one of items 41 to 53, wherein the branched nanoparticles comprise triethylamine associated with the metal-organic framework. (Item 55) 55. The method according to any one of items 41 to 54, wherein the liquid is at least 99% by volume methanol. (Item 56) 56. The method of any one of items 41 to 55, wherein the combining step comprises mixing the liquids at a temperature of 15°C or greater and 60°C or less for at least 10 minutes. (Item 57) 56. Any of items 41 to 55, wherein the polydentate ligand comprises at least two carboxylate groups. 1. The method according to any one of claims 1 to 9. (Item 58) 58. The method according to any one of items 41 to 57, wherein the metal ion is a transition metal ion. (Item 59) 59. The method according to any one of items 41 to 58, wherein the metal ion is copper. (Item 60) 60. The method according to any one of items 41 to 59, wherein the metal-organic framework is HKUST-1. (Item 61) 59. The method of any one of items 41 to 56 and 58 to 59, wherein the polydentate ligand comprises an imidazolate. (Item 62) 62. The method according to any one of items 41 to 58 and 61, wherein the metal ion is zinc. (Item 63) 63. The method of any one of items 41 to 56, 58 to 59, and 61 to 62, wherein the metal-organic framework is a zeolite-like imidazolate framework. (Item 64) 64. The method of any one of items 41 to 56, 58 to 59, and 61 to 63, wherein the metal-organic framework is ZIF-8. (Item 65) 65. The method of any one of items 41 to 64, wherein the branched nanoparticles form an at least partially permeable network in the mixed matrix membrane. (Item 66) 66. The method of any one of items 41 to 65, wherein the device is capable of separating a portion of the first gas from a gas mixture comprising a first gas and a second gas. (Item 67) 67. The method of any one of items 41 to 66, wherein the CO2 plasticization pressure point of the mixed matrix membrane is at least 1.5 times the CO2 plasticization pressure point of a substantially identical membrane lacking the metal-organic framework.

Claims

[Claim 1] Energy consumption.

Images