Efficient graphene film for carbon capture and preparation method and application thereof

Abstract

The invention relates to a preparation method of a graphene membrane with N-functionalized pores for selective gas separation, and also relates to a preparation method and application of the graphene membrane. Particularly, the graphene membrane provided by the invention has relatively high CO2 permeability and CO2 / N2 selectivity with attractive force.

Description

Technical Field

[0001] The present invention generally belongs to the field of gas selective separation filters, and is particularly suitable for separating gas mixtures, especially in the context of carbon capture resulting from the separation of carbon dioxide and nitrogen, such as separating from waste gas or wastewater. The present invention more specifically relates to filters using atomically thick graphene porous membranes. Background Art

[0002] Traditional gas separation processes such as cryogenic distillation, absorption, and adsorption have been successfully applied in many fields such as air fractionation, carbon capture, natural gas desulfurization, biogas treatment, and olefin / paraffin separation. These processes mainly rely on heat energy and achieve separation by utilizing the differences in volatility, solubility, and binding energy between gas molecules. However, due to the capital-intensive characteristics of their equipment, they are usually only suitable for medium and large-scale production (Feron et al., 2009, Energy Procedia, Vol. 1, No. 1, pp. 1067 - 1074). However, driven by increasingly stringent environmental regulations and carbon emission tax policies, there is an urgent need to develop energy-efficient and economical gas separation technologies. Membrane-based separation processes do not rely on heat energy and exhibit significant energy efficiency advantages in various gas separation scenarios (Khalilpour et al., 2015, J.Clean.Prod., Vol. 103, pp. 286 - 300), and their environmental friendliness is prominent (no chemical input and no by-products generated) (Sholl et al., 2016, Nature, Vol. 532, No. 7600, pp. 435 - 437). Such processes can be conveniently integrated into existing industrial facilities due to their simple structure and steady-state operation.

[0003] High-performance membrane materials can effectively reduce the energy consumption loss in carbon capture because they avoid the thermodynamic energy consumption requirements. Developing membrane materials with both high selectivity and high CO2 permeability is crucial for reducing the equipment footprint and capture cost. Current cutting-edge membrane technologies include: polymer thin film composites (Zhang et al., 2021, J. Memb. Sci., Vol. 638, 119696), metal-organic frameworks (Qiao et al., 2019, Nat. Mater., Vol. 18, 163 - 168), stacked nanosheet structures (Zhou et al., 2017, Nat. Mater., Vol. 8, No. 1), and the design of functional layers to promote CO2 transport (Chen et al., 2022, J. Memb. Sci., Vol. 645, 120195; Han et al., 2022, J. Memb. Sci. Vol. 2, No. 1, 100014; Marius et al., 2022, Science, Vol. 376, No. 6588, 90 - 94). These technologies have achieved breakthroughs in CO2 / N2 separation performance through material selection and structure optimization. The membrane layer design strategies for carbon capture can be systematically classified into two technical paths: material screening and structural engineering.

[0004] In dense polymer membranes, a polymer thin film is used as the selective separation layer. Compared with N2, CO2 has higher solubility and diffusion coefficients in most polymer membranes. Due to the smaller size of CO2 molecules, its diffusion coefficient in the adsorbed phase is higher. Solubility is determined by dispersion and electrostatic interactions. Polymers containing polar groups (such as ether or amine groups) have a high affinity for CO2, and thus can produce a high CO2 / N2 selectivity. Such materials have been commercially applied to CO2 capture membranes, but their performance is limited by the inherent properties of the polymers. The advanced membrane materials prepared by this method currently exhibit limited permeation rates (1,000 - 3,000 GPU; 1 GPU = 3.35×10 -10 mol m -2 s -1 Pa -1) Meanwhile, the CO2 / N2 selectivity is 20 - 100. In addition, there is a phenomenon of chain segment packing aging in nanoporous polymers, and their free volume will gradually decrease over time (Tiwari et al., 2014, Polymer, 55, 5788 - 5800). The thin films prepared from nanoporous crystalline materials (zeolites and metal - organic framework materials MOFs) have an ordered structure and well - defined pore size characteristics. By regulating the pore size parameters of zeolites and MOFs, a pore design beneficial to the rapid diffusion of CO2 can be achieved. Although such thin films have ideal potential in the field of CO2 / N2 separation, there is currently a lack of a simple method to prepare ultrathin film materials with both high CO2 permeation rate (Babu et al., 2019, Adv. Mater., Vol. 31, 1900855; Kim et al., 2017, J. Mater. Chem. A, Vol. 5, 11246 - 11254).

[0005] The thin film materials prepared by stacking two - dimensional nanosheets (such as graphene oxide nanosheets) have a transport path defined by the interlayer gap. Such materials are constructed by layer - by - layer deposition of nanosheets with micron - scale dimensions and a monolayer thickness of only 1 - 2 nm. The size of the interlayer gap can be precisely controlled by chemical regulation means, and this characteristic enables them to exhibit excellent selectivity. However, due to the highly tortuous and long - range nature of the transport path formed by the interlayer gap, the CO2 permeation flux still remains at a relatively low level (100 - 500 GPU) (Wang et al., 2016, Energy Environ. Sci., Vol. 9, 3107 - 3112).

[0006] Currently, facilitated transport membranes are gradually realizing commercial applications in carbon capture. Their core advantage lies in the high - selectivity separation of the CO2 / N2 system, which stems from the strong reaction mechanism between CO2 and the flowing (or fixed) carriers in the membrane, thus achieving the selective transport of CO2. The reported selectivity can reach the order of 100 - 1000, but the CO2 permeation flux is usually low (100 - 1000 GPU).

[0007] Therefore, although there are reports of high CO2 / N2 selectivity membrane materials (such as facilitated transport membranes), there are still major challenges in achieving the simultaneous improvement of ultra - high flux (>1000 GPU) and high selectivity. The CO2 permeation flux is restricted by the thickness limit of the selective layer. It is a challenge to reduce the thickness of the selective layer to less than 50 - 100 nm while avoiding pinhole defects that cause a decrease in selectivity.

[0008] Two-dimensional (2D) films with CO2-selective nanopores enable fast and selective gas transport. Monolayer porous graphene is an ideal selective layer because its atomically thin pores can maximize gas transport at specific pore sizes. The transport mechanism of gases through atomically thin pores depends on the relative relationship between molecular size and pore size. When the pore size is large (>2 nm), molecular effusion is the main transport path. Effusion can generate extremely high permeation fluxes, but it sacrifices gas selectivity, which is determined by the molecular mass ratio. In contrast, for sub-nanometer pores, the transport rate depends on the energy barrier that molecules experience during pore crossing. Studies have shown that this size-dependent energy barrier can effectively distinguish different gas molecules (Zhao et al., 2019, *Sci. Adv.*, Vol. 5, No. 1, eaav1851; Sun et al., 2021, *Nat. Commun.*, Vol. 12, No. 1, p. 7170).

[0009] However, given the extremely small size difference between CO2 and N2 precise pore introduction techniques are required to achieve a narrow pore size distribution. This is particularly difficult for membrane materials that require a high pore density to achieve high permeation fluxes. Therefore, recent research progress in graphene lattice engineering has achieved limited CO2 / N2 selectivity (20) and high CO2 permeation fluxes (10,000 GPU).

[0010] Amine functionalization of graphene oxide has been studied as a CO2 adsorption strategy (Zhou et al., 2019, *J. Memb. Sci.*, Vol. 573, 184-191; Zhang et al., 2019, *J. Memb. Sci.*, Vol. 570-571, 343-354). However, such materials have not been reported to have the characteristics of atomically thin membrane structures, high permeation fluxes, or high CO2 / N2 selectivity.

[0011] Molecular simulation studies have revealed the interaction mechanism between CO2 and porous graphene: by theoretically constructing oxygen-crown nanopores (i.e., 18-crown-6 nanopores) in the graphene layer, after replacing the carbon atoms at the pore edges with oxygen atoms, the charge distribution around the pores makes it affinity for CO2, while repelling N2 and O2 molecules. Compared with graphene membranes with all-carbon atoms (unmodified oxygen end groups), nitrogen end-group modified graphene nanopores, or larger oxygen end-group modified nanopores, graphene membranes containing 18-crown-6 nanopores show potential advantages in CO2 separation / capture and may achieve faster transport rates (Luan et al., 2022, *ACS Nano*, Vol. 16, No. 4, pp. 6274–6281).

[0012] However, although the construction of graphene crown ether nanopores can be theoretically achieved, it is quite challenging to obtain crown nanopores suitable for CO2 capture and filtration in experiments for the following reasons: There is currently no known method to precisely introduce the crown ether nanopore structure into graphene.

[0013] Therefore, although the development of membrane materials with high CO2 permeation flux is highly attractive and promising for carbon capture applications because it can significantly reduce the required membrane area and thus lower the equipment investment cost, to date, the development of these membranes has been hindered by the above-mentioned technical limitations. Summary of the Invention

[0014] The general objective of the present invention is to provide an efficient gas selective filter using a graphene membrane for gas separation (such as CO2 / N2 separation).

[0015] One specific objective of the present invention is to provide an efficient gas selective filter for CO2 capture.

[0016] Provide a preferred gas selective filter that has high selectivity for pairs of molecules of similar size (such as CO2 and N2), and is suitable not only for low-concentration CO2 feed but also for high-concentration CO2 feed compatible with combined heat and power (CHP) plants.

[0017] It is particularly advantageous to provide a gas selective filter that has an attractive CO2 / N2 selectivity, from about 20 to about 2,000, especially higher than 20 even for low-concentration CO2 feed (such as 0.5 - 1% CO2).

[0018] It is advantageous to provide a gas selective filter that has a large CO2 permeability, especially exceeding 200, particularly exceeding 50,000 GPU, while having an attractive CO2 / N2 selectivity.

[0019] It is advantageous to provide a gas selective filter made of inorganic materials that has a rigid structure and does not densify over time, thereby extending the service life of the membrane to more than 5 years.

[0020] An object of the present invention is to provide a gas selective filter comprising a graphene membrane, and a method for preparing a gas selective filter comprising a graphene membrane, the filter being low-cost, having good gas selectivity, and high CO2 capture performance.

[0021] It is advantageous to provide a simple and scalable method for preparing a gas selective filter comprising a graphene membrane, which can achieve the combination of the above-mentioned properties.

[0022] The object of the present invention can be achieved by providing a gas selective separation filter as claimed in claim 10 and its use as claimed in claim 16.

[0023] The present invention discloses a method for preparing a gas selective separation filter, comprising the following steps:

[0024] a) providing a supported nanoporous single-layer graphene membrane, the membrane comprising nanopores with a pore density of about 10 10 cm -2 to about 10 13 cm -2 and the nanopores are modified with O functional groups, the O / C ratio being about 0.01 to 0.5, and the single-layer graphene membrane being supported on a mechanical support;

[0025] b) performing N-functionalization of the graphene nanopores of the supported nanoporous single-layer graphene membrane;

[0026] c) optionally, transferring the obtained supported N-functionalized graphene membrane from the mechanical support to a porous mechanical support.

[0027] The present invention also discloses a gas selective filter comprising an N-functionalized graphene membrane, which comprises N-functionalized nanopores, and on average 16 to 20 pyridine Ns at each pore edge.

[0028] Furthermore, the present invention also discloses a use of a gas selective filter comprising the graphene membrane of the present invention for gas separation, in particular for separating N2 and / or CO2.

[0029] In a preferred embodiment, for the gas selective filter of the present invention, the CO2 permeability of the graphene membrane is about 200 to 50,000 GPU, depending on the porosity of the graphene.

[0030] In a preferred embodiment, for the gas selective filter of the present invention, the N2 permeability of the graphene membrane is about 3 to about 2,500 GPU, depending on the porosity of the graphene (e.g., 100 GPU).

[0031] In a preferred embodiment, for the gas selective filter of the present invention, the CO2 / N2 is from about 20 to about 2,000, depending on the CO2 concentration and the pore size of the graphene.

[0032] Furthermore, the present invention also discloses a use of a gas selective filter comprising the graphene membrane of the present invention for gas separation, in particular for separating N2 and / or CO2.

[0033] Other features and advantages of the present invention will be apparent from the claims, the detailed description and the drawings. Description of the Drawings

[0034] Figure 1 Shows the method for preparing the gas selective separation filter of the present invention.

[0035] Figure 2 Shows the chemical properties and structure of the N-functionalized graphene film of the present invention, which contains the pyridine N-substituted nanopores described in Example 3. (a) Schematic diagram of the N-functional groups (pyridine N substituted at the pore edge and -NH2 connected near the pore edge) on the monolayer graphene film and its CO2 adsorption route. (b) XPS spectra of N-functionalized graphene loaded with -NH2 groups and their derivatives, pyridine N, and Py.CO2. (c) Peak shift of the adsorption curve and desorption curve (after heat treatment at 150 °C) of N-functionalized graphene. (d) Change in N-functional group density before and after heating at 150 °C. Heating results in an increase in the density of -NH2 and pyridine N, but a decrease in the density of carbamate, ammonium salt, and Py.CO2. (e) Change in the density of N-functional groups and HCO3 - during the heating process at 150 °C, indicating that the desorption process follows the zwitterionic reaction mechanism. (f) Schematic diagram of the reaction mechanism of NH3 with epoxy and semiquinone groups in the graphene lattice (adapted from Vacchi et al., 2016, Nanoscale, Vol. 8, No. 28, pp. 13714–13721; Kolle et al., 2021, Chem. Rev., Vol. 121, No. 13, pp. 7280–7345).

[0036] Figure 3 Shows the CO2 adsorption-desorption phenomenon on N-functionalized graphene as described in Example 3. (a) Schematic diagram, (b) corresponding STM image, and (c) three-dimensional (3D) topography STM images of the pyridine N-substituted nanopores in the filled state, empty state, and refilled state during desorption and adsorption processes. In NAP-XPS, during the desorption and adsorption processes at a CO2 pressure of 20 mbar, the percentage changes of -NH2 and its derivatives (d) and pyridine N and Py.CO2 (e). (f) Quantitative analysis of the percentage change of N-functional groups during CO2 adsorption and desorption in NAP-XPS, indicating that the quantitative change of N-functional groups is similar to the CO2 adsorption reaction. The percentage data of each N-functional group are extracted from the XPS spectra.

[0037] Figure 4 Shows the AC-HRTEM and Raman spectroscopy characterization of N-functionalized graphene as described in Example 3. (a) EDS distribution maps (HAADF, carbon, nitrogen, and oxygen elements) in the AC-HRTEM image and (b) EDS spectrum of N-functionalized graphene, confirming the presence of N-functional groups on the graphene surface. (c) Raman spectra of pristine graphene and N-functionalized graphene functionalized at 20 °C and 80 °C. (d) I2D / I G Relationship diagram of intensity ratio and G peak position and (e) relationship diagram of 2D peak position and G peak position. The data are all extracted from Raman spectra. (f) Under the irradiation of TEM electron beam (electron dose 4.7×10 5 e - -2 ), the changes of clusters on O3-treated graphene and N-functionalized graphene indicate that the N-clusters on N-functionalized graphene have higher stability under electron beam.

[0038] Figure 5 Shows the gas transport and carbon capture performance characteristics of the N-functionalized graphene film as described in Example 3. (a) Permeation results of the graphene film for 20% CO2 / N2 mixed gas before and after 1.5 hours or 24 hours of N-functionalization treatment at 20 °C. Single gas permeabilities (b) and gas pair ideal selectivities (c) of the graphene film before and after N-functionalization treatment indicate that N-functional groups and CO2 reduce the pore size, resulting in a decrease in permeability and an increase in selectivity. (d) Single gas permeation results of N-functionalized graphene at 30 °C, 60 °C, and 100 °C, showing preferential adsorption of CO2. (e) Comparison of CO2 / N2 mixed gas separation performance with advanced post-combustion capture membranes available. The performance upper limit of the polymer membrane is shown in the figure as a reference (assuming a selective layer thickness of 1 μm). The black target area represents membranes with a separation factor higher than 20 and a CO2 permeability higher than 1000 GPU. (f) Stability test of N-functionalized graphene for carbon capture shows that the performance remains stable after 8 regeneration cycles. All gases were measured at a feed pressure of 2 bar.

[0039] Figure 6 Shows the CO2 adsorption and gas transport characteristics of the N-functionalized graphene of the present invention as described in Example 3 in a low-concentration CO2 atmosphere. (a) The experimental Py.CO2 coverage extracted from XPS results is in good agreement with the theoretical Langmuir single-site isotherm. (b) Normalized gas permeation results of N-functionalized graphene measured at different CO2 partial pressures. The results are in good agreement with the gas transport model. (c) Carbon capture performance of N-functionalized graphene measured at 12 mbar and 8.4 mbar CO2 partial pressures; (d) Photograph of a centimeter-scale N-functionalized film on a flat polymer support. The red line marks the film edge. Detailed implementation mode

[0040] The term "graphene film" refers to a layer of graphene, especially monolayer graphene, such as that prepared by the CVD method. For example, the thickness range of monolayer graphene film is about 0.34 to 1 nm. However, in the embodiments of the present invention, the graphene film may also comprise bilayer graphene or a partial bilayer graphene structure. It should be noted that achieving highly uniform monolayer graphene in the surface region of the film may not be highly efficient for industrial-scale film manufacturing.

[0041] The term "N-functionalization" refers to the introduction of N atoms in the form of pyridine N into the graphene lattice. In the present invention, N-functionalized nanopores are achieved by the reaction of O-functional groups (such as semiquinone group (C=O), epoxy group) on the nanoporous monolayer graphene film with a nitrogen source (such as ammonia, nitrogen plasma or N-containing small molecules, such as hydrazine, ethylenediamine, etc.).

[0042] The term "O-functional group" refers to the groups obtained by O3 treatment, including semiquinone group, ether group and epoxy group.

[0043] The term "sacrificial support layer" refers to a support suitable for the graphene film (such as copper, nickel, platinum or other metal substrates on which monolayer graphene can be synthesized), especially a non-porous support, which can be removed before or after the graphene film is transferred to a structural (mechanical) support.

[0044] The pore density described in the present invention is usually measured by high-resolution transmission electron microscopy (HRTEM), and the pore size (van der Waals gap), O / C ratio and the average number of nitrogen atoms at the pore edge are determined by the combined technique of HRTEM and X-ray photoelectron spectroscopy (XPS). XPS is a common method for analyzing the elemental (O, C) composition in surface science.

[0045] The term "membrane performance" refers to the combination of gas permeability and selectivity. In the field of gas separation, it is generally considered that a CO2 permeability of 1000 GPU and a CO2 / N2 selectivity of 20 or more indicate good membrane performance. In addition, an O2 / N2 selectivity of 3 or more is also regarded as good membrane performance (see Kiwon et al., 2019, *Angew. Chem. Int. Ed.*, Vol. 131, pp. 16542-16546).

[0046] According to a specific embodiment, a method for preparing a gas selective filter is provided, comprising:

[0047] a) providing a supported nanoporous monolayer graphene film, the film comprising a pore density of about 10 10 cm -2 to about 10 13 cm -2The nanopores, which are modified with O-functional groups and have an O / C molar ratio of about 0.01 to about 0.5, and the single-layer graphene film is supported on a mechanical support;

[0048] b) N-functionalize the graphene nanopores of the supported nanoporous single-layer graphene film;

[0049] c) Optionally, transfer the obtained supported N-functionalized graphene film from the mechanical support to a porous mechanical support.

[0050] According to a specific embodiment, the mechanical support is a sacrificial support (such as an annealed copper foil).

[0051] If the single-layer graphene film is prepared on a sacrificial support, the sacrificial support needs to be replaced with a porous mechanical - non-sacrificial support before or after the N-functionalization treatment, and the replacement is achieved through the following steps: i) coat the surface of the single-layer graphene film on the sacrificial support with a porous support, such as a polymer layer having a high gas permeability (about 5×10 4 to 6×10 6 GPU) as described in the present invention; then ii) remove the sacrificial support by wet etching.

[0052] According to a specific embodiment, the porous mechanical support is a thin layer (usually with a thickness of 100 to about 1000 nm) porous polymer film or a nanoporous carbon film, with a porosity of 10% to about 50%.

[0053] Typical porous mechanical supports can be selected from the group consisting of: poly[1-(trimethylsilyl)propyne] (PTMSP) layer, amorphous fluoropolymer layer such as TEFLON AF 2400, nanoporous carbon (NPC) film, carbon nanotube network, or a combination thereof.

[0054] According to another specific embodiment, the supported nanoporous single-layer graphene film contains O-functional group-modified nanopores, which can be obtained by the following method: introducing O-functional groups into the graphene lattice by subjecting the nanoporous single-layer graphene film to O3 treatment. Typically, introducing O-functional group-modified nanopores into the single-layer graphene film can be achieved by a precisely controlled O3 millisecond etching process. For example, see Hsu et al., 2021, *ACS Nano*, Vol. 15, No. 8, pp. 13230 - 13239). For example, first expose the single-layer graphene at an O3 pressure of 250 °C and 13.5 torr to obtain the target pore density and size (such as a pore density of about 10 10 to about 10 13 cm -2, the van der Waals gap within the pores is less than 1 nm, typically about 0.25 to about 1 nm); subsequently, O-functionalization is carried out at 20 °C for 1 - 2 hours to obtain O-functionalized porous single-layer graphene (usually the O / C molar ratio is about 0.01 to about 0.5, preferably about 0.01 to about 0.2).

[0055] Another method is to treat single-layer graphene with O3 at about 25 to about 80 °C for about 0.01 to about 10 hours to generate pores, and then heat at above 100 °C (usually about 100 °C to about 200 °C) for about 1 hour for O-functionalization of the pores, and a nano-porous single-layer graphene film functionalized with O functional groups with an O / C molar ratio of about 0.01 to about 0.5 can be obtained.

[0056] According to a specific embodiment, to achieve the above O-functionalization, single-layer graphene can be treated with O3 in O2 with an O3 content of 1 - 20% at the functionalization temperature (0 - 80 °C) and reaction time (0.01 - 10 hours).

[0057] The present invention is based on the following discovery: when at least 16 - 20 O-functional groups are provided at the edge of each pore, these groups will react with the nitrogen atoms of a nitrogen source (such as NH3) to form pyridine groups.

[0058] As shown in the figure, especially with reference to Figure 1 , a schematic diagram of a preparation method of a gas selective filter is provided.

[0059] More specifically, Figure 1 The steps of the illustrated embodiment include:

[0060] a1) Provide a single-layer graphene film on a sacrificial support

[0061] According to a specific embodiment, a single-layer graphene film can be prepared on a copper foil by chemical vapor deposition (CVD) to obtain a single-layer graphene film on a mechanical support.

[0062] a2) Form nano-pores in the single-layer graphene film on the sacrificial support by O3 etching

[0063] Subsequently, precise controlled O3 millisecond etching treatment is carried out on the single-layer graphene film on the sacrificial support to introduce nano-pores into the single-layer graphene film in a well-controlled manner.

[0064] Generally, the supported nano-porous single-layer graphene film needs to be treated with an O3 / O2 mixed gas (for example, 1 or 20 wt%, 1 - 1000 ml / min, treated at about 20 °C to about 80 °C for about 0.01 to about 10 hours).

[0065] According to a specific embodiment, the obtained nanoporous single-layer graphene film needs to be further treated at high temperature in an H2 atmosphere to smooth the Cu, facilitating the subsequent removal of the sacrificial support layer.

[0066] According to a preferred embodiment, a single-layer graphene film with nanopores is obtained (the pore density ranges from 10 10 to 10 13 cm -2 and the van der Waals gap in the pores is less than 1 nm).

[0067] a3) Perform O-functionalization of the graphene nanopores on the supported nanoporous single-layer graphene film

[0068] According to a specific embodiment, perform epoxy functionalization of the graphene lattice on the supported nanoporous single-layer graphene film, for example, by treating with O3 for about 0.01 to about 10 hours (e.g., about 30 to about 60 minutes).

[0069] Generally, place the supported nanoporous single-layer graphene film in an O3 / O2 mixed gas (e.g., 1 or 20 wt%, 1 - 1000 ml / min, at a temperature of about 100 to about 200 °C for about 1 hour).

[0070] According to a specific embodiment, a nanoporous single-layer graphene film is obtained, the film contains nanopores (the pore density is in the range of 10 10 cm -2 to 10 13 cm -2 and the van der Waals gap in the pores is less than 1 nm), the nanopores are O-functionalized, and the O / C molar ratio is about 0.01 to 0.5, wherein the single-layer graphene film is supported on a sacrificial mechanical support. The pore density of the nanoporous single-layer graphene film and the degree of O-functionalization of the pores can be determined by scanning tunneling microscopy (STM) XX and X-ray photoelectron spectroscopy (XPS), respectively.

[0071] a4) Prepare a single-layer graphene film on a porous mechanical support layer

[0072] According to another specific embodiment, before removing the sacrificial support layer, first coat a mechanical reinforcement layer on the single-layer graphene film on the sacrificial support layer. In particular, before removing the sacrificial support layer by wet etching, coat (e.g., by spin-coating) a porous polymer layer on the single-layer graphene film on the sacrificial support, such as those mentioned in the present invention.

[0073] According to a specific embodiment, the sacrificial support is a copper foil, and a single-layer graphene film on the copper foil is spin-coated with a layer of poly[1-(trimethylsilyl)propyne] (PTMSP) of about 0.2 to 1.0 μm.

[0074] According to a specific embodiment, the spin-coating of the single-layer graphene film on the sacrificial support is carried out using a gas-permeable polymer solution, and the concentration of the polymer in a solvent (such as toluene) is 1.25 to about 3 wt% (such as 1.25 wt%).

[0075] According to another specific embodiment, the sacrificial support is removed from the single-layer graphene film coated with the polymer, and then the obtained single-layer graphene film coated with the polymer is transferred onto a porous mechanical support such as porous tungsten (for example, the porosity ranges from 2% to 80%).

[0076] b) N-functionalize the graphene nanopores of the supported nanoporous single-layer graphene film

[0077] According to another specific embodiment, a mechanically supported nanoporous single-layer graphene film containing O-functionalized nanopores with an O / C molar ratio of about 0.01 to 0.5 is placed in a gaseous NH3 atmosphere, thereby introducing pyridine groups at the edges of the graphene pores.

[0078] According to another specific embodiment, the N-functionalization reaction is carried out in saturated ammonia vapor.

[0079] According to another specific embodiment, the N-functionalization reaction is carried out at a temperature of about 5 to 100 °C, typically 20 to 80 °C.

[0080] Typically, the N-functionalization reaction lasts for about 0.01 to 72 hours, such as 0.5 to 24 hours.

[0081] For example, the reaction is carried out for 24 hours at 20 °C and about 0.8 bar, or for 1.5 hours at 80 °C and about 3.6 bar.

[0082] The N-functionalization of the graphene film is achieved by the reaction of NH3 with the O-functional groups, forming 16 to 20 pyridine Ns at the edge of each pore and introducing amino groups around the pores.

[0083] According to another specific embodiment, an N-functionalized nanoporous single-layer graphene film obtainable by the method of the present invention is provided.

[0084] According to another specific embodiment, an N-functionalized nanoporous single-layer graphene film is provided, wherein the nanopores are N-functionalized and 16 to 20 pyridine Ns are introduced at the edge of each pore.

[0085] According to one specific embodiment, the gas selective filter in the present invention has such a pore size distribution that the pores in graphene have van der Waals gaps less than 1 nm and greater than 0.25 nm.

[0086] According to another specific embodiment, an N-functionalized nanoporous single-layer graphene membrane is provided, wherein the density of N-functionalized pores is about 10 10 cm -2 to 10 13 cm -2 .

[0087] According to another specific embodiment, an N-functionalized nanoporous single-layer graphene membrane is provided, wherein pyridine functionalization accounts for about 1% to about 80% of the total N-functionalization of the nanopores (for example, about 16% N-functionalization at 25 °C and about 50.3% N-functionalization at 80 °C). Generally, the degree of pyridine functionalization can be determined by X-ray photoelectron spectroscopy (XPS).

[0088] According to another specific embodiment, an N-functionalized nanoporous single-layer graphene membrane is provided, wherein pyridine functionalization accounts for about 10% to about 55% of the total N-functionalization of the nanopores.

[0089] According to another specific embodiment, after removing the sacrificial support and providing enhanced support by known techniques described in, for example, previous reports (Huang et al., 2018, Nat. Commun., Vol. 9, p. 2632) and WO 2019 / 175162, the N-functionalized graphene membrane can be assembled into a gas filter module.

[0090] According to one specific embodiment, the gas selective filter in the present invention can preferably be used for carbon capture (such as CO2 / N2 separation).

[0091] According to one specific embodiment, the gas selective filter of the present invention has an N2 permeability of about 3 to about 2500 GPU, depending on the porosity of the graphene (for example, 100 GPU).

[0092] According to one specific embodiment, the gas selective filter of the present invention has a CO2 permeability of about 200 to about 50,000 GPU, depending on the porosity of the graphene and the CO2 concentration in the feed, for example, from 2500 to about 20,000 GPU, preferably from 8000 to about 18,000 GPU (for example, 10,000 GPU).

[0093] According to a specific embodiment, the gas selective filter of the present invention has a CO2 / N2 selectivity of about 20 to about 2,000 (e.g., 100), depending on the CO2 concentration in the feed and the pore size of the graphene.

[0094] In a preferred embodiment, the gas selective filter of the present invention has a CO2 / N2 selectivity of about 100 to about 2,000 (e.g., 400) at low CO2 feed (such as 0.5 - 1% CO2).

[0095] In a preferred embodiment, the gas selective filter of the present invention has a CO2 / N2 selectivity of about 20 to about 200 (e.g., 100) at high CO2 feed (such as 10 - 20% CO2).

[0096] The gas selective filter of the present invention has significant CO2 permeability and CO2 / N2 selectivity, and can be used as a valuable tool for capturing CO2 from flue gases (such as in the steel and cement industries), because this combination of properties will significantly reduce the energy consumption and cost of carbon capture.

[0097] In these examples, the binding of CO2 to two-dimensional graphene pores containing pyridine N-substituted nanopores and amines was fully supported spectroscopically and directly observed microscopically. It is believed that compared with N2, the binding of CO2 to the pyridine N groups at the pore edges facilitates the competitive transport of CO2, thus resulting in an attractive combination of high CO2 permeability and high CO2 / N2 selectivity.

[0098] High permeability can reduce the membrane area required to process a certain volume of gas mixture, thus reducing the investment cost of the separation process. High CO2 / N2 selectivity is crucial for low-feed-pressure separation applications such as post-combustion capture.

[0099] After the description of the present invention, the following examples are for illustration only and not for limitation.

[0100] Examples

[0101] Example 1: General method for preparing graphene membranes of the present invention

[0102] The method for preparing the gas selective filter of the present invention is as Figure 1 shown in A and B and is described in detail as follows.

[0103] a) Provide a supported nanoporous single-layer graphene membrane, the membrane having a pore density of 10 10 cm -2 to 10 13 cm -2Nanopores within a certain range, with a van der Waals gap in the pores less than 1 nm, having O functional groups, an O / C ratio of 0.01 - 0.2, and the monolayer graphene membrane is supported on a mechanical support.

[0104] The above monolayer graphene membrane can provide sacrificial or non - sacrificial mechanical support. Typical non - sacrificial mechanical supports include polymeric supports such as poly[1 - (trimethylsilyl)propyne] (PTMSP), polytetrafluoroethylene (PTFE), known as TEFLON TM , or nanoporous carbon membranes, carbon nanotube networks, or combinations of the above materials.

[0105] Typical sacrificial mechanical supports include annealed copper foils, and the monolayer graphene membrane is deposited on the copper foil by methods such as CVD.

[0106] If the above monolayer graphene membrane is placed on a sacrificial support, the sacrificial support can be replaced with a non - sacrificial mechanical support by i) coating a support layer (such as a polymeric hydrophobic layer) on the monolayer graphene membrane on the sacrificial support, and then ii) removing the sacrificial support by wet etching.

[0107] The supported nanoporous monolayer graphene membrane containing O - functionalized nanopores can be obtained by subjecting the nanoporous monolayer graphene membrane to O3 treatment to introduce O functional groups onto the graphene lattice.

[0108] b) Subject the supported nanoporous monolayer graphene membrane to N - functionalization of graphene nanopores

[0109] Then, the above - mentioned supported nanoporous monolayer graphene membrane is subjected to N - functionalization of graphene nanopores by reacting the O functional groups (such as epoxy groups) of the nanoporous monolayer graphene membrane with a nitrogen source (such as ammonia or nitrogen plasma or N - containing small molecules such as hydrazine, ethylenediamine, etc.). For example, the N - functionalization of nanopores is carried out under the action of gaseous ammonia. The gaseous ammonia can be provided in the form of gaseous ammonia or saturated ammonia vapor formed in a mixture of NH3 and methanol (such as 7N NH3 / methanol). For large - scale production, pure gaseous ammonia in a gas cylinder can be directly used for N - functionalization. N - functionalization of the membrane by gas - phase reaction can be achieved in large - batch or roll - to - roll processes, as described in (Bae et al., 2010, Nat. Nanotechnol., Vol. 5, 574 - 578).

[0110] The reaction can be carried out at a temperature of about 5 to 100 °C, preferably around room temperature, about 10 to about 80 °C (such as about 20 °C).

[0111] The reaction can be carried out at an ammonia pressure of 10 mbar to about 50 bar, usually about 1 bar to 10 bar.

[0112] The functionalization reaction with N is based on the following reaction route: NH3 reacts with epoxy groups on the graphene lattice to form primary amines (-NH2) through ring-opening chemical reactions (Vacchi et al., 2016, ibid.). -NH2 forms carbamate (-NHCOO-) and -NH3 through zwitterionic reactions with CO2 and H2O + (Equations 1 and 2) (Kolle et al., 2021, ibid.). The addition of pyridine and graphitic N is achieved by the reaction of semiquinone groups at the pore edges with NH3 (Wang et al., 2016, J. Phys. Chem., Vol. 120, No. 10, pp. 5673 - 5681). In the case of single vacancy defects, the semiquinone groups will be replaced by N atoms bonded to three adjacent carbon atoms Figure 2 (f)). This results in the doping of N (graphitic N) on the graphene lattice, which heals the single vacancy defects (Bigras et al., 2020, 2D Mater. Appl., Vol. 4, No. 1, p. 42). On the other hand, pores larger than single vacancies should produce pyridine N. The smaller shoulder peak at ~403.1 eV is expected to be Py.CO2 (CO2 adsorbed at pyridine N sites).

[0113] Example 2: Method for preparing a graphene membrane according to the present invention

[0114] The method for preparing a gas selective filter according to the present invention is described as follows.

[0115] a) Provide a single-layer graphene membrane on a sacrificial support

[0116] Synthesize a single-layer graphene membrane on a copper foil by chemical vapor deposition (CVD) as follows. Use an annealed copper foil to synthesize single-layer graphene (SLG) by a low-pressure CVD process (Rezaei et al., 2020, J. Memb. Sci., Vol. 612, 118406). Place the copper foil (Strem Chemicals Inc., purity 99.9%, 50 μm) into a furnace heated to 1'000 °C in a 700 torr CO2 atmosphere to remove organic contaminants. Then evacuate the CO2, and then introduce a mixture of H2 and Ar (the ratio of H2 and Ar is 1:10) into the furnace at 700 torr. Raise the furnace temperature to 1075 °C and maintain it for 4 h after the pressure stabilizes. Finally, cool the furnace to 1'000 °C at a rate of 0.1 °C min -1 -1. Finally, obtain the copper foil. Subsequently, introduce 24 sccm of CH4 and 8 sccm of H2 into the furnace, and the system pressure is 460 mtorr. After reacting for 30 minutes, turn off the CH4 flow, and let the furnace cool naturally. Take out the generated single-layer graphene film from the furnace for further processing.

[0117] b) Forming nanopores in the single-layer graphene film on the sacrificial support by O3 etching

[0118] Then, pores are formed in the obtained single-layer graphene film in two steps: high-temperature O3 etching to incorporate pores, and low-temperature O3 exposure to further enlarge the pores. This is a well-controlled O3 millisecond etching process (Hsu et al., 2021, ibid.) to add nanopores to the single-layer graphene film in a well-controlled manner. Then, the obtained nanoporous single-layer graphene film is further treated at high temperature in an H2 atmosphere to smooth the Cu for subsequent removal of the sacrificial support.

[0119] In short, the synthesized single-layer graphene / copper is placed in a self-made millisecond gasification reactor, which is connected to a vacuum pump. The system continuously pumps air to exclude gases. The reactor is heated to 250 °C with an H2 flow rate of 20 sccm. Then, the Ar gas flow is interchanged with the H2 gas flow to maintain an inert environment. A controlled ozone flow (9 wt% O3 by mole) is added to the millisecond gasification reactor. After reaching the required O3 conditions (maximum pressure: 150 torr, etching time: 3 s), the system is naturally cooled to room temperature with an Ar flow.

[0120] The obtained single-layer graphene film has nanopores (3.0×10 12 cm -2 , and the van der Waals gap in the pores is less than 1 nm). Then, the O3-treated single-layer graphene film supported on copper is treated at high temperature (such as 500 °C) for 60 minutes in an H2 atmosphere.

[0121] c) Performing O-functionalization of graphene nanopores on the supported nanoporous single-layer graphene film

[0122] The supported nanoporous single-layer graphene film is treated with O3 to achieve epoxy O-functionalization (such as epoxy functionalization) of the graphene lattice.

[0123] In short, the obtained single-layer graphene film supported on copper is placed in a chamber with an Ar gas flow to purge the residual air. Then, the Ar gas flow is turned off, and an O3 / O2 mixed gas flow (9 wt%) is injected at 20 °C. After the required reaction time (0.5 - 2 hours), the O3 gas flow is interchanged with the Ar gas flow. Then, the O3-treated graphene on the copper foil is taken out of the chamber.

[0124] The oxidation of graphene under controlled conditions forms O clusters around the pores, where the chemical properties of O are epoxy, ether, and semiquinone groups (Li et al., 2022, *JACS Au*, Vol. 2, pp. 723 - 730). The latter only exists at the pore edges, restricted by the maximum number of covalent bonds of C atoms. A nanoporous single-layer graphene membrane is obtained, which includes nanopores with a pore density between 10 10 cm -2 to 10 13 cm -2 , the van der Waals gap in the pores is less than 1 nm, functionalized with epoxy groups, and the molar O / C ratio is between 0.01 - 0.2, where the single-layer graphene membrane is supported on a sacrificial mechanical support. As shown in Figure 2 (f), in the obtained nanoporous single-layer graphene membrane, the nanopores are surrounded by O functional groups, and the O functional groups react with NH3 to form pyridine and -NH2 groups around the pores.

[0125] d) Remove the sacrificial support to prepare a single-layer graphene film on a porous mechanical support layer

[0126] Coat a mechanical reinforcement layer on the single-layer graphene film on the sacrificial support, and then perform wet etching to remove the sacrificial support.

[0127] Briefly, graphene staying on a copper foil is spin-coated (1'000 rpm for 30 s, then 2'000 rpm for 30 s) with a gas-permeable polymer (e.g., a solution of 1.25 wt% poly[1-(trimethylsilyl)propyne] (PTMSP) in toluene). The resulting PTMSP film (~250 nm thick) is dried in a fume hood at room temperature for 12 h and then dried under vacuum for another 12 h. Then, the single-layer graphene film coated with PTMSP on copper is floated in an aqueous 1M FeCl3 etching solution for 30 minutes to remove the copper. This leaves a floating graphene / PTMSP film on the water surface, which is then transferred to a porous support as follows:

[0128] The obtained PTMSP-coated single-layer graphene film (without the sacrificial support) is soaked in a 1M hydrochloric acid bath and a deionized water bath for 1 hour each. The washed film is scooped up by a porous tungsten support (bearing 5-μm laser-drilled holes), as described previously (Huang et al., 2018, *Nat.*, Vol. 9, No. 1, p. 2632). Before measurement, the obtained supported single-layer graphene membrane is dried on the tungsten support in a fume hood for at least 12 hours. An O-functionalized nanoporous single-layer graphene membrane is obtained, where the single-layer graphene membrane is supported on a non-sacrificial porous mechanical support.

[0129] e) N-functionalize the graphene nanopores of the supported nanoporous single-layer graphene membrane

[0130] Place the obtained nanoporous single-layer graphene membrane with epoxy group-functionalized nanopores supported by, for example, a porous tungsten support, into a conical flask containing an ammonia / methanol solution (7N), and conduct a gas-phase NH3 functionalization reaction. Before sealing the flask, evacuate the air inside the flask with a pump to ensure that the NH3 partial pressure reaches the required concentration.

[0131] React the saturated ammonia vapor with the supported nanoporous graphene at 20 °C or 80 °C for the required time. For example, react at 20 °C and about 0.8 bar for about 24 hours, or react at 80 °C and about 3.6 bar for about 1.5 hours. The N-functionalization of the graphene membrane is carried out by gradually replacing oxygen atoms with nitrogen atoms at the edges of the nanopores. In this step, pyridine groups are incorporated into the pore edges. In addition, -NH2 groups are also incorporated.

[0132] After the reaction time ends, place the supported N-functionalized graphene membrane in a vacuum for 12 hours, and then heat it to 150 °C in a gas permeation device to remove the residual solvent. Cool the obtained membrane to the temperature required for measurement.

[0133] Then, the obtained membrane can be loaded into a membrane module for gas permeation studies as described in Example 2, where the membrane module consists of four-inch Swagelok VCR fittings, achieving a leak-free seal between metal and metal.

[0134] Example 3: Characterization of the obtained N-functionalized graphene membrane

[0135] Characterize the obtained N-functionalized graphene membrane by the following techniques: Measure the N-functionalized graphene sample with X-ray photoelectron spectroscopy (XPS) to confirm the presence of N-functional groups, and analyze it with the O-functional group single-layer graphene as a reference. After confirming the absence of N-functional groups on the graphene surface, place the O-functionalized single-layer graphene in NH3 vapor for N-functionalization. The presence of N-functional groups on the N-functionalized single-layer graphene is confirmed by XPS. On the functionalized membrane obtained by the method of the present invention, deposit O-functionalized graphene (step a4) or N-functionalized graphene membrane (step b) on the tungsten foil, use the monochromatic Kα line (1486.6 eV) of an aluminum X-ray source, and set the analyzer to a pass energy of 20 eV.

[0136] The preparation method of the single-layer graphene membrane is as described in Example 1 ( Figure 1The O-functionalized graphene steps a to a4 and N-functionalized graphene steps a to b) in [reference], and transfer is carried out during the membrane manufacturing process (etching away copper in an FeCl3 solution and then transferring the membrane onto a porous tungsten support). Before XPS measurement, the graphene sample membrane of the present invention is rinsed in toluene (three batches of toluene solution, 2 hours for each batch) and acetone solvent to remove the PTMSP reinforcement layer. Then, before loading the sample into the XPS chamber, the sample is heated in Ar gas at 150 °C to remove residual solvent. During measurement, the sample is electrically grounded to the sample stage and measured.

[0137] The adsorption-desorption measurement is carried out in near-ambient pressure XPS (NAP-XPS). A fresh N-functionalized graphene sample is placed in the NAP-XPS chamber, and the initial spectrum is measured before the desorption experiment (heating to 150 °C). Then, the N-functionalized graphene sample is heated to 150 °C under vacuum (10 - 9 mbar) conditions to desorb CO2 and H2O. The desorbed N-functionalized graphene sample is measured for the desorption spectrum.

[0138] Then, the desorbed sample is measured again. Subsequently, 20 mbar of CO2 is introduced into the NAP-XPS chamber at 30 °C, and CO2 adsorption is carried out for the required time (30 minutes). Then, the XPS chamber is evacuated before each measurement, and the sample is allowed to stand at 30 °C for 1 hour. Subsequently, the corresponding spectra are collected. The sample is further desorbed by heating at the required temperature for 1 hour in an ultra-high vacuum chamber (UHV). Then, the spectra are measured. All spectral fittings are carried out using CasaXPS software.

[0139] The sample treated with NH3 is analyzed by X-ray photoelectron spectroscopy (XPS), and a broad N1s peak appears ( Figure 2 b). The deconvolution of the high-resolution N1s spectrum ( Figure 2b) The addition of pyridine N (398.1 eV) was confirmed (Li et al., 2009, J. Am. Chem., Vol. 131, No. 43, 15939 - 15944). Graphitic N and primary amine (-NH2) were also detected at 401.7 eV (He et al., 2014, Angew. Chemie Int. Ed., Vol. 53, No. 36, pp. 9503 - 9507) and 399.5 eV (Compton et al., 2010, Adv. Mater., Vol. 22, No. 8, 892 - 896), respectively. In addition, functional groups formed by the adsorption of atmospheric CO2 on pyridine N and -NH2 sites were also detected. These groups are pyridine N related to CO2 (hereinafter referred to as Py.CO2, ~403.1 eV) (He et al., 2014, ibid.), carbamate (-NHCOO-, ~400.3 eV) (Compton et al., 2010, ibid.), and ammonium (-NH3 + , ~401.7 eV) (Navaee et al., 2015, RSC Adv., Vol. 5, No. 74, 59874 - 59880). The 400.3 eV peak can also be attributed to pyrrolidine N (Tian et al., 2020, Nat., Vol. 11, No. 1, 388). However, the mass balance of the total density of -NH2 and its derivatives (carbamate and ammonium) during adsorption and desorption indicates that the amount of pyrrolidine N is negligible because pyrrolidine N involves a five - membered ring (5 - MR), while the graphene lattice has a honeycomb structure composed of six - membered rings (6 - MR).

[0140] Energy - dispersive X - ray spectroscopy (EDS, Figure 4 a) also confirmed the incorporation of N into the graphene lattice. Quantitative analysis of the EDS spectrum showed that N in the lattice accounted for approximately 9% of the total atoms (C and N), which is consistent with the XPS data. In sharp contrast to the rapid gasification of oxygen clusters in graphene, no gasification of clusters by the beam was observed, indicating that N functional groups have higher stability than O functional groups( Figure 4 (f)).

[0141] Raman spectroscopy was used to probe the doping of pyridine N in the graphene lattice. As is well known, heteroatom doping changes the Fermi level and electron carrier density of the graphene lattice, which can be studied by Raman spectroscopy. As the doping density increases, the intensity ratio of the 2D peak to the G peak (I2D / IG) decreases, and the position of the G peak, wG, shifts (Froehlicher et al., 2015, Phys. Rev. B, Vol. 91, 205413).

[0142] Raman characterization was performed on the pristine graphene film and N-functionalized graphene film transferred onto tungsten. The transfer and preparation of the samples were the same as those for the XPS samples. The measurements were carried out using single-point data acquisition or mapping with a Renishaw micro-Raman spectrometer equipped with a blue laser (λL = 457 nm, EL = 2.71 eV). The Raman data were analyzed using MATLAB. To calculate the height and position of the D, G, and 2D peaks, the background was subtracted from the Raman data using the least-squares curve fitting tool.

[0143] Raman Spectroscopy Modeling

[0144] Functionalization and doping on the graphene lattice can change the Fermi level (electron carrier density) (Beams et al., 2015, J. Phys. Condens. Matter, Vol. 27, No. 8, p. 083002). A decrease (increase) in the Fermi level results in p-type (n-type) doping, thus affecting the position of the G and 2D peaks and the intensity of the 2D peak. Therefore, the change in the peak is an indicator of the doping concentration dependence. The following is the principle of Raman shift.

[0145] 1. As the doping concentration increases, the G peak shifts to the right.

[0146] 2. As the doping concentration increases, the intensity of the 2D peak is suppressed. However, the G peak is insensitive to it. Therefore, I 2D / I G will decrease as the electron carrier density concentration increases.

[0147] To establish a doping concentration model by Raman spectroscopy, graphene was transferred onto tungsten, and Raman spectroscopy measurements were carried out using a blue laser (457 nm). At least 50 points were collected for each sample. After subtracting the spectral background, the characteristics of the G and D peaks, such as intensity, peak position, and FWHM (full width at half maximum), were fitted with Gaussian curves.

[0148] The blue shift of the 2D peak indicates that O3-treated graphene and N-functionalized graphene are p-type doped (Froehlicher et al., 2015, Phys. Rev. B, Vol. 91, 205413). For p-type doping, regardless of the presence of defects on the graphene lattice, the G peak shift can be used to calculate the doping concentration 11 . The Fermi level of p-type doped graphene can be estimated by Equation 1.

[0149] E F = -18Δω G -83 (1)

[0150] where E F represents the change in the Fermi level caused by functionalization, in meV; Δω GIndicates the shift of the G peak, in cm -1 .

[0151] The ω of pristine graphene, O3-treated graphene, and N-functionalized graphene (20 and 80 °C) G are 1584, 1588, 1588, and 1592, respectively. The E of O3-treated graphene and N-functionalized graphene (20 and 80 °C) F are -155.6, -155.6, and -226.0 meV, respectively. The electron carrier density (m -3 ) is converted from E F (Table S3) by Equation 2.

[0152]

[0153] where n is the electron carrier density (cm -3 ), ν F is the Fermi velocity (10 6 m / s), and h is the Planck constant.

[0154] Since the thickness of graphene is 0.35 nm, the electron carrier density of graphene (cm -2 ) can be obtained. XPS measurements show that N-functionalized graphene treated at 20 °C and 80 °C contains 9% and 20% N, respectively. We note that each graphitic N doping contributes -0.42 mobile carriers to the graphene lattice. Assuming that the contribution values of N functional groups to mobile carriers are similar, the N-functionalized graphene at 20 °C (80 °C) should carry 1.2×10 14 (2.2×10 14 ) cm -2 of mobile carriers. In addition, it is also noted that the presence of O functional groups on the graphene lattice also helps to increase the electron carrier density. Therefore, the results of Raman spectroscopy are similar to the values calculated from XPS results.

[0155] The I 2D / I G ratio of graphene treated with NH3 decreases from above 2 (pristine lattice) to below 1 ( Figure 5 d), while w G also undergoes a blue shift. When the concentration of pyridine N increases, i.e., by raising the NH3 reaction temperature to 80 °C, this shift increases. These observations confirm that pyridine N induces p-doping of the graphene lattice. Based on the dependence of w G on the Fermi level, the electron carrier densities when treating NH3 at 20 and 80 °C are estimated to be 2.9×10 14 and 5.3×10 14 cm -2。The results of Raman are similar to those of XPS.

[0156] Aberration-corrected high-resolution transmission electron microscopy (AC-HRTEM) analysis was performed using a double-corrected Titan Themis 60-300 (FEI) equipped with a Wein-type monochromator. Monolayer graphene specimens were prepared on silicon nitride or gold TEM grids with laser-drilled holes of 0.6 - 1.2 μm. Paraffin was used as a support layer to transfer graphene onto the microscope grid. Then, it was removed from the grid by dissolving paraffin in heptane. The grids were annealed in a CO2 atmosphere at 900 °C to remove hydrocarbon contaminants in the atmosphere. Subsequently, the monolayer graphene on the TEM grid was treated with O3 at 66 °C for 10 minutes. Subsequently, it was heated in an Ar gas environment at 200 °C to form pores in the graphene lattice. O functionalization was carried out on graphene to introduce O functional groups. Then N functionalization was performed to add N functional groups at the pore edges.

[0157] During the acquisition process, an incident electron beam of 80 keV was used to suppress knock-on damage. AC-HRTEM images were processed with a high-pass filter and a Gaussian filter to reduce noise and improve contrast. Compositions were obtained using energy-dispersive spectroscopy (EDS) at 80 kV.

[0158] Scanning tunneling microscopy (STM) imaging experiments were carried out in a cryogenic STM (CreaTec Fischer & Co. GmbH). To avoid oxidation of Cu during ammonia treatment, HOPG was used instead of graphene on Cu. O3 and NH3 treatments were the same as those for the above graphene treatment. Before loading HOPG into the STM chamber, it was heated to 150 °C in an Ar gas environment to remove surface contamination. The imaging conditions were 77 and 4.2 K. Before imaging, an STM probe was prepared by cutting a commercial Pt / Ir wire (platinum: 90 wt%, diameter 0.25 mm; Alfa Aesar). WSXM software (Horcas et al., 2007, Rev. Sci., 2007, Vol. 78, No. 1, 13705).

[0159] The FEI Teneo (field emission 0.2 - 30 keV) scanning electron microscope with an acceleration voltage of 1–5 kV and a working distance of 2.5 - 4.0 mm was used to observe the cross-sectional morphology of the film. Three samples were imaged for the average value and standard deviation of the film thickness.

[0160] CO2 adsorption and desorption analysis on N-functionalized graphene

[0161] Interestingly, the adsorption of CO2 on the pores can be observed by LTSTM imaging ( Figure 3b). For this purpose, a highly oriented pyrolytic graphite (HOPG) substrate was used for O3 treatment to avoid the rough surface caused by oxidation. Then, N-functionalization was carried out on the O3-treated graphene to obtain N-functionalized HOPG. The resulting pyridine N-substituted pores were integrated into the highly oriented pyrolytic graphite (HOPG) substrate. The substrate was exposed to the atmosphere for 10 minutes and then inserted into the LTSTM at 4K for imaging. Several light clusters could be observed in the image ( Figure 3 b).

[0162] These clusters have a size of 2 - 3 nm and a density of 3.0×10 12 cm -2 , which is comparable to the size and density of the clusters on the oxidized lattice before pyridine substitution. These clusters are basic pores surrounded by oxygen functional groups (the superstructure of epoxy trimers) formed by lattice oxidation (Li et al., 2022, ibid.). Some of the epoxy groups in the clusters react with NH3 to be converted into -NH2. Interestingly, although the clusters were observed to have a ring structure before NH3 treatment, which is consistent with the presence of pores in the center of the clusters, no ring structure was observed after NH3 treatment ( Figure 3 b). A possible explanation is that the pores may be occupied by CO2, thus masking the pores during imaging. To confirm this, the specimen was heated to 150 °C to desorb CO2. In fact, clusters with a ring morphology with cavities appeared after heating. To confirm whether this morphological change was due to reversible CO2 desorption or irreversible cluster gasification, the clusters were exposed to the atmosphere for 10 minutes again and the sample was imaged again. Indeed, the resulting clusters looked filled ( Figure 3 b). It can be easily seen from the three-dimensional height profile ( Figure 3 c) that the clusters showed evacuation and filling phenomena when heated and exposed to the atmosphere, respectively. The cluster area accounts for about 9% of the total area. Since the ratio of N and O is higher than this ratio, the clusters are likely formed by the stacking of functional groups.

[0163] The reversible adsorption and desorption of CO2 on the pyridine N-substituted pores can also be demonstrated in the ultra-high vacuum chamber of near ambient pressure XPS (NAP-XPS, Figure 2 d). The specimen was exposed to a CO2 environment at 30 °C and 20 mbar for 30 minutes for CO2 adsorption. The desorption experiment was carried out by heating the specimen at 150 °C in UHV for 1 hour. Repeated adsorption and desorption cycles showed that the adsorption of CO2 was quantitatively reversible ( Figure 3 d - e). No capacity loss was observed. The total number of pyridine N and -NH2 sites remained unchanged. During adsorption (desorption), the increase (decrease) in the Py.CO2 density was numerically similar to the decrease (increase) in the pyridine N density ( Figure 3f). -NH2 and its derivatives (-NHCOO- and -NH3 + ) are also the case. Since there is no water in the ultra-high vacuum chamber, the formation of HCO3 - was not observed.

[0164] XPS data show that the N sites account for 9% of the C sites in graphene, i.e., the N density is 3.8×10 14 cm -2 . According to the STM images, this is consistent with the fact that the clusters occupy about one-tenth of the graphene area. Further mass balance shows that pyridinic N accounts for 13.3% of the total N sites, so the pyridinic site density is 5.0×10 13 cm -2 . Given that the pore (cluster) density is 3.0×10 12 cm -2 , this corresponds to an average of 16 pyridinic N per pore.

[0165] The incorporation of N into the graphene lattice can be explained as follows. Pyridinic N and graphitic N are generated by the reaction of NH3 with the semiquinone group at the pore edge ( Figure 2 (e)). The -NH2 group is generated by the ring-opening chemical reaction of NH3 with the epoxy group.

[0166] Pyridinic N acts as a Lewis base and undergoes a complexation reaction with CO2. The -NH2 sites chemically adsorb CO2 to form derivatives (Reactions 1 and 2, Figure 2 a). Briefly, two adjacent -NH2 groups bind to CO2 to form -NHCOO- and -NH3 + (Reaction 1). In the presence of adsorbed water, HCO3 - is also formed (Reaction 2). In fact, the C1s XPS spectrum of the sample shows a shoulder peak at 289.0 eV, which can be attributed to HCO3 - (Lao et al., 2019, Angew. Chemie Int. Ed., Vol. 58, No. 16, pp. 5432 - 5437; Bezerra et al., 2014, Appl. Sci., Vol. 314, pp. 314 - 321).

[0167]

[0168] Heating to 150 °C for 60 minutes in the ultra-high vacuum (UHV) chamber of the XPS allows the desorption of CO2 associated with the pyridinic N sites. This is manifested by the complete disappearance of the Py.CO2 peak in the N1s spectrum ( Figure 2 ). The increase in the pyridinic N site density is quantitatively consistent with the disappearance of Py.CO2 ( Figure 2e). Heating also results in a decrease in the intensities of the -NHCOO- and -NH3 + peaks in the N1s spectrum and the HCO3 - peak in the C1s spectrum. Similarly, the increase in the density of -NH2 sites is quantitatively consistent with the loss of -NHCOO- and -NH3 + sites. These observations suggest that at 150 °C, CO2 is completely desorbed from the pyridine N sites, and at the same time, the CO2 chemisorbed on the -NH2 sites is also partially restored. This observation is consistent with the physisorption and chemisorption on the pyridine N sites and -NH2 sites.

[0169] Desorption of removed CO2 and H2O

[0170] After desorption, the intensities of -NHCOO-, -NH3 + / graphitized N, and Py·CO2 were analyzed. The intensity of CO2 decreased from 27.3%, 46.5%, and 4.9% to 18.8%, 35.7%, and 0.0%, respectively; at the same time, the peaks of -NH2 and pyridine N showed opposite trends ( Figure 2 d). In addition, the N1s peak shifted to the right, indicating a change in the chemical bond. This indicates that desorption removed some CO2 and H2O from the graphene surface.

[0171] CO2 adsorption-desorption cycle

[0172] It is expected that the directions of the zwitterionic reactions (Reactions 1 and 2) are to the right. Recovery was observed after CO2 desorption. Quantitative analysis ( Figure 2 d) showed that after CO2 adsorption, the amounts of -NHCOO-, -NH3 + / graphitized N, and Py·CO2 increased to 26.8%, 43.6%, and 7.4%, respectively. In contrast, the peaks of -NH2 and pyridine N decreased to 13.3% and 8.9%. Interestingly, the results showed consistency with the stoichiometry of Reactions 1 and 2. In addition, the percentage change in Py·CO2 was equal to that of pyridine N after CO2 adsorption. This indicates that the peak at ~403.1 eV is Py·CO2.

[0173] The N-functionalized graphene was heated to 150 °C under vacuum (10 -9 mbar) conditions for desorption. XPS measurements were carried out after the temperature was cooled to 30 °C. The N-functionalized graphene showed a reversible trend. -NHCOO-, -NH3 +The percentages of graphitized N and Py·CO2 decreased to 18.9%, 35.5%, and 0.0%, respectively. On the other hand, the peaks of -NH2 and pyridine N increased to 29% and 16.6%, respectively. The second CO2 adsorption–desorption cycle was carried out. The results were similar to those of the first CO2 adsorption cycle. The reversible trend indicates that a reversible zwitterionic reaction occurred during the adsorption–desorption process of N-functionalized graphene. Meanwhile, this also demonstrates the robustness of XPS fitting.

[0174] The overall chemisorption coverage of CO2 can be obtained from the total amount of N functional groups adsorbed by CO2. Among them, -NHCOO - was 26.9%, -NH3 + was 43.5, and Py·CO2 was 7.4. Therefore, the amounts of CO2 adsorbed through Reaction 1 and Reaction 2 were 26.9% and 16.6%, respectively. Considering that the adsorption amount of CO2 on pyridine N was 7.4%, the total adsorption amount of CO2 involved 50.9% of N atoms. Since N atoms accounted for 9% of the N-functionalized lattice and the carbon density was 3.8×10 15 cm -2 , the coverage of carbon dioxide was 1.7×10 14 molecules cm -2 .

[0175] Study on the temperature dependence of further desorption

[0176] By directly heating the N-functionalized graphene sample, the N functional groups can be adjusted. In the previous section, heating the N-functionalized graphene can desorb H2O and CO2 in the graphene lattice. As the temperature increases, the signals of -NHCOO - and -NH3 + gradually weaken. Desorption provides information on the amounts of graphitized N and -NH3 + . Since the percentage of -NH3 + must be greater than or equal to the percentage of -NHCOO− due to the two zwitterionic reactions, the amount of graphitized N (≤4.0) can be subtracted from the results at 150 °C.

[0177] Effect of NH3 treatment and its concentration

[0178] To understand the effect of pyridine N, in this invention, NH3 treatment was directly carried out on the nanoporous single-layer graphene film supported on 250 nm thick poly[1-(trimethylsilyl)propyne] (PTMSP). In this way, the changes in gas transport properties can be detected. As a control, the NH3 treatment was carried out on the independent PTMSP film. No change in the gas transport behavior of the PTMSP film was observed.

[0179] In contrast, for the obtained graphene membranes, in the gas permeation experiments described below, a sharp increase in CO2 / N2 permeability and a decrease in CO2 permeability were observed after NH3 treatment. The increase in selectivity is related to the treatment time, and the longer the treatment time, the greater the increase in selectivity, which is consistent with the increase in the concentration of pyridyl groups with the prolongation of the reaction time. When collecting gas permeation data as a function of kinetic diameter, a decrease in the permeability of all gases (He, H2, CO2, O2, N2, and CH4) was observed. However, the permeability of N2 decreased more significantly. Therefore, attractive H2 / N2, CO2 / N2, and O2 / N2 selectivities can be achieved along with relatively high permeabilities ( Figure 5 b and c). A possible reason for the decrease in permeability and the increase in gas pair selectivity is the narrowing of the pore diameter due to chemisorption (at -NH2 sites) or physical adsorption of CO2 (at pyridine N sites) near the pore orifice. Due to the strong adsorption of CO2 discussed above, it was found that the transport of all gases except CO2 is activated when measuring the gas transport as a function of permeation temperature.

[0180] NH3 treatment was carried out using a 7N NH3 / CH3OH solution. The concentration of NH3 in the gas phase was estimated by vapor-liquid equilibrium (VLE). The total pressure (P) of NH3 / CH3OH and the molar ratio of NH3 in the gas phase (y1) were extracted by curve fitting based on the literature data at different treatment temperatures (20 and 80 °C) (Feng et al., 199, *J. Chem. Eng. Data*, Vol. 44, No. 3, pp. 401 - 404; et al., 2007, *J. Chem. Eng. Data*, Vol. 52, No. 5, pp. 1653 - 1659). The relationship between the P point and the molar composition of NH3 in the liquid phase (x1) was fitted with a fifth-order polynomial equation, and the relationship between y1 and x1 was fitted with a linear equation. Then, P and y1 were obtained based on the known x1.

[0181] In the 20 °C VLE system, the molar concentration of NH3 in the gas phase (y1) was 0.94, and P in the 7N NH3 / CH3OH solution was 0.089 MPa. On the other hand, in the 80 °C VLE system, y1 was 0.77 and P was 0.47 MPa. Higher temperature (80 °C) and higher pressure promoted the doping of N, resulting in a higher total N percentage in N-functionalized graphene (9% and 20% at 20 °C and 80 °C, respectively). In addition, the percentage of pyridine N in the sample treated at 80 °C increased to 50.3% compared with 15.9% of the sample treated at 20 °C.

[0182] Gas Permeation

[0183] Single-component and mixed gas permeation tests were carried out in a self-made permeation module. The permeation device consists of a mass flow controller (MFC) and a mass spectrometer (MS, Hiden Analytical, HPR-20). The mass flow controller and the mass spectrometer were calibrated with an error of no more than 5%. The membrane module consists of a four-inch Swagelok VCR fitting, achieving a leak-free seal between metal and metal. In all measurements, the pressure difference between the permeate side and the feed side was maintained at 1 bar, and Ar gas (15 sccm, 1 bar) was used as the sweep gas to transport the permeated gas to the mass spectrometer. In the mixture experiments, CO2 with molar contents of 20%, 50%, and 80% in the N2 and CH4 mixture was used on the feed side. Saturated water vapor was introduced into the system together with the CO2 / N2 mixture (20 / 80%). The MS was used to track the typical gas permeation experiment in real time. Then, the performance of the membrane was calculated and reported based on the extracted steady-state results.

[0184] The permeability J of gas i is calculated according to the formula.

[0185] J = Q / (A·ΔP i ) (3)

[0186] where Q is the molar flow rate of gas i through the membrane, A is the active membrane area, and ΔP i is the transmembrane pressure difference of component i. The ideal selectivity of two gases i and l is calculated by dividing the permeability of gas i by the permeability of gas l. For the mixed gas permeation test, the separation factor of two gases i and l is calculated using Equation (4).

[0187]

[0188] To understand the application of pyridine N-substituted graphene membranes in post-combustion capture, CO2 / N2 mixtures with different concentrations of CO2 (20, 50, 80 vol%) in N2 were tested at 30 °C. Satisfactory separation performance was obtained in all cases, and the CO2 / N2 separation coefficient increased with the increase of the CO2 component, which may be due to the competitive adsorption of CO2 on N2. For the feed containing 20% CO2, the separation factor was continuously observed to be between 45 - 61. In addition, the N-functionalized graphene with a CO2 permeability close to 10000 GPU had a CO2 / N2 separation factor in the range of 23 - 85 ( Figure 5 e). For example, when the CO2 permeability was 12420 GPU, the CO2 / N2 separation factor could reach 52. A higher feed temperature (60 °C) could also produce attractive performance, with a CO2 / N2 separation factor of 22 and a CO2 permeability of 11800 GPU.

[0189] These membranes showed stable performance over several weeks. After continuous testing for 19.5 days at 30 °C under 20% CO2 / N2 mixed gas conditions, the CO2 permeability of the membrane gradually decreased as the CO2 / N2 separation factor increased. However, simply heating the membrane at 150 °C for 30 minutes could fully restore its performance. This indicates that the flux decline might be due to pore blockage events caused by fouling, and heating at 150 °C could successfully desorb the contaminants. To simulate flue gas conditions, a 20% (v / v) CO2 / N2 mixed gas and 2 bar saturated water vapor were used in the feed. Under these conditions, the CO2 permeability of the membrane was 17700 GPU at 30 °C, and the CO2 / N2 separation factor was 31. The membrane was also tested in a mixed gas with saturated water vapor at 60 °C. Under these conditions, the CO2 permeability was 10000 GPU, and the CO2 / N2 separation factor was 25. The membrane showed stable performance over several weeks( Figure 5 f). XPS measurements confirmed that in in-situ XPS, the N functional groups on the graphene lattice were retained after two heating and CO2 adsorption cycles. Membranes stored in the laboratory for more than a year (445 days) showed similar performance (CO2 concentration of 20%, feed pressure = 2 bar, temperature T = 30 °C (Table 1)).

[0190] Table 1

[0191]

[0192] To further understand the effect of CO2 adsorption on membrane performance, the density of the Py.CO2 complex was calculated using XPS, thus obtaining the functional relationship between the saturation of pyridine N sites and CO2 concentration. At a CO2 partial pressure of 20 mbar, the coverage of pyridine N was close to 45%. At a CO2 pressure of 1 bar, saturation of all sites (pyridine N and -NH2) was observed. Fitting these data to the Langmuir single-site adsorption isotherm gave an equilibrium constant K eq of 4.4×10 -4 Pa -1 ( Figure 5 a). At low CO2 pressures, the sharp increase in pyridine N sites was also reflected in the increase in CO2 permeability with increasing CO2 feed pressure( Figure 6 b).

[0193] When the pressure of CO2 is below 0.2 bar, the permeability of CO2 increases sharply. The permeability at low pressure (12 mbar) is several times that at 1 bar. This can be explained by the gas transport model described below. The number of gas molecules adsorbed on the pore edges determines the translocation rate and permeability. According to the Langmuir single-site isotherm, the amount of CO2 adsorbed on pyridine N-substituted nanopores is a function of pressure. Therefore, a pressure-dependent CO2 transport model can be derived.

[0194] Calculating the activation energy for gas transport in nanopores

[0195] The transport of gas in graphene nanopores proceeds via an activated transport mechanism. By fitting the gas flow rate varying with temperature using the Arrhenius relationship, E act-app the apparent activation energy can be obtained.

[0196]

[0197] where A0 is the pre-exponential factor of the Arrhenius term, C0 is the pore density, and E act-app is the activation energy for the migration of gas molecules in the pores (E act ) and the sum of the adsorption energy of gas molecules on the graphene lattice (E sur ). The gas permeation data of N-functionalized graphene membranes were calculated at 30, 60, and 100 °C.

[0198] Gas transport model through pyridine N-substituted nanopores

[0199] The transport of gas molecules on the graphene membrane follows the translocation mechanism. The gas transport is assumed to be first-order kinetics, i.e.,

[0200]

[0201] where k trans is the translocation coefficient of gas molecules, P is the permeability of each pore, N pore is the number of gas molecules at the pore opening, and p is the pressure difference between the feed and the permeate.

[0202] The amount of CO2 adsorbed on pyridine N-substituted nanopores can be expressed by the Langmuir single-site isotherm.

[0203]

[0204] where a is a constant coefficient and K eq is the equilibrium constant.

[0205] Combining equations 1 and 2, the permeability of CO2 can be obtained as follows:

[0206]

[0207] By fitting the pressure-dependent gas permeability with Equation 3, the equilibrium constant ( Figure 5 b) can be calculated. The obtained K eq is 5.1×10 -4 Pa -1 .

[0208] In addition, the Henry's coefficient H pore can be expressed as

[0209]

[0210] Therefore, the permeability of CO2 can be obtained from Equation 5

[0211] P = ak trans H pore (5)

[0212] The permeability is proportional to the Henry's coefficient, and the Henry's coefficient increases in the low-pressure region. This results in an increase in the permeability of CO2 in the low-pressure region.

[0213] The adsorption of CO2 on pyridine N is simulated by the Langmuir single-site isotherm ( Figure 6 a). The functional relationship between the coverage of CO2 and pressure is as follows:

[0214]

[0215] where θ pore is the coverage of CO2.

[0216] The coverage of CO2 on pyridine N is obtained from the XPS data measured at different CO2 partial pressures. K eq (4.4×10 - 4 Pa -1 ) is obtained by fitting the experimental results with the Langmuir single-site isotherm.

[0217] Subsequently, the gas permeation experimental data are in good agreement with the model. The obtained K eq is 5.1×10 -4 Pa -1 , which is consistent with the above theoretical K eq value. This indicates that the transport of CO2 in the nanopores substituted with pyridine N follows the pressure-dependent gas translocation model. Due to the increase in the permeability of low-concentration CO2, extremely high performance can be obtained. The CO2 permeability of one of the membranes reaches 50000 GPU, and the CO2 / N2 separation factor reaches 30. Other membranes with lower oxidation degrees enable the CO2 / N2 separation factor to reach 360, and the CO2 permeability is 2010 GPU( Figure 6 c).

[0218] The method of the present invention includes graphene oxidation and NH3 treatment. This method is scalable mainly because it involves treating graphene with gases, treating graphene with O3 to increase porosity, and treating graphene with NH3 to increase pyridine groups. Centimeter-scale membranes were prepared using a mild oxidation method. When the CO2 partial pressure in the feed gas is close to 1%, these membranes also show excellent performance (CO2 permeability of 6570 GPU, CO2 / N2 selectivity exceeding 2000, Figure 6 d).

[0219] High CO2 permeability at low CO2 concentrations is very attractive for CO2 capture from point source emissions such as building exhaust gases and coal power plants with CO2 concentrations between 0.2% and 7%. This is also attractive for CO2 capture from aluminum production and natural gas combined cycles, as the CO2 concentrations in the flue gas are 1 - 2% and 3 - 4% respectively (Wang et al., 2020, Frontiers in Energy Research, Vol. 15, pp. 1 - 24).

[0220] Techno-economic analysis of membranes using extremely diluted feeds (0.5% and 1%) shows capture penalties of $70 / ton and $45 / ton respectively.

[0221] In summary, these data indicate that the N-functionalized nanoporous single-layer graphene membranes of the present invention and their production methods contribute to the efficient carbon capture of the membranes. Through reaction with O functional groups, N-functionalization preferentially occurs near the pore edges and at the pore edges. The presence and content of N functional groups (>10 12 cm -2 ) were confirmed by XPS, EDS, and Raman spectroscopy, and the results were in good agreement. The filling and empty pore structures driven by reversible CO2 adsorption-desorption on pyridine N-substituted nanopores were observed using LTSTM. The CO2 adsorbed at the pore edges and the N functional groups located at the pore edges prevent the penetration of N2.

[0222] The post-combustion capture performance of these membranes was compared with state-of-the-art membranes, including non-functionalized nanoporous graphene, graphene modified with amine-rich polymers or ionic liquids, polymers, microporous polymers, and facilitated transport membranes ( Figure 5 e). Table 2 below lists the comparison of CO2 / N2 mixture separation performance of the N-functionalized graphene membranes of the present invention with other pore-engineered membranes at a CO2 ratio of 20% in the feed.

[0223] Table 2

[0224]

[0225]

[0226]

[0227] * indicates the ideal selectivity.

[0228] 20: Merkel et al., 2010, *J. Memb. Sci.*, Vol. 359, No. 1, pp. 126 - 139; 21: Wang et al., 2017, *Angew. Chemie Int. Ed.*, Vol. 56, No. 45, pp. 14246 - 14251; 22: Karunakaran, 2017, *J. Mater. Chem. A*, Vol. 5, No. 2, pp. 649 - 656; 23: Kim et al., 2013, *Science* (80 -.), Vol. 342, No. 6154, pp. 91 - 95; 24: Zhou et al., 2017, ibid.; 25: Fu et al., 2018, *Nat. Commun.*, Vol. 9, No. 1, p. 990; 26: Kim et al., 2013, *J. Memb. Sci.*, 428, pp. 218 - 224; 27: Pang et al., 2020, *J. Memb. Sci.*, 612, 118443; 28: Zhang et al., 2021, ibid.; 29: Guo et al., 2020, *Nano Lett.*, 20, pp. 7995 - 8000; 30: Marius et al., 2022, ibid.; 31: Yang et al., 2020, *Chem*, Vol. 6, No. 3, pp. 631 - 645; 32: Sutrisna et al., 2018, *J. Mater. Chem. A*, Vol. 6, No. 3, pp. 918 - 931; 33: Sutrisna et al., 2017, *J. Memb. Sci.*, 524, pp. 266 - 279; 34: Kim et al., 2018, *J. Mater. Chem. A*, Vol. 6, No. 17, pp. 7668 - 7674; 35: Dai et al., 2012, *J. Memb. Sci.*, 401 - 402, pp. 76 - 82; 36: Nikolaeva et al., 2017, *J. Mater. Chem. A*, Vol. 5, No. 37, pp. 19808 - 19818; 37: Qiao et al., 2019, ibid.; 38: Fu et al., 2016, *Energy Environ. Sci.*, Vol. 9, No. 2, pp. 434 - 440; 39: Xie et al., 2018, *Energy Environ. Sci.*, Vol. 11, No. 3, pp. 544 - 550; 40: Li et al., 2013, *J. Memb. Sci.*, 436, pp. 121 - 131; 41: Scofield et al., 2016, *J. Memb. Sci.*, 499, pp. 191 - 200; 42: Hsu et al., 2021, supra.

[0229] The membranes of the present invention not only have attractive post-combustion capture performance but also perform excellently in dilute CO2 sources. The pyridine-N-substituted pores will lay the foundation for a variety of applications for separating molecules based on chemical and charge affinity, including ion-ion separation (DuChanois et al., 2022, Sci. Adv. Chem. Soc., Vol. 143, No. 13, pp. 5080-5090), desalination (Zhao et al., 2021, Nat. Mater, Vol. 20, No. 11, pp. 1551-1558), and nanofiltration (Ling et al., 2022, Sci. Adv., Vol. 3, No. 4, e1601939).

[0230] Furthermore, the N-functionalized graphene membranes of the present invention showed similar carbon capture performance after 445 days, indicating that the present invention can easily achieve a typical service life of 5 years.

Claims

1. A method for preparing a gas selective separation filter, comprising the following steps: a) Provide a supported nanoporous single-layer graphene membrane, the membrane comprising nanopores with a pore density of about 10 10 cm -2 to about 10 13 cm -2 , and the nanopores are modified with O functional groups, the O / C ratio is about 0.01 to 0.5, and the single-layer graphene membrane is supported on a mechanical support; b) N-functionalize the graphene nanopores of the supported nanoporous single-layer graphene membrane; c) Optionally, transfer the resulting supported N-functionalized graphene membrane from the mechanical support to a porous mechanical support.

2. The method according to claim 1, characterized in that, The supported nanoporous single-layer graphene is N-functionalized by reacting the O-functional groups (such as epoxy groups) of the O-functionalized nanoporous single-layer graphene membrane with a nitrogen source.

3. The method according to claim 1 or 2, characterized in that, The N-functionalization reaction is carried out under saturated ammonia vapor.

4. The method according to any one of the preceding claims, characterized in that, The N-functionalization reaction is carried out at a temperature of about 5 °C to about 100 °C, usually at a temperature of 20 °C to about 80 °C.

5. The method according to any one of the preceding claims, characterized in that, The N-functionalization reaction is carried out for about 0.01 to 72 hours, for example 0.5 to about 24 hours.

6. The method according to any one of the preceding claims, characterized in that, The N-functionalization reaction is carried out at about 10 mbar to about 50 bar, usually at a pressure of about 1 bar to 10 bar.

7. The method according to any one of the preceding claims, characterized in that, Including nanopores with a pore density of about 10 10 cm -2 to about 10 13 cm -2 The supported nanoporous single-layer graphene film having nanopores and functionalized with O functional groups with a molar O / C ratio of about 0.01 to about 0.5 is obtained by ozone treatment of the nanoporous single-layer graphene film, for example, by treating with O3 for about 0.5 to about 2 hours.

8. The method according to claim 8, characterized in that, The O3 treatment includes placing the nanoporous single-layer graphene membrane in an O3 / O2 mixed gas stream.

9. A gas selective filter, comprising a graphene membrane having N-functionalized nanopores, wherein the nanopores are obtained by the method according to any one of the preceding claims.

10. A gas selective filter, comprising an N-functionalized nanoporous single-layer graphene membrane, wherein the nanopores are N-functionalized and 16 to 20 pyridine Ns are introduced at the edge of each pore.

11. The gas selective filter according to any one of claims 9 or 10, comprising an N-functionalized nanoporous single-layer graphene membrane, the pore size distribution of which makes the van der Waals gap of the graphene pores less than 1 nm.

12. The gas selective filter according to any one of claims 10 to 11, comprising an N-functionalized nanoporous single-layer graphene membrane, wherein the density of the N-functionalized pores is about 10 10 cm -2 to 10 13 cm -2 .

13. The gas selective filter according to any one of claims 9 to 12, wherein the CO2 permeability of the graphene membrane is from about 200 GPU (6.7x10 -8 mol m -2 s -1 Pa -1 ) to about 50,000 GPU (1.675x10 -5 mol m -2 s-1 Pa -1 ), for example from 2'500 to about 20'000 GPUs.

14. The gas selective filter according to any one of claims 9 to 13, wherein the CO2 / N2 selectivity of the graphene membrane is about 20 to about 2'000 (e.g., 100), depending on the CO2 concentration in the feed and the pore size of the graphene.

15. Use of the gas selective separation filter according to any one of claims 9 to 14 for separating gases, in particular for separating N2 from CO2.

Images