Realization Of Ultra-selective Molecular Sieve Gas Separation Membranes Through Multi-covalent Cross-linking Of Microporous Polymer Mixtures

Thank you for visiting Nature. The browser version you are using has limited support for CSS. For the best experience, we recommend that you use a newer version of the browser (or turn off the compatibility mode in Internet Explorer). At the same time, to ensure continued support, we will display sites without styles and JavaScript.

Nature Communications Volume 12, Article Number: 6140 (2021) Cite this article

High-performance membranes that exceed the upper limit of traditional permeation selectivity are attractive for advanced gas separation. In this context, microporous polymers are receiving more and more attention due to their excellent permeability, however, they exhibit moderate selectivity that is not conducive to separating gas mixtures of similar sizes. Here, we report a method for designing polymer molecular sieve membranes, which is designed by multi-covalent cross-linking of bromomethyl polymer with inherent microporosity and Tröger base, while achieving high permeability Sexuality and selectivity. By adjusting the reaction temperature, reaction time and oxygen concentration, polymer chain scission, rearrangement and thermal oxidation cross-linking reactions occur, and ultra-selective gas separation is achieved. After heat treatment at 300 °C for 5 hours, the membrane exhibited O2/N2, CO2/CH4 and H2/CH4 selectivities as high as 11.1, 154.5 and 813.6, respectively, exceeding the most advanced upper limit. This design strategy represents a generalizable method to create molecular sieve polymer membranes with great potential for high-performance separation processes.

Energy related to the separation and purification of industrial gases, fine chemicals and water currently accounts for 10-15% of the country's total energy consumption1. This number is expected to triple by 20502,3. Global energy shortages, climate change, and rapid population growth have stimulated the exploration of energy-saving technologies such as gas separation, water purification, and energy production4,5,6. Due to the advantages of low energy consumption, small footprint and easy operation7,8,9, membrane-based separation technology is expected to meet the needs of energy-saving separation processes. Currently, membrane-based gas separation mainly relies on synthetic polymer materials with customizable gas transmission characteristics. Ultra-permeable and selective polymer membranes have become the key factors for achieving the most ideal gas product recovery and purity in practical industrial applications.

In order to improve the gas separation performance of polymer membranes, various material synthesis and design strategies have been proposed and extensively studied, including grafting large groups to adjust the stiffness of the polymer 10, 11, 12, 13, and constructing by integrating molecular sieve fillers The hybrid membrane is converted to polymers 14, 15, 16, and the polymer structure is modified at the microscopic level by external stimuli (for example, heat, light, and oxygen)17,18,19. For example, rigid polymers with high free volume, such as inherently microporous polymer (PIM) 20, Tröger’s base polymer (TB) and polyimide-containing large groups 21, have proven to be more breathable than commercial polymers. The material film is several orders of magnitude higher, such as the matrix 22. The accumulation of broken polymer chains in these rigid polymers results in the formation of solid microcavities, allowing rapid gas diffusion without significantly reducing selectivity. In addition, polymer matrices combined with inorganic or organic molecular sieves (such as covalent organic frameworks, metal-organic frameworks, and graphene oxide) provide an alternative for obtaining membranes with enhanced gas separation properties23,24. Subsequently, the cross-linking or rearrangement of the aforementioned polymer systems caused by chemical, light or heat treatment provides a further method to tailor the microstructure of the polymer 17, 18, 19. Although all these methods have demonstrated the potential to prepare membranes with good gas separation performance, reducing the pore size distribution and enhancing molecular sieve performance through polymer microstructure engineering are still key challenges for high-performance gas separation membranes.

Here, we have developed a method for preparing multi-covalent cross-linked microporous membranes, which has the advantages of mild membrane processing temperature and excellent molecular sieving performance. Unlike the above-mentioned research on cross-linked polymers through external stimuli, which mainly involve intramolecular cross-linking reactions and it is difficult to fine-tune the polymer microstructure, we have established oxygen-induced chain scission, polymer segment rearrangement, and in-situ/polymer Interchain crosslinking to build a hypercrosslinked network in a microporous polymer system at moderate temperatures.

Our strategy is specifically designed for polymer blends of bromomethylated PIM (PIM-BM) and TB. They were wisely selected as prototypes of microporous polymer systems to provide both intermolecular and intramolecular interactions Link reaction site (Figure 1). The nucleophilic coupling reaction between the reaction site of the CH2Br group of PIM-BM and the tertiary amino group of TB in the PIM-BM/TB mixture plays a key role in creating a pre-crosslinking network before the oxidative crosslinking reaction. This pre-crosslinking reaction is expected to significantly affect the fluidity of the polymer chain. When exposed to ppm-level O2, polymer chain scission occurs above 250 °C, thereby enhancing the molecular sieving performance of the membrane. It is believed that the multiple cross-linking reactions that occur gradually at different temperatures help to form a molecular sieve structure in the cross-linked PIM-BM/TB membrane (denoted as XPIM-BM/TB). The resulting XPIM-BM/TB membrane exhibits ultra-high selectivity for technically important gas pairs (such as H2/CH4, CO2/CH4, and O2/N2). Compared with the previously reported cross-linked PIM-118 and oxidative cross-linked PIM-BM, TB (Supplementary Figure 21) or PIM-117, our membrane exhibits the highest CO2/CH4 selectivity, up to 154.5, CO2 gas The permeability is 68 baler. This fact indicates that the intramolecular crosslinking reaction that occurs in pure TB or PIM-1 alone will not produce the polymer membrane structure required to achieve ultra-high gas selectivity. In addition, the H2/CH4 selectivity of this membrane is as high as 813.6, and the H2 permeability is 358 Barrer. Compared with the intramolecular cross-linked PIM-BM, the performance of XPIM-BM/TB membrane far exceeds the current permeation selectivity of many gas pairs (such as CO2/CH4, H2/CH4, H2/N2, O2/N2) Upper limit. We attribute such high separation performance to the integrated multi-covalent cross-linking reaction (including self-cross-linking in PIM-BM and intermolecular/intramolecular in polymer blends (PIM-BM and TB) Cross-linking) fine-tune the pore size distribution in the cross-linked membrane. In any case, these membranes exhibit attractive gas separation performance and can achieve the super-selective separation of industrially relevant gas pairs discussed in this work.

a The chemical structure of PIM-BM and TB. b The proposed cross-linking mechanism between PIM-BM and TB.

The N atom in the TB part can interact with the -CN group in the PIM, which significantly improves the miscibility of the PIM and TB polymer. Therefore, this work uses N-bromosuccinimide bromomethylating agent (represented by PIM-BM-x, where x is the degree of bromomethylation, x=70, Supplementary Figure 1-3)25. Then PIM-BM-70 and rigid TB polymer (Supplementary Figures 4 and 5) with excellent mechanical properties (PIM-BM/TB) (Supplementary Figure 19a and Supplementary Table 3) were selected as the framework to develop the cross-linked microporous membrane.

The physically mixed PIM-BM-70 and TB polymer exhibit excellent miscibility, and can be easily made into a transparent film by dissolving in chloroform and casting on a glass plate. The resulting PIM-BM/TB film is cross-linked by heat treatment in a temperature window of 120-300°C in nitrogen containing ppm-level oxygen for different time periods. The cross-sectional morphology of the film was characterized by scanning electron microscopy (SEM). As shown in the SEM images in Figures 2a-f, all films showed a smooth and non-macroporous surface regardless of the thermal crosslinking temperature. As the heat treatment temperature increased from 120°C to 300°C, the color of the film changed from the original yellow to brownish yellow, and then to black, as shown in Figure 2g. The PIM-BM/TB treated at 200°C for 20 hours is partially insoluble in common organic solvents such as chloroform (Supplementary Figure 10), which easily dissolves the original PIM-BM/TB. After being treated at a high temperature of 250 to 300 °C, the PIM-BM/TB membrane becomes completely insoluble in chloroform, as shown in Figure 2g.

a, c, e Cross-sectional images of PIM-BM/TB, PIM-BM/TB-250 °C-10 h and PIM-BM/TB-300 °C-5 h. b, d, f Surface images of PIM-BM/TB, PIM-BM/TB-250 °C-10 h and PIM-BM/TB-300 °C-5 h. g PIM-BM/TB photos processed at different temperatures.

We have proposed three possible chemical cross-linking mechanisms that may occur during heating, that is, tertiary amines react with bromomethyl to form quaternary ammonium salts, alkylation reactions, and oxidative cross-linking reactions in the PIM-BM/TB film, as shown in the figure Shown in Figure 1 and the supplementary figure. 7-9.

In the first case, we found that the reaction sites between the CH2Br group of PIM-BM and the tertiary amino group of TB play a key role in the PIM-BM/TB mixture. The -CH2Br group in the repeating unit of PIM-BM reacts with the tertiary amino group in the TB polymer. X-ray photoelectron spectroscopy (XPS) results (Figure 3a-e) confirmed that the covalent C-Br bond in the fresh PIM-BM/TB membrane was gradually converted to Br-containing through the nucleophilic coupling reaction of tertiary amine and bromomethyl Of salt. Initial heat treatment. The reaction takes place at a temperature of 120–300 °C. It was found that the degree of nucleophilic coupling reaction between C-Br bond and tertiary amine quantified by XPS results increased from 12% to 40% as the reaction temperature increased from 120°C to 300°C (Supplementary Table 2). Therefore, the main reaction mechanism proposes that part of the tertiary amino group in the TB polymer and the bromomethyl group in the PIM-BM (bromomethylated PIM) are converted into a quaternary ammonium salt [N (R)3]CH2RBr-( Supplementary Figure 7). Based on this reaction mechanism, based on the XPS results, the crosslinking degrees of XPIM-BM/TB-250 °C-10 h and XPIM-BM/TB-300 °C-5 h membranes are estimated to be 25% and 25%, respectively. 40%.

XPS spectrum of Br 3d of fresh PIM-BM/TB. b XPS spectrum of Br 3d at PIM-BM/TB-120 °C-20 h. c XPS spectrum of PIM-BM/TB-200 °C-20 h Br 3d. d XPS spectrum of PIM-BM/TB-250 °C-10 h Br 3d. e XPS spectrum of XPIM-BM/TB-300 °C-5 h Br 3d. f FTIR spectrum of fresh and hot cross-linked PIM-BM/TB.

Regarding the second reaction mechanism, the membrane is subjected to an inert gas containing ppm-level oxygen in the temperature range of 250-300°C. In this case, we propose a possible thermal cross-linking of C-Br bond and benzene ring through alkylation reaction pathway. In addition to the above coupling reaction (Supplementary Figures 8 and 9), HBr is also produced as a gaseous state. by-product. As evidenced by the corresponding ionic current of HBr from TG/mass spectrometry (Supplementary Figure 11a), for the original PIM-BM/TB, HBr was released from 230 to 300°C, and the crosslinking mechanism was confirmed by the alkylation reaction. In contrast, for the cross-linked XPIM-BM/TB-250 °C-10 h, the amount of HBr released during the alkylation reaction is significantly reduced because most of the CH2Br groups are involved in the cross-linking with the benzene ring reaction. TG-MS did not detect the HBr signal of XPIM-BM/TB-300 °C for 5 hours (it is difficult to detect trace amounts of HBr), indicating that as the reaction proceeds in the polymer blend, most of the C-Br group The group was consumed. XPS results also confirmed the consumption of C-Br group, where the ratio of the Br 3d core signal of C-Br (70.3 eV) to Br− (68.3 eV) in XPS spectra is for the film heat-treated at 300°C Significantly reduced by 5 hours (Figure 3e). Under the alkylation reaction, the XPIM-BM/TB-250 °C-10 h and XPIM-BM/TB-300 °C-5 h membrane cross-linking degrees are estimated to be 22% and 45%, respectively, according to the XPS results. In addition, as shown by the Fourier Transform Infrared Spectroscopy (FTIR) of the film (Figure 3f), the characteristic C-Br peak of the XPIM-BM/TB film treated at a temperature of ≥ 250 °C changes at around 660 cm-1. It is not so strongly compared with the original PIM-BM/TB, which further supports the consumption of C-Br groups during the cross-linking reaction.

In the third stage, we proposed an oxidative cross-linking mechanism, that is, PIM-BM/TB film undergoes thermal oxidation at a temperature of 250-300°C in the presence of ppm oxygen. For pure PIM-117, this phenomenon was also observed in the temperature range of 300–450 °C. Oxygen plays a key role in the partial breakdown of polymer chains into polymer fragments. The cleavage of the spiro bond of PIM-BM and the methylene group connected to the N atom of TB may be the main decomposition step. Part of the main chain is oxidized, resulting in the formation of COOH groups. After that, the broken polymer chains are assumed to be thermally rearranged into an energy-favorable state. At the same time, chemical reactions occur between reactive groups, including oxidation-induced free radicals, CH2Br, and nitrogen-containing groups in the polymer chain, leading to extensive covalent crosslinking (Figure 1). The degree of oxidized polymer chain shear can be adjusted by changing the oxygen concentration in the purge gas, which will be discussed later.

Please note that the IR spectrum of the cross-linked PIM-BM/TB film at 300°C shows that the functional organic groups of the sample have obvious peaks, indicating the polymer properties of the film (Figure 3f). In order to further clarify the state of the membrane, elemental analysis was performed using a treatment temperature of up to 550°C, which is widely used for pyrolysis of membranes (Supplementary Table 15) 33. The carbon content and C/O ratio of the membrane treated at 300 °C are very close The value of the untreated film is significantly lower than the value of the carbonized sample at 550 °C. The polymer state of PIM-1 membranes crosslinked at temperatures up to 385°C is also reported in the literature. Based on these evidences, it is believed that the crosslinked film maintains a polymer state rather than carbonization.

As mentioned above, despite the different trends observed in the characterization of PIM-BM/TB films at different temperatures, the three possible cross-linking mechanisms are essentially coupled, all of which contribute to the formation of a cross-linked network within the film . The cross-linking reaction mechanism in the decoupled PIM-BM/TB membrane is worthy of further study to clarify the influence of each individual reaction on the membrane structure and performance.

A wide range of technologies further characterize the physical and chemical properties of the membrane. As shown in the TGA curve, the dependence of the weight loss on the crosslinking temperature indicates that the crosslinked film has higher thermal stability than the uncrosslinked PIM-BM/TB blend (Figure 4a and Supplementary Figure 11a) ). In fact, XPIM-BM/TB-300 °C-5 h membrane begins to decompose at ~400 °C, which is significantly higher than the decomposition temperature of untreated PIM-BM/TB membrane or treatment at 200 °C of ~250 °C. X-ray diffraction (XRD) spectra of PIM-BM/TB before and after thermal crosslinking showed that all polymers were amorphous (Figure 4b). With reference to the XRD spectrum of the original film, the peak at an angle of 11.9° (d-space value is 7.43 Å) is attributed to the loosely packed polymer chains 26,27. As the degree of cross-linking increases, the broad peak at 11.9° shifts slightly to a higher angle (ie smaller d-space value), which means that the cross-linking reaction tightens the chain spacing and may enhance the molecular sieving performance of the membrane . The molecular modeling comparison of PIM-BM/TB and XPIM-BM/TB shows that as the degree of crosslinking increases, the free volume fraction decreases (Supplementary Table 1), which is consistent with the PALS results.

a Thermal stability of PIM-BM/TB and cross-linked PIM-BM/TB. b XRD spectrum. c Mechanical properties of polymer film.

A typical film stress-strain graph (Figure 4c) depicts the decrease in elongation at break and tensile strength as the degree of crosslinking increases, indicating that the ductility of XPIM-BM/TB film is higher than that of untreated PIM-BM/ TB decreases. For example, the ultimate yield stress of the original film at a strain of 8.8% is 43 MPa, while the film treated at 200 °C for 20 hours has a tensile stress of 19 MPa at a strain at break of 1.9%. When the temperature was further increased to 300°C, although the elongation at break decreased, the film still maintained a high mechanical stress of 22 MPa (Supplementary Table 3). The mechanical strength test proved the potential of using our method to make strong membranes.

As demonstrated by the above XRD test, the formation of covalent cross-links tends to tighten the polymer chain, resulting in a decrease in the d-spacing in the polymer matrix. In order to further understand the microstructure, a combination of simulation and experiment methods based on the MELT program and PALS results was used to characterize the pore size distribution of the membrane. As shown in the molecular simulation in Figure 5a, the rigid polymer chains in PIM-BM/TB are stacked disorderly, resulting in the formation of irregularly shaped free volumes. The molecular modeling comparison between the original PIM-BM/TB and the cross-linked PIM-BM/TB shows that effective stacking occurs after cross-linking (Figure 5a, b). The compaction effect caused by this crosslinking is consistent with the fractional free volume (FFV) simulation (Supplementary Table 1). For example, using H2 as a structural probe, for uncrosslinked PIM-BM/TB and crosslinked XPIM-BM/TB-300 °C-5 membranes, the calculated free volume fraction was reduced from 0.228 to 0.187, respectively.

a Representative chain conformation in cross-linked PIM-BM/TB from computer modeling results. b 3-D view of cross-linked PIM-BM/TB modeling structure in amorphous battery (300 °C-5 h) (battery size: 30 × 30 × 30 A; density: ~1.223 g/cm3; Gray-Van der Waals surface; dark gray-Connolly surface with a hole radius of 1.45 A). c The pore size distribution of PALS.

Figure 5c shows the free volume distribution generated from the MELT program based on the PALS results. The mean free volume radius shifts to a lower value during crosslinking. Therefore, as described in Table S1, the free volume fraction of the cross-linked film is less than that of the untreated sample. In addition, heat treatment at 250 and 300 °C results in a narrower pore size distribution of the untreated membrane, which indicates that the chain movement in the cross-linked form of the membrane is more restricted and the pores are smaller28. In fact, cross-linking not only tightens the inner pores of the membrane, but also adjusts the width of the ultra-micropores connecting adjacent cavities, allowing smaller gas molecules (such as H2 and CO2) to selectively diffuse while eliminating larger ones. The gas molecules such as N2 and CH4, which will be discussed in the following section.

In order to explore the gas transmission characteristics in the membrane, first use gas molecules to perform a single gas permeation on the original polymer membrane and the thermally crosslinked membrane, including H2 (2.89 Å), CO2 (3.3 Å), O2 (3.46 Å), N2 (3.64) Å) and CH4 (3.8 Å), the temperature is 35 °C, and the feed pressure is 50 psia. The gas permeability and ideal selectivity of the membrane are shown in Table 1. The original PIM-BM/TB membrane exhibits high permeability and moderate gas selectivity, which is consistent with the reported PIMs series membranes. As expected, due to the shrinkage of the pore structure as described above, the cross-linking reaction caused a sharp drop in the gas permeability of XPIM-BM/TB. In particular, the gas permeability of large gas molecules (CH4, N2) decreases more significantly than that of small gas molecules (such as O2, H2, CO2), as shown in Figure 6a. However, the gas selectivity increases significantly during thermal crosslinking. In addition, regardless of the crosslinking temperature, the order of gas permeability in XPIM-BM/TB is H2>CO2>O2>N2>CH4. This trend is consistent with the order of aerodynamic diameters, indicating that these membranes have molecular sieving properties. In fact, cross-linked membranes with effective stacked chains show significantly enhanced molecular sieving capabilities. For example, the H2/CH4 selectivity of a representative XPIM-TM/TB membrane prepared after heating at 250 °C for 10 hours increased from 17.2 to 118.6, and the CO2/CH4 selectivity increased from 17.9 to 54.7.

a Gas permeability as a function of dynamic diameter. b Use Robeson upper limit CO2/CH4 separation. c H2/CH4 separation and upper limit.

Plot the CO2/CH4 and H2/CH4 separation data with the current upper limit and compare with the literature. Figures 6b and c show very high gas selectivity compared with other reported PIM-based membranes. Compared with the polymer gas separation membranes used in industry (such as polysulfone, PSF), XPIM-BM/TB membranes show significantly higher selectivity, while maintaining a higher order of magnitude of permeability, far exceeding the current The upper limit is 29,30,31. It is worth noting that the H2/CH4 selectivity of XPIM-BM/TB membrane treated at 300 °C for 5 hours is as high as 813.6, which is 47 times higher than that of uncrosslinked PIM-BM/TB, which is the highest value. One reported comparable polymer membranes. Due to the rigid structure of tightly packed chains in these membranes, significantly high H2/CH4 permeation selectivity is achieved. The reduction of H2 after heat treatment is smaller than that of CH4, indicating that XPIM-BM/TB has formed an ideal narrow pore structure, and H2 preferentially penetrates CH4; thereby obtaining ultra-high H2/CH4 selectivity.

Other notable gas pairs, including O2/N2 and H2/N2 are described in Supplementary Figure 18. Similar to H2/CH4, the PIM-BM/TB membrane cross-linked at 300 °C-5 h shows quite high O2/N2 and H2/N2 selectivity, which greatly exceeds the gas separation limit of traditional polymer membranes. Compared with H2/CH4 and CO2/CH4, the separation process of O2/N2 is much more difficult, because O2 and N2 have similar kinetic diameters with only Angstrom differences. Nevertheless, cross-linked membranes show great promise in O2/N2 separation because the separation performance is much higher than the Robeson upper limit. As the reaction temperature increased, the O2/N2 selectivity increased significantly from 3.8 to 11.1, and the O2 permeability decreased from 423 Barrer to 18 Barrer. In this work, an impressive O2/N2 selectivity of over 11 was achieved, which is comparable to the most selective O2/N2 separation membrane material summarized by Robeson.

In any case, the super-selective XPIM-BM/TB membrane developed in this work is the best reported in the literature for separating gas pairs (including CO2/CH4, H2/CH4, H2/N2 and O2/N2) One of the membranes (Figure 6 and Supplementary Figures 17 and 18).

One of the important advantages of the PIM-BM/TB polymer membrane is its versatility in terms of customized microporous structure and gas separation performance. The gas transmission characteristics closely related to the membrane pore structure can be precisely adjusted by controlling the cross-linking reaction temperature, reaction time and atmosphere. Figure 7 shows the effects of reaction temperature, reaction time, and oxygen concentration in the purge gas on the gas permeability and selectivity of the cross-linked PIM-BM/TB. Obviously, as the crosslinking temperature increases from 80°C to 300°C, the permeability gradually decreases, while the selectivity increases significantly (Figure 7a, b). In fact, the cross-linked membrane shows a greater decrease in gas permeability of larger molecules (such as CH4) when cross-linked, while the decrease in smaller molecules (such as H2) is smaller (Table 1). Correspondingly, the H2/CH4 gas pair with the largest kinetic diameter difference is the most sensitive to the reaction temperature. The gas pairs explored in this study (ie H2/CH4, H2/N2, CO2/CH4, CO2/N2 and O2/N2 ).

a gas permeability and b gas selectivity as a function of reaction temperature. All samples were heat treated at 200 ppm O2 for 10-20 hours at the set point temperature, except for the samples which were annealed at 300°C for 5 hours. c Gas permeability and d gas selectivity as a function of the reaction time of the cross-linked membrane treated at 250 °C. For membranes treated at 250 °C for 10 hours, e gas permeability and f gas selectivity as a function of oxygen concentration.

In addition to the reaction temperature, the gas separation performance can also be easily adjusted by changing the crosslinking time. For membranes reacted at the same reaction temperature of 250°C, increasing the crosslinking time from 5 hours to 20 hours will result in an increase in gas selectivity and a decrease in gas permeability. For example, the CO2 permeability decreased from 431 Barres to 149 Barres, and the CO2/CH4 selectivity increased from 26.9 to 79.9 (Supplementary Table 6). We further studied the influence of oxygen concentration on the gas separation performance of the crosslinked membrane. The temperature was 250°C and the reaction duration was 10 hours, as shown in Figure 7e, f. For example, when the oxygen concentration is increased from inert gas to 21,000 ppm, the CO2 gas permeability will decrease. In particular, the gas selectivity of CO2/CH4 reaches a maximum of 54.7 at 200 ppm oxygen. The results in Figure 7e and f clearly prove that the oxygen concentration plays a key role in the thermal oxidative cross-linking reaction. Based on the above discussion, adjusting the crosslinking temperature, time, and oxygen concentration can easily produce polymer membranes with attractive gas separation properties. In order to further adjust the membrane performance, the mixing ratio of PIM-BM/TB and the degree of bromomethylation are other potential factors to improve the gas transmission performance, which is currently being explored in our laboratory.

In order to explore the practical applicability of the membrane under aggressive feed conditions, the membrane was subjected to high CO2 feed pressures. In the case of uncrosslinked PIM-BM/TB membranes, due to non-idealities under high pressure, exposure of the membrane to high pressure pure CO2 leads to a decrease in gas permeability (Figure 8a), which is common in highly porous PIM. By increasing the pressure to 500 psia, the selectivity of CO2/CH4 will decrease. On the other hand, compared with the original membrane, the permeability drop of the cross-linked membrane is much smaller, indicating that the structure in the membrane is quite stable. In addition, PIM-BM/TB-300 °C-5 h was tested with an equimolar CO2/CH4 gas mixture (Figure 8c, d). The mixed gas permeation results show that the heat-treated PIM-BM/TB membrane has attractive mixed CO2/CH4 separation characteristics. The CO2 permeability tested at 100 psi is 165 Barrer, and the CO2/CH4 selectivity is 110. By further increasing the feed pressure to 500 psi, both CO2 permeability and CO2/CH4 tend to decrease; however, the crosslinked membrane maintains a CO2 permeability higher than 93 Barrer and a CO2 permeability higher than 66 at a maximum pressure of 500 psi. CO2/CH4 selectivity, showing the performance required for operation under aggressive conditions.

a The relationship between the permeability of pure CO2 and the feed pressure. b The relationship between CO2/CH4 ideal selectivity and feed pressure. c The relationship between CO2 permeability and feed pressure for separating equimolar CO2/CH4 gas mixture. d CO2/CH4 selectivity and separation of feed pressure for equimolar CO2/CH4 gas mixture. e Relationship between pure gas permeability of XPIM-BM/TB-250 °C-10 h and aging time. f The relationship between the gas selectivity of XPIM-BM/TB-250 °C-10 h and the aging time.

In order to evaluate the long-term stability of the membrane, the physical aging of XPIM-BM/TB treated at 250°C for 10 hours within 360 days was studied. The gas permeability of XPIM-BM/TB gradually decreases during the aging process, while the gas selectivity increases. As shown in the Robertson diagram (Figure 6b, c), the aging XPIM-BM/TB membrane still exhibits excellent gas separation performance, which is much higher than the upper limit of the Robertson CO2/CH4 and H2/CH4 in 2008.

In this work, we describe a method for designing microporous polymer blend membranes through multi-covalent cross-linking of PIM-BM/TB. We have proposed three possible cross-linking mechanisms in PIM-BM/TB membranes, which depend on the cross-linking temperature: (a) tertiary amine reacts with bromomethyl to form quaternary ammonium salt; (b) alkyl group at high temperature Chemical reaction; (c) Thermal oxidative crosslinking reaction after polymer chain scission and rearrangement. Complicated internal and mutual cross-linking reactions occur simultaneously between PIM-BM and TB.

The cross-linked PIM-BM/TB molecular sieve membrane with customizable porosity exhibits the required gas separation performance for industrially important gas pairs. The membrane showed an ultra-high H2/CH4 selectivity of 813.6 in the cross-linked XPIM-BM/TB at 300°C for 5 hours, while maintaining an H2 permeability of 358 Barrer. More importantly, for a variety of gas pairs, including H2/CH4, CO2/CH4, H2/N2, and O2/N2, the cross-linked membrane significantly exceeds the upper limit of current conventional polymer membranes. In the future, by controlling the degree of bromomethylation and optimizing the mixing ratio of PIM-BM and TB in the polymer synthesis process, the physical and gas separation performance of the microporous polymer membrane developed in this work can be further customized.

Overall, the strategies in this work provide a way to design and manufacture promising molecular sieve membranes for high-performance separations. This progressive cross-linking method in which multiple cross-linking reactions occur in different temperature ranges has great potential for manufacturing super-selective membranes for H2 recovery and CO2 capture. This concept is expected to be applicable to other commercially available polymers currently being developed in our laboratory, such as polysulfone, polyimide, etc.

A dense membrane is prepared by casting a filtered solution of equimolar PIM-BM and Tröger's Base (TB) in chloroform on a clean glass substrate. After slowly evaporating the solvent within 2 days, a dry free-standing membrane was obtained, exposed to methanol for overnight soaking, and further dried in a vacuum oven at 70°C for 24 hours. The thickness of the PIM-BM/TB film is approximately 50 μm (±10 μm).

The thermal analysis of PIM-BM/TB fresh and heat-treated membranes was performed in TGA to study the thermal degradation under nitrogen atmosphere. The polymer film was dynamically heated from room temperature to 100°C at a rate of 5°C/min, and kept under a nitrogen atmosphere for 30 minutes, and then heated at a rate of 5°C/min to 800°C. TG-MS was performed in TGA Q50 V20.10 Build 36 under 200 ppm O 2 equilibrated with nitrogen.

The membrane of PIM-BM/TB was cross-linked in a CenturionNeytechQex vacuum furnace at a balance of 200 ppm O2 and nitrogen. Purge the vacuum furnace for 60 minutes, then raise the temperature between 80 and 300°C at a rate of 3°C/min and maintain it for 5-20 hours. After the thermal crosslinking treatment process, the film was cooled to room temperature at a rate of 3°C/min in a furnace for further study. The membrane is labeled "XPIM-BM/TB-Temperature (h)", for example, XPIM-BM/TB-80°C-20h.

XRD is used to study the change of d-spacing. The results were recorded on a Bruker AXS GADDS device, using copper radiation with a wavelength of 1.54 Å (voltage: 40 kV, current: 30 mA). The d-spacing is calculated according to Bragg's law (d = λ/2 sin θ). XPS is used to monitor the chemical changes of PIM-BM/TB fresh and thermally cross-linked PIM-BM/TB membranes. They were recorded on a HSi spectrometer (Thermo Fisher ESCALAB 250 xi., England) using a monochromatic Al Kɑ X-ray source (1486.6 eV photons) under full vacuum with a constant residence time of 100 ms and a passing energy of 40 eV. The anode voltage and anode current are 15 kV and 10 mA, respectively. All nuclear level spectra are obtained at a photoelectron emission angle of 90° to the sample. In order to compensate for the surface charging effect, all binding energies (BE) are referenced to the C1s hydrocarbon peak of 284.8 eV. The stoichiometry of surface elements is determined by the peak area ratio, and the accuracy is within ±5%.

The film was analyzed by SEM using Hitachi S5500 microscope. The polymer film broke and was coated with a thin layer of gold. The molecular weight of the polymer was measured using gel permeation chromatography (GPC, Shimadzu LC-20A) with Ultrastyragel column and tetrahydrofuran (THF) as the eluent flowing at a flow rate of 1 mL/min. FTIR measurement is performed using the attenuated total reflection mode (FTIR-ATR) and Perkin-Elmer Spectrum 2000 FTIR spectrometer. Each sample was scanned 32 times. The DX-2700 machine used Cu Kα radiation at 30 mA and 40 kV for wide-angle X-ray scattering at 0.03 steps per second. The tensile test of the polymer film was performed on an Instron-1211 (Instron Co., USA) mechanical tester at a crosshead speed of 1 mm/min. The polymer film is cut into slices with an effective length of 10 cm and a width of 1 cm. The exact value is determined by high-resolution photos and calibration of known lengths. The average value of Young's modulus is derived from the initial slope. The tensile strength at break and the elongation at break were also measured. The positron annihilation experiment is carried out by using fast-fast coincidence PALS. The 22Na source is used as the positron source. The activity of the 22Na source is about 10 mCi. Kapton film is used to encapsulate the dried 22Na source. Cut the film into 1 cm × 1 cm slices. The thickness of the test slice is about 1.5 mm. Two pieces of the same sample are sandwiched between a 20 μCi positron source (22Na) and sealed with two pieces of 7 μm Kapton film. The positron lifetime spectrum of single crystal Ni is used as a reference to subtract the source components of positron annihilation in the Kapton film and 22Na. The positron lifetime (τ) is obtained from the time difference between the emission of the born gamma ray (1.28 MeV) and the annihilation photon (0.511 MeV).

The pure gas permeation test is carried out at a temperature of 35 °C, the feed pressure is up to 500 psi, and a constant volume variable pressure device is used. Use the same constant volume variable pressure device to measure the permeation characteristics of the mixed gas in the same membrane cell. Expose the membrane to a certified gas mixture of CO2/CH4 (50/50 vol%) with a feed pressure of up to 500 psi and a temperature of 35 °C. The gas composition was analyzed by gas chromatograph (GC-7820A, Agilent).

Molecular dynamics (MD) simulations are constructed by the Forcite module in the Materials Studio software package (Accelrys Inc., CA, USA). In a cube simulation box, four polymer chains (2 PIM-BM-70% polymer chains and 2 TB polymer chains) with 10 repeating units were constructed. The initial density before crosslinking is 0.5 g/cm3, and the target density is 1.177 g/cm3. The force field is PCFF. The Berendsen algorithm with a decay constant of 0.1 ps is used to control the temperature and pressure of each box. The specific procedures before crosslinking are as follows: (1) Energy minimization; (2) 50 ps NVT-MD simulation at 600 K; (3) 100 ps NPT-MD simulation, 600 K, 1 GPa; (4) 100 ps NPT -MD simulation, 298.15 K, 1 GPa; (5) 100 ps NPT-MD simulation, 298.15 K, 0.1 MPa; (6) 50 ps NVT-MD at 298.15 K. The Ewald summation method is used to calculate non-bonded interactions with an accuracy of 0.001 kcal/mol.

The final equilibrium structure is used to crosslink Br with N or C atoms, and a cube simulation box with crosslinked polymers is used as the initial structure for molecular dynamics simulation. The specific procedures after crosslinking are as follows: (1) 100 ps NPT-MD simulation, 2 GPa 298.15 K; (2) 100 ps NPT-MD simulation, 298.15 K, 0.1 MPa; (3) 50 ps NVT at 298.15 K -MD simulation. The force field and other parameters are the same as those used before crosslinking.

The data supporting the results of this study can be obtained from the corresponding author upon reasonable request.

Low, Z., Budd, PM, McKeown, NB & Patterson, DA Gas permeability, physical aging and its mitigation effect in high free volume glass polymers. Chemistry Rev. 118, 5871–5911 (2018).

Nugent, P. etc. Porous material with the best adsorption thermodynamics and kinetics for CO2 separation. Nature 495, 80–84 (2013).