Imidazolium grafted anion exchange membrane, preparation method and application of imidazolium grafted anion exchange membrane in non-aqueous flow battery
By preparing a three-dimensional multi-level topological structure of alkylimidazolium-grafted crosslinked polybenzimidazole anion exchange membrane, the problems of insufficient resistance to organic solvents and insufficient ionic conductivity of non-aqueous flow battery membranes were solved, achieving efficient ion transport and stable energy storage performance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING TECH UNIV
- Filing Date
- 2026-01-22
- Publication Date
- 2026-04-17
AI Technical Summary
Non-aqueous flow battery separators are insufficient in terms of resistance to organic solvents, ionic conductivity, and permeability of active materials, making it difficult to meet the requirements of high energy density energy storage.
A three-dimensional, multi-level topological alkyl imidazolium-grafted crosslinked polybenzimidazole anion exchange membrane was prepared by a combination of solvent-inducible phase separation and chemical crosslinking. The ionic conductivity and organic solvent resistance of the membrane were improved by imidazolium cation grafting.
It achieves an ionic conductivity of up to 2.14 mS·cm⁻¹ and extremely low active material permeability, with an energy efficiency improvement of about 35% and a coulombic efficiency improvement of about 13%. The membrane maintains structural integrity in organic solvents and exhibits good cycle stability.
Smart Images

Figure CN121873401A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to imidazolium-grafted anion exchange membranes, their preparation methods, and their applications in non-aqueous flow batteries, belonging to the field of flow battery technology. Background Technology
[0002] The large-scale utilization of intermittent energy sources such as solar and wind power urgently requires supporting high-capacity energy storage technologies to achieve supply and demand balance. Flow batteries, especially vanadium redox flow batteries (VFBs), are highly promising grid-scale energy storage technologies. However, the development of vanadium redox flow batteries is limited by high material costs and relatively low energy density (typically 25-30 Wh·L⁻¹). In contrast, non-aqueous flow batteries (NAFBs) have a wider electrochemical stability window (up to 4V, far exceeding the 1.5V of aqueous systems) and a richer selection of materials, giving them significant advantages in the field of high-energy-density energy storage.
[0003] As a core component of flow batteries, the separator has become a research hotspot due to its dual function of inhibiting cross-permeation of active materials and promoting rapid charge carrier transport. However, the development of separators for non-aqueous flow batteries faces numerous challenges. These separators must simultaneously meet stringent requirements such as excellent resistance to organic solvents, high ionic conductivity in the organic solvent phase, and low permeability of active materials. Furthermore, the inherent defects of non-aqueous electrolytes, such as high viscosity and poor ionic conductivity, further exacerbate the difficulty of achieving these requirements. Currently, commercially available polyolefin separators such as Celgard and Fumasep are widely used in non-aqueous systems, but these separators generally suffer from insufficient ion selectivity or easy swelling in organic solvents. Therefore, the preparation of separators that combine excellent resistance to organic solvents, low permeability of active materials, and high ionic conductivity is crucial for advancing the technology of non-aqueous flow batteries. Summary of the Invention
[0004] This invention proposes an alkylimidazolium-grafted crosslinked polybenzimidazole (PBI) anion exchange membrane with a three-dimensional multi-level topology. Figure 1 The cross-linking process endows the membrane with excellent stability against organic solvents, while the subsequent imidazolium cation grafting strategy significantly improves the ionic conductivity of the membrane in organic solutions. Combined in-situ and ex-situ characterization with molecular dynamics (MD) simulations revealed that the grafted imidazolium cations effectively promote anion hopping conduction. Furthermore, the excellent performance of the non-aqueous flow battery validates that the cross-linked structure allows the membrane to maintain structural integrity during long-term operation in organic solutions.
[0005] An anion exchange membrane includes: a porous base membrane composed of polybenzimidazole (PBI); the porous base membrane being chemically cross-linked; and an alkylimidazolium cation covalently grafted onto the polybenzimidazole via a linking group.
[0006] The linking group is xyleneyl.
[0007] The alkyl group in the alkylimidazolium cation is a C1-C8 straight-chain or branched alkyl group.
[0008] The alkyl group is selected from ethyl, butyl, or hexyl.
[0009] The porous base membrane has a hierarchical pore structure comprising micropores and mesopores; the pore size of the micropores is 4-6 Å, preferably 4.5-5.5 Å.
[0010] The anion exchange membrane has an ionic conductivity in N,N-dimethylformamide electrolyte greater than 0.4 mS·cm⁻¹, preferably 0.9-2.5 mS·cm⁻¹.
[0011] The permeability of the anion exchange membrane to 10-methylphenthiazide is less than 5 × 10⁻ 9 cm²·s⁻¹, and / or a permeability to 2,1,3-benzothiadiazole less than 2 × 10⁻ 8 cm²·s⁻¹.
[0012] The anion exchange membrane has an ion exchange capacity of 0.4-1.5 meq·g⁻¹.
[0013] A method for preparing an anion exchange membrane according to any one of the claims includes the following steps:
[0014] a) Preparation of polybenzimidazole porous membranes via phase inversion method;
[0015] b) Crosslink the polybenzimidazole porous membrane obtained in step a) using a chemical crosslinking agent;
[0016] c) The cross-linked membrane is reacted with a linker, and then with an alkylimidazolium cation, thereby grafting alkylimidazolium cations onto the membrane.
[0017] The phase inversion method described in step a) includes: dissolving polybenzimidazole in an organic solvent to form a casting solution, and immersing the casting solution in a non-solvent coagulation bath for phase inversion; preferably, the organic solvent is N,N-dimethylacetamide; more preferably, the mass percentage concentration of polybenzimidazole in the casting solution is 10-20 wt%.
[0018] The non-solvent coagulation bath is a mixture of alcohol solvents; preferably, the mixture of alcohol solvents consists of isopropanol and ethanol; more preferably, the volume ratio of isopropanol to ethanol is 1:4 to 1:24.
[0019] The chemical crosslinking agent in step b) is persulfate; preferably, the chemical crosslinking agent is potassium persulfate; more preferably, the crosslinking treatment is carried out at 10-40°C for 12-36 hours; and the concentration of the persulfate aqueous solution is 3-8 wt%.
[0020] The linker in step c) is α,α′-dihalo-p-xylene; preferably, the linker is α,α′-dibromo-p-xylene.
[0021] In step c), the reaction with the linker is carried out at 60-100°C for 6-48 hours; and / or the reaction with alkylimidazolium is carried out at 40-80°C for 12-36 hours; the concentration of the linker in the solvent is 2-5 wt%.
[0022] The alkylimidazolium mentioned in step c) is selected from at least one of 1-ethylimidazolium, 1-butylimidazolium, and 1-hexylimidazolium; preferably, in the reaction with the alkylimidazolium, the concentration of the alkylimidazolium in the solvent is 0.01-0.2 mol / L.
[0023] The application of the anion exchange membrane in non-aqueous flow batteries.
[0024] A method for promoting anion transport in a non-aqueous electrolyte, comprising:
[0025] An anion exchange membrane is provided, the anion exchange membrane comprising a cross-linked polybenzimidazole porous matrix and a plurality of imidazole on the matrix covalently grafted thereon; the anion exchange membrane is contacted with a non-aqueous electrolyte containing the anions;
[0026] The imidazolium cation adsorbs the anion through coulombic interactions and promotes the hopping migration of the anion between adjacent imidazolium cation sites.
[0027] The imidazolium cation is an alkyl imidazolium cation, and its alkyl side chain disrupts the instantaneous stable coordination between the imidazolium cation and the anion through its own flexible movement, thereby promoting the hopping migration of the anion.
[0028] The alkyl side chain is a C1-C8 alkyl group, preferably butyl.
[0029] The anion is tetrafluoroborate (BF4⁻) anion.
[0030] By means of the method, the migration number of the anion in the membrane is increased to 0.95 or more, preferably 0.99 or more.
[0031] A method for improving the selective transport capability of polybenzimidazole-based anion exchange membranes for anions in non-aqueous electrolytes includes the following steps: covalently grafting multiple alkylimidazolium cations onto a cross-linked polybenzimidazole porous matrix to form an anion adsorption site network composed of the alkylimidazolium cations within the membrane; wherein the site network is used to enrich anions through Coulomb interactions upon contact with the non-aqueous electrolyte and to provide a transport path for the anions dominated by hopping migration.
[0032] The grafting of the alkylimidazolium cation improves the dielectric constant of the anion exchange membrane at 10⁻² Hz.
[0033] The method enables the anion exchange membrane to achieve an ion selectivity of 1×10⁻⁶. 5 The ion selectivity coefficient is defined as the ratio of ionic conductivity to the permeability of the active material, where S·s·cm⁻³ or higher.
[0034] The application of any of the methods described herein in non-aqueous flow batteries is intended to improve the energy efficiency and cycle stability of the batteries.
[0035] The beneficial effects of this invention are as follows: An integrated preparation strategy combining solvent-induced phase separation, chemical crosslinking, and imidazolium grafting successfully constructed an alkylimidazolium-grafted crosslinked polybenzimidazole anion exchange membrane with a three-dimensional multi-level topology. Through precise control of the membrane's pore structure and surface properties, a highly efficient ion transport network was constructed. The resulting membrane exhibits extremely low active material permeability, a high DMF phase ionic conductivity of 2.14 mS·cm⁻¹, and excellent resistance to organic solvents. In DMF-based non-aqueous flow battery systems, the performance of this membrane far surpasses that of the commercially available Celgard 2500 membrane, with an energy efficiency improvement of approximately 35% and a coulombic efficiency improvement of approximately 13%, and it can stably cycle for more than 350 cycles. Attached Figure Description
[0036] Figure 1 : A schematic diagram of the preparation process of this patent.
[0037] Figure 2 Digital photographs of the original PBI membrane and the K-xPBI membrane.
[0038] Figure 3 Digital photos of PBI and K-xPBI membranes after being soaked in DMF for 7 days.
[0039] Figure 4 : 1H time-domain low-field nuclear magnetic resonance spectrum of K-xPBI film with H2O as probe molecule.
[0040] Figure 5Characterization of K-19PBI-Im membranes grafted with different alkyl imidazole side chains. (a) Schematic diagram of ion transport in K-19PBI-Im membranes grafted with different imidazole side chains (including ethylimidazolium (EIm), butylimidazolium (BIm), and hexylimidazolium (HIm)). (b) High-resolution N1s X-ray photoelectron spectroscopy (XPS) of the K-19PBI-Im membrane. (c) 1H time-domain low-field nuclear magnetic resonance (LF-NMR) spectra of the K-19PBI and K-19PBI-Im membranes. (d) Electrolyte absorbance of the K-19PBI and K-19PBI-Im membranes. (e) BF4⁻ anion permeability of the K-19PBI and K-19PBI-Im membranes in DMF electrolyte.
[0041] Figure 6 FTIR spectrum of K-19PBI-Im membrane.
[0042] Figure 7 : SEM images of the surface and cross-section of the K-19PBI-Im diaphragm.
[0043] Figure 8 Digital photograph of K-19PBI-Im diaphragm after 7 days of soaking in DMF.
[0044] Figure 9 Ionic conductivity of K-19PBI membrane and K-19PBI-Im membrane.
[0045] Figure 10 Tensile strength and elastic modulus of K-19PBI-Im diaphragm.
[0046] Figure 11 Swelling degree of K-19PBI diaphragm and K-19PBI-Im diaphragm.
[0047] Figure 12 Energy efficiency (EE) of K-19PBI-Im separator in NAFB at a current density of 5 mA·cm⁻².
[0048] Figure 13Characterization of the K-19PBI-yBIm membrane. (a) Cross-sectional scanning electron microscope (SEM) image of the K-19PBI-0.1BIm membrane, with the inset showing a photograph of the sample. (b) Electrolyte uptake of the K-19PBI-yBIm membrane. (c) BF4⁻ anion permeability of the K-19PBI-yBIm membrane in DMF electrolyte. (d) MPT and (e) BTD permeability of the K-19PBI-yBIm membrane and the commercial Celgard 2500 membrane in DMF electrolyte. (f) Comparison of ionic conductivity to active material permeability of reported NAFB membranes evaluated in similar electrolytes. The dashed line represents membrane selectivity (the ratio of ionic conductivity to active material permeability).
[0049] Figure 14 Battery performance of separators in non-aqueous flow batteries (NAFB). (a) Schematic diagram of NAFB structure using BTD (negative electrolyte) and MPT (positive electrolyte) as redox couple. (b) Coulombic efficiency (CE), (c) voltage efficiency (VE), and (d) energy efficiency (EE) of Celgard2500, K-19PBI, and K-19PBI-yBIm separators at a current density of 5 mA cm⁻². (e) Long-cycle stability of NAFB using K-19PBI-0.1BIm separator. (f) Comparison of energy efficiency (EE) and cycle number of different separators in NAFB using phenothiazine derivatives or 2,1,3-benzothiadiazole as active materials.
[0050] Figure 15 Mechanism of imidazole cations. (a) TEA⁺ and BF4⁻ transference numbers of K-19PBI and K-19PBI-BIm membranes. (b) Broadband dielectric spectrum of K-19PBI-yBIm membrane. (c) In-situ FTIR spectrum of K-19PBI-0.1BIm in TEABF4 / DMF solution. (d) Interaction energy between BF4⁻ anion and grafted membrane obtained by molecular dynamics (MD) simulations (Lennard-Jones and Coulomb). (e) Representative snapshots of MD simulations: BF4⁻ (green), TEA⁺ (red), DMF solvent (translucent), K-PBI (purple), and K-PBI-BIm (yellow). (f) Diffusion coefficient and root mean square displacement (MSD) curves of the membranes.
[0051] Figure 16 Broadband dielectric spectrum of K-19PBI-yBIm film: Effect of BIm content on (a) dielectric constant at 10⁻²Hz and (b) rate of decrease at mid-to-high frequencies.
[0052] Figure 17Local magnified FTIR spectrum of K-19PBI-0.1BIm membrane before testing (1000–2000 cm⁻¹ region)
[0053] Figure 18 Representative snapshots of molecular dynamics simulations at different time points. Color coding: BF4⁻ (green), TEA⁺ (red), DMF solvent (translucent), K-PBI segment (purple), K-PBI-BIm segment (yellow).
[0054] Figure 19 Schematic diagram of the transition and transport of BF4⁻ anions in the K-19PBI-BIm membrane. Detailed Implementation
[0055] Non-aqueous flow batteries (NAFBs) require membranes with excellent resistance to organic solvents and high ionic conductivity in the organic solvent phase to withstand the inherent corrosive effects of non-aqueous electrolytes. This invention designs an imidazolium-grafted anion exchange membrane (AEM) that combines excellent resistance to organic solvents, a three-dimensional hierarchical topological porous structure, and rapid anion hopping conduction characteristics in the organic solvent phase. This membrane exhibits an anionic conductivity as high as 2.1 mS·cm⁻¹ in N,N-dimethylformamide (DMF) electrolyte, while also possessing excellent barrier properties, with permeabilities to the positive electrode electrolyte 2,1,3-benzothiadiazole and the negative electrode electrolyte 10-methylphenthiazide as low as 1.6 × 10⁻⁻⁻⁻⁶. 8 cm²・s⁻¹ and 3.8×10⁻ 9 cm²・s⁻¹. Furthermore, this membrane exhibits excellent stability in highly polar organic solvents. After 350 cycles at a current density of 5 mA・cm⁻² in a non-aqueous flow battery system, the average energy efficiency (EE) remains above 66.1%, far exceeding that of the commercial Celgard membrane (average EE of 48.8% after only 70 cycles). This provides an innovative strategy for the design of next-generation non-aqueous flow battery membranes, offering the dual advantages of excellent resistance to organic solvents and high ionic conductivity.
[0056] Example 1
[0057] Porous polybenzimidazole (PBI) membranes with a 3D hierarchical topology were prepared using a non-solvent-induced phase separation (NIPS) process, with the following steps: A 15 wt% casting solution was prepared by dissolving PBI polymer in DMAc and continuously mechanically stirred until a transparent and homogeneous dispersion was obtained. The prepared solution was allowed to stand at room temperature for 24 h to remove trapped bubbles before casting. Subsequently, the solution was uniformly cast onto a clean glass plate using a doctor blade (gap set to 200 μm). The wet membrane was immediately immersed in a phase inversion bath composed of isopropanol (IPA) and ethanol (EtOH) at volume ratios of 1:4, 1:9, 1:14, 1:19, and 1:24. The resulting membranes were labeled xPBI, where x corresponds to the proportion of EtOH in the mixed solution (i.e., x = 4, 9, 14, 19, 24). After NIPS, the membrane was carefully peeled from the glass substrate, thoroughly rinsed with IPA to remove residual solvent, and stored in IPA for subsequent characterization and modification.
[0058] Example 2: Preparation of K-xPBI membrane
[0059] Porous PBI membranes were immersed in a 5 wt% K₂S₂O₈ aqueous solution and treated at room temperature for 24 h to induce chemical crosslinking. Subsequently, the treated membranes were removed and thoroughly rinsed repeatedly with deionized water to remove residual K₂S₂O₈. The resulting chemically crosslinked PBI membranes were labeled K-xPBI, where x = 4, 9, 14, 19, 24 (representing the proportion of ethanol in the aforementioned phase inversion bath).
[0060] Example 3: Preparation of K-19PBI-Im membrane
[0061] Under optimized reaction parameters, α,α′-dibromo-p-xylene (DBX) was chemically grafted onto a K-19PBI membrane. The specific steps were as follows: First, a 3 wt% DBX solution was prepared by dissolving DBX powder in anhydrous acetonitrile. Then, the K-19PBI membrane was immersed in the aforementioned DBX solution and reacted at 80°C under reflux for 24 h to ensure sufficient grafting efficiency. After the reaction, the membrane was thoroughly washed with isopropanol (IPA) to remove unreacted DBX and residual byproducts, and then stored in IPA for subsequent modification. The resulting DBX-functionalized membrane was labeled K-19PBI-DBX.
[0062] Subsequently, the K-19PBI-DBX membrane underwent secondary grafting with ethylimidazolium (EIm), butylimidazolium (BIm), or hexylimidazolium (HIm). Specifically, the membrane was immersed in an NMP solution (0.1 mol / L) of 1-ethylimidazolium, 1-butylimidazolium, or 1-hexylimidazolium, and reacted at 60°C for 24 h with continuous stirring. The corresponding alkylimidazolium-grafted membranes are collectively referred to as the K-19PBI-Im series, and individual samples were named K-19PBI-EIm, K-19PBI-BIm, and K-19PBI-HIm, respectively.
[0063] To prepare K-19PBI-yBIm membranes (where y = 0.05, 0.1, 0.15, representing the molar concentration of 1-butylimidazole in NMP), K-19PBI-DBX membranes were immersed in a series of 1-butylimidazole / NMP solutions with gradient concentrations (0.05, 0.1, and 0.15 mol / L) and reacted under the same conditions (60 °C, 24 h) to investigate the effect of butylimidazole on membrane performance.
[0064] The main testing methods are as follows:
[0065] The electrolyte absorbance and swelling degree of the membrane were tested according to standard methods. The membrane was immersed in 1 MTEABF4 / DMF solution at room temperature for 24 hours. After immersion, excess solution on the surface of the wet membrane was gently blotted with lint-free paper, and its wet doping weight was recorded immediately. Subsequently, the diaphragm was vacuum dried at 120°C for 24 hours, and its dry weight was measured. Then, the dried membrane was thoroughly cleaned repeatedly with DMF to remove residual TEABF4, followed by vacuum drying at 120°C for 24 hours to obtain the dedoped dry weight. DMF , and polymers ( The mass fraction is calculated using the following formula:
[0066]
[0067]
[0068]
[0069] For swelling degree measurement, the diaphragm was cut into 2×2cm square samples. First, its dry length was measured and recorded. The dry film was then immersed in pure DMF at room temperature until swelling equilibrium was reached. After immersion, excess solution was gently blotted from the surface of the wet film with lint-free paper, and its wet length was immediately measured. The degree of swelling of the diaphragm is calculated using the appropriate formula. Degree of swelling
[0070]
[0071] in and These are the lengths of the wet film and the dry film, respectively.
[0072] Permeability of active substances
[0073] The permeability of the active material through the membrane was determined using a diffusion cell apparatus. The diffusion cell consisted of two chambers: the left chamber contained 50 mL of 0.1 M active material + 1.0 MTEABF4 solution (50 mL), while the right chamber contained 50 mL of blank 1.0 MTEABF4 / DMF solution. To minimize the effects of concentration polarization, magnetic stir bar was placed in both chambers for continuous stirring throughout the test. Every 5 hours, 4 mL of solution was sampled from the right chamber, and an equal volume of fresh 1.0 MTEABF4 / DMF solution was immediately added to maintain a constant volume. The concentration of the permeated active material in the sampled solution was quantitatively analyzed using a UV-Vis spectrophotometer. The permeability coefficient of the active material was calculated using the following formula:
[0074]
[0075] Among them, V R This indicates the constant volume of the solution in the right chamber (50 mL in this work); C R (t) represents the concentration (mol / L) of the active substance in the right chamber at time t; C L is the initial concentration (mol / L) of the active substance in the left chamber, which is considered a constant during the test period for simplified calculation; A and L are the effective permeation area (cm²) and thickness (cm) of the diaphragm, respectively; P is the permeation coefficient (cm²·s⁻¹) of the active substance.
[0076] The permeability of BF4⁻ anions across the membrane was measured using a similar diffusion cell method. In this case, the left chamber contained 50 mL of 1 M TEABF4 solution, and the right chamber contained 50 mL of DMF. The concentration change of BF4⁻ in the right chamber was monitored by tracking the conductivity of the solution using a conductivity meter, and the permeability coefficient of BF4⁻ was calculated using the same formula.
[0077] Wideband dielectric relaxation
[0078] The dielectric constant (ε) of the membrane was characterized using an impedance analyzer based on the alternating current (AC) impedance method. Before testing, the membrane sample was cut into circular pieces with a diameter of 14 mm. These circular pieces were pretreated by immersing in 1 MTEABF4 / DMF electrolyte for 24 h to achieve balanced doping, and then the surface was gently wiped dry with lint-free paper to remove excess solution. Metal electrodes were uniformly attached to both sides of the pretreated membrane to ensure good contact. During testing, an AC voltage signal was applied, and the corresponding current response was recorded to calculate key dielectric parameters, including the dielectric constant (ε). The frequency range was set from 0.01 Hz to 1 MHz to comprehensively capture the dielectric behavior of the membrane and reveal the variation of dielectric parameters in the low- to mid-high frequency region.
[0079] Ion transport number
[0080] The ion transport number of the membrane was studied using an electrochemical workstation. The membrane was assembled into an H-type electrolytic cell filled with different concentrations of TEABF4 / DMF electrolyte (0.1M / 0.3M), and current-voltage (IV) curves were recorded. To eliminate potential interference from electrode-solution interface polarization and liquid junction potential between the two electrolytes, two Ag / AgNO3 reference electrodes were used, each connected to the half-cell via a salt bridge filled with 1M TEABF4 solution. Under these optimized conditions, the open-circuit voltage (V0) of the device was equal to the diffusion potential (V_d) caused by the electrolyte concentration gradient, which was calculated using formula (6):
[0081]
[0082] Where R, T, F, t_BF4⁻, t_TEA⁺, and Δ are the gas constant (J·mol⁻¹·K⁻¹), Kelvin temperature (K), Faraday constant (C·mol⁻¹), BF4⁻ transport number, TEA⁺ transport number, and concentration gradient, respectively. It is worth noting that if the TEABF4 concentration gradient is high (e.g., 1 / 3 M), the mean ionic activity coefficient Δ of the TEABF4 solution must be considered.
[0083] Ion exchange capacity
[0084] The ion exchange capacity (IEC, meq / g) of the membrane was determined by the Mohr method as follows: First, a circular membrane sample (19 mm in diameter) in Br⁻ form was immersed in 0.1 M NaOH solution and 0.1 M NaCl solution, respectively, to convert it to OH⁻ and Cl⁻ forms sequentially. Each immersion step was maintained at room temperature for 24 h, and the immersion solution was changed every 12 h to ensure complete ion exchange. After the ion conversion process was completed, the membrane was removed and thoroughly rinsed with deionized (DI) water until the rinse water was neutral (to remove residual electrolytes), and then vacuum dried at constant weight for more than 12 h. Subsequently, the dried membrane was immersed in 50 mL of 0.05 M NaNO₃ solution and gently stirred for 24 h. Simultaneously, an indicator solution was prepared by mixing approximately 5 mL of K₂CrO₄ solution (5 wt%) with 150 mL of DI water. This indicator solution was then added to the NaNO₃ solution containing the membrane. Finally, titrate the entire mixture with a standardized 0.05 mg NO3 solution until a continuous brick-red precipitate appears, which is the titration endpoint.
[0085]
[0086] Where C AgNO3 V represents the concentration of the AgNO3 solution (0.1M). AgNO3 W_d is the volume of AgNO3 used in the titration, and W_d is the weight of the dry film. The reported data is the average of three experiments.
[0087] Ionic conductivity
[0088] The ionic conductivity of the membrane was evaluated using electrochemical impedance spectroscopy (EIS) on an electrochemical workstation. Prior to testing, the membrane sample was pre-immersed in 1 MTEABF4 / DMF electrolyte at room temperature for 24 h to achieve doping equilibrium, and then the surface was gently wiped dry with lint-free paper to remove excess solution. The pretreated membrane was then sandwiched between two stainless steel (SS) blocking electrodes to assemble a symmetrical coin cell structure (SS / Membrane / SS), ensuring tight contact between the membrane and the electrodes. EIS measurements were performed under the following conditions: frequency range of 10 Hz to 1 MHz, sinusoidal voltage amplitude of 10 mV, and room temperature. The ionic conductivity (σ) was calculated from the obtained EIS spectra using the following formula:
[0089]
[0090] Where L refers to the thickness of the diaphragm, R b A is the bulk resistance of the diaphragm, and A is the area of the inert electrode.
[0091] In-situ Fourier transform infrared spectroscopy
[0092] In-situ Fourier transform infrared (FTIR) spectra were acquired using a spectrometer equipped with a liquid attenuated total reflectance (ATR) accessory. All measurements were performed under a dry atmosphere to avoid moisture interference. The diaphragm sample was equilibrated in DMF for 30 min prior to spectral acquisition. Subsequently, a DMF solution of tetraethylammonium tetrafluoroborate (TEABF4) was added to the test system, and the background spectrum was immediately acquired and used as a baseline. The spectral acquisition protocol was set as follows: one spectrum was acquired every 15 s for the first 5 min; after this initial phase, spectra were acquired every 5 min. All spectra were recorded in absorption mode and compared with the pre-acquired background spectrum. The acquired spectral data were saved in text format for subsequent data processing.
[0093] The time-dependent spectral evolution is presented in ascending order from bottom to top, corresponding to spectra recorded from 15 s to 150 min. The relative absorbance change (ΔA) is defined as ΔA = A t -A0, where A0 refers to the baseline absorbance, A t This represents the time-dependent absorbance of the diaphragm at different measurement intervals.
[0094] Molecular dynamics simulation
[0095] The molecular structure was optimized at the B3LYP-D3(BJ) / 6-31+G(d,p) level using Gaussian 16A.03 software, and vibrational analysis was performed to ensure the absence of imaginary frequencies. RESP charges were calculated using Multiwfn 3.8 (dev) software, and GAFF force field parameters were obtained using the Acpype code. The polymer was represented by MMFF94 charges. Molecular dynamics simulations were performed using GROMACS 2021.7 software. Long-range electrostatic interactions were handled using the Particle Mesh Ewald (PME) method with a Coulomb cutoff radius of 1.2 nm. Van der Waals (vdW) interactions were handled using a force-switching method, where forces smoothly decayed to zero between 0.9 and 1.2 nm to reduce cutoff noise. Hydrogen-related bonds were constrained using the LINCS algorithm, and dispersion corrections were applied to energy and pressure. All simulation systems first undergo energy minimization using the steepest descent method with 5000 steps and a time step of 1 fs, followed by a 1 ns NPT pre-equilibrium run using a leapfrog MD integrator, and then a long (100 ns) unconstrained production run with a time step of 2 fs. Throughout all simulations, a velocity recalibration (V-rescale) thermostat (where τ...) is used. t =1.0ps) to maintain the temperature at 298.15K, and through a C-rescale isotropic constant voltage regulator (where τ = 1.0ps) pThe pressure was maintained at 1 bar (0.5 ps). Three-dimensional periodic boundary conditions (PBC) were used. The results were visualized using VMD 1.9.3 software.
[0096] Non-aqueous flow battery (NAFB) single cell performance
[0097] The separator was sandwiched between two carbon felt (SGL) electrodes and clamped by two graphite plates. The entire electrode-separator assembly was then fixed between two polytetrafluoroethylene (PTFE) plates to ensure structural stability. Both positive and negative electrode electrolytes were 10 mL of a DMF-based mixed solution containing 1 MTEABF4, 0.1 MBTD, and 0.1 MMPT. The SGL graphite felt (area: 2 cm × 2 cm) served as the positive and negative electrodes, while the graphite plates acted as current collectors for both electrodes. During battery testing, the anolyte was purged with nitrogen to eliminate oxygen interference, and the flow rate of both electrolytes was set to 50 mL·min⁻¹. Coulombic efficiency (CE), voltage efficiency (VE), and energy efficiency (EE) were calculated using the following formulas:
[0098]
[0099]
[0100]
[0101] The membrane characterization and test results are as follows:
[0102] The initial membrane was prepared using the non-solvent-induced phase separation (NIPS) method, and then crosslinked and modified with potassium persulfate (K2S2O8). The crosslinking effect generates a self-healing effect by reducing the intermolecular spacing of polybenzimidazole molecules, endowing the membrane with excellent resistance to organic solvents and outstanding barrier properties. Figure 2 and Figure 3 A cross-linked membrane with a three-dimensional hierarchical topological porous structure was obtained through controlled processing. A typical K-xPBI membrane synthesis process is as follows:
[0103]
[0104] Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) confirmed the successful preparation of the K-xPBI membrane.
[0105] The pore size distribution of the membrane was characterized using low-field nuclear magnetic resonance (LF-NMR) technology. Tests showed that a large number of micropores and mesoporous structures exist within the membrane channels. Figure 4Among them, the micropore size of the K-19PBI membrane was optimized to 5.2 Å, corresponding to the highest CO2 adsorption capacity. Based on the combined results of ionic conductivity, mechanical properties and battery performance tests, K-19PBI was selected as the substrate membrane for subsequent grafting modification.
[0106] To further improve the ionic conductivity of the membrane in organic solvents, α,α'-dibromo-p-xylene (DBX) was used as a grafting agent to graft three imidazolium groups with different alkyl chain lengths—1-ethylimidazolium (EIm), 1-butylimidazolium (BIm), and 1-hexylimidazolium (HIm)—onto the surface of the K-19PBI membrane. The synthesis steps of the K-19PBI-Im membrane are as follows, where R represents an alkyl side chain (including ethyl, butyl, and hexyl).
[0107]
[0108] The resulting imidazolium-grafted septum (K-19PBI-Im) Figure 5 b, Figure 6 The increased intensity of the band at 2800-2900 cm⁻¹ attributable to the CH stretching vibration of the alkyl chain confirms the successful grafting of different alkyl chains. This not only preserves the three-dimensional hierarchical porous structure of the substrate membrane, but also... Figure 5 c Figure 7 It also maintains excellent stability against organic solvents. Figure 8 The imidazole functional group significantly improved the electrolyte absorption rate of the membrane. Figure 5 d), thereby achieving a faster ion transport rate ( Figure 5 (e). Among the three K-19PBI-Im membranes, K-19PBI-BIm exhibits the most balanced overall performance, with the highest ionic conductivity (e). Figure 9 ), mechanical properties ( Figure 10 ), swelling degree ( Figure 11 ) and battery efficiency ( Figure 12 All were at their optimal levels. Therefore, this study further prepared K-19PBI-yBIm membranes with different butylimidazolium (BIm) grafting degrees (y represents the molar concentration of 1-butylimidazolium in N-methylpyrrolidone (NMP) solution) and conducted in-depth research on their performance.
[0109] The K-19PBI-yBIm membrane retains the three-dimensional hierarchical topological pore structure of the substrate membrane (Figure 13a), and with the increase of BIm grafting degree, the key performance indicators of the membrane show a regular optimization: the ion exchange capacity (IEC) increases from 0.49 meq·g⁻¹ to 1.23 meq·g⁻¹, the electrolyte absorption rate (Figure 13b) and ion transport rate (Figure 13c) are significantly improved, and the mechanical properties are further enhanced. The ionic conductivity test results show (Table 1) that the K-19PBI-0.15BIm membrane has an ionic conductivity as high as 2.14 mS·cm⁻¹ in DMF solution, which is much higher than that of the K-19PBI substrate membrane (0.098 mS·cm⁻¹) and the commercial Celgard 2500 membrane (0.45 mS·cm⁻¹).
[0110] Table 1
[0111]
[0112] In addition to high ionic conductivity, this series of membranes also exhibits extremely low permeability of active materials (d in Figure 13). Figure 13 (e). Notably, the K-19PBI-0.1BIm membrane exhibits significantly superior barrier properties compared to the commercially available Celgard 2500 membrane, with permeability to 10-methylphenhiazine (MPT) and 2,1,3-benzothiadiazole (BTD) as low as 3.8 × 10⁻ ... 9 cm²・s⁻¹ and 1.6×10⁻ 8 cm²・s⁻¹. For example... Figure 13 As shown in f, the prepared K-19PBI-BIm separator outperforms most reported non-aqueous flow battery separators in terms of overall performance, exhibiting both excellent ionic conductivity and active material barrier properties, with an ion selectivity as high as 1×10⁻⁶. 5 S・s・cm⁻³ (defined as the ratio of ionic conductivity to the permeability of the active material).
[0113] A non-aqueous flow battery was assembled under a nitrogen atmosphere using MPT as the negative electrode electrolyte, BTD as the positive electrode electrolyte, and a DMF solution of tetraethylammonium tetrafluoroborate (TEABF4) as the supporting electrolyte. The battery performance of the separator was then evaluated. Figure 14 (a).
[0114] Compared to commercially available Celgard membranes and ungrafted membranes, the K-19PBI-yBIm series membranes exhibit superior battery performance, with both voltage efficiency (VE) and energy efficiency (EE) increasing with increasing BIm grafting degree (Figure 14b, Figure 14c). However, excessively high grafting degree leads to excessively high electrolyte absorption by the membrane, causing severe swelling and consequently a decrease in coulombic efficiency (CE). Figure 14 (d) The decay of coulombic efficiency is a key factor restricting the long-term cycle stability of batteries and a core indicator that needs to be balanced in the design of separators. Based on the excellent overall battery performance, a K-19PBI-0.1BIm separator was selected for long-term cycle performance testing. As shown in Figure 14e, after 350 cycles at a current density of 5 mA·cm⁻², the average energy efficiency of this separator remained at around 65.2%, demonstrating excellent long-term cycle stability. In contrast, the commercial Celgard 2500 separator failed after only 70 cycles under the same conditions, with an average energy efficiency of only 48.8%. The above results indicate that the grafting of butylimidazolium side chains creates an efficient channel for ion transport, effectively improving the ionic conductivity of the separator in organic solvents. Compared with previously reported separators for non-aqueous flow batteries using similar electrolyte systems, the separator designed in this study has significant advantages in both cycle stability and energy efficiency (Figure 14f).
[0115] Mechanism of action of imidazolium cations:
[0116] To clarify the mechanism of action of imidazolium cations in BIM-grafted membranes, this study employed broadband dielectric spectroscopy (BDS) for analysis. In this system, membrane-electrode interface polarization and membrane bulk polarization occur simultaneously. Both polarizations coexist in the low-frequency range (10⁻²~10³ Hz), while in the mid-to-high-frequency range (10³~10⁻² Hz), they increase. 6 At low frequencies (Hz), the polarization of the diaphragm body is dominant. Under low-frequency conditions, the dielectric constant (ε′) of the diaphragm increases with the increase of the BIm grafting degree (Figure 15a, ). Figure 16 This is because a higher BIM grafting degree promotes the enrichment of more conductive active ions within the membrane, thereby enhancing interfacial and bulk polarization effects. Under medium- and high-frequency conditions, a rapidly alternating electric field hinders ion migration to the electrode interface, causing the dielectric constant to decay at a faster rate with increasing BIM grafting degree (Figure 16b). Therefore, it can be inferred that the enrichment of conductive active ions within the membrane originates from the adsorption of BF4⁻ anions by imidazolium cations.
[0117] To further investigate the nature of this adsorption, in-situ Fourier transform infrared spectroscopy (FTIR) was used for characterization. Over time, the Br⁻ anions adsorbed by the imidazolium cations within the membrane were gradually replaced by BF⁻ anions. The relative change in absorbance was defined as ΔA = A. t - A0 (A0 is the baseline absorbance of the membrane after equilibration in DMF, see Figure 17 for details; A t (The absorbance values are at different time points). This process is accompanied by significant spectral changes: the intensity of the BF bond stretching vibration peak of the BF4⁻ anion (1049 cm⁻¹ and 787 cm⁻¹) gradually increases, while the characteristic vibration peak of the imidazolium ring (1655~1660 cm⁻¹) undergoes a blue shift and its intensity changes. These spectral changes are attributed to the interaction between the imidazolium cation and the BF4⁻ anion, which drives the redistribution of electron cloud density within the imidazolium ring. After approximately 120 minutes, the spectral changes tend to stabilize. Figure 18 This indicates that the system has reached adsorption equilibrium.
[0118] Furthermore, the effect of the interaction between imidazolium cation and BF4⁻ anion on ion conduction was investigated. This was achieved by measuring the transport numbers of tetraethylammonium cation (TEA⁺) and BF4⁻ anion (…). Figure 15 (c) This study enabled a quantitative analysis of the contributions of cations and anions to ion transport. The results showed that the anion transport number of the K-19PBI-0.15BIm membrane was as high as 0.9942, indicating that this interaction can effectively promote BF4⁻ anion transport, making anion transport the dominant mode of ion transport.
[0119] Molecular dynamics (MD) simulations were used to analyze the interaction energy and diffusion coefficient before and after grafting imidazolium cations. The simulation results showed that the interaction energy between the imidazolium cations and BF4⁻ anions in the grafted membrane was significantly negative and fluctuated significantly over time (Figure 15, d), confirming a strong Coulomb-dominated interaction between the two, which drives the hopping migration of BF4⁻ anions between imidazolium cations. The mean square displacement (MSD) curve (Figure 15, e) showed that the diffusion coefficient of the BF4⁻ anion increased from 8.37 × 10⁻⁻⁻⁴ ... 7 cm²・s⁻¹ increased to 10.96×10⁻ 7 cm²・s⁻¹( Figure 15 (f). This enhanced diffusion performance stems from the intrinsic flexibility of the alkyl side chain, which disrupts the stable coordination structure between the imidazolium cation and the BF4⁻ anion, allowing the anion to rapidly jump to the adjacent imidazolium cation site (f). Figure 18Compared to ungrafted membranes, the K-19PBI-BIm membrane exhibits significantly improved performance. The core mechanism lies in the imidazolium cation driving intersite hopping migration of the BF4⁻ anion in organic solvents via coulombic interactions. Figure 19 The synergistic effect of imidazolium cations and BF4⁻ anions effectively promotes anion conduction, which is the core reason why this membrane performs excellently in non-aqueous flow batteries.
Claims
1. An anion exchange membrane, characterized in that, include: A porous base membrane composed of polybenzimidazole (PBI); the porous base membrane is chemically cross-linked; And alkylimidazolium cations covalently grafted onto the polybenzimidazole via linking groups.
2. The anion exchange membrane according to claim 1, characterized in that, The linking group is xyleneyl.
3. The anion exchange membrane according to claim 1 or 2, characterized in that, The alkyl group in the alkylimidazolium cation is a C1-C8 straight-chain or branched alkyl group.
4. A method for preparing an anion exchange membrane as described in any one of claims 1-3, characterized in that, Includes the following steps: a) Preparation of polybenzimidazole porous membranes via phase inversion method; b) Crosslink the polybenzimidazole porous membrane obtained in step a) using a chemical crosslinking agent; c) The cross-linked membrane is reacted with a linker, and then with an alkylimidazolium cation to graft alkylimidazolium onto the membrane.
5. The preparation method according to claim 4, characterized in that, The phase inversion method in step a) includes: dissolving polybenzimidazole in an organic solvent to form a casting solution, and immersing the casting solution in a non-solvent coagulation bath for phase inversion; preferably, the organic solvent is N,N-dimethylacetamide; the mass percentage concentration of polybenzimidazole in the casting solution is 10-20 wt%; the chemical crosslinking agent in step b) is persulfate; and the crosslinking treatment is carried out at 10-40°C for 12-36 hours.
6. The preparation method according to claim 4, characterized in that, The linker in step c) is α,α′-dihalo-p-xylene; in step c), the reaction with the linker is carried out at 60-100°C; the reaction with the alkylimidazolium is carried out at 40-80°C; the alkylimidazolium is selected from at least one of 1-ethylimidazolium, 1-butylimidazolium, and 1-hexylimidazolium; in the reaction with the alkylimidazolium, the concentration of the alkylimidazolium in the solvent is 0.01-0.2 mol / L.
7. The application of the anion exchange membrane according to any one of claims 1-3 in a non-aqueous flow battery.
8. A method for promoting anion transport in a non-aqueous electrolyte, characterized in that, include: An anion exchange membrane is used, the anion exchange membrane comprising a cross-linked polybenzimidazole porous matrix and a plurality of imidazolium cations covalently grafted onto the matrix; the anion exchange membrane is contacted with a non-aqueous electrolyte containing the anions; wherein the imidazolium cations adsorb the anions through coulombic interactions and promote the hopping migration of the anions between adjacent imidazolium cation sites.
9. The method according to claim 8, characterized in that, The imidazolium cation is an alkyl imidazolium cation, and its alkyl side chain disrupts the transient stable coordination between the imidazolium cation and the anion through its own flexible movement, thereby promoting the hopping migration of the anion; the alkyl side chain is a C1-C8 alkyl group; the anion is a tetrafluoroborate (BF4⁻) anion; through the method, the migration number of the anion in the membrane is increased to 0.95 or more, preferably 0.99 or more.
10. A method for improving the selective transport capability of polybenzimidazole-based anion exchange membranes for anions in non-aqueous electrolytes, characterized in that, The method includes the following steps: covalently grafting multiple alkylimidazolium cations onto a cross-linked polybenzimidazole porous matrix to form an anion adsorption site network composed of the alkylimidazolium cations within the membrane; wherein, the site network is used to enrich anions through Coulomb interactions upon contact with a non-aqueous electrolyte, and to provide a transport path for the anions dominated by hopping migration; the grafting of the alkylimidazolium cations improves the dielectric constant of the anion exchange membrane at 10⁻² Hz; the method results in an ion selectivity of the anion exchange membrane reaching 1 × 10⁻⁶. 5 The ion selectivity coefficient is defined as the ratio of ionic conductivity to the permeability of the active material, where S·s·cm⁻³ or higher.