Home Journals Progress in Chemistry
Progress in Chemistry

Abbreviation (ISO4): Prog Chem      Editor in chief: Jincai ZHAO

About  /  Aim & scope  /  Editorial board  /  Indexed  /  Contact  / 
Review

Preparation of Covalent Organic Framework Membranes by Interfacial Polymerization

  • Jiansong Liu ,
  • Guida Pan ,
  • Feng Zhang ,
  • Wei Gao ,
  • Juntao Tang , * ,
  • Guipeng Yu , *
Expand
  • College of Chemistry and Chemical Engineering,Central South University,Changsha 410083,China
* (Guipeng Yu);
(Juntao Tang)

Received date: 2024-07-10

  Revised date: 2024-09-30

  Online published: 2025-04-30

Supported by

Hunan province Funds for Distinguished Young Scientists(No. 2022JJ10080),the National Science Foundation of China(52103275)

Hunan province Funds for Distinguished Young Scientists(No. 2022JJ10080),the National Science Foundation of China(52173212)

Abstract

In recent years,covalent organic frameworks(COFs)have emerged as focal points in the research of membrane materials. Distinguished by their distinctive porous structures and structural versatility,COFs offer a promising avenue for advancement in membrane applications compared to conventional polymeric materials. This article delves into diverse interfacial systems,systematically detailing the methodologies for fabricating high-performance COF membranes via interfacial polymerization. The mechanisms underlying membrane formation across various interfacial systems and the strategies for precisely controlling the membrane structure will be elucidated. Furthermore,the intricate relationship between the membrane structure and application performance will be summarized. The challenges and perspectives in this field will be highlighted in the last part of this review.

Contents

1 Introduction

2 Gas/liquid interface polymerization

2.1 Langmuir-Blodgett method

2.2 Surfactant-mediated

3 Liquid/liquid interface polymerization

3.1 Regulation of the system

3.2 Additive-mediated

3.3 Optimizing synthetic conditions

4 Liquid/solid interface polymerization

5 Solid/gas interface polymerization

6 Applications of COF membrane

6.1 Water resource treatment

6.2 Gas separation and storage

6.3 Membrane catalysis

6.4 Electric device

7 Conclusion and outlook

Cite this article

Jiansong Liu , Guida Pan , Feng Zhang , Wei Gao , Juntao Tang , Guipeng Yu . Preparation of Covalent Organic Framework Membranes by Interfacial Polymerization[J]. Progress in Chemistry, 2025 , 37(5) : 686 -697 . DOI: 10.7536/PC240705

1 Introduction

Porous polymer materials are a class of porous materials that have rapidly emerged in recent years, combining to some extent the advantages of inorganic porous materials and traditional polymeric materials[1]. Among them, covalent organic frameworks (COFs) have attracted sustained academic attention due to their abundant pores, adjustable pore sizes, and high specific surface areas[2-5]. The tailored structures and tunable channels of COFs make them ideal candidates for constructing separation membranes and overcoming the trade-off effect. Meanwhile, the unique skeleton structures of COFs also facilitate the development of proton-conducting membranes and semiconductor thin films. However, the insoluble and infusible characteristics of COF materials significantly restrict their processability, posing challenges in preparing precise and highly efficient functional membranes, thus limiting further applications of COF materials. Therefore, obtaining self-supporting, defect-free, and large-area covalent organic framework thin films through novel preparation methods has become a major research focus in recent years.
At present, the main preparation methods for covalent organic framework membranes include blending method, in situ growth method, layer-by-layer stacking, and interfacial polymerization. Among these, the blending method results in poor uniformity of the dispersed particles, easily causing uneven microstructures in the composite membrane. In addition, the compatibility between the polymer matrix and filler particles is difficult to control, which may lead to defects. In situ polymerization generally forms membranes on the substrate surface; however, this method suffers from poor controllability during the membrane formation process, affecting the membrane quality. The layer-by-layer stacking method produces nanosheets whose size cannot be precisely controlled, and the regularity of pores can be easily damaged during the exfoliation process. In contrast, interfacial polymerization uses a well-defined interface as a template to guide the growth of COF membranes, effectively confining the reaction at the apparent phase boundary under mild conditions.
Interface aggregation utilizes a specific two-phase interface as the reaction system, thereby effectively confining the reaction to the phase interface and assembling the membrane in a confined environment[16-18]. By precisely controlling the physical and chemical properties of the two-phase interface, fine regulation of key physicochemical properties of the membrane such as thickness, pore size, hydrophilicity, and charge characteristics can be achieved[19]. This method demonstrates unique advantages by enabling structural control of the membrane at the molecular level, thus achieving efficient separation of specific molecules or ions.
As shown in Fig. 1, in recent years, researchers in the field have made numerous innovative and distinctive efforts in preparing COF membranes via interfacial polymerization. However, a systematic summary of the structures and properties of COF membranes under different interfacial systems is still lacking, and the understanding of the process and film-formation mechanisms remains insufficiently comprehensive. Therefore, this review will discuss key influencing factors during the interfacial polymerization process by focusing on four interfacial systems: liquid/gas, liquid/liquid, liquid/solid, and solid/gas. This analysis aims to elucidate preparation strategies and regulation methods for high-performance COF membranes, while briefly explaining the impacts of pore structure, crystallinity, and charge characteristics of COF membranes on subsequent application performance. Finally, this article will address the challenges faced in fabricating high-performance COF membranes through interfacial polymerization and provide an outlook on future development trends in this field. Through this review, we aim not only to provide references for research and application of COF membranes but also to inspire discussions and draw attention to the future development of high-performance COF membranes.
图1 界面聚合方法制备COF膜的时间历程图

Fig.1 Time course diagram of COF film preparation by interfacial polymerization method

2 Interfacial Polymerization at Gas/Liquid Interfaces

2.1 Langmuir-Blodgett Method

The gas/liquid interface is the region where gas and liquid phases come into contact. Researchers mainly control chemical reactions at the interface by altering characteristics such as interfacial tension, wettability, and gas solubility within the gas/liquid interface. In the preparation of COF membranes via gas/liquid interfacial polymerization, the Langmuir-Blodgett (LB) method is commonly employed. This method involves dissolving amphiphilic molecules in a solvent and arranging the molecules in a controlled manner at the gas-liquid interface to assemble them into a membrane. COF functional membranes prepared through LB interfacial polymerization exhibit features including large-scale ordered crystalline networks, long-range order, stable pores, and absence of membrane compression deformation (Table 1).
表1 气/液界面聚合制备的各COF膜

Table 1 Each COF membrane prepared by gas/liquid interfacial polymerization

Membrane Name Performances Thickness SBET(m2/g) Application ref
TFP-DHF
2DCOF
Cut-off molecular weight of 900 Da 2.9±0.3 nm 285 Molecular sieving 21
Truxene 2DP Cut-off molecule size of 1.3 nm 2~3 nm - Dye separation 22
COF-2,5-Ph Single crystal domain size in μm level 220 nm - Thin-film transistor 26
In 2015, Zhang Wei's research group[20] first utilized the air/water interface (Langmuir-Blodgett method), starting from simple aromatic triamines and dialdehyde building blocks, to synthesize two-dimensional covalent organic framework thin films via dynamic imine chemistry, as shown in Figure 2. They characterized the obtained COF monolayer films using optical microscopy (OM), scanning electron microscopy (SEM), and atomic force microscopy (AFM), confirming the formation of large-area (millimeter-scale), monomolecular-layer-thick aromatic polyimine thin films. Furthermore, they employed tip-enhanced Raman spectroscopy to characterize the chemical structure of the prepared films, with density functional theory simulations verifying the accuracy of the peak assignments. Their work opens new possibilities for the synthesis of customizable 2D polymers and systematic studies on their structure-property relationships.
图2 Langmuir-Blodgett法制备COF薄膜[20]

Fig.2 Preparation of COF thin films by Langmuir-Blodgett method[20]. Copyright 2015,John Wiley and Sons

Inspired by the research achievements of Zhang Wei's group, Lai Zhiping's group[21] synthesized large-area crystalline 2D COF membranes via the Langmuir-Blodgett method in 2018. This COF membrane was prepared through the condensation reaction of 2,4,6-triformylphloroglucinol (TFP) and 9,9-dihexylfluorene-2,7-diamine (DHF), exhibiting a tilted AA-stacked hexagonal lattice with an approximate pore size of 1.41 nm. Compared to amorphous membranes fabricated using the same process, the TFP-DHF 2D COF membrane demonstrated clearly defined porous structures and significantly enhanced solvent permeability by approximately two-fold. All these results indicate the great potential of two-dimensional COF membranes for high-performance organic solvent nanofiltration applications.
Subsequently, the Lai Zhiping research group[22] further utilized hexaalkylated triaminotriamine [Tru(NH2)3] and terephthalaldehyde (TPA) as reactive monomers to successfully synthesize large-area imine bond-linked 2D COF films at the air/water interface using the Langmuir-Blodgett method. Studies showed that by adjusting the surface pressure during polymerization, the thickness of the COF film could be precisely controlled within a range of 0.85–1.7 nm. When the surface pressure was slightly higher than that required for forming a densely packed monolayer, a dense membrane with a thickness of approximately two to three molecular layers was successfully obtained, and the film morphology was confirmed through SEM and AFM analyses. They also demonstrated that the dynamic imine chemistry-based self-correction mechanism helps enhance the short-range order and crystallinity of the film. Through filtration tests, they found that the membrane exhibited excellent desalination performance for both monovalent salts (NaCl) and divalent salts (MgSO4). Additionally, it demonstrated effective size-selective separation for large organic molecules (bromothymol blue), achieving a rejection molecular size of approximately 1.3 nm, which is close to the predicted pore size of the membrane.

2.2 Surfactant-Mediated

Unlike the LB method described above, introducing surfactants represents a new strategy for preparing two-dimensional COF membranes at the gas/liquid interface. The length of surfactant molecules can influence the crystallinity of the membrane; for instance, the crystallinity of two-dimensional polymers largely depends on the ordering of small-molecule surfactants[23]. Additionally, because of the weak intermolecular interactions among low molecular weight substances, the assembly of surfactant molecules is susceptible to external conditions such as concentration and solution pH[24-25]. The research group led by Zhi-Kun Zheng[26] previously proposed a macromolecular surfactant-mediated approach for synthesizing 2D COFs thin films with large single-crystal domains. Compared with small-molecule surfactants, macromolecules connected via covalent bonds provide a more stable system. Meanwhile, these covalent bonds offer long-range order along the polymer chains as well as excellent chemical stability.
Subsequently, the research group of Zheng Zhikun[27] utilized a poly(sodium 4-styrenesulfonate) (PSS)-mediated synthetic approach to achieve controllable synthesis of organic two-dimensional material films with large single-crystal domains, thereby providing a general method for preparing oriented 2D COF films with large single-crystal domains at the air-water interface. They employed ordered arrays of charged groups to stabilize precursors and intermediates carrying opposite charges; negatively charged PSS diffuses at the air/water interface, guiding the assembly, polymerization, and crystallization of protonated monomers. Using 5,10,15,20-tetrakis(4-aminophenyl)-21H,23H-porphyrin (TAPP) and 2,5-dihydroxyterephthaldehyde (2,5-Ph) as monomers, they prepared two-dimensional COF films with an average single-crystal domain size of approximately 3.57 ± 2.57 μm2.
In addition, the research group of Xinliang Feng successfully fabricated monolayer 2D COF thin films (4) and multilayer 2D COF thin films (5) through a Schiff-base condensation reaction between two porphyrin monomers and 2,2-dihydroxyterephthaldehyde at the air-water interface, by introducing intramolecular hydrogen bonds into the COF membranes. The prepared monolayer 2D COF thin film (4) not only exhibits a large size (4 inches) but also outstanding mechanical properties. In addition to investigating the application of COF membranes in separation, this research group further demonstrated that the monolayer 2D COF can serve as an active semiconducting layer in thin-film transistors (TFTs), while the multilayer 2D COF membrane (5) based on cobalt porphyrin can efficiently catalyze water splitting for hydrogen production.

3 Liquid/Liquid Interface Polymerization

The liquid/liquid interface refers to the contact region between two immiscible liquids. Controlling chemical reactions occurring at the liquid/liquid interface typically involves considering factors such as interfacial tension between the two phases, liquid viscosity, and molecular interactions within the liquids, as shown in Figure 3. For the synthesis of COF membranes, the liquid/liquid interface is the most commonly used interface. Strategies for modulating the liquid/liquid interface include adjusting the interface system, utilizing additive mediation, and enhancing the membrane formation process. Controlling the diffusion rates of monomers and the reaction rate at the interface are key factors in fabricating continuous and stable high-crystallinity COF membranes (Table 2).
图3 液/液界面聚合示意图

Fig.3 Liquid/liquid interface polymerization schematic

表2 液/液界面聚合制备的COF膜

Table 2 COF membranes prepared by liquid/liquid interfacial polymerization

Membrane Name Performances Thickness SBET
(m2/g)
Application ref
TaPa-Py Membrane of 64 cm2 24 nm 1145 Nanofiltration 29
t-TpMa-COF High permeability(H+)of 1015.7 mol·L-1·g-1 71 μm - Liquid deacidification 30
TFB-BD Permeability(acetonitrile)of 523 L·m-2·h-1·bar-1 2 μm 1096 Dye separation 31
COF-DhTG Rejection(NaCl)of 93.0%-93.6%; Permeability of
1.7~3.7 L·m-2·h-1·bar-1
5.3 μm 24 Desalination 32
TFB-BD Dye rejection(RhB)>98.0% 4 μm 99.5 Dye separation 33
Tp-Byp Permeability(acetonitrile)of 339 L·m-2·h-1·bar-1 75 nm 1151 Dye separation 34
TAPB-PDA Pore size of 1.5 nm 2.5 nm~100 μm 563 Nanofiltration 37
COF-JLU2 Pore size(0.49~0.51 nm) 50~400 nm 353 Nanofiltration 38
3D SCOF Proton conductivity of 843 mS·cm-1; Power density of 21.2 W·m-2; Selectivity of 0.976 1 μm 49 Proton conduction 39
Acidamide-COF Permeability of 482.3 L·m-2·h-1·bar-1; Dye rejection(methylene blue)>99%) 29.56 μm 984 Dye separation 40
COF-MeCN Permeability of 9.4 L·m-2·h-1·bar-1; Rejection(Na2SO4)of 85% 115 nm 610 Desalination 44
TpPa/PSF Dye rejection(Congo red)of 99.5%; Permeability of
50 L·m-2·h-1·bar-1
500 nm - Dye separation 45
TFPM-HZ/PAN Permeability of 44 L·m-2·h-1·bar-1; Operation time of 1000 h 250 nm 1000 Organic solvent nanofiltration 46
COF-DT Dye rejection(Alcian bule)of 98.6% 395 nm 1546 Dye separation 47
TAPB-PDA Permeability(DETHz-Tb)of 0.2 L·m-2·h-1·bar-1 70 μm 1269 Dye separation 48
TpPa-75-250-2 Selectivity(NaCl/Adriamycin)of 41.8; Permeability of
48.09 L·m-2·h-1·bar-1
78 nm - Antibiotic desalination 49

3.1 Regulation of Interface Systems

Shaffer's research group synthesized large-area, ultrathin 2D COF membranes with dimensions up to 64 cm2 using 2,4-pyridinediamine (Pa-Py) and benzidine (Pa-Bz) as reaction monomers with 2,4,6-triformylphloroglucinol (Tp) at the toluene/water interface. By depositing onto seven different substrates, COF films with thickness of 24 nm were obtained consistently, demonstrating substrate-independent fabrication[29]. They achieved angstrom-precise control over pore size by employing monomers with different molecular sizes. This method overcame some limitations of the Langmuir-Blodgett technique for synthesizing COF membranes, solving the problem of requiring halogenated solvents incompatible with polymeric substrates. Performance testing revealed that the organic solvent nanofiltration flux of these COF membranes was an order of magnitude higher than that of commercial polymer membranes.
Malijian research group used Tp and m-phenylenediamine (Ma) as reaction monomers to prepare a Turing structure t-TpMa-COF membrane at the water/chloroform interface[30], as shown in Fig. 4. The team observed that the two sides of the prepared membrane exhibited different morphologies: the chloroform side displayed a dense Turing structure similar to the surface of lotus leaves with hydrophobic properties, while the water and acetonitrile side showed a porous structure with superhydrophilicity. In addition, the excellent acid stability, thermal stability, and irradiation stability of this membrane enable it to withstand extreme environments such as high temperature, high acidity, and strong radiation during post-treatment of waste fuels. Meanwhile, the t-TpMa membrane exhibits exceptional capability for intercepting metal ions in complex multi-ion systems, achieving rejection rates as high as 100%. Under conditions of an initial acidity of 5 mol·L-1 HNO3, the membrane reduced the filtrate acidity by more than 1 mol·L-1 within 24 h, with a hydrogen ion permeation flux reaching 1015.7 mol·L-1·m-2.
图4 制备具有双向各向异性图灵结构的t-TpMa-COF膜的示意图[30]

Fig.4 Schematic representation of the preparation of t-TpMa-COF membranes with bidirectional anisotropic Turing structures[30]. Copyright 2021,John Wiley and Sons

However, the interfacial systems introduced in the aforementioned studies have some drawbacks, such as the use of volatile organic solvents, whose evaporation can interfere with the liquid-liquid interface and affect the assembly of COF membranes. Meanwhile, the properties of the organic solvent-water interface are difficult to precisely regulate. To overcome these limitations, Lai Zhiping's research group[31] employed ionic liquids as alternative organic solvent phases and, using a diffusion-controlled strategy, fabricated highly crystalline free-standing COF membranes at the IL-H2O interface, exhibiting a surface area over four times that of COF membranes synthesized at the dichloromethane-H2O interface. In this process, the diffusion rate of the 4,4'-biphenyldiamine (BD) monomer was controlled by hydrogen bonding between the catalyst PTSA and the amine, whereas the diffusion rate of the TFB monomer was regulated by high-viscosity ILs containing long alkyl chains. Benefiting from their exceptionally high crystallinity and uniform pore size, these COF membranes demonstrated remarkable permeability to acetonitrile, acetone, water, methanol, and ethanol, while simultaneously exhibiting excellent selective rejection of dyes.
Aqueous two-phase systems (ATPS) are organic solvent-free systems used for the formation of multi-level structures and the preparation of functional materials. Compared to oil-water systems, ATPS possess fully aqueous compositions, eliminating environmental pollution caused by solvent volatilization. The Jiang Zhongyi research group[32] first reported an interfacial polymerization strategy based on ATPS for preparing COF membranes. Aqueous solutions containing polyethylene glycol (PEG) and dextran (Dex) undergo phase separation, forming ATPS. Researchers dispersed aldehyde and amine monomers into separate phases and precisely controlled the properties of the COF-DhTGCl membrane by adjusting factors such as the Dex/PEG mass ratio, pH value, and reaction time. Due to the similar thermal capacities of the two phases, heat generated at the interface (from imine bond formation) dissipates easily, resulting in a flat and dense structure on the membrane surface. The prepared COF membrane exhibited high NaCl rejection rates ranging from 93.0% to 93.6%, with a permeability of 1.7 to 3.7 L·m-2·h-1·bar-1. The Zheng Liqiang research group[33] proposed a system using temperature-responsive ionic liquid (IL)-based two-phase systems for synthesizing COF membranes. After forming a biphasic aqueous interface at the appropriate temperature, the monomers and catalyst were separately dispersed into the IL-rich and water-rich phases. Due to the high viscosity of the IL-rich phase and hydrogen bonding in the water-rich phase, the diffusion of both aldehyde and amine monomers could be simultaneously controlled, leading to reactions occurring at the aqueous biphasic interface. By optimizing the interfacial properties, crystalline COF membranes with ultra-high solvent flux and excellent dye rejection performance were obtained.

3.2 Additive-Mediated

Thermodynamic control conditions often lead to the formation of amorphous polymers due to faster reaction rates. To address this issue, in 2017, the Banerjee group[34] introduced a salt-mediated approach (amine-p-toluenesulfonic acid (PTSA) salt), as shown in Figure 5. They reacted Tp with four different-sized amine monomers to prepare four COF films: Tp-Bpy, Tp-Azo, Tp-Ttba, and Tp-Tta. In this method, hydrogen bonding between PTSA and amine slows the diffusion rate of monomers in the aqueous phase, favoring crystallization under thermodynamic control conditions. PXRD patterns of these COF films confirmed their high crystallinity and showed good agreement with simulated AA stacking models. Using this strategy, they further synthesized four COF films with different pore sizes at room temperature. The highly crystalline two-dimensional COF membranes could be retained at the liquid-liquid interface and were easily transferred onto various substrates including glass surfaces, metals, and porous grids. Notably, the crystal structure and porosity of the free-standing films exhibited no differences compared to those of the corresponding powder materials.
图5 胺-对甲苯磺酸(PTSA)盐介导调控制备高结晶性COF膜的示意图[34]

Fig.5 Schematic diagram of amine-p-toluenesulfonic acid(PTSA)salt-mediated modulation for the preparation of highly crystalline COF membranes[34]. Copyright 2017,American Chemical Society

Alshareef's research group also employed a salt-mediated approach in their study, preparing COF films on a glass substrate using two immiscible solvents: water and chloroform[35]. The team utilized a three-layer solvent process with 4,4'-azodianiline and Tp as precursors to synthesize COF films containing azo functional groups (Tp-azo). In this system, the bottom layer consisted of a chloroform solution of Tp, while the top layer was an aqueous solution of azo-p-toluenesulfonic acid (PTSA) salt. An intermediate layer of deionized water further separated these two layers, preventing premature mixing of the solutions and thereby slowing down the chemical reaction rate. After formation at the interface between the two phases, the COF film could deposit onto the glass substrate located at the bottom of the organic phase.
Unlike the amine-p-toluenesulfonic acid salt strategy used by the Banerjee group, the Dichtel group had previously demonstrated that Sc(OTf)₃ is a highly active catalyst for the synthesis of imine-based COFs in their earlier work[36]. The synthesized powder samples exhibited excellent crystallinity and surface area at room temperature. This catalyst shows high water tolerance and can accelerate imine formation, with the polymerization restricted exclusively to the interface even when both monomers are dissolved in the organic phase. Typically, the two monomers initially polymerize to form an amorphous network, which subsequently becomes long-range ordered under thermodynamic control through reversible exchange processes of the imine bonds. Given the significant impact of Sc(OTf)₃ catalyst on the formation of imine-type COFs, the Dichtel group employed an interfacial polymerization strategy by dissolving multifunctional amines and aldehydes in the upper organic phase and the Lewis acid Sc(OTf)₃ in the lower aqueous phase, ultimately yielding highly crystalline, free-standing COF films at the interface between the two phases[37]. This method provides a new approach for synthesizing continuous 2D COF films linked via imine bonds, where the lateral size of the film is determined by the size of the polymerization vessel and the thickness is controlled by the initial monomer concentration, ranging from 2.5 nm to 100 μm. They transferred the COF membrane onto a polyethersulfone support and demonstrated through testing that the resulting COF-PES membrane exhibited strong retention performance toward rhodamine WT.
Jiang Zhongyi's research group[38] proposed a Brønsted acid-mediated one-step self-assembly method. As shown in Figure 6, a Brønsted acid and 2,4,6-triformylphloroglucinol (Tp) are added to the organic phase, while hydrazine hydrate (Hh) is introduced into the aqueous phase, allowing polymerization to occur at the interface between the two phases. The Brønsted acid acts as a multifunctional medium that facilitates the transformation of COFs from amorphous to crystalline, ensures confined film growth at the interface, and regulates the assembly behavior of COF subunits, thereby playing a crucial role in controlling the microstructural evolution of the COF-JLU2 membrane. Specifically, the Brønsted acid controls error correction and structural reconstruction of the COF by slowing down the overall reaction, thus endowing the membrane with high crystallinity. Furthermore, they established a correlation among the membrane structure, separation performance, and the partition coefficient (logP) of the Brønsted acid. When logP ranges between 1.0 and 3.0, COF membranes with continuous active layers, tunable thickness (50–400 nm), and small pore sizes (0.49–0.51 nm) can be fabricated. These pore sizes exceed the kinetic diameter of n-butanol (0.48–0.50 nm), and the resulting COF-JLU2 membrane exhibits significantly superior n-butanol dehydration performance compared to previously reported pervaporation membranes.
图6 Brønsted酸介导的一步界面聚合在聚合物载体上制备COF-JLU2膜的示意图[38]

Fig.6 Schematic illustration of Brønsted acid mediated one-step interfacial polymerization to fabricate a COF-JLU2 membrane on a polymeric support[38]. Copyright 2012,Royal Society of Chemistry

To further investigate the effect of acids on the membrane growth process, Jiang Zhongyi's research group[39] also employed a dual-acid-mediated interfacial polymerization strategy to fabricate three-dimensional sulfonic acid-functionalized COF membranes (3D SCOF) for efficient and selective ion transport. The researchers fixed octanoic acid as the organic solvent and introduced various acids, including trifluoroacetic acid, p-toluenesulfonic acid, and acetic acid, into the aqueous phase to carry out different dual-acid interfacial polymerization processes. The team found that acetic acid exhibited an appropriate diffusion rate within the reaction system; by coordinating the diffusion and reaction processes of the monomers, the reaction could be confined to the interface region, thereby obtaining a continuous and dense membrane at the interface. The resulting 3D SCOF membrane possesses highly interconnected ion transport channels, ultra-microporous pore size (0.97 nm), and abundant sulfonic acid groups (ion exchange capacity up to 4.1 mmol·g-1), exhibiting a high proton conductivity of 843 mS·cm-1 at 90 °C.
The above-mentioned works all involve COF membranes connected by imine bonds. Due to the dynamic characteristics within COFs linked via imine bonds, their chemical stability under harsh conditions is relatively low. Zhao Xin's research group[40] fully combined the characteristics of amide bonds and COFs, utilizing an interfacial polymerization strategy with tris(4-aminophenyl)benzene (TAPB) and benzene-1,3,5-tricarbonyl chloride (TFB) as reactive monomers and p-toluenemethanesulfonic acid as a catalyst. They first synthesized an imine-linked COF membrane, which was then treated with potassium peroxymonosulfate to obtain, for the first time, an amide-linked COF membrane. Compared with imine- and amine-linked COF membranes fabricated from the same monomers, the amide bond-connected COF exhibited better solution permeability, separation selectivity, and mechanical strength. Moreover, benefiting from the hydrogen bonding interactions between amide bonds, it demonstrated improved stability.
Another approach to address the issue of instability is the synthesis of COFs linked by irreversible bonds. The Li Ming group combined the Suzuki coupling reaction with an interfacial polymerization strategy, developing a method for synthesizing 2D COF membranes linked by C—C bonds under mild reaction conditions and with a broad range of monomers.[41] They synthesized two types of COF membranes, 2D CCOF1 and 2D CCOF2, which exhibited large lateral dimensions while maintaining internal order. Due to the broad compatibility of the Suzuki coupling reaction, this strategy can be effectively applied to the synthesis of various novel 2D CCOF membranes, and may also be extended to other similar reactions, such as the Sonogashira and Heck reactions.
Zhou's research group[42] investigated the influence of solvent polarity during COF crystallization and found that highly polar solvents exhibit greater hydrogen-bond donating ability, which enhances the crystallinity of COFs. Dichtel's group[43] also reported that introducing nitrile-functionalized co-solvents can interact with covalent bonds, thereby promoting the growth of COF microcrystals. Xu Zhi's research group[44] added acetonitrile as a solvent inducer into an aqueous phase. As shown in Figure 7, the introduced acetonitrile enhances the interaction between the TpPa network and the solvent, effectively generating low-crystallinity TpPa nanoparticles, thus reducing the strong coupling between interfacial reaction and self-assembly processes. These nanoparticles can undergo reversible bond exchange and achieve higher crystallinity. They demonstrated through X-ray diffraction and nitrogen adsorption-desorption characterizations that the resulting membranes exhibit significantly improved crystallinity and porosity compared to membranes prepared without using a solvent inducer, thereby greatly enhancing the desalination and permeability performance of TpPa membranes.
图7 溶剂诱导界面聚合制备COF薄膜的方案示意图[44]

Fig.7 Scheme illustration of solvent-induced interfacial polymerization to prepare COF films[44]. Copyright 2022,Elsevier

3.3 Reinforced Membrane Preparation Process

In previous studies, the prepared COF films usually need to be transferred onto porous substrates, a process that is often complicated and time-consuming. In addition, the adhesion between thin COF layers and substrates has also attracted widespread attention. To address these challenges, the research group of Wang Yong [45] developed a strategy for directly synthesizing COF films at polymer matrix interfaces. They sequentially deposited COF precursors of Tp and paraphenylenediamine (Pa) onto a PSF ultrafiltration substrate, enabling direct growth of a COF film integrated with the substrate within 1 min due to a moderate reaction rate. Using this method, they successfully fabricated highly permeable TpPa/PSF composite membranes that outperformed those prepared using other strategies, demonstrating the feasibility of improving COF films based on polymer matrix interfaces. They further extended this strategy [46] by introducing a porous support at the interface between organic and aqueous phases. Specifically, tetra(4-formylphenyl)methane (TFPM) and hydrazine hydrate (HZ) were dissolved in ethyl acetate, while Sc(OTf)· was dissolved in water, and the porous PAN membrane served as a buffer layer to achieve controlled diffusion of Sc(OTf)· from water to ethyl acetate, resulting in TFPM-HZ/PAN membranes with crystallinity. Notably, the unique 3D topological design of the membrane with ultrasmall pore size and remarkable pore connectivity enhances both selectivity and permeability simultaneously, surpassing state-of-the-art organic solvent nanofiltration (OSN) membranes in permeability-selectivity performance.
The research group of Juliping [47] employed interfacial polymerization through reverse diffusion, introducing aqueous phases containing TAPB and HAc, as well as an oil phase containing DMTP, into opposite sides of a self-made diffusion cell separated by a hydrolyzed PAN membrane. The hydrolyzed PAN substrate, due to its hydrophilic surface, acted as a passivating agent, allowing TAPB to slowly diffuse from the aqueous phase into the oil phase. During interfacial polymerization, the monomer concentration dominated the formation of the COF-DT thin film. Increasing the monomer concentration enabled more TAPB molecules to diffuse onto the growing layer's surface within the same reaction duration. The synthesized COF membrane exhibited tunable solvent permeability and solute rejection rate, with thickness showing linear dependence.
Due to the lack of effective methods to control and monitor the membrane growth process in real time, as the liquid/liquid interface merely acts as an inert physical carrier interface, Su Bin's research group[48] first utilized the interfacial potential difference existing between different interfaces, employing the liquid/liquid interface as a proton pump. By adjusting the potential difference across the interface, protons are transported from the aqueous phase to the organic phase, thereby promoting interfacial polymerization for the preparation of COF membranes. Based on the presence of interfacial potential differences, all conventional electrochemical methods are applicable to the liquid/liquid interface; thus, the growth of COF membranes can be monitored in real time via interfacial double-layer capacitance measurements, and ion transfer currents can be used for in-situ detection of the membrane permeability. Using this method, they successfully fabricated three types of centimeter-sized, highly crystalline, free-standing imine COF membranes with different pore sizes and surface functionalities under room temperature conditions. Their defect-free and well-ordered porous structures were confirmed through molecular permeation and ultrafiltration tests, demonstrating that this potential difference regulation method holds great promise for studying interface growth mechanisms and enabling scalable synthesis of functional free-standing COF membranes.
To address the complexity and time-consuming nature of the transfer process, Meng Hong's and Dong Liangliang's research groups reported a novel green and industrially adaptable scratching-assisted interfacial polymerization (SAIP) technique for the fabrication of scalable and uniform TpPa-COF membranes. This process utilizes nontoxic, low-volatile ionic liquids (ILs) as the continuous phase, replacing traditional organic solvents, to enable interfacial synthesis of the TpPa COF layer on a support membrane. This approach not only enhances the environmental sustainability of the membrane formation process and the mechanical strength of the resulting membrane, but also allows the production of large-area continuous membranes (19×25 cm2) with a thickness of approximately 78 nm within just 2 minutes.

4 Liquid/Solid Interface Polymerization

Compared with other synthetic methods, polymerization and crystallization at the interface require longer time, leading to a significant trade-off between membrane thickness and defect structure. The self-supporting COF membranes produced via liquid/liquid interfacial polymerization usually require tedious optimization of solvents to form stable two phases, allowing for the synergistic matching of diffusion, reaction, and crystallization processes at appropriate temperatures. In contrast, the liquid/solid interface inherently provides a stable solid phase that eliminates surface tension, while enabling regulation of surface properties to control and optimize chemical reactions, as shown in Figure 8. Specifically, compared with liquid/liquid interfacial polymerization, the liquid/solid interfacial polymerization confines the reaction to the restricted liquid/solid interface, simultaneously enhancing the self-healing and self-inhibition effects during the polymerization process (see Table 3).
图8 液/固界面聚合的示意图

Fig.8 Schematic representation of polymerization at the liquid/solid interface

表3 液/固界面聚合制备的各COF膜

Table 3 Each COF membrane prepared by liquid/solid interface polymerization

Membrane Name Performances Thickness SBET
(m2/g)
Application ref
COFDT Rejection(NaCl)of 99.99%; Permeability of 220 L·m-2·h-1 300~500 nm 790 Desalination 50
TaPa Rejection(NaCl)of 99.96%; Permeability of 92 kg·m-2·h-1 85 nm 252 Desalination 51
TTA-DHTA Young's modulus of 25.9±0.6 GPa; 4~150 nm - Ultrafiltration 52
The Feng Xiao research group[50] reported a strategy for preparing membrane distillation membranes composed of vertically aligned channels with hydrophilic gradients by engineering defects in covalent organic framework membranes through the removal of imine bonds. They dissolved the monomers in ethyl acetate and prepared dense COFDT thin films with varying thicknesses (300~500 nm) on glass substrates, followed by alkaline etching to obtain free-standing thin films. Characterization using atomic force microscopy revealed differences in hydrophilicity between the two sides of the COF membrane; the side attached to the glass substrate exhibited a smooth surface, while the other side showed relatively higher roughness. By employing a diffusion etching strategy, they achieved the fabrication of COF thin films with gradient variations in pore size and hydrophilic/hydrophobic environments within the pores along the depth direction at a confined solid/liquid interface. Molecular dynamics simulations further demonstrated the enhanced evaporation effect of water inside the confined nanopores. This COF membrane achieved ultra-high flux desalination, reaching a flux of up to 220 L·m-2·h-1 while maintaining a NaCl rejection rate of 99.99%.
Jiang Zhongyi's research group[51] reported an electrochemical interfacial polymerization strategy for preparing ultrathin COF membranes. In this strategy, the concentration of electrons on the cathode causes a deprotonation reaction to occur at a confined solid/liquid interface, providing favorable kinetics and thermodynamics for interfacial polymerization. During the electrochemical process, the non-coated regions exhibit higher current density, which promotes preferential polymerization of COF monomers in these areas, endowing the membrane with self-healing capability and a defect-free structure. Additionally, the pre-deposited insulating layer reduces the current density and suppresses the electrochemical deprotonation reaction, exerting a self-inhibitory effect after forming a continuous TpPa membrane. By adjusting the electrochemical time and voltage, they could conveniently control both the self-healing and self-inhibition effects, achieving precise regulation of membrane thickness and defect control. X-ray diffraction characterization revealed that the characteristic diffraction peak at 4.7° was relatively sharp, indicating that the TpPa membrane obtained after 4 h of electrochemical reaction time exhibited high crystallinity. When tested using a 7.5 wt% NaCl saline solution, the resulting COF membrane demonstrated excellent performance with a permeate flux of 92 kg·m-2·h-1 and a salt rejection rate of 99.96%.
At the liquid/gel interface system, an anomalous diffusion liquid layer forms on the immersed gel surface. Compared to air/liquid or liquid/liquid interfaces, this system enables rapid formation of an adjustable ultrathin sealed super-diffusion liquid layer between the liquid and gel, providing a robust interface for chemical reactions. Wang Dong's research group[52] proposed utilizing this system to combine the advantages of other interfaces for preparing high-quality free-standing COF films, as shown in Figure 9. When water droplets contact the hydrogel in an oil phase, an ultra-spreading water layer forms on the hydrogel surface. Consequently, they introduced amine monomers and aldehyde monomers into the hydrogel and oil phase, respectively, observing that these reactive monomers could diffuse into the thin super-diffusion water layer, resulting in uniform COF films. The prepared COF films exhibit controllable thickness ranging from 4 to 150 nm and crystallinity with certain orientation, with pore sizes of approximately 3.4 nm. AFM indentation tests revealed that the COF films have a Young's modulus of 25.9±0.6 GPa, indicating excellent mechanical properties. These COF membranes can not only be applied in nanofiltration separation but also demonstrate high performance in sensing applications.
图9 液/凝胶界面制备COF薄膜示意图[52]

Fig.9 Schematic diagram of COF films prepared at the liquid/gel interface[52]. Copyright 2018,American Chemical Society

5 Interfacial Polymerization at Solid/Gas Interface

The solid/gas interface is the contact region between a solid and a gas. Due to its stable solid phase, researchers can control the diffusion rate of the gas phase at the solid/gas interface to achieve stable chemical reactions. Compared to other interface systems, solid/gas interfacial polymerization offers two distinct advantages: first, the reaction rate can be increased by raising the temperature without disturbing the interface; second, the presence of monomers in the static solid phase confines the reaction to occur only at the interface. This solid/gas interfacial polymerization strategy has significant advantages in addressing the kinetic mismatch between polymerization and crystallization and in reducing membrane thickness (Table 4).
表4 固/气界面聚合制备的各COF膜

Table 4 COF membrane prepared by solid/gas interface polymerization

Membrane
Name
Performances Thickness SBET
(m2/g)
Application ref
TFP-PDA Permeability(acetonitrile)of 583 L·m-2·h-1·bar-1; Rejection(Alcian bule)>98% 9 nm 610 Nanofiltration 53
TpPa-1/Alumina Permeability(ethanol)of 200 L·m-2·h-1·bar-1; MWCO of 700 Da 100~500 nm - Nanofiltration 54
Jiang Zhongyi's research group[53] demonstrated a solid/gas interfacial polymerization method, in which they used TFP anchored on a support as the solid monomer and p-phenylenediamine (PDA) dispersed in the gas phase as the gaseous monomer to fabricate COF membranes, as shown in Figure 10. Due to the high diffusion rate and kinetic energy of PDA vapor, the membrane can be prepared within 9 h, which is 8 times faster than previously reported methods for synthesizing similar COF membranes. Compared with liquid-phase molecular monomers, gaseous-phase molecular monomers possess higher kinetic energy, generate higher vapor pressure at elevated temperatures, and exhibit faster diffusion rates, thereby accelerating both the polymerization process and homogeneous nucleation at the solid/gas interface, ultimately shortening the time required for COF formation. This approach effectively addresses and overcomes the temporal mismatch between rapid IP reaction and its relatively slow crystallization process. X-ray diffraction characterization revealed that when the reaction time reached 9 h, the crystalline peak at 4.7° became very sharp and exhibited significantly higher relative intensity compared to the peak at 27°, indicating highly crystalline membrane formation at this stage. By adjusting the TFP concentration, the membrane thickness could be reduced to 120 nm, and the resulting TFP-PDA membrane exhibited ultra-high permeability toward both water and acetonitrile.
图10 固/气界面制备COF薄膜示意图[53]

Fig.10 Schematic diagram of COF films prepared at the soild/gas interface[53]. Copyright 2020,American Chemical Society

Xu Rong's research group[54] reported a novel gas/gas-solid (V/V-S) method for growing ultrathin COF membranes on the inner cavity surface of alumina hollow fibers. In this method, two monomers were vaporized and introduced into a polydopamine-modified alumina substrate as a gaseous mixture (V/V), enabling efficient polymerization and crystallization at the confined gas-solid curved interface, thereby producing the TpPa-1/Alumina COF membrane. By adjusting the number of growth cycles, they controlled the COF membrane thickness between 100 and 500 nm. This COF membrane exhibited excellent stability and performance over 80 h in continuous cross-flow organic solvent nanofiltration, with a methanol permeability of approximately 200 L·m-2·h-1·bar-1, while the dye molecular weight cutoff (MWCO) was about 700 Da.

6 Covalent Organic Framework Membranes for Application

COF films have been widely applied in fields such as water resource treatment, gas separation and storage, catalysis, sensors, and electronic devices due to their advantages of well-ordered and stable pore channels, tunable pore size, and high specific surface area.

6.1 Water Resource Treatment

COF membranes are primarily applied in water resource treatment, including seawater desalination and wastewater treatment, involving technologies such as microfiltration, ultrafiltration, nanofiltration, reverse osmosis, membrane distillation, and pervaporation. Researchers aim to develop COF liquid separation membranes with high permeability and selectivity by adjusting properties such as pore size, crystallinity, specific surface area, and membrane thickness. Meanwhile, efforts continue to enhance the hydrophilicity and fouling resistance of COF membranes. For instance, introducing sulfonic acid groups on the pore walls of COF membranes[55] can create abundant negatively charged sites, thereby achieving high water flux and effectively adsorbing pollutants or separating multivalent salt ions, dye molecules, and organic salt ions, which contributes to environmental remediation and resource recovery.

6.2 Gas Separation and Storage

COF membranes have demonstrated high efficiency, low energy consumption, and convenient operation in applications such as gas separation and storage. Due to their rich and uniform pore structures, COFs can serve as gas separation membranes that surpass the Robeson upper bound limit, enabling energy-saving and efficient molecular sieving. Researchers have employed strategies such as introducing side groups, staggered stacking, and fabricating composite layers[56] to reduce membrane pore sizes for selective gas separation (H2/CO2, H2/CH4). Additionally, by incorporating polymers with intrinsic microporosity[57], the synergistic effect of precise size-based sieving through PIM channels and rapid molecular transport via COF channels has been applied to nitrogen gas separation (CO2/N2). The development of COF membranes suitable for gas separation plays a significant role in achieving energy conservation and emission reduction, enhancing economic benefits, and promoting an ecological civilization system characterized by harmonious coexistence between humans and nature.

6.3 Membrane Catalysis

The effectiveness of COF powders in catalyzing various transformations has been widely demonstrated by numerous studies. The tunability of COFs allows catalytic sites to be deliberately positioned within robust, open membrane pore channels. Compared to traditional batch processes, membrane reactors are known for their high efficiency and superior operational flexibility. Researchers have leveraged the molecular tunability of COFs to expand their applicability in membrane reactors[58], a performance that significantly surpasses that of corresponding homogeneous catalysts and COF powders. In membrane catalysis, the increased reactant concentration and rapid removal of water generated within the membrane greatly accelerate the reaction, reduce the apparent activation energy, and thereby enhance both the efficiency and selectivity of catalytic reactions. Preparing COFs into membranes for catalytic applications holds great promise in fields such as organic synthesis, environmental catalysis, and energy conversion.

6.4 Electronic Devices

COFs possess potential applications in functional electronics due to their unique advantages such as abundant π-electron arrays, high electron mobility, tunable bandgap, and good crystallinity. When COFs are fabricated into membranes for sensor applications, the porous structure and unique affinity for water molecules enable rapid mass transfer and effective utilization of water-binding sites[59]. In addition, COF membranes can achieve efficient charge transfer efficiency owing to their structures favorable for charge transport, thus finding applications in organic electronics and optoelectronic devices[60], particularly in fields such as organic photovoltaics, light-emitting diodes, and supercapacitors.

7 Conclusion and Prospect

The regular pore structure, easily adjustable pore size, and high porosity of COFs make them high-performance membrane materials applicable to various fields. This article mainly summarizes and discusses several representative works through four interfacial polymerization systems, focusing on the preparation strategies and applications of COF membranes. Furthermore, from an application perspective, this article briefly analyzes the structure-property relationship between the hierarchical structures and performance of COF membranes.
The development of high-performance COF membranes still faces challenges, including how to completely transfer the prepared COF membranes and directly fabricating free-standing COF membranes at a large scale without substrate support. These issues will be the focus of future research. In addition, to enhance their universality, more in-depth studies on the following aspects are required.
(1) Synthesis and structural optimization of novel covalent organic framework membranes: Utilizing topological principles and dynamic covalent chemistry theory to explore new structural units and bonding methods for preparing novel covalent organic frameworks, and investigating universal approaches to obtain highly stable structures.
(2) External Field Regulation and Interfacial Film Formation Mechanism: Explore external field-regulated monomer diffusion and assisted film formation strategies, such as introducing electric or magnetic fields to precisely control the stacking and assembly of COF membranes. Meanwhile, thoroughly investigate the film formation mechanism at the interface, and utilize simulation calculations to analyze how pore irregularities influence membrane charge properties, crystallinity, and other characteristics.
(3) Industrialized Production and Green Optimization: Design membrane fabrication devices adaptable to large-scale industrial production and expand novel application scenarios such as biomimetic recognition and osmotic power generation, aiming at scaling up and transferring COF membranes. Meanwhile, implement green optimization processes to reduce the volatilization of organic solvents and the use of toxic reagents, thereby enhancing production efficiency and environmental sustainability.
[1]
Dou H Z, Xu M, Wang B Y, Zhang Z, Wen G B, Zheng Y, Luo D, Zhao L, Yu A P, Zhang L H, Jiang Z Y, Chen Z W. Chem. Soc. Rev., 2021, 50(2): 986.

[2]
Geng K Y, He T, Liu R Y, Dalapati S, Tan K T, Li Z P, Tao S S, Gong Y F, Jiang Q H, Jiang D L. Chem. Rev., 2020, 120(16): 8814.

[3]
Côté A P, Benin A I, Ockwig N W, O’Keeffe M, Matzger A J, Yaghi O M. Science, 2005, 310(5751): 1166.

[4]
Waller P J, Gándara F, Yaghi O M. Acc. Chem. Res., 2015, 48(12): 3053.

[5]
Huang N, Zhai L P, Coupry D E, Addicoat M A, Okushita K, Nishimura K, Heine T, Jiang D L. Nat. Commun., 2016, 7: 12325.

[6]
Liu Y, Wang M W, Lu H L, Chen J Q. Chem. Ind. Eng. Prog., 2021, 40(08): 4360.

(刘祎, 汪明旺, 吕宏凌, 陈金庆. 化工进展, 化工进展, 2021, 40(08): 4360.).

[7]
Zhu J Y, Chen T T, Han S Q, Zhang Y T. J. Zhengzhou Univ. Eng. Sci., 2023, 44(1): 13.

(朱军勇, 陈田田, 韩双乔, 张亚涛. 郑州大学学报(工学版), 2023, 44(1): 13.).

[8]
Yang H, Wu H, Pan F S, Li Z, Ding H, Liu G H, Jiang Z Y, Zhang P, Cao X Z, Wang B Y. J. Membr. Sci., 2016, 520: 583.

[9]
Wang C B, Li Z Y, Chen J X, Li Z, Yin Y H, Cao L, Zhong Y L, Wu H. J. Membr. Sci., 2017, 523: 273.

[10]
Kandambeth S, Biswal B P, Chaudhari H D, Rout K C, Kunjattu H S, Mitra S, Karak S, Das A, Mukherjee R, Kharul U K, Banerjee R. Adv. Mater., 2017, 29(2): 1603945.

[11]
Halder A, Karak S, Addicoat M, Bera S, Chakraborty A, Kunjattu S H, Pachfule P, Heine T, Banerjee R. Angew. Chem. Int. Ed., 2018, 57(20): 5797.

[12]
Zhang W X, Zhang L M, Zhao H F, Li B, Ma H P. J. Mater. Chem. A, 2018, 6(27): 13331.

[13]
Zhang X K, Li H, Wang J, Peng D L, Liu J D, Zhang Y T. J. Membr. Sci., 2019, 581: 321.

[14]
Wang H J, Zhai Y M, Li Y, Cao Y, Shi B B, Li R L, Zhu Z, Jiang H F, Guo Z Y, Wang M D, Chen L, Liu Y W, Zhou K-G, Pan F S, Jiang Z Y. Nat. Commun., 2022, 13: 7123.

[15]
Wang R, Zhou Y S, Zhang Y, Xue J, Caro J, Wang H H. Adv. Mater., 2022, 34(44): 2204894.

[16]
Qu Y Q, Cha Y Q, Jia H G, Xu S P, Zhang M Y. China Plastics, 2022, 36(11): 150.

(蘧延庆, 查雨欣, 贾宏葛, 徐双平, 张明宇. 中国塑料, 中国塑料, 2022, 36(11): 150.).

[17]
Xian W P, Zuo X H, Zhu C J, Guo Q, Meng Q W, Zhu X C, Wang S, Ma S Q, Sun Q. Nat. Commun., 2022, 13: 3386.

[18]
Yang J L, Tu B, Zhang G J, Liu P C, Hu K, Wang J R, Yan Z, Huang Z W, Fang M N, Hou J J, Fang Q J, Qiu X H, Li L S, Tang Z Y. Nat. Nanotechnol., 2022, 17(6): 622.

[19]
Yuan S S, Li X, Zhu J Y, Zhang G, Van Puyvelde P, Van der Bruggen B. Chem. Soc. Rev., 2019, 48(10): 2665.

[20]
Dai W Y, Shao F, Szczerbiński J, McCaffrey R, Zenobi R, Jin Y H, Schlüter A D, Zhang W. Angew. Chem. Int. Ed., 2016, 55(1): 213.

[21]
Shinde D B, Sheng G, Li X, Ostwal M, Emwas A H, Huang K W, Lai Z P. J. Am. Chem. Soc., 2018, 140(43): 14342.

[22]
Gadwal I, Sheng G, Thankamony R L, Liu Y, Li H F, Lai Z P. ACS Appl. Mater. Interfaces, 2018, 10(15): 12295.

[23]
Liu K J, Qi H Y, Dong R H, Shivhare R, Addicoat M, Zhang T, Sahabudeen H, Heine T, Mannsfeld S, Kaiser U, Zheng Z K, Feng X L. Nat. Chem., 2019, 11(11): 994.

[24]
de Miguel G, Hosomizu K, Umeyama T, Matano Y, Imahori H, Pérez-Morales M, Martín-Romero M T, Camacho L. J. Colloid Interface Sci., 2011, 356(2): 775.

[25]
Togashi D M, Romão R I S, Gonçalves da Silva A M, Sobral A J F N,, Costa S M B. Phys. Chem. Chem. Phys., 2005, 7(22): 3874.

[26]
Ou Z W, Liang Z H, Dong X, Tan F L, Gong L, Zhao P, Wang H L, Liu W, Zheng Z K. Chin. J. Chem., 2021, 39(12): 3322.

[27]
Ou Z W, Liang B K, Liang Z H, Tan F L, Dong X, Gong L, Zhao P, Wang H L, Zou Y H, Xia Y J, Chen X D, Liu W, Qi H Y, Kaiser U, Zheng Z K. J. Am. Chem. Soc., 2022, 144(7): 3233.

[28]
Sahabudeen H, Qi H Y, Glatz B A, Tranca D, Dong R H, Hou Y, Zhang T, Kuttner C, Lehnert T, Seifert G, Kaiser U, Fery A, Zheng Z K, Feng X L. Nat. Commun., 2016, 7: 13461.

[29]
Shevate R, Shaffer D L. ACS Nano, 2022, 16(2): 2407.

[30]
Zhou L H, Li X F, Cao K C, Jia Z M, Long H H, Li Y, Tao G H, Liu N, Zhang J, Ma L J. Adv. Funct. Mater., 2022, 32(9): 2108178.

[31]
Gao S Q, Li Z Y, Yang Y Y, Wang Z Z, Wang Y L, Luo S J, Yao K S, Qiu J K, Wang H Y, Cao L, Lai Z P, Wang J J. ACS Appl. Mater. Interfaces, 2021, 13(30): 36507.

[32]
Wang H J, Zhao J S, Li Y, Cao Y, Zhu Z T, Wang M D, Zhang R N, Pan F S, Jiang Z Y. Nano-Micro Lett., 2022, 14: 216.

[33]
Liang L P, Wang Y Q, Ren W R, Gao X P, Zheng L Q, Lu F. ACS Appl. Polym. Mater., 2023, 5(11): 9477.

[34]
Dey K, Pal M, Rout K C, Kunjattu H S, Das A, Mukherjee R, Kharul U K, Banerjee R. J. Am. Chem. Soc., 2017, 139(37): 13083.

[35]
Hota M K, Chandra S, Lei Y J, Xu X M, Hedhili M N, Emwas A H, Shekhah O, Eddaoudi M, Alshareef H N. Adv. Funct. Mater., 2022, 32(23): 2201120.

[36]
Matsumoto M, Dasari R R, Ji W, Feriante C H, Parker T C, Marder S R, Dichtel W R. J. Am. Chem. Soc., 2017, 139(14): 4999.

[37]
Matsumoto M, Valentino L, Stiehl G M, Balch H B, Corcos A R, Wang F, Ralph D C, Mariñas B J, Dichtel W R. Chem, 2018, 4(2): 308.

[38]
Wang H J, Chen L, Yang H, Wang M D, Yang L X, Du H Y, Cao C L, Ren Y X, Wu Y Z, Pan F S, Jiang Z Y. J. Mater. Chem. A, 2019, 7(35): 20317.

[39]
Zhu T H, Kong Y, Lyu B H, Cao L, Shi B B, Wang X Y, Pang X, Fan C Y, Yang C, Wu H, Jiang Z Y. Nat. Commun., 2023, 14: 5926.

[40]
Lu Y, Zhou Z B, Qi Q Y, Yao J, Zhao X. ACS Appl. Mater. Interfaces, 2022, 14(32): 37019.

[41]
Zhou D, Tan X Y, Wu H M, Tian L H, Li M. Angew. Chem. Int. Ed., 2019, 58(5): 1376.

[42]
Elewa A M, EL-Mahdy A F M, Hassan A E, Wen Z H, Jayakumar J, Lee T L, Ting L-Y, Mekhemer I M A, Huang T F, Elsayed M H, Chang C L, Lin W-C, Chou H H. J. Mater. Chem. A, 2022, 10(23): 12378.

[43]
Smith B J, Parent L R, Overholts A C, Beaucage P A, Bisbey R P, Chavez A D, Hwang N, Park C, Evans A M, Gianneschi N C, Dichtel W R. ACS Cent. Sci., 2017, 3(1): 58.

[44]
Wu Y L, Wang Y X, Xu F, Qu K, Dai L H, Cao H Y, Xia Y S, Lei L F, Huang K, Xu Z. J. Membr. Sci., 2022, 659: 120799.

[45]
Wang R, Shi X S, Xiao A K, Zhou W, Wang Y. J. Membr. Sci., 2018, 566: 197.

[46]
Shi X S, Zhang Z P, Yin C C, Zhang X, Long J H, Zhang Z, Wang Y. Angew. Chem. Int. Ed., 2022, 61(36): e202207559.

[47]
Guo H K, Jiang J X, Fang C J, Zhu L P. J. Mater. Chem. A, 2023, 11(36): 19374.

[48]
Yang R J, Liu S S, Sun Q, Liao Q B, Xi K, Su B. J. Am. Chem. Soc., 2022, 144(26): 11778.

[49]
Pan Y, Liu H H, Huang Z Q, Zhang W H, Gao H Q, Liang L J, Dong L L, Meng H. Angew. Chem. Int. Ed., 2024, 63(4): e202316315.

[50]
Zhao S, Jiang C H, Fan J C, Hong S S, Mei P, Yao R X, Liu Y L, Zhang S L, Li H, Zhang H Q, Sun C, Guo Z B, Shao P P, Zhu Y H, Zhang J W, Guo L S, Ma Y H, Zhang J Q, Feng X, Wang F C, Wu H G, Wang B. Nat. Mater., 2021, 20(11): 1551.

[51]
Wang M D, Wang Y T, Zhao J Y, Zou J Y, Liang X, Zhu Z T, Zhu J S, Wang H J, Wang Y H, Pan F S, Jiang Z Y. Angew. Chem. Int. Ed., 2023, 62(13): e202219084.

[52]
Hao Q, Zhao C Q, Sun B, Lu C, Liu J, Liu M J, Wan L J, Wang D. J. Am. Chem. Soc., 2018, 140(38): 12152.

[53]
Khan N A, Zhang R N, Wu H, Shen J L, Yuan J Q, Fan C Y, Cao L, Olson M A, Jiang Z Y. J. Am. Chem. Soc., 2020, 142(31): 13450.

[54]
Siow W J S, Chong J Y, Ong J H, Kraft M, Wang R, Xu R. Angew. Chem. Int. Ed., 2024, 63(32): e202406830.

[55]
Liu X, Wang J R, Shang Y X, Yavuz C T, Khashab N M. J. Am. Chem. Soc., 2024, 146(4): 2313.

[56]
Fan H W, Peng M H, Strauss I, Mundstock A, Meng H, Caro J. Nat. Commun., 2021, 12: 38.

[57]
Wu X Y, Tian Z Z, Wang S F, Peng D D, Yang L X, Wu Y Z, Xin Q P, Wu H, Jiang Z Y. J. Membr. Sci., 2017, 528: 273.

[58]
Zheng L P, Zhang Z Q, Lai Z Z, Yin S J, Xian W P, Meng Q W, Dai Z F, Xiong Y B, Meng X J, Ma S Q, Xiao F S, Sun Q. Nat. Commun., 2024, 15: 6837.

[59]
Liu X Q, Huang R R, Peng L Y, Yang J L, Yan J B, Zhai B B, Luo Y, Zhang C, Tan S W, Liu X Y, Ding L P, Fang Y. Angew. Chem. Int. Ed., 2025, 64(2): e202414472.

[60]
Bonaccorso F, Sun Z, Hasan T, Ferrari A C. Nat. Photonics, 2010, 4(9): 611.

Outlines

/