English

• Water is the material basis for human survival and development, and is also an environmental medium that is highly susceptible to interference and pollution from human activities. With the rapid development of industrialization, pollutants in water threaten human health and ecosystems. Organic pollutants are an important part of water pollutants, such as organic dyes, organic fluorides, organic pesticides and antibiotics [1]. Currently, the search for purification strategies for the efficient removal of novel organic pollutants from water remains a serious challenge.

  In recent years, various materials such as activated carbon [2], graphene oxide [3], porous organic polymers (POPs) [4], metal-organic frameworks (MOFs) [5], and covalent organic frameworks (COFs) [6,7] have been used for the efficient removal of pollutants from water. Among them, COFs are an emerging class of porous crystalline organic materials. Due to their ordered geometry, high porosity, low density, high specific surface area, as well as clear structure and good chemical stability, they are considered to be one of the most promising materials for environmental remediation applications. However, the currently produced COFs exists in powder form, which limits its practical application [8]. To overcome this problem, many methods have recently been reported to mold COFs into macroscopic porous materials such as aerogels and membranes and other objects [9-11]. Among them, aerogels are a class of extremely porous and low-density solid materials distinguished by a unique 3D interconnected pore network [12] that facilitates mass transport of guest molecules through their matrix [13].

  Currently, several COF-based aerogels, such as pure COF aerogel [14-17], COF/cellulose aerogel [8], COF/chitosan aerogel [18-25], and COF/graphene aerogel [26-32], have good application prospects in environmental remediation. For instance, Li and co-works have first reported a series of porous COF-based chitosan aerogels for removing pollutants from water [18]. Zhang and co-works have prepared structurally stable ultralight COF/cellulose hybrid aerogels for the first time [33]. Zamora's team has developed a simple three-step method for the preparation of COF aerogels based on sol-gel transformation, solvent exchange and supercritical CO₂ drying [13]. Li and co-works have prepared pure COF aerogels with imine linkages using scandium trifluoromethanesulfonate as a catalyst, which have prospective applications in efficient iodine uptake [34]. In summary, there are generally two methods for preparing COF aerogels. The first one is the mechanical blending of nano COF powder and polymer, which may result in the phenomenon of COF powder shedding. The second is the method of COF self-supporting to form COF pure aerogels, which is not universal. Addressing these issues remains a huge challenge. In contrast, the COF composite aerogels prepared by in-situ synthesis method on substrates could combine the advantages of both and avoid the shortcomings.

  Based on that, graphene oxide (GO) is chosen as an ideal substrate for assembling extended structures due to its hydrophilic surface and large specific surface area [35-38], and it has been shown to exhibit good applications in water treatment. Moreover, GO has been successfully used to in-situ mediate the formation of 3D multistage porous aerogel materials by means of 2D COF nanosheets [26,35,37]. However, their poor mechanical property, higher density and ordinary adsorption capacity limit their recycling and practical applications. Firstly, the selection of suitable COF is very important for adsorption performance, and it is promising to optimize the adsorption effect by modulating the functional groups of COF. Secondly, the addition of suitable cross-linking agents is expected to solve the mechanical problem. Sodium alginate (SA) is a natural biomolecule with the -COOH group, which is non-toxic, non-harmful and biocompatible [39-41], the introduction of SA is expected to significantly improve the mechanical properties. The structural characteristic of SA makes it will form hydrogen bonds with GO and COF components thereby increasing chemical cross-linking interactions [39,42-45]. Therefore, there is an urgent need to develop superelastic and ultralight COF/GO aerogel materials with excellent adsorption capacity.

  In this study, ultralight COF/SA/RGO (CSR) aerogels with three-dimensional (3D) network structure are prepared by in-situ hydrothermal method. COFs grow in-situ on the graphene surface during the hydrothermal reaction process, and strong hydrogen-bond interactions between SA and graphene oxide sheets strengthen the 3D network, ultimately forming CSR composite aerogels. In addition, CSR aerogels with different COFs types and contents can be obtained by varying the COFs ligand and the ratio of COFs in the composite aerogels. The prepared CSR aerogels have ultra-light density and ultra-high elasticity. The loading of COF in CSR-S-3 and CSR-N-1.6 aerogels could reach 63.83% and 47.27%, respectively. Using the same mass of CSR aerogels and COFs powder, the CSR aerogels show faster extraordinary adsorption performance and convenient use in the removal of organic dyes and antibiotics. This study provides an effective strategy for synthesizing novel COFs aerogel adsorbents.

  The synthesis principle and process of COF powder and CSR aerogels are shown in Figs. 1a-c, Fig. 2a and Figs. S1-S4 (Supporting information). In order to find suitable reaction conditions of COF aerogels, four individual COFs are firstly synthesized under hydrothermal conditions. The X-ray diffraction (XRD) patterns prove that the COFs have high crystallinity (Figs. S1-S4 in supporting information), indicating that the crystalline COFs are successfully prepared [46-48]. Then, graphene is chosen as the substrate to induce in-situ growth of COFs nanosheets on graphene in aqueous solution. Sodium alginate is introduced to chemically cross-link with graphene through hydrogen bonding for the purpose of increasing intermolecular interactions and thus enhancing the mechanical properties of the material (Fig. 2b). During the hydrothermal reaction, graphene oxide is partially reduced to reduced graphene oxide (RGO). Moreover, aldehyde ligand Tp as the organic linker reacts with the amino ligand Pa-SO₃H (Pa-NO₂) in a Schiff base reaction to form COF. It is worth noting that COF nanosheet in-situ grows uniformly along the surface of 2D graphene nanosheets to form a heterojunction of the two phases. Furthermore, a black hydrogel is obtained and washed with methanol, water, methanol and water to remove unreacted reagents. Finally, an ultralight CSR aerogel with a multistage porous structure is formed after freeze-drying.

  Figure 1

  

  Figure 1.  (a) Synthesis schematic for the preparation of CSR aerogels. (b) COFs monomers and catalysts used in this work. (c) Structures and images for four different COF gels.

  DownLoad: Full-Size Img PowerPoint

  Figure 2

  

  Figure 2.  (a) Synthesis procedure of CSR via the hydrothermal method. (b) Hydrogen bonding interactions between sodium alginate and GO.

  DownLoad: Full-Size Img PowerPoint

  The effect of different groups on COFs aerogels molding is explored by varying the amino ligands. Four amino monomers (Pa-SO₃H, Pa-NO₂, Pa, BD) are selected to prepare aerogels. When Pa and BD are used for in-situ growth to form composites, it fails to form aerogel varying the addition amount of raw ligand. In contrast, when Pa-SO₃H and Pa-NO₂ are used for in-situ growth on graphene, well-formed aerogels CSR-S and CSR-N are obtained. This may be due to the fact that the side chain groups enhance the non-covalent force between COFs and graphene oxide. In the preparation of CSR-S and CSR-N aerogels, the volume and weight of the aerogels increase with increasing COFs content under the condition of constant GO and SA content. The mass of aerogel is 15 mg when the COFs content is 0 mg and named as SR. For CSR-S aerogel, as the mass of Tp increases from 5.33 mg to 10.67 mg and 16 mg, the mass of CSR-S aerogels increases from 30 mg to 45 mg and 60 mg, and the aerogels are named as CSR-S-2, CSR-S-3 and CSR-S-4, respectively. For CSR-N aerogel, when the mass of Tp increases from 2.99 mg to 5.97 mg, the mass of CSR-N aerogels increases from 15 mg to 24 mg, and the aerogels are named CSR-N-1.2 and CSR-N-1.6. With the increase of COFs addition, their density decreases and their volume increases, which may be due to the fact that the addition of COFs also cooperates with graphene to support the construction of 3D networks. The X in the above-mentioned CSR-X represents the weight ratio of the formed aerogels to the base SR aerogel. According to the Fig. S5 (Supporting information), it can be seen that the CSR-S-4 aerogel is not well molded after hydrothermal synthesis, and the aerogel fractures and tends to collapse after freeze-drying. Therefore, CSR-S-3 aerogel is used for subsequent performance testing, and the optimum ratio of CSR-S aerogel is CSR-S-3. The optimal ratio of CSR-N aerogel is CSR-N-1.6, and the specific additions are shown in Tables S1 and S2 (Supporting information). Unless otherwise indicated, CSR-S aerogel refers to CSR-S-3 aerogel and CSR-N aerogel refers to CSR-N-1.6 aerogel.

  Scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) mapping shows that both COF-S (Figs. 3a and b) and COF-N (Figs. 3c and d) have a similar rod-like morphology with a diameter of about 500 nm and a length of a few micrometers. The CSR-S-3 aerogel has a very low density of about 7.5 mg/cm³, and can be easily held on the fur (Fig. 3e). As shown in Fig. S7 (Supporting information), the micro-morphological configurations of both RGO and SA/RGO (SR) aerogels exhibit similar porous structures. The thickness of the monolayer of SA/RGO aerogel is significantly thicker than that of RGO, which is attributed to the introduction and cross-linking of sodium alginate that enhance the bending stiffness of graphene walls. The microstructure of the CSR aerogels is observed by SEM to analyze the reason for the low density. It could be observed from SEM images that the extended and interconnected nanosheets form a 3D sponge-like structure with a rough surface and many folds (Figs. 3f and g, Fig. S6d in Supporting information). The aerogels have highly open multilevel pores, including mesopores and macropores. The corresponding EDS mapping images show that C, N, O, S elements of CSR-S-3 (Fig. 3h) and C, N, O of CSR-N-1.6 (Fig. S6e in Supporting information) are uniformly distributed throughout the region, further demonstrating that the graphene nanosheets are completely and uniformly covered by the COFs.

  Figure 3

  

  Figure 3.  SEM and the corresponding elemental mapping images of COF-S (a, b) and COF-N (c, d). (e) A photograph of an ultralight CSR-S-3 aerogel standing on a setaria viridis. (f-h) SEM images of CSR-S-3 and the corresponding EDS plots. (i) XRD images of CSR-S-3 and COF-S. (j) FT-IR spectrum of SR, COF-S and CSR-S-3. (k) Stress-strain curves of CSR-S-3 aerogel at different strains (inset: images of COF aerogel before and after being compressed). (l, m) TEM images of CSR-S-3. (n) AFM image and the corresponding height profiles of CSR-S-3. (o) Density comparative plots of the COF aerogels.

  DownLoad: Full-Size Img PowerPoint

  The feasibility of hydrothermal synthesis of aerogels is verified by powder X-ray diffraction. The PXRD analysis of the RGO aerogel shows major peak at 23.6°. After the addition of SA during the aerogel synthesis process, a strong diffraction peak appears at around 24.6° in the SA/RGO aerogel (Fig. S11 in Supporting information). Compared to the RGO aerogel, the position of the main diffraction peak of the RGO/SA aerogel is shifted by 1°. This indicates that the distance between graphite flake layers is reduced due to the introduction of SA, which may result in the improvement of the mechanical properties of the aerogel. The appearance of a peak at 4.6° for the two COFs powder and their corresponding CSR-S-3 and CSR-N-1.6 aerogels coincides with the simulated X-ray diffraction images, indicating the in-situ formation of COFs in the aerogel during the hydrothermal process (Fig. 3i and Fig. S6b in Supporting information). Fourier transform infrared spectroscopy (FT-IR) shows that the strong characteristic peaks at 1227 cm^-1 (C-N) and 1572 cm^-1 (C=C) for CSR-S-3, which are attributed to the formation of β-ketamine linkage framework structure. In addition, the absorption bands near 993 cm^-1, 1018 cm^-1 and 1433 cm^-1 are attributed to the stretching of O=S=O (Fig. 3j) [35]. The peaks of CSR-N-1.6 at 1571 cm^-1 (C=C), 1225 cm^-1 (C-N) and 1432 cm^-1 (N=O) also prove the successful synthesis of COFs (Fig. S6c in Supporting information). The FTIR spectrum of COFs powder and CSR aerogels show similar peaks, further validating the successful synthesis of COFs (Fig. 3j and Fig. S6c). Fig. 3k shows the stress-strain curves of CSR-S-3 aerogel under 50% and 70% deformation, and it can be clearly seen that after releasing the pressure, the aerogel is able to quickly recover to near-original morphology, indicating that the aerogel has good mechanical properties.

  Transmission electron microscopy (TEM) images show a rough surface with many wrinkles (Figs. 3l and m, Fig. S8 in Supporting information). Moreover, no isolated COFs crystal rods or clusters are observed in the CSR aerogels, suggesting that 2D COFs sheets grow in-situ along the surface of 2D reduced graphene oxide nanosheets. To further investigate the coverage of the in-situ growth of COFs, atomic force microscopy (AFM) measurements are performed on CSR aerogels. For the CSR aerogels, the average thickness of the COF-S loaded graphene is about 4 nm in CSR-S-3 (Fig. 3n) and about 3.6 nm in CSR-N-1.6 (Fig. S9 in Supporting information), which exhibit an increase in thickness compared to the SA/RGO aerogel (2 nm) (Fig. S10 in Supporting information). This increase in thickness can be attributed to the uniform growth of several layers of COFs on the graphene oxide surface. These results further demonstrate that the 2D COFs are grown in-situ along the surface of the 2D reduced graphene oxide nanosheets. Besides, the CSR-N-1.6 aerogel has a density of only 5 mg/cm³, which is the lowest among the known reports COFs aerogels (Fig. 3o, Fig. S6a and Table S3 in Supporting information).

  X-ray photoelectron spectroscopy (XPS) shows a complete view of the surface chemistry of COF, SR, and CSR (Fig. S12 in Supporting information). Compared with SR, the XPS measurement spectrum of COF and CSR clearly show four visible peaks of C, N, O, and S. Fig. S13 (Supporting information) shows C 1s XPS spectra of SR, the sharp peak at 283 eV corresponds to the major sp² carbon atom, revealing a well-preserved conjugated structure. Meanwhile, two small peaks at 284.8 eV and 287.1 eV are attributed to the remaining incompletely oxidized C—O and C=O bonds. As shown in Fig. S14 (Supporting information), the N 1s of COF-S and CSR-S-3 shows the -NH- and -NH₂⁺- peaks centered at 398.8 and 400.9 eV, and S 2p shows the peaks of -SO₃H and -SO₃- centered at 166.8 and 168.2 eV, respectively. As shown in Fig. S15 (Supporting information), the N 1s of COF-N and CSR-N-1.6 also shows two similar peaks spiked -NH- and -NO₂. The high-resolution XPS spectrum of CSR have the same type of N and S species as the COFs powder, further validating the growth of COFs on graphene. In addition, the COFs loading in the CSR aerogels is calculated by elemental analysis (Tables S4 and S5 in Supporting information). Calculated in terms of S, the COFs loading of CSR-S-3 is 63.83%. In terms of N, the COFs loading of CSR-N-1.6 is 47.27%. Thermogravimetric analysis (TGA) shows the thermal stability of CSR in a nitrogen environment, and the weight loss of COF-S below 120 ℃ is attributed to water evaporation. Besides, the thermal stability of both CSR aerogels is intermediate between those of COFs and SR, with weight retention in the range of 40%−70% after heating to 800 ℃ (Figs. S16 and S17 in Supporting information). These results further confirm the formation of ultrathin COFs on graphene templates.

  N₂ adsorption-desorption experiments are performed at 77 K to determine the specific surface areas of SR, COFs and CSR. The BET specific surface areas of COF-S and COF-N are 81 m²/g and 230 m²/g (Figs. S18d and S18e in Supporting information), respectively. The lower BET of COF-S can be attributed to the reduced crystallinity and the presence of -SO₃H groups in the channels. The specific surface areas of CSR-S-3 and CSR-N-1.6 aerogels are 56 m²/g and 55 m²/g (Figs. S18b and S18c in Supporting information), respectively. The lower BET of CSR-N-1.6 aerogel is more obvious, probably due to partial pore blockage and weaker interactions between nanosheets. The specific surface area of the substrate SR aerogel is only 27 m²/g due to the strong π-π stacking between graphene sheets (Fig. S18a in Supporting information). As expected, the specific surface area of the CSR is intermediate between the measured values of pure COFs and SR. The introduction of COFs greatly increases the surface area of the SR aerogel, and the larger surface area contributes the adsorption capacity.

  Mechanical stability is an important factor in aerogels applications because aerogels may be subjected to continuous stress and deformation. During the synthesis process, sodium alginate is added to the graphene oxide solution in order to increase the elasticity of the composites. The SA/RGO (SR) aerogels obtained by adding different amounts of sodium alginate are prepared separately. With the increase of SA addition, the solution became more and more clarified and the volume of the aerogel decreases slightly, indicating that the addition of SA helps to bind the RGO together.

  The mechanical properties of SR aerogels with different SA additions are analyzed by compression tests, and the stress-strain curves at loading and unloading show that the elastic properties of the aerogels increase with the increase of SA content. When the additions of SA are 0 and 5 mg, the aerogels could completely rebound after 50% compression, while when the additions of SA are elevated again, the aerogels would fracture after compression and the volume is also significantly reduced (Fig. S19 in Supporting information). Finally, 5 mg of SA is chosen to be added to the aerogels. After the addition of 5 mg SA, the maximum pressure that the SR aerogel could withstand increased stress from 6.8 kPa to 13.4 kPa when the aerogel is undergoing 50% strain. The results show that the addition of SA could improve its mechanical properties (Fig. S19). Meanwhile, the CSR aerogels could fully rebound without fracture after 50% and 70% compression (Figs. S20-S23 in Supporting information). As shown in Movie S1 (Supporting information), the two CSR aerogels instantly rebound after being subjected to a 100 g weight. Moreover, they also fully rebound after manual compression, and this excellent elasticity is due to the 3D network structure formed during the hydrothermal process.

  In order to evaluate the adsorption capacity of aerogels for pollutants in water, adsorption experiments are carried out. In the adsorption experiment, methylene blue (MB), rhodamine B (RhB) and methylene orange (MO) dyes are selected as adsorbates (Fig. S24 in Supporting information), of which the first two are cationic organic dyes and the latter is an anionic organic dye [49]. The initial dye concentration is 20 mg/L for three organic dyes. As shown in Figs. 4a-c, the CSR-N-1.6 aerogels adsorb 90%, 97.75%, and 84.13% of MB, RhB, and MO within 5 min, indicating that the CSR-N-1.6 aerogel has good adsorption capacity for all three organic dyes. In addition, the CSR-N-1.6 aerogel reaches adsorption equilibrium within 10 and 3 min after contacting with MB and RhB organic dyes, while the equilibrium times of COF-N powder for MB and RhB are 30 min (Figs. 4d and e). Furthermore, CSR-N-1.6 also shows good adsorption performance for anionic organic dyes MO, which could be attributed to H-bonding and π-π interactions [26]. CSR-N-1.6 could reach 87.13% removal for MO dye within 5 min, while COF-N powder only reached 62.57% removal at 120 min (Fig. 4f). As shown in Fig. S25 (Supporting information), the equilibrium times of CSR-S-3 aerogel for three organic dyes (MB, RhB, MO) are 3 min, 1 min, and 5 min, respectively, and the final removal rates could reach > 99%. Surprisingly, CSR-S-3 could also reach 97.53% removal rate for RhB within 1 min, which is comparable to that of COF-S powder for 5 min. The above phenomena indicates that the CSR aerogels can greatly reduce the time required for equilibration.

  Figure 4

  

  Figure 4.  UV–vis absorption spectra of CSR-N-1.6 in the presence of aqueous solutions of (a) MB, (b) RhB and (c) MO. Removal efficiency of CSR-N-1.6, COF-N and SR for (d) MB, (e) RhB and (f) MO (inset images from left to right: dye stock solution, solution after SR adsorption, solution after COF-N adsorption and solution after CSR-N-1.6 adsorption). PSO kinetic plots of CSR-N-1.6, COF-N and SR for (g) MB, (h) RhB and (i) MO solutions. (j) Zeta potential values of SR, COF and CSR. (k) Continuous flow device prepared by CSR-S-3 aerogels for MB filtration. (l) UV–vis absorption spectra before and after flowing through of MB solution.

  DownLoad: Full-Size Img PowerPoint

  For practical applications, the short time for equilibrium adsorption is an important parameter for evaluating adsorbents [49]. The adsorption kinetics are studied at different time intervals at an initial dye concentration of 10 mg/L. As shown in Figs. S26-S28 (Supporting information), among the five materials (SR, CSR-S-3, CSR-N-1.6, COF-S and COF-N), CSR aerogels show higher adsorbance of dye and faster adsorption rate at the same time. It can be seen that the adsorption kinetics of CSR aerogel are much faster than that of SR aerogel. This may be due to the porous nature of COFs and stronger interaction force with organic dyes.

  The adsorption kinetics of CSR-S-3 aerogel for organic dyes adsorption are calculated using a pseudo second-order (PSO) model. The fitted parameters for the rate equations can be found in Tables S6-S8 (Supporting information). From the tables, it can be seen that the CSR aerogels have greater adsorption capacity and larger values of k₂ (adsorption rate constant) compared to COFs and SR, indicating the stronger interaction between the adsorbent and the adsorbate. The PSO equations fit the experimental data well and the correlation coefficients (R²) are close to unity (Figs. 4g-i), indicating that the adsorption kinetics of organic dyes on the COFs and aerogels are in agreement with the PSO model. Besides, the influence of pH on the stability and performance of material is also tested. As shown in Figs. S2b and S3b (Supporting information), the XRD peak shapes of both COFs remain essentially unchanged after treatment in aqueous solutions at different pH values for 12 h, indicating that the materials maintain good crystallization under harsh conditions. Taking methylene blue (MB) as an example, the removal efficiency of CSR-S and CSR-N under different pH conditions is tested. As shown in Fig. S29 (Supporting information), both aerogels exhibit good adsorption performance on the dye (MB) at different pH levels, indicating that CSR aerogels can adsorb organic dye well in harsh environments.

  Meanwhile, COF-1 without group modification is prepared for comparison, in order to further investigate the adsorption mechanism. As shown in Figs. S30 and S31 (Supporting information), the removal rates of cationic organic dyes MB and RhB by COF-1 during the equilibrium adsorption time is 76% and 90%, respectively. In comparison, the removal rates of sulfonate-modified COF-S and nitro-modified COF-N are 99.8%, 99.6% for MB and 84.5%, 92% for RhB, respectively. Through these results, it is shown that the adsorption of cationic organic dyes is significantly enhanced by the COFs modified with groups. Furthermore, the surface charge properties of SR, COFs and CSR are evaluated using zeta potential measurements (Fig. 4j). The zeta potential of SR is −17.2 mV, which can be attributed to the presence of a small number of oxygen-containing groups on its surface. The COF-S sample has a negative potential of −19.7 mV due to the large amount of electronegative -SO₃H, and its corresponding CSR-S-3 aerogel has a potential of −18.7 mV. The COF-N powder and the CSR-N-1.6 aerogel have the potential of −18.8 mV and −18.1 mV, respectively. These results show that electrostatic interactions between CSR aerogels and cationic organic dyes could improve their adsorption properties.

  As shown in Movie S2 (Supporting information) and Fig. 4k, CSR aerogels are prepared as simple flow-through double-layer filters for continuous and efficient removal of organic dyes due to their structural integrity and good mechanical stability. The anisotropic porous channels in the composite aerogels provide pathway for liquid transport, the COFs provides sufficient adsorption sites for contaminant removal. Taking CSR-S-3 aerogel as an example, the prepared filter removes up to 99% of MB under the action of gravity (Fig. 4l). For continuous removing for organic dyes, a peristaltic pump is connected for continuous transfer. Surprisingly, the filter still has a very high dye removal efficiency of 95% after 2 h of filtration (Fig. S32 in supporting information).

  The improved performance of CSR aerogels compared to COFs powder can be explained as follows. Firstly, since the ultra-low-density composite aerogels have 3D continuous macroporous channels, contaminants can diffuse into the channels rapidly, thus increasing the rapid removal of various organic dyes by the CSR aerogels. Moreover, in CSR composite aerogels, SR and RGO aerogels synergistically provide negative charge on the material surface, and COFs and RGO provide conjugated structure. This facilitates the binding of organic pollutant molecules to the adsorbent through electrostatic attraction, intermolecular H-bonds and π-π interactions. Therefore, CSR composite aerogels are excellent material for ultrafast capture of organic dyes in water remediation.

  Tetracyclines and quinolone antibiotics are the main antibiotics in water bodies, with a combined share of 70% [50]. Tetracycline hydrochloride (TC) and ciprofloxacin hydrochloride (CIP), as common antibiotics, seriously impact human health and environmental quality [51]. In this experiment, CSR aerogels are applied as an adsorbent to remove common TC and CIP organic antibiotic contaminants from contaminated water. Usually, the adsorption of antibiotic contaminants can be divided into two phases, first through a phase of rapid adsorption, followed by a phase of adsorption equilibrium. The adsorption kinetic profiles of CSR-S-3, COF-S and SR are shown in Figs. 5a and d, from which it can be observed that the equilibrium adsorption times of CSR-S-3 for TC and CIP are 60 and 10 min. Besides, those of COF-S and SR are 120 and 240 min for TC and 90 and 240 min for CIP.

  Figure 5

  

  Figure 5.  (a, d) Adsorption kinetics of CSR-S-3, COF-S, and SR for TC and CIP solutions. (b, e) PSO kinetic plots of CSR-S-3, COF-S and SR on TC and CIP solutions. (c, f) Adsorption isotherms of CSR-S-3, COF-S and SR for TC and CIP. (g) Cycling performance of CSR-S-3 aerogel for CIP. (h) XRD images of CSR-S-3 after adsorption of TC and CIP cycles. (i) FT-IR spectrum of CSR-S-3 before and after adsorption of TC and CIP. (j) Fluorescence spectra before and after CSR aerogels adsorption. (k) UV-vis absorption spectra of CIP aqueous solutions in the presence of CSR-S-3 and CSR-N-1.6. (l) CSR-S-3 filter application for CIP.

  DownLoad: Full-Size Img PowerPoint

  In addition, the adsorption kinetics of TC and CIP adsorption are calculated using the PSO model. The fitted parameters for the rate equations can be found in Tables S9 and S10 (Supporting information). The PSO equations fit the experimental data well and the correlation coefficients (R²) are close to unity, indicating that the kinetics of adsorption of TC and CIP on the synthesized COFs and aerogels are in agreement with the PSO model (Figs. 5b and e). As expected, CSR-S-3 aerogel takes less time to reach adsorption equilibrium. This may be attributed to the presence of 3D macroporous adsorption channels in the low-density CSR-S-3 aerogel, which allows antibiotic contaminants to approach the binding site more easily.

  In order to analyze the maximum adsorption capacity of antibiotic contaminants after equilibrium, adsorption isotherms of different adsorbents are fitted using Langmuir model. As shown in Figs. 5c and f, the maximum adsorption capacity of CSR-S-3 is calculated to be about 332.7 mg/g for TC and 405.7 mg/g for CIP. The maximum adsorption amounts of COF-S and SR are 232 mg/g and 37 mg/g for TC, and 367 mg/g and 19.3 mg/g for CIP. It can be seen that the adsorption amounts of CSR aerogels are all higher than those of matrix SR aerogel and COFs powder. Removal rate constants (k₂) are calculated based on the commonly used kinetic PSO model, and it can be seen that the removal rate of CSR-S-3 is significantly higher than that of COFs powder (Tables S9-S11 in Supporting information). From the table, it can be seen that CSR aerogels have greater adsorption capacity and faster adsorption rate compared to COF powder and SR aerogel. As shown in Fig. S33 (Supporting information), the adsorption amounts of CSR-N-1.6 for TC and CIP are 72.7 mg/g and 132.1 mg/g, while the adsorption amounts of COF-N are 58.5 mg/g and 103.5 mg/g. These results indicate that the adsorption performance of CSR-N-1.6 for antibiotics is improved compared with that of SR and COF-N. In conclusion, the adsorption of antibiotics can be significantly improved by making COFs into SR aerogels.

  The recoverability of adsorbents is an important factor in evaluating their practical application. Cyclic stability tests reveal that CSR-S-3 aerogel shows great potential as a candidate material for antibiotic removal in practical application. Under the same experimental conditions, CSR-S-3 aerogel is tested for cyclic adsorption desorption in 20 mg/L of CIP. The result shows that the removal efficiency decreases somewhat after cycling, but still exceeds 75% after 5 cycles (Fig. 5g). Moreover, the XRD patterns of the recycled CSR aerogels are almost identical to the original (Fig. 5h). The CSR adsorbents show better reusability, suggesting the potential application of CSR aerogels for adsorbing antibiotics in environmental remediation.

  In order to explore the adsorption mechanism, the FT-IR spectra of CSR-S-3 aerogel before and after adsorption of TC and CIP are studied. TC and CIP molecules contain a large number of C=C and C-O bonds. As shown in Fig. 5i, upon adsorption of TC and CIP, the C-N bond vibrational bands are red-shifted, probably caused by hydrogen bonding interactions between -OH, -NH₂ of TC (or -COOH of CIP) with -SO₃H, -NH-, -OH of CSR-S-3 during the adsorption process. The emergence of a new vibrational band with a peak at 1080 cm^-1 is caused by C-O stretching vibrations, confirming the adsorption of TC and CIP. This is because COF-S does not contain C-O and the C-O content in SR is minimal. Moreover, CSR aerogels are negatively charged and TC is positively charged, which can be adsorbed by electrostatic interactions.

  On the other hand, CIP itself can emit 430 nm fluorescence peak at the exication of 315 nm [52,53]. Therefore, the adsorption effect of CSR aerogels on CIP can be rapidly detected by fluorescence. When CSR-S and CSR-N are added to 20 mg/L of CIP, about 89.6% and 75% of the total fluorescence intensity of CIP is quenched (Fig. 5j). Moreover, the removal of CIP by CSR-S and CSR-N is also known to be 99.9% and 75% by UV spectrophotometer (Fig. 5k). Furthermore, CIP solution with high concentration of 100 mg/L is also colorless in the naked eye under 365 nm UV lamp irradiation after adding CSR-S-3 to the solution, indicating that the CIP in the solution has been adsorbed by the adsorbent. In addition, the filter device prepared above is also utilized to rapidly detect the removal of CIP by fluorescence. As shown in Fig. 5l, CIP emits its bright blue light under UV light and the filter solution becomes transparent after passing through the filter, which indicates that CIP is effectively adsorbed in the aerogels. The fluorescence quenching of CIP is due to the fluorescent molecules being adsorbed and carried away by the CSR adsorbent, thus greatly diminishing the fluorescence in the adsorbed solution.

  Most polluted water contains a variety of co-existing substances, which may affect the adsorption process. In order to evaluate the versatility of CSR aerogels, the removal efficiency of CSR-S aerogel is determined for four mixed solutions of MB-TC, MB-CIP, RhB-TC and RhB-CIP. As shown in Figs. 6a-d and Fig. S34 (Supporting information), the curves of time-dependent removal efficiency and adsorption kinetic of CSR-S-3 aerogel for each pollutant in the mixed solution reveal its remarkable performance. Specifically, the removal of each pollutant could reach about 90% after 5 min of exposure to the mixed solution. Among them, in the MB-TC and RhB-CIP mixed solution, the adsorption equilibrium time of CSR-S-3 for dyes and antibiotics in the mixed solution is only 5 min. While the MB-CIP mixed solution requires 10 min and RhB-TC requires 30 min. After 60 min of adsorption, the removal of all pollutants could reach > 99.7%. These results confirm that CSR-S-3 has good adsorption versatility and can simultaneously and efficiently remove different organic pollutants in practical water treatment.

  Figure 6

  

  Figure 6.  Adsorption kinetics of CSR-S-3 for (a) MB-TC solution, (b) MB-CIP solution, (c) RhB-TC solution and (d) RhB-CIP solution. (e) Oil contact angles for CSR-S-3. (f, g) Pictures of aerogel before and after dropping oil (lubricant) into CSR-S-3. (h) Absorption efficiency graph of CSR-S-3 for various solvents. (i) Adsorption mechanism of CSR aerogels for organic pollutants. (j) Radar image of properties of CSR aerogels.

  DownLoad: Full-Size Img PowerPoint

  CSR aerogels are expected to effectively adsorb organic pollutants due to their high pore structure, low density and good mechanical stability. As shown in Fig. 6e, CSR-S has very good lipophilicity, and one drop of oil droplet (lubricant) on the aerogel surface is adsorbed all the way in 3 s. The CSR-S-3 aerogel is tested for its ability to absorb organic solvents, using lubricant as an example. As shown in Movie S3 (Supporting information), when lubricant is continuously dripped onto the surface of the aerogel, the aerogel adsorbs it at a very fast rate and can adsorb 150 times its own weight. After adsorption, the aerogel is not deformed and the liquid does not leak out (Figs. 6f and g). CSR-S-3 aerogels are tested for adsorption of various other organic solvents, and the results show that CSR-S-3 aerogels have a strong adsorption capacity of about 40–150 times their own weight (Fig. 6h), depending on the solvent tested. The excellent adsorption capacity is attributed to the low density and high porosity of the aerogel, as well as its amorphous 3D network. Therefore, CSR-S-3 aerogel also has great application prospects for adsorption of organic solvents. Overall, the two COF aerogels show excellent performance individually. Based on the above results, a schematic diagram of the possible mechanism for antibiotic adsorption by CSR is shown in Fig. 6i. The adsorption mechanism can be attributed to electrostatic interactions, hydrogen bonding and π-π interactions. As shown in Fig. 6j, CSR-N-1.6 and CRS-S-3 aerogel has lower density and exhibits better adsorption performance for dyes and antibiotics compared to those of COF powder, showing that CSR aerogels have good comprehensive performance and have great application prospects in adsorption of pollutants.

  In summary, two types of CSR aerogels are prepared by in-situ growth of COF nanosheets on the GO surface, and the effects of different ligands of COFs on the molding and properties of composite aerogels are explored. Sodium alginate improves the mechanical properties of the aerogels, and interconnected graphene nanosheets provide 3D macroporous channels and act as a substrate to support the homogeneous growth of COFs nanosheet. The synthesized CSR aerogels have a 3D multistage porous structure, excellent mechanical properties, and ultra-low densities (approximately 7.5 mg/cm³ for CSR-S-3 and 5 mg/cm³ for CSR-N-1.6). Notably, the density of CSR-N-1.6 is the lowest that has been reported. The 3D CSR aerogels exhibit excellent adsorption properties for organic pollutants (organic dyes and antibiotics) with fast kinetics and good reusability. The adsorption isotherms and kinetics of this composite aerogels are in good agreement with the Langmuir model and the pseudo-second-order model, respectively. In addition, the device of the simple double-layer filter can realize efficient and quick removal of organic pollutants in the water. Furthermore, it is also possible to identify the presence of antibiotics in contaminated water by fluorescence detection, and to visually recognize the presence of antibiotics. This work provides a new perspective on the removal of pollutants in environmental remediation.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Shiyan Ai: Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization. Yaning Xu: Writing – review & editing, Supervision, Formal analysis, Conceptualization. Hui Zhou: Methodology, Formal analysis. Ziwei Cui: Methodology, Formal analysis. Tiantian Wu: Writing – review & editing, Supervision. Dan Tian: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

  Acknowledgment

  The authors acknowledge the financial support provided by the National Natural Science Foundation of China (Nos. 22175094, 21971113).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110761.

  
  
• 1. [1]

     M. Najafi, S. Abednatanzi, P. Gohari Derakhshandeh, et al., Coordin. Chem. Rev. 454 (2022) 214332. doi: 10.1016/j.ccr.2021.214332

  2. [2]

     X. Lei, Q. Tang, Y. Zheng, et al., Nat. Sustain. 6 (2023) 816–826. doi: 10.1038/s41893-023-01101-z

  3. [3]

     N. Yousefi, X. Lu, M. Elimelech, N. Tufenkji, Nat. Nanotechnol. 14 (2019) 107–119. doi: 10.1038/s41565-018-0325-6

  4. [4]

     S. Yu, F. Liu, J. Tian, et al., Chem. Soc. Rev. 13 (2023) 5444.

  5. [5]

     S. Dhaka, R. Kumar, A. Deep, et al., Coordin. Chem. Rev. 380 (2019) 330–352. doi: 10.1016/j.ccr.2018.10.003

  6. [6]

     D. Liu, P. Duan, W. Yang, et al., Chem. Eng. J. 430 (2022) 132817. doi: 10.1016/j.cej.2021.132817

  7. [7]

     P. Duan, D. Liu, W. Yang, et al., Rare. Metals. 41 (2022) 1323–1331. doi: 10.1007/s12598-021-01884-0

  8. [8]

     Z. Zhang, X. Shi, X. Wang, Z. Zhang, Y. Wang, Sep. Purif. Technol. 309 (2023), 123108. doi: 10.1016/j.seppur.2023.123108

  9. [9]

     X. Yaning, W. Tiantian, C. Ziwei, et al., J. Environ. Chem. Eng. 11 (2023) 110337. doi: 10.1016/j.jece.2023.110337

  10. [10]

      Y. Xu, T. Wu, Z. Cui, et al., Sep. Purif. Technol. 303 (2022) 122275. doi: 10.1016/j.seppur.2022.122275

  11. [11]

      Z. Cui, J. Wu, T. Wu, et al., Sep. Purif. Technol. 319 (2023) 124112. doi: 10.1016/j.seppur.2023.124112

  12. [12]

      S. Ahankari, P. Paliwal, A. Subhedar, H. Kargarzadeh, ACS Nano. 15 (2021) 3849–3874. doi: 10.1021/acsnano.0c09678

  13. [13]

      J.A. Martin-Illan, D. Rodríguez-San-Miguel, O. Castillo, G. Beobide, et al., Angew. Chem. Int. Ed. 60 (2021) 13969–13977. doi: 10.1002/anie.202100881

  14. [14]

      D. Zhu, Y. Zhu, Q. Yan, et al., Chem. Mater. 33 (2021) 4216–4224. doi: 10.1021/acs.chemmater.1c01122

  15. [15]

      X. Li, Z. Jia, J. Zhang, et al., Chem. Mater. 34 (2022) 11062–11071. doi: 10.1021/acs.chemmater.2c03117

  16. [16]

      C. Xiao, Y. Yao, X. Guo, et al., Small 20 (2024) 2311881. doi: 10.1002/smll.202311881

  17. [17]

      H. Ji, M. Li, G. Yan, et al., ACS Appl. Mater. Interfaces 30 (2023) 36738–36747. doi: 10.1021/acsami.3c08351

  18. [18]

      F. Li, L. Ding, B. Yao, et al., J. Mater. Chem. A 6 (2018) 11140. doi: 10.1039/c8ta03275k

  19. [19]

      L. Ding, B. Yao, F. Li, et al., J. Mater. Chem. A 7 (2019) 4689. doi: 10.1039/c8ta12046c

  20. [20]

      W. Gong, C. Liu, H. Shi, et al., J. Mater. Chem. C 7 (2022) 4689.

  21. [21]

      K. Liu, J. Yang, J. Liu, et al., Chem. Eng. J. 472 (2023) 144966. doi: 10.1016/j.cej.2023.144966

  22. [22]

      J. Wang, C. Liu, X. Kan, et al., Green. Chem. 22 (2021) 1150.

  23. [23]

      W. Wu, F. Li, B. Yao, et al., J. Hazard. Mater. 433 (2022) 128831. doi: 10.1016/j.jhazmat.2022.128831

  24. [24]

      X. Luo, C. Wang, G. Huang, et al., Sep. Purif. Technol. 298 (2022) 121616. doi: 10.1016/j.seppur.2022.121616

  25. [25]

      B. Yao, W. Wu, L. Ding, et al., J. Org. Chem. 83 (2021) 3024–3032. doi: 10.1021/acs.joc.0c02423

  26. [26]

      N. An, Z. Guo, J. Xin, et al., J. Mater. Chem. A 9 (2021) 16824. doi: 10.1039/d1ta04313g

  27. [27]

      Y. Jiang, Z. Zhang, D. Chen, et al., Adv. Mater. 34 (2022) 2204250. doi: 10.1002/adma.202204250

  28. [28]

      C. Li, J. Yang, P. Pachfule, et al., Nat. Commun. 11 (2020) 4712. doi: 10.1038/s41467-020-18427-3

  29. [29]

      W. Luo, M. Zou, L. Luo, et al., ACS Appl. Mater. Interfaces 49 (2022) 55098–55108. doi: 10.1021/acsami.2c17041

  30. [30]

      Z. Wang, J. Li, S. Liu, G. Shao, X. Zhao, Nanoscale 14 (2022) 16944. doi: 10.1039/d2nr04378e

  31. [31]

      P. Zhu, S. Li, S. Zhou, et al., Chem. Eng. J. 42 (2020) 127559.

  32. [32]

      C. Li, P. Guggenberger, S. Han, et al., Angew. Chem. Int. Ed. 134 (2022) e202206564. doi: 10.1002/ange.202206564

  33. [33]

      Z. Zhang, X. Shi, X. Wang, et al., Sep. Purif. Technol. 309 (2023) 123108. doi: 10.1016/j.seppur.2023.123108

  34. [34]

      X. Li, Z. Jia, J. Zhang, et al., Chem. Mater. 34 (2022) 11062–11071. doi: 10.1021/acs.chemmater.2c03117

  35. [35]

      C. Li, S. Cao, J. Lutzki, et al., J. Am. Chem. Soc. 144 (2022) 3083–3090. doi: 10.1021/jacs.1c11689

  36. [36]

      J. Sun, A. Klechikov, C. Moise, et al., Angew. Chem. Int. Ed. 57 (2017) 1034.

  37. [37]

      C. Wang, F. Liu, J. Chen, et al., Energy. Storage. Mater. 32 (2020) 448–457. doi: 10.1016/j.ensm.2020.07.001

  38. [38]

      A. Khan, B. Nilam, C. Rukhsar, et al., Tungsten 351 (2022) 391–418. doi: 10.69554/yakk1498

  39. [39]

      H. Guo, A. Zhang, H. Fu, et al., Chem. Eng. J. 453 (2023) 139633. doi: 10.1016/j.cej.2022.139633

  40. [40]

      S. Han, Q. Wu, J. Zhu, et al., Chem. Eng. J. 471 (2023) 144511. doi: 10.1016/j.cej.2023.144511

  41. [41]

      Z. Qin, X. Chen, Y. Lv, et al., Chem. Eng. J. 451 (2022) 138586.

  42. [42]

      H. Ma, L. Yu, Z. Li, et al., Small 52 (2023) 2304877.

  43. [43]

      J. Xu, G. Wang, L. Zhu, et al., Chem. Eng. J. 455 (2022) 140704.

  44. [44]

      S. Bi, L. Hou, Y. Lu, Chem. Eng. J. 406 (2020) 126778.

  45. [45]

      X. Min, Z. Wu, T. Wei, et al., ACS Energy Lett. 8 (2023) 3147–3153. doi: 10.1021/acsenergylett.3c00682

  46. [46]

      S. Karak, S. Kandambeth, B.P. Biswal, et al., J. Am. Chem. Soc. 135 (2017) 1856–1862. doi: 10.1021/jacs.6b08815

  47. [47]

      X. Zhong, Q. Ling, P. Kuang, B. Hu, Chem. Eng. J. 467 (2023) 143415. doi: 10.1016/j.cej.2023.143415

  48. [48]

      H. Fan, J. Gu, H. Meng, et al., Angew. Chem. Int. Ed. 57 (2018) 3837–4102. doi: 10.1002/anie.201802719

  49. [49]

      M. Peydayesh, M.K. Suter, S. Bolisetty, et al., Adv. Mater. 32 (2020) 1907932. doi: 10.1002/adma.201907932

  50. [50]

      M.A. Cook, G.D. Wright, Sci. Transl. Med. 14 (2022) 657.

  51. [51]

      S.S. Rameshwar, B. Sivaprakash, N. Rajamohan, et al., Environ. Chem. Lett. 21 (2023) 2995–3022. doi: 10.1007/s10311-023-01623-0

  52. [52]

      X. Zheng, M. wang, S. Zhang, et al., Water. Res. 260 (2024) 121941. doi: 10.1016/j.watres.2024.121941

  53. [53]

      C. Li, F. Wang, X. Xu, et al., Chem. Eng. J. 455 (2023) 140841. doi: 10.1016/j.cej.2022.140841

• 

  Figure 1  (a) Synthesis schematic for the preparation of CSR aerogels. (b) COFs monomers and catalysts used in this work. (c) Structures and images for four different COF gels.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Synthesis procedure of CSR via the hydrothermal method. (b) Hydrogen bonding interactions between sodium alginate and GO.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  SEM and the corresponding elemental mapping images of COF-S (a, b) and COF-N (c, d). (e) A photograph of an ultralight CSR-S-3 aerogel standing on a setaria viridis. (f-h) SEM images of CSR-S-3 and the corresponding EDS plots. (i) XRD images of CSR-S-3 and COF-S. (j) FT-IR spectrum of SR, COF-S and CSR-S-3. (k) Stress-strain curves of CSR-S-3 aerogel at different strains (inset: images of COF aerogel before and after being compressed). (l, m) TEM images of CSR-S-3. (n) AFM image and the corresponding height profiles of CSR-S-3. (o) Density comparative plots of the COF aerogels.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  UV–vis absorption spectra of CSR-N-1.6 in the presence of aqueous solutions of (a) MB, (b) RhB and (c) MO. Removal efficiency of CSR-N-1.6, COF-N and SR for (d) MB, (e) RhB and (f) MO (inset images from left to right: dye stock solution, solution after SR adsorption, solution after COF-N adsorption and solution after CSR-N-1.6 adsorption). PSO kinetic plots of CSR-N-1.6, COF-N and SR for (g) MB, (h) RhB and (i) MO solutions. (j) Zeta potential values of SR, COF and CSR. (k) Continuous flow device prepared by CSR-S-3 aerogels for MB filtration. (l) UV–vis absorption spectra before and after flowing through of MB solution.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a, d) Adsorption kinetics of CSR-S-3, COF-S, and SR for TC and CIP solutions. (b, e) PSO kinetic plots of CSR-S-3, COF-S and SR on TC and CIP solutions. (c, f) Adsorption isotherms of CSR-S-3, COF-S and SR for TC and CIP. (g) Cycling performance of CSR-S-3 aerogel for CIP. (h) XRD images of CSR-S-3 after adsorption of TC and CIP cycles. (i) FT-IR spectrum of CSR-S-3 before and after adsorption of TC and CIP. (j) Fluorescence spectra before and after CSR aerogels adsorption. (k) UV-vis absorption spectra of CIP aqueous solutions in the presence of CSR-S-3 and CSR-N-1.6. (l) CSR-S-3 filter application for CIP.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Adsorption kinetics of CSR-S-3 for (a) MB-TC solution, (b) MB-CIP solution, (c) RhB-TC solution and (d) RhB-CIP solution. (e) Oil contact angles for CSR-S-3. (f, g) Pictures of aerogel before and after dropping oil (lubricant) into CSR-S-3. (h) Absorption efficiency graph of CSR-S-3 for various solvents. (i) Adsorption mechanism of CSR aerogels for organic pollutants. (j) Radar image of properties of CSR aerogels.

  下载: 全尺寸图片 幻灯片
•