English

• Water pollution is one of the biggest global problems and affects the safety of animals and human beings around the world. Conventional pollutants such as heavy metals and organic pollutants, as well as bacteria are widely spread in water pollution. Especially, mercury contamination is ubiquitous in water by natural processes, human activities, and industrial manufactures. Unlike organic pollutants, mercury is non-degradable through the ecosystem and can be biomagnified 10⁶–10⁷ times in organisms, thereby posing global environmental and health risks [1]. Once adsorbed into the human body, mercury can severely damage the organs because of its strong affinity to proteins and enzymes [2], thus leading to serious health issues. Meanwhile, contamination by bacteria in water is an another major reason for poor water quality [3]. The presence of pathogenic strains of bacteria can increase the risk of infection and is harmful to the surrounding environment. According to the World Health Organization (WHO) report, billions of people lack access to safe and clean water supply and millions of people die from diseases related to the polluted water each year [4]. Thus, effectively removing mercury and bacteria from wastewater is very important and highly desirable.

  The conventional methods for water treatment often involve the use of adsorbents and disinfectants separately, which are usually laborious and wasteful [5]. It is very attractive to develop multifunctional synthetic adsorbent materials, which can combine the adsorption and antibacterial properties in a single system to enable sustainable environmental applications. Covalent organic frameworks (COFs) are emerging crystalline materials, which show pre-designable functionality, regular pore structure, excellent chemical stability, and large surface area [6,7]. Especially, two-dimensional COFs hold these significant properties as excellent adsorbents for mitigating environmental problems [8,9]. Their ordered π-columnar arrangements not only ensure fully accessible environment for adsorbates but also enable the functional groups close to each other, which is helpful improving the adsorption. Therefore, COFs are highly promising candidates for constructing multifunctional adsorbents.

  In recent years, various types of predesigned COFs with specific functions have been developed and show great application potential in water treatment [10-14]. When used for mercury adsorption, it is found that the easily accessible and specific recognition sites can help to promote the adsorption, and sulfur species are effective groups for the removal of mercury from water [15-21]. Porphyrin and its derivatives have been constructed as functional candidates in the photodynamic therapy (PDT) treatment of photosensitizer, due to their high photo-sensitivity, high molar absorption coefficient, and excellent stability [22-24]. Recent studies show that functional porphyrin can be directly embedded into COF-frameworks for photodynamic inactivation of bacteria [25-29]. Although the existing COFs have been functionalized with specific functional groups for water treatment, few COF materials are capable of removing both heavy metals and pathogenic microbes simultaneously. To date, the exploration of multifunctional COFs for selectively removing Hg²⁺ and effectively killing bacteria in the meantime has been rarely achieved. Therefore, developing a novel COF-platform combining the selective Hg²⁺ adsorption sites with the sunlight-driven sterilization property is particularly compelling for water treatment.

  Herein, we construct a multifunctional COF, namely DmtaTph, for water treatment (Scheme 1). The resultant functionalized COF with superior crystallinity and high porosity, shows high stability and surface area. The introduction of the methylthio units in the DmtaTph offers high-affinity sites for selective and efficient adsorption of Hg²⁺. At the same time, due to the coexistence of photosensitive porphyrin motifs, the obtained COF displays sunlight-driven sterilization property. Further research has been conducted on the adsorption mechanism of DmtaTph for Hg²⁺ and its photodynamic inactivation of bacteria. Moreover, the adsorption and antibacterial performance of DmtaTph in wastewater have also been investigated. Overall, with the properties of highly selective and efficient Hg²⁺ adsorption and remarkable antibacterial capability, DmtaTph is extremely promising for water treatment in an energy and resource-saving way.

  Scheme 1

  

  Scheme 1.  The schematic diagram of the experimental system. (a) Synthetic route of DmtaTph by the condensation of Tph and Dmta. (b) Removal of Hg²⁺ and sunlight-driven sterilization from water using DmtaTph.

  DownLoad: Full-Size Img PowerPoint

  The multifunctional COF DmtaTph was prepared using two precursors 2,5-di(methylthio)terephthalaldehyde (Dmta) and 5,10,15,20-tetrakis(4-aminophenyl)porphyrin (Tph) via solvothermal synthesis. Methylthio and porphyrin groups were reasonably integrated into the COF-platform to construct DmtaTph, which was a brownish-red precipitate. The molecular structure of DmtaTph was verified by Fourier transform infrared (FT-IR) and solid state ¹³C NMR spectroscopy. The FT-IR spectrum showed C=N stretching at 1614 cm⁻¹ (Fig. S1 in Supporting information), corresponding to the characteristic stretching vibration of imine bonds. The C=O stretching vibration of Dmta at 1678 cm⁻¹ and NH₂ vibration band of Tph at 3200−3500 cm⁻¹ were significantly reduced, suggesting that they were almost completely reacted. Solid state ¹³C NMR spectroscopy of DmtaTph revealed an apparent up-field shift of the −SCH₃ resonances in the spectrum, together with a characteristic peak of ca. 160 ppm assigned to the carbon atom of CH═N bond (Fig. S2 in Supporting information). These results indicated the successful synthesis of DmtaTph.

  The crystallinity of DmtaTph was characterized by powder X-ray diffraction (PXRD) analysis with Cu Kα radiation. It exhibited the main characteristic peaks with the 2θ values of 3.5°, 5.0°, and 7.0°, which were assigned to the (010), (100), and (020) facets, respectively (Fig. 1a). To assess the crystalline framework structure of DmtaTph, a possible 2D models with AA eclipsed and AB staggered stacking layers were built using the Material Studio software. As displayed in Fig. 1a, the experimental PXRD pattern of DmtaTph (red) matched well with the simulated pattern of AA model (green), but differed from that of AB model (Fig. S3 in Supporting information), demonstrating the π-eclipsed stacking of the layers in DmtaTph. Furthermore, the AA stacking mode was subjected to Pawley refinement to generate a theoretical PXRD pattern (Fig. 1a, black), which was in excellent agreement with the experimental data (Fig. 1a, red), with only minor variations (Fig. 1a, blue). After a geometrical energy minimization based on the AA model, DmtaTph adopted the P1 space group with following unit cell parameters: a = 26.35 Å, b = 35.61 Å, c = 3.57 Å, α = β = 90°, γ = 135.1° (Fig. 1b and Table S1 in Supporting information). The π-π stacking distance of DmtaTph between two successive layers was found to be 3.57 Å.

  Figure 1

  

  Figure 1.  The characterization of DmtaTph. (a) Comparison between the experimental profile (red) and Pawley refined profile (black), the simulated pattern for eclipsed (AA) stacking mode (green), and the refinement difference (blue). (b) Top view and side view of DmtaTph in AA stacking model (C, dark gray; N, blue; S, yellow). (c) N₂ adsorption isotherm of DmtaTph at 77 K. (d) HR-TEM image of DmtaTph and its magnified view of the selected red square area showing a periodic structure.

  DownLoad: Full-Size Img PowerPoint

  The porosity of DmtaTph was examined through nitrogen adsorption/desorption measurements at 77 K. As shown in Fig. 1c, DmtaTph presented a typical type Ⅰ isotherm curve characterized with a sharp increase at P/P₀ < 0.1, indicating its microporous structure. According to the Brunauer-Emmett-Teller (BET) model, it has the high BET surface area of 434.32 m²/g. Based on the single-point measurement at P/P₀ = 0.95, the pore volume was calculated to be 0.27 cm³/g. Nonlocal density functional theory (NLDFT) was employed to estimate the pore size distributions of DmtaTph, providing the average pore width of 2.2 nm (Fig. S4 in Supporting information), which coincided with the simulated crystal structure (Fig. 1b). The high specific surface area and regular pore structure are not only sufficient for the access of O₂, but also enable the methylthio units of DmtaTph to have an excellent enrichment capability of Hg²⁺.

  Scanning electron microscopy (SEM) was utilized to visualize the morphology of DmtaTph, which indicated the formation of aggregated vesicles with different sizes (Fig. S5 in Supporting information). The corresponding elemental mapping analysis demonstrated that the C, N, and S elements were uniformly distributed over DmtaTph, indicating its homogeneous structure (Fig. S6 in Supporting information). In addition, the crystalline nature of DmtaTph was further verified by high-resolution transmission electron microscopy (HR-TEM). The results showed that it was highly crystalline with a 2D square lattice net (Fig. 1d). And its magnified view indicated a layer distance of 3.52 Å in DmtaTph, which was in agreement with the simulated structure model (3.57 Å, Fig. 1b). Thermogravimetric analysis (TGA) revealed that there was no obvious weight loss of DmtaTph up to ca. 400 ℃, implying its remarkable thermal stability (Fig. S7 in Supporting information). Additionally, the chemical stability of DmtaTph was also evaluated. When immersed it in various organic solvents (THF, CH₂Cl₂, CH₃CN, acetone, and methanol), aqueous NaOH (0.1 mol/L) and HCl (0.01 mol/L) solutions, we found that the diffraction peaks of DmtaTph in PXRD patterns were remained (Fig. S8 in Supporting information), indicating its excellent stability in various environments.

  The pH can influence the surface charge of the adsorbent, thereby regulating electrostatic interactions with the adsorbate. We initially measured the zeta potential of DmtaTph in the range of pH 2–7 to determine its surface charge properties (Fig. S9 in Supporting information). The point of zero charge of DmtaTph was determined to be 4.4. When pH > 4.4, the surface of DmtaTph becomes negatively charged, which facilitates the adsorption of cations via electrostatic attraction. Subsequently, we investigated the pH effect on the absorption capability of DmtaTph with Hg²⁺ concentration of 500 mg/L for 12 h. As illustrated in Fig. 2a, the adsorption amount of Hg²⁺ (Eq. S1 in Supporting information) remained above 400 mg/g within the pH range of 2–7 due to the high affinity of methylthio units to Hg²⁺. The adsorption capacity is maximum at pH 5. At low pH values (pH < 4.4), the protonation on the surface of the adsorbent caused a positive zeta potential (Fig. S9), increasing electrostatic repulsion with Hg²⁺ and thereby reducing the adsorption capacity of DmtaTph. Conversely, with the increase of pH (pH > 5), the hydrolysis of Hg²⁺ intensified [30], diminishing the availability of Hg²⁺ in aqueous solution and further decreasing the adsorption capacity of DmtaTph. Therefore, in the following study, the optimal adsorption pH was fixed at 5.0 to improve the adsorption efficiency and avoid hydrolysis.

  Figure 2

  

  Figure 2.  The adsorption of Hg²⁺ by DmtaTph. (a) The pH effect on Hg²⁺ adsorption. (b) Hg²⁺ adsorption isotherm with Langmuir model at 288, 298, and 308 K. (c) Adsorption kinetics fitted by pseudo-first-order, pseudo-second-order and Elovich models. (d) The removal efficiencies of DmtaTph towards various anions and cations.

  DownLoad: Full-Size Img PowerPoint

  To reveal the adsorption process between DmtaTph and Hg²⁺, the adsorption isotherms were studied at 288, 298, and 308 K, using Hg²⁺ concentrations ranging from 10 mg/L to 1400 mg/L. As shown in Fig. 2b, the adsorption capacity of DmtaTph increased distinctly upon raising the Hg²⁺ concentrations and incubation temperature. The equilibrium adsorption isotherms were fitted with the Langmuir and Freundlich models, and their equations and fitting parameters were provided in Eqs. S2 and S3, and Table S2 (Supporting information). As shown in Fig. 2b, Figs. S10 and S11 (Supporting information), the adsorption isotherm was well-fitted with the Langmuir model, reflecting monolayer adsorption behavior of the adsorption process. The calculated maximum Hg²⁺ adsorption capacity of DmtaTph at 288, 298, and 308 K were 587.3, 657.9, and 748.6 mg/g, respectively, which are comparable to many other reported COFs (Table S3 in Supporting information). To gain a deep understanding of the adsorption process, thermodynamic parameters were calculated using Eqs. S4 and S5 (Supporting information) [31]. The plots of lnK^ϴ against 1/T were displayed in Fig. S12 (Supporting information), and the detailed calculation were listed in Table S4 (Supporting information). The calculated ΔG^ϴ values were –16.63 kJ/mol at 288 K, −17.50 kJ/mol at 298 K, and −18.77 kJ/mol at 308 K. The negative values indicated that the adsorption process of Hg²⁺ on DmtaTph was spontaneous. The ΔH^ϴ value was 14.45 kJ/mol, indicating an endothermic nature of the adsorption process, which primarily relied on physical adsorption.

  To obtain the real-time rate of Hg²⁺ decontamination from polluted water, we explored the kinetics of Hg²⁺ uptake using DmtaTph at 298 K. As shown in Fig. 2c, the adsorption capacity increased quickly for the first 1 min, and the Hg²⁺ adsorption increased steadily until reached the plateau at about 60 min, revealing a fast adsorption rate of Hg²⁺ by DmtaTph. The kinetic data were analyzed using several models: Pseudo-first-order, pseudo-second-order, Elovich, and intra-particle diffusion models (Fig. 2c, Fig. S13, Eqs. S6-S9 and Table S5 in Supporting information). The pseudo-second-order kinetic model was the most suitable for describing the adsorption process, suggesting a chemical chelation between the methylthio functional groups and Hg²⁺. The distribution coefficient K_d of DmtaTph for Hg²⁺ was calculated to be 4.683 × 10⁵ mL/g (Eq. S10 in Supporting information), indicating its strong affinity to Hg²⁺ [32]. The intra-particle diffusion model illustrated the Hg²⁺ adsorption process on DmtaTph in three distinct stages (Fig. S13). Initially, Hg²⁺ was rapidly adsorbed onto the exterior surface of DmtaTph. Subsequently, intra-particle diffusion of Hg²⁺ within DmtaTph occurred, followed by further diffusion through the smaller pores of DmtaTph until adsorption equilibrium [33]. The slower second step played a significant role in controlling the adsorption process, demonstrating that the intra-particle diffusion within DmtaTph was the rate-limiting factor in achieving equilibrium.

  The Hg²⁺ removal capability of DmtaTph was assessed in a competitive environment. As shown in Fig. 2d, DmtaTph exhibited very high removal efficiency for Hg²⁺, but relatively low removal efficiency for other common anions and cations, including NO₃⁻, Cl⁻, Br⁻, Na⁺, Mg²⁺, K⁺, Ca²⁺, Mn²⁺, Fe³⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Cd²⁺, and Pb²⁺. Compared with the multifunctional material APCOF-1 reported previously [29], DmtaTph possessed high substrate selectivity and adsorption capacity for Hg²⁺ removal. The high selectivity of DmtaTph for Hg²⁺ could be expounded by Pearson's soft-hard acid-base theory between the sulfur and Hg²⁺ [34].

  Besides high adsorption capacity and remarkable selectivity, DmtaTph shows excellent performance in recyclability. By using excess Na₂S, DmtaTph could be easily desorbed. As demonstrated in Fig. S14 (Supporting information), the adsorption efficiency for Hg²⁺ retained above 84% after five consecutive adsorption-desorption cycles, indicating the stable nature of DmtaTph. After 5 cycles, the regenerated DmtaTph was further characterized by FT-IR and PXRD. The unchanged FT-IR spectra revealed that it did not show significant negative effect on the DmtaTph structure (Fig. S15 in Supporting information). The PXRD pattern indicated that the crystallinity was reduced but still retained after the adsorption of Hg²⁺ (Fig. S16 in Supporting information).

  We further studied the adsorption mechanism of DmtaTph for Hg²⁺. The SEM element mapping images were used to monitor the changes of element contents before and after adsorption. As displayed in Fig. S17 (Supporting information), a large amount of Hg element was observed after soaking in HgCl₂ aqueous solution for 12 h. The distribution of Hg elements was consistent with that of S elements in DmtaTph. The above results verified the high affinity between chelating thioether sites and Hg²⁺. To gain further investigation of the Hg²⁺ adsorption by DmtaTph, we performed X-ray photoelectron spectroscopy (XPS) measurements. In the XPS spectrum of DmtaTph (Fig. 3a), five distinct peaks were found ascribed to the characteristic peaks of S 2p, S 2s, C 1s, N 1s, O 1s, respectively, where the oxygen peak may be caused by exposure to some surface oxygen in the air [35]. After the adsorption of Hg²⁺, new peaks of Hg 4f and Hg 4d appeared in the XPS spectrum of DmtaTph−Hg²⁺, revealing the successful adsorption of Hg²⁺ on DmtaTph. In the Hg 4f spectrum (Fig. 3b), the Hg 4f_7/2 and Hg 4f_5/2 peaks of DmtaTph−Hg²⁺ changed 0.31 and 0.34 compared with those of HgCl₂, confirming the adsorption of Hg²⁺ on DmtaTph. In S 2p spectrum of DmtaTph (Fig. 3c), only two peaks at 164.29 and 163.10 eV were observed before adsorption, which were assigned to S 2p_1/2 and S 2p_3/2, respectively [36]. After Hg²⁺ adsorption, a new peak at 166.36 eV emerged due to the complexes formed between S and Hg [37]. Additionally, there were 0.45 and 0.43 eV shifts for S 2p_1/2 and S 2p_3/2 peaks, respectively, attributed to the interaction between S and Hg²⁺. However, N 1s XPS spectra of DmtaTph and DmtaTph−Hg²⁺ showed that the position of three fitted peaks was almost unchanged (Fig. 3d), indicating no involvement of N in adsorption. These XPS results confirmed that the efficient Hg²⁺ adsorption was primarily due to the strong coordination between the soft heavy Hg²⁺ and the soft methylthio groups in DmtaTph.

  Figure 3

  

  Figure 3.  The Hg²⁺ adsorption mechanism of DmtaTph. (a) XPS survey spectra of DmtaTph before and after Hg²⁺ adsorption. (b) Hg 4f spectra of HgCl₂ and DmtaTph−Hg²⁺. (c) S 2p spectra of DmtaTph before and after Hg²⁺ adsorption. (d) N 1s spectra of DmtaTph before and after Hg²⁺ adsorption.

  DownLoad: Full-Size Img PowerPoint

  Photodynamic therapy (PDT) has been recognized as an effective antibacterial method due to its efficiency, broad-spectrum, and overcoming drug resistance [38]. It mainly relies on the cytotoxic singlet oxygen (¹O₂), which destroys target pathogens by the photosensitizer in the presence of oxygen with light irradiation [39,40]. Porphyrin-based species present excellent photochemical properties and act as classical and ideal photosensitizers in PDT due to their low dark toxicity and high extinction coefficient [25-28]. Thus, porphyrin-derived DmtaTph is a very attractive candidate as a photosensitizer to demonstrate the antibacterial activity.

  The photodynamic property of DmtaTph with and without light irradiation was studied. Using 9,10-anthracenediyl-bis(methylene)dimalonic acid (ABDA) as the ¹O₂ indicator, the ¹O₂ generation behavior was determined via measuring the ABDA absorbance (λ = 398 nm) over time. The pure ABDA solution in water showed slight decrease in absorbance after 30 min irradiation by visible light (~50 mW/cm², λ > 400 nm) or natural sunlight (~50 mW/cm²) (Figs. S18−S20 in Supporting information). In contrast, the ABDA absorbance in the presence of DmtaTph significantly decreased by 98.7% and 99.6% after 30 min irradiation with visible light and natural sunlight, respectively (Figs. S20−S22 in Supporting information), proving the ¹O₂ generation ability and the antibacterial capacity of DmtaTph under light irradiation.

  The antibacterial ability of DmtaTph was evaluated with Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) as the representative Gram-negative and Gram-positive bacteria. As exhibited in Fig. 4a, no obvious bacteria death was observed without DmtaTph, and most of the Mueller Hinton (MH) agar plates were covered with bacterial colonies. Although bacterial colonies in the control groups under light irradiation alone were slightly reduced (Figs. 4b and c), numerous colonies still grew in the absence of DmtaTph. When DmtaTph was irradiated with visible light (~50 mW/cm², λ > 400 nm), 98.9% of E. coli and 99.4% of S. aureus were killed, and almost no colony formed on the MH agar plates, implying that the porphyrin-derived photodynamic behavior played a key role in the antibacterial activity. In addition, the time course of visible light irradiation was tested at each set time interval. The number of CFU of bacteria continuously reduced upon raising the irradiation time (Figs. S23−S25 in Supporting information), and no bacterial colonies were observed on the MH agar plates at 1.5 h. Thus, the irradiation time of light was kept at 1.5 h. Additionally, the antibacterial activity of DmtaTph in the presence of natural sunlight was also examined (Fig. 4). The data indicated that 98.9% of E. coli and 99.6% of S. aureus were killed under sunlight irradiation (~50 mW/cm²) for 1.5 h, confirming its excellent antibacterial behavior in natural light. Additionally, the recyclability of DmtaTph was also studied after the photodynamic inactivation of bacteria under sunlight irradiation. After each cycle, DmtaTph was separated by centrifugation, washed with water and ethanol, dried under vacuum, and then reused for the next antibacterial cycle. Notably, the antibacterial effects against E. coli (≥93%) and S. aureus (≥92%) were still significant after five cycles (Fig. S26 in Supporting information). Moreover, the PXRD patterns of reused DmtaTph maintained the main peaks (Fig. S27 in Supporting information), indicating that DmtaTph was stable during the sterilization process.

  Figure 4

  

  Figure 4.  The antibacterial performance of DmtaTph. (a) Images of E. coli and S. aureus treated with or without DmtaTph under different conditions. (b) E. coli viability under different conditions. (c) S. aureus viability under different conditions.

  DownLoad: Full-Size Img PowerPoint

  To obtain deeper insights into the mechanism of photodynamic inactivation by DmtaTph, electron paramagnetic resonance (EPR) spectroscopy was employed with trapping agents of 5,5-dimethyl-1-pyrroline N-oxide (DMPO) and 2,2,6,6-tetramethyl-1-piperidine (TEMP) to determine the type and intensity of the free radicals. As depicted in Fig. 5a, a distinctive three-line signal (1:1:1) corresponding to TEMP-¹O₂ adducts was clearly observed under visible light irradiation, which was absent under dark condition. In contrast, no characteristic four-line signal (1:2:2:1) of DMPO-^•OH was observed under the same conditions. Additionally, a weak four-line signal (1:1:1:1) corresponding to DMPO-^•O₂^- was detected. Based on the above analysis, we concluded that the primary active species involved in this process was ¹O₂. Furthermore, the inactivation of bacteria by the ¹O₂ produced from DmtaTph was further confirmed by fluorescence microscopy. The probe 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA), which can be oxidized into fluorescent dichlorofluoresce (DCF) by ROS, was applied to evaluate the generation of ¹O₂ in bacteria. As displayed in Fig. 5b, negligible fluorescence was seen when E. coli and S. aureus were irradiated with light in the absence of DmtaTph and kept in dark in the presence of DmtaTph, respectively. In contrast, after treatment with DmtaTph under visible light (~50 mW/cm², λ > 400 nm), strong and obvious green fluorescence was observed in both S. aureus and E. coli, implying that massive ¹O₂ was produced in bacteria.

  Figure 5

  

  Figure 5.  The photodynamic inactivation mechanism of DmtaTph to bacteria. (a) EPR spectra using TEMP and DMPO as spin trap agents under dark and visible light. (b) Fluorescence images of ¹O₂ generation in E. coli and S. aureus treated with or without DmtaTph under dark and visible light. (c) Illustration of the possible antibacterial mechanism of DmtaTph.

  DownLoad: Full-Size Img PowerPoint

  Overall, the whole bacterial eradication process is illustrated in Fig. 5c. DmtaTph initially adsorbs on the surface of bacteria. When DmtaTph is irradiated with light, electrons in the ground state (S₀) of DmtaTph are excited and transferred to the excited singlet state (S₁) of DmtaTph. Subsequently, the photogenerated electrons in S₁ can flow to the triplet state (T₁) by intersystem crossing (ISC), or return to S₀ by internal conversion (IC) or fluorescence emission (FE) [39,40]. Then, DmtaTph in T₁ further combines with ³O₂ to yield ¹O₂, which quickly spreads to the surrounding adherent bacteria. The porous structure of DmtaTph further facilitates the diffusion of ³O₂ and ¹O₂ through the pores. As a consequence, E. coli and S. aureus are effectively killed via the photodynamic processes.

  To further assess the potential of DmtaTph in practical applications, we investigated the Hg²⁺ removal and antibacterial performance of DmtaTph in simulating water samples and real wastewater. The results are listed in Table 1. It was found that DmtaTph successfully adsorbed over 98% of initial Hg²⁺ from simulated samples 1–4 after treatment for 2 h. Additionally, DmtaTph exhibited significant antibacterial activities under natural sunlight (~50 mW/cm²), with a killing efficiency exceeding 99.9% (Fig. S28 in Supporting information). These findings demonstrated that DmtaTph had the potential to simultaneously remove both Hg²⁺ and bacteria from wastewater without any interference between two actions. In the case of real wastewater 5, which contained an initial Hg²⁺ concentration of 15.4 µg/L and the total concentration of 166.7 mg/L for other ions (Table S6 in Supporting information), DmtaTph displayed a high removal efficiency for Hg²⁺ (99.4%), with a residual Hg²⁺ concentration of 0.09 µg/L. The results confirmed that DmtaTph could efficiently remove Hg²⁺ from wastewater even in the presence of complex ion mixtures.

  Table 1

  

  Table 1.  The removal efficiencies of DmtaTph for Hg²⁺ and bacteria in wastewater (n = 3).

  DownLoad: CSV

  No.	Species	Initial concentration	Residue concentration	Removal efficiency
  1	Hg2+ E. coli	96.2 µg/L 102 CFU/mL	0.63 µg/L 0	99.3% 100%
  2	Hg2+ E. coli	962.2 µg/L 104 CFU/mL	16.47 µg/L 2	98.3% 99.9%
  3	Hg2+ S. aureus	96.2 µg/L 102 CFU/mL	0.75 µg/L 0	99.2% 100%
  4	Hg2+ S. aureus	962.2 µg/L 104 CFU/mL	18.82 µg/L 1	98.0% 99.9%
  5	Hg2+	15.4 µg/L	0.09 µg/L	99.4%

  In this work, both methylthio and porphyrin groups were successfully incorporated into the COF-platform to construct a multifunctional COF, DmtaTph. This material not only displayed selective removal of Hg²⁺ with a high adsorption capacity of 657.9 mg/g at 298 K, but also showed superior antibacterial effects toward E. coli (98.9%) or S. aureus (99.6%) under sunlight irradiation. Specifically, the adsorption of Hg²⁺ and antimicrobial activity were independent and did not influence each other, allowing DmtaTph to effectively achieve dual purification functions in wastewater treatment. Mechanistic studies revealed that the strong coordination between the S species and Hg²⁺ was the main driving force for high Hg²⁺ capture. The inactivation of bacteria by the ¹O₂ was produced from DmtaTph with the assistance of light irradiation. Furthermore, DmtaTph showed high selectivity, stability, and recyclability, and it was successfully applied in removal of Hg²⁺ and bacteria in wastewater. Therefore, it demonstrates great potential in simultaneously reducing the contamination of heavy metal ions and pathogens in wastewater. Our work also provides a unique strategy for construction of adsorbent materials for environmental remediation, and the multi-functionalization of the COF-platform is worthy of further exploration.

  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

  Mei Zhao: Writing – original draft, Data curation. Fengyang Zhao: Investigation, Data curation. Jiantao Ping: Validation, Funding acquisition. Wenli Wu: Software. Lingxi Zhao: Software. Xinyue Luan: Data curation. Li Yu: Validation. Shuhua Liu: Visualization. Yongxian Guo: Writing – review & editing, Funding acquisition. Juyoung Yoon: Supervision. Qiongzheng Hu: Writing – review & editing, Funding acquisition.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (Nos. 22106078, 62105175), High-end Foreign Experts Recruitment Plan (No. G2023024007L), Shandong Provincial Natural Science Foundation (Nos. ZR2022YQ12, ZR2021QB031, ZR2021QF058), and the Science, Education, and Industry Integration Pilot Project for Talent Research at Qilu University of Technology (Shandong Academy of Sciences) (No. 2024RCKY028).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110782.

  
  
• 1. [1]

     M. McNutt, Science 341 (2013) 1430. doi: 10.1126/science.1245924

  2. [2]

     I. Onyido, A.R. Norris, E. Buncel, Chem. Rev. 104 (2004) 5911–5930. doi: 10.1021/cr030443w

  3. [3]

     P.K. Pandey, P.H. Kass, M.L. Soupir, S. Biswas, V.P. Singh, AMB Express 4 (2014) 51. doi: 10.1186/s13568-014-0051-x

  4. [4]

     S. Singh, R. Garg, A. Jana, C. Bathula, S. Naik, M. Mittal, Ceram. Int. 49 (2023) 7308–7321.

  5. [5]

     U. Baig, M. Faizan, M. Sajid, J. Clean. Prod. 302 (2021) 126735. doi: 10.1016/j.jclepro.2021.126735

  6. [6]

     A.P. Côté, A.I. Benin, N.W. Ockwig, et al., Science 310 (2005) 1166–1170. doi: 10.1126/science.1120411

  7. [7]

     K.T. Tan, S. Ghosh, Z. Wang, et al., Nat. Rev. Methods Primer 3 (2023) 1. doi: 10.1038/s43586-022-00181-z

  8. [8]

     J.L. Segura, M.J. Mancheño, F. Zamora, Chem. Soc. Rev. 45 (2016) 5635–5671. doi: 10.1039/C5CS00878F

  9. [9]

     J. Wang, S. Zhuang, Coord. Chem. Rev. 400 (2019) 213046. doi: 10.1016/j.ccr.2019.213046

  10. [10]

      M.B. Asif, S. Kim, T.S. Nguyen, J. Mahmood, C.T. Yavuz, J. Am. Chem. Soc. 146 (2024) 3567–3584. doi: 10.1021/jacs.3c10832

  11. [11]

      X. Zhang, G. Li, D. Wu, et al., Biosens. Bioelectron. 145 (2019) 111699. doi: 10.1016/j.bios.2019.111699

  12. [12]

      Z. Xia, Y. Zhao, S.B. Darling, Adv. Mater. Interfaces 8 (2021) 2001507. doi: 10.1002/admi.202001507

  13. [13]

      G. Li, Y. Wu, C. Zhong, Y. Yang, Z. Lin, Chin. Chem. Lett. 35 (2024) 108904. doi: 10.1016/j.cclet.2023.108904

  14. [14]

      N. Liu, L. Shi, X. Han, et al., Chin. Chem. Lett. 31 (2020) 386–390. doi: 10.1016/j.cclet.2019.06.050

  15. [15]

      H.L. Qian, M.S. Zhu, M.L. Du, X.Q. Ran, X.P. Yan, J. Hazard. Mater. 427 (2022) 128156. doi: 10.1016/j.jhazmat.2021.128156

  16. [16]

      Y. Zhang, H. Li, J. Chang, et al., Small 17 (2021) 2006112. doi: 10.1002/smll.202006112

  17. [17]

      N. Huang, L. Zhai, H. Xu, D. Jiang, J. Am. Chem. Soc. 139 (2017) 2428–2434. doi: 10.1021/jacs.6b12328

  18. [18]

      Q. Sun, B. Aguila, J. Perman, et al., J. Am. Chem. Soc. 139 (2017) 2786–2793. doi: 10.1021/jacs.6b12885

  19. [19]

      S.Y. Ding, M. Dong, Y.W. Wang, et al., J. Am. Chem. Soc. 138 (2016) 3031–3037. doi: 10.1021/jacs.5b10754

  20. [20]

      M. Hussain, N. Maile, K. Tahir, et al., Chem. Eng. J. 446 (2022) 137410. doi: 10.1016/j.cej.2022.137410

  21. [21]

      W. Wang, M. Gong, D. Zhu, et al., Environ. Sci. Ecotechnology 14 (2023) 100236. doi: 10.1016/j.ese.2023.100236

  22. [22]

      M. Ethirajan, Y. Chen, P. Joshi, R.K. Pandey, Chem. Soc. Rev. 40 (2011) 340–362. doi: 10.1039/B915149B

  23. [23]

      J. Tian, B. Huang, M.H. Nawaz, W. Zhang, Coord. Chem. Rev. 420 (2020) 213410. doi: 10.1016/j.ccr.2020.213410

  24. [24]

      K.S. Ndlovu, M.J. Moloto, K.E. Sekhosana, T.T.I. Nkambule, M. Managa, Environ. Sci. Pollut. Res. 30 (2023) 11210–11225.

  25. [25]

      L.G. Ding, S. Wang, B.J. Yao, et al., J. Mater. Chem. A 10 (2022) 3346–3358. doi: 10.1039/d1ta08743f

  26. [26]

      F.L. Meng, H.L. Qian, X.P. Yan, Talanta 233 (2021) 122536. doi: 10.1016/j.talanta.2021.122536

  27. [27]

      J. Hynek, J. Zelenka, J. Rathouský, et al., ACS Appl. Mater. Interfaces 10 (2018) 8527–8535. doi: 10.1021/acsami.7b19835

  28. [28]

      C. Zhang, J. Guo, X. Zou, et al., Adv. Healthcare Mater. 10 (2021) 2100775. doi: 10.1002/adhm.202100775

  29. [29]

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

  30. [30]

      M. Zhao, Y.S. Guo, G.D. Fu, et al., Microchem. J. 171 (2021) 106855. doi: 10.1016/j.microc.2021.106855

  31. [31]

      L. Zhang, Y.H. Li, L.H. Meng, et al., Sep. Purif. Technol. 341 (2024) 126928. doi: 10.1016/j.seppur.2024.126928

  32. [32]

      G.E. Fryxell, Y. Lin, S. Fiskum, et al., Environ. Sci. Technol. 39 (2005) 1324–1331. doi: 10.1021/es049201j

  33. [33]

      Y.H. Li, M.Y. Liu, Y.W. Wei, C.C. Wang, P. Wang, Environ. Sci.: Nano 10 (2023) 672–682. doi: 10.1039/D2EN01035F

  34. [34]

      Y. Liu, Y. Zhang, X. Jiang, et al., Coordin. Chem. Rev. 513 (2024) 215896.

  35. [35]

      Q. Fu, T. Zhang, X. Sun, et al., Chem. Eng. J. 454 (2023) 140154.

  36. [36]

      L. Wang, J. Wang, Y. Wang, F. Zhou, J. Huang, J. Hazard. Mater. 429 (2022) 128303.

  37. [37]

      X. Gao, M. Li, Y. Zhao, Y. Zhang, Chem. Eng. J. 378 (2019) 122096.

  38. [38]

      X. Hu, H. Zhang, Y. Wang, et al., Chem. Eng. J. 450 (2022) 138129.

  39. [39]

      T.X. Luan, L. Du, J.R. Wang, et al., ACS Nano 16 (2022) 21565–21575. doi: 10.1021/acsnano.2c10423

  40. [40]

      A. Schlachter, P. Asselin, P.D. Harvey, ACS Appl. Mater. Interfaces 13 (2021) 26651–26672. doi: 10.1021/acsami.1c05234

• 

  Scheme 1  The schematic diagram of the experimental system. (a) Synthetic route of DmtaTph by the condensation of Tph and Dmta. (b) Removal of Hg²⁺ and sunlight-driven sterilization from water using DmtaTph.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  The characterization of DmtaTph. (a) Comparison between the experimental profile (red) and Pawley refined profile (black), the simulated pattern for eclipsed (AA) stacking mode (green), and the refinement difference (blue). (b) Top view and side view of DmtaTph in AA stacking model (C, dark gray; N, blue; S, yellow). (c) N₂ adsorption isotherm of DmtaTph at 77 K. (d) HR-TEM image of DmtaTph and its magnified view of the selected red square area showing a periodic structure.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  The adsorption of Hg²⁺ by DmtaTph. (a) The pH effect on Hg²⁺ adsorption. (b) Hg²⁺ adsorption isotherm with Langmuir model at 288, 298, and 308 K. (c) Adsorption kinetics fitted by pseudo-first-order, pseudo-second-order and Elovich models. (d) The removal efficiencies of DmtaTph towards various anions and cations.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The Hg²⁺ adsorption mechanism of DmtaTph. (a) XPS survey spectra of DmtaTph before and after Hg²⁺ adsorption. (b) Hg 4f spectra of HgCl₂ and DmtaTph−Hg²⁺. (c) S 2p spectra of DmtaTph before and after Hg²⁺ adsorption. (d) N 1s spectra of DmtaTph before and after Hg²⁺ adsorption.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  The antibacterial performance of DmtaTph. (a) Images of E. coli and S. aureus treated with or without DmtaTph under different conditions. (b) E. coli viability under different conditions. (c) S. aureus viability under different conditions.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  The photodynamic inactivation mechanism of DmtaTph to bacteria. (a) EPR spectra using TEMP and DMPO as spin trap agents under dark and visible light. (b) Fluorescence images of ¹O₂ generation in E. coli and S. aureus treated with or without DmtaTph under dark and visible light. (c) Illustration of the possible antibacterial mechanism of DmtaTph.

  下载: 全尺寸图片 幻灯片

  Table 1.  The removal efficiencies of DmtaTph for Hg²⁺ and bacteria in wastewater (n = 3).

  No.	Species	Initial concentration	Residue concentration	Removal efficiency
  1	Hg2+ E. coli	96.2 µg/L 102 CFU/mL	0.63 µg/L 0	99.3% 100%
  2	Hg2+ E. coli	962.2 µg/L 104 CFU/mL	16.47 µg/L 2	98.3% 99.9%
  3	Hg2+ S. aureus	96.2 µg/L 102 CFU/mL	0.75 µg/L 0	99.2% 100%
  4	Hg2+ S. aureus	962.2 µg/L 104 CFU/mL	18.82 µg/L 1	98.0% 99.9%
  5	Hg2+	15.4 µg/L	0.09 µg/L	99.4%

  下载: 导出CSV
•