An Anionic Porous Indium-Organic Framework with Nitrogen-Rich Linker for Efficient and Selective Removal of Trace Cationic Dyes

Metal-organic frameworks (MOFs) with porosity and functional adjustability have great potential for the removal of organic dyes in the wastewater. Herein, an anionic porous metal-organic framework (MOFs) [Me2NH2]2In2[(TATAB)4(DMF)4]·(DMF)4(H2O)4 (HDU-1) was synthesized, which is constructed from a [In(OOC)4]− cluster and a nitrogen-rich linker H3TATAB (4,4′,4″-s-triazine-1,3,5-triyltri-p-aminobenzoic acid). The negatively charged [In(OOC)4]− cluster and uncoordinated –COOH on the linker result in one unit cell of HDU-1 having 8 negative sites. The zeta potential of -20.8 mV dispersed in pure water also shows that HDU-1 possesses negatively charged surface potential. The high electronegativity, water stability, and porosity of HDU-1 can facilitate the ion-exchange and Coulombic interaction. As expected, the HDU-1 exhibits high selectivity and removal rates towards trace cationic dyes with suitable size, such as methylene blue (MB) (96%), Brilliant green (BG) (99.3%), and Victoria blue B (VB) (93.6%).


Introduction
As raw materials for papermaking, printing and dyeing, textile, and other industries, organic dyes with certain toxicity are often discharged with industrial wastewater, which not only causes harm to the natural ecological environment, but also seriously threatens human health. In fact, the United Nations has chosen "Clean water and sanitation" as one of the Sustainable Development Goals for 2030. However, most organic dyes are difficult to remove from wastewater because of excellent chemical stability. Furthermore, some dyes can form toxic, mutagenic, and carcinogenic intermediates through hydrolysis and oxidation reactions [1,2]. Therefore, the adsorption and removal of dyes before discharge is particularly important for the purpose of the protection of human health and ecological environments. Many organic dye treatment technologies, such as adsorption, membrane separation, photocatalytic degradation, biocatalytic degradation, microwave irradiation, etc., have been widely used in wastewater treatment [3,4]. Among these technologies, adsorption is taken into account to be one of the most feasible and economical methods to deal with organic dye water pollution. However, traditional adsorbents, such as active carbon, zeolite, and silica, show weak adsorption efficiency and low selective separation toward trace dye molecules, which greatly limits their further practical application [5,6]. Therefore, the exploitation of novel and efficient adsorbents for the removal of trace dyes is urgent and necessary.
As a new class of porous organic-inorganic hybrid material, metal-organic frameworks (MOFs) are constructed from metal ions/clusters and organic linkers through coordination bond . There are currently about 100,000 MOFs in the Cambridge database, far Based on the above design guidance, we designed and synthesized a stable anionic porous MOF [Me 2 NH 2 ] 2 In 2 [(TATAB) 4 (DMF) 4 ]·(DMF) 4 4 ] − cluster and a nitrogen-rich liker H 3 TATAB (4,4 ,4 -s-triazine-1,3,5-triyltri-p-aminobenzoic acid) (Scheme S1), which can selectively capture and separate trace cationic organic dyes. The negatively charged [In(OOC) 4 ] − cluster and uncoordinated -COOH on the linker result in one unit cell of HDU-1 having 8 negative sites. The zeta potential of −20.8 mV dispersed in pure water also shows that HDU-1 possesses negatively charged surface potential. Therefore, the high electronegativity, water stability, and porosity of HDU-1 can facilitate the ion-exchange and Coulombic interaction. As expected, the HDU-1 exhibits high selectivity and removal rates towards trace cationic dyes with suitable size, such as methylene blue (MB), Brilliant green (BG), and Victoria blue B (VB) under the synergetic promotion of ion-exchange, Conulombic interaction, size-dependent, and π-π stacking interaction between host-guest.

Results and Discussion
Utilizing adenine as a guide agent, the HDU-1 was synthesized by solvent-thermal reaction choosing nitrogen-rich tricarboxylic acid ligand H 3 TATAB and InCl 3 in DMF/H 2 O/1,4dioxane for 3 days (Figure 1a). The scanning electron microscopy and optical photograph show that HDU-1 is a light yellow spindle crystal ( Figure S1). The single-crystal X-ray diffraction (SCXRD) analyses revealed that HDU-1 crystallizes in the monoclinic space group C2/c, with a = 16.8151 (12)  Due to the flexibility of the -NH-between the benzene and triazine ring, the three benzene rings and triazine of the H 3 TATAB are not oriented in the same plane in the framework. The linker displays irregular shape and only two carboxyl groups for each ligand are coordinated by [In(COO) 4 ] (Figure 1c). Thus, the symmetry of the whole framework reduces from tetragonal to monoclinic. Each [In(COO) 4 ] cluster is connected by four linkers, and each linker is linked by two isolated [In(COO) 4 ] clusters, to construct a 3D porous structure with uncoordinated carboxyl groups exposed to the channels (Figure 1d). There are a one-dimensional irregular channels along the c axis of approximately 3 × 6 Å 2 , taking into account the van der Waals radii. Noteworthily, the negatively charged behavior of the framework comes from a large amount of uncoordinated -COOH exposed to the channels and the negatively charged cluster [In(COO) 4 ] − . The counteraction [Me 2 NH 2 ] + ions from decomposition of DMF solvent located within the channels are used to balance the negative charge of the skeleton. The zeta potential of −20.8 mV dispersed in pure water also shows that HDU-1 possesses negatively charged surface potential ( Figure 2a). Moreover, the high density hydrogen bonds exist in channels which come from H atom on -NH-with O on DMF from channels, O on -COOH, and N on triazine ring, respectively ( Figure S2).
The thermal stability for HDU-1 was proved by the thermogravimetric analysis with a heating rate 5 • C/min under N 2 atmosphere. As shown in Figure 2b, the TGA curve shows that HDU-1 gradually loses solvent molecules of DMF and H 2 O before 225 • C, and is eventually thermally stable up to about 300 • C under N 2 atmosphere. Finally, the metal oxides and amorphous carbon were formed. Elemental analysis for HDU-1 was measured and shows: Found (wt%): C, 51. 27 4 by the chemical formula from crystal structure and elemental analysis (EA). The solvent content from molecular formula is consistent with the results of loss solvent molecules for TGA. As represented in Figure 2c, the PXRD peaks of as-synthesized HDU-1 are consistent with the simulated one obtained from single crystal data, indicating the pure phase of the obtained sample. HDU-1 was soaking in water  Figure 2c and Figure S3). It is found that the crystals could maintain the stability in various environments, which may be attributed to the large number of hydrogen bonds and the equilibrium ions in the framework.     The thermal stability for HDU-1 was proved by the thermogravimetric analysis with a heating rate 5 °C/min under N2 atmosphere. As shown in Figure 2b, the TGA curve shows that HDU-1 gradually loses solvent molecules of DMF and H2O before 225 °C, and is eventually thermally stable up to about 300 °C under N2 atmosphere. Finally, the metal oxides and amorphous carbon were formed. Elemental analysis for HDU-1 was measured and shows: Found  (Figures 2c and S3). It is found that the crystals could maintain the stability in various environments, which may be attributed to the large number of hydrogen bonds and the equilibrium ions in the framework.
HDU-1 with the high-density electronegative sites and open/modified channels has a potential to remove and selectively separate the contaminant by ion-exchange and Coulombic interaction. We selected the different charged and configuration trace dyes (5 ppm) to evaluate the removal property of HDU-1, and the sizes and shapes of the dyes mole- HDU-1 with the high-density electronegative sites and open/modified channels has a potential to remove and selectively separate the contaminant by ion-exchange and Coulombic interaction. We selected the different charged and configuration trace dyes (5 ppm) to evaluate the removal property of HDU-1, and the sizes and shapes of the dyes molecules as shown in Scheme S2. The methyl orange (MO − ) and congo red (CR − ) exhibits -1 charge and methylene blue (MB + ), rhodamine B (RhB + ), rhodamine 6G (R6G + ), Brilliant green (BG + ), and Victoria blue B (VB + ) exhibits +1 charge, respectively. In addition, the sizes of five kinds of cationic dyes are totally different. A small amount of HDU-1 of 10 mg was immersed in a 15 mL various dyes aqueous solution. Then, the removal performance for organic dyes from aqueous solution was measured via UV-visible adsorption spectra. The results reveal that the anionic framework has no adsorption capacity at all for cationic dyes, such as MO − and CR − (Figure 3). There was no change in the color of the methyl orange solution (Figure 4). Among these five cationic organic dyes, HDU-1 can efficiently remove the MB + , BG + , and VB + completely ( Figure 5), and the removal processes can be observed by the color therewith disappeared of the solutions (Figure 4). While for the cationic dyes RhB + and R6G + , obvious absorption peaks are still observed after 24 h, indicating that HDU-1 can only adsorb a fraction of RhB + and R6G + (Figures 4 and 6).
In order to further evaluate the adsorption rates of HDU-1 for trace cationic organic dyes, the concentration changes of RhB + , R6G + MB + , BG + , and VB + in aqueous solution with time induced by HDU-1 are showed in Figure 4d. The removal rate of HDU-1 can reach 82.3% after 25 min, and the solution is nearly clarified after 180 min for VB + . While the MB + and BG + display just about 54.0% and 56.3%, removal efficiency after 25 min and 20 min, respectively. In addition, the removal efficiency of HDU-1 for MB and BG molecules is also comparable to those research in reported MOFs NOTT-210 [38], InOF [36], ZJU-24-0.89 [39], Cu 2 (L)(H 2 O) 2 [40], and JOU-11 [20]. These results indicated that HDU-1 has different adsorption rate VB + > BG + > MB + > RhB + > R6G + for the cationic organic dyes with different size and shape. The faster removal efficiency may be attributed to the stronger π-π interaction of the naphthalene ring on the Victoria blue B (VB + ) with the linker of HDU-1. Although the peak intensity of PXRD patterns ( Figure S4) of HDU-1 after dye-adsorption decreased, the structure of the crystals did not collapse during adsorption process. This is also proved by the fact that the adsorption peak of 1600 cm −1 for C=O stretching vibration of linker belonging to carboxyl group in the FT-IR spectroscopy does not change before and after dye-adsorption. The adsorption ability of HDU-1 for these trace cationic organic dyes may be attributed to the following aspects: (a) HDU-1 is anionic framework with high electronegative and surface potential can effective trap trace cationic organic dyes because of the Coulombic interaction and ion-exchange. (b) The open/modified channels with suitable size give opportunity to trap trace cationic organic dyes. (c) The π-π stacking interaction of aromatic rings between organic dye molecules and organic linker is conducive to adsorption. Furthermore, the pseudo-first order and pseudo-second order adsorption kinetics of MB + , BG + , VB + , R6G + , and RhB + on HDU-1 are analyzed via the time-dependent adsorption capacity. As showed in Figure S6, the fitting results display that the kinetics data are better matched with the pseudo-second-order kinetic model. R PEER REVIEW 6 of 12 mg was immersed in a 15 mL various dyes aqueous solution. Then, the removal performance for organic dyes from aqueous solution was measured via UV-visible adsorption spectra. The results reveal that the anionic framework has no adsorption capacity at all for cationic dyes, such as MO − and CR - (Figure 3). There was no change in the color of the methyl orange solution (Figure 4). Among these five cationic organic dyes, HDU-1 can efficiently remove the MB + , BG + , and VB + completely ( Figure 5), and the removal processes can be observed by the color therewith disappeared of the solutions (Figure 4). While for the cationic dyes RhB + and R6G + , obvious absorption peaks are still observed after 24 h, indicating that HDU-1 can only adsorb a fraction of RhB + and R6G + (Figures 4 and 6).    Molecules 2023, 28, x FOR PEER REVIEW 6 of 12 mg was immersed in a 15 mL various dyes aqueous solution. Then, the removal performance for organic dyes from aqueous solution was measured via UV-visible adsorption spectra. The results reveal that the anionic framework has no adsorption capacity at all for cationic dyes, such as MO − and CR - (Figure 3). There was no change in the color of the methyl orange solution (Figure 4). Among these five cationic organic dyes, HDU-1 can efficiently remove the MB + , BG + , and VB + completely ( Figure 5), and the removal processes can be observed by the color therewith disappeared of the solutions (Figure 4). While for the cationic dyes RhB + and R6G + , obvious absorption peaks are still observed after 24 h, indicating that HDU-1 can only adsorb a fraction of RhB + and R6G + (Figures 4 and 6).     In order to further evaluate the adsorption rates of HDU-1 for trace cationic organic dyes, the concentration changes of RhB + , R6G + MB + , BG + , and VB + in aqueous solution with time induced by HDU-1 are showed in Figure 4d. The removal rate of HDU-1 can reach 82.3% after 25 min, and the solution is nearly clarified after 180 min for VB + . While the MB + and BG + display just about 54.0% and 56.3%, removal efficiency after 25 min and 20 In order to further evaluate the adsorption rates of HDU-1 for trace cationic organic dyes, the concentration changes of RhB + , R6G + MB + , BG + , and VB + in aqueous solution with time induced by HDU-1 are showed in Figure 4d. The removal rate of HDU-1 can reach 82.3% after 25 min, and the solution is nearly clarified after 180 min for VB + . While the MB + and BG + display just about 54.0% and 56.3%, removal efficiency after 25 min and 20 On the basis of the difference in dye adsorption capacity described above, the HDU-1 is capable of selectively separating dye molecules by two modes. The fresh prepared crystals were soaked in a mixed solution of two kind of dyes (1:1 in mole ratio). First, charge-dependent selective adsorption was investigated in a mixture of MO − /MB + and MO − /BG + . As shown in Figure 7a, the trace cationic organic dyes were almost completely removed by HDU-1 (adsorption rate of 96.3% after 7 h for BG + and 94.0% after 24 h for MB + ), while the peak intensity of UV-visible spectra for MO − after 7 h had no significant change. At the same time, the color of the mixed solution changed (Figure 8) also confirmed the selective adsorption of cationic organic dyes based on charge-dependent. The peak intensity of UV-visible spectra of MO − still gone down after 24 h for the mixed MB + /MO − solution, possibly due to the adsorption effect of adsorbed MB + in channels on MO − with small molecule size as time extends. Second, HDU-1 can selectivly remove the cationic organic dyes with different sizes and shape based on size-dependent mode. As shown in Figure 7, MB + and BG + were completely adsorbed while RhB + were hardly removed because the molecular sizes of linear MB + of 1.57 × 0.78 × 0.40 nm 3 and subtriangular BG + of 1.81 × 1.33 × 0.63 nm 3 are smaller than that of RhB + of 1.85 × 1.34 × 0.87 nm 3 . The color of the mixed MB + /RhB + and BG + /RhB + solution changed from purple to dark pink (Figure 8), demonstrating the selective adsorption of size-dependent. In a word, HDU-1 can be taken into account as a candidate adsorption porous material for the selective removal of cationic organic dye from wastewater. dependent adsorption capacity. As showed in Figure S6, the fitting results display that the kinetics data are better matched with the pseudo-second-order kinetic model.
On the basis of the difference in dye adsorption capacity described above, the HDU-1 is capable of selectively separating dye molecules by two modes. The fresh prepared crystals were soaked in a mixed solution of two kind of dyes (1:1 in mole ratio). First, charge-dependent selective adsorption was investigated in a mixture of MO − /MB + and MO − /BG + . As shown in Figure 7a, the trace cationic organic dyes were almost completely removed by HDU-1 (adsorption rate of 96.3% after 7 h for BG + and 94.0% after 24 h for MB + ), while the peak intensity of UV-visible spectra for MOafter 7 h had no significant change. At the same time, the color of the mixed solution changed (Figure 8) also confirmed the selective adsorption of cationic organic dyes based on charge-dependent. The peak intensity of UV-visible spectra of MO − still gone down after 24 h for the mixed MB + /MO − solution, possibly due to the adsorption effect of adsorbed MB + in channels on MO − with small molecule size as time extends. Second, HDU-1 can selectivly remove the cationic organic dyes with different sizes and shape based on size-dependent mode. As shown in Figure 7, MB + and BG + were completely adsorbed while RhB + were hardly removed because the molecular sizes of linear MB + of 1.57 × 0.78 × 0.40 nm 3 and subtriangular BG + of 1.81 × 1.33 × 0.63 nm 3 are smaller than that of RhB + of 1.85 × 1.34 × 0.87 nm 3 . The color of the mixed MB + /RhB + and BG + /RhB + solution changed from purple to dark pink (Figure 8), demonstrating the selective adsorption of size-dependent. In a word, HDU-1 can be taken into account as a candidate adsorption porous material for the selective removal of cationic organic dye from wastewater.  The reusability plays an important role for the practical application of M als. In order to study the regeneration and reuse of HDU-1, we immersed the ing VB + molecules in ethanol with a small amount of HCl (10 mM) for 24 h a perature to desorb VB + dyes. As shown in Figure S7, it was observed the efficiency still achieves over 99% after three cycles. The regeneration per HDU-1 is further proved by the images of HDU-1 loading with dyes before a tion (inset of Figure S7a), which confirms that the HDU-1 can be an ideal re rous material.
Thermogravimetric analyses (TGA) were carried out on a Netzsch TGA a heating rate of 5 °C/min in N2 atmosphere. Elemental analyses for C, H, carried out on a Flash EA1112 micro elemental analyzer. Powder X-ray diffrac patterns were recorded in the 2θ range from 5 to 50° on a Shimadzu XRD 700 eter with Cu Kα (λ = 1.542 Å) radiation at room temperature. Infrared spectr collected in the wavenumber range from 4000 to 500 cm −1 on Thermo Fisher spectrometer using KBr pallets. The zeta potential was carried out on Malve Nano ZS90. The UV−vis spectra were performed in the wavelength range fro nm on a Shimadzu UV-3600 spectrophotometer. The reusability plays an important role for the practical application of MOFs materials. In order to study the regeneration and reuse of HDU-1, we immersed the HDU-1 loading VB + molecules in ethanol with a small amount of HCl (10 mM) for 24 h at room temperature to desorb VB + dyes. As shown in Figure S7, it was observed the regeneration efficiency still achieves over 99% after three cycles. The regeneration performance of HDU-1 is further proved by the images of HDU-1 loading with dyes before and after elution (inset of Figure S7a), which confirms that the HDU-1 can be an ideal renewable porous material.

X-ray-Crystallography
The single crystal measurement of HDU-1 was executed on Bruker APEX-II diffractometer coupled to CCD detector with graphite-monochromatic Mo Kα radiation (λ = 0.71073 Å) and Atlas detector at 273 K. The unit cell parameters and data were determined and collected directly by CrysAlisPro program. The structure of HDU-1 was solved by the direct methods and refined by the full matrix least square method of SHELX program package. All non-hydrogen atoms were directly located according to Fourier diffraction points and refined by anisotropy. H atoms on the ligand were added by theoretical model. The disordered solvent molecules and ions in the crystal channel were computed by PLATON software/SQUEEZE subroutine CCDC: 2159854.

Dye Adsorption Measurements
Seven different organic dyes with different electric charge, kinetic diameter, and molecular structure were selected to explore the adsorption property and adsorption mechanism of HDU-1. In order to study adsorption kinetic, HDU-1 was dried overnight under vacuum at 25 • C and kept in a desiccator. Then, the fresh prepared crystals of 10 mg were weighed precisely. Then, the adsorbents were added into the solution (15 mL

Conclusions
In summary, an anionic functional metal-organic frameworks HDU-1 has been designed and constructed from a nitrogen-rich tricarboxylic acid ligand and In(III) ion. The HDU-1 possesses high-density electronegative sites and uncoordinated-COOH on the linker. The zeta potential of −20.8 mV dispersed in pure water also shows that HDU-1 possesses negatively charged surface potential. The negative charge nature and porosity of HDU-1 can facilitate the ion-exchange and Coulombic interaction. Our results exhibit that HDU-1 can effectively remove the trace cationic organic dyes of MB + , VB + , and BG + from aqueous solution based on charge-dependent, size-dependent, and π-π stacking interaction between host-guest. The removal rate of HDU-1 can reach 82.3% after 25 min, and the solution is nearly clarified after 180 min for VB + . At the same time, MB + and BG + can be separated by HDU-1 from the mixed solution of MO − /BG + , MO − /MB + , BG + /RhB + , MB + /RhB + . Hence, HDU-1 can be taken into account as a candidate adsorption porous material for the selective removal of cationic organic dye from wastewater.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/molecules28134980/s1, Scheme S1: Chemical structure of H 3 TATAB; Scheme S2. Molecular structures of seven organic dyes; Figure S1: (a) SEM image of HDU-1 and (b) optical photograph of HDU-1; Figure S2: The hydrogen bonds which come from H atom on -NH-with O on DMF from channels (a), O on -COOH (b) and N on triazine ring (c), respectively; Figure S3. PXRD patterns of stability test of HDU-1 in pH solution for different time and PBS solution for 24; Figure S4. PXRD comparison of HDU-1 before and after adsorption of dyes; Figure S5. FT-IR comparison of HDU-1 before and after adsorption of dyes; Figure S6. The pseudo-first order and pseudo-second order adsorption kinetics of MB + , BG + , VB + , R6G + and RhB + on HDU-1 via the time-dependent adsorption capacity; Figure S7. (a) The UV-vis spectra of VB absorption/desorption of HDU-1 (inset: image of VB solution after adsorption/desorption for four times), (b) Regeneration performance of HDU-1 for VB adsorption; Table S1. Molecular size of the dyes.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.