Chinese Chemical Letters  2019, Vol. 30 Issue (1): 234-238   PDF    
An anionic In(III)-based metal-organic framework with Lewis basic sites for the selective adsorption and separation of organic cationic dyes
Qi Yang, Bin Wang, Ya Chen, Yabo Xie*, Jianrong Li    
Beijing Key Laboratory for Green Catalysis and Separation and Department of Chemistry and Chemical Engineering, College of Environmental and Energy Engineering, Beijing University of Technology, Beijing 100124, China
Abstract: In this paper, a new anionic metal-organic framework, [In(PBPTTBA)] [(CH3)2NH2] (BUT-29) has been synthesized through the reaction of tetratopic acid ligand with double Lewis pyridine sites, 4,4',4",4'"-(4,4'-(1,4-phenylene)bis(pyridine-6,4,2-triyl)) tetrabenzoic acid (H4PBPTTBA) and In(NO3)2·5H2O and fully characterized by single-crystal X-ray diffraction (SXRD), powder X-ray diffraction (PXRD), thermogravimetric analysis (TGA), infrared spectroscopy (IR), and elemental analysis (EA). BUT-29 can be used as an efficient adsorbent for the selective removal of organic cationic dyes in N, N'-dimethylformamide (DMF) solution. The adsorption capacities of BUT-29 toward methylene blue and crystal violet at 298 K can reach 1119 mg/g and 832 mg/g, respectively. Moreover, the adsorbed dyes can be released in the DMF solution of LiNO3 gradually.
Keywords: Anionic metal-organic framework     Lewis basic sites     Selective adsorption     Separation     Organic dye    

Dyes are widely used in many industrial manufacturing processes, such as textiles, plastics, paper, printing, and so on. However, the industrial wastewater containing dyes molecules are usually discharged into water directly without purification, leading to serious water pollution. In addition, most dyes molecules are immune toward light and oxidation so that they are difficult to degrade in nature [1-3]. Therefore, the removal of dyes from water is essential. Till now, the removal of organic dyes from water is mainly based on traditional methods such as photocatalysis degradation [4-6] and membrane separation [7]. Compared with the above-mentioned methods, the adsorption technology based on physical absorption is more promising because of its low energy-cost and user-friendly control. As a new class of porous materials, metal-organic frameworks (MOFs), formed by organic ligands and inorganic nodes through coordination bond have been intensively investigated for many applications such as gas storage/separation [8-10], solid separation [11], ion exchange [12], catalysis [13-18] and sensing [19, 20], due to their advantages of ultrahigh porosity, high surface area, and tunable functionalities [21-23]. Specifically, as many published works showed that MOFs have shown excellent performances in the removal of dyes [24-28].

Ionic MOFs (I-MOFs) is a unique subclass of MOFs and consists of charged frameworks and extra-framework counter ions [29, 30]. Strong electrostatic interactions between the charged framework and guest molecules are beneficial to increase the adsorption capacity and improve the efficiency of separation processes. Since organic dyes are divided into electrically positive, negative, and neutral charged ones, I-MOFs have been considered as promising candidates for selective adsorption and separation of organic dyes [30, 31].

In this work, a tetratopic acid ligand with double Lewis pyridine sites, 4,4',4",4'"-(4,4'-(1,4-phenylene)bis(pyridine-6,4,2-triyl)) tetrabenzoic acid (H4PBPTTBA) has been designed and synthesized. The assembly of H4PBPTTBA with In(NO3)2·5H2O in DMF yielded a new anionic MOF, [In(PBPTTBA)][(CH3)2NH2] (BUT-29, where BUT = Beijing University of Technology). BUT-29 can rapidly and selectively adsorb organic cationic dyes including acriflavine hydrochloride (AH), acridine red (AR), safranine O (SO), methylene blue (MB), crystal violet(CV) and rhodamine 6G (R6G) through an ion-exchange process. Organic anionic and neutral dyes such as congo red (CR), orange G, naphthol yellow S (NYS), orange II, acid fuchsin (AF), methyl orange (MO), methyl yellow (MY), tartrazine, and isatin cannot be adsorbed by BUT-29. Furthermore, adsorption isotherms measured at 298 K demonstrate that the maximum adsorption amounts of MB and CV in BUT-29 are 1119 and 832 mg/ g, respectively, and BUT-29 could be fully reused after washing with DMF solution of LiNO3 several times. These results indicate that BUT-29 is promising adsorbents for efficient capture of MB and CV from waste water.

Solvothermal reaction of H4PBPTTBA and In(NO3)2·5H2O in the presence of HBF4 as competing reagents in DMF at 120 ℃ for 48 h yielded rhombic shaped crystals of BUT-29. Its phase purity has been characterized by PXRD. As shown in Fig. S3 (Supporting information), the experimental PXRD pattern match well with the one simulated from the single-crystal data, indicating that BUT-29 is in the pure phase. The TGA plot of the prepared sample of BUT-29 is showed in Fig. S1 in Supporting information, confirming that BUT-29 is stable up to ca. 370 ℃. In the FT-IR spectra of BUT-29, slight blue shifts of carbonyl group characteristic bands compared with corresponding ligands were observed, illustrating the metal coordination of carboxylate groups in these ligands (Fig. S2 in Supporting information).

Single-crystal X-ray diffraction reveals BUT-29 crystallizes in an orthorhombic chiral space group of C2/c. In the structure of BUT-29, the InIII ion adopts a tetrahedral linkage geometry through coordinating to eight O atoms from four carboxylate groups of four different PBPTTBA4- ligands (Fig. 1a). The In-O distances are in the range of 2.2617(1) Å to 2.2905(1) Å, being comparable to those of reported indium complexes [32, 33]. As shown in Fig. 1c, the carboxylate groups in PBPTTBA4- ligand exhibit chelating coordination mode, with each one coordinates to one InIII ion. Thus, one InIII connect to four PBPTTBA4- ligands and one PBPTTBA4- ligand connect to four InIII atoms to form a 3D framework with one dimensional (1D) rhombic channel, the edge length of which is 16.8 Å (atom to atom distance). Furthermore, two such frameworks are mutually interpenetrated (Fig. 1d). After removal of free solvent molecules, the total solvent-accessible volume of BUT-29 is estimated to be 71.3% by PLATON. From the topological viewpoint, each In(COO)4 node and PBPTTBA4- ligand can be viewed as 4-connected nodes, the 3D structure of BUT-29 can thus be simplified as a 4,4-c net with the point symbol of (42 × 84), corresponding to a Pts-type topology. It should be noted that in the framework of BUT-29, every InIII atom connect with four carboxylic groups, thus, in its pore, there should exist countercations to balance the charge of the framework. However, due to the disordered nature, these counter-cations could not be identified through single-crystal X-ray diffraction analysis. We speculate that these counter-cations should be (CH3)2NH2+, which is a common by-product in the solvothermal reaction when DMF was used as solvent. EA analysis was then carried out, which showed extra N content inside the pore, which implied the presence of (CH3)2NH2+. Furthermore, in the TGA curve of the assynthesized BUT-29, there exist a 7.1% weight loss between 200 ℃ and 340 ℃, which is well corresponding to the presence of one (CH3)2NH2+ ion per formula (Fig. S1).

Fig. 1. (a) View of the [In(CO2)4] node; (b) View of the single open framework in BUT-29 along c axis, showing large rhombic channels (c) coordination mode of the PBPTTBA4- ligand; (d) View of the 3D framework of BUT-29 along c axis, showing two-fold interpenetration of the network

Based on the large rhombic channels and cationic framework of BUT-29, we sought to explore its application in the removal of organic dyes. Fifteen dyes including AH, AR, SO, MB, CV, R6G, CR, orange G, NYS, orange II, AF, MO, MY, Tartrazine, and Isatin were checked. Freshly prepared BUT-29 was well activated and then immersed in DMF solutions of these dyes at room temperature. The adsorption abilities of BUT-29 toward these dyes were determined by UV–vis spectroscopy. As shown in Figs. 2ad and Figs. S4 (Supporting information), nearly all the cationic dyes can be removed within 15 min, while the anionic and neutral dyes could not be absorbed. To future demonstrate the selective adsorption of BUT-29 toward cationic dyes, the MOF samples were soaked in the binary mixtures of MB/MO, MB/RB with the concentration ratio is 50/50, respectively. The results were as expected: only the MB was absorbed by the BUT-29 in the solution mixture and the solutions exhibited the color of MO or RB at last and the color of the BUT-29 changed from yellow to blue (inset photograph of Figs. 2e and f). Moreover, compared to other absorbents using a few days, the absorption rate of cationic dyes by BUT-29 was really fast, and nearly 100% of dye molecules can be absorbed within 10 min. The high adsorption rate is attributed to the large size of the cage windows (12.4 ×7.1 Å) and negative charged framework.

Fig. 2. UV–vis spectra changes of DMF solutions of (a) MB, (b) MO, (c) CV, and (d) AF; selective adsorption of BUT-29 toward MB in the binary mixtures of (e) MB/MO and (f) MB/RB with the concentration ratio is 50/50; inset shows the digital photograph of the color change of BUT-29 before and after organic dye adsorption

In order to verify whether the absorption process toward cationic dyes is caused by ionic interactions between the cationic dye and framework, dye-releasing experiments were performed both in pure DMF and a saturated DMF solution of LiNO3, respectively. The release process was monitored through UV–vis spectroscopy. The results demonstrate that the cationic dye molecules in MB@BUT-29 and CV@BUT-29 can be gradually released in the presence of LiNO3, whereas they are hardly released without LiNO3 (Fig. 3). These results suggest that the absorption can be assigned to the ionic interaction of the dyes with the anionic framework. In addition, the absorbed dyes can be released within 20 min Figs. 3c and d).

Fig. 3. UV–vis spectra monitored the process of dye release of MB@BUT-29 (a) and CV@BUT-29 (b) in a saturated DMF solution of LiNO3; comparison of dye release of MB@BUT-29 (c) and CV@BUT-29 (d) in pure DMF and in a saturated DMF solution of LiNO3

In the above experiments, it was showed that MB and CV can be completely adsorbed within 6 min and 4 min in BUT-29, respectively Figs. 2a and c). Thus, we further explored the adsorption isotherms of MB and CV in BUT-29 at 298 K. As shown in Fig. 4a, the maximum adsorption amounts of MB and CV in BUT-29 are 1119 and 832 mg/g, respectively. These values are comparable and even higher than those in other porous materials reported so far (Table S2 in Supporting information). In addition, we further measured the of FT-IR spectra of MB loaded BUT-29 and found the appearance of an extra characteristic peak in the wavenumber of 1323 cm-1, which should be the stretching vibration peak of C-N bond of MB molecules, indicating that MB molecules are indeed adsorbed into the pores of BUT-29 (Fig. S5 in Supporting information). The superior performances of BUT-29 in MB and CV adsorptions could be ascribed to their large specific surface areas, suitable pore size, as well as the ionic interactions between the cationic dye and anionic framework structure of BUT-29. Moreover, Langmuir [34] and Freundlich [35] models were used to fit and examine above adsorption isotherms respectively. The related parameters are given in Figs. S6 and S7 in Supporting information. Obviously, the data are well fitted by the Langmuir model, indicating a homogeneous and monolayer adsorption occurring in BUT-29 with a finite number of identical sites.

Fig. 4. (a) Adsorption isotherms of BUT-29 toward MB and CV (adsorption conditions: at 298 K, 50 mL of solution, 15 mg of MOFs, contact time of 4 h); (b) Cyclic application of BUT-29 in adsorbing CV (adsorption conditions: 298 K, 50 mL solution, 15 mg MOFs, and 4 h adsorption time)

In order to reduce the cost, reusability of an adsorbent is important for its practical applications. To explore the reusability of BUT-29, the MOF sample after CV adsorption was regenerated by dispersing it in saturated DMF solution of LiNO3, and then, the regenerated MOF again used to absorb CV. As shown in Fig. 4b, after 4 repeated cycles, BUT-29 almost regained its initial adsorption capacities toward CV, demonstrating its high stability and good reusability. In addition, as confirmed from PXRD pattern, BUT-29 retained crystallinity after cyclic tests (Fig. S3).

In summary, a new In(III)-based MOF, BUT-29, has been designed, synthesized, and used in the selective adsorption of organic cationic dyes in DMF solution. BUT-29 has a 3D framework with one dimensional (1D) rhombic channel, which is occupied by (CH3)2NH22+ counter ions. BUT-29 can be used as an excellent candidate for the selective adsorption of organic cationic dyes and its maximum adsorption amounts toward MB and CV are 1119 mg/g and 832 mg/g, respectively, comparable or higher than those in other porous materials reported so far. The adsorption isotherms of BUT-29 toward MB and CV can be fitted by the Langmuir model, indicating a homogeneous and monolayer adsorption process. In addition, BUT-29 can be fully reused after washing with DMF solution of LiNO3 several times.


This work was financially supported by the National Natural Science Foundation of China (NSFC, No. U1407119).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at

P. Waranusantigul, P. Pokethitiyook, M. Kruatrachue, E.S. Upatham, Environ. Pollut. 125 (2003) 385-392. DOI:10.1016/S0269-7491(03)00107-6
S.K. Kansal, A. Kumari, Chem. Rev. 114 (2014) 4993-5010. DOI:10.1021/cr400093w
V.K. Garg, R. Gupta, A.B. Yadav, R. Kumar, Bioresour. Technol. 89 (2003) 121-124. DOI:10.1016/S0960-8524(03)00058-0
P. Mahata, G. Madras, S. Natarajan, J. Phys. Chem. B 110 (2006) 13759-13768.
F. Wang, F.L. Li, M.M. Xu, et al., J. Mater Chem. A 3 (2015) 5098-5916.
M. Dai, H.X. Li, J.P. Lang, CrystEngComm 17 (2015) 4741-4753. DOI:10.1039/C5CE00619H
I. Koyuncu, Desalination 143 (2002) 243-253. DOI:10.1016/S0011-9164(02)00263-1
Z.R. Herm, E.D. Bloch, J.R. Long, Chem. Mater. 26 (2014) 323-338. DOI:10.1021/cm402897c
M.P. Suh, H.J. Park, T.K. Prasad, D.W. Lim, Chem. Rev. 112 (2012) 782-835. DOI:10.1021/cr200274s
M. Pera-Titus, Chem. Rev. 114 (2014) 1413-1492. DOI:10.1021/cr400237k
D. Liu, J.P. Lang, B.F. Abrahams, J. Am. Chem. Soc 133 (2011) 11042-11045. DOI:10.1021/ja203053y
C.K. Brozek, M. Dinca, Chem. Soc. Rev. 43 (2014) 5456-5467. DOI:10.1039/C4CS00002A
J.C. Jiang, O.M. Yaghi, Chem. Rev. 115 (2015) 6966-6997. DOI:10.1021/acs.chemrev.5b00221
M.Y. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 112 (2012) 1196-1231. DOI:10.1021/cr2003147
D. Liu, Z.G. Ren, H.X. Li, et al., Angew. Chem. Int. Ed. 49 (2010) 4767-4770. DOI:10.1002/anie.201001551
D.S. Deng, H. Guo, G.H. Kang, CrystEngComm 17 (2015) 1871-1880. DOI:10.1039/C4CE02204A
X.W. Wang, H. Guo, M.J. Liu, X.Y. Wang, D.S. Deng, Chin. Chem. Lett. 25 (2014) 243-246. DOI:10.1016/j.cclet.2013.11.045
D.S. Deng, H. Guo, B.M. Ji, New J. Chem. 41 (2017) 12611-12616. DOI:10.1039/C7NJ02021J
L.E. Kreno, K. Leong, O.K. Farha, et al., Chem. Rev. 112 (2012) 1105-1125. DOI:10.1021/cr200324t
D.S. Deng, L.L. Liu, B.M. Ji, G.J. Yin, C.X. Du, Cryst. Growth Des. 12 (2012) 5338-5542. DOI:10.1021/cg300900m
H. Furukawa, K.E. Cordova, M. O'Keeffe, O.M. Yaghi, Science 341 (2013) 1230444. DOI:10.1126/science.1230444
W. Lu, Z. Wei, Z.Y. Gu, et al., Chem. Soc. Rev. 43 (2014) 5561-5593. DOI:10.1039/C4CS00003J
M. O'Keeffe, O.M. Yaghi, Chem. Rev. 112 (2012) 675-702. DOI:10.1021/cr200205j
M.L. Ma, J.H. Qin, C. Ji, et al., J. Mater Chem. C 2 (2014) 1085-1093.
Q. Chen, Q. He, M. Lv, et al., Appl. Surf. Sci. 327 (2015) 77-85. DOI:10.1016/j.apsusc.2014.11.103
Y.X. Shi, W.X. Li, H.H. Chen, et al., Chem. Commun. 53 (2017) 5515-5518. DOI:10.1039/C7CC02704D
D. Liu, F.F. Lang, X. Zhou, et al., Inorg. Chem. 56 (2017) 12542-12550. DOI:10.1021/acs.inorgchem.7b02034
B. Wu, W.H. Zhang, Z.G. Ren, J.P. Lang, Chem. Commun. 51 (2015) 14893-14896. DOI:10.1039/C5CC05990A
Y. Han, J.R. Li, Y.B. Xie, G.S. Guo, Chem. Soc. Rev. 43 (2014) 5952-5981. DOI:10.1039/C4CS00033A
G. Nickerl, A. Notzon, M. Heitbaum, et al., Cryst. Growth Des. 13 (2013) 198-200. DOI:10.1021/cg301347t
B.Q. Song, X.L. Wang, Y.T. Zhang, et al., Chem. Commun. 51 (2015) 9515-9518. DOI:10.1039/C5CC02639C
E.Y. Cho, J.M. Gu, I.H. Choi, et al., Cryst. Growth Des. 14 (2014) 5026-5033. DOI:10.1021/cg5005837
L. Liu, X.N. Zhang, Z.B. Han, et al., J. Mater Chem. A 3 (2015) 14157-14164. DOI:10.1039/C5TA00986C
J. Galán, A. Rodríguez, J.M. Gómez, S.J. Allen, G.M. Walker, Chem. Eng. J. 219 (2013) 62-68. DOI:10.1016/j.cej.2012.12.073
L. Wang, Z. Huang, M. Zhang, B. Chai, Chem. Eng. J. 189 (2012) 168-190.