A new approach for ultra-high adsorption of cationic methylene blue in a Zr-sulfonic-based metal-organic framework.

A series of Zr-sulfonic-based metal-organic frameworks have been synthesized by the solvothermal method, namely VNU-17 and VNU-23. Particularly, VNU-17 and VNU-23 adopt the sulfonate group (SO3 -) moieties densely packed within their structure, which can efficiently uptake MB+ from wastewater. The maximum adsorption capacity for MB+ onto VNU-23 is up to 1992 mg g-1 at pH = 7, which is more than five times that of activated carbon and possesses the highest value among all the reported MOF materials. In addition, VNU-23 retains the adsorption uptake of MB for at least five cycles. The adsorption isotherms and kinetic studies reveal that MB+ dye adsorption onto VNU-23 fits a Langmuir isotherm and the pseudo second order kinetic model. Furthermore, the ultra-high adsorption capacity of VNU-23 for MB dye can be accounted for by the suitable pore/channel size together with electrostatic attraction and π-π interactions. These results indicate that VNU-23 can be utilized as a promising candidate for removing MB+ from an aqueous medium. This journal is © The Royal Society of Chemistry.

Introduction

With the rapid development of industry and population growth, organic dyes have become crucial contaminants in water, which cause imbalance in ecosystems and affect human health.[1,2] Cationic dyes, such as methylene blue (MB), raise severe environmental concerns due to their various applications in dyeing cotton, wool, silk, and coloring paper.[3] Hence, removing organic dyes from wastewaters is extremely necessary before discharging them into the natural environment. There are many approaches to remove these pollutants, such as catalysis,[4-6] advanced oxidation processes,[7] flocculation,[8] membrane filtration,[9] and electrolysis.[10] Apart from these methods, adsorption has emerged as the most efficient technique to purge hazardous substances from wastewater due to its fleetness, low cost, high yield, insignificant dangerous by-products, and simplicity.[11-16] Different adsorbents have been utilized for the mentioned purpose, such as activated carbon,[17,18] metal oxide,[19] chitosan,[20] and carbon nanotube.[21] Nevertheless, there are several drawbacks to using these adsorbents such as slow adsorption kinetics, tricky generation, thermal and chemical instability, and low specific surface area. Finding a general strategy for designing novel materials to improve the adsorption performance has been attracting much attention.

Metal–organic frameworks (MOFs) are crystalline and porous materials, which adopt permanent porosity, the modular essence of constitution and synthesis. Compared with other adsorbents, MOFs have many dominant features, such as high ordered structure, good thermal and chemical stability, low density, and large specific surface area. Consequently, MOFs can be added to an assorted arrangement of functional groups and guest molecules. It allows MOFs to be extraordinary scaffolds in various utilizations.[22] As expected, MOFs have been employed as efficient adsorption materials for the removal of dyes. Besides, increased stability in water and facile modification of the ligand are the unique properties of MOFs in the adsorption of pollutants. By adding positive and negative charge groups into MOFs, the possibility of removal of contaminants from wastewater has been significantly enhanced through the electrostatic interaction mechanism.[23-27]

However, only a few MOFs display a high removal level of pollutants, leading to poor application in the treatment of practical wastewater.[23,28] Moreover, there is no specific strategy in the design and synthesis of MOFs that can create an ultra-high adsorption capacity for organic dyes. With this in mind, we expect that introducing dense charge carrier moieties such as sulfonate groups (–SO3−) inside the structure of MOFs to enhance the interaction between MB+ cations and SO3− groups via electrostatic attraction can lead to an excellent adsorption capacity for MB+.

With all of this taken into consideration, we developed an approach for synthesizing a series of Zr-sulfonic-based MOFs, constructed from Zr6O8(H2O)8(COO)8 cluster and sulfonic-rich linkers such as 4-sulfonaphthalene-2,6-dicarboxylic acid (H3SNDC) and 4,8-disulfonaphthalene-2,6-dicarboxylic acid (H4SNDC), namely VNU-17 and VNU-23. As in our previous studies,[29,30] VNU-17 and VNU-23 possessed the bcu topology (Fig. 1b, c), highlighted by suitable pore windows for introducing MB via electrostatic attraction (Fig. S1†). As expected, VNU-23 exhibited a maximum adsorption capacity (qm) for MB of about 1992 mg g−1 at pH = 7, with adsorption uptake maintained without any appreciable loss for at least five cycles (Fig. 6). These findings demonstrated the adsorption performance for MB for systematically fine-tuning such materials for use as absorbents in practical conditions.

Fig. 1

The architecture of DUT-52 is constructed from 12-connected Zr6O4(OH)8(NDC)6 to form a structure adopting the fcu topology (a), which is depicted from a CCDC number of 937877.[31] The design of VNU-17, Zr6O8(H2O)8(HSNDC)4, and VNU-23, Zr6O8(H2O)8(H2SNDC)4, is depicted (b and c) with 8-connected bcu topology, which is constructed from the cif information of previously reported VNU-17 and VNU-23 with the CCDC number of 1540864 and 1585550, respectively.[29,30] Atom colors: Zr, blue polyhedra; C, black; O, red; S, yellow. All H atoms are omitted for clarity.

Fig. 6

The reusability of VNU-23 in the adsorption process of MB.

Experimental

Synthesis of 4-sulfonaphthalene-2,6-dicarboxylic acid (H3SNDC) linker

A mixture of 2,6-naphthalenedicarboxylic acid (1 g, 4.63 mmol) and 10 mL of oleum (SO3 in concentrated H2SO4, 10 wt%) was added to a 100 mL glass flask and then stirred at 130 °C for 1 day. The solution was cooled to room temperature and the product was dissolved in 100 mL of distilled water, filtered and precipitated by 50 mL of concentrated HCl (37 wt%). Next, the solid was filtered, washed with 100 mL of concentrated HCl and dried at 80 °C under vacuum for 1 day to give a pure linker at 80% yield. 1H-NMR (DMSO-d6, 500 Hz): δ = 9.51 (s, 1H); 8.52 (s, 1H); 8.48 (s, 1H); 8.10 (d, 1H, J = 0.02 ppm); and 7.99 ppm (d, 1H, J = 0.02 ppm) (Fig. S2†).

Synthesis of 4,8-disulfonaphthalene-2,6-dicarboxylic acid (H4SNDC) linker

A mixture of 2,6-naphthalenedicarboxylic acid (1 g, 4.63 mmol) and 10 mL of fuming sulfuric acid was introduced into a 100 mL glass flask and continuously stirred and refluxed at 150 °C for 1 day. Next, fresh fuming sulfuric acid (6 mL) was added and then stirred and refluxed at 170 °C for 2 days. The reaction mixture was then cooled to room temperature, and the mixture was slowly dropped into 200 mL of distilled water, filtered and precipitated by 100 mL of concentrated HCl (37 wt%). Finally, the solid was filtered, washed with 200 mL of concentrated HCl and dried under vacuum at 80 °C for 1 day to yield a gray-white powder. 1H-NMR (DMSO-d6, 500 Hz): δ = 9.52 (s, 2H); 8.47 (s, 2H) (Fig. S2†).

Synthesis of VNU-17

A mixture containing ZrOCl2·8H2O (0.052 g, 0.162 mmol), H3SNDC linker (0.054 g, 0.183 mmol), formic acid (1.3 mL), and DMF (8 mL) was heated at 120 °C for 2 days. The mixture was then cooled to room temperature. The solid was centrifugated and washed with DMF for 3 days (6 mL per day). The material was then immersed in a MeOH/H2O (v/v = 4/1) solution of H2SO4 0.3 M for 3 days (0.3 M H2SO4, 5 times per day). Next, the material was washed to pH = 5 with an excess amount of methanol/H2O (v/v = 4/1). The solid was exchanged with MeOH for 2 days (5 × 5 mL per day), centrifugated, and activated under vacuum at 80 °C for 1 day to yield a pure acidified VNU-17 (82% yield, based on Zr4+).

Synthesis of VNU-23

ZrOCl2·8H2O (0.0393 g, 0.122 mmol), H4SNDC linker (0.0496 g, 0.132 mmol), formic acid (1.5 mL), and DMF (6 mL) were ultrasonically dissolved for 10 min in a 20 mL Pyrex vial. The mixture was then heated at 120 °C for 2 days. After cooling down to room temperature, the gray-white powder was collected by centrifugation.

Subsequently, the product was washed with DMF for 3 days (6 mL per day). Next, this material was soaked in a methanol/H2O (v/v = 4/1) solution of H2SO4 for 2 days (0.3 M H2SO4, 5 times per day). The sample was washed to pH = 5 with an excess amount of methanol/H2O (v/v = 4/1). Finally, this material was exchanged with MeOH for 2 days (5 × 5 mL per day), centrifugated, dried and activated under vacuum at 80 °C for 1 day to yield an acidified VNU-23 (70% yield, based on Zr4+).

Batch adsorption experiment

The MB adsorption studies were carried out at room temperature under a constant stirring rate (500 rpm). An adsorbent dosage of VNU-23 of 5–30 mg was introduced to the initial concentration of MB of 20 mg L−1. To study the effect of initial pH on the adsorption behaviour, 500 mg L−1 of MB was chosen as the initial concentration with the dosage of VNU-23 of about 0.25 g L−1 after stirring for 24 h. The initial pH (3–10) of MB was adjusted by the addition of 0.01 M NaOH and 0.01 M HCl solutions and recorded using a pH meter. The spectrum of MB solutions was analysed on a UV-vis spectrophotometer at a maximum wavelength of 663 to 664 nm. The concentration of MB was determined by using a calibration curve made from the standard solutions (Fig. S3 and S4†).

The adsorption capacity (mg g−1) at equilibrium (qe) and time t (q) and removal percentage (R%) in the case of MB were calculated by using the following equations:where C0, C and Ce are the concentration of MB at initial, t and equilibrium time, respectively. V (mL) is the volume of the solution and m (mg) symbolizes the adsorbent mass.

To investigate the adsorption isotherms, 25 mg of VNU-23 was introduced to 100 mL of the MB solution (100–1600 mg L−1) at pH = 7. The MB concentrations remaining in the supernatant solution after 24 h were measured.

Reusability

To study the reusability of the materials, a series of adsorption–desorption cycles were conducted. After the complete adsorption of MB at the specific conditions, the materials were centrifugated and then immersed and stirred in ethanol at pH = 3 as an effective desorption medium. Subsequently, the materials were dried under vacuum at 80 °C for 24 h, and then tested for the next adsorption experiment.

Results and discussion

Characterization of Zr-sulfonic-based MOF

As in previous reports,[29,30] VNU-17 and VNU-23 are successfully synthesized from a mixture containing ZrOCl2·8H2O salt and H3SNDC for VNU-17 or H4SNDC for VNU-23 in N,N-dimethylformamide (DMF) in the presence of formic acid as a modulator and heating at 120 °C for 1 day. To gain insight into the adsorption ability of these materials, we recall the structure of VNU-17 and VNU-23. They crystallize in the tetragonal space group, I4/m, with the unit cell of a = b = 17.75 and c = 22.48 Å. Particularly, VNU-17 and VNU-23 are iso-reticular, generated by 8-connected Zr6O8(H2O)8(–CO2)8, combined by SNDC3− and SNDC4− linker, respectively (Fig. 1b and c). This leads to a 3D structure based on bcu topology, different from that of DUT-52 (Fig. 1a).[31] Interestingly, VNU-17 and VNU-23 possess the dense SO3− groups within the framework, suitable for the adsorption of MB+via electrostatic attraction. Meanwhile, these groups exhibit a repulsive interaction towards a negative moiety like methyl orange (MO−) (see Fig. 2b). The pure phases of the as-synthesized DUT-52, VNU-17 and VNU-23 are further evidenced by powder X-ray diffraction (PXRD) analyses (Fig. 2a), which are in good agreement with the previous works.[29,30] Additionally, to generate a perfect adsorption space within the structure, these materials are immersed in H2SO4 (0.3 M) to remove dimethylammonium ions, which appears by the decomposition of DMF solvent in the synthetic process, and washed with an excess amount of the solvents to pH = 5. Also, Fourier transform infrared (FT-IR) and thermogravimetric analysis (TGA) are conducted, which confirm the presence of the sulfonate groups by the broad peaks at 1178 and 1190 cm−1 (Fig. S6†), and reveal the thermal stability of the materials up to 450 °C (Fig. S12†). Subsequently, we demonstrate that the chemical stability of the Zr-sulfonic-based MOFs in water is retained after 30 days, as evidenced by PXRD analysis (Fig. S8†). It is noted that the sulfonic-based MOFs lose the long-range structural order upon activation.[29,30] This is also in line with the N2 adsorption analysis for the VNU-23 material (Fig. S14†); the BET surface area of the activated VNU-23 is lower than the theoretically predicted value. This phenomenon can be attributed to the flexibility of the SO3H group after recovery, affecting the structural order of the materials.[32,33] However, the crystallinity is fully restored after immersing in water (Fig. S9†). The presence of MB molecules within VNU-23 is also confirmed by TGA analysis (Fig. S13†). Accordingly, the thermal gravimetric curves indicate a weight loss of about 5 wt% from room temperature to 300 °C, which is attributed to the loss of coordinated water, followed by a steep drop from 300 to 600 °C. This steep drop is assigned to the loss of MB molecules and framework decomposition (Fig. S13†).

Fig. 2

Powder X-ray diffraction analysis of as-synthesized VNU-17 (red) and as-synthesized VNU-23 (blue) in comparison with as-synthesized DUT-52 (black) (a); the plausible adsorption mechanism of MB+ and MO− on VNU-23 (b). Herein, the structure of VNU-23 is illustrated from the cif information with the CCDC number of 1585550.[30] Atom colors: Zr6O8(H2O)8 cluster, blue polyhedra; C, black; O, red; N, blue; S, yellow; H, gray.

In order to inspect the effect of the number of SO3H groups within the MOF on the MB adsorption capacity, a series of specific experiments are conducted. In detail, 10 mg each of DUT-52, VNU-17 and VNU-23 is introduced into 15 mL of MB (200–1000 mg L−1) at an initial pH range of 4 to 5. Fig. 3a reveals that the adsorption capacities for MB (1000 mg L−1) on DUT-52, VNU-17 and VNU-23 are 20.140, 314.75 and 1290.5 mg g−1, respectively. The increase in the adsorption capacity upon raising the number of SO3H groups within VNU-17 and VNU-23 can be attributed to the rise in the attraction forces between the highly negatively charged SO3− moieties in the MOF and the positively charged MB molecules. In contrast, DUT-52 exhibits a very poor adsorption capacity for MB due to the absence of the SO3H groups within the structure. Interestingly, we illustrate that the SO3− moieties of VNU-23 can strongly attract the MB+ ions and repulse the MO− ions as evidenced by the UV-vis spectrum (Fig. 3b). Specifically, the maximum peak at 464 nm of MO is significantly unchanged after 5 min, while the maximum peak at 664 nm of MB has almost disappeared after stirring for 15 min (Fig. 3b). Based on these data, VNU-23 is chosen for the following adsorption experiments in this main paper. In addition, as a full comparison and evaluation on the effect of the number of SO3H groups, the kinetic studies, adsorption isotherm models, and reusability of VNU-17 are presented in Section S9.†

Fig. 3

Effect of the number of SO3H groups within the structures of DUT-52, VNU-17 and VNU-23 on the adsorption capacity of MB [m = 10 mg, VMB = 15 mL, t = 4 h, pH = 4–5] (a); the adsorption property of VNU-23 in a mixture of MB and MO [m = 10 mg, C0(MB) = 10 mg L−1, C0(MO) = 20 mg L−1, Vmixture = 20 mL, t = 0–15 min] (b); effect of pH medium on the adsorption capacity and removal efficiency of MB on VNU-23 [m = 10 mg, C0 = 500 mg L−1, VMB = 15 mL, t = 24 h] (c); effect of VNU-23 dosage on removal efficiency of MB [m = 2–30 mg, C0 = 20 mg L−1, VMB = 100 mL, pH = 7, t = 24 h] (d).

Effect of pH

Fig. 3c shows that the adsorption capacity and the removal efficiency of MB are raised with the increase in the pH value from 3 to 7. This behaviour can be accounted for by the increasing pH (3 to 7) of the MB solution leading to an increase in the magnitude of the negative charges on the surface of VNU-23, which in turn raises the electrostatic interaction between the SO3− groups inside the MOF and MB+. However, the removal and the adsorption capacity are reduced at a pH range from 8 to 10. This can be explained by the high competition between OH− ions and SO3− components with MB+ at a pH > 7, resulting in a decreased tendency of MB adsorption on VNU-23. Thus, the optimal pH of 7 is selected for the subsequent studies.

Effect of adsorbent dose

The effect of adsorbent dose on the adsorption of MB on VNU-23 is given in Fig. 3d. The removal efficiency rises with increasing adsorbent dose and reaches 93% with a dosage of 0.2 g L−1. This result indicates that other active sites become available on raising the adsorbent dose, allowing more adsorbate ions to adhere to it.

Adsorption isotherm models

To study the essence of the merge between VNU-23 and MB, the Langmuir, Freundlich, Temkin, and Dubinin–Radushkevich isotherm models have been employed for the interacting liquid–solid phase.

The linear forms of the Langmuir and Freundlich models are displayed by eqn (4) and (5).[34]where qe and Ce are the MB removal amount and MB concentration at equilibrium, respectively, KL is the Langmuir constant, qm is the theoretical maximum capacity of MB adsorption, and KF and n are the Freundlich constants.

The Temkin model is also employed by eqn (6),where R is the gas constant, b and kT are the constant of adsorption heat and the constant of Temkin, respectively, and T is the adsorption temperature.

Furthermore, the isotherm of Dubinin–Radushkevich (DR) is investigated by eqn (7),where ε is a constant, kDR and C are the constant of DR isotherm and the adsorption energy per adsorbent molecule, respectively.

Fig. 4b, c and S19† reveal the linear plots of the Langmuir, Freundlich, Temkin and DR isotherms for the adsorption of MB on VNU-23, respectively. The model parameters, acquired by using the models to the experimental data, are also given in Table 1.

Fig. 4

Effect of initial concentration on the adsorption capacity of MB onto VNU-23 [m = 20 mg, VMB = 100 mL, C0 = 200–1600 mg L−1, pH = 7, t = 24 h] (a). Data fitting with the adsorption isotherm models: Langmuir (b) and Freundlich (c).

The parameters calculated from isotherm models for the adsorption of MB over VNU-23

Isotherms	Paramaters	Value
Langmuir	q m (mg g−1)	2000
	K L (L mg−1)	0.011
	R 2	0.999
Freundlich	1/n	0.230
	K F (mg g−1 (L g−1)1/n)	433.2
	R 2	0.975
Temkin	b	7.805
	k T (L mg−1)	0.558
	R 2	0.967
DR	C (mol2 J2)	3 × 10−5
	k DR (mg g−1)	1623.6
	R 2	0.755

As given in Table 1, the equilibrium data are fitted better by the Langmuir model (R2 = 0.999) than by the Freundlich model (R2 = 0.975), Temkin model (R2 = 0.967) and DR model (R2 = 0.755) for the adsorption of MB onto VNU-23, suggesting that only a layer of the MB molecules is formed at the adsorption sites of the adsorbent. Hence, it can be supposed that the monolayer Langmuir adsorption isotherm is the corresponding model to describe MB adsorption onto VNU-23. It is noteworthy that the theoretical value of qm (2000 mg g−1) determined from the Langmuir model is much close to the experimental value (1992 mg g−1, see Fig. 4a).

Also, the separation factor RL, which is a vital parameter derived from the Langmuir model, is considered a credible magnitude for evaluating the adsorption favourability, where C0 is the initial MB concentration and KL is the Langmuir constant.

It is noted that the RL value is found to be less than 1 and greater than zero, which can indicate the strong interactions between the active sites of SO3− and MB+ ions. Also, the 1/n value obtained from the Freundlich model indicates the favourable process of MB over VNU-23.

Adsorption kinetics

To investigate the rate of adsorption mechanism of MB on VNU-23, studies on the kinetics of adsorption are conducted by the pseudo first order and pseudo second order models. The types of these models are determined by eqn (9) and (10), respectively. In addition, the intra-particle diffusion model is utilized for evaluating the MB adsorption ability onto VNU-23, exhibited by eqn (11).where q and qe are the MB uptake amounts at t and equilibrium, respectively, kf and ks are the rate constants of pseudo first and second order, respectively, ki is the rate constant of the intra-particle diffusion, and c is the constant exhibiting the thickness of the boundary layer.

The kinetic models and parameters are shown in Fig. 5 and Table 2. It is noted that the experimental value best fits the pseudo second order (R2 = 0.999) than the pseudo first order (R2 = 0.882). Significantly, the pseudo second order has a higher affinity between the experimental and calculated adsorption capacities. In fact, the pseudo second order is considered as the adsorption process controlled by the rate-limiting stage of chemical adsorption, including the electron transfer and electron sharing process.[35,36]

Fig. 5

The kinetic curve for the adsorption of MB on VNU-23 [m = 5 mg, VMB = 20 mL, C0 = 50 mg L−1, pH = 7] (a). Data fitting with the adsorption models: pseudo first order (b), pseudo second order (c) and intra-particle diffusion model (d).

Pseudo first order and second order model parameters for the adsorption of MB onto VNU-23

Kinetic models	Paramaters	Value
	q e,exp (mg g−1)	159.8
Pseudo first order	q e,cal (mg g−1)	104.3
	k f (min−1)	0.026
	R 2	0.882
Pseudo second order	q e,cal (mg g−1)	166.7
	k s (10−4) (g mg−1 min−1)	4.912
	R 2	0.999

Furthermore, the obtained results in Fig. 5d indicate that the adsorption is achieved through three stages.[37-39] In detail, the rate constant values (ki) in Table 3 of the intra-particle diffusion process point out that ki1 > ki2 > ki3 can be attributed to the change in the MB diffusion rate during three stages.

Intra-particle diffusion model parameters for the adsorption of MB onto VNU-23

Intra-particle diffusion	Paramaters	Value
First stage	k i1	23.15
	R 2	0.999
Second stage	k i2	2.492
	R 2	0.987
Third stage	k i3	1.883
	R 2	0.939

The first stage is fast and assigned to the boundary layer formed on the external surface of the VNU-23 framework, while the second stage is slower and can be ascribed to the entrance of the MB molecules into the internal surface. Finally, in the third stage, the MB molecules diffuse slowly into the pores and interact with the active sites of VNU-23 till the equilibrium is reached. Although the last stage is slow, it is not the controlling stage of the adsorption process, as the straight line does not pass through the origin.

The reusability of VNU-23 has been studied through five cycles for MB removal. The adsorption uptake of MB onto VNU-23 is indicated in Fig. 6. A removal efficiency of 97% is retained after five cycles. It is worth mentioning that VNU-23 is easily reused after all the cycles of the adsorption process without loss of initial weight.

Furthermore, the achievable regeneration of VNU-23 is evidenced by the FT-IR spectra and PXRD analysis in Section S10.† After the adsorption of MB, the characteristic vibrations of MB are found in MB⊂VNU-23 (see Fig. S7†). However, after immersing in ethanol, these signals disappear, as shown in Fig. S21.† Also, a high degree of correspondence in the PXRD patterns is observed for the VNU-23 samples before and after the desorption of MB (Fig. S20†). These represent the ability to utilize VNU-23 as a potential and highly effective reusable candidate for practical applications in the removal of organic dyes from an aqueous environment.

A comparison of the maximum adsorption capacity (qm) of VNU-23 with that of other reported adsorbents is represented in Table 4. It is seen that VNU-23 exhibits the highest adsorption capacity for MB (qm = 1992 mg g−1) as compared with the other adsorbents. The high adsorption capacity can be explained by the effective attraction interactions of SO3− moieties inside VNU-23 with MB+ cation. According to the obtained results, VNU-23 is chosen as an efficient and reusable adsorbent for removing cationic dyes from wastewater.

The maximum adsorption capacities of the MB cation onto the different reported adsorbents

Materials	q m (mg g−1)	Ref.
UiO-66/MIL-101(Fe)-GOCOOH	448.7	37
MIL-100 (Fe)	736.2	24
Ce(iii)-doped UiO-66	145.3	40
Ce(iii)-doped UiO-67	398.9	41
MIL-101-SO3H	351.0	42
Amino-MIL-101(Al)	762.0	43
MOF-235	252.0	44
NH2- MIL-125	405.6	44
Er-MOF	192.3	45
VNU-17	579.6	This work
VNU-23	1992	This work

Plausible adsorption mechanism for VNU-23

The functional groups, the pore window/channel sizes, surface areas, and pore volumes of MOFs can influence the adsorption possibility of materials.[46] It is noted that the pore window and cage of VNU-23 are sufficient for MB to enter along the channels in the Oz axis (Fig. S1†). Based on the SO3H groups existing within the structure of VNU-23, the adsorption mechanism of MB can be supposed to have several interactions, incorporating electrostatic interaction and π–π stacking. To demonstrate the mentioned hypothesis, we have performed a series of additional experiments. Firstly, the FT-IR spectra of MB⊂VNU-23 display unique peaks, centered at 1661 and 1333 cm−1, that are absent in the spectrum of VNU-23. These are assigned to the vC and vC–N–C vibrational modes of MB molecules, respectively (Fig. S7†). Notably, the peak at 1556 cm−1 for VNU-23 is attributed to the CC vibrations in the benzene ring. However, for MB⊂VNU-23, this peak exhibits a difference from the peak at 1480 cm−1. Such a change maybe from the π–π interactions between the channel wall and the conjugate planes of MB molecules.[47] Moreover, the two adsorption bands for MB⊂VNU-23 centered at 1135 and 1176 cm−1 correspond to the sulfonic groups' vibrational modes. These bands display a shift to the lower frequency in comparison with the original peaks of the sulfonic groups within VNU-23, which suggests that the mechanism of adsorption of MB onto VNU-23 includes electrostatic interactions between the SO3− groups and MB+. This is in good accordance with the previous report.[48]

To further confirm the stated interactions, the PXRD patterns of dyes⊂VNU-23 have been investigated. Recalling the evidence that the structural order of VNU-23 is lost after activation (see Fig. S9†), we expect that if the dye molecules are successfully diffused into the pores of VNU-23, its structure will be retained upon activation. Hence, PXRD analyses are conducted for the effective evaluation of whether the dye molecules are loaded into the pores of VNU-23.[29,30] Consequently, as given in Fig. S10,† the PXRD pattern shows that MB⊂VNU-23 (absorbed at pH = 7) retains its crystallinity after activation, thus proving the diffusion of MB within the internal pores of VNU-23 via electrostatic and π–π interactions. Also, Rietveld refinement is carried out to match the simulated and experimental PXRD patterns of MB⊂VNU-23, to obtain a deeper insight into the loading of MB within the pores of VNU-23 (Fig. S11†).

To verify the specific π–π interactions inside the pores, VNU-23 is soaked in MB solution at pH = 3, which facilitates the formation of neutral MB molecules (pKa = 3.8).[49] The PXRD analysis of activated MB⊂VNU-23 (absorbed at pH = 3) is in reasonable conformity with that of the simulated VNU-23, resulting in the conclusion that the neutral MB molecules are loaded into the channel wall of VNU-23 through π–π interactions (Fig. S10†). Similarly, rhodamine B (RhB+, 6.8 × 11.8 × 15.8 Å), as a cationic dye, is loaded onto VNU-23 at pH = 7. The size of the RhB+ molecules is unsuitable for entering the pores of VNU-23, driving to a poor adsorption capacity for RhB onto VNU-23 (ca. 42 mg g−1), which is attributed to the adsorption of RhB onto the external surface of the framework. This results from the loss of structural order for the RhB⊂VNU-23 sample (Fig. S10†).

Conclusion

In summary, a series of experiments for the adsorption of MB onto Zr-sulfonic-based MOFs have been performed. It is noted that the maximum adsorption capacity for MB onto VNU-23 is found to be 1992 mg g−1 at pH = 7. This value is much higher than that of other MOFs and most the other porous materials reported previously. The equilibrium data are studied using the Langmuir, Freundlich, Temkin, and Dubinin–Radushkevich isotherm models. The results exhibit that the experimental data are efficiently correlated by the Langmuir adsorption isotherm. In addition, the kinetic mechanism is studied by using the pseudo first order, pseudo second order, and intra-particle diffusion models. The obtained results reveal a good agreement between the experimental data and the pseudo second order kinetic model. Here, the MB adsorption over VNU-23 can be explained to be through electrostatic attraction and π–π interactions. Furthermore, the recyclability tests indicate a good capability of reusing VNU-23 for many cycles with no significant reduction in the adsorption capacity, confirming the practical applications of our work.

Conflicts of interest

The authors maintain that they have no conflict of interest for this communication.