Adsorption behavior of magnetic amino-functionalized metal–organic framework for cationic and anionic dyes from aqueous solution

Abstract

As organic dyes are a major group of water pollutants, the development of materials for removal of dyes is of great significance for the environment. Here we report a kind of magnetic metal–organic framework (MOF) for the adsorption of different kinds of anionic and cationic dyes. Magnetic NH₂-MIL-101(Al) was synthesized and characterized with Fourier-transform infrared spectrometry (FT-IR), physical property measurement system (PPMS), X-ray diffraction (XRD), a transmission electron microscope (TEM), a thermogravimetric analyzer (TGA) and nitrogen adsorption/desorption measurements. Adsorption isotherms and thermodynamic studies indicated that the maximum adsorption capacities of malachite green (MG) and indigo carmine (IC) were 274.4 and 135 mg g⁻¹, respectively. The mechanisms of interaction such as electrostatic interaction, hydrogen bonding, π–π stacking interaction, and hydrophobic interaction are discussed for the adsorption of cationic and anionic dyes onto magnetic NH₂-MIL-101(Al). Also, magnetic separation in this method shortened separation time, and the magnetic material can be reused at least five times without obvious decrease in the removal efficiency. Moreover, the synthesis of the magnetic NH₂-MIL-101(Al) is easy to control and the good solvent stability makes this material promising as a novel adsorbent for the adsorption and removal of dyes from aqueous solution.

---

1. Introduction

Organic dyes are important products, widely used in textile, paper, rubber and plastics, etc. They have posed a significant threat to the environment and creatures' health due to the fact that they have a certain toxicity and can even be carcinogenic.¹ It is thus of vital importance to solve the problem of pollution caused by organic dyes. However, wastewater containing dyes is very difficult to treat, since the dyes are recalcitrant organic molecules, resistant to aerobic digestion, and are stable to light, heat and oxidizing agents.² There are numerous conventional methods that can be used for removing organic dyes from wastewater, including biotransformation,³ coagulation,⁴ membrane filtration,⁵ adsorption⁶ and advanced oxidation.⁷ Currently, adsorption is the most common treatment for effective dyestuff removal because it is simple to operate, high performing and cost-effective. In the past few years, a number of adsorbents have been reported to eliminate dyes, such as agricultural wastes,⁸ inorganic materials⁹ and polymers.¹⁰

Recently, metal–organic frameworks (MOFs) have met with great interest, owing to their enormous variety of interesting molecular topologies¹¹ and a wide range of potential applications as functional materials. MOFs have some advantages as adsorbents in that their structures and pore sizes are tunable.¹² Further, they contain polar and polarisable bonds, and can even bear open (or, in aqueous solution, water-coordinated) coordination sites on metal atoms.¹³ Thus, the highly regular channel structures and controllable pore sizes of MOFs permit their applications in adsorption of guest molecules/ions, such as dyes or lanthanide ions. As potential organic dye adsorbents, however, porous MOFs have received much less attention than the well-known activated carbons. As far as we know, only a few examples of porous MOFs being used for organic dye adsorption have been reported to date.^14,15

Among the numerous MOFs reported so far, the most widely used mesoporous MOFs are represented by the MIL-100 and MIL-101 types (MIL stands for Material of Institute Lavoisier). The huge variety of possible applications¹⁶ of these MOFs is based upon their mesoporous cavities (Fig. S1†), large surface areas, the presence of accessible metal sites and their comparatively high thermal and chemical robustness.^17–19 For example, the unfunctionalised MIL-101(Cr) has been used as a sorbent for the cationic dye malachite green⁹ and anionic dye xylenol orange.²⁰ In addition, the presence of coordinatively unsaturated metal sites in MIL-101 allows its use as a mild Lewis acid²¹ and, more importantly, allows its postfunctionalization via grafting of active species.²² NH₂-MIL-101(Fe) has been reported and applied in drug delivery²³ and medical imaging,²⁴ where the relatively low thermal and chemical stability shown by this framework does not seem to affect its outstanding performance. Considering other types of application, the development of materials with a similar topology and stability as MIL-101(Cr) but presenting functional organic sites is needed.

Meanwhile, chemically synthesized magnetic composites are of particular interest for their unique magnetic properties which mean that they can be effectively separated under an external magnetic field and recycled for reuse. Magnetic materials based on SiO₂,²⁵ zeolites,²⁶ mesoporous CeO₂ (ref. 27) and activated carbon²⁸ have been successfully investigated. Considering the superior properties of MOFs and their successful development in a wide range of research fields, the design and synthesis of magnetic MOFs are especially desirable.

Herein, we demonstrate a novel scheme to fabricate magnetic NH₂-MIL-101(Al), which combines the magnetic properties of Fe₃O₄ nanoparticles with the MOF's functional properties. The resulting material shows a high chemical and thermal stability. Moreover, owing to the presence of amines, magnetic NH₂-MIL-101(Al) displays outstanding adsorption properties and has been found to be a versatile adsorbent for removing various dyes in aqueous solution. We chose different kinds of dyes (Fig. S2†) to investigate the mechanisms of interactions, such as electrostatic interaction, hydrogen bonding, π–π stacking interaction and hydrophobic interaction. The influence of the molecular size for the dye's adsorption is also investigated. The prepared magnetic NH₂-MIL-101(Al) composites offer superparamagnetic properties and excellent adsorption capability for the fast magnetic removal of dyes from water samples.

2. Experimental

2.1. Reagents and instruments

The standards of methyl orange (MO), methylene blue (MB), malachite green (MG), and indigo carmine (IC) were purchased from Aladdin (Shanghai, China). Iron(III) chloride hexahydrate (FeCl₃·6H₂O), iron(II) chloride tetrahydrate (FeCl₂·4H₂O), ammonia (NH₃·H₂O), tetraethyl orthosilicate (TEOS), 3-aminopropyltriethoxysilane (KH550), sodium hydroxide (NaOH), and hydrochloric acid (HCl) were obtained from Kermel (Tianjin, China). Aluminum chloride hexahydrate (AlCl₃·6H₂O) and 2-aminoterephthalic acid were purchased from Sigma (St. Louis, MO, USA). N,N-Dimethylformamide (DMF), ethanol, and methanol were obtained from Fuyu (Tianjin, China). Stock solutions of cationic and anionic dyes (1000 mg L⁻¹) were prepared by dissolving the solid in water, and keeping it at 4 °C in a refrigerator. High purity water was obtained from a Milli-Q water system (Millipore, Billerica, MA, USA). All other reagents were of analytical grade.

The magnetic MOFs were characterized by a transmission electron microscope (TEM, H7650, Hitachi, Japan), physical property measurement system (PPMS, Quantum Design Instrument, San Diego, CA, USA), Fourier-transform infrared spectrometry (FT-IR360, Nicolet, Madison, WI, USA), XRD-600 diffractometer (Shimadzu, Kyoto, Japan), thermogravimetric analyzer (TGA; Pyris 1, PerkinElmer, Waltham, MA, USA) and a Brunauer–Emmett–Teller (BET) surface area instrument (Quadrasorb SI-MP, Quantachrome, Florida, UK). The dye concentrations were determined using the absorbance of the solutions after getting the UV spectra of the solution with a spectrophotometer (Shimadzu UV spectrophotometer, UV-1800). The adsorption experiments were carried out at constant temperature controlled by a SHA-B shaking table (Shengtang, Tianjin, China). A KQ5200E ultrasonic apparatus (Kunshan Instrument, Kunshan, China) was used to ensure uniform dispersion.

2.2. Preparation of the amino-functionalized Fe₃O₄ nanoparticles

The Fe₃O₄ nanoparticles were prepared by a coprecipitation method. FeCl₂·4H₂O (0.01 mol) and FeCl₃·6H₂O (0.02 mol) were dissolved in 100 mL water. The mixture was stirred vigorously while the temperature increased to 80 °C, and then 10 mL of ammonia was added into the solution. The mixture was stirred for 1 h at 500 rpm under an N₂ stream. After that, the magnetic particles were separated by a permanent magnet and washed with water until the pH was 7. The product was then dried at 60 °C for further use.

The magnetic Fe₃O₄ (0.5 g) obtained was dispersed into a mixed solution of water and ethanol (1 : 1, v/v). Then 3 mL of NH₃·H₂O and 5 mL of TEOS were added under stirring at 40 °C for 24 h. The products were reacted with 3-aminopropyltriethoxysilane (2 mL) in 100 mL of water to obtain amino-functionalized Fe₃O₄ nanoparticles under continuous agitation for 6 h at 60 °C. The resultant nanoparticles were collected and washed with ethanol and water. Then the solid was finally dried at 80 °C under vacuum overnight.

2.3. Preparation of magnetic MOFs

The 2-aminoterephthalic acid (0.56 g) was dissolved in 30 mL of DMF and AlCl₃·6H₂O (0.51 g) was dissolved in 40 mL of DMF. The 2-aminoterephthalic acid solution was dropped into the solution of Al³⁺ to prepare the pre-assembly solution. Then the amino-functionalized Fe₃O₄ nanoparticles were immediately added under ultrasound and maintained for 30 min to ensure uniform dispersion. The resulting mixture was placed in a Teflon-lined reactor and heated for 10 h at 120 °C. The powder obtained was washed several times with DMF, ethanol and water in order to remove the impurities. To remove organic species trapped within the pores, the product was activated in methanol at 80 °C for 10 h and then dried under vacuum at 60 °C for further use.

2.4. Adsorption experiments

In a typical adsorption experiment, the adsorbent (10 mg) was weighed precisely and put in the aqueous dye solutions (5 mL) with concentrations from 10 to 1000 mg L⁻¹. The dye solutions containing the adsorbents were mixed well by a shaking table and maintained for a fixed time (10 min to 120 min) at room temperature. After adsorption, the solutions were separated from the adsorbents with a magnet, and the dye concentrations were detected. To get the thermodynamic parameters of adsorption, such as ΔG⁰ (free energy change), ΔH⁰ (enthalpy change) and ΔS⁰ (entropy change), the adsorptions were further carried out at 25, 35 and 45 °C. To determine the adsorption capacity at various pH values, the pH of the dye solutions was adjusted with 0.1 M HCl or 0.1 M NaOH aqueous solution.

3. Results and discussions

3.1. Fabrication of magnetic NH₂-MIL-101(Al)

The framework of the MOF is based on the coordination interaction between the carboxyl group of 2-aminoterephthalic acid and the Al(III) of AlCl₃·6H₂O. S. Jung et al.²⁹ confirmed an assumption that a certain number of the carboxyl groups are exposed on the crystalline surface and can be conjugated with other compounds via chemical interaction. Accordingly, in our research, the Fe₃O₄ nanoparticles were first functionalized with amino groups, which can form chemical bonds with carboxyl groups exposed on the surface of MOF.

Previous studies proved that the functionalization of an MOF could improve its adsorption ability to target molecules. MOF functionalized with ethylenediamine sourced from NH₂ was used for the adsorption of naproxen and clofibric acid.³⁰ In this study, NH₂-MIL-101(Al) is built from supertetrahedral building units formed by aminoterephthalate ligands and trimeric Al(III) octahedral clusters and thus bears an amino group on every linker,³¹ rather than using the post synthesis grafting method that was demonstrated by Zubair Hasan and co-workers.³⁰

When the MIL-101(Al) has been endowed with magnetism and amine functionalization, the resulting composite material, magnetic NH₂-MIL-101(Al) not only has a porous structure, it also has rapid separation and good adsorption ability.

3.2. Characterization of magnetic NH₂-MIL-101(Al)

The FT-IR spectrum of magnetic MOF is shown in Fig. 1a. The characteristic peak at 602 cm⁻¹ was assigned to the Fe–O bond. The observed feature around 1089 cm⁻¹ was the Si–O–Si bond of magnetic MOF. The peak at 3431 cm⁻¹ corresponded to an O–H group. The peak at 1398 cm⁻¹ was attributed to the existence of O–H bending vibration. The adsorption band around 2973 cm⁻¹ unveiled the stretching vibration of the C–H group in CH₃. The characteristic band of the amino group at 1623 cm⁻¹ was attributed to the N–H bending (scissoring) vibration.

Fig. 1 FT-IR spectrum (a), magnetization curve (b), TEM images (c and d) and thermogravimetric curve (e) of magnetic NH₂-MIL-101(Al).

The saturation magnetization of the composites was performed in an applied magnetic field at room temperature. As shown in Fig. 1b, the magnetization curve showed a superparamagnetic property. The magnetic saturation value of the magnetic NH₂-MIL-101(Al) composites is 18.73 emu g⁻¹. This value demonstrated that the composite has strong magnetism.

The TEM image of magnetic NH₂-MIL-101(Al) (Fig. 1c) showed a homogeneous surface morphology with uniform particle size. Moreover, the diameter of magnetic NH₂-MIL-101(Al) (Fig. 1c) increases markedly compared with crude Fe₃O₄ nanoparticles (Fig. 1d), which revealed that the NH₂-MIL-101(Al) layer was successfully attached to Fe₃O₄.

To reveal the thermal stability of magnetic NH₂-MIL-101(Al), a thermo-gravimetric experiment was carried out with pure single crystal samples under N₂ atmosphere conditions at a rate of 10 °C min⁻¹ over a range of 40–850 °C. As shown in Fig. 1e, the first main weight-loss was due to the dehydration of the material up to 120 °C. These weight losses, starting at roughly 380 °C for magnetic NH₂-MIL-101(Al), mark the degradation of the framework. This degradation temperature compares well to the value reported in the literature.³²

The XRD patterns of the parent materials (NH₂-MIL-101(Al), Fe₃O₄@SiO₂) and the magnetic NH₂-MIL-101(Al) are presented in Fig. 2a. The experimental XRD pattern shows the as-prepared magnetic NH₂-MIL-101(Al) composites match well with those of both NH₂-MIL-101(Al) and Fe₃O₄@SiO₂ nanospheres, confirming that the magnetic NH₂-MIL-101(Al) composites obtained are composed of crystalline NH₂-MIL-101(Al) and Fe₃O₄@SiO₂ nanospheres. The existence of Fe₃O₄@SiO₂ nanospheres and NH₂-MIL-101(Al) was further confirmed by energy-dispersive X-ray microanalysis (Fig. 2b). Elemental C, O, N, Al, Si and Fe were detected in the magnetic NH₂-MIL-101 (Al) composites.

Fig. 2 Powder XRD pattern (a), energy-dispersive X-ray microanalysis (b), BJH pore size distributions (c) and N₂ adsorption–desorption isotherms (d) of magnetic NH₂-MIL-101(Al).

To further characterize the porosity of the magnetic NH₂-MIL-101(Al), nitrogen-adsorption experiments were performed at 77 K. As can be seen from Fig. 2c, the pore-size distribution of the as-prepared magnetic NH₂-MIL-101(Al) samples calculated from the desorption branch of the N₂ isotherm by the Barrett–Joyner–Halenda (BJH) model was 3.88 nm. As shown in Fig. 2d, NH₂-MIL-101(Al) and magnetic NH₂-MIL-101(Al) exhibited a type IV isotherm with a distinct hysteresis loop in the relative pressure P/P₀ range of 0.5–1.0, suggesting a mesoporous structure. The BET surface area measured from the nitrogen isotherms was 285.61 m² g⁻¹. Compared with the NH₂-MIL-101(Al) (392.14 m² g⁻¹), the BET surface areas of the composites significantly decreased due to the incorporation of Fe₃O₄ nanospheres.

3.3. Adsorption isotherms

The adsorption isotherms of dyes onto magnetic NH₂-MIL-101(Al) were measured at three different temperatures. We take MG and IC as the representatives to study the adsorption properties of cationic and anionic dyes. When the MG and IC concentrations were at 1000 mg L⁻¹, the adsorption capacities of magnetic NH₂-MIL-101(Al) for MG and IC were 274.4 mg g⁻¹ and 135 mg g⁻¹ at 298 K (Fig. 3a), respectively. The adsorption isotherm models are fundamental for describing the interactive behavior between adsorbate and adsorbent and important for investigating mechanisms of adsorption. In this study, Langmuir,³³ Freundlich,³⁴ and Dubinin–Radushkevich³⁵ models were used to describe the adsorption equilibrium data derived from the adsorption of dyes onto magnetic NH₂-MIL-101(Al) at different temperatures (298, 308 and 318 K). Equilibrium adsorption isotherms were studied with dye concentrations ranging from 10 to 1000 mg L⁻¹ with a fixed adsorbent mass and pH.

Fig. 3 Adsorption isotherms (a), Langmuir (b), Freundlich (c) and Dubinin–Radushkevich (d) plots of the isotherms for MG and IC adsorption onto magnetic NH₂-MIL-101(Al) at 298 K.

(a) Langmuir isotherm. The Langmuir model (eqn (1) and (2)), which provides a basic view of adsorption, is usually valid for surfaces with a finite number of identical sites.³⁶ It has been successfully applied to many pollutant adsorption processes from aqueous solution.^37,38 The linear form of the Langmuir model is represented as follows:^33,38where C_e is the liquid-phase concentrations of dye at equilibrium (mg L⁻¹), q_e the amount of dye adsorbed at equilibrium (mg g⁻¹), Q₀ the maximum monolayer adsorption capacity (mg g⁻¹), and b the Langmuir constant (L mg⁻¹ or L mol⁻¹). A linear plot of (C_e/q_e) vs. C_e is obtained from the Langmuir model, as shown in Fig. 3b. All of the correlation coefficients, R², for the isotherms were higher than 0.993 at the three temperatures, indicating that the adsorption of dyes onto magnetic NH₂-MIL-101(Al) can be adequately described by the Langmuir isotherm model.

From the value of b deduced from the Langmuir model, the equilibrium parameter (R_L) was calculated using the following equation:³⁹

where C₀ is the initial dye concentration. The value of R_L indicates whether the isotherm is unfavorable (R_L > 1), linear (R_L = 1), favorable (0 < R_L < 1) or irreversible (R_L = 0).³⁹ The R_L values for dyes adsorption onto magnetic NH₂-MIL-101(Al) were less than one and greater than zero, showing a favorable adsorption (Table 1).

Table 1 Adsorption isotherm parameters derived from the Langmuir model, the Freundlich model and Dubinin–Radushkevich model at different temperatures

Dyes	T (K)	Langmuir model			Freundlich model			Dubinin–Radushkevich model
		Q0 (mg g^−1)	RL	R^2	n	KF (mg g^−1)	R^2	E (kJ mol^−1)	QD–R (mg g^−1)	R^2
IC	298	137.0	0.03–0.73	0.997	4.60	30.14	0.948	0.381	98.16	0.659
	308	131.6	0.02–0.72	0.998	4.66	29.61	0.948	0.366	96.02	0.668
	318	128.2	0.02–0.71	0.998	4.70	29.17	0.947	0.360	94.20	0.675
MG	298	285.7	0.02–0.68	0.993	3.05	38.63	0.976	1.25	156.19	0.561
	308	277.8	0.02–0.68	0.994	3.08	38.16	0.977	1.30	154.02	0.562
	318	270.3	0.02–0.68	0.994	3.12	37.51	0.977	1.35	151.00	0.562

---

(b) Freundlich isotherm. The Freundlich model (eqn (3) and (4)) is used to describe adsorption onto heterogeneous surfaces with different functional groups on the surface and several adsorbent–adsorbate interactions.⁴⁰ The Freundlich model can be represented in linear form as follows:³⁴where C_e is the liquid-phase concentrations of dye at equilibrium (mg L⁻¹) and q_e the amount of dye adsorbed at equilibrium (mg g⁻¹). K_F and 1/n are Freundlich constants, where n indicates the degree of to which an adsorption process is favorable and K_F (mg g⁻¹) (L mg⁻¹)^1/n is the adsorption capacity of the adsorbent. K_F and 1/n can be determined from the linear plot of ln q_e vs. ln C_e (Fig. 3c). The Freundlich constants and correlation coefficients at different temperatures are listed in Table 1.

In general, as the K_F increases, the adsorption capacity of the adsorbent increases. If n < 1, this means poor adsorption, from one to two means moderately difficult adsorption, and from two to ten means good adsorption.⁴¹ All of the n values obtained from the Freundlich model are more than unity, indicating that adsorption of dyes onto the magnetic NH₂-MIL-101(Al) is favorable (Table 1).

(c) Dubinin–Radushkevich isotherm. The nature of adsorption (physical or chemical) was also analyzed by the Dubinin–Radushkevich (D–R) isotherm. The D–R isotherm model (eqn (4)–(6)) is generally applied to express the adsorption mechanism with a Gaussian energy distribution onto a heterogeneous surface. The D–R isotherm model can be expressed as:³⁵

ln q_e = ln Q_D–R − βε² (4)

where q_e is the amount of dye adsorbed at equilibrium (mg g⁻¹), Q_D–R is the theoretical adsorption capacity (mg g⁻¹), β is the constant of the sorption energy and ε is Polanyi potential, which is expressed by the following equation:³⁵where C_e is the liquid-phase concentrations of dye at equilibrium (mg L⁻¹), R is the ideal gas constant (8.314 × 10⁻³ kJ K⁻¹ mol⁻¹) and T is the absolute temperature (K). A linear plot of ln q_e versus ε² is obtained from the model, as shown in Fig. 3d.

The energy of adsorption is the free energy of the transfer of 1 mol of solute from infinity (in solution) to the surface of the adsorbent. The mean value of adsorption energy E, can be calculated from the constant of the sorption energy (D–R pattern) as follows:³⁵

The magnitude of E (kJ mol⁻¹) is used for estimating the type of adsorption mechanism. If this value is between 8 and 16 kJ mol⁻¹, the adsorption process is controlled by a chemical mechanism, while for E < 8 kJ mol⁻¹, the adsorption process proceeds through a physical mechanism.⁴² The calculated values of E (Table 1) suggested that the adsorption of dye occurs via physical adsorption. However, the correlation coefficients of the Dubinin–Radushkevich isotherm were lower than 0.68, which means that this mode is not suitable for the study of dye adsorption onto magnetic NH₂-MIL-101(Al).

In conclusion, the correlation coefficients of the Freundlich isotherm and the Dubinin–Radushkevich isotherm are much lower than those for the Langmuir isotherm, suggesting that the Langmuir isotherm model is the best model to fit the experimental data.

3.4. Adsorption kinetics

The adsorption kinetics is an important parameter for designing adsorption systems and is required for selecting the optimum operating conditions for a pilot-scale process. In this work, as can be seen from Fig. 4a, the adsorption capacity was enhanced rapidly in the beginning, and then slowed down and reached equilibrium. In order to investigate the adsorption kinetics of dyes onto magnetic NH₂-MIL-101(Al), two models have been studied: the pseudo-first-order kinetic model and pseudo-second order kinetic model. The linear form of the pseudo-first-order equation is given as follows:⁴³

ln(q_e − q_t) = ln q_e − k₁t (7)

where q_e and q_t are the amounts of dye adsorbed (mg g⁻¹) at equilibrium and at time t (min), respectively, and k₁ is the rate constant of pseudo-first-order kinetics (min⁻¹). Values of k₁ were calculated from the plots of ln(q_e − q_t) vs. time (Table 2 and Fig. 4b). From Table 2, the lineal regression coefficients obtained from the pseudo-first-order kinetic model were found to be low. Furthermore, there were significant differences between the calculated and experimental q_e values, indicating that the first order model does not reproduce the adsorption kinetics of dyes onto magnetic NH₂-MIL-101(Al).

Fig. 4 Kinetics adsorption (a), pseudo-first-order (b) and pseudo-second-order kinetics (c) for the adsorption of MG and IC onto magnetic NH₂-MIL-101(Al).

Table 2 Adsorption kinetics parameters for the adsorption of MG and IC onto the magnetic NH₂-MIL-101(Al)

Sample	Pseudo-first-order			Pseudo-second-order
	qe,cal (mg g^−1)	k1 (min^−1)	R^2	qe,cal (mg g^−1)	k2 (g mg^−1 min^−1)	R^2
IC	13.01	1.0 × 10^−2	0.958	38.91	5.27 × 10^−3	0.999
MG	16.48	3.01 × 10^−2	0.979	48.78	4.05 × 10^−3	0.999

---

The experimental data was also examined by the pseudo-second-order kinetic model which is given by the following equation:^43,44

where k₂ is the rate constant of pseudo-second-order kinetics (g mg⁻¹ min⁻¹). The rate constant, the amounts of dye adsorbed and the R² values are showed in Table 2. The plots of t/q_t vs. time show excellent linearity (Fig. 4c). The results show that the pseudo-second-order kinetic model fits the experimental data better with linear regression coefficients of 0.99 (R² > 0.99). Furthermore, the calculated q_e values for the pseudo-second-order kinetic model show good agreement with the experimental q_e values.

3.5. Thermodynamic analysis

The thermodynamic parameters that must be considered to determine the process are Gibbs free energy change (ΔG⁰) (eqn (10)), enthalpy change (ΔH⁰) and entropy change (ΔS⁰) (eqn (11)) due to the transfer of a unit mole of solute from solution on the solid–liquid interface. These parameters were calculated using the following equations:

ΔG = −RT ln K₀ (10)

where R is the ideal gas constant (8.314 × 10⁻³ kJ mol⁻¹ K⁻¹), K₀ is the adsorption equilibrium constant, C_s is the amount of dye adsorbed per mass of magnetic NH₂-MIL-101(Al) (mg g⁻¹), C_e is the dye concentration in solution at equilibrium (mg L⁻¹), ΔG⁰ is the change in Gibbs free energy (kJ mol⁻¹), ΔH⁰ is the enthalpy of adsorption (kJ mol⁻¹) and ΔS⁰ is the entropy of adsorption (J mol⁻¹ K⁻¹). The results of the thermodynamic calculations are shown in Table 3. The value of ln K₀ at a certain temperature was obtained by plotting ln(C_s/C_e) against C_s and extrapolating C_s to zero based on eqn (9).⁹ The negative values of change in Gibbs free energy (ΔG⁰) shown in Table 3 indicated the feasibility of the process and the spontaneous nature of the adsorption under the experimental conditions used. ΔH⁰ and ΔS⁰ can be determined from the slope and the intercept of the linear plot of ln K₀ vs. 1/T, as shown in Fig. 5, and the results are also shown in Table 3. The enthalpy change (ΔH⁰) in Table 3 was positive, which suggested that the adsorption process was an endothermic reaction. The disruption of the magnetic NH₂-MIL-101(Al) structure upon exposure to the dyes explains the significantly unfavorable enthalpy upon adsorption, as at least some of the metal–carboxylate bonds that held the structure together were broken and weakened. However, a positive ΔS⁰ is favorable for spontaneous adsorption of dyes onto magnetic NH₂-MIL-101(Al), which may result from the release of more water molecules desolvated from the dye molecules after adsorption onto magnetic NH₂-MIL-101(Al).⁹ Based upon the values of ΔH⁰ and ΔS⁰ for the reaction, the destruction of the ordered MOF structure, and the dissolution of Al³⁺ ions, appear to be the driving force for the interaction.¹³

Table 3 Thermodynamic parameters for the adsorption of MG and IC onto the magnetic NH₂-MIL-101(Al)

Sample	Temperature (K)	ln K0	ΔG^0 (kJ mol^−1)	ΔH^0 (kJ mol^−1 K^−1)	ΔS^0 (J mol^−1 K^−1)
IC	298	6.259	−15.51	2.61	60.80
	308	6.295	−16.12
	318	6.325	−16.72
MG	298	5.103	−12.64	2.20	49.80
	308	5.129	−13.13
	318	5.159	−13.64

---

Fig. 5 Van't Hoff plot to get the ΔH⁰ and ΔS⁰ of IC (a) and MG (b) onto the magnetic NH₂-MIL-101(Al).

3.6. Effect of pH

The pH value of the dye solution is recognized as an important factor in the adsorption process, which influences the surface charge, dissociation of functional groups on the active sites, and degree of ionization of the adsorbents and dyes. In order to further investigate the adsorption mechanism, we chose four kinds of dyes to discuss the effect of pH. The existing forms and distribution curves of these dyes under different pH conditions are shown in Table S1.†

The zeta potential of the absorbent is the one of the factors which might have an important influence on the adsorption capacity. The zero point charge (pH_PZC) is defined as the point at which the zeta potential on the surface of the adsorbent is zero. In aqueous solutions, when pH < pH_PZC, the surface charge of a solid is positive. In contrast, it is negative at pH > pH_PZC. The pH_PZC of magnetic NH₂-MIL-101(Al) shown in Fig. 6a was 10. We investigated the effect of pH on the adsorption of dyes onto magnetic NH₂-MIL-101(Al) (Fig. 6b and c) and studied the adsorption mechanism.

Fig. 6 The pH effect on the zeta potential of magnetic NH₂-MIL-101(Al) (a), effect of the solution pH on the amount of adsorbed anionic (b) and cationic (c) dyes onto magnetic NH₂-MIL-101(Al).

The efficiency of removal of adsorbed MO leveled off in the pH range of 4–9, and decreased with further increase of pH from 9 to 12. When pH > 4, MO mainly exists in the form of anions, has an electrostatic interaction with magnetic NH₂-MIL-101(Al) carrying positive charge at pH 4–10, but has electrostatic repulsion with adsorbents carrying negative charge at pH > 10. At pH = 3, a part of the MO mainly exists as amphoteric molecules, which reduces the electrostatic interaction with materials. The interaction between IC carrying negative charge and magnetic NH₂-MIL-101(Al) was similar to MG.

Compared with the anionic dyes, the cationic dyes showed a contrasting trend. The adsorption efficiency of MG onto magnetic NH₂-MIL-101(Al) increased as the pH increased from 3 to 5, then leveled off in the pH range of 5–9. Because of the hydrolysis of MG dye in basic solution, the influence of pH (10–12) on the adsorption capacity of MG was not evaluated further. MG (pK_a = 10.3) is protonated in an acidic medium and deprotonated at higher pH.^45,46 The removal of MG from the solution showed a fast increase between pH 3 and 5 due to the decreasing positive charge on the surface of magnetic NH₂-MIL-101(Al), which weakens the electrostatic repulsion between the cationic MG molecules and the adsorbent surface. At further pH, MG showed high adsorption under conditions with pH > 5, which might be explained by π–π stacking because both the adsorbate and the adsorbent have benzene rings.

MB is another kind of cationic dye existing in the form of two species in the pH range from 3 to 12. Around pH 3, the main form existing is MBH²⁺. As the pH increases, the main form existing is MB⁺. When pH < 10, the removal efficiency of MB leveled off. As the solution pH increased from 10 to 12, the negative charges on the surface of magnetic NH₂-MIL-101(Al) increased, which was favorable for the adsorption of positively charged MB due to electrostatic attraction.

3.7. Adsorption mechanism

According to the analysis of adsorption kinetics, isotherms, and thermodynamics, it can be considered that the adsorption was related to the chemical action between the dyes and the magnetic NH₂-MIL-101(Al). In addition to the electrostatic interaction, there may be other forces between adsorbent and adsorbate.

First, due to the sphere of influence of –NH₂ involved in the surface of magnetic NH₂-MIL-101(Al), it can easily combine with the adsorbate through hydrogen bonding,^47,48 which promotes the adsorption of dyes onto magnetic NH₂-MIL-101(Al). In addition, there was a potentiality in the structure of magnetic NH₂-MIL-101(Al), where the –OH in an octahedral cluster of three AlO₅(OH) units can combine with dimethylamino (–N(CH₃)₂) to form hydrogen bonds.⁴⁹

In this research, the dyes are compounds endowed with π–π conjugate structure and would have strong affinity with the magnetic NH₂-MIL-101(Al) which is rich in aromatic rings. The π–π accumulation force benefits the adsorption of dyes, which resulted in the high adsorption capacity of the dyes.

Hydrophobic interactions are often observed during adsorption of organics from aqueous media. In this work, with the same kind of dye, the adsorption of MO/MG is higher than that of IC/MB, which may contribute to the difference in the partition coefficient P (log P) (Table S2†). The adsorption amounts increased with an increase in the log P values, indicating the potentially main role of hydrophobic interactions.

Based on the above discussion, we think that the electrostatic interaction is the main force between adsorbent and adsorbate, because when they have opposite charges, the dye adsorption was satisfied. However, forces such as hydrogen bonds, π–π accumulation force, and hydrophobic interactions also exist, because when adsorbent and adsorbate have the same charge, magnetic NH₂-MIL-101(Al) also has a certain interaction with dyes.

MOFs are well ordered and highly porous crystalline materials and one of their unique characteristics is the ability to tune their pore size without altering the structural properties. To check the effect of mesoporosity, the adsorption of MG was conducted over virgin materials and meso-structured magnetic NH₂-MIL-101(Al). As shown in Fig. 7, it was observed that MG was completely adsorbed in 2 h by the meso-structured NH₂-MIL-101(Al) and magnetic NH₂-MIL-101(Al), whereas no apparent adsorption of MG onto Fe₃O₄, Fe₃O₄@SiO₂ was observed after 2 h of adsorption; this clearly indicates the effect of mesopores in accelerating adsorption. The extended structure of amino-MIL-101(Al) is composed of two types of quasi-spherical mesoporous cages formed by 12 pentagonal and 16 hexagonal faces with windows 1.2 and 1.6 nm, respectively, in diameter.²⁹ The measured dimensions of organic dye molecules from the calculation are shown in Table S2.† For the positively charged dye MG⁺, its two relatively short molecular dimensions along the x and y directions are 9.9 and 4.2 Å, respectively, which are much smaller than the diameter of 12.0 Å. As a result, the access of dye MG⁺ into the channel of magnetic NH₂-MIL-101(Al) is allowed. The same mechanism is suitable for the other kinds of dyes.

Fig. 7 The adsorption of MG with different kinds of adsorbent materials.

3.8. Effect of ionic strength on the adsorption of dyes

Wastewater containing dyes commonly has a high electrolyte concentration, and thus the effect of the electrolyte on the removal of the dyes needed to be clarified to evaluate the real value of magnetic NH₂-MIL-101(Al) in the removal of dyes.

To investigate the salt effect on the adsorption of dyes, NaCl was used to adjust the solution salinity. The salt could lead to a high ionic strength, which might affect the adsorption of dyes onto the adsorbent. As shown in Fig. 8, the equilibrium adsorption ability for the MB, MO, and IC decreased slightly as the concentration of NaCl increased from 0 to 0.1 mol L⁻¹. The result indicated that there is some competitive adsorption between these dyes and metal ions, but the effect is very small.

Fig. 8 Effect of the concentration of NaCl on the adsorption of MG, MB, MO and IC onto magnetic NH₂-MIL-101(Al).

In contrast, for MG, the equilibrium adsorption ability increased slightly as the concentrations of NaCl increased from 0 to 0.1 mol L⁻¹ (Fig. 8). One of the investigation conditions for ionic strength is that the solution is neutral, thus the MG also existed as neutral molecules. The increase in MG adsorption capacity may be due to the salting out effect.

3.9. Recycling of adsorbents

Repeated availability is an important factor for adsorbents. The desorption of cationic dyes (MG and MB) from loaded magnetic NH₂-MIL-101(Al) was achieved using 0.1 mol L⁻¹ of HCl/ethanol (10 : 90, v/v), and the desorption of anionic dyes (IC and MO) from loaded magnetic NH₂-MIL-101(Al) was achieved using 0.01 mol L⁻¹ of NaOH/methanol (10 : 90, v/v).⁵⁰ In this study, the reusability of the magnetic NH₂-MIL-101(Al) was estimated after solvent washing of the used magnetic material. As shown in Fig. 9, the removal rates of MG, IC, MO and MB by magnetic NH₂-MIL-101(Al) slightly decrease in successive cycles, gradually stabilizing at about 93%, 88%, 90% and 92%, respectively. The good reusability of spent magnetic NH₂-MIL-101(Al) demonstrates the potential of magnetic NH₂-MIL-101(Al) for the adsorption and removal of dyes.

Fig. 9 Reusability of magnetic NH₂-MIL-101(Al) for the adsorptive removal of four different dyes.

3.10. Comparison of magnetic NH₂-MIL-101(Al) with other adsorbents for the adsorption of dyes

Table 4 shows the adsorption capacities for MG and IC using different adsorbents. The adsorbent used in this study shows a higher adsorption capacity than the most adsorbent materials reported in the literature. Meanwhile, the materials obtained can also be adjusted by pH to adsorb different kinds of dyes from aqueous solution. Moreover, because of the magnetic properties, magnetic NH₂-MIL-101(Al) is attractive for its rapid and efficient adsorption, which indicated that the material is superior to other adsorbents.

Table 4 Comparison of adsorption of MG and IC with other reported systems

	Adsorbent	Qm (mg g^−1)	Temperature (K)	References
MG	MIL-100	485	323	9
	MIL-100	266	303	9
	MIL-53	34.9	303	9
	Commercial powder activated carbon	149	303	9
	Bamboo-based activated carbon	264	303	51
	Cyclodextrin-based material	91.9	298	45
	Oil palm trunk fiber	149	303	52
	Natural zeolite	24.5	298	53
	Chitosan bead	93.6	303	54
	Magnetic NH2-MIL-101(Al)	274.4	298	This work
IC	Rice husk ash	29.3	303	55
	Zeolitic material	32.8	303	56
	Activated sewage sludge	60.0	298	57
	Pyrolysis of sewage sludge	92.8	303	56
	Nanocomposite hydrogels	370.4	298	58
	Magnetic NH2-MIL-101(Al)	135.0	298	This work

---

4. Conclusions

In summary, we have reported the adsorption behavior of magnetic amino-functionalised MOF for cationic and anionic dyes from aqueous solution in view of the adsorption isotherms, thermodynamics, kinetics, and the regeneration of the sorbent. The adsorption follows the pseudo-second-order kinetics model. The best-fit adsorption isotherm was achieved with the Langmuir model. Magnetic NH₂-MIL-101(Al), compared with other adsorbents, shows relatively high adsorption amounts for cationic and anionic dyes, probably mainly due to the electrostatic interaction; however, π–π stacking interaction and hydrogen bonding cannot be ruled out. The high adsorption capacity, good solvent stability, and excellent reusability mean that magnetic NH₂-MIL-101(Al) is promising as a novel adsorbent for the adsorption and removal of cationic and anionic dyes from aqueous solution.

Acknowledgements

This work was supported by Open Project of State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (No. ES201607) and the National Natural Science Foundation of China (No. 21205010).

References

1. S. B. Wang, Y. Boyjoo, A. Choueib and Z. H. Zhu, Water Res., 2005, 39, 129–138.
2. Q. Sun and L. Yang, Water Res., 2003, 37, 1535–1544.
3. A. N. Kulkaini, A. A. Kadam, M. S. Kachole and S. P. Govindwar, J. Hazard. Mater., 2014, 276, 461–468.
4. B. Y. Shi, G. H. Li, D. S. Wang, C. H. Feng and H. X. Tang, J. Hazard. Mater., 2007, 143, 567–574.
5. J. W. Lee, S. P. Choi, R. Thiruvenkatachari, W. G. Shim and H. Moon, Water Res., 2006, 40, 435–444.
6. D. Mahanta, G. Madras, S. Radhakrishnan and S. Patil, J. Phys. Chem. B, 2008, 112, 10153–10157.
7. W. X. Chen, W. Y. Lu, Y. Y. Yao and M. H. Xu, Environ. Sci. Technol., 2007, 41, 6240–6245.
8. D. S. Sun, X. D. Zhang, Y. Wu and X. Liu, J. Hazard. Mater., 2010, 181, 335–342.
9. S. H. Huo and X. P. Yan, J. Mater. Chem., 2012, 22, 7449–7455.
10. Z. L. Xie, X. Huang, M. M. Titirici and A. Taubert, RSC Adv., 2014, 4, 37423–37430.
11. H. Y. Zhao, Z. Jin, H. M. Su, J. L. Zhang, X. D. Yao, H. J. Zhao and G. S. Zhu, Chem. Commun., 2013, 49, 2780–2782.
12. E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown and J. R. Long, Science, 2012, 335, 1606–1610.
13. E. Haque, V. Lo, A. I. Minett, A. T. Harris and T. L. Church, J. Mater. Chem. A, 2014, 2, 193–203.
14. H. N. Wang, F. H. Liu, X. L. Wang, K. Z. Shao and Z. M. Su, J. Mater. Chem. A, 2013, 1, 13060–13063.
15. M. L. Ma, J. H. Qin, C. Ji, H. Xu, R. Wang, B. J. Li, S. Q. Zang, H. W. Hou and S. R. Batten, J. Mater. Chem. C, 2014, 2, 1085–1093.
16. W. Xuan, C. Zhu, Y. Liu and Y. Cui, Chem. Soc. Rev., 2012, 41, 1677–1695.
17. T. Wittmann, R. Siegel, N. Reimer, W. Milius, N. Stock and J. Senker, Chem.–Eur. J., 2015, 21, 314–323.
18. G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé and I. Margiolaki, Science, 2005, 309, 2040–2042.
19. S. Bernt, V. Guillerm, C. Serre and N. Stock, Chem. Commun., 2011, 47, 2838–2840.
20. C. Chen, M. Zhang, Q. X. Guan and W. Li, Chem. Eng. J., 2012, 183, 60–67.
21. A. Henschel, K. Gedrich, R. Kraehnert and S. Kaskel, Chem. Commun., 2008, 35, 4192–4194.
22. D. Y. Hong, Y. K. Hwang, C. Serre, G. Ferey and J. S. Chang, Adv. Funct. Mater., 2009, 19, 1537–1552.
23. P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J. F. Eubank, D. Heurtaux, P. Clayette, C. Kreuz, J. S. Chang, Y. K. Hwang, V. Marsaud, P. N. Bories, L. Cynober, S. Gil, G. Ferey, P. Couvreur and R. Gref, Nat. Mater., 2010, 9, 172–178.
24. K. M. L. Taylor-Pashow, J. Della Rocca, Z. G. Xie, S. Tran and W. B. Lin, J. Am. Chem. Soc., 2009, 131, 14261–14263.
25. H. Chen, C. Deng and X. Zhang, Angew. Chem., Int. Ed., 2010, 49, 607–611.
26. S. Mendoza-Bello, R. A. Morales-Luckie, L. Flores-Santos, J. P. Hinestroza and V. Sanchez-Mendieta, J. Nanopart. Res., 2012, 14, 1242–1251.
27. T. Tsoncheva, J. Roggenbuck, D. Paneva, M. Dimitrov, I. Mitov and M. Fröba, Appl. Surf. Sci., 2010, 257, 523–530.
28. Z. C. Kadirova, K. Katsumata, T. Isobe, N. Matsushita, A. Nakajima and K. Okada, Appl. Surf. Sci., 2013, 284, 72–79.
29. S. Jung, Y. Kim, S. Kim, T. H. Kwon, S. Huh and S. Park, Chem. Commun., 2011, 47, 2904–2906.
30. Z. Hasan, E. J. Choi and S. H. Jhung, Chem. Eng. J., 2013, 219, 537–544.
31. E. Stavitski, M. Goesten, J. Juan-Alcaniz, A. Martinez-Joaristi, P. Serra-Crespo, A. V. Petukhov, J. Gascon and F. Kapteijn, Angew. Chem., Int. Ed., 2011, 50, 9624–9628.
32. P. E. Serra-Crespo, V. Ramos-Fernandez, J. Gascon and F. Kapteijn, Chem. Mater., 2011, 23, 2565–2572.
33. I. Langmuir, J. Am. Chem. Soc., 1916, 38, 2221–2295.
34. H. M. F. Freundlich, J. Phys. Chem., 1906, 57, 1100–1107.
35. M. M. Dubinin, E. D. Zaverina and L. V. Radushkevich, J. Phys. Chem., 1947, 2, 1351–1362.
36. Y. F. Huang, Y. Q. Wang, Q. S. Zhao, Y. Li and J. M. Zhang, RSC Adv., 2014, 4, 47921–47924.
37. M. R. Malekbala, S. Hosseini, S. Kazemi-Yazdi, S. Masoudi Soltani and M. R. Malekbala, Chem. Eng. J., 2013, 229, 257–266.
38. T. Wang, K. Kailasam, P. Xiao, G. Chen, L. Chen, L. Wang, J. Li and J. Zhu, Microporous Mesoporous Mater., 2014, 187, 63–70.
39. K. Hall, L. Eagleton, A. Acrivos and T. Vermeulen, Ind. Eng. Chem. Fundam., 1966, 5, 212–223.
40. Y. Yu, B. N. Murthya, J. G. Shapter, K. T. Constantopoulos, N. H. Voelcker and A. V. Ellis, J. Hazard. Mater., 2013, 260, 330–338.
41. M. S. Chiou and H. Y. Li, J. Hazard. Mater., 2002, 93, 233–248.
42. M. Kousha, E. Daneshvar, M. Salar-Sohrabi, M. Jokar and A. Bhatnagar, Chem. Eng. J., 2012, 192, 67–76.
43. B. Sarkar, Y. Xi, M. Megharaj and R. Naidu, Appl. Clay Sci., 2011, 51, 370–374.
44. Y. S. Ho and G. McKay, Process Biochem., 1999, 34, 451–465.
45. G. Crini, H. N. Peindy, F. Gimbert and C. Robert, Sep. Purif. Technol., 2007, 53, 97–110.
46. I. D. Mall, V. C. Srivastava, N. K. Agarwal and I. M. Mishra, Colloids Surf., A, 2005, 264, 17–28.
47. Q. Chen, Q. He, M. Lv, Y. Xu, H. Yang, X. Liu and F. Wei, Appl. Surf. Sci., 2015, 327, 77–85.
48. B. Liu, F. Yang, Y. Zou and Y. Peng, J. Chem. Eng. Data, 2014, 59, 1476–1482.
49. I. Ahmed and S. H. Jhung, J. Hazard. Mater., 2015, 283, 544–550.
50. Y. Y. Jia, Q. Jin, Y. Li, Y. X. Sun, J. Z. Huo and X. J. Zhao, Anal. Methods, 2015, 7, 1463–1470.
51. B. H. Hameed and M. I. El-Khaiary, J. Hazard. Mater., 2008, 157, 344–351.
52. B. H. Hameed and M. I. El-Khaiary, J. Hazard. Mater., 2008, 154, 237–244.
53. R. P. Han, Y. Wang, Q. Sun, L. L. Wang, J. Y. Song, X. T. He and C. C. Dou, J. Hazard. Mater., 2010, 175, 1056–1061.
54. K. Yurdakoc, Z. Bekci, C. Ozveri and Y. Seki, J. Hazard. Mater., 2008, 154, 254–261.
55. U. R. Lakshmi, V. C. Srivastava, I. D. Mall and D. H. Lataye, J. Environ. Manage., 2009, 90, 710–720.
56. E. Gutiérrez-Segura, M. Solache-Rios and A. Colin-Cruz, J. Hazard. Mater., 2009, 170, 1227–1235.
57. M. Otero, F. Rozada, L. F. Calvo, A. I. García and A. Morán, Dyes Pigm., 2003, 57, 55–65.
58. M. Dalaran, S. Emik, G. Güçlü, T. B. İyim and S. Özgümüş, Desalination, 2011, 279, 170–182.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07567c

---