Facile synthesis of methyl propylaminopropanoate functionalized magnetic nanoparticles for removal of acid red 114 from aqueous solution

Abstract

Methyl propylaminopropanoate-coated Fe₃O₄@SiO₂ nanoparticles (Fe₃O₄@SiO₂–MPAP NPs) was prepared as a novel adsorbent and used for the removal of acid red 114 (AR-114) from aqueous solution through hydrogen bonding interactions between the sulfonate groups of the dye and the ester carbonyl groups of the adsorbent. Characterization of the obtained new adsorbent was achieved by FT-IR, XRD and TEM. According to the experimental results, about 95.8% of AR-114 was removed from aqueous solution at the adsorbent amount of 0.24 g L⁻¹ at pH = 2 in 60 min. The adsorption kinetics and isotherm results demonstrate that the kinetics and equilibrium adsorptions can be well-described by the pseudo-second-order kinetics and Langmuir models, respectively. Thermodynamics studies depicted that the adsorption of hazardous dye AR-114 was spontaneous and endothermic with randomness of the process. Furthermore, the Fe₃O₄@SiO₂–MPAP NPs could be simply recovered by an external magnet and it exhibited striking recyclability and reusability for five cycles. Such functional nanoparticles are appropriate adsorbents for removal of dyes containing sulfonate groups from aqueous solutions.

---

1. Introduction

Synthetic dyes, with complex aromatic structures, are present in wastewater streams from many industries including textile, leather, food, cosmetics, electronics, paper, rubber, plastic, printing and pharmaceutical.^1–3 These dyes and their degradation species are toxic in nature and have potentially carcinogenic and mutagenic effects on human beings, micro-organisms and aquatic life.^4–10 A wide range of physical and chemical procedures such as electrochemical degradation, ozonation, oxidation, photocatalysis, ion-exchange, membrane separation, adsorption and biological process have been developed for the treatment of industrial wastewater containing dyes before they are delivered to the environment.^11–30 Among these possible techniques for water treatment, adsorption has been considered as a more feasible process for most pollutant removal due to its low cost, simplicity of design, ease of operation without harmful residues and the possibility of regeneration of the adsorbent.^31–33 Currently, a variety of potential adsorbents have been employed for the removal a large amount of molecule dyes from water.^34–36

Magnetic nanoparticles (M-NPs) have received considerable attention as a new adsorbent with large surface area and small diffusion resistance for the adsorb contaminants such as metals, dyes and gases.^37–40 Furthermore, M-NPs can be separated easier and faster with an external magnet without centrifugation or filtration.⁴¹ In the current study, for the first time the use of Fe₃O₄@SiO₂ coated with new ester-terminal ligands as a sorbent for removal of acid red 114 (AR-114) was studied. Recently, ester linkages have been demonstrated for effective removal of phenolic compounds from wastewater mainly through the hydrogen bonding interaction between the hydroxyl group of phenolic compounds and the carbonyl group of ester moiety.^42,43 Hydrogen bonds are principle noncovalent interactions to construct supramolecular nanoporous architectures since they are highly selective and directional.⁴⁴ In addition, the hydrogen bonded materials have appealing properties and have promise in fields such as filtration, separation, adsorption, ion conductivity, enantioselective separation and molecular recognition.⁴⁵

Acid red 114 (AR-114), as a bis-azo biphenyl dye, with carcinogenic effects has been widely used in large quantities in dyeing of woven fabrics and wool and cotton textiles and cause many environmental troubles.^46,47 The azo dyes with an aromatic structure, are resistant to aerobic digestion, and are stable to oxidizing agents. The aim of this study is to modify the surface of Fe₃O₄@SiO₂ NPs with methyl propylaminopropanoate (MPAP) as a new magnetic adsorbent for significant removal of AR-114 from aqueous solutions through the hydrogen bonding interaction between the sulfonate groups of AR-114 and the ester carbonyl groups of adsorbent. The effect of principle factors including pH, contact time, the sorbent amount and dye concentration were studied. The kinetic and thermodynamic parameters of adsorption were also analyzed. The results obtained could be useful in further application of modified NPs, and effective way for treatment of dye containing wastewaters.

2. Results and discussion

In continuation to our prior work,^9,10,48 herein, the modified magnetic NPs Fe₃O₄@SiO₂–MPAP was synthesized by the steps shown in Scheme 1. Firstly, Fe₃O₄ NPs were prepared by co-precipitation of Fe²⁺ and Fe³⁺ ions in basic solution⁴⁹ and then in order to avoid possible aggregation or oxidation of the Fe₃O₄ NPs surfaces, a layer of SiO₂ using sol–gel process, through the hydrolysis of TEOS was coated on Fe₃O₄ NPs surfaces and successfully, Fe₃O₄@SiO₂ core–shell microspheres were prepared.⁵⁰ Subsequently, Fe₃O₄@SiO₂ core–shell was reacted with 3-aminopropyltriethoxysilane (APTES) as a spacer to obtain amino-functionalized Fe₃O₄@SiO₂–APTES.⁵⁰ Finally, methyl propylaminopropanoate-modified Fe₃O₄@SiO₂ (Fe₃O₄@SiO₂–MPAP) was prepared by the reaction of methyl acrylate (MA) with aminopropyl grafted Fe₃O₄@SiO₂ during the double Michael addition in EtOH solution and the success of this immobilization was monitored with FT-IR, XRD and TEM.

Scheme 1 (a) Preparation of Fe₃O₄@SiO₂–MPAP NPs and (b) main interactions between Fe₃O₄@SiO₂–MPAP NPs and AR-114.

2.1. Fourier transform infrared spectroscopy

The FT-IR spectra of Fe₃O₄ (black line), Fe₃O₄@SiO₂ (violet line), Fe₃O₄@SiO₂–APTES (red line) and Fe₃O₄@SiO₂–MPAP (blue line) are shown in Fig. 1. The typical absorption peak for Fe₃O₄ at 570 cm⁻¹ is related to the stretching vibration of Fe–O bond and two bonds at 3423 and 1633 cm⁻¹ are attributed to the symmetric and asymmetric stretch and bending vibrations of O–H, respectively which is attached to the surface iron atoms. For Fe₃O₄@SiO₂, the strong broad band at 1089 cm⁻¹ with a shoulder at 1209 cm⁻¹ are assigned to the asymmetric stretching bonds of Si–O–Si in SiO₂. The weaker bands at 796, 466 and 952 cm⁻¹ correspond to the Si–O–Si symmetric stretch, Si–O–Si or O–Si–O bending modes and Si–O symmetric stretch, respectively. The band at 567 cm⁻¹ is an indication of the presence of Si–O–Fe. In the spectrum of Fe₃O₄@SiO₂–APTES, the characteristic absorptions band at 1000, 3417 and 2869–2906 cm⁻¹ are corresponded to C–N, N–H and C–H stretching modes of the alkyl chain, respectively and the N–H bending mode appeared at 1627 cm⁻¹ that confirm the successful coating of APTES on Fe₃O₄@SiO₂. In the FT-IR spectrum of the Fe₃O₄@SiO₂–MPAP, the absorption band at 1722 and 1012 cm⁻¹ was attributed to the C=O and C–O stretching of the ester groups. Also, the absence of the N–H stretching and bending modes in the spectrum, clearly show that the Michael additions proceed to completion and two ester groups have been successfully coated on the surface of Fe₃O₄@SiO₂ NPs.

Fig. 1 The FT-IR spectra of Fe₃O₄ (black line), Fe₃O₄@SiO₂ (violet line), Fe₃O₄@SiO₂–APTES (red line) and Fe₃O₄@SiO₂–MPAP (blue line).

2.2. X-ray diffraction

The X-ray diffraction (XRD) patterns of Fe₃O₄ (black line), Fe₃O₄@SiO₂ (red line), Fe₃O₄@SiO₂–APTES (blue line) and Fe₃O₄@SiO₂–MPAP (green line) are shown in Fig. 2. The sharp diffraction peaks with 2θ at 30.3°, 35.6°, 43.3°, 57.2°, and 62.5° are observed, which indicate that the Fe₃O₄ particles have highly crystalline cubic spinel structure of the magnetite and the same sets of characteristic peaks were also observed for Fe₃O₄@SiO₂, Fe₃O₄@SiO₂–APTES and Fe₃O₄@SiO₂–MPAP indicating the stability of the crystalline phase of Fe₃O₄ NPs during silica coating and surface amino- and ester-functionalization.⁵¹

Fig. 2 The XRD patterns of Fe₃O₄ (black line), Fe₃O₄@SiO₂ (red line), Fe₃O₄@SiO₂–APTES (blue line) and Fe₃O₄@SiO₂–MPAP (green line).

The average crystallite size D was determined by the Scherrer formula, D = Kλ/(β cos θ), where λ = 1.54 Å is the wavelength of Cu-K_α radiation used, β is the full width at half-maximum (FWHM) intensity of the diffraction line, θ is the Bragg angle for the measured hkl peak and K is a constant equal to 0.94. The particle sizes of the magnetite calculated using the Scherrer equation were 12.1, 13.8, 15.7 and 18.0 nm for Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂–APTES and Fe₃O₄@SiO₂–MPAP, respectively.⁵² These values are in good agreement with the TEM images.

2.3. Transmission electron microscopy image

Fig. 3a–d show the transmission electron microscopy (TEM) images of the Fe₃O₄@SiO₂–MPAP and Fe₃O₄ microspheres. Fig. 3a–c show that the MPAP (with light color) is immobilized on the surface of Fe₃O₄@SiO₂ NPs (with dark color), obviously. The pure Fe₃O₄ NPs (Fig. 3d) appear to be almost spherical in shape with a diameter range about <15 nm. The TEM analysis demonstrates that the presence of SiO₂–MPAP shell increases the size of magnetite NPs.³⁰ The synthesized Fe₃O₄@SiO₂–MPAP NPs showed a spherical shape with an average diameter of about <30 nm, however, the NPs tended to aggregate to large particle.

Fig. 3 The TEM micrographs of synthesized (a and b) Fe₃O₄@SiO₂–MPAP NPs, (c) Fe₃O₄@SiO₂–MPAP core–shell and (d) Fe₃O₄ NPs.

2.4. Effect of pH on adsorption

The pH of the aqueous solution is the very crucial parameter for its effect on the adsorption property of adsorbent and adsorbate. The effect of initial pH on the AR-114 adsorption by Fe₃O₄@SiO₂–MPAP surfaces at different values, ranging 2–11 and with a constant time 120 min is shown in Fig. 4. The initial concentration of dye and adsorbent dosage were kept fixed at 20 mg L⁻¹ and 0.2 g L⁻¹, respectively. As was expected, the results demonstrated that the uptake amount strongly depends on solution pH. Maximum removal efficiency of AR-114 onto Fe₃O₄@SiO₂–MPAP was observed at the initial pH = 2 (86.25%), and decreased dramatically when solution pH increased and turned to alkaline, which could be due to the ionization of sulfonic groups of AR-114 at pH above the pK_a. The pK_a value of benzenesulfonic acid is 2.1 at 298 K and change to sulfonate anion at pH > pK_a with hydrophilic nature. Therefore, sulfonate groups of AR-114 existed as molecular form (sulfonic form) in the range of pH < pK_a and adsorbed effectively onto the Fe₃O₄@SiO₂–MPAP surface through van der Waals interactions. In addition, the intermolecular hydrogen bonds occurred between hydroxyl groups of adsorbate and the carbonyl and methoxy groups of adsorbent as an adsorption driving force to catch the AR-114 molecules from water (Scheme 1b).⁴² Therefore, pH = 2 was chosen as the optimum pH for all further adsorption experiments. In order to obtain information about the surface charge of the adsorbent, the point of zero charge (pH_ZPC) was determined. The plot of pH difference versus initial pH of the solutions is illustrated in Fig. 5, which shown the point of zero charge is around 6. This means that at pH values below 6, the Fe₃O₄@SiO₂–MPAP surface has a net positive charge, while at pH greater than 6, the surface has a net negative charge and the electrostatic repulsive force exists between adsorbent and adsorbate. Hence, the acidic pH facilitate the adsorption of AR-114 onto Fe₃O₄@SiO₂–MPAP surface and maximum removal percentage and adsorption capacity take place at pH = 2.

Fig. 4 The effect of pH on the adsorption of AR-114 on Fe₃O₄@SiO₂–MPAP in different time interval ([AR-114]₀ = 20 mg L⁻¹ and adsorbent dosage = 0.2 g L⁻¹).

Fig. 5 Determination of the pH of point of zero charge (pH_ZPC).

2.5. Effect of adsorbent dose on adsorption

The adsorption of dye on Fe₃O₄@SiO₂–MPAP was investigated by changing the quantity of adsorbent range of 0.1–0.3 g L⁻¹, with the dye concentration of 20 mg L⁻¹, room temperature (25 ± 1 °C) and pH of 2.0 for different time intervals (0–120 min). The results in Fig. 6, show that with the increase in adsorbent dosage from 0.1–0.3 g L⁻¹, the percentage adsorption increases from 53.71 to 97.34% over the entire contact time due to increased active sites. Furthermore, the rate of removal of AR-114 at all dosages is primarily rapid in the first stage of contact time and then it is gradually slowed until reactions reach equilibrium at about 60 min. The rapid adsorption at the initial contact time is attributed to the large quantity of free active sites on the Fe₃O₄@SiO₂–MPAP surface and convenient accessibility of them for AR-114 molecules.⁵³ Since the removal efficiency of AR-114 does not significant difference between dosage 0.24 and 0.3 g L⁻¹, further bath adsorption experiments were carried out at 0.24 g L⁻¹. In fact, the removal efficiency enhanced from 55.24 to 95.75% by increasing the contact time from 0.5 to 60 min at pH = 2 and adsorbent dosage equal to 0.24 g L⁻¹.

Fig. 6 The effect of adsorbent dose on the adsorption of AR-114 on Fe₃O₄@SiO₂–MPAP in different time interval ([AR-114]₀ = 20 mg L⁻¹ and pH = 2).

2.6. Effect of contact time on adsorption

To know the equilibration time for maximum adsorption and the kinetics of the adsorption process, the adsorption of AR-114 on the Fe₃O₄@SiO₂–MPAP surface was studied as a function contact time as shows in Fig. 7. The sorption of AR-114 increases fast in the first and almost 90% adsorption was completed within 10 min on the Fe₃O₄@SiO₂–MPAP surface, and then slowed down until the sorption process reaches equilibrium after 60 min. In general, the removal process is quite fast and 30 min is enough to achieve equilibrium. In this section, the concentration of AR-114 was 20 mg L⁻¹, pH = 2, temperature 298 K and adsorbent dosage 0.24 g L⁻¹. The dye adsorption processes are shown in the inset of Fig. 7.

Fig. 7 The effect of contact time on the adsorption of AR-114 on Fe₃O₄@SiO₂–MPAP in different contact time interval. The inset figures show photographs of aqueous solutions of AR-114 before (0 min vortex) and after adsorption (2–15 min vortex). ([AR-114]₀ = 20 mg L⁻¹, adsorbent dosage = 0.24 g L⁻¹ and pH = 2).

2.7. Effect of initial dye concentration on adsorption

The adsorbate initial concentration acts as an important driving force to overcome the mass transfer resistance of dye between the aqueous and the solid phases.⁵⁴ The effects of initial dye concentrations on the rate of adsorption by Fe₃O₄@SiO₂–MPAP were studied in the range from 10 to 100 mg L⁻¹ (Fig. 8). By increasing the initial dye concentration the percentage of dye removal decreased, these results suggest that the vacant sites on the adsorbent are saturated by dye molecules.

Fig. 8 The effect of initial dye concentrations on the adsorption of AR-114 on Fe₃O₄@SiO₂–MPAP in different contact time interval (adsorbent dosage = 0.24 g L⁻¹ and pH = 2).

2.8. Adsorption kinetics

Adsorption kinetic experiments were analyzed at different AR-114 concentration = 10, 20, 40, 60, 80 and 100 mg L⁻¹, constant adsorbent dosage = 0.24 g L⁻¹ and pH = 2. The pseudo-first-order, pseudo-second-order and intra-particle-diffusion models were applied in order to find an efficient model for the description of adsorption mechanism. The pseudo-first-order kinetic model is expressed as follow:⁵⁵

ln(q_e − q_t) = ln(q_e) − k₁t (1)

where q_e and q_t are the amount of dye adsorbed at equilibrium (mg g⁻¹) and amount of dye adsorbed at time t (mg g⁻¹), respectively; k₁ (min⁻¹) is the equilibrium rate constants of the pseudo-first-order adsorption. Using this equation, the values of k₁ and q_e were calculated from the slope and intercept of the plot of ln(q_e − q_t) versus t, respectively.⁵⁶

The pseudo-second-order kinetic model can be expressed as follow:

where k₂ is the equilibrium rate constant of pseudo-second-order adsorption (g mg⁻¹ min⁻¹). The slope and intercept of the plot of versus t were used to calculate the second-order rate constant k₂.

The possibility of intra-particle diffusion resistance affecting adsorption was investigated using the intra-particle diffusion model as:⁵⁷

q_t = k_pt^1/2 + C (3)

where k_p (mg g⁻¹ min^−1/2) is the intra-particle diffusion rate constant which can be evaluated from the slope of the linear plot of q_t versus t^1/2, and C (mg g⁻¹) is intercept. Values of C give an idea about the thickness of the boundary layer: the larger the intercept, the greater the boundary layer effect (Fig. 9).

Fig. 9 The linear plots of pseudo-second-order model on the AR-114 removal by Fe₃O₄@SiO₂–MPAP in different time and concentration (pH = 2 and adsorbent dosage = 0.24 g L⁻¹).

The best-fit model was selected based on the linear regression correlation coefficient R² values. The kinetic parameters for the removal AR-114 at different initial concentrations by three models are summarized in Table 1. The results show that first-order kinetic and intra-particle diffusion models are not suitable for the present adsorption system due to low correlation coefficients R². For the first-order kinetic model very difference exists between q_e,exp and q_e,cal that demonstrate a poor pseudo-first-order fit to the experimental data. Therefore, it is necessary to fit the experimental data to another model. The kinetic data for AR-114 adsorption showed the best fitting (R² = 0.9999) with the pseudo-second-order model and the calculated q_e values also agree very well with the experimental data. Moreover, when the initial AR-114 concentration increased from 10 to 100 mg L⁻¹, the value of k₂ (g mg⁻¹ min⁻¹) and R² for the pseudo-second-order model were decreased from 0.062 to 0.0071 g mg⁻¹ min⁻¹ and 0.9998 to 0.9996, respectively. Also, q_e,cal (mg g⁻¹) increased from 46.08 to 108.7 mg g⁻¹. This result indicated that adsorption data were in agreement with this model. Furthermore, the curve-fitting plots of pseudo-second-order kinetic model gave a straight line, passed through the origin and confirmed the best fit of this kinetic model in the present system (Fig. 9). Based on the kinetic model obtained, a chemisorptions bond takes place during adsorption of AR-114 on the Fe₃O₄@SiO₂–MPAP NPs surface. For the intra-particle diffusion model the value of C was calculated as 67.162 mg g⁻¹ (C ≠ 0), indicating that intra-particle diffusion is not the only rate-limiting step for AR-114 adsorption and the sorption process is rather complex and involves more than one diffusive resistance.

Table 1 Kinetic parameters for AR-114 adsorption onto Fe₃O₄@SiO₂-MPA

Kinetic models	[AR-114]0 (mg L^−1)
	10	20	40	60	80	100
qe,exp (mg g^−1)	46.5	83.71	104.08	102.69	98.76	108.71
Pseudo-first-order
k1 (min^−1)	0.0144	0.0303	0.0382	0.0656	0.0397	0.0309
qe,cal (mg g^−1)	14.17	12.33	6.03	5.79	4.73	6.45
R^2	0.4092	0.369	0.5013	0.6382	0.5988	0.3083
Pseudo-second-order
k2 (g mg^−1 min^−1)	0.062	0.0253	0.0099	0.0097	0.0081	0.0071
qe,cal (mg g^−1)	46.08	83.33	103.09	104.17	99.01	108.7
R^2	0.9998	0.9999	0.9981	0.9999	0.9995	0.9996
Intra-particle diffusion
kp (mg g^−1 min^−1/2)	0.8195	1.9076	3.1404	4.2669	3.4093	4.1794
C0 (mg g^−1)	39.006	67.162	75.635	66.697	68.336	71.158
R^2	0.4595	0.5183	0.4704	0.6133	0.6016	0.7457

---

2.9. Adsorption isotherm studies

The evaluation of adsorption isotherms are important for developing a model for interaction of adsorbent–adsorbate and provide inclusive information about the nature of interactions.^58,59 The adsorption isotherms of AR-114 on the Fe₃O₄@SiO₂–MPAP with 20 mg L⁻¹ AR-114 as an initial concentration using various adsorbent dosages (0.04–0.24 g L⁻¹) at pH = 2 for 72 h at different temperatures are given in Fig. 10, and the equilibrium adsorption data were evaluated according to the renowned models Langmuir and Freundlich isotherms. The Langmuir isotherm supposes that monolayer adsorption take place at binding sites with homogenous energy levels, without interactions between adsorbed molecules and transmigration of adsorbed molecules onto adsorption surface. The Langmuir equations can be expressed as:⁶⁰orwhere C_e is the equilibrium concentration of the AR-114 solution (mg L⁻¹), q_e is the adsorption capacity at equilibrium (mg g⁻¹), k_l is the constant related to free energy of adsorption (L mg⁻¹), and q_m is the maximum adsorption capacity at monolayer coverage (mg g⁻¹).

Fig. 10 The (a) Langmuir isotherm and (b) Freundlich isotherm on the AR-114 removal by Fe₃O₄@SiO₂–MPAP at different temperatures (pH = 2, [AR-114]₀ = 20 mg L⁻¹, adsorbent dosage = 0.04–0.24 g L⁻¹ and time = 72 h).

The Freundlich isotherm is a practical equation that supposes heterogeneous adsorbent surface with its adsorption sites at changeable energy levels.⁶¹ The corresponding equations are commonly represented by:

q_e = k_fC_e^1/n (6)

ork_f (mg¹⁺ⁿ g⁻¹ L⁻ⁿ) and n are the Freundlich constants characteristics of the system, indicating the adsorption capacity and the adsorption intensity, respectively. If the value of 1/n is lower than 1, it indicates a normal Freundlich isotherm; if not, it is indicative of cooperative adsorption.⁶²

The calculated isotherm parameters from both models are shown in Table 2. The adsorption of AR-114 was fit to the Langmuir isotherm model better than Freundlich with the higher R² (0.9913). It indicates that the adsorption occurred at specific homogeneous sites within the adsorbent forming monolayer coating of AR-114 at the surface of the absorbent. The Freundlich constant 1/n was smaller than 1, indicating a high adsorption intensity (Table 2). Table 1, also shows that the maximum monolayer adsorption capacity (q_m) of AR-114 by Fe₃O₄@SiO₂–MPAP was in 293 K and decreased above it. In this temperature, interaction between the AR-114 and Fe₃O₄@SiO₂–MPAP is probably larger due to the presence of a higher number of potential adsorption sites. However, at high temperatures, the interaction between AR-114 and Fe₃O₄@SiO₂–MPAP is weakened due to decreased of hydrogen bonds and van der Waals interactions.⁶³

Table 2 Adsorption isotherm parameters for AR-114 adsorption on the Fe₃O₄@SiO₂–MPAP

T (K)	Langmuir isotherm				Freundlich isotherm
	qm (mg g^−1)	k1 (L mg^−1)	R^2	RL	kf (mg^1+n g^−1 L^−n)	n	R^2
293	105.2632	1.2025	0.9913	0.0399	71.1377	7.5075	0.7108
308	70.4225	1.5778	0.9839	0.0307	88.5523	25.1889	0.1927
318	75.7576	2.129	0.9741	0.023	63.6502	18.4162	0.1815

---

The adsorption isotherm process favorability was also evaluated using the dimensionless separation factor (R_L) that were calculated using the following equation:⁶⁴

The adsorption process can be defined as favorable (0 < R_L < 1), unfavorable (1 < R_L), linear (R_L = 1) and irreversible in nature (R_L = 0).⁶⁵ In this study, the value of R_L calculated for the adsorption of AR-114 by Fe₃O₄@SiO₂–MPAP fall between 0 and 1 (R_L = 0.0399), therefore, the adsorption of AR-114 onto the adsorbent appears to be a favorable process.

2.10. Thermodynamic studies

Thermodynamics parameters such as standard enthalpy change (ΔH°), standard Gibbs free energy change (ΔG°) and standard entropy change (ΔS) were estimated to check the temperature effects on the adsorption of AR-114 on the Fe₃O₄@SiO₂–MPAP NPs. The thermodynamic parameters can be evaluated with the Van't Hoff equations:⁶⁶

ΔG° = −RT ln K (9)

ΔG° = ΔH° + TΔS° (11)

where K is adsorption equilibrium constant (L mol⁻¹), R the universal gas constant (8.314 J mol⁻¹ K⁻¹) and T the absolute temperature (K).

Thermodynamic experiments were performed at different temperature from 293 to 318 K at constant adsorbent dosage (0.24 g L⁻¹) and at pH = 2. The values of ΔH° and ΔS° was determined by plotting ln K versus 1/T (Fig. 11), and the ΔG° values were obtained from eqn (11). Table 3 presents the thermodynamic results at various temperatures. The positive value of ΔH° indicates that the adsorption is endothermic in nature. The positive value of ΔS° shows the higher degree of disorder at the solid–solution interface during the adsorption process. Furthermore, the negative value ΔG° confirms the spontaneous nature of sorption and affinity of Fe₃O₄@SiO₂–MPAP for the AR-114 dye and the increase of these values with the temperature increasing indicate a higher adsorption with increase in the temperatures. Since the reaction process is endothermic, this can be attributed to increased surface coverage at higher temperature, expansion and creation of reactive and active sites.

Fig. 11 The Van't Hoff plot for ln K vs. 1/T ([AR-114]₀ = 20 mg L⁻¹, adsorbent dosage = 0.24 g L⁻¹ and pH = 2).

Table 3 Thermodynamics parameters of AR-114 adsorption on the Fe₃O₄@SiO₂–MPAP

ΔH° (kJ mol^−1)	ΔS° (J mol^−1 K^−1)	ΔG° (kJ mol^−1)
		293 K	308 K	318 K
250.31	60.42	−0.45	−1.17	−2

---

2.11. Desorption and recycling studies

Several (adsorption–desorption) regeneration cycles with Fe₃O₄@SiO₂–MPAP NPs were performed by 20 mL of NaOH 0.01 M solution, as shown in Fig. 12. In fact, the AR-114 molecules on the surface of Fe₃O₄@SiO₂–MPAP NPs could be replaced with hydroxyl ions of the basic solution in adsorbent washing step. The result showed the recovery values were not significantly decreased and reused for at least five successive removal processes with removal efficiency higher than 80% that display the stability of adsorbent. Fig. 12d shows the TEM image of the Fe₃O₄@SiO₂–MPAP after five adsorption–desorption cycles that indicate the Fe₃O₄@SiO₂–MPAP NPs was not significantly altered during the regeneration process. Also, after five cycles of the desorption–adsorption process, the high magnetic sensitivity of Fe₃O₄@SiO₂–MPAP NPs still retained and was collected from the solution using a magnet of 1.4 T (right inset in Fig. 12a–c). Therefore, the Fe₃O₄@SiO₂–MPAP NPs can be potentially used as a magnetic adsorbent for further dye adsorption from water.

Fig. 12 The removal efficiency during five cycles of Fe₃O₄@SiO₂–MPAP. The inset figures show aqueous solutions of AR-114 (a) before adsorption, (b) after adsorption, (c) after five cycles of desorption–adsorption and (d) the TEM image of the Fe₃O₄@SiO₂–MPAP after five adsorption–desorption cycles.

A detailed understanding of dye mineralization by bio- or chemical degradation is required for the conversion of recovered AR-114 to inorganic compounds such as NO₃, CO₂ and H₂O.^67–69 Experiments are planned to further examine the pathways in the complete detoxification of this recovered dye.

2.12. A comparison with other adsorbents

A comparison between results of the performance of Fe₃O₄@SiO₂–MPAP with other adsorbents on removal of dyes in the previous studies is presented in Table 4.^24,70–81 As the results show, the absorption capacity of Fe₃O₄@SiO₂–MPAP NPs due to hydrogen bond interaction could be comparable or even better than the other adsorbent. In Fe₃O₄@SiO₂–MPAP NPs compared to Fe₃O₄ NPs (entry 18), a thin and dense SiO₂ layer along with a desired thickness of MPAP shell can protect the iron oxide core from leaching out under acidic conditions.⁸² Hence, the stability of Fe₃O₄@SiO₂–MPAP NPs increases. It is noticeable that the higher adsorption capacity of SBA-3/PEHA (entry 9) is due to electrostatic interaction and hydrogen bond formation between the surface of the adsorbent and AR-114, simultaneously.²⁴ Additionally, the adsorption capacity of Fe₃O₄@SiO₂–NH₂ as a result of electrostatic interaction is lower than Fe₃O₄@SiO₂–MPAP.⁷² Also, the adsorption capacity of Fe₃O₄@SiO₂ on removal of Congo red (CR) as a AR-114 derivative is lower than Fe₃O₄@SiO₂–MPAP NPs (entry 19).⁸⁰ These results demonstrate that the hydrogen bond formation plays more important function in AR-114 adsorption than electrostatic interaction. Therefore, Fe₃O₄@SiO₂–MPAP NPs could be considered as an alternative and efficient adsorbent for removing organic compounds with hydrogen bonding ability such as amino-, hydroxyl- and sulfonate-substituted dyes, phenols and amines because of their high adsorption capacities. Furthermore, the magnetic properties of Fe₃O₄@SiO₂–MPAP make it more efficient adsorbent for the removal of contaminant from aqueous solution.

Table 4 Comparison of various adsorbents for the removal of dye

Adsorbents	Dye	qm (mg g^−1)	Adsorbent dosage (g L^−1)	Concentration (mg L^−1)	Best fit isotherm	Ref.
a Acid-activated Eichornia crassipes (an activated plant biomass).b Kattamanakku tree leaf powder carbon.c Ceiba pentradenta wood waste-activated carbon@H3PO4.d Ipomea carnea stem waste@phosphoric acid.e Activated pongam seed shells.f Activated cotton seed shells.g Activated sesame seed shells.h Pentaethylene hexamine functionalized SBA-3.i Magnetite/silica/pectin nanoparticles.j Methyl orange.k Eriochrome black T.l Magnetite/pectin nanoparticles.m Chitosan coated magnetic mesoporous silica nanoparticles.n Methylene blue.o Cetyltrimethylammonium bromide coated magnetite nanoparticles.p Disperse red 167.q Disperse blue 183.
Activated carbon	AR-114	103.73	0.25	250	Toth	70
AAEC^a	AR-114	112.34	1.25	100	Langmuir	71
KTC^b	AR-114	450.02	2	40	Freundlich	72
CPAC^c	AR-114	69.45	0.10	20	Langmuir	73
ICAC^d	AR-114	80.86	0.10	20	Langmuir	74
Activated carbon-Pp^e	AR-114	204.08	3	100	Langmuir & Freundlich	75
Activated carbon-Cp^f	AR-114	153.85	3	100	Langmuir & Freundlich	75
Activated carbon-Sp^g	AR-114	102.04	3	100	Langmuir & Freundlich	75
SBA-3/PEHA^h	AR-114	1000	0.20	100	Freundlich	24
MSP NPs^i	MO^j	26.75	2	100	Freundlich & Sips	30
MSP NPs	EBT^k	65.35	2	100	Sips	30
MP NPs^l	MO	27.22	2	100	Freundlich	30
MP NPs	EBT	72.35	2	100	Freundlich	30
CMMSNs^m	MB^n	43.03	0.2	20	Freundlich	76
CTAB–Fe3O4 NPs^o	DR-167^p	2000	—	80	—	77
CTAB–Fe3O4 NPs	DB-183^q	2000	—	80	—	77
Fe3O4@SiO2	MB	—	1	20	—	78
Fe3O4	AR-114	111	1	100	Freundlich	79
Fe3O4@SiO2	CR	50.54	3	30	Langmuir	80
Fe3O4@SiO2–NH2	AR-114	84.75	0.40	20	Langmuir	81
Fe3O4@SiO2–MPAP	AR-114	105.26	0.24	20	Langmuir	This work

---

3. Experimental

3.1. Materials and apparatus

The following reagents were purchased from Merck and used without further modification: ferric chloride hexahydrate (FeCl₃·6H₂O), ferrous chloride tetrahydrate (FeCl₂·4H₂O), ammonium hydroxide (NH₃, 25 wt%), tetraethylorthosilicate (TEOS), 3-aminopropyl-trimethoxysilane (APTES), and methyl acrylate. Acid red 114 (AR-114) was purchased from Sigma-Aldrich. A pH meter (Metrohm, model 713, Swiss) was used for pH measurements. The UV-Vis absorption spectra in the range of 200–700 nm were measured with a Shimadzu UV-2100 spectrometer. Spectral data were obtained using quartz cuvettes with 1 cm optical path length. The FT-IR spectra for the samples were obtained using Shimadzu FT-IR-8900 spectrophotometer by using KBr pellets. The X-ray diffraction (XRD) patterns were recorded in air at ambient temperature by Phillips (pw-1840) X-ray diffractometer with Cu-Kα radiation source (λ = 1.54056 Å) at 40 kV voltage and 25 mA current in a wide angle range (2θ = 20–70). The TEM samples were prepared by suspending the nanoparticles in EtOH followed by sonication for several minutes. One drop of the dilute nanoparticles/EtOH suspension was placed on a carbon-coated holey TEM copper grid and was dried in air. The dried grid was then loaded into a double tilt sample holder. The sample was then examined with a Philips CM-20 STEM equipped with a Gatan UltraScan 1000 CCD camera and an energy dispersive X-ray spectrometer: INCA Energy TEM 200. TEM images were taken at 200 kV.

3.2. Preparation of magnetic nanoparticles of Fe₃O₄@SiO₂–MPAP

Fe₃O₄@SiO₂–MPAP NPs were synthesized in the following steps (Scheme 1): first, Fe₃O₄ nanoparticles were synthesized by co-precipitation process of a mixture of the FeCl₃·6H₂O and FeCl₂·4H₂O (molar ratio 1 : 2) according to the previous work,⁴⁹ and then were coated with silica layers via a sol–gel method.⁵⁰ Subsequently, aminopropyl modified silica coated Fe₃O₄ magnetic nanoparticles (Fe₃O₄@SiO₂–APTES) was synthesized according to the procedure reported in the previous literature.⁵⁰ The obtained Fe₃O₄@SiO₂–APTES (6.9 mmol, 6 g) was suspended in dry EtOH (300 mL) and stirred heavily for 30 min under N₂ gas. Then, a solution of methyl acrylate (140 mmol, 11.82 g) in dry EtOH (100 mL) was added drop-wise via a dropping funnel under mechanical stirring. The reaction mixture was vigorously stirred at 50 °C for 5 days at N₂ atmosphere. The resulted mixture was cooled and the solid was separated magnetically and then washed with absolute EtOH several times (3 × 50 mL). The precipitate was then dried in oven at 50 °C for 8 h and the black nanopowder Fe₃O₄@SiO₂–MPAP was obtained.

3.3. Adsorption studies

The adsorptive removal of AR-114 was carried by batch experiments. Tests were conducted in 500 mL conical flasks containing 20 mg L⁻¹ AR-114 solution in a water bath to elucidate the values of the test parameters including solution pH (2–11), dye concentration (10–100 mg L⁻¹), temperature (298 K), contact time (0–120 min) and absorbent dosage (0.1–0.3 g L⁻¹). The pH of the solution was adjusted by dropwise addition of 0.1 M HCl or 0.1 M NaOH. After each removal condition experiments, the samples were separated from the dye solution with a permanent magnet and the residual dye molecules concentrations in the solution were determined by UV-Vis spectrophotometer at 515 nm for AR-114.

3.4. Desorption and reusability experiments

Desorption was investigated separately for 20 mg L⁻¹ AR-114 in 500 mL. The sorbent was rinsed with 20 mL of 0.01 M NaOH in H₂O that was sufficient for complete desorption of the dye. In order to increase the recoveries; 10 min vortex was used after each washing step. The UV-Vis test of the upper layer solution after vortex was applied to make sure all the AR-114 was recovered. The concentration of each eluents was measured using the obtained standard curve from spectrophotometry method. For reusability test, the sorbent was washed 5 times with 20 mL of 0.01 NaOH aqueous solutions (each step with 4 mL solvent and 10 min vortex) after dye extraction. Then, the sorbent was rinsed with H₂O to remove the NaOH excess and dried at 40 °C in an oven.

4. Conclusions

Here, for the first time diester-modified Fe₃O₄@SiO₂–MPAP NPs have been successfully prepared by surface modification of Fe₃O₄@SiO₂ with methyl acrylate and used as a new adsorbent for the removal of AR-114 from aqueous solution. The adsorption kinetics, isotherms and thermodynamics were examined in detail. The adsorption followed pseudo-second-order kinetics. The equilibrium data fitted well the Langmuir isotherm and thermodynamics studies depicted that the adsorption of hazardous dye AR-114 was spontaneous and endothermic with randomness of the process. The experimental results showed that the Fe₃O₄@SiO₂–MPAP NPs could be utilized as a promising and efficient adsorbent for the environmental cleanup.

Acknowledgements

This study was supported in part by the Research Committee of University of Guilan & Iran Nanotechnology Initiative Council (INIC).

Notes and references

1. V. S. Mane, I. D. Mall and V. C. Srivastava, Dyes Pigm., 2007, 73, 269.
2. J. W. Lee, S. P. Choi, R. Thiruvenkatachari, W. G. Shim and H. Moon, Dyes Pigm., 2006, 69, 196.
3. M. Akgül, J. Hazard. Mater., 2014, 267, 1.
4. O. J. Hao, H. Kim and P. C. Chiang, Crit. Rev. Environ. Sci. Technol., 2000, 30, 449.
5. H. S. Rai, M. S. Bhattacharyya, J. Singh, T. K. Bansal, P. Vats and U. C. Banerjee, Crit. Rev. Environ. Sci. Technol., 2005, 35, 219.
6. L. Zhang, H. Zhang, W. Guo and Y. Tian, Appl. Clay Sci., 2014, 93, 85.
7. S. Liu, Y. Ding, P. Li, K. Diao, X. Tan, F. Lei, Y. Zhan, Q. Li, B. Huang and Z. Huang, Chem. Eng. J., 2014, 248, 135.
8. P. Hadi, J. Guo, J. Barford and G. McKay, Environ. Sci. Technol., 2016, 50, 5041.
9. N. O. Mahmoodi, A. Ghavidast, S. Mirkhaef and M. A. Zanjanchi, Dyes Pigm., 2015, 118, 110.
10. A. Ghavidast, N. O. Mahmoodi and M. A. Zanjanchi, J. Mol. Liq., 2014, 198, 128.
11. S. Raghu, C. W. Lee, S. Chellammal, S. Palanichamy and C. A. Basha, J. Hazard. Mater., 2009, 171, 748.
12. S. Singh, S. L. Lo, V. C. Srivastava and A. D. Hiwarkar, J. Environ. Chem. Eng., 2016, 4, 2911.
13. P. Kariyajjanavar, N. Jogttappa and Y. A. Nayaka, J. Hazard. Mater., 2011, 190, 952.
14. M. M. Haque, W. T. Smith and D. K. Y. Wong, J. Hazard. Mater., 2015, 283, 164.
15. E. Hu, X. Wu, S. Shang, X. M. Tao, S. X. Jiang and L. Gan, J. Cleaner Prod., 2016, 112, 4710.
16. S. M. d. A. G. U. d. Souza, K. A. S. Bonilla and A. A. U. d. Souza, J. Hazard. Mater., 2010, 179, 35.
17. P. Aravinda, H. Selvaraj, S. Ferro and M. Sundaram, J. Hazard. Mater., 2016, 318, 203.
18. H. Y. Zhu, R. Jiang, Y. Q. Fu, R. R. Li, J. Yao and S. T. Jiang, Appl. Surf. Sci., 2016, 369, 1.
19. M. Akgül and A. Karabakan, Microporous Mesoporous Mater., 2011, 145, 157.
20. N. Rajic, D. Stojakovic, S. Jevtic, N. Z. Logar, J. Kovac and V. Kaucic, J. Hazard. Mater., 2009, 172, 1450.
21. L. Yao, L. Zhang, R. Wang, S. Chou and Z. Dong, J. Hazard. Mater., 2016, 301, 462.
22. V. P. Mahida and M. P. Patel, RSC Adv., 2016, 6, 21577.
23. B. S. Kaith, R. Jindal and R. Sharma, RSC Adv., 2015, 5, 43092.
24. M. Anbia and S. Salehi, Dyes Pigm., 2012, 94, 1.
25. M. B. Gholivand, Y. Yamini, M. Dayeni, S. Seidi and E. Tahmasebi, J. Environ. Chem. Eng., 2015, 3, 529.
26. M. Shirzad-Siboni, A. Khataee and S. W. Joo, J. Ind. Eng. Chem., 2014, 20, 610.
27. M. Shirzad-Siboni, S. J. Jafari, O. Giahi, I. Kim, S. M. Lee and J. K. Yang, J. Ind. Eng. Chem., 2014, 20, 1432.
28. M. Shirzad-Siboni, A. Khataee, A. Hassani and S. Karaca, C. R. Chim., 2015, 18, 204.
29. H. Chaudhuri, S. Dash and A. Sarkar, New J. Chem., 2016, 40, 3622.
30. O. A. Attallah, M. A. Al-Ghobashy, M. Nebsen and M. Y. Salem, RSC Adv., 2016, 6, 11461.
31. S. Wang and H. Wu, J. Hazard. Mater., 2006, 136, 482.
32. D. Shao, G. Hou, J. Li, T. Wen, X. Ren and X. Wang, Chem. Eng. J., 2014, 255, 604.
33. Y. B. Song, X. D. Song, C. J. Cheng and Z. G. Zhao, RSC Adv., 2015, 5, 87030.
34. B. O. Okesola and D. K. Smith, Chem. Commun., 2013, 49, 11164.
35. H. P. C. van Kuringen, G. M. Eikelboom, I. K. Shishmanova, D. J. Broer and A. P. H. J. Schenning, Adv. Funct. Mater., 2014, 24, 5045.
36. H. P. C. van Kuringen, D. J. Mulder, E. Beltran, D. J. Broer and A. P. H. J. Schenning, Polym. Chem., 2016, 7, 4712.
37. Y. He, Z. Cheng, Y. Qin, B. Xu, L. Ning and L. Zhou, Mater. Lett., 2015, 151, 100.
38. A. R. Tehrani-Bagha, M. Gharagozlou and F. Emami, J. Environ. Chem. Eng., 2016, 1530.
39. M. Shanehsaz, S. Seidi, Y. Ghorbani, S. M. R. Shoja and S. Rouhani, Spectrochim. Acta, Part A, 2015, 149, 481.
40. Y. Pan, J. Wang, C. Sun, X. Liu and H. Zhang, J. Hazard. Mater., 2016, 309, 65.
41. B. Saha, S. Das, J. Saikia and G. Das, J. Phys. Chem. C, 2011, 115, 8024.
42. X. Qiu, N. Li, X. Ma, S. Yang, Q. Xu, H. Li and J. Lu, J. Environ. Chem. Eng., 2014, 2, 745.
43. B. Pan, B. Pan, W. Zhang, Q. Zhang, Q. Zhang and S. Zheng, J. Hazard. Mater., 2008, 157, 293.
44. D. Gonzalez-Rodriguez and A. P. H. J. Schenning, Chem. Mater., 2011, 23, 310.
45. H. P. C. van Kuringen and A. P. H. J. Schenning, in Hydrogen bonding in supramolecular nanoporous materials, ed. Z. T. Li and L. Z. Wu, Springer-Verlag, Berlin Heidelberg, 2015, ch. 2, pp. 43–67.
46. P. Jha, R. Jobby and N. S Desai, J. Hazard. Mater., 2016, 311, 158.
47. N. Thinakaran, P. Panneerselvam, P. Baskaralingam, D. Elango and S. Sivanesan, J. Hazard. Mater., 2008, 158, 142.
48. N. O. Mahmoodi, A. Ghavidast, M. Ashkan, M. Mamaghani, M. A. Zanjanchi, K. Tabatabaeian and A. Arabanian, Synth. React. Inorg., Met.-Org., Nano-Met. Chem., 2016, 46, 1552.
49. A. Mohagheghian, R. Vahidi-Kolur, M. Pourmohseni, J. K. Yang and M. Shirzad-Siboni, Water, Air, Soil Pollut., 2015, 226, 320.
50. A. Alizadeh, M. M. Khodaei, M. Beygzadeh, D. Kordestani and M. Feyzi, Bull. Korean Chem. Soc., 2012, 33, 2546.
51. X. Huang, G. Wang, M. Yang, W. Guo and H. Gao, Mater. Lett., 2011, 65, 2887.
52. B. D. Cullity, Elements of X-ray Diffraction, Addison-Wesley, Reading, MA, 1967.
53. D. Karadag, M. Turan, E. Akgul, S. Tok and A. Faki, J. Chem. Eng. Data, 2007, 52, 1615.
54. Y. Önal, C. Akmil-Başar and Ç. Sarıcı-Özdemir, J. Hazard. Mater., 2007, 146, 194.
55. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361.
56. E. Bulut, M. Özacar and İ. A. Şengil, J. Hazard. Mater., 2008, 154, 613.
57. S. Senthilkumaar, P. Kalaamani, K. Porkodi, P. R. Varadarajan and C. V. Subburaam, Bioresour. Technol., 2006, 97, 1618.
58. I. Langmuir, J. Am. Chem. Soc., 1916, 38, 2221.
59. N. Kannan and M. M. Sundaram, Dyes Pigm., 2001, 51, 25.
60. M. A. K. M. Hanafiah, W. S. W. Ngah, S. H. Zolkafly, L. C. Teong and Z. A. A. Majid, J. Environ. Sci., 2012, 24, 261.
61. H. Freundlich, Über die Adsorption in Lösungen, ed. W. Engelmann, 1906.
62. L. Fan, Y. Zhang, X. Li, C. Luo, F. Lu and H. Qiu, Colloids Surf., B, 2012, 91, 250.
63. J. S. Piccin, G. L. Dotto and L. A. A. Pinto, Braz. J. Chem. Eng., 2011, 28, 295.
64. L. Ai, C. Zhang and Z. Chen, J. Hazard. Mater., 2011, 192, 1515.
65. L. Xiong, Y. Yang, J. Mai, W. Sun, C. Zhang, D. Wei, Q. Chen and J. Ni, Chem. Eng. J., 2010, 156, 313.
66. X. Y. Huang, H. T. Bu, G. B. Jiang and M. H. Zeng, Int. J. Biol. Macromol., 2011, 49, 643.
67. J. J. Lin, X. S. Zhao, D. Liu, Z. G. Yu, Y. Zhang and H. Xu, J. Hazard. Mater., 2008, 157, 541.
68. J. T. Spadaro, M. H. Gold and V. Renganathan, Appl. Environ. Microbiol., 1992, 58, 2397–2401.
69. M. Neamtu, I. Siminiceanu, A. Yediler and A. Kettrup, Dyes Pigm., 2002, 53, 93.
70. G. McKay, A. Mesdaghinia, S. Nasseri, M. Hadi and M. Solaimany Aminabad, Chem. Eng. J., 2014, 251, 236.
71. N. Rajamohan, M. Rajasimman, R. Rajeshkannan and B. Sivaprakash, J. Eng. Sci. Technol., 2013, 8, 778.
72. G. Revathi, S. Ramalingam, P. Subramanian, A. Ganapathi and R. Valliappan, Orient. J. Chem., 2010, 26, 1385.
73. K. Geetha, N. Velmani and P. S. Syed Shabudeen, J. Chem. Pharm. Res., 2014, 6, 26.
74. K. Geetha, N. Velmani, A. Manimaran, K. Mabel Hebsuba and S. Karthikeyan, J. Environ. Nanotechnol., 2016, 5, 39.
75. N. Thinakaran, P. Panneerselvam, P. Baskaralingam, D. Elango and S. Sivanesan, J. Hazard. Mater., 2008, 158, 142.
76. Y. Li, Y. Zhou, W. Nie, L. Song and P. Chen, J. Porous Mater., 2015, 22, 1383.
77. A. A. Rajabi, Y. Yamini, M. Faraji and F. Nourmohammadian, Nanochemistry Research, 2016, 1, 49.
78. A. Fisli, S. Yusuf, S. Ridwan, Y. K. Krisnandi and J. Gunlazuardi, Adv. Mater. Res., 2014, 896, 525.
79. G. Kaykioglu and E. Gunes, J. Nanomater., 2016, 2016, 1.
80. P. Wang, X. Wang, S. Yu, Y. Zou, J. Wang, Z. Chen, N. S. Alharbi, A. Alsaedi, T. Hayat, Y. Chen and X. Wang, Chem. Eng. J., 2016, 306, 280.
81. M. Mohammadi Galangash, Z. Niyazi Kolkasaraei, A. Ghavidast and M. Shirzad-Siboni, in preparation.
82. X. Zhao, Y. Shi, T. Wang, Y. Cai and G. Jiang, J. Chromatogr. A, 2008, 1188, 140.

---