Adsorption studies of methyl orange on an immobilized Mn-nanoparticle: kinetic and thermodynamic

Abstract

In this paper, a novel nano-adsorbent containing Mn-nanoparticle decorated organo-functionalized SiO₂–Al₂O₃ mixed-oxide was introduced as a new scavenger of dyes such as methyl orange. The SiO₂–Al₂O₃ mixed-oxide was functionalized with a Schiff base ligand and thereafter, in the next step, Mn-nanoparticles were prepared over the organo-functionalized SiO₂–Al₂O₃ mixed-oxide. The synthesized materials were characterized by several methods, such as FT-IR spectroscopy, UV-vis, CHN elemental analysis, SEM, TEM, ICP-OES, EPR and XPS. The contact time to obtain equilibrium for maximum adsorption was 15 min. EPR and XPS of the Mn ions evidenced that most of the covalently bonded active sites of the nano-adsorbent are in the form of Mn(III) ions. The heterogeneous Mn(III) ions were found to be an effective adsorbent for the removal of methyl orange ions from solution. The adsorption process was spontaneous and endothermic in nature and followed a pseudo-second-order kinetic model.

---

1. Introduction

Dye pollution of wastewater is an environmental problem urgently needing to be solved. The sources of aquatic dye pollution are broad, including numerous industries like the plastics, paper, textile and also cosmetics industries which utilize dyes to color their goods and 15% of the dyes are lost during the dyeing process. Dyes usually have a synthetic origin and complex chemical structure which make them persistent in light, oxidation and biodegradable processes.¹ Indeed, numerous dyes have been developed for their chemical invariability and do not suffer from biochemical degradation easily. Nowadays, more than one hundred thousand kinds of commercial dyes are utilized with a production of over nine million tons annually.^1,2 Therefore, improving a sustainable method for the removal of synthetic dyes has long been a challenge for scientists. In particular, many efforts have been made to develop several kinds of materials to remove dyes from drinking water.

The following conventional methods are used in dye removal from wastewater: coagulation and flocculation, oxidation or ozonation, membrane separation, and adsorption.² Adsorption processes have been reported to be low-cost promising alternatives for the treatment of dyes present in wastewater. The use of activated carbons, modified clays, polymeric resins, waste materials, and zeolites as adsorbents have also been described.^1,2 For most natural adsorbents, such as clays, zeolites, agricultural waste, and chitosan, their particles show negatively charged surfaces and consequently, excellent adsorption properties for cationic organic compounds can be obtained. However, these systems generally have some disadvantages in dye removal, such as the difficulty in separating these powdery natural adsorbents, except by high speed centrifugation, from the treated effluent, nonresistance against acid solutions, their poor mechanical strengths and that they are less adaptable to a wide range of dye wastewaters limits their practical applications. However, most of these adsorbents either do not have considerable adsorption capacities or need relatively long adsorption contact times, e.g., from several hours to a couple of days (Table 1). It is, therefore, desirable to develop effective adsorbents with short contact times for the removal of dye ions from aqueous solutions. Activated carbon is the most prevailing adsorbent for this process because of its high surface area, high adsorption capacity, and high degree of surface reactivity. However, it is expensive and must be regenerated on a regular basis. Inorganic supports present several advantages with respect to activated carbon, including stability, high surface area, possible reuse, relative rapidity in reaching equilibrium, better high mechanical resistance, easy modifications and a higher concentration of chelating groups on the surface, and they are often much cheaper than their organic counterparts. For comparative purposes, the adsorption capacities of several adsorbents for methyl orange are summarized in Table 1.

Table 1 Adsorption capacities of MO on various adsorbents

Adsorbents	Qmax (mg g^−1)	Ref.
Ferrocene-Si–Al	381	3
λ-Fe2O3/MWCNTs/chitosan	60.5–66.1	4
Hypercrosslinked chloromethylated PS adsorbent functionalized with formaldehyde carbonyl groups (HJ-1)	70.9–76.9	5
Mesoporous magnetic Co-NPs/carbon nanocomposites	380	6
Acid modified carbon coated monolith	132.7	7
Calcined layered double hydroxides	200	8
Alkali-activated multiwalled carbon nanotubes	149	9
Pinecone derived activated carbon	404.4	10
Carbon nanotubes	35.4–64.7	11
Chitin/alginate magnetic nano-gel beads (MCAs)	107.5	12
Porous spongy CSGO monolith	567.07	13
Chitosan/MgO composite	60	14
Cu@Cu2O	344.8	15
Ultrafine coal powder	5.24	16
Modified ultrafine coal powder	5.56	16
Chitosan	34.83	17
Calcined Lapindo volcanic mud	333.3	18
Mesoporous Y-Fe2O3/SiO2 nanocomposites calcined	476	19
Y-Fe2O3/SiO2/chitosan composite	34.29	20
MgNiAl layered double hydroxides	375	21
Protonated cross-linked chitosan	89.30	22
Mesoporous carbon CMK-3	294.1	23
De-oiled soya	16.66	24
Bottom ash	3.61	24
Metal–organic framework (MOF)	194	25
Ni-containing ordered mesoporous carbons	107.1	26
Hypercrosslinked polymer	72.9	27
Banana peel	21	28
Ammonium-functionalized silica nanoparticle	105.4	29
Palygorskite clays	39	30
700 °C treated palygorskite clays	98	30
Mn@Si/Al	571	This work

---

In this paper, the application of stable, covalently immobilized Mn-nanoparticles (manganese is one of the most versatile metals, playing a central catalytic role in the environment) on the surface of a SiO₂–Al₂O₃ mixed-oxide host by a linker approach has been studied as a novel nano-adsorbent for capturing methyl orange from aqueous solution. Methyl orange (MO), with an IUPAC name of sodium 4-[(4-dimethylamino) phenyl diazenyl] benzene sulfonate, is a typical water-soluble anionic dye and has a harmful effect on living organisms within a short period of exposure.

2. Experimental

2.1. Materials

All the chemicals were purchased from Merck and Aldrich and used without further purification, except for the solvents, which were treated according to standard methods. Doubly distilled water was used throughout.

2.2. Preparation of the organometallic functionalized SiO₂–Al₂O₃ mixed-oxide

SiO₂–Al₂O₃ (1/1) was used as the support. This support was prepared by a sol–gel method as follows: aluminum tri-sec-butylate (97%) and tetraethyl orthosilicate (98%) were used as the precursors, and 2,4-pentandione (H-acac) as the complexing agent. Appropriate amounts of aluminum tri-sec-butylate and tetraethyl orthosilicate were dissolved in n-butanol and the solution was heated to 60 °C. The components were thoroughly mixed, cooled down to room temperature, and then H-acac was added. The resultant clear solution was hydrolyzed with deionized water (11.0 mol H₂O/mol alkoxide). The solution was left overnight to hydrolyze the alkoxides, yielding a transparent gel. The transparent gel was dried at 110 °C to remove water and any solvent, and then it was calcined at 500 °C for 5 h to remove the organic materials. The support is referred to as SiO₂–Al₂O₃ (1 : 1). The SiO₂–Al₂O₃ (1 : 1)-supported 2-aminoethyl-3-aminopropyl-trimethoxysilane (2-AE-3-APTMS) (Scheme 1) was prepared by refluxing 5.2 g of SiO₂–Al₂O₃ (1 : 1) that was activated at 550 °C for 6 h under air with 3.5 mL (0.0195 mol) of 2-AE-3-APTMS in dry dichloromethane (100 mL) for 24 h. The solid was filtered and washed with methanol and dichloromethane, and dried at 100 °C under vacuum for 6 h. The functionalized SiO₂–Al₂O₃ (1 : 1) mixed-oxide that was prepared with the linker is identified hereafter by Si/Al-pr-NH-et-NH₂. Methyl-2-pyridylketone was added to a suspended solution of Si/Al-pr-NH-et-NH₂ in dry methanol. The mixture was refluxed for 24 h to prepare a Schiff base (vide Scheme 1) on the surface of the mixed-oxide (a bi-dentate ligand).

Scheme 1 Immobilization procedure of the organometallic functionalized SiO₂–Al₂O₃ mixed-oxide.

The adsorbent containing Mn-nanoparticles was obtained by stirring 0.5 g of the hybrid material, the Si–Al mixed-oxide-Schiff base ligand, with Mn(OCOCH₃)₂·4H₂O (5.4 mmol), and LiCl (8.5 mmol) in 30 mL of ethanol at reflux for 24 h. Then, the resulting material (brown powder) was filtered off, washed with copious amounts of ethanol and methanol and dried under vacuum at 60 °C.

2.3. Characterization

Diffuse reflectance spectra were recorded on a JASCOV-550 UV-Vis spectrophotometer. Fourier transform IR spectra were measured using a JASCO FT/IR (680 plus) spectrometer. The spectra of the solids were obtained using KBr pellets. The vibrational transition frequencies are reported in wave numbers (cm⁻¹). Approximately 40–50 mg of material, which had been previously calcined in air at 298 K for 4 h, with 10–15 times KBr was pressed (for 3 min at 15 metric tons cm⁻² pressure under approximately 10–2 Torr vacuum) into a self-supporting wafer of 0.9 mm diameter. The FT-IR spectra of the synthesized materials were recorded at a resolution of 4 cm⁻¹ and a scan number of 16. Elemental analysis was performed by a CHNO-Rapid Heraeus elemental analyzer (Wellesley MA). Chemical analyses were carried out by inductively coupled plasma optical emission spectroscopy (ICP-OES) using a Shimadzu ARL 34000 instrument (spectroflamed; typically, 30 mg of the sample was dissolved in 500 μL of a 40% HF solution, 4 mL of a 1 : 4 HCl : H₂SO₄ solution and 45 mL of H₂O). Nitrogen (99.999%) adsorption experiments have been performed at −196 °C using a volumetric apparatus (Quanta chrome NOVA automated gas sorption analyzer). Before the adsorption experiments, the sample was de-gassed at 120 °C for 16 h. The specific surface areas are calculated from the BET method. Transmission electron microscopy (TEM) was carried out on powder samples with a Tecnai F30TEM operating at an accelerating voltage of 300 kV. In addition, energy dispersive X-ray analysis was conducted on each sample. XPS (small area X-ray photoelectron spectroscopy) data were recorded with a PHI-5702 Multi-Technique System. Electron paramagnetic resonance (EPR) spectra were recorded with a Bruker EMX spectrometer operating at X-band (m = 9.42 GHz) frequency and a 100 kHz field-modulation. The spectra were recorded at 80 K using a Bruker BVTB 3500 variable temperature controller. The magnetic field was calibrated with a Bruker ER 035 M NMR gaussmeter and the microwave-frequency was calibrated with a frequency counter fitted in a Bruker ER 041 XG-D microwave bridge unit.

2.4. Adsorption measurements

Different MO ion concentrations were freshly prepared in a solution of deionized water. Sorption experiments were carried out in batch conditions: 0.07 g of the Mn nano-adsorbent was shaken up with 20 mL of the organic pollutant at a concentration of between 0.75 and 1000 mg L⁻¹, and at a controlled temperature of 25 °C (each reaction was repeated three times, and displayed a relative standard deviation lower than 1.64%). The time required to work under equilibrium conditions was determined by preliminary kinetic measurements. The kinetic tests of MO showed no significant variation in sorption after 24 h. After centrifugation at 3000 rpm for 5 min, the liquid phase was separated and the solute concentration was determined at λ_max = 464.0 nm using a JASCOV-550 instrument.

The amount adsorbed was calculated as:

q_e = V(C_o − C_e)/m (1)

where C_o and C_e are the initial and equilibrium liquid-phase concentrations (mg L⁻¹) of the adsorbate; V is the volume of the solution (L); and m is the amount of adsorbent (g). Eqn (1) assumes that the change in the volume of the bulk liquid phase is negligible as the solute concentration is small and the volume occupied by the adsorbent is also small. The amount of the dye adsorbed on the sample was calculated based on a previously determined calibration curve.

3. Results and discussion

3.1. Characterization of Mn@Si/Al

The structure of the obtained organometallic-modified Si–Al mixed-oxide was confirmed by elemental analysis, BET (N₂ adsorption–desorption technique), FT-IR spectroscopy, UV-Vis, SEM, TEM, ICP-MS, EPR and XPS. The nitrogen sorption measurement of the modified Si–Al mixed-oxide confirms the presence of the Mn-complex attached to the modified Si–Al mixed-oxide (Table 2). The considerable decrease in the specific surface area (S_BET) after functionalization is due to the presence of organometallic groups, which block the access of nitrogen molecules into the structure of the SiO₂–Al₂O₃ mixed-oxide. The nitrogen and Mn contents of the organometallic-modified Si–Al mixed-oxide (Mn@Si/Al) were determined by elemental analysis and ICP, respectively. Consequently, the content of the immobilized Mn@Si/Al was computed. The results are listed in Table 2. From these data, we could calculate that the aluminosilicate support bears 4.1 mmol g⁻¹ of the Schiff base and 3.0 mmol g⁻¹ of Mn ions. In fact, during the condensation and immobilization, the dosage of methyl-2-pyridylketone was excessive to minimize the amount of the untreated residual linker on the support. A change in the color of the resultant powder can also be visualized during the reactions (Scheme 1).

Table 2 Chemical composition and physicochemical properties of the organometallic functionalized SiO₂–Al₂O₃ mixed-oxide

Catalyst	Elemental analyses^b (wt%)		Organic functional group (mmol g^−1 mixed-oxide)^c	Immobilized Mn–Schiff base-complex (mmol g^−1 mixed-oxide)^d	Structural parameters^e
	N	Mn			Surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore diameter (Ao)
a Molar ratio of Si–Al was 60 : 40, determined from EDX analysis.b Nitrogen was estimated from the elemental analyses. Mn content determined from ICP analysis.c Determined from the N-content.d Determined from the Mn-content.e The pore size calculated using the BJH method.
SiO2–Al2O3 mixed-oxide^a	—	—	—	—	243	0.028	20
Si/Al-pr-NH-et-N = methyl-2-pyridylketone	6.2	—	4.4	—	131	0.014	16
Si/Al-pr-NH-et-N = methyl-2-pyridylketone-Mn	5.7	3.0	4.1	0.56	114	0.014	16

---

3.1.1. IR spectroscopy. Infrared wave numbers (cm⁻¹) of significant valence vibrations, which are collected in Fig. 1, are helpful for the identification of the immobilized ligand and the complex. The SiO₂–Al₂O₃ mixed-oxide showed characteristic FT-IR peaks at 2900–3800, 1260–1060 and 600–900 cm⁻¹ due to the O–H bonds of the (Si–Al)O₄ units, adsorbed water molecules and M–O–M stretching vibrations,³¹ respectively, and the bands at 2851–2921 cm⁻¹ are assigned to the stretching mode of the remaining –CH₂ groups. 2-AE-3-APTMS-SiO₂–Al₂O₃ showed bands at 2851–2921 cm⁻¹, which strongly suggests that the SiO₂–Al₂O₃ mixed-oxide has been successfully modified by amine spacer groups. Likewise, the N–H deformation peaks at 1540–1560 cm⁻¹ confirm the successful functionalization of the Si–Al mixed-oxide with 2-AE-3-APTMS. In accordance with the previously reported data,³² the C=N (Schiff base) absorption was assigned at 1635 cm⁻¹, the vibration of C=N from the pyridine group appeared in the region of 1571–1569 cm⁻¹ (Fig. 1),³³ and the peak at 1436 cm⁻¹ can be assigned to the C=C stretching vibration of the pyridine group. The FT-IR spectrum of the Schiff base clearly shows the C–H vibrations of Py groups at 3070–3060 cm⁻¹, which further confirms the presence of pyridine groups on the Si–Al mixed-oxide after immobilization of the ligand.

Fig. 1 FT-IR spectra the organometallic functionalized SiO₂–Al₂O₃ mixed-oxide in the region of 4000–400 cm⁻¹. Legends: SiO₂–Al₂O₃ mixed-oxide (A), Si/Al-pr-NH-et-NH₂ (B), Si/Al-pr-NH-et-N = methyl-2-pyridylketone (C), Si/Al-pr-NH-et-N = methyl-2-pyridylketone-Mn (D).

In the infrared spectrum of the pristine SiO₂–Al₂O₃, the structural bands at 780 cm⁻¹ correspond to the stretching vibration modes of the M–O, M–O–M, and O–M–O bonds, where M means metal atoms, and O means oxygen atoms. Furthermore, it can be seen that there are complex bands in the 780 cm⁻¹ region, attributed to ring vibrational modes. It is interesting to note that the bands related to the M–O lattice vibrations changed in intensity at around 780 cm⁻¹, which further indicated that an organo-functional group was successfully immobilized on SiO₂–Al₂O₃. However, for the spectrum of Mn@Si/Al, it can be easily seen that the bands are similar to those of SiO₂–Al₂O₃ in the range from 900 to 600 cm⁻¹, indicating that the M–O lattice vibrations showed a small change. Further evidence for this coordination mode was provided by the ν(Mn–N) band at ca. 412 cm⁻¹.³⁴ The FT-IR results demonstrate the formation of the Mn-complex, immobilized on the Si–Al mixed-oxide through the 2-AE-3-APTMS linker (Fig. 1).

3.1.2. Dynamic reflectance UV-Vis spectroscopy. The anchoring of the synthesized materials on the solid surface was also followed by DR-UV-Vis spectroscopy of the resulting adsorbent. Thus, the UV-Vis spectrum of the Si–Al mixed-oxide only had a side-band adsorption near 244 nm, while the spectrum of Si/Al-pr-NH-et-N = methyl-2-pyridylketone-Mn (Mn@Si/Al) was dominated by strong absorptions in the 250–305 nm region due to the π → π* and n → π* transitions of the ligand (Fig. 2). Furthermore, Mn@Si/Al exhibited broad and weak bands around 391 nm, probably attributable to ligand-to-metal charge transfer transitions, similar to metallosalen compounds.³⁵ Several low intensity asymmetric broad bands appeared above 500 nm in the visible region at λ = 530–680 nm and are attributed to the d → d transitions expected for manganese complexes with a square pyramidal geometry, (d_xz → d_x²_−y²), (d_yz, d_xy → d_x²_−y²) and (d_z² → d_x²_−y²) which are similar to the related metal (salen) compounds described in the literature.³⁶ The DR-UV-Vis and FT-IR spectra reveal that the Mn@Si/Al nano-adsorbent is synthesized upon the coordination of Mn ions to the organo-functional groups. The charge transfer bands, the stretching frequency at 1635 cm⁻¹ (Fig. 1) and the elemental analyses (Table 1) clearly confirm that the Schiff base ligands and C=N groups are not affected or destroyed through the immobilization on the functionalized Si–Al mixed-oxide.

Fig. 2 UV-Vis diffuse reflectance spectra of the organo-functionalized SiO₂–Al₂O₃ mixed-oxide and immobilized manganese nano-mediator. Legends: SiO₂–Al₂O₃ mixed-oxides (A), Si/Al-pr-NH-et-N = methyl-2-pyridylketone (B), Si/Al-pr-NH-et-N = methyl-2-pyridylketone-Mn (C).

3.1.3. SEM and TEM study. Scanning electron microscope images of the nanosized SiO₂–Al₂O₃ mixed-oxide and Mn@Si/Al are shown in Fig. 3. It could be seen that the nanoparticles appearance and size were similar, demonstrating that the nanoparticles of the SiO₂–Al₂O₃ mixed-oxide have good mechanical stability and they have not been destroyed during the whole modification. The TEM micrograph of the modified mixed-oxides, Fig. 3, shows that the involved Mn ions confined onto the organo-functionalized nanosized SiO₂–Al₂O₃ mixed-oxide are well dispersed (the Mn ions did not accumulate and are dispersed as Mn-nanoparticles) (see Fig. 4).

Fig. 3 SEM micrographs of Mn@Si/Al (left) and TEM micrograph of Mn@Si/Al (right).

Fig. 4 Mn-nanoparticles; TEM micrographs of Mn@Si/Al plus a carbon nanotube paste electrode (the dark spots are Mn-nanoparticles).

3.1.4. EPR spectroscopy. EPR spectroscopy has been employed to confirm the oxidation state and the chemical environment of the Mn species present in the sample. The extreme spin–spin coupling of an even number of unpaired electrons combined with an extremely short relaxation time makes Mn(III) ions hard to monitor by EPR spectroscopy. For Mn(II) ions with an electron spin of S = 5/2, there are five different ΔM = 1 transitions, where M is the electron spin quantum number. For an isotropic g factor, transitions other than ΔM = +1/2 ↔ −1/2, forcefully depend on the orientation with respect to the magnetic field and are typically not monitored in powder samples. The ΔM = +1/2 ↔ −1/2 transition splits into a sextet by the hyperfine splitting of the Mn nuclear spin I = 5/2. The +1/2, m ↔ −1/2, m transitions with Δm = 0, where m is the nuclear spin quantum number, are allowed transitions that give a sextet spectrum, and the ΔM = 1, Δm = 1 transitions are forbidden transitions that show weak lines between the lines of the sextet.

The X-band EPR spectrum of the immobilized Mn-nanoparticles on the functionalized SiO₂–Al₂O₃ mixed-oxide measured at room temperature is indicated in Fig. 5. The EPR spectrum shows that some of the Mn ions are in other oxidation states, because Mn(III) exhibits no EPR signal. The observed EPR signals, characteristic of Mn(II)/Mn(IV), indicate that, at least, part of the Mn ions are present in the +2 or +4 oxidation state upon immobilization on the functionalized SiO₂–Al₂O₃ mixed-oxide. Thus, next to Mn(III), both Mn(II) and Mn(IV) species must be present, and the observed EPR spectrum must be due to one or both of these species (the reported g values for Mn(IV) are slightly less than 2, whereas those of Mn(II) are slightly above 2).

Fig. 5 EPR spectrum of Mn@Si/Al.

The EPR spectrum appears to have a broad feature centered at around g = 2, with wide wings (such wings are typical for a distribution of the zero-field splitting parameters, and have been described for Mn(II) in glass environments,³⁷ or for frozen solutions of Mn(II) phosphate complexes).³⁸ In fact, the peak broadening and wings are due to dipole–dipole interactions,³⁹ and consequently, they are predominant in the spectra of higher loaded samples (samples containing more than 1.5 wt% Mn).⁴⁰

The EPR spectrum of Mn@Si/Al has a lower intensity in comparison with the other spectra reported in the literature. On the other hand, the inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis of the samples indicates that the nanoadsorbent has a high loading of manganese. This contradiction might be explained by stating that Mn(III) species (expected to be EPR silent under these conditions) are probably the most likely product of the reaction between Mn(II) and Si/Al-pr-NH-et-N = methyl-2-pyridylketone.

Therefore, Mn(III) is the main oxidation state which is immobilized on the functionalized SiO₂–Al₂O₃ mixed-oxide and probably MO was mainly captured by Mn(III). It is believed that during the immobilization of Mn(II) over the modified SiO₂–Al₂O₃ under air conditions (in the presence of atmospheric oxygen), the oxidation state of Mn(II) was easily oxidized and began to prefer the Mn(III) state (in fact, oxidation of Mn(II) gives a stable cation: Mn(III)). The EPR peak broadening of Mn@Si/Al could be due to the presence of Mn(II) and Mn(IV) in its structure (g = 1.95). Microcrystalline nano-catalysts often yield characteristic EPR spectra with features distinctly different from those observed for magnetically isolated ions. The detailed features of the resonances depend strongly on the shape, size distributions and on the magnetic properties of the particles. Consequently, with regard to the EPR spectrum and TEM photo, the broad feature around g_av = 2.03 is probably characteristic of super paramagnetic relaxation as it is concerned with the magnetic moment of the whole particle and either may be due to the highly dispersed Mn species on the nano-adsorbent (Fig. 5).

3.1.5. XPS. XP spectroscopic analysis of the heterogeneous catalyst was used to study the characterization of the oxidation state of the immobilized Mn-nanoparticles on the surface of the modified-SiO₂–Al₂O₃. This technique has previously been shown to provide valuable information regarding the chemical state of the catalytically active sites in different catalysts.⁴¹ The XPS spectrum of the Mn catalyst produced a Mn 2p_3/2 and a Mn 2p_1/2 peak at 643.01 and 654.61 eV, respectively, and is depicted in Fig. 6. However, there are two satellite peaks at 647.5 and 659.4 eV beside the main peak of Mn 2p_3/2 and Mn 2p_1/2 in Mn@Si/Al. These results indicate that most of the Mn ions on the surface of the support are in the +3 oxidation state (due to the binding energy (BE) of 643.01 eV for Mn 2p_3/2, which is attributed to the Mn(III) state of the nanoparticles).

Fig. 6 XP spectroscopy of Mn@Si/Al. The Mn 2p_3/2 component is fit by multiplet peaks labeled 1, 2 and 3, with peak 3 corresponding to the shake-up satellite. The Mn 2p_1/2 component is fit by two peaks and a satellite peak.⁴²

3.2. Adsorption study

Preliminary experiments showed that the nano-sized SiO₂–Al₂O₃ has a low capacity to remove MO dye from wastewater. So, in this work, various pretreatment processes are designed for the nano-sized SiO₂–Al₂O₃ with the aim of enhancing the MO adsorption capacity by altering the chemical modification to give more active sites and/or by obtaining a suitable physical structure of the available functional groups.

3.2.1. Effect of the initial pH on methyl orange ion adsorption. The effect of pH plays an important role on the active sites of the nano-adsorbent as well as the dye speciation during the adsorption reaction.^1–20 In order to evaluate the influence of the pH on the adsorption capacity of Si/Al and Mn@Si/Al, experiments were carried out at an initial concentration of 105 mg L⁻¹ and in the pH range 1.0–13.0 (Fig. 7).

Fig. 7 Effect of pH on MO ion sorption from aqueous solution.

As shown in Fig. 7, the modified Si–Al nano-sized mixed-oxide with organometallic groups indicates higher MO capturing capacity than the unmodified Si–Al at various pH values. The results indicate that the pH has a significant influence on the adsorption of MO onto Mn@Si/Al. A similar observation was published in the literature for different adsorbents.^10–30 As the initial pH increased to 6.8, the sorption capacity increased to more than 97.3% and decreased at the initial pH values above 6.5 (>48.6%). The lowest MO sorption capacity of Mn@Si/Al (48.6%) was found at an initial solution pH of 13.0, which may be due to partial dissolution of the immobilized Mn ions at this solution pH (data not published).

At acidic pH, protons can combine with one of the nitrogens in a nitrogen–nitrogen double bond, which can generate a reddish colored MO dye solution, that is, MO is protonated and hence the electrostatic repulsion interaction between protonated MO and the positively charged Mn@Si/Al active sites results in a decline in the percentage of adsorption (the dissociation constant pK_a for MO is 3.46, so MO molecules were predominantly present as monovalent anions at this equilibrium pH). However, decreasing MO adsorption in basic pH may be attributed to the competition of OH⁻ with the MO ions for the adsorption sites on the Mn@Si/Al (the active sites, Mn(III) ions, are closely associated with hydroxyl ions OH⁻, that is, restricting the approach of MO ions as a result of the repulsive force) and therefore fewer groups are available for MO to bind with. Therefore, the possible mechanism of MO adsorption may be considered as the strong electrostatic interaction between the positive adsorption active sites of the adsorbent and the negatively charged MO (Scheme 2). This pH-dependent trend has also been observed for the adsorption of similar anionic dyes onto TiO₂,⁴³ activated carbon modified with Y(III) ions⁴⁴ and the copper(II) complex of dithiocarbamate-modified starch,⁴⁵ confirming the interaction of the anionic dyes with metal centers, which directs the adsorption process.

Scheme 2 The proposed mechanism for the heterogenization of MO by Mn@Si/Al.

To further study the oxidation state of the immobilized Mn ions at different pH values, the ESR spectrum of Mn@Si/Al at pH 2.0 was measured (Fig. 8) and compared with those recorded at pH 7.0 (Fig. 5). Among the methods used in this study [Fourier transform infrared (FT-IR), diffuse reflectance UV-Vis (DRS) and electron paramagnetic resonance (EPR)], EPR was found to be sensitive to the oxidation state of the manganese atom at different pH values. The ESR spectra of Mn@Si/Al at pH values lower than 4.0 show similar features to that of Mn@Si/Al at pH values higher than 4.0 but the intensity of the ESR spectra of Mn@Si/Al at pH > 4.0 is less than that recorded at pH < 4.0. The intensity of the ESR spectrum of Mn@Si/Al at pH 2.0 shows a high intensity with a g value of about 2 compared to those prepared at pH > 4.0, which is consistent with Mn(II) (g = 2.0) in the environment of distorted octahedral symmetry. Therefore, for the sample at pH > 4.0, there might be more oxidation of Mn(II) to Mn(III). Indeed, Mn(III) sites are also on Mn@Si/Al as they are more inclined to be incorporated into the framework than Mn(II).⁴⁶ In order to introduce a greater number of Mn(III) sites on Mn@Si/Al, the conversion of Mn(II) to Mn(III) is required. Since this oxidation can be carried out by dissolved O₂, pH values higher than 4.0 might be more effective than pH 2.0 or 3.0 to dissolve more oxygen and to increase oxidation. However, these results revealed that better heterogenization of Mn(II) on the active sites of Schiff base ligands can be obtained at pH values lower than 4.0. Thus, increasing the concentration of Mn(III) on Mn@Si/Al is feasible if the experimental pH is raised to 4.0. All the above experiments clearly reveal that the acidity of the solution influences the electronic structure of the immobilized Mn-nanoparticle.

Fig. 8 EPR spectrum of Mn@Si/Al at pH 2.0.

3.2.2. Effect of the adsorbent dosage. The adsorbent dosage is one of the most important factors as it presents the capacity of the adsorbent for a given initial amount of the adsorbate. To determine the correlation of MO sorption on adsorbent dosage, various dosages (0.010–1.00 g) of Mn@Si/Al, at a controlled temperature of 25 ± 1 °C, were added into 10 mL of a 105.0 mg L⁻¹ MO solution without any changes to the solution pH (Fig. 9). The relation between the adsorbent dosage and the removal percentage of MO by Mn@Si/Al is shown in Fig. 9, from which it can be seen that the removal percentage of MO increased on increasing the adsorbent dosage, probably due to an increase in the number of adsorption sites. As the adsorbent dosage increased from 0.010 to 0.15 g, the removal efficiency of MO ions increased significantly from 46. 0% to 99.3%.

Fig. 9 Effect of adsorbent dosage on the sorption of MO ions by Mn@Si/Al from aqueous solution.

However, the higher adsorbent dose results in a lower removal capacity of Mn@Si/Al (94.6%). It is believed that at the low adsorbent dosage, the dispersion of Mn@Si/Al nanoparticles in an aqueous solution is better, that is, all of the active sites on the adsorbent surface are entirely uncovered, which could accelerate the approachability of MO molecules to a large number of the adsorbent active sites. Thus, the adsorption on the active sites is saturated quickly, resulting in a high removal capacity. On the other hand, at higher adsorbent dosages, the accessibility of adsorbent active sites with higher energy decreases and a larger fraction of the active sites with lower energy become occupied, leading to a decrease in the adsorption capacity. Furthermore, increasing the adsorbent dosage enhances the chance of collision between the adsorbent nanoparticles and hence creates particle aggregation, inducing a decline in the total surface area and an increase in the diffusion path length, which both result in a decrease in the amount of removal capacity of MO from aqueous solution. Therefore, a 0.07 g adsorbent dosage was chosen as the optimal dosage for the rest of the study.

3.2.3. Effect of the initial concentration on the uptake. The adsorption results obtained for Si–Al and Mn@Si/Al upon varying the initial MO ion concentration (0.75–1000 mg L⁻¹) are illustrated in Fig. 10, where it can be seen that Mn@Si/Al was demonstrated to be the most efficient in that more than 97.0% of MO was removed after 15 min, whereas only 4.2% of MO was removed by Si–Al. The Si–Al utilized in this investigation is a nano-sized mixed-oxide and it shows low efficiency in the removal of MO, but provided good suspensibility and dispersibility for organometallic groups when removing MO from aqueous solution and it improved the reactivity of Mn@Si/Al.

Fig. 10 Equilibrium absorption of MO by Si–Al and Mn@Si/Al at 25 °C. Dashed line represents the fitting curve using the Langmuir adsorption model.

When the initial MO concentrations were increased in the presence of Mn@Si/Al, the removal efficiency of the dye reached 98.0% and 53.5% for 150 and 1000 mg L⁻¹ initial dye concentrations, respectively. The relative increase in the loading capacity of the sorbent with increasing MO concentrations is probably due to the interaction between the dye and the adsorbent, which provides the vital driving force to defeat the resistance to the mass transfer of MO ions between the aqueous solution and Mn@Si/Al. The observed enhancement in the MO uptake on increasing the initial dye ion concentration could be due to an increase in electrostatic interactions (relative to covalent interactions), which involves active sites of progressively lower affinity for MO up to the saturation point. On the other hand, the higher removal efficiency of Mn@Si/Al at a low MO initial concentration could be related to the high ratio of initial mole numbers of MO to the available active sites on the surface area and therefore, the fractional adsorption is dependent on the initial concentration.

3.2.4. Adsorption isotherms. Equilibrium adsorption isotherms are known to be very important when it comes to understanding adsorption mechanisms. Thus, the experimental adsorption equilibrium data of MO dye on Mn@Si/Al were fitted by applying the Langmuir and Freundlich isotherm models, which are typically used for aqueous-phase adsorption (Table 3).^1–30 These adsorption models give a representation of the adsorption equilibrium between an adsorbate in solution and the surface active sites of the adsorbent. The Langmuir and Freundlich adsorption isotherms can be expressed by eqn (2) and (3), respectively.

q_e = q_mK_LC_e/1 + K_LC_e (2)

q_e = K_FC_e^1/n (3)

where q_e (mg g⁻¹) is the specific equilibrium amount of the adsorbate, C_e (mg L⁻¹) is the equilibrium concentration of the adsorbate, q_m is the maximal adsorption capacity, K_L is the Langmuir isotherm constant (L mg ⁻¹), K_F is the Freundlich isotherm constant (mg g⁻¹ (mg L⁻¹)ⁿ) and n is an indicator of the adsorption effectiveness. The constant n gives an idea of the degree of heterogeneity in the distribution of energetic centers and is related to the magnitude of the adsorption driving force. High n values therefore indicate a relatively uniform surface, whereas low values mean high adsorption at low solution concentrations. Furthermore, low n values indicate the existence of a high proportion of high-energy active sites.

Table 3 Fitting of the parameters of the experimental results to the Langmuir and Freundlich equation parameters

System	qm (mg g^−1)	KL (L mg^−1)	KF (mg g^−1 (mg L^−1)^n)	n	R^2	Sorption model
Mn@Si/Al	571.2	2.73 × 10^−2			0.9927	Langmuir
			0.448	0.294	0.6927	Freundlich

---

The Langmuir equation relates the coverage of molecules on a solid surface to the concentration of a medium above the solid surface at a fixed temperature. Adsorption is limited to monolayer coverage, and intermolecular forces decrease with the distance from the adsorption surface, whereas the Freundlich model supposes that the adsorption surface is heterogeneous, that interaction between adsorbed molecules can occur, and that multilayer adsorption is possible. The Langmuir and Freundlich adsorption isotherms exhibit an approximately linear relationship for Mn@Si/Al. The data obtained from Mn@Si/Al revealed that the Langmuir isotherm model correlated better (R² > 0.9827) than the Freundlich isotherm (Table 3 and Fig. 10), using the experimental data for the adsorption equilibrium of MO ions by the modified SiO₂–Al₂O₃ mixed-oxide. Thereby, it is realistic to infer that the adsorption active sites on the modified Si–Al are not mostly energetically heterogeneous (non-uniform surface), and this is a major aspect in this model. Likewise, the maximum adsorption capacity (q_m) for Mn@Si/Al (571.2 mg g⁻¹), as obtained using the Langmuir isotherm, was much higher than the values found for most other adsorbents reported in the literature,^3–30 thereby also suggesting that there is a high thermodynamic stability for MO heterogenization at the active sites of Mn@Si/Al (Table 3).

3.2.5. Adsorption kinetics. The effect of the shaking time (0–240 min) on the adsorption of MO (150 mg L⁻¹) onto Mn@Si/Al (0.07 g), in a solution with pH 6.5 at 25 °C is shown in Fig. 11, from which it can be seen that the amount of adsorption increases on increasing the contact time. Studies of the adsorption kinetics of MO removal revealed that the majority of the dye (98.0%) was removed within the first 0–15 min of contact with the modified nano-adsorbent. The percentage of maximum adsorption was 99.8% at 240 min. Indeed, the fast adsorption during the initial stage is probably due to the high concentration gradient between the adsorbate in solution and that on the adsorbent as there are a high number of vacant sites available during this period, while the obtained plateau after 15 min relates to a slow rate of adsorption which could be due to an agglomeration of MO molecules on the Mn@Si/Al active sites. Therefore, in order to optimize the adsorption process, the adsorption isotherms for the remaining initial concentrations were obtained for a contact time of 15 min.

Fig. 11 The adsorption kinetics of Mn@Si/Al for MO at room temperature. The inset shows the pseudo-second order plot for the adsorption.

In order to determine and interpret the mechanisms of the MO adsorption processes and the main parameters governing sorption kinetics, empirically obtained kinetic sorption data were fitted to the pseudo-first-order, pseudo-second-order, and intra-particle diffusion models shown in eqn (4) and (5), and (6), respectively. Each of these models has been widely used to describe metal and organic sorption on several sorbents.^1,2 From the linear form of these three models, equations can be written as follows:

Pseudo-first-order equation: q_t = q_e[1 − exp(−k₁t)] (4)

Pseudo-second-order equation: q_t = k₂q_e²t/1 + k₂q_et (5)

Intra-particle diffusion equation: q_t = k_intt^1/2 (6)

The initial adsorption rate (h) can be determined from the k₂ and q_e values using

h = k₂q²_e (7)

where k₁, k₂ and k_int are the adsorption rate constants of first, and second order kinetic and intra-particle diffusion models, in min⁻¹, L (mg min)⁻¹ and mg g⁻¹ min^−1/2, respectively; q_e and q_t, in mg g⁻¹, are the equilibrium adsorption uptake (at time t = ∞) and the adsorption uptake (at time t), respectively.

The calculated kinetics parameters for the adsorption of MO dye ions onto Mn@Si/Al at an initial concentration of 150 mg g⁻¹ are presented in Table 4, where it can be seen that the pseudo-second-order equation appeared to be the best-fitting model compared to those for the other two equations (the correlation coefficient is extremely high for the pseudo-second-order equation of Mn@Si/Al; R² > 0.999). The value of q_e,cal also appeared to be very close to the experimentally observed value of q_e,exp. The plot of the linear form of the pseudo-second-order for the adsorption of MO ions is shown in the inset of Fig. 11. Similar results have been reported for the adsorption of dyes onto different adsorbents in the literature.^1–10

Table 4 Kinetic parameters for the adsorption of MO by Mn@Si/Al

Pseudo first order constants	Pseudo second order constants	Intra-particle diffusion constants
k1 (min^−1) = 0.623	k2 (g mg^−1 min^−1) = 0.0065	kint (mg g^−1 min^−1/2) = 1.7743
q1 (mg g^−1) = 144.9	q2 (mg g^−1) = 149.2	R^2 = 0.7486
R^2 = 0.8563	h (mg g^−1 min^−1) = 144.9
	R^2 = 1

---

This consistency in the experimental data with the pseudo-second-order kinetic model indicates that the rate limiting step for the adsorption of MO ions on the organometallic-functionalized SiO₂–Al₂O₃ mixed-oxide is chemical adsorption (Scheme 2). As a result, the adsorption of the MO dye onto the nano-adsorbent may be considered to involve two processes with initial adsorption rate of 144.9 mg (g min)⁻¹ over Mn@Si/Al. Although this adsorption rate is related to the content and type of active adsorption site on the matrix of the adsorbent, Mn(III) sites are the main reactive groups for the removal of MO ions from aqueous solution. A larger advantage of the pseudo-second-order model is that it predicts the behavior over the whole range of the adsorption process.

3.2.6. Effect of the temperature on the uptake. Furthermore, the equilibrium adsorption capacity of MO onto the favored adsorbent, Mn@Si/Al, were studied at a range of different temperatures of 15, 25, 40, 60 and 80 °C in pH = 6.5 (Fig. 12). The increase in the temperature of the solutions of MO from 15 to 80 °C leads to an increase in the adsorption capacity of Mn@Si/Al, thereby indicating that the adsorption of the dye onto the active sites of Mn@Si/Al is endothermic, possibly due to the availability of more such active sites and the enlargement and activation of the adsorbent surface at higher temperatures. It could also be due to the increased mobility of MO dye ions from the bulk solution towards the adsorbent surface, thereby enhancing the accessibility to the adsorbent active sites.

Fig. 12 Effect of temperature on the adsorption of MO by Mn@Si/Al at different initial concentrations.

To gain a better understanding of the effect of rising temperature on the adsorption of the dye ions onto Mn@Si/Al, three basic thermodynamic parameters were studied: the Gibbs free energy of adsorption (ΔG°), the enthalpy change (ΔH°), and the entropy change (ΔS°).

The thermodynamic parameters ΔG°, ΔS° and ΔH° for this adsorption process were determined using the following equation.

ΔG = −RT ln K (8)

where K is the thermodynamic equilibrium constant. The value of K can be determined by plotting ln(q_e/C_e) against q_e and extrapolating to zero, where q_e is the adsorbed MO ion concentration at equilibrium and C_e is the equilibrium concentration of MO ions in solution. The effect of temperature on the thermodynamic constant was determined by eqn (9)

dln K/dt = ΔH°/RT² (9)

The ΔH° and ΔS° values were calculated from the slope and intercept of the linear plot of ln K vs. 1/T, as shown in Fig. 13.

ln K = ΔS°/R − ΔH°/RT (10)

and the Gibbs free energy is given by eqn (11), where ΔG° is the free energy change (J mol⁻¹) and R and T are the universal constant (8.314 J mol⁻¹ K⁻¹) the absolute temperature (K), respectively.

ΔG° = ΔH° − TΔS° (11)

Fig. 13 Plot of ln K vs. 1/T for the MO adsorption onto Mn@ Si/Al.

The corresponding values of the thermodynamic parameters are presented in Table 5. This shows that ΔH° and ΔS° are positive for all the experiments and ΔG° is negative in all the systems. The positive values of ΔH° reveal that the adsorption process is endothermic in nature and thus, the removal of MO increases with increasing temperature. The positive value of ΔS° reveals an increase in randomness and an increase in the degrees of freedom at the Mn@Si/Al–solution interface during the immobilization of the MO dye ions on the active sites of the adsorbent, where methyl orange is a bulk molecule in comparison with H₂O, so some water molecules could be desorbed during the adsorption of a methyl orange molecule. This process led to a partial liberation of the solvated MO ions from the solvent molecules before adsorption (liberation of water molecules from solvated-MO), thereby enabling commonality of randomness and spontaneity in the system.^1,2 The necessity of a large amount of heat to remove the MO dye ions from the solution makes the sorption process endothermic. A positive ΔS° (disorder of the system) was observed on Mn@Si/Al, which indicates that the MO ions lose most of their water of solvation. This is also supported by the positive ΔH° value of MO sorption onto Mn@Si/Al. The positive value of the standard enthalpy change for MO ion sorption indicates the endothermic nature of the adsorption. It was also observed that with an increase in temperaturem, the value of ΔG decreases, which indicates that the sorption process is spontaneous and thermodynamically favorable by an increase in temperature (Table 5). In fact, the value of ΔG° for physisorption is in the range −20 > ΔG° < 0 kJ mol⁻¹, but chemisorption is between −400 > ΔG° < −80 kJ mol⁻¹.⁴⁷ The calculated ΔG° values based on eqn (14) were more than −40 kJ mol⁻¹ for most cases (Table 5). Therefore, the ΔG° values suggest that the adsorption of MO ions onto Mn@Si/Al is a chemisorption process, as the adsorption mechanism is proposed in Scheme 2.

Table 5 Thermodynamic parameters for the adsorption of MO onto Mn@Si/Al as a function of temperature

Initial MO concentration	ΔH°	ΔS°	ΔG°/kJ mol^−1
mg L^−1	(J mol^−1)	(J mol^−1 K^−1)	288	298	313	333	353
150	95.45	348.11	−100.16	−103.64	−108.86	−115.82	−122.79
250	80.00	298.58	−85.911	−88.896	−93.375	−99.347	−105.31
350	67.25	240.98	−69.334	−71.744	−75.359	−80.179	−84.998
450	55.18	199.84	−57.498	−59.497	−62.494	−66.491	−70.488
550	32.84	119.45	−34.368	−35.563	−37.355	−39.744	−42.133
650	31.66	113.26	−32.587	−33.719	−35.418	−37.683	−39.949
750	23.15	82.66	−23.782	−24.609	−25.849	−27.502	−29.155
850	20.93	73.30	−21.089	−21.822	−22.921	−24.387	−25.853
1000	16.73	57.33	−16.494	−17.067	−17.927	−19.074	−20.220

---

To prove the proposed mechanism in Scheme 2, FT-IR spectra of Mn@Si/Al and MO-loaded Mn@Si/Al samples, before and after the adsorption process, were recorded in the range 4000–400 cm⁻¹ (Fig. 14). FT-IR is often used to study the active sites of the adsorbents and to identify the interactions of those groups responsible for the dye adsorption. The adsorption ability of Mn@Si/Al for MO dye ions from aqueous solution results from the strong interaction of the active sites of the immobilized Mn-nanoparticles towards the dye. In the equilibrated sample of Mn@Si/Al with a MO solution, extra bands at 1729, 1606, 1447, 1421, 620 and 577 cm⁻¹ were observed, showing the presence of MO anchored to the active sites of the modified nano-adsorbent. The peak at 1729 cm^{− 1} is enhanced in Mn@Si/Al–MO. This may belong to a coordinated sulfonate group, and generally appears when the bond between the metal and oxygen is strong (Mn–O–SO₂–R, Scheme 2).⁴⁸ The modes of MO ion coordination with the immobilized Mn(III) is evidenced through the O atoms of the sulfonate group appearing as a new band at around 577 cm⁻¹ (Fig. 14). Indeed, the new absorption peaks observed in the range of 560 to 660 cm⁻¹ are the characteristic absorption peaks of a “Metal–O” bond, which demonstrates the existence of the Mn–O bond (Mn–O–SO₂–R, Scheme 2).⁴⁹ In addition, the kinetics results proved that the adsorption obeys a pseudo second-order kinetics model, which indicates that the adsorption involves chemisorption.

Fig. 14 FT-IR spectra of Mn@Si/Al before and after adsorption of MO from aqueous solution. Legends: Mn@Si/Al (A), Mn@Si/Al after MO adsorption (B).

In order to provide further clarification into the mechanistic aspects of the MO adsorption onto Mn@Si/Al, a desorption study was investigated. The use of water and ethanol solutions for the MO desorption are ineffective, even after 48 h. Very low desorption (does not exceed 7% of the presorbed amount) with these solutions suggests that metal–dye complexes were formed on Mn@Si/A. The high basic solution (above 0.01 mol L⁻¹) did help in breaking the chelating interaction, in which the hydroxyl groups replaced the MO to bind Mn@Si/A.

4. Conclusion

This work shows that a mixed-oxide modified with Mn-nanoparticles could find application as an adsorbent for the removal of hazardous dye ions from aqueous solutions. The synthesized material was verified by TEM, SEM, EPR, XPS, ICP, BET, UV-vis and FT-IR analysis. Several adsorption conditions, such as initial dye concentration, contact time, adsorbent dosage and pH on the adsorption capacity of MO will influence the dye adsorption. The kinetic data indicates that the pseudo-second order kinetic model was found to be well suited and provide a high degree of correlation with the experimental data for the adsorption process of MO on Mn@Si/Al. The immobilization of Mn-nanoparticles produces a positive charge on the surface (Mn(III)), which will provide a strong attraction to the negatively charged MO ions. Adsorption of the MO dye on Mn@Si/Al is an endothermic process. The strong chemical binding between the dye and the active sites of the nano-adsorbent results in much less desorption of MO from the solid adsorbent surface.

Acknowledgements

Thanks are due to the Iranian Nanotechnology Initiative for supporting this work.

References

1. L. Zhou, C. Gao and W. Xu, ACS Appl. Mater. Interfaces, 2010, 2, 1483.
2. T. Robinson, G. McMullan, R. Marchant and P. Nigam, Bioresour. Technol., 2001, 77, 247.
3. M. Arshadia, F. Salimi Vahid, J. W. L. Salvacion and M. Soleymanzadeh, Appl. Surf. Sci., 2013, 280, 726.
4. H. Y. Zhu, R. Jiang, L. Xiao and G. M. Zeng, Bioresour. Technol., 2010, 101, 5063.
5. J. H. Huang, K. L. Huang, S. Q. Liu, A. T. Wang and C. Yan, Colloids Surf., A, 2008, 330, 55.
6. P. Zhang, Q. An, J. Guo and C. C. Wang, J. Colloid Interface Sci., 2013, 389, 10.
7. W. Cheah, S. Hosseini, M. A. Khan, T. G. Chuah and T. S. Y. Choong, Chem. Eng. J., 2013, 215–216, 747.
8. Z. M. Ni, S. J. Xia, L. G. Wang, F. F. Xing and G. X. Pan, J. Colloid Interface Sci., 2007, 316, 284.
9. J. Ma, F. Yu, L. Zhou, L. Jin, M. X. Yang, J. S. Luan, Y. H. Tang, H. B. Fan, Z. W. Yuan and J. H. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 5749.
10. M. R. Samarghandi, M. Hadi, S. Moayedi and F. B. Askari, Iran. J. Environ. Health Sci. Eng., 2009, 6, 285–294.
11. Y. J. Yao, B. He, F. F. Xu and X. F. Chen, Chem. Eng. J., 2011, 170, 82–89.
12. G. X. Li, Y. M. Du, Y. Z. Tao, H. B. Deng, X. G. Luo and J. H. Yang, Carbohydr. Polym., 2010, 82, 706.
13. Y. Wang, X. Liu, H. Wang, G. Xi, W. Huang and R. Song, J. Colloid Interface Sci., 2014, 416, 243.
14. Y. Haldorai and J.-J. Shim, Appl. Surf. Sci., 2014, 292, 447.
15. T. Kou, Y. Wang, C. Zhang, J. Sun and Z. Zhang, Chem. Eng. J., 2013, 223, 76.
16. L. I. U. Zhuannian, Z. H. O. U. Anning, W. A. N. G. Guirong and Z. H. A. O. Xiaoguang, Chin. J. Chem. Eng., 2009, 17(6), 942.
17. T. K. Saha, N. C. Bhoumik, S. Karmaker, M. G. Ahmed, H. Ichikawa and Y. Fukumori, Journal of Water Resource and Protection, 2010, 2, 898.
18. A. J. Aishah, T. S. Sugeng, A. Hazirah, N. D. Rahim, M. Arif and A. Aziz, J. Hazard. Mater., 2010, 181, 755.
19. W. Deligeer, Y. W. Gao and S. Asuha, Appl. Surf. Sci., 2011, 257, 3524.
20. H. Y. Zhu, R. Jiang, Y. Q. Fu, J. H. Jiang and L. Xiao, Appl. Surf. Sci., 2011, 258, 1337.
21. Z. B. Hassina, B. Mokhtar and A. Loubna, Chem. Eng. J., 2012, 187, 142.
22. R. Huang, Q. Liu, J. Huo and B. Yang, Arabian J. Chem., 2013 DOI:10.1016/j.arabjc.2013.05.017.
23. N. Mohammadi, H. Khani, V. Kumar Gupta, E. Amereh and S. Agarwal, J. Colloid Interface Sci., 2011, 362, 457.
24. A. Mittal, A. Malviya, D. Kaur, J. Mittal and L. Kurup, J. Hazard. Mater., 2007, 148, 229.
25. E. Haque, J. Eun Lee, I. Tae Jang, Y. Kyu Hwang, J.-S. Chang, J. Jegal and S. Hwa Jhung, J. Hazard. Mater., 2010, 181, 535.
26. Y. Tian, X. Wang and Y. Pan, J. Hazard. Mater., 2012, 213–214, 361.
27. J.-H. Huang, K.-L. Huang, S.-Q. Liu, A.-T. Wang and C. Yan, Colloids Surf., A, 2008, 330, 55.
28. G. Annadurai, R. S. Juang and D. J. Lee, J. Hazard. Mater., 2002, B92, 263.
29. J. Liu, S. Ma and L. Zang, Appl. Surf. Sci., 2013, 265, 393.
30. H. Chen, A. Zhong, J. Wu, J. Zhao and H. Yan, Ind. Eng. Chem. Res., 2012, 51, 14026.
31. X. S. Zhao, G. Q. Lu and X. Hu, Microporous Mesoporous Mater., 2000, 41, 37.
32. Y. Luo and J. Lin, Microporous Mesoporous Mater., 2005, 86, 23.
33. R. Wang, B. Gao and W. Jiao, Appl. Surf. Sci., 2009, 255, 4109.
34. R. Mukhopadhyay, S. Bhattacharjee and R. Bhattacharyya, J. Chem. Soc., Dalton Trans., 1994, 2799.
35. J. A. Cubillos and W. F. Hölderich, Rev. Fac. Ing., Univ. Antioquia, 2007, 41, 31.
36. J. Tedim, S. Patrício, R. Bessada, R. Morais, C. Sousa, M. B. Marques and C. Freire, Eur. J. Inorg. Chem., 2006, 3425.
37. D. L. Griscom and R. E. Griscom, J. Chem. Phys., 1967, 47, 2711.
38. D. Y. Zhao and D. Goldfarb, in Zeolites: A Refined Tool for Designing Catalytic Sites, ed. L. Bonneviot and S. Kaliaguine, Elsevier, Amsterdam, 1995, p. 181.
39. G. Brouet, X. Chen, C. W. Lee and L. Kevan, J. Am. Chem. Soc., 1992, 114, 3720.
40. W. S. Kijlstra, E. K. Poels, A. Bliek, B. Weckhuysen and R. A. Schoonheydt, J. Phys. Chem. B, 1997, 101, 309.
41. B. J. Aronson, C. F. Blanford and A. Stein, J. Phys. Chem. B, 2000, 104, 449.
42. V. Di Castro and G. Polzonetti, J. Electron Spectrosc. Relat. Phenom., 1989, 48, 117.
43. C. Bauer, P. Jacques and A. Kalt, Chem. Phys. Lett., 1999, 307, 397.
44. H. Tamai, T. Yoshida, M. Sasaki and H. Yasuda, Carbon, 1999, 37, 983.
45. R. Cheng, S. Ou, B. Xiang, Y. Li and Q. Liao, Langmuir, 2010, 26(2), 752.
46. Q. Zhang, Y. Wang, S. Itsuki, T. Shishido and K. Takehira, J. Mol. Catal. A: Chem., 2002, 188, 189.
47. R. S. Juang and J. Y. Chung, J. Colloid Interface Sci., 2004, 275, 53.
48. C. N. R. Rae, Chemical Applications of Infrared Spectroscopy, Academic Press, New York, 1963, pp. 337. 364.
49. W. Ma, F. Q. Ya, M. Han and R. Wang, J. Hazard. Mater., 2007, 143, 296.

---