sodium alginate

International Journal of Biological Macromolecules 116 (2018) 607–619

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules journal homepage: http://www.elsevier.com/locate/ijbiomac

Synthesis of zeolite/nickel ferrite/sodium alginate bionanocomposite via a co-precipitation technique for efficient removal of water-soluble methylene blue dye Mahsa Bayat, Vahid Javanbakht ⁎, Javad Esmaili ACECR Institute of Higher Education (Isfahan Branch), Isfahan 84175-443, Iran

a r t i c l e

i n f o

Article history: Received 13 February 2018 Received in revised form 30 April 2018 Accepted 2 May 2018 Available online 05 May 2018 Keywords: Nanocomposite Alginate Methylene blue

a b s t r a c t In this study, we sought to synthesize magnetic nanocomposite of zeolite/nickel ferrite through co-precipitation method and modify its surface by sodium alginate to enhance its methylene blue adsorption capacity and to prevent its oxidation. Nanocomposite characteristics were investigated by SEM, VSM, XRD and FTIR analyses. The results indicate that nanocomposite synthesis and modification has been completely successful. Adsorption thermodynamics, kinetics, and isotherms were examined and parameters were optimized by Minitab software using experimental design method, response surface methodology and Box-Behnken design. The highest capacity of methylene blue adsorption from the aqueous solution obtained at optimal pH of 5, the initial dye concentration of 10 mg/L and an adsorbent amount of 0.03 g was about 54.05 mg/g. Analyzing kinetic data of adsorption experiments confirmed that adsorption process complies with the pseudo-second-order kinetic model. Assessing equilibrium isotherm data at different temperatures showed that these data are in good agreement with Langmuir isotherm model. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Today, environmental pollution is considered to be a global challenging problem. Toxic wastewaters generated by different industries have detrimental effects on water resources, soil fertility, aquatic organisms and ecosystems integrity [1]. Dye pollutants are introduced to the environment by various industries such as textiles, pulp, and paper, pharmaceuticals and leather industries which use different colored chemicals [2]. Colors are one of the most dangerous chemical compounds found in industrial effluents [3]. These compounds cause allergies, dermatitis, skin irritation, cancer, as well as genetic mutations in humans [4,5]. Methylene blue is the most common color combination used for cotton, wool and silk dyeing. Inhalation of this compound can lead to respiratory impairment, while direct exposure to it can cause permanent damages to human and animal eyes, local burns, increased sweating and mental disorders [6]. Various dye removing methods including coagulation and flocculation, chemical oxidation, reverse osmosis, and adsorption have been investigated [7]. Adsorption is a sustainable process and an effective way to remove pollutants from water, which far outstrips other methods because of its low cost, high flexibility, easy design and manufacturing and higher sensitivity to toxic substances [8]. In the recent decades, a lot of efforts have been ⁎ Corresponding author. E-mail address: [email protected] (V. Javanbakht).

https://doi.org/10.1016/j.ijbiomac.2018.05.012 0141-8130/© 2018 Elsevier B.V. All rights reserved.

made to produce and use inexpensive materials to remove heavy metals and organic pollutants from water and industrial wastewaters. Zeolites are one of the minerals widely used for contaminants removal. Zeolites are hydrated crystalline aluminous silicates of alkaline and alkalineearth metals, especially sodium, calcium, magnesium, strontium, and barium that have three-dimensional networks. This adsorbent is widely used to remove contaminants because of its high surface area, spherical shape, and high cation exchange capacity. Clinoptilolite is one of the most commonly used natural zeolites [9]. Due to the dispersion of zeolite particles in the solution and separation difficulties, magnetizing adsorbents to facilitate their removal from adsorbent-containing suspensions and pollutant-containing solutions is of interest. In the recent years, the separation of magnetic adsorbents using magnet has replaced the other methods such as sedimentation, centrifugation, and filtration [10]. Magnetic zeolites are considerably used in the water and wastewater treatment industry to remove pollutants such as phthalates, benzene, and heavy metals. In this study, nickel ferrite nanoparticles which grant magnetic properties to the adsorbents are used. That is to say, once settling in a magnetic field, they acquire the magnetic property needed for adsorption and separation; a property which is very interesting in removing adsorbent after the adsorption process. Ferrites are referred to a group of magnetic materials that have iron oxide as their main constituent and their metallic ions are arranged in two separate sub-networks in an antiparallel manner. High magnetism, low coercivity, and high specific electrical resistivity are the main characteristics

608

M. Bayat et al. / International Journal of Biological Macromolecules 116 (2018) 607–619

of ferrites [11]. Despite the advances in the synthesis of magnetic nanoparticles, preventing the accumulation of magnetic nanoparticles has remained an important issue. The main problem is caused by nanoparticles susceptibility to oxidation, which increases as the size of the particle decreases. Therefore, proposing an effective method to establish the chemical stability of nanoparticles is essential. Protecting nanoparticles with different polymers such as sodium alginate can be achieved by an impermeable layer that prevents oxygen from reaching the nanoparticle surface. Sodium alginate is an inexpensive, available, non-toxic brown seaweed-derived biopolymer which is effective in removing contaminants from the aqueous environment. This polymer has attracted a lot of attention because of properties such as biodegradability and natural nature and has been used for dye and even heavy metals removal [12]. Among different studies various low cost adsorbents have been used for methylene blue from wastewaters such as activated carbon, lignocellulosic biomasses, and natural zeolites, but due to the separation difficulties of these adsorbent powders from the dye solution, magnetizing them with magnetic iron oxide nanoparticles to facilitate their separation is investigated [13–17]. Magnetic iron oxide nanoparticles have high chemical activity and are easily oxidized in the air generally resulting in loss of magnetism. Therefore, providing proper surface coating such as natural polymers to keep the stability of magnetic iron oxide nanoparticles is very important [18]. In this research, natural zeolite

Clinoptilolite was magnetized by Nickel Ferrite (NiFe2O4) magnetic nanoparticles with precipitation of Ni2+ and Fe3+ by NaOH solution as precipitator and then, the surface of the obtained nanocomposite was coated with Sodium Alginate as a natural polymer and subsequently, its efficiency in methylene blue removal from aqueous solution was evaluated. To investigate the parameters affecting the methylene blue adsorption process (concentration of dye, pH, and adsorbent amount) and optimization of them, response surface methodology and BoxBehnken design were used. Furthermore, different kinetic and isotherm models and thermodynamic studies were also studied. 2. Experimental methods 2.1. Materials To prepare nickel ferrite nanoparticles, Iron (III) chloride hexahydrate (FeCl3·6H2O) and nickel (II) nitrate hexahydrate (Ni(No3)2·6H2O) both purchased from Merck Inc., 98% sodium hydroxide (NaOH) obtained from SAMCHUNG Co. Ltd. and distilled water was used. Nanocomposite was prepared using powdered Clinoptilolite with a mean particle size of 1 μm and the chemical formula of KNa2Ca2 (Al7Si29) O72·24H2O which was acquired from central Alborz mountain range located in Semnan province (see Fig. 1). 37% hydrochloric acid (HCl) was purchased

Fig. 1. Schematic of (a) synthesis of the magnetic nanocomposite, (b) crosslinking the sodium alginate chains by CaCl2 crosslinker.


609

from SAMCHUNG Co. Crosslinking calcium chloride (CaCl2) and methylene blue (C16H18N3SCl) acquired from Merck Co. Sodium alginate (C5H7O4COONa) was purchased from Sigma Aldrich Co.

The residual concentration of methylene blue was calculated using the equation of the line derived from the calibration curve. The adsorption capacity can be calculated using the following Eq. (1):

2.2. Preparation of clinoptilolite/nickel ferrite nanocomposite

qe ¼

Before nanocomposite synthesis, zeolite was processed and amended. 5 g of zeolite was rested in 5 M HCl solution for 24 h. Then, the processed product was rinsed three times with distilled water to allow its pH to be neutralized and the acid to be completely washed. Since zeolite is a humidity adsorbent it is very difficult to dry it at room temperature, so it was dried in the oven at low temperature. To synthesize nanocomposite, the amended zeolite, with an amount equivalent to the processed product, was added to water and was then put on the stirrer for 10 min at room temperature. 2 M FeCl3·6H2O, 1 M Ni (NO3)·6H2O and 8 M NaOH solutions were separately prepared using distilled water. Afterwards, salt solutions were added to zeolite suspension. Then, sodium hydroxide solution was slowly added to the solution of zeolite and salts as a precipitation agent under semi-anaerobic conditions at 80 °C. The injection took about an hour, and the light brown color of the solution eventually turned into dark brown. The pH of the solution was continuously increased by adding NaOH. Once injection was finished, the solution was allowed to reach room temperature, and after precipitation, it was rinsed thrice with distilled water and ethanol. After a 15 min centrifugation, it was placed in an oven at 80 °C for 8 h. The product was a brown powder which was finally put in a furnace at 600 °C for 7 h to form nickel ferrite magnetic nanoparticles. The final product was zeolite nanocomposite magnetized by nickel ferrite nanoparticles.

ðC0 −Ce Þ V M

where q is adsorbent capacity (mg/g), C0 is the initial concentration (mg/l), Ce is the equilibrium concentration (after adsorption) (mg/l), V is the solution volume (ml), and M denotes adsorbent mass (g) [19]. 2.6. Box-Behnken experimental design and optimization by RSM The most appropriate method to determine the best conditions for pollutants removal from domestic and industrial wastewaters is to optimize the factors affecting the pollutants removal process. Optimization of dye removal process by changing 3 independent factors including the adsorbent amount, pH, and initial dye concentration, was estimated using response surface methodology (RSM) and Box-Behnken design (BBD). RSM is composed of various experimental techniques which evaluate the relationship between measured response and empirical factors with respect to different indices [20]. The general equation of RSM is in the form of Eq. (2). In this equation ×1, ×2, …, Xk are input variables, β0 is independent parameter of regression equation, βis, βiis, and βijs (i = 1, 2, …, k; j = 1, 2, …, k) are linear effect parameters, quadratic equation effects parameters, and interactive effects parameters. ε is random error and Y is the predicted Response [21]. Y ¼ β0 þ

2.3. Preparation of clinoptilolite/nickel ferrite/sodium alginate nanocomposite 0.1 g of zeolite magnetized by nickel ferrite nanoparticles was poured into a 5% w/w CaCl2 solution and was placed on a stirrer for 1 h. Then, 0.2 g of sodium alginate with a weight ratio of 1:5 to adsorbent was dissolved in 30 cm3 of distilled water and was then added drop by drop to the solution of adsorbent and calcium chloride during a 1 h period and in the meantime, the solution was mixed using a stirrer. After the injection, the solution was placed on a magnet and was washed three times with distilled water after precipitation to remove excess unreacted calcium chlorides. Eventually, the product was dried in an oven at 70 °C. The final product was a powdered nanocomposite amended with sodium alginate.

ð1Þ

k X

βi X i þ

i¼1

k X i¼1

βii X i 2 þ

k X k X

βij X i X j þ ε

ð2Þ

i¼1 j¼1

This method consists of Doehlert matrix, central composite design, three-level full factorial and Box-Behnken design (BBD). Box-Behnken and Doehlert matrix are more effective than central composite design and three-level factorial design [22]. The Box-Behnken is a design for rotatable multivariable optimization for quadratic equation estimation based on three levels of −1, 0, and 1 [23]. The number of experiments in this design can be obtained from Eq. (3) where k is the number of independent parameters, C0 is the number of central points and N is the number of experiments which is less than the one obtained from (4) for central composite design [22]. N ¼ 2kðk−1Þ þ C 0

ð3Þ

N ¼ 2k þ 2k þ C 0

ð4Þ

2.4. Characterization The crystalline and phase structures of nanocomposite were determined using a radiation of Cu Kα (λ = 0.154) in the range of 10–80 (2Ɵ) with JEOL JDX-8030 diffractometer (X-Ray Diffraction (XRD)). To determine the type of synthetic materials and their functional groups, and to ensure the success of nanocomposite preparation, Fourier Transform Infrared Spectrometer Jasco-4200 (FTIR) was used. Magnetic properties of nanocomposite and its saturation magnetization were determined using vibrating magnetic magnetometer (VSM, MPMS-5 SQUID). Surface properties of amended natural zeolite and zeolite/nickel ferrite nanocomposite modified by sodium alginate were assessed using scanning electron microscope (SEM, Zeiss EVO 50electron). 2.5. Methylene blue removal study Methylene blue solutions were prepared by dissolving methylene blue in distilled water with concentrations of 10–50 mg/l. 0.1 M NaOH and 0.1 M HCl solutions were used to adjust pH between 2 and 6. A certain amount of magnetic nanocomposite was dispersed in 25 ml of methylene blue solution. The resulting solution was stirred at 200 rpm. The magnetic adsorbent was then removed from the solution using a magnet and the adsorption rate was measured by a spectrophotometer.

In this research, the design of experiments was carried out using MINITAB 17 which is a suitable software for analysis of experimental data. In the designed statistical model, appropriate levels were designated for each factor and after that, a specified number of experiments were randomly designed and carried out. Since there were 3 independent factors and 3 replications at center points, 15 experiments were performed in this study. Depending on the chosen parameters, the optimal value was evaluated using regression equations and RSM diagrams analysis. Table 1 shows the factors and experimental levels for each of them.

Table 1 Parameters and experimental levels for each parameter in Box-Behnken experimental design. Levels

Operational parameters Adsorbent dosage (g) Dye concentration (mg/l) pH

Low

Center

High

−1 0.01 10 2

0 0.03 30 5

+1 0.05 50 8

610


Fig. 2. SEM images of (a) amended Zeolite; (b) Zeolite/Nickel Ferrite/Alginate nanocomposite.

3. Results and discussion 3.1. Characterization investigations 3.1.1. SEM observations SEM images were used to assess surface properties and also the appearance of amended natural zeolite and Zeolite/Nickel Ferrite nanocomposite modified by sodium alginate. As can be

seen, the general structure of ferrite- and alginate-coated zeolite has not changed. The morphology of the amended zeolite illustrated in Fig. 2a indicates numerous pores on the surface. The SEM analysis of alginate-amended nanocomposite depicted in Fig. 2b shows that the adsorbent has a non-uniform and heterogeneous surface that its alginate coating leads to separation of the particles and prevents the agglomeration of Zeolite/Nickel Ferrite.

Fig. 3. Magnetization curve of Zeolite/Nickel Ferrite/Sodium Alginate nanocomposite.


611

Fig. 4. XRD results of (a) Clinoptilolite, (b) Nickel Ferrite nanoparticles, (c) Clinoptilolite/Nickel Ferrite nanocomposite, and (d) Clinoptilolite/Nickel Ferrite/Sodium Alginate nanocomposite.

3.1.2. VSM results The magnetic property is one of the main properties of the synthesized nanocomposite which is characterized by VSM analysis (Fig. 3). According to magnetometry results, the synthetic sample S-shaped curve which implies that nanocomposite is superparamagnetic. The horizontal line represents zero saturation magnetization of zeolite which reaches about 12 emu/g after combining with nickel ferrite. In addition, the saturation magnetization of the nanocomposite sample amended with sodium alginate is about 6 emu/g which indicate the ease of nanocomposite separation from aqueous suspension after

nanocomposite using in adsorption process which is considered as one of the applicable advantages of nanocomposite magnetized by nickel ferrite. 3.1.3. X-ray diffraction results The XRD analysis determines the crystalline properties and the formation of the zeolite/nickel ferrite/alginate nanocomposite. The X-ray diffraction pattern in Fig. 4 shows the formation of nickel ferrite. The distinctive peaks at 2θ = 30.41, 35.69, 43.45, 37.30, 54.05, 57.48 and 63.08 represented respectively by (2,2,0), (3,1,1), (0,0,4), (4,2,2),

Fig. 5. FTIR spectrum of (a) Clinoptilolite, (b) Magnetic nanoparticles of Nickel Ferrite, (c) Sodium alginate, and (d) Clinoptilolite/Nickel Ferrite/Alginate nanocomposite.

612


(1,1,5) and (4,4,0) (Reference code: 03-0875) indicate the crystal structure of nickel ferrite [24]. The zeolite/nickel ferrite nanocomposite sample is the result of the coprecipitation of metal oxides and composite subsequent to nickel ferrite precipitation on the zeolite surface. The XRD analysis of the sample shows the crystal structure of the nanocomposite while the peaks at 2θ = 35.69, 63.21 and those observed at 2θ = 22.41 and 25.74 indicate the presence of nickel ferrite and zeolite, respectively [25]. XRD analysis of amended zeolite/nickel ferrite/sodium alginate nanocomposite in Fig. 4d shows that, in addition to the zeolite peaks, the nickel ferrite peaks have also appeared in the amended nanocomposite without causing any damage to the zeolite peaks which suggests that the amended nanocomposite has retained its original structure. Using the Debye-Scherrer formula (Eq. (5)), the average size of the particles can be obtained, which in this study was calculated to be 48 nm: D¼

Kλ βCOSθ

ð5Þ

In this equation K is the factor of crystal shape (0.9), λ is the wavelength of the X-ray tube (if Cu Kα radiation is used, λ = 0.154 nm), β is peak's full width at half maximum height, θ is the diffraction angle, and D is the average size of particles (nm).

3.1.4. FTIR result Infrared spectroscopy is based on the absorption of radiation and the evaluation of the vibrational motions of molecules and polyatomic ions. The infrared spectrum of zeolite clinoptilolite is shown in Fig. 5. This spectrum shows that absorption peak lied at 3420.14 cm−1 is related to the outer surface chemical groups of the clinoptilolite crystal structure, which overlaps wide and strong peak appeared at 3000–3420.14 cm−1 region that is associated with the stretching vibrations caused by O\\H. The peak observed at 1641.13 cm−1 is related to bending and stretching vibrations of O\\H bond. The absorption peaks centered at 797.42–1500 cm−1 regions result from the bending and stretching vibrations of Si\\O or Al\\O bonds in the Clinoptilolite structure [26]. In Fig. 5b the strong absorption peaks at 464.76 cm−1 and 608.43 cm−1 which belong to Fe\\O bond show that ferrite is formed. In general, the wavelengths below 700 are associated with the Fe\\O bond. The bands appeared at 400–650 cm−1 region are related to the stretching vibrations of Fe\\O bond in tetrahedral and octahedral sites [27]. As can be seen in Fig. 5c the absorption bands of the carboxyl, ether, and hydroxyl groups have appeared well in the infrared spectrum of sodium alginate. Since the stretching bands of alginate O\\H bond have appeared at 3000–3600 cm−1 region, it can be concluded that the peak at 3449.06 cm−1 belongs to this bond. Bands observed at 2932.23 cm−1 can be attributed to the stretching vibrations of C\\H

Fig. 6. (a) Methylene blue adsorption kinetics, concentration (mg/l) decay and adsorbent capacity (mg/g) against time, (b) Pseudo-first-order kinetic model, (c) Pseudo-second-order kinetic model (d) Elovich model (e) Intraparticle diffusion model (0.025 g of adsorbent, initial Methylene blue concentration = 20 mg/l, pH = 6, solution volume = 50 ml, shaking speed = 200 rpm, and contact time = 240 min). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


aliphatic groups. Asymmetric and symmetric absorption bands of the carboxylate group of sodium alginate are shown by peaks centered at 1623.77 and 1411.64 cm−1, respectively. Finally, the bands observed at 1090, 1026.91 and 946.877 cm−1 are attributed to the stretching vibrations of C\\O bonds of etheric groups in the structure [28]. According to the infrared spectrum of the Clinoptilolite/Nickel Ferrite/Alginate nanocomposite depicted in Fig. 5d, not only zeolite and nickel ferriterelated peaks but also sodium alginate-related bands appeared at 1634.34, 2920.06 and 3402.78 cm−1 indicate that zeolite/nickel ferrite nanocomposite is successfully coated by alginate. 3.1.5. Estimation of the adsorbent surface area The nanocomposite surface area was measured by methylene blue adsorption as a monolayer on solid sorbents [29]. According to the adsorption experimental data, the specific surface area of the nanocomposite was calculated using the following equation [29]: S¼

QNA Scs 10−20 mMw

ð6Þ

In this equation S, Q, NA, Scs, MW, and m are the nanocomposite specific surface area in m2/g, the mass of methylene blue adsorbed (g), the Avogadro's number (6.02 × 1023 mol−1), the methylene blue molecular

613

cross-section (197.2°A2), the molecular weight of methylene blue (373.9 g mol−1), and the mass of adsorbent (g), respectively. According to this equation, the surface area of the nanocomposite was calculated about 219.07 m2/g. 3.2. Adsorption studies 3.2.1. Modeling and determination of kinetic parameters Chemical kinetics explains the reaction pathways needed to reach the equilibrium, while chemical equilibrium doesn't provide any information regarding these pathways and reaction rates. Adsorption kinetics represents the strong relationships between chemical and physical properties of adsorbent, adsorbate, and the effects of adsorption mechanism. The adsorption process mechanisms including chemical reaction, diffusion-controlled reaction, and mass transfer were assessed using several kinetic models under different laboratory conditions [30]. Pseudo-first-order kinetic model is defined by Eq. (6) in which qe and qt are adsorption capacity at equilibrium and at the time t (mg g−1), respectively and k1 is the rate constant of pseudo-firstorder adsorption (min−1) ln ðqe −qt Þ ¼ lnqe −k1 t

ð7Þ

Fig. 7. (a) the experimental data isotherm, (b) the effect of temperature on adsorption capacity, (c) Langmuir isotherm model, (d) Freundlich isotherm model, (e) Temkin isotherm model, (e) Dubinin-Radushkevich isotherm model (0.025 g of adsorbent, initial Methylene blue concentration = 10–50 mg/l, pH = 6, solution volume = 50 ml, shaking speed = 200 rpm, and contact time = 120 min). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

614


The pseudo-second-order kinetic model is represented in linear form as Eq. (8) where K2 expresses the constant of pseudo-secondorder adsorption rate.

4.5

t 1 t ¼ þ qt k2 qe2 qe

3.5

ð8Þ

3

The pseudo-first-order and pseudo-second-order models illustrate diffusion in the external film, intraparticle diffusion, and interaction between adsorbate and adsorbent's functional groups [25,31]. Intraparticle diffusion is the adsorptive transfer from solution phase to the adsorbent phase which is expressed by Eq. (9): qt ¼ ki t 0:5 þ C

ð9Þ

where, ki and C illustrate the constant of intraparticle motion and the constant of diffusion rate in the boundary layer, respectively. Elovich model as Eq. (10) represents the kinetics of chemical adsorption by heterogeneous adsorbents. In this equation, α is constant of initial adsorption/desorption rate and β is activation energy of the adsorbent [18]. qt ¼

1 1 ln ðαβ Þ þ β β

4

ð10Þ

The adsorption of 20 mg l−1 dye on the synthesized nanocomposite was evaluated at different times. The equilibrium concentration and adsorbent capacity against time are shown in Fig. 6. As can be seen, during the first 10 min, approximately 85% of methylene blue is adsorbed and as time passes adsorption percentage and capacity increases while adsorption rate declines because a layer of adsorbed molecules covered the surface of adsorbent and a repulsive force between adsorbent surface charges and like dye charges in the solution is developed. This increment in the capacity of adsorption continues until the adsorption equilibrium is reached and adsorbent becomes saturated. Given the adsorbent shape, it takes about 120 min for the reaction to reach equilibrium and after that, a steady trend appears in the adsorption curve. The values of determination factors (R2) obtained from kinetic models of methylene blue adsorption revealed that the dye adsorption adjusts more with the pseudo-second-order kinetic model. Furthermore, intraparticle diffusion as a transitional process was used to investigate the adsorption mechanism. Fig. 6e illustrates that the process of dye adsorption is more than one diffusion step. In addition, slopes of three straight lines Ki,1 N Ki,2 N Ki,3 represents three stages of dye adsorption at the outer surface, the inner surface, and adsorption equilibrium, respectively [18]. In Table 2, the values of different constants and R2 for each linear model are presented.

2.5 2

20ppm

1.5

30ppm

1

40ppm

0.5

50ppm

0 0.003

0.00305 0.0031 0.00315 0.0032 0.00325 0.0033 0.00335 0.0034

Fig. 8. Thermodynamic diagram of methylene blue adsorption (Van't Hoff plot) by Zeolite/ Ferrite Nickel/Alginate adsorbent (0.025 g of adsorbent, initial Methylene blue concentration = 10–50 mg/l, pH = 6, solution volume = 50 ml, shaking speed = 200 rpm, and contact time = 120 min). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

where qmax and kl are constant adsorption of Langmuir and the maximum capacity of adsorption (mg/g), respectively, and dimensionless parameter RL expresses the adsorption process type in which is defined by eq. (13): RL ¼

1 1 þ K L C0

ð13Þ

where, C0 denotes the initial concentration of adsorbed molecules and RL specifies the nature of adsorption.0 b RL b 1 indicates desirable isotherm, RL = 0 and 1 represent linear and irreversible isotherms, and the RL N 1 shows undesirable isotherm [25]. Freundlich model expresses an adsorption on heterogeneous adsorbent surfaces and adsorption sites with different energies. This model is shown by the following linear form [33]: ln qe ¼ ln k f þ

1 lnce n

ð14Þ

Alginate chains

3.2.2. Methylene blue adsorption isotherm The plot of the capacity of adsorption against the equilibrium dye concentration at a constant temperature represents the isotherm of adsorption in which can be illustrated by several models. Important information about the mechanism of the adsorption, characteristics of adsorbent surface and bonds can be obtained from parameters of these models. Equilibrium data of experiments at different temperatures were analyzed by models of Freundlich, Langmuir, DubininRadushkevich (DR), and Temkin. Langmuir isotherm represents a monolayer and reversible interaction between adsorbates and the homogeneous adsorbent surfaces with the same sites [32]. Langmuir model expresses in forms of Eq. (11) and Linear Eq. (12): qe ¼

qmax kl ce ð1 þ kl ce Þ

ce 1 1 ¼ þ qe kl qmax qmax

ð11Þ

ð12Þ

Methylene blue molecule Fig. 9. The possible schematic of interaction between adsorbent surface and methylene blue molecules. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


615

(17): 70

1 E ¼ pffiffiffiffiffiffi 2β

60

ð17Þ

50

Temkin model assumes that with increasing the adsorbent surface covering, adsorption temperature declines linearly rather than logarithmically in which expresses by:

40 30 20

qe ¼

10 0 0

10

20

30

40

50

60

70

80

Fig. 10. Predicted values against experimental data.

Where kf and n are the constant of Freundlich model for the maximum capacity of adsorption and the heterogeneous factor, respectively. DR model is used to distinguish physical and chemical adsorptions and expresses as: ln qe ¼ ln qm −βɛ 2

ð15Þ

where qmax is the maximum adsorption capacity, β is a constant related to the average energy of adsorption and ε is defined as: 1 ɛ ¼ RT ln 1 þ ce

ð6Þ

Where R is universal gas constant (8.314 J/mol k) and T is temperature. Activation energy (E) in the DR model can be calculated from eq.

RT RT ln ðAT Þ þ ln ce bT bT

ð18Þ

where bT and A are constants related to adsorption heat and adsorption potential, respectively. In Fig. 7a, the adsorption capacity of methylene blue is plotted against the equilibrium concentration of dye at various temperatures. Temperature is another important parameter in the physical-chemical processes study because the adsorption capacity of adsorbent changes as temperature rises. If the adsorption percentage increases with temperature, it suggests that the adsorption process is endothermic. This may be explained by the fact that the mobility of dye molecules and the number of active sites increase as temperature increases, but according to Fig. 7b adsorption capacity reduction with increasing temperature indicates that the adsorption process is exothermic. This suggests that increasing temperature leads to the reduction of adsorption forces between the dye molecules and the active sites at the adsorbent surface decrease which in turn results in the reduction of adsorption percentage [34]. Figs. 7c–f show Langmuir, Freundlich, Temkin, and (DR) isotherms. The parameters of various methylene blue adsorption isotherm models with zeolite/nickel ferrite/alginate nanocomposite as adsorbent are reported in Table 3. As shown in the results, the adsorption of methylene blue on the synthesized nanocomposite complies more with Langmuir isotherm model which suggests a monolayer adsorption on the homogeneous surfaces with identical sites [35]. Moreover, the maximum capacity of Langmuir adsorption model was 36.50 mg g−1 and the RL varies in distance 0 to 1, which expresses the desirable isotherm.

Fig. 11. (a) The normal probability plots versus the remaining amounts of methylene blue, (b) the residual sum of squares versus experimental data of methylene blue adsorption, (c) the standardized residual plots versus the experiment running order. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

616


Fig. 12. Three-dimensional graph of methylene blue adsorption capacity as a function of (a) initial concentration of adsorbate and adsorbent amount, (b) initial concentration of adsorbate and pH (c) pH and adsorbent amount. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.2.3. Adsorption thermodynamic studies Thermodynamic parameters such as Gibbs free energy (ΔG), enthalpy (ΔH) and entropy (ΔS), are among the most important properties of adsorption processes in practical applications. ΔG can be obtained by the following equation: ΔG ¼ −RT ln K d

ð19Þ

where R is universal gas constant, T is temperature (K) and kd is the amount of adsorbed methylene blue (mg/g) to its residual concentration in solution (mg/l) [36]. Negative values of ΔG at different temperatures show that adsorption process is spontaneous. The values of ΔH and ΔS can be determined using Van't Hoff equation and by plotting the ln Kd against 1/T (Fig. 8): ln K d ¼

ΔG ΔH ΔS ¼− þ RT RT R

ð20Þ

kd ¼

qe ce

ð21Þ

According to Table 4, negative values of ΔG for methylene blue adsorption indicates the spontaneous nature of adsorption process and temperature increasing leads to decreasing the ΔG and increasing the spontaneity of adsorption process. The negative values of ΔH suggest that adsorption on zeolite/ferrite nickel/alginate is exothermic. The negative values of ΔS indicate a decrease in irregularities and the decreased likelihood of adsorbent accidental collisions with a solution. The negative entropy changes reflect the fact that the degree of freedom at solid-solution reduces during the adsorption [37]. 3.2.4. Comparing the capacity of methylene blue adsorption by various adsorbents The adsorption capacities of various adsorbents and the synthesized adsorbent for methylene blue adsorption are compared in Table 5 which

Fig. 13. Response optimization graph.

M. Bayat et al. / International Journal of Biological Macromolecules 116 (2018) 607–619 Table 2 The kinetic values of methylene blue adsorption on Zeolite/Ferrite Nickel/Alginate adsorbent. Pseudo-first-order kinetic model

Pseudo-second-order kinetic model

qe (exp.) (mg/g)

K1 qe (calc.) (1/min) (mg/g)

R2

18.78

0.0144

0.98 5945/853

1.9208

Elovich kinetic model

K2 (g/(mg min))

h (mg/(g min))

qe (calc.) (mg/g)

R2

212.472

18.9035

1

Table 4 Calculated values of thermodynamic equilibrium constant and free energy. C0 (mg/l)

T (K)

ΔH (KJ/mol)

ΔS (J/mol)

G (KJ/mol)Δ

20

298 313 328 298 313 328 298 313 328 298 313 328

−25.12

−44.4497

−37.58

−91.8531

−34.97

−86.8979

−22.97

−52.4106

−10.68 −10.93 −9.200 −10.00 −9.29 −7.20 −8.83 −8.30 −6.18 −7.02 −7.30 −5.38

30

Intraparticle diffusion kinetic model −1

(g/mg)β

(mg/g min

2.109

5.61e + 14

)α

R

2

0.89

Kt,1(mg/g min0/5)

Kt,2

Kt,3

C

5.4579

0.1888

0.0252

11.02

40

50

shows that synthesized nanocomposite has a good adsorption capacity. Table 6 represents a comparison between the capacity of methylene blue adsorption by the adsorbent and its components. As can be seen, adding nickel ferrite nanoparticles to zeolite and its subsequent modification and coating with sodium alginate increases methylene blue adsorption capacity which is significant compared to its components. Interaction forces between the N+ groups of methylene blue molecules and carboxylate (COO−) groups of alginate as active surface sites, can be a possible reason for dye adsorption (Fig. 9.). On the other hand, the OH group on the Clinoptilolite surface could bind with the N+ groups of methylene blue molecules through hydrogen bonding or via electrostatic interaction.

3.3. Box-Behnken design of experiments The experimental data and predicted responses by Minitab17 software for methylene blue adsorption process on Zeolite/Ferrite Nickel/ Alginate nanocomposite are shown in Table 7. 15 tests were designed randomly and central point tests were repeated three times in order to error estimation of experiments. Once the experiments were carried out, the response surface regression was used to determine the optimal model explaining each response behavior in the adsorption process and to determine the independent variables and considerable interactions. To this end, Analysis of variance (ANOVA(was used as a mean to analyze the responses and to verify the model's validity. In this method, the importance of the effects of a certain parameter and inter-parametric interactions on the process is examined through some tests. ANOVA determines the importance of the model. The statistical significance of each parameter is specified by the F and P-values reported in the table. The higher F-values and P-Values b0.05 indicate the considerable significance of model terms and imply that the studied parameter influences the desired response. As shown in Table 8, P-value is 0.002 which indicates the significance of the model. The linear terms of initial dye concentration and adsorbent amount, the quadratic terms of the adsorbent amount and the interactional terms of the initial concentration of the solution and adsorbent Table 3 Equilibrium isotherms of methylene blue adsorption using Zeolite/Ferrite Nickel/Alginate adsorbent. Langmuir isotherm model

Freundlich isotherm model

Temp.

qmax

Kl

R2

RL

n

Kf

R2

25 40 55

54.054 49.75 44.843

1.293 1.608 0.721

0.99 0.97 0.97

0.01–0.07 0.01–0.05 0.02–0.1

2.05 2.94 2.70

26.62 26.76 17.95

0.91 0.99 0.99

Temkin isotherm model

DR isotherm model

Temp.

BT

AT

R2

β

qm

R2

E

25 40 55

11.803 7.658 7.888

12.722 49.685 13.515

0.99 0.92 0.96

7.00E-08 2.00E-08 5.00E-08

41.808 33.642 30.984

0.98 0.81 0.83

2.67 5 3.16

617

amount were the terms of the model that had a significant effect. The polynomial equation derived from the model is as follows:

qe ¼ 4:0 þ 4:81pH−1072x þ 1:188C−0:345 pH pH þ 20697x x þ 0:00130 C C−41:7 pH x þ 0:0257 pH C−23:49 x C ð22Þ

The compliance and accuracy of the predicted model were assessed by R2 or determination coefficient. The closer R2 is to 1, the better is prediction and response and the more robust is the predicted model. The value of R2 was calculated to be 0.9768 indicating that the model is suitable and can explain 97.68% of the variability in the response. The high value of adjusted determination coefficient or R2adj was 0.9268 which highlights the significance and accuracy of the proposed model in fitting the experimental data. The predicted values of methylene blue adsorption capacity are plotted against the experimental data in Fig. 10 which shows that the experimental data are dispersed close to a straight line indicating the goodness of fit between experimental and predicted data. The experimental data of adsorption capacity of methylene blue reported in Table 8 were analyzed using Minitab17.

3.3.1. Analysis of statistical charts In order to authenticate the obtained model, statistical charts shown in Fig. 11 can be used. Fig. 11a indicate plots of normal probability for BBD. This plot shows the normality level of residual response of methylene blue adsorption. The response distribution around the straight regression line should be linear, otherwise, the model normality hypothesis is void. As shown in this Figure, the points are linearly distributed that is indicative of good fitting of the model to experimental data. Residual sum of squares can also be used to verify the validity of the proposed model. In this plot, if the pattern of the residual sum of squares was randomly distributed on both sides of zero line, the predicted model is acceptable. This plot for adsorption of methylene blue by zeolite/nickel ferrite/sodium alginate nanocomposite is shown in Fig. 11b which depicts the validation of the predicted model. Fig. 11c illustrated plots of standardized residual in terms of running order of the experiments in which the points are distributed randomly around the straight central line.

Table 5 Maximum adsorption capacity of methylene blue using various adsorbents. Adsorbent

Adsorption (mg/g)

Ref.

Zeolite (ZD & ZS) Treated zeolite Alginate/activated carbon beads Nikel-Zinc ferrite nanoparticles Zeolite/ferrite nickel/alginate nanocomposite

13.49–11.13 28.6–42.7 38.9 36 54.05

[38] [39] [40] [41] This work

618


Table 6 Comparison of the adsorption capacity of methylene blue by the adsorbent and its components in the same condition (0.01 g of adsorbent, initial Methylene blue concentration = 20 mg/l, pH = 6, solution volume = 50 ml, shaking speed = 200 rpm, and contact time = 120 min). Adsorbent

Adsorption (mg/g)

Treated zeolite Zeolite/ferrite nickel nanocomposite Zeolite/ferrite nickel/alginate nanocomposite

12.154 17.572 19.147

3.3.2. Study of parameters interaction The effect of changes in the initial concentrations of the pollutant and adsorbent amount on the methylene blue adsorption capacity is shown in Fig. 12a. As can be seen, the adsorption capacity of methylene blue decreases as the adsorbent amount is reduced to 0.01 g while an increase in the pollutant initial concentration to 50 mg/L leads to a rise in the adsorption capacity to over 60 mg/g. Since adsorbents have a negative charge on their surface in basic solutions, the cations are easily adsorbed due to the opposite electrical charge while the adsorption of anions is slower [38]. In this study, the effect of pH on the capacity of methylene blue adsorption by zeolite/nickel ferrite/sodium alginate was investigated in the pH range between 2 and 8. At low pHs, a high concentration of hydrogen ions stimulates the protonation of the functional groups. As a result, adsorbents become more positively charged, which prevents the adsorption of cationic dyes on the adsorbent. In other words, additional hydrogen ions compete against cationic molecules in an attempt to be adsorbed on active sites, so the adsorption capacity is greatly reduced at lower pHs. When pH increases, the number of available positively charged sites decreases implying that the number of negatively charged sites increases. As the surface of adsorbents become more negatively charged, the interactions between adsorbents and cationic dye molecules also increase, so the adsorption capacity increases at higher pHs. Increasing the adsorbent amount results in an increase in the number of available adsorption sites and subsequently an increase in adsorption amount (removal percentage). However, it reduces the adsorbate amount per unit mass of adsorbent. Increasing the adsorbent amount can increase adsorbent's specific surface area and adsorption sites. But as can be seen in Fig. 12a, the adsorption capacity of methylene blue decreases as a result of adsorbent amount increment which might be due to adsorption sites remaining unsaturated during the adsorption process and the number of available adsorption sites increasing with the adsorbent amount. This reduced adsorption capacity might also stem from the adsorbent accumulation and some active sites overlapping at high adsorbent amounts [39]. In Fig. 12b, the three-dimensional plot of the effects of pH and initial dye concentration on the methylene blue adsorption rate is shown. As can

Table 7 Experimental design results. Run no.

pH Adsorbent dosage (g)

Dye concentration (mg/l)

Experimental results (mg/g)

Predicted results (mg/g)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

5 8 5 2 8 5 8 5 2 5 5 2 5 2 8

10 10 10 10 30 50 50 30 30 30 30 30 50 50 30

3.97 6.57 17.72 5.75 11.69 70.80 30.56 19.23 32.03 19.21 19.10 10.01 19.47 23.58 43.73

8.69 3.92 19.68 1.71 9.62 66.07 34.60 19.20 34.10 19.20 19.20 9.30 17.50 26.23 44.41

0.05 0.03 0.01 0.03 0.05 0.01 0.03 0.03 0.01 0.03 0.03 0.05 0.05 0.03 0.01

Table 8 Analysis of variance of surface response of experimental design results. Source

Degree Sum of of squares freedom

Mean square

F-value

P-value

Model Linear pH Dye concentration Adsorbent dosage Quadratic pH × pH Dye concentration × dye concentration Adsorbent dosage × adsorbent dosage Interaction Dye concentration × pH Adsorbent dosage × pH Adsorbent dosage × dye concentration Error Residual Pure error Cor total

9 3 1 1 1 3 1 1

4046.19 3353.98 56.09 1523.76 1774.13 304.61 35.54 1.00

449.58 1117.99 56.09 1523.76 1774.13 101.54 35.54 1.00

20.70 51.48 2.58 70.16 81.69 4.68 1.64 0.05

0.002 0.000 0.169 0.000 0.000 0.065 0.0257 0.839

1

252.61

252.61

11.63

0.019

3 1 1 1

387.60 9.50 25.09 353.01

129.20 9.50 25.09 353.01

5.95 0.44 1.16 16.25

0.042 0.538 0.332 0.010

5 3 2 14

108.60 21.72 108.59 36.20 0.00 0.00 4154.79

19,108.59 0.000

be observed, with simultaneously increasing the pollutant initial concentration and pH, adsorption capacity also increases such that with increasing pH up to 8 and pollutant initial concentration up to 50 mg/l, adsorption capacity amounts to about 30 mg/g. Adsorption capacity increases by increasing the initial concentration of the pollutant. This phenomenon can be explained by the fact that the driving force of mass transfer is directly related to concentration increment which causes adsorption sites to be well surrounded and pollutant to penetrate into deeper zones of the adsorbent and ultimately leads to more adsorption and increased equilibrium adsorption capacity. The initial concentration of dye provides a significant driving force to overcome the total resistance originated from the mass transfer between the liquid and solid phases. As can be seen, an increase in the initial concentration of dye enhances the amount of adsorbed dye as well, thus, it can be said that the removal of dye is affected by its initial concentration. In Fig. 12c, the adsorption capacity of methylene blue is depicted as a function of pH and the adsorbent amount which shows that adsorption decreases by increasing the adsorbent amount from 0.01 to 0.05 g. It is also observed that the removal of methylene blue at pH values below 8 is not ideal. Therefore, according to the aforementioned reasons, the highest adsorption capacity can be obtained by reducing the adsorbent amount and increasing the pH. To increase the adsorption capacity, reducing the adsorbent amount is more effective than increasing pH. 3.3.3. Optimization of adsorption process To achieve the maximum capacity of dye adsorption the optimal condition should be investigated. The goal of numerical optimization is to find the best conditions for the adsorption process so that the influencing parameters values are determined in such a way that response optimized. The maximum removal efficiency, relationships between the variables and their optimal conditions can be determined using numerical optimization. Minitab17 software uses the response optimizer tool to easily determine the optimal conditions for each parameter. As shown in Fig. 13, pH = 8, an adsorbent amount of 0.01 and initial concentration of 50 mg L−1 are the optimal conditions to achieve the maximum adsorption capacity. In this optimal condition, an adsorption of 69.61 mg g−1 is predicted. 4. Conclusions In this research, nickel ferrite magnetic nanoparticles were precipitated on the Clinoptilolite for its magnetization and then coated with


sodium alginate. The obtained Clinoptilolite/Nickel Ferrite/Alginate nanocomposite was characterized by SEM, XRD, FTIR, and VSM analyses. Since the nanocomposite was shown 16 emu g−1 saturation magnetization, it was easily and quickly separated from solution using a simple magnet. The kinetics of the adsorption process was evaluated by pseudo-first-order, pseudo-second-order, Elovich and intraparticle diffusion models, and the pseudo-second-order model exhibited the highest consistency with the kinetic data. Investigating the adsorption isotherm using Langmuir, Freundlich, Temkin, and DR models revealed that adsorption process corresponds better with Langmuir model, and the maximum adsorption capacity is 54.05 mg/g at 25 °C. To achieve the optimal conditions for methylene blue adsorption, the RSM and BBD were used for experiments. The maximum capacity of methylene blue adsorption obtained under optimal conditions of the adsorbent amount of 0.01 g, pH = 8, and initial methylene blue concentration of 50 mg L−1 was 69.61 mg/g. The synthesized superparamagnetic nanocomposite can be used as a suitable adsorbent for adsorption of methylene blue from aqueous solutions. Acknowledgment Financial support of this work by ACECR Institute of Higher Education (Isfahan Branch) is gratefully appreciated. Conflict of interest The authors declare that they have no conflict of interest. References [1] J. Paul, K. Rawat, K. Sarma, S. Sabharwal, Decoloration and degradation of reactive Red-120 dye by electron beam irradiation in aqueous solution, Appl. Radiat. Isot. 69 (7) (2011) 982–987. [2] S. Rajgopalan, Water pollution problem in the textile industry and control, in: R.K. Trivedy (Ed.), Pollution Management in Industries, Environmental Publications, Karad, India 1995, pp. 21–44. [3] B. Royer, N.F. Cardoso, E.C. Lima, J.C. Vaghetti, N.M. Simon, T. Calvete, R.C. Veses, Applications of Brazilian pine-fruit shell in natural and carbonized forms as adsorbents to removal of methylene blue from aqueous solutions—kinetic and equilibrium study, J. Hazard. Mater. 164 (2) (2009) 1213–1222. [4] D.S. Brookstein, Factors associated with textile pattern dermatitis caused by contact allergy to dyes, finishes, foams, and preservatives, Dermatol. Clin. 27 (3) (2009) 309–322. [5] P.A. Carneiro, G.A. Umbuzeiro, D.P. Oliveira, M.V.B. Zanoni, Assessment of water contamination caused by a mutagenic textile effluent/dyehouse effluent bearing disperse dyes, J. Hazard. Mater. 174 (1) (2010) 694–699. [6] M. Rafatullah, O. Sulaiman, R. Hashim, A. Ahmad, Adsorption of methylene blue on low-cost adsorbents: a review, J. Hazard. Mater. 177 (1) (2010) 70–80. [7] N. Mallick, Biotechnological potential of immobilized algae for wastewater N, P and metal removal: a review, Biometals 15 (4) (2002) 377–390. [8] M. Anbia, S. Salehi, Removal of acid dyes from aqueous media by adsorption onto amino-functionalized nanoporous silica SBA-3, Dyes Pigments 94 (1) (2012) 1–9. [9] G. Crini, Non-conventional low-cost adsorbents for dye removal: a review, Bioresour. Technol. 97 (9) (2006) 1061–1085. [10] L. Ai, H. Huang, Z. Chen, X. Wei, J. Jiang, Activated carbon/CoFe2O4 composites: facile synthesis, magnetic performance and their potential application for the removal of malachite green from water, Chem. Eng. J. 156 (2) (2010) 243–249. [11] R. Suresh, P. Moganavally, M. Deepa, Synthesis and characterization of nickel ferrites nanoparticles, National Conference on Nanomaterials for Environmental [NCNER2015] 2015, pp. 113–116. [12] P. Finotelli, M. Morales, M. Rocha-Leao, E. Baggio-Saitovitch, A. Rossi, Magnetic studies of iron (III) nanoparticles in alginate polymer for drug delivery applications, Mater. Sci. Eng. C 24 (5) (2004) 625–629. [13] A. Nasrullah, A. Bhat, A. Naeem, M.H. Isa, M. Danish, High surface area mesoporous activated carbon-alginate beads for efficient removal of methylene blue, Int. J. Biol. Macromol. 107 ( (2018) 1792–1799.

619

[14] M. Boumediene, H. Benaïssa, B. George, S. Molina, A. Merlin, Characterization of two cellulosic waste materials (orange and almond peels) and their use for the removal of methylene blue from aqueous solutions, Maderas, Ciencia y tecnología 17 (1) (2015) 69–84. [15] Y.-S. Ho, R. Malarvizhi, N. Sulochana, Equilibrium isotherm studies of methylene blue adsorption onto activated carbon prepared from Delonix regia pods, J. Environ. Prot. Sci. 3 (2009) 111–116. [16] V. Ponnusami, S. Vikram, S. Srivastava, Guava (Psidium guajava) leaf powder: novel adsorbent for removal of methylene blue from aqueous solutions, J. Hazard. Mater. 152 (1) (2008) 276–286. [17] H. Benaïssa, Z.S. Bereksi, INFLUENCE OF SOME EXPERIMENTAL PARAMETERS ON METHYLENE BLUE SORPTION KINETICS FROM SYNTHETIQUE AQUEOUS SOLUTIONS USING THISTLE STALKS AS A LOW-COST SORBENT: EXPERIMENTAL AND MODELLING STUDIES. [18] W. Wu, Q. He, C. Jiang, Magnetic iron oxide nanoparticles: synthesis and surface functionalization strategies, Nanoscale Res. Lett. 3 (11) (2008) 397. [19] J.D. Seader, E.J. Henley, Separation Process Principles, 2011. [20] F. Ghorbani, H. Younesi, S.M. Ghasempouri, A.A. Zinatizadeh, M. Amini, A. Daneshi, Application of response surface methodology for optimization of cadmium biosorption in an aqueous solution by Saccharomyces cerevisiae, Chem. Eng. J. 145 (2) (2008) 267–275. [21] M.Y. Can, Y. Kaya, O.F. Algur, Response surface optimization of the removal of nickel from aqueous solution by cone biomass of Pinus sylvestris, Bioresour. Technol. 97 (14) (2006) 1761–1765. [22] S.C. Ferreira, R. Bruns, H. Ferreira, G. Matos, J. David, G. Brandao, E.P. da Silva, L. Portugal, P. Dos Reis, A. Souza, Box-Behnken design: an alternative for the optimization of analytical methods, Anal. Chim. Acta 597 (2) (2007) 179–186. [23] M.A. Bezerra, R.E. Santelli, E.P. Oliveira, L.S. Villar, L.A. Escaleira, Response surface methodology (RSM) as a tool for optimization in analytical chemistry, Talanta 76 (5) (2008) 965–977. [24] K. Maaz, S. Karim, A. Mumtaz, S. Hasanain, J. Liu, J. Duan, Synthesis and magnetic characterization of nickel ferrite nanoparticles prepared by co-precipitation route, J. Magn. Magn. Mater. 321 (12) (2009) 1838–1842. [25] V. Javanbakht, S.M. Ghoreishi, N. Habibi, M. Javanbakht, A novel magnetic chitosan/ clinoptilolite/magnetite nanocomposite for highly efficient removal of Pb (II) ions from aqueous solution, Powder Technol. 302 (2016) 372–383. [26] H. Jahangirian, M. Shah Ismail, M.J. Haron, R. Rafiee-Moghaddam, K. Shameli, S. Hosseini, K. Kalantari, R. Khandanlou, E. Gharibshahi, S. Soltaninejad, Synthesis and characterization of zeolite/Fe3O4 nanocomposites by green quick precipitation method, Dig. J. Nanomaterials Bios 4 (8) (2013) 4. [27] H. Moradmard, S.F. Shayesteh, P. Tohidi, Z. Abbas, M. Khaleghi, Structural, magnetic and dielectric properties of magnesium doped nickel ferrite nanoparticles, J. Alloys Compd. 650 (2015) 116–122. [28] H. Daemi, M. Barikani, M. Barmar, Variations in calcium and alginate ions concentration in relation to the properties of calcium alginate nanoparticles, Sci. Technol. 26 (1) (2013) 25–32. [29] A. Baraka, P. Hall, M. Heslop, Melamine–formaldehyde–NTA chelating gel resin: synthesis, characterization and application for copper (II) ion removal from synthetic wastewater, J. Hazard. Mater. 140 (1) (2007) 86–94. [30] M.V. Dinu, E.S. Dragan, Evaluation of Cu2+, Co2+ and Ni2+ ions removal from aqueous solution using a novel chitosan/clinoptilolite composite: kinetics and isotherms, Chem. Eng. J. 160 (1) (2010) 157–163. [31] Y. Tan, M. Chen, Y. Hao, High efficient removal of Pb (II) by amino-functionalized Fe3O4 magnetic nano-particles, Chem. Eng. J. 191 (2012) 104–111. [32] Y.C. Chang, D.H. Chen, Adsorption kinetics and thermodynamics of acid dyes on a carboxymethylated chitosan-conjugated magnetic nano-adsorbent, Macromol. Biosci. 5 (3) (2005) 254–261. [33] X. Yang, B. Al-Duri, Kinetic modeling of liquid-phase adsorption of reactive dyes on activated carbon, J. Colloid Interface Sci. 287 (1) (2005) 25–34. [34] N. Tripathi, Cationic and anionic dye adsorption by agricultural solid wastes: a comprehensive review, J. Appl. Chem. 5 (2013) 91–108. [35] M. Rafatullah, O. Sulaiman, R. Hashim, A. Ahmad, Adsorption of copper (II), chromium (III), nickel (II) and lead (II) ions from aqueous solutions by meranti sawdust, J. Hazard. Mater. 170 (2) (2009) 969–977. [36] Y.-M. Hao, C. Man, Z.-B. Hu, Effective removal of Cu (II) ions from aqueous solution by amino-functionalized magnetic nanoparticles, J. Hazard. Mater. 184 (1) (2010) 392–399. [37] X. HUANG, N.-Y. GAO, Q.-l. ZHANG, Thermodynamics and kinetics of cadmium adsorption onto oxidized granular activated carbon, J. Environ. Sci. 19 (11) (2007) 1287–1292. [38] M. Anbia, S.A. Hariri, Removal of methylene blue from aqueous solution using nanoporous SBA-3, Desalination 261 (1) (2010) 61–66. [39] S. Cengiz, L. Cavas, Removal of methylene blue by invasive marine seaweed: Caulerpa racemosa var. cylindracea, Bioresour. Technol. 99 (7) (2008) 2357–2363.