Removal of sudan dyes from aqueous solution by magnetic carbon nanotubes: Equilibrium, kinetic and thermodynamic studies

Removal of sudan dyes from aqueous solution by magnetic carbon nanotubes: Equilibrium, kinetic and thermodynamic studies

G Model JIEC-2144; No. of Pages 5 Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx Contents lists available at ScienceDirect Jour...

651KB Sizes 0 Downloads 25 Views

G Model

JIEC-2144; No. of Pages 5 Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec

Removal of sudan dyes from aqueous solution by magnetic carbon nanotubes: Equilibrium, kinetic and thermodynamic studies Xiaoshan Sun, Haijian Ou, Changfeng Miao, Ligang Chen * Department of Chemistry, College of Science, Northeast Forestry University, 26 Hexing Road, Harbin 150040, China

A R T I C L E I N F O

Article history: Received 10 April 2014 Received in revised form 15 July 2014 Accepted 20 July 2014 Available online xxx Keywords: Carbon nanotubes Magnetic separation Sudan dyes Adsorption

A B S T R A C T

In this study, magnetic carbon nanotubes (MCNTs) have been synthesized by hydrothermal synthesis of Fe3O4 nanoparticles onto carbon nanotubes, and characterized by Fourier transform infrared spectroscopy, transmission electron microscope and physical property measurement system. They were used as adsorbents for the removal of sudan I, sudan II, sudan III, and sudan IV from aqueous solution. The effects of experimental parameters on adsorption were fully studied. The results showed that 5 mg MCNTs can remove sudan dyes sufficiently from 20 mL aqueous solution within 60 min at pH in the range of 4–7. Moreover, the existence of NaCl favored the adsorption. The sorbents can be reused more than ten times with 2% acetic acid ethanol as desorption solution. The equilibrium, kinetic and thermodynamic studies were investigated. The resultant kinetic data were well described by the pseudosecond-order model. The Freundlich model fitted sudan dyes sorption isotherms better than the Langmuir model. The thermodynamic data calculated from the temperature dependent sorption isotherms demonstrated that sudan dyes sorption on the MCNTs was a spontaneous process. ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.

Introduction Sudan dyes are kind of coloring additives in fuels, waxes, plastics, floor and shoe polishes [1]. The dyes are harmful to human health. The International Agency for Research on Cancer (IARC) has classified the sudan dyes as category 3 carcinogens [2]. Because of their low cost and the enhancement of products appearance, sudan dyes are still used in food stuffs unfortunately [3,4]. And their discharge into water could cause environmental pollutions [5]. Although there are many researchers presented some methods for the determination of sudan dyes in food stuffs. Liu et al. extracted and separated sudan dyes in chilli powder by cloud point extraction [6]. Chailapakul et al. determined sudan dyes in soft drink with electrochemical detection [7]. Zhao et al. extracted sudan dyes in tomato sauce and sausage with molecularly imprinted polymers [8]. There were few researches about the removal of sudan dyes from environment water. Jiang et al. removed sudan dyes from water with C18-functional

* Corresponding author. Tel.: +86 451 82190679/+86 451 82198244; fax: +86 451 82191571. E-mail address: [email protected] (L. Chen).

ultrafine magnetic silica nanoparticles [9], but their studies did not involve the adsorption mechanism, kinetic and thermodynamic characteristics. Recent studies [10–18] indicated that carbon nanotubes (CNTs) had high sorption capacity for inorganic and organic pollutants since they were found in 1991 [19], due to their unique physical, chemical and mechanical properties, such as hydrogen storage [20], composite materials [21] or nano-electronics [22]. However, the pristine CNTs are difficult to separate from the aqueous solution due to their small size [23]. The magnetic separation technique based on small magnetic particles was introduced in 1973 and from then on more and more attentions have been paid to its development and application [9]. Magnetic separation has advantages such as its easy phase separation with aqueous solutions and its capability of treating large amount of wastewater within a short time [24,25]. In this study, magnetic carbon nanotubes (MCNTs) were synthesized by hydrothermal synthesis of Fe3O4 nanoparticles onto CNTs. The MCNTs were characterized by Fourier transformed infrared spectroscopy (FT-IR), transmission electron microscope (TEM), and physical property measurement system (PPMS). Adsorption equilibrium, kinetic and thermodynamic characteristics of sudan dyes adsorption on MCNTs were studied. The effects

http://dx.doi.org/10.1016/j.jiec.2014.07.034 1226-086X/ß 2014 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: X. Sun, et al., J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.07.034

G Model

JIEC-2144; No. of Pages 5 X. Sun et al. / Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx

2

of experimental parameters, such as the amount of magnetic sorbent, shaking time, ionic strength, pH, and desorption solvent have been fully investigated. Experimental Chemical and equipment The standards of sudan I, sudan II, sudan III and sudan IV were obtained from Aladdin (Shanghai, China). The CNTs used in this study were purchased from Nanoport (Shenzhen, China). Chromatographic grade acetonitrile was obtained from Fisher (Pittsburgh, PA, USA). Methanol, ethanol, sodium acetate anhydrous, iron(III) chloride hexahydrate (FeCl36H2O), sodium hydroxide, sodium chloride, glycol, nitric acid and acetic acid, hydrochloric acid were of analytical grade and obtained from Kermel (Tianjin, China). Chromatographic analysis was performed on a LC-15C high performance liquid chromatograph with a UV detector (Shimadzu, Kyoto). A Zorbax SB-C18 column (250 mm  4.6 mm I.D., 5 mm) was used as an analytical column (Agilent, Palo Alto, CA, USA). The adsorption experiments were carried out at constant temperatures controlled by a SHA-B shaking table (Shengtang, Jintan, China). In the process of preparing MCNTs, KQ5200E ultrasonic apparatus (Kunshan Instrument, Kunshan, China) and DF-101S magnetic stirrer (Yarong Instrument, Zhengzhou, China) were used. The MCNTs were characterized by a FT-IR 360 Fourier transform infrared spectroscopy (Nicolet, Madison, WI, USA), H-7650 transmission electron microscope (Hitachi, Japan), physical property measurement system (Quantum Design Instrument, San Diego, CA, USA) and themogravimetric analyzer (Pyris 1, Perkin Elmer, Waltham, MA, USA).

Results and discussion Characterization of MCNTs composites Fig. 1a displays the TEM image of MCNTs. It was observed that iron oxide nanoparticles were successfully attached onto the surface of CNTs. FT-IR spectra of CNTs and MCNTs were compared in Fig. 1b. Most of absorption signals were matched between the two materials: 1384 or 1379 cm1 (vibration peak of –CH3), 1649 or 1640 cm1 (vibration peak of C5 5C), 2923 or 2920 cm1 (vibration peak of –CH2–) and 3410 or 3420 cm1 (vibration peak of O–H). The signal at 579 cm1 (vibration peak of Fe–O) [26] was only found in the MCNTs, which indicates Fe3O4 was combined

Preparation of magnetic carbon nanotubes Firstly, CNTs were purified with 3 mol/L HNO3 at 95 8C for 4 h under a magnetic stirring. Then the purified CNTs were washed to neutral pH with pure water, dried at 60 8C for 15 min. Subsequently, 0.2 g of activated CNTs were suspended in 50 mL of mixed solution containing sodium acetate anhydrous (1.8 mg), glycol (40 mL) and FeCl36H2O (0.7 g). Then the mixture was heated at 200 8C for 12 h. Ultimately the MCNTs were magnetically separated and washed repeatedly with ethanol three times. The prepared material was then dried at 60 8C for further use. General adsorption procedure Batch adsorption experiments were conducted using 50 mL centrifuge tubes containing 5 mg MCNTs and 20 mL sudan dyes solution. After ionic strength and pH were adjusted to designated values, the tubes were shaken at 298 K for 60 min, the solid/liquid phases were separated under strong external magnetic field. Then the concentration of sudan dyes in suspensions was determined by HPLC. The column temperature was set at room temperature, and the injection volume was 20 mL. The mobile phase was acetonitrile and the flow rate was 1.0 mL/min. The detection wavelength was 505 nm. The retention time of sudan dyes were 3.383, 4.901, 5.801, 9.245 min, respectively. Kinetic experiments were performed using a series of 50 mL centrifuge tubes containing MCNTs and sudan dyes solution in a time range of 1 to 100 min. Adsorption isotherm studies were carried out with initial concentrations of sudan dyes varying between 1 and 25 mg/L (20 mL sample volume), and the experimental temperatures were controlled at 298, 308 and 318 K, respectively. The thermodynamic parameters for the adsorption process were determined at each temperature.

Fig. 1. (a) The TEM images of MCNTs; (b) Fourier transformed infrared spectra of CNTs and MCNTs; (c) magnetization curve of MCNTs.

Please cite this article in press as: X. Sun, et al., J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.07.034

G Model

JIEC-2144; No. of Pages 5 X. Sun et al. / Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx

100 80 60 Ads. %

with CNTs. The percentage of Fe3O4 in MCNTs is 28.3% which was obtained through thermogravimetric analysis. The PPMS was applied to characterize the magnetic property of the prepared materials (Fig. 1c). As shown in Fig. 1c, the magnetization curves showed its superparamagnetic property. The saturation magnetization of MCNTs is 22.4 emu/g. As shown in Fig. 1c (inset), in the presence of an external magnetic field, the MCNTs were attracted to the wall of vial in comparison with a black homogeneous dispersion existed without an external magnetic field.

3

40 sudan I sudan II sudan III sudan IV

20 0

Adsorption study

0

20

40

60

80

100

Time (min)

Effect of MCNTs dosage The effect of adsorbent dosage was studied by varying MCNTs amount from 1 to 20 mg. The corresponding result was shown in Fig. 2. The experimental results revealed that the removal efficiencies of sudan dyes increased gradually with increasing amounts of MCNTs from 1 to 5 mg. With increasing adsorbent dosage, more surface area was available for adsorption due to the increase in active sites on the surface of MCNTs and thus making easier penetration of adsorbate to adsorption sites [27]. Effect of the contact time The time needed for the interaction between the adsorbate and adsorbent is crucial. Hence, it is important to study the effect of contact time on the removal of the sudan dyes. Fig. 3 showed the effect of contact time on the adsorption of four sudan dyes onto MCNTs from aqueous solutions. In general, the adsorption of sudan dyes increased significantly within the first 10 min. The rapid increase of removal efficiency in the initial stages indicated that there are plenty of readily accessible sites [28]. When the contact time was up to 60 min, the adsorption of sudan dyes remains constant. Effect of pH of solution sample Solution pH is related to the adsorption mechanism and reflects the various adsorbent–adsorbate physico-chemical interactions [29,30]. In this study, the effect of pH was studied by varying the pH in the range of 3–11. The effect of solution pH is shown in Fig. 4. The efficient adsorption of sudan dyes was obtained in the pH range of 4–7. The adsorption decreased with increasing pH from 7 to 11. The adsorption is lower at pH 3 than at pH 7. Usually, adsorbability depends on the characteristics of adsorbent and adsorbate. At pH 3, MCNTs carried smaller positive charge [31], the repulsive energy between MCNTs was smallest resulting in a highly compacted aggregation structure of MCNTs, which was unfavorable for sudan dyes adsorption. When pH gradually increased to 4, MCNTs surface was negatively charged

Fig. 3. Effect of the contact time on adsorption of sudan dyes.

and the repulsive energy between MCNTs increased [32,33], so the adsorption increased accordingly. Furthermore, sudan dyes could have different charges on different sites depending on solution pH. While sudan dyes showed high adsorption under conditions with 4 < pH < 7, most of sudan molecules carried no net electrical charge, which makes them hardly have electrostatic attraction or repulsion with MCNTs, the increase of pH from 4 to 7 had no significant effect on the adsorptive affinity of sudan in the experiment. When the pH was greater than 7, the number of positively charged sites on MCNTs decreased and the number of negatively charged sites increased, part of sudan molecule exists as anion, and the adsorption was significantly impeded. Effect of ionic strength The influence of ionic strength on the adsorption of sudan dyes is critical because it can create different adsorption situations between the MCNTs surfaces and the sudan dyes are either attractive or repulsive [30]. So in order to expose the influence of ionic strength, the adsorption experiment was conducted by changing the ionic strength using NaCl of 0, 1%, 5% and 10%. In Fig. 5, an increase of adsorption capacity was observed when NaCl existed in solvent. The reason could be attributed to variations of the solubility of sudan dyes in water. With adding salt into solutions, the affinity of sudan dyes to water phase would decrease and lead to higher adsorption efficiency. Adsorption isotherms The equilibrium isotherms are very important for understanding the adsorption systems. Frequently, the Langmuir adsorption isotherm and the Freundlich isotherm equations were available for analyzing experimental sorption equilibrium data. The Langmuir

100

100

90

90

sudan I sudan II sudan III sudan IV

70 60 50

Ads. %

Ads. %

80 80

70 sudan I sudan II sudan III sudan IV

60 50 40

0

4

8

12

16

20

Amount of MCNTs (mg) Fig. 2. Effect of the amount of MCNTs on adsorption of sudan dyes.

0

1

2

3

4

5

6

7

8

9

10 11 12

pH Fig. 4. Effect of pH values of sample solution on adsorption of sudan dyes.

Please cite this article in press as: X. Sun, et al., J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.07.034

G Model

JIEC-2144; No. of Pages 5 X. Sun et al. / Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx

4

Thermodynamic study 100

The sorption behaviors of different concentrations of sudan dyes onto MCNTs were critically investigated at 298, 308 and 318 K, respectively. Thermodynamic parameters were calculated from following equations [23]:

Ads. %

90

80

DG ¼ RTln K c

sudan I sudan II sudan III sudan IV

70

where R is the universal gas constant (8.314 J/(mol K)), T is temperature (K) and Kc is the distribution coefficient. Gibbs free energy change of adsorption (DG8) was calculated using ln Kc values for different temperatures. The Kc values were calculated using following equation:

60 0

2

4

6

8

10

Percentage of sodium chloride (%) Fig. 5. Effect of ionic strength on adsorption of sudan dyes.

Kc ¼

adsorption isotherm assumes monolayer adsorption on to a surface containing a finite number of adsorption sites, with uniform strategies of adsorption and no transmigration of adsorbate in the plane of the surface. In contrast to Langmuir, the Freundlich isotherm is based on the assumption that the adsorption occurs on heterogeneous sites with non-uniform distribution of energy levels [34]. The Langmuir equation [35] can be represented as follows: Ce 1 Ce ¼ þ qe bqm qm

(1)

where Ce (mg/L) is equilibrium concentration of sudan dyes in solution, qe (mg/g) is the adsorption capacity of sudan dyes adsorbed at equilibrium, qm (mg/g) is maximum amounts of sudan dyes adsorbed per unit mass of adsorbent, b (L/mg) is a constant related to the heat of adsorption. The slope and intercept of linear plots of Ce/qe against Ce yield the values of 1/qm and 1/bqm for Eq. (1). The Fruendlich equation [36] can be represented as follows: ln qe ¼ ln K F þ

  1 ln C e n

(2)

where KF and 1/n are empirical constants. These constants can also be evaluated from the intercept and slope of the linear plot of ln qe versus ln Ce. The values of the isotherm parameters are presented in Table 1. It was found that the adsorption of sudan dyes on MCNTs correlated well (R2 > 0.9) with the equation as compared to the Langmuir equation (R2 < 0.82) under the studied concentration range. Therefore, the Freundlich model fitted sudan dyes sorption isotherm better than the Langmuir model. Table 1 Isotherm parameters of Langmuir and Freundlich model for the adsorption of sudan dyes onto MCNTs. Sample

Sudan I

Sudan II

Sudan III

Sudan IV

T (K) Langmuir model

298 308 318 298 308 318 298 308 318 298 308 318

(3)

Freundlich model

qe Ce

(4)

where Ce is the equilibrium concentration of sudan dyes, qe is the amount of sudan dyes adsorbed per unit weight of MCNTs at equilibrium concentration (mg/g). The enthalpy change (DH8) and entropy change (DS8) of adsorption were estimated from the following equation: ln K c ¼

DS o R



DH o

(5)

RT

According to Eq. (5), DH8 and DS8 parameters can be calculated from the slope and intercept of the plot of ln Kc versus 1/T, respectively. The thermodynamic parameters were summarized in Table 2. The negative DH8 suggests that the sorption process of sudan dyes is exothermic. Adsorption kinetics The kinetics of sudan dyes adsorption on MCNTs were performed by contacting with sudan dyes solution. The adsorption capacity was enhanced rapidly in the initial time, and then slowed down and reached equilibrium finally. In order to analyze the adsorption kinetic models, pseudo-first-order and pseudo-secondorder [37] were applied to the experiment data. The pseudo-firstorder model is expressed as follows: ln ðqe  qt Þ ¼ ln qe  k1 t

(6)

where qe (mg/g) and qt (mg/g) refer to the amount of sudan dyes adsorbed at equilibrium and any time t (min), respectively. k1 (1/ min) is the rate constant of pseudo-first order which can be calculated from the plot of ln (qe  qt) versus t. The pseudo-second-order model is expressed as follows: t 1 t ¼ þ qt k2 q2e qe

(7)

Table 2 The thermodynamic parameters for the adsorption of sudan dyes onto MCNTs.

qm (mg/g) R2

aL (L/mg) Kf

n

R2

Sample

T (K)

ln Kc

DG8 (kJ/mol)

DH8 (kJ/mol)

DS8 (kJ/(mol K))

26.5252 21.9298 19.3050 24.0385 20.0000 19.0476 23.0415 19.9203 18.9753 22.7790 19.8020 18.8679

1.1636 0.4043 0.2680 0.4837 0.3259 0.2214 0.3012 0.2505 0.1901 0.2454 0.2453 0.2051

1.4804 1.6186 1.5898 1.6005 1.6521 1.6013 1.4415 1.4802 1.4393 1.4397 1.4633 1.4548

0.9658 0.9071 0.9231 0.9677 0.9518 0.9359 0.9617 0.9518 0.9404 0.9562 0.9502 0.9519

Sudan I

298 308 318 298 308 318 298 308 318 298 308 318

3.892 3.178 2.023 3.147 2.667 2.249 2.351 2.117 1.754 2.208 2.048 1.658

9.642 8.138 6.667 7.798 6.830 5.946 5.824 5.420 4.638 5.471 5.243 4.384

55.345

0.153

35.398

0.093

23.435

0.059

21.567

0.054

0.5520 0.4689 0.5176 0.6450 0.5982 0.5177 0.8145 0.5618 0.4960 0.5432 0.5545 0.5432

13.6154 5.8194 3.9040 6.7517 4.5073 3.4120 4.7660 3.6947 2.9288 4.1475 3.5008 3.0429

Sudan II

Sudan III

Sudan IV

Please cite this article in press as: X. Sun, et al., J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.07.034

G Model

JIEC-2144; No. of Pages 5 X. Sun et al. / Journal of Industrial and Engineering Chemistry xxx (2014) xxx–xxx

5

Table 3 Kinetic parameters for the adsorption of sudan dyes onto MCNTs. Sample

Sudan Sudan Sudan Sudan

I II III IV

Pseudo-first-order order

Pseudo-second-order 2

k1 (1/min)

qe(cal) (mg/g)

R

qe(exp) (mg/g)

k2 (g/(mg min))

qe(cal) (mg/g)

R2

qe(exp) (mg/g)

0.187 1.986 0.067 0.063

0.435 0.958 0.530 0.412

0.985 0.540 0.895 0.642

1.000 0.971 0.951 0.921

0.642 1.018 0.32 1.022

1.069 0.987 0.946 0.908

1.000 0.999 0.998 0.998

1.000 0.971 0.951 0.921

where k2 is the rate constant of second order adsorption. The slope and intercept of plots of t/qt versus t were used to calculate the second rate constant k2. The pseudo-first-order rate constant (k1), pseudo-second order rate constant (k2), correlation coefficients (R2) along with the experimental and calculated uptake capacity (qe) are presented in Table 3. The adsorption parameters for adsorption kinetics were determined at different times by the pseudo-first-order and pseudo-second-order. The values of R2 for the pseudo-secondorder model (R2 > 0.920) were higher than that of pseudo- firstorder (R2 > 0.642). The values qe(cal) estimated from the pseudofirst-order kinetic model showed significant difference compared with experimental values qe(exp), indicating that the adsorption process did not follow the pseudo-first-order adsorption expression. The pseudo-second-order kinetic model was feasible to describe the adsorption process of sudan dyes on the MCNTs. Desorption and regeneration To investigate the possibility of regeneration of the MCNTs, the desorption experiments were performed. The results showed that ethanol could be used for the regeneration of the MCNTs and the desorption ratio of the dyes was increased when acetic acid was added into ethanol. When 2% acetic acid ethanol was used as the desorption solvent, the adsorbents can be reused 10 times without decreasing obviously. Conclusions In this study, MCNTs were successfully coated with Fe3O4. They were used for the removal of sudan dyes from aqueous solution based on magnetic separation technique conveniently, fast and efficiently. The new adsorbent was prepared easily and utilized conveniently. Magnetic separation in the method shortened separation times, and the MCNTs can be reused at least ten times without obvious decrease in the removal efficiency. The test results of kinetic and thermodynamic studies would be useful for removal of sudan dyes from water. Acknowledgements This work was supported by the Fundamental Research Funds for the Central Universities (no. 2572014EB06) and Northeast

Forestry University Student Research Training Program (no. KY2013018)

References [1] R. Rebane, I. Leito, S. Yurchenko, K. Herodes, J. Chromatogr. A. 1217 (2010) 2747. [2] R. Noguerol-Cal, J.M. Lo´pez-Vilariˇno, G. Ferna´ndez-Marta´nez, L. Barral-Losada, M.V. Gonza´lez-Rodrı´guez, J. Chromatogr. A 1179 (2008) 152. [3] C.Y. Long, Z.B. Mai, Y.F. Yang, B.H. Zhu, X.M. Xu, L. Lu, X.Y. Zou, J. Chromatogr. A 1216 (2009) 8379. [4] F.J. Lo´pez-Jime´nez, S. Rubio, D. Pe´rez-Bendito, Food Chem. 121 (2010) 763. [5] R.D. Ambashta, M. Sillanpa¨a¨, J. Hazard. Mater. 180 (2010) 38. [6] W. Liu, W.J. Zhao, J.B. Chen, M.M. Yang, Anal. Chim. Acta 605 (2007) 41. [7] O. Chailapakul, W. Wonsawat, W. Siangproh, K. Grudpan, Y.F. Zhao, Z.W. Zhu, Food Chem. 109 (2008) 876. [8] C.D. Zhao, T. Zhao, X.Y. Liu, H.X. Zhang, J. Chromatogr. A 1217 (2010) 6995. [9] C.Z. Jiang, Y. Sun, X. Yu, L. Zhang, X.M. Sun, Y. Gao, H.Q. Zhang, D.Q. Song, Talanta 89 (2012) 38. [10] J.D. Orbell, L. Godhino, S.W. Bigger, T.M. Nguyen, L.N. Ngeh, J. Chem. Educ. 74 (1977) 1446. [11] X. Peng, Z. Luan, Z. Di, Z. Zhang, C. Zhu, Carbon 43 (2005) 880. [12] X.K. Wang, C.L. Chen, J.Z. Du, X.L. Tan, D. Xu, S.M. Yu, Environ. Sci. Technol. 39 (2005) 7084. [13] X.K. Wang, C.L. Chen, W.P. Hu, A.P. Ding, D. Xu, X. Zhou, Environ. Sci. Technol. 39 (2005) 2856. [14] C.L. Chen, X.K. Wang, Ind. Eng. Chem. Res. 45 (2006) 9144. [15] D. Xu, X.L. Tan, C.L. Chen, X.K. Wang, J. Hazard. Mater. 154 (2008) 407. [16] H. Yan, A. Gong, H. He, J. Zhou, Y. Wei, L. Lv, Chemosphere 62 (2006) 142. [17] X.L. Tan, M. Fang, X.K. Wang, J. Nanosci. Nanotechnol. 8 (2008) 5624. [18] Y.S. Minaberry, G.J. Gordillo, Chemosphere 78 (2010) 1356. [19] S. Iijima, Nature 354 (1991) 56. [20] Y. Suttisawat, P. Rangsunvigit, B. Kitiyanan, M. Williams, P. Ndungu, M.V. Lototskyy, A. Nechaev, V. Linkov, S. Kulprathipanja, Int. J. Hydrogen Energy 34 (2009) 6669. [21] S. Hrapovic, K.B. Male, Y. Liu, J.H.T. Luong, Anal. Lett. 41 (2008) 278. [22] V. Sgobba, D.M. Guldi, Chem. Soc. Rev. 38 (2009) 165. [23] W. Konicki, I. Pełech, E. Mijowska b, I. Jasin´ ska, Chem. Eng. J. 210 (2012) 87. [24] C.H. Setchell, J. Chem. Technol. Biotechnol. 35 (1985) 175. [25] M. Takafuji, S. Ide, H. Ihara, Z. Xu, Chem. Mater. 16 (2004) 1977. [26] J. Ma, W. Liu, S.P. Zhang, J.T. Zhao, W.L. Liu, J. Alloys. Compd. 509 (2011) 7895. [27] A. Sari, M. Tuzen, D. Citak, M. Soylak, J. Hazard. Mater. 149 (2007) 283. [28] T.R. Bastami, M.H. Entezari, Chem. Eng. J. 210 (2012) 510. [29] S. Babel, T.A. Kurniawan, Chemosphere 54 (2004) 951. [30] Z. Aksu, F. Go¨nen, Process Biochem. 39 (2004) 599. [31] B. Wang, J.L. Gong, G.M. Guang, W.J. Zhou, J. Environ. Sci.—China 28 (11) (2008) 1009. [32] G.D. Sheng, D.D. Shao, X.M. Ren, X.Q. Wang, J.X. Li, Y.X. Chen, X.K. Wang, J. Hazard. Mater. 178 (2010) 505. [33] S. Zhang, T. Shao, S.S.K. Bekaroglu, T. Karanl, Water Res. 44 (2010) 2067. [34] J.J. Yin, R. Chen, Y.S. Ji, C.D. Zhao, G.H. Zhao, H.X. Zhang, Chem. Eng. J. 157 (2010) 466. [35] Langmuir, J. Am. Chem. Soc. 40 (1918) 1361. [36] M.F. Freundlich, J. Phys. Chem. 57 (1906) 385. [37] K.V. Kumar, J. Hazard. Mater. 137 (2006) 1538.

Please cite this article in press as: X. Sun, et al., J. Ind. Eng. Chem. (2014), http://dx.doi.org/10.1016/j.jiec.2014.07.034