Adsorption of anionic dyes from aqueous solution by iron oxide nanospheres

Adsorption of anionic dyes from aqueous solution by iron oxide nanospheres

G Model JIEC-1635; No. of Pages 7 Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx Contents lists available at ScienceDirect Jour...

1MB Sizes 0 Downloads 14 Views

G Model

JIEC-1635; No. of Pages 7 Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

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

Adsorption of anionic dyes from aqueous solution by iron oxide nanospheres Maryam Khosravi, Saeid Azizian * Department of Physical Chemistry, Faculty of Chemistry, Bu-Ali Sina University, Hamedan, Iran

A R T I C L E I N F O

Article history: Received 23 June 2013 Accepted 19 October 2013 Available online xxx Keywords: Iron oxide nanospheres Adsorption Isotherm Adsorption kinetics Anionic dyes

A B S T R A C T

Magnetic iron oxide nanospheres were fabricated for removal of anionic dyes from aqueous solution. The iron oxide nanospheres are ferromagnetic and can be separated by an external magnetic field. The iron oxide has been synthesized with the solvothermal method. The prepared product was characterized by SEM and XRD methods. The adsorption of Reactive Orange (RO) and Reactive Yellow (RY) on to iron oxide nanospheres was investigated from both equilibrium and kinetic point of view and the results were modeled with appropriate equilibrium and kinetic models. The influences of solution pH and temperature on the removal efficiency were studied too. ß 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.

1. Introduction Among the different pollutants released to the environment from industries, dyes can be considered as one of the most dangerous contaminants. Most of the dyes are toxic and must be removed before discharge in to receiving-streams since they can reduce light penetration and increase the chemical oxygen demand (COD) [1]. Since dyes are stable, recalcitrant, colorant, and even potentially carcinogenic and toxic, their release in to the environmental poses serious environmental, esthetical and health problem [2]. Control of dye pollutants in water is an important measure in environmental protection. Various physical methods like adsorption [3] membrane separation, catalytic oxidation [4] and chemical methods such as chlorination and ozonation [5] have been used for dye removal. Among these, adsorption is a well-known separation process and is widely used to remove certain classes of chemical pollutants like dyes from aqueous solution. In this regard, much attention has recently been paid to the use of nanosized material due to their high surface area. For example silver nanoparticle loaded activated carbon [6] ordered mesoporous alumina [7], TiO2 nanoparticles [8,9] have been used as adsorbent for dye removal from aqueous solution. During the past decade various types of iron oxide have attracted attention because of the reactive surface. The magnetic properties of iron oxide particles allow the fast magnetic separation after the adsorption process. Therefore, the magnetic particles can be a very good choice for adsorption of various dyes.

* Corresponding author. Tel.: +98 8118282807; fax: +98 8118380709. E-mail addresses: [email protected], [email protected] (S. Azizian).

Zhong et al. synthesized flower like iron oxide nanostructures and used for water treatment [10]. Porous iron oxide nanospheres was used recently by Zhu et al. [11] for the catalytic degradation of Xylenol Orange. Different methods have been used to synthesize porous iron oxide. Zhong et al. synthesized porous Fe3O4 by reflux at 195 8C and calcination at 500 8C under N2 protection for 3 h [10]. Liu et al. synthesized porous Fe3O4 nanoparticles via a simple hydrothermal method [12]. In the present work the porous iron oxide nanospheres was synthesized by using Zhu et al. method [11] and its ability as adsorbent was investigated for removal of anionic dyes with azo functional group from aqueous solution. Our adsorption studies include kinetics and equilibrium. The effects of pH and temperature on the adsorption efficiency were studied too. 2. Experimental 2.1. Materials FeCl36H2O (99%), polyvinyl pyrrolidone (PVP), sodium acetate trihydrate (99.5%), ethylene glycol (99.5%) were purchased from Merck Co. Dyes (Reactive Orange 13 and Reactive Yellow 15, Fig. 1) were obtained from Alvan Sabet company. 2.2. Synthesis of adsorbent Iron oxide nanosphere was prepared by Zhu et al. reported method [11]. Briefly, 0.75 g Fe3O46H2O, 0.5 g PVP and 0.41 g NaAc3H2O were dissolved in 25 ml ethylene glycol. The mixture was stirred for 2 h until it become homogeneous and it was

1226-086X/$ – see front matter ß 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jiec.2013.10.040

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

2

O

ONa

and shaking speed of 150 rpm for 24 h to reach equilibrium condition. Then, the solution was analyzed for residual RO and RY concentration using UV–vis spectrophotometer at 489 nm and 415 nm, respectively. The amount of adsorbed RO and RY per unit mass of adsorbent at equilibrium qe (mg/g) were calculated by using the following equation:

S O

O S NaO

HO

N O

N

N

N O

Cl

N N

NH2

S NaO

qe ¼

O

C0  Ce V W

(2)

where Ce (mg/l) is the equilibrium concentrations of dye.

Reactive Orange 13

2.6. Effect of pH

H3C H3C NaO3SOH2CH2CO2S

N

N N

The solution pH is an important parameter that affects the adsorption performance of dye molecules. In this study, 5 ml of dyes solutions of concentration 40 mg/l in the pH range of 4–10.5 (using HCl and NaOH) were prepared (pH meter OHAUS-STARTER 3000). Then 5 mg of iron oxide was added and the solutions were shaken for 4 h (150 rpm and 25 8C). Then, the solutions were analyzed with UV/vis spectrophotometer. The removal percentage of dye was calculated by:

SO3Na

N

OCH3

OH

Reactive Yellow 15 Fig. 1. The chemical structure of anionic dyes used in adsorption studies.

%Re ¼

subsequently transferred in to a Teflon-lined stainless-steel autoclave. The autoclave was maintained at 220 8C for 3 h. The black product was collected and washed with ethanol for several times, and then dried in a oven at 60 8C for 24 h. The weight of obtained dry sample is about 0.4 g. 2.3. Characterization of adsorbent The prepared iron oxide particles were characterized by scanning electron microscopy (SEM) (Hitachi-Japan-S4160) and X-ray diffraction (XRD) (ADP2000 ITALSTRUCTURE Italia) methods. The Fourier transform infrared (FT-IR) spectra were recorded with a Tensor 65 spectrometer (Perkin Elmer) using a KBr wafer with the wave number ranging 400–4000 cm1. 2.4. Kinetic adsorption studies The kinetic studies were carried out at three different initial concentrations of RO (20, 40, 80 mg/l) and RY (20, 50, 80 mg/l) with pH 7.1 and 7.2, respectively. For such studies a series of, 5 ml of dye solution was added to 5 mg of adsorbent. The samples were shaken at 25 8C and 150 rpm. At different time intervals the iron oxide was separated by a strong permanent magnet and the concentration of the residual pollutant in the solution was analyzed by UV-vis spectrophotometer (PG Instrument LTD model T80). The amount of pollutant adsorbed at each time interval per unit mass of the adsorbent qt (mg/g), was calculated by the following equation: qt ¼

C0  Ct V W

(1)

where C0 (mg/l) is initial liquid-phase concentration of the pollutant and Ct (mg/l) is its concentration at time t. V (l) is the volume of the solution and W (g) is the mass of the adsorbent. 2.5. Equilibrium adsorption studies The equilibrium studies were carried out by adding 5 ml of different initial concentration (6–120 mg/l) of RO and RY solution to 5 mg of adsorbent. The samples were shaken in a shaker thermostat (n-BioTek, NB-300) in a isothermal condition of 25 8C

ðA0  AÞ  100 A0

(3)

where A0 is the initial absorbance and A is the equilibrium absorbance at maximum wavelength for each dye. 2.7. Point of zero charge pH The pH of point of zero charge (pHpzc) of adsorbent is an important characteristic that determined by Faria et al. method [13]. In a typical method, 10 ml NaCl (0.01 M) solution with pH value between 4 and 11.1 (using HCl and NaOH) was added to 30 mg of iron oxide nanospheres and agitated for 24 h in a shaker (150 rpm and 25 8C). The point zero charge (pHpzc) was determined by plotting final pH, versus initial pH of solution. 2.8. Utilization of used adsorbent In order to investigate that how many times, iron oxide nanosphere can be used as adsorbent, the following experiments were performed. First, the used adsorbent without any treatment was added to a new dye solution (20 mg/l) and the removal percentage was obtained after 5 min (Run 2). Then the adsorbent which was used in the previous section was separated and added to another new dye solution and the removal percentage was calculated (Run 3). Second, the used adsorbent was washed with distilled water several times and used as adsorbent for other new dye solution. Third, the used adsorbent was washed with NaOH solution (0.01 M) for 45 min and then used for dye removal from new solutions. 3. Results and discussion 3.1. Characterization of adsorbent The morphology of the prepared iron oxide was studied by scanning electron microscopy (SEM). Fig. 2(a) shows the SEM image of the prepared iron oxide which clearly shows that the particles are spherical with radius less than 75 nm. Each particle consist of a number of smaller nanoparticles as like as Fig. 2(b). These results are in agreement with previous reported data [11]. The X-ray diffraction (XRD) pattern of the prepared iron oxide is shown in Fig. 3. The XRD pattern shows five characteristic peaks at 2u = 30.18, 35.88, 43.18, 57.38 and 638 that can be indexed to (2 2 0),

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

3

a 2923.03

3446.25

b

3900

3400

572.75

1423.3 1288.52 1668.21

2900

2400

1900

1400

900

400

wavenumber (cm -1) Fig. 4. FT-IR spectra of iron oxide nanospheres (a) and PVP (b).

O H2 C C n* H PVP

N *

(2 0 2)

(1 1 0)

Fe3O4

Fig. 5. Schematic illustration of the interaction between iron oxide nano-sphere and PVP.

Fe3 O4 -Fe2 O3

(a)

35 20 mg/l 80 mg/l

40 mg/l

25 qt (mg/g)

(4 4 0)

(5 1 1)

30 (4 0 0)

Intensity

(2 2 0)

(3 1 1)

Fig. 2. (a) SEM images of iron oxide nanospheres and (b) schematic of iron oxide nanospheres.

20 15 10

5

15

25

35

45

55

65

5

2 (degree)

0

Fig. 3. The XRD spectra of the prepared iron oxide nanosphere.

0

10

20

30

40

50

60

70

40

50

60

70

t (min)

(b)

20 mg/l 30

50 mg/l

80 mg/l

25

qt (mg/g)

(3 1 1), (4 0 0), (5 1 1) and (4 4 0) planes of Fe3O4 [10,11] and there are two peaks at 2u = 37.68 and 43.98 resulting from the (1 1 0) and (2 0 2) faces of a-Fe2O3 [10,14]. Therefore the obtained XRD pattern indicates that the prepared iron oxide nanospheres consist of both Fe3O4 and a-Fe2O3. The IR spectra of iron oxide nanospheres (curve a) and the capping agent PVP (curve b) are shown in Fig. 4. The peak near 3446 cm1 refer to the O–H stretching vibration while at 2923 cm1 belongs to the symmetrical stretching C–H and the absorption peak at 1668 cm1 can be assigned to the C5 5O stretching. The vibrating absorption peaks at 1423and 1288 cm1are from the asymmetric C–H. It is clear that the spectrum of the obtained iron oxide (Fig. 4(a)) is similar to that of PVP (Fig. 4(b)). This result indicates that the PVP is coordinated to the prepared iron oxide nanospheres, which is similar to the Zhu et al. [11] reported. The possible interaction between the obtained Fe3O4 nanocrystals and PVP is schematically shown in Fig. 5 [11].

35

20 15 10 5 0 0

10

20

30

t (min) Fig. 6. Experimental kinetic data for the adsorption of (a) RO and (b) RY by iron oxide nanospheres at different initial concentrations. The solid lines represent the predicted values by FL-PSO model for RO and FL-PFO model for RY.

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

4

Table 1 Obtained kinetic model parameters for the adsorption of RO onto iron oxide nanospheres. Kinetic model References C0 = 20 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21] C0 = 40 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21] C0 = 80 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21]

Equation

qe (mg/g)

k

k1 (1/s)

k2 (g/mgs)

F2

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ 2 a qt ¼ kqe t =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

15.01 16.13 16.20 – 15.41 16.28 15.60

– – – – 5.8  103 0.0420 0.5770

0.4603 3.9  104 – – – –

– 0.0418 – – – – –

– – 0.9993 – – – –

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ 2 a qt ¼ kqe t =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

19.42 20.72 21.16 – 21.50 28.69 24.50

– – – – 1  106 0.0282 0.7390

1.1571 – 7.7  104 – – – –

– 0.0808 – – – – –

– – 0.9994 – – – –

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ qt ¼ kq2e t a =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

25.42 27.22 27.74 – 28.33 32.39 29.42

– – – – 1.8069 0.0366 0.9025

1.4296 – 8.4  104 – – – –

– 0.0712 – – – – –

– – 0.9994 – – – –

a

– – – 73.250 – – – – – – 562.82 – – – – – – 1340.3 – – –

b

n

a

r2

– – – 0.4491 – – –

– – – – 2.05 – –

– – – – – 0.95 0.62

0.9777 0.9939 0.9936 0.9722 0.9901 0.9940 0.9920

– – – 0.4251 – – –

– – – – 4.25 – –

– – – – – 0.35 0.28

0.9070 0.9673 0.9580 0.9982 0.9937 0.9989 0.9985

– – – 0.3456 – – –

– – – – 5.11 – –

– – – – – 0.46 0.34

0.9607 0.9748 0.9666 0.9911 0.9951 0.9958 0.9924

b

n

a

r2

– – – 0.4869 – – –

– – – – 3.45 – –

– – – – – 0.53 0.42

0.9220 0.9743 0.9658 0.9962 0.9991 0.9980 0.9990

Table 2 Obtained kinetic model parameters for the adsorption of RY onto iron oxide nanospheres. Kinetic model References C0 = 20 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21] C0 = 50 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21] C0 = 80 mg/l PFO [15] PSO [15,16] MOE [18] Elovich [17] MPnO [20] FL-PSO [21] FL-PFO [21]

Equation

qe (mg/g)

k

k1 (1/s)

k2 (g/mgs)

F2

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ qt ¼ kq2e t a =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

13.01 14.09 14.62 – 14.69 17.67 15.44

– – – – 2.5  105 0.0336 0.5526

0.4984 – 4.15  104 – – – –

– 0.0525 – – – – –

– – 0.9992 – – – –

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ qt ¼ kq2e t a =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

18.63 19.79 20.22 – 20.07 22.89 20.62

– – – – 1.7  105 0.0358 0.6851

0.6573 – 5.7  104 – – – –

– 0.0512 – – – – –

– – 0.9993 – – – –

– – – 141.36 – – –

– – – 0.3768 – – –

– – – – 3.48 – –

– – – – – 0.58 0.44

0.9265 0.9766 0.9700 0.9863 0.9982 0.9949 0.9968

qt = qe(1  exp( k1t)) qt ¼ k2 q2e t=ð1 þ k2 qe tÞ qt = qe((1  exp( k1t))/(1  F2exp( k1t))) qt = (1/b)ln(1 + abt) n1 1=n qt ¼ qe ð1  enkqe t Þ 2 a qt ¼ kqe t =ð1 þ kqe t a Þ a qt ¼ qe ½1  ekt 

23.80 25.46 26.10 -– 26.52 31.96 27.88

– – –

0.8063 – 6  104 – – – –

– 0.0450 – – – – –

– – 0.9993 – – – –

– – – 228.55 – – –

– – – 0.3030 – – –

– – – – 4.11 – –

– – – – – 0.47 0.47

0.9103 0.9696 0.9601 0.9935 0.9986 0.9967 0.9976

1  106 0.2388 0.6808

3.2. Adsorption kinetics modeling The kinetics of adsorption for the treatment of dye-containing effluents has already been studied. Study of adsorption kinetics is important because the rate of adsorption (which is one of the criteria for efficiency of adsorbent) and also the mechanism of adsorption can be concluded from kinetic studies. Fig. 6 shows the variation of the amount of adsorbed (qt) as a function of time. The data presented in Fig. 6 indicate that the rate of adsorption and also the amount of ad-species increase with increase of solute concentration. The rate of adsorption for both dyes is high at initial times of adsorption. For both dyes most of adsorption takes

a

– – – 52.702 – – –

place within 10 min which indicate that the rate of dye adsorption by iron oxide nanospheres is high. In order to analyze the adsorption kinetics of RO and RY by iron oxide nanospheres the pseudo-first order (PFO) [15], pseudo-second order (PSO) [15,16], Elovich [17], mixed 1,2-order (MOE) [18], pseudo-n-order (PnO) [19], modified pseudo-n-order (MPnO) [20], and recently presented models including fractal-like pseudo-first order (FL-PFO) and fractal-like pseudo-second order (FL-PSO) [21] models, were tested. The result of fitting is listed in Tables 1 and 2. Higher correlation coefficient values and similarity of qe,cal and qe,exp values revealed applicability of FL-PSO kinetic model for RO and FL-PFO kinetic model for RY. Based on the obtained correlation

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

(a)

coefficient of adsorption in these systems is dependent to the time [21]. The physical meaning of this time dependency is changing of reaction path with time [21].

35 30

qe (mg/g)

25

3.3. Adsorption isotherms modeling

20 15 10 5 0

0

10

20

30

40

50

60

70

80

90

90

100

Ce (mg/l)

(b)

30 25 20

qe (mg/g)

5

15 10 5 0

0

10

20

30

40

50

60

70

80

Ce (mg/l) Fig. 7. Adsorption isotherm of (a) RO and (b) RY onto the iron oxide nanospheres.

coefficient values, r2, the MPnO model describes the adsorption of RY by iron oxide better than the other kinetic models, but its high n value is not acceptable and therefore the best model for kinetic of adsorption RY by iron oxide is the FL-PFO model. The result of fitting by these kinetic models are shown in Fig. 6. As mentioned the adsorption kinetics of RO and RY on to iron oxide nanospheres follows fractal like kinetic models. This means that the rate

The equilibrium relationship between the quantity of adsorbed per unit of adsorbent (qe) and residual solution concentration (ce) at a constant temperature is known as the adsorption isotherm. Adsorption isotherms describe how adsorbate interact with sorbent materials. Fig. 7 shows the adsorption isotherm of RO and RY. The equilibrium data was modeled with Langmuir (L) [22], Freundlich (F) [22], Redlich-Peterson (R-P) [22], Toth [22], Langmuir–Freundlich (L–F) [23] and Extended Langmuir (E-L) [24] models (Tables 3 and 4). The parameters of these models were calculated and summarized in Tables 3 and 4. According to the results, the Langmuir–Freundlich model was best fitted to the equilibrium data. This means that adsorbent provides a heterogeneous surface. The results of fitting by this isotherm are shown in Fig. 7. The obtained adsorption capacities for RO and RY are 32.50 and 25.04 mg/g, respectively. The difference between adsorption capacity of RY and RO may be related to the presence of higher number of functional groups on the RY structure. Although the previous reports for adsorption of RO by different adsorbents shows [25–28] higher adsorption capacity than iron oxide nanosphere but the rate of adsorption on the iron oxide nanospheres is higher than the other adsorbent and this case is useful in some of practical applications. 3.4. Effect of solution pH We have investigated the removal percentage of RO and RY by iron oxide nanospheres at different pH. The effect of solution pH on the adsorption of dye molecules are shown in Fig. 8. The removal percentage for the RO and RY (anionic dyes) decreases with the increase of pH values. As observed in Fig. 9, the point of zero charge for iron oxide nanospheres is 3.8. As shown in Fig. 8 the pH range is between 4 and 11.1, i.e. higher than the pHpzc. In other words at pH > pHpzc the surface charge is negative and by increasing of pH the negative charge on the surface increases. Therefore for both RY and RO dyes by increasing of pH from 4 to 11.1 the removal efficiency decreases

Table 3 Obtained isotherm model parameters for the adsorption of RO onto iron oxide nanospheres. Isotherm model References

Equation

qm (mg/g)

KL (l/mg)

KF (l mg(1(1/n))/g)

KR (l/g)

1/n

aR (l/mg)b

bT

b

a

r2

L [22] F [22] R-P [22] Toth [22] L–F [23] E-L [24]

qe = qmKLCe/(1 + KLCe) 1=n qe ¼ K F Ce b qe ¼ K R C e =ð1 þ aR Ce Þ 1=nT nT qe ¼ qm bT C e =ð½1 þ ðbT C e Þ  Þ 1=n 1=n qe ¼ qm KCe =ð1 þ KCe Þ pffiffiffiffiffiffiffiffiffiffiffi qe ¼ qm K L C e =ð1 þ K L C e þ a K L C e Þ

30.51 – – 32.90 32.50 33.23

2.67 – – – 1.64 9.08

– 19.52 – – – –

– – 123.59 – – –

– 0.12 – 0.1 0.62 –

– – 4.81

– – 0.95

– –

– –

– – – 1.83 – –

– –

– – 1.56

0.9844 0.9581 0.9947 0.9972 0.9974 0.9971

Table 4 Obtained isotherm model parameters for the adsorption of RY onto iron oxied nanospheres. Isotherm model References

Equation

qm (mg/g)

KL (l/mg)

KF (l mg(1(1/n))/g)

KR (l/g)

1/n

aR (l/mg) b

bT

b

a

r2

L [22] F [22] R-P [22] Toth [22] L–F [23] E-L [24]

qe = qmKLCe/(1 + KLCe) 1=n qe ¼ K F Ce b qe ¼ K R C e =ð1 þ aR Ce Þ 1=nT nT qe ¼ qm bT C e =ð½1 þ ðbT C e Þ  Þ 1=n 1=n qe ¼ qm KCe =ð1 þ KCe Þ pffiffiffiffiffiffiffiffiffiffiffi qe ¼ qm K L C e =ð1 þ K L C e þ a K L C e Þ

25.47 – – 25.10 25.04 25.00

0.61 – – – 0.59 0.56

– 12.10 – – – –

– – 15.38 – – –

– 0.17 – 0.53 1.13 –

– – 0.6022

– – – 1.14 – –

– – 1.00

– –

0.9942 0.9140 0.9940 0.9948 0.9952 0.9943

– –

– –

– – 0.11

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

6

80 RO

70

RY

60

% Re

50 40 30 20 10 0 3

4

5

6

7

8

9

10

11

12

pH Fig. 8. Effect of solution pH on the removal percentage of RO and RY by iron oxide nanospheres.

12 10

pHf

8 6 4

pHzpc

2 0 2

4

6

8

10

12

pHi

Fig. 11. Dye removal percentage by the used adsorbent, without treatment and treated with water or NaOH solution (a) RO and (b) RY.

Fig. 9. pH variation in terms of initial pH of solution for determination of point of zero charge pH.

because of repulsion forces between anionic dyes and negative surface charge. Therefore for both of these dyes the optimum pH for higher removal from aqueous solution is around 4. 3.5. Effect of temperature The effect of temperature on dyes removal efficiency by iron oxide nanospheres were examined at four temperature (25, 35, 45 and 55 8C). The experiments were performed at concentration of 20 mg/l RO and RY. In the studies 5 mg of iron oxide nanospheres with 5 ml of dye solutions were placed in a shaker (150 rpm) for

1 h. Then the solutions were analyzed with UV/vis spectrophotometer for residual concentration. Fig. 10 shows the removal percentage of RO and RY by iron oxide at different temperatures. It is clear that for both dyes the removal percentage decrease with temperature. By increasing of temperature from 25 to 55 8C the removal percentage of RO decrease from 98% to 88% and RY decrease from 75% to 55%. These results indicate that the adsorption of these dyes onto iron oxide nanospheres is exothermic. Since the effect of temperature on the removal percentage of RY is higher than RO, it can be suggested that the adsorption of RY onto iron oxide nanospheres is more exothermic than RO. 3.6. Performance of used adsorbent Fig. 11 shows the performance of used adsorbent without treatment and treated with water or NaOH solution, for removal of RO form aqueous solution. As this figure shows the adsorbent can be regenerated by NaOH solution moderately. 4. Conclusion

Fig. 10. Effect of temperature on the removal percentage of RO and RY by iron oxide nanospheres.

In the present study, iron oxide nanospheres have been synthesized with solvothermal method and used as an effective adsorbent for the removal of anionic dyes from aqueous solutions. The efficiency of the prepared iron oxide nanospheres as adsorbent was tested by adsorption of anionic dyes including RO and RY. The equilibrium adsorption studies show that the equilibrium data follow Langmuir–Freundlich isotherm which indicate that the surface is heterogeneous. The obtained maximum adsorption capacity for RO and RY are 32.5 and 25.0 mg/g, respectively. The kinetic studies reveals that the rate of adsorption onto iron oxide

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040

G Model

JIEC-1635; No. of Pages 7 M. Khosravi, S. Azizian / Journal of Industrial and Engineering Chemistry xxx (2013) xxx–xxx

nanospheres is rapid and mainly take place within 10 min. Following of adsorption kinetic data by fractal-like models indicates that the rate coefficient of adsorption is time dependent, i.e. the paths of adsorption changes with time. The optimum pH for removal of these dyes by iron oxide nanospheres is about 4, i.e. acidic solution. The adsorption of RY and RO onto iron oxide nanospheres is exothermic. Finally, since the prepared iron oxide nanospheres has a strong magnetic character and can be easily separated by a magnet from solution, it is a excellent candidate as easy to use adsorbent. References [1] R. Gong, M. Li, C. Yang, Y. Sun, J. Chen, Journal of Hazardous Materials 121 (2005) 247. [2] A. Gill, F.C.C. Assis, S. Albeniz, S.A. Korili, Chemical Engineering Journal 168 (2011) 1032. [3] M. Hejazifar, S. Azizian, H. Sarikhani, Q. Li, D. Zhao, Journal of Analytical and Applied Pyrolysis 92 (2011) 258. [4] H. Xu, M. Prasad, Y. Liuc, Journal of Hazardous Materials 165 (2009) 1186. [5] Y.M. Slokar, A.M. Le Marechal, Dyes Pigments 37 (1998) 335. [6] M. Ghaedi, B. Sadeghian, A. AmiriPebdani, R. Sahraei, A. Daneshfar, C. Duran, Chemical Engineering Journal 187 (2012) 133. [7] B. Yahyaei, S. Azizian, Chemical Engineering Journal 209 (2012) 589. [8] S. Jafari, S. Azizian, B. Jaleh, Journal of Industrial and Engineering Chemistry 18 (2012) 2124.

7

[9] S. Jafari, S. Azizian, B. Jaleh, Colloids and Surfaces A 384 (2011) 618. [10] L.S. Zhong, J.S. Hu, H.P. Liang, A.M. Cao, W.G. Song, L.G. Wan, Advanced Materials 18 (2006) 2426. [11] M. Zhu, G. Diao, Journal of Physical Chemistry C 115 (39) (2011) 18923. [12] X. Liu, Q. Hu, Z. Fang, Q. Wu, Q. Xie, Langmuir 25 (13) (2009) 7244. [13] P.C.C. Faria, J.J.M. Orfao, M.F.R. Pereira, Water Research 38 (2004) 2043. [14] T. Belin, N. Millot, N. Bovet, M. Gailhanou, Journal of Solid State Chemistry 180 (2007) 2377–2385. [15] S. Azizian, Journal of Colloid and Interface Science 276 (2004) 47. [16] Y.S. Ho, G. Makay, Process Safety and Environmental Protection 76 (1998) 332. [17] B. Saha, C. Orvig, Chemical Reviews 254 (2010) 2959. [18] A.W. Marczewski, Applied Surface Science 256 (2010) 5145. [19] A. Ozer, Journal of Hazardous Materials 141 (2007) 753. [20] S. Azizian, R. NaviriFallah, Applied Surface Science 256 (2010) 5153. [21] M. Haerifar, S. Azizian, Journal of Physical Chemistry C 116 (2012) 13111. [22] S. Basha, Z.V.P. Murthy, B. Jha, Industrial and Engineering Chemistry Research 47 (2008) 980. [23] S. Azizian, M. Haerifar, J. Basiri-Parsa, Chemosphere 68 (2007) 2040. [24] P. Zhang, L. Wang, Separation and Purification Technology 70 (2010) 367. [25] M. Ghaedi, F. Karimi, B. Barazesh, R. Sahraei, A. Daneshfar, Journal of Industrial and Engineering Chemistry 19 (2013) 756. [26] S. Hajati, M. Ghaedi, F. Karimi, B. Barazesh, R. Sahraei, A. Daneshfar, Journal of Industrial and Engineering Chemistry (2013) http://dx.doi.org/10.1016/ j.jieec.2013.05.015. [27] M. Auta, B.H. Hameed, Journal of Industrial and Engineering Chemistry (2013) http://dx.doi.org/10.1016/j.jiec.2013.06.013. [28] M. Ghaedi, A.M. Ghaedi, F. Abdi, M. Roosta, R. Sahraei, A. Daneshfar, Journal of Industrial and Engineering Chemistry (2013) http://dx.doi.org//10.1016/ j.jiec.2013.06.008.

Please cite this article in press as: M. Khosravi, S. Azizian, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.10.040