Adsorption of Pb(II) and Cu(II) from aqueous solution on magnetic porous ferrospinel MnFe2O4

Journal of Colloid and Interface Science 367 (2012) 415–421

Contents lists available at SciVerse ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

Adsorption of Pb(II) and Cu(II) from aqueous solution on magnetic porous ferrospinel MnFe2O4 Yueming Ren a, Nan Li a, Jing Feng a, Tianzhu Luan b,⇑, Qing Wen a, Zhanshuang Li a, Milin Zhang a a

Key Laboratory of Superlight Materials & Surface Technology, Ministry of Education, College of Material Science and Chemical Engineering, Harbin Engineering University, Harbin 150001, PR China b The First Affiliated Hospital of Harbin Medical University, Harbin 150001, PR China

a r t i c l e

i n f o

Article history: Received 15 July 2011 Accepted 10 October 2011 Available online 25 October 2011 Keywords: MnFe2O4 Magnetic Porous Pb(II) and Cu(II) Adsorption

a b s t r a c t The adsorption of Pb(II) and Cu(II) from aqueous solution on magnetic porous ferrospinel MnFe2O4 prepared by a sol–gel process was investigated. Single batch experiment was employed to test pH effect, sorption kinetics, and isotherm. The interaction mechanism and the regeneration were also explored. The results showed that Pb(II) and Cu(II) removal was strongly pH-dependent with an optimum pH value of 6.0, and the equilibrium time was 3.0 h. The adsorption process could be described by a pseudo-second-order model, and the initial sorption rates were 526.3 and 2631.5 lmol g1 min1 for Pb(II) and Cu(II) ions, respectively. The equilibrium data were corresponded well with Langmuir isotherm, and the maximum adsorption capacities were 333.3 and 952.4 lmol g1 for Pb(II) and Cu(II) ions, respectively. The adsorbed Pb(II) and Cu(II) ions were in the form of the complex with oxygen in carboxyl and hydroxyl groups binding on the surface of magnetic porous MnFe2O4. The sorbent could be reused for five times with high removal efficiency. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction Heavy metals are important contaminants in the liquid wastes of large number of industries such as plating, paint and dyes, glass operations, lead batteries, electroplating, mining, and smelters [1]. Pb(II) and Cu(II), in particular, are among the most common pollutants found in industrial effluents. According to the US Environmental Pollution Agency, they are highly toxic and can cause a variety of negative effects on human health even at low dosages, for example, anemia, encephalopathy, hepatitis, and the nephritic syndrome [2]. Therefore, it is necessary to reduce Pb(II) and Cu(II) from the source water. Conventional technologies for the removal of heavy metals from wastewater include chemical precipitation, ion exchange, electrochemical removal, membrane and microbe separation, adsorption, etc. [3–5]. Among these methods, adsorption is an attractive approach in groundwater and drinking water treatment, due to high removal efficiency and without yielding harmful by-products [6]. However, the adsorbents of heavy metal with high adsorption capacity, fast adsorption–desorption kinetics, easy fixation, and separation from water are in great demand. Recently, considerable attention has been paid to the investigation of different types of low-cost adsorbents especially using ⇑ Corresponding author. Fax: +86 451 85555058. E-mail address: [email protected] (T. Luan). 0021-9797/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2011.10.022

metal oxide, such as aluminum oxide, iron oxide, and manganese oxide [7,8]. The iron oxides including Fe3O4 and c-Fe2O3 [9,10] have been reported to be effective adsorbents for metal ion removal because of their affinity and consequent selectivity in the adsorption process. However, masses of the metal oxides present relative low adsorption capacities, low adsorption rate, narrow optimum pH ranges, and difficult recycling, which limit their applicability [11]. Spinel ferrites with the general formula MeFe2O4 (Me = Mn, Co, Ni, Cu, Mg, etc.) have high magnetic permeability and low magnetic losses, which are widely used in electrical and practical applications of information storage system, ferrofluid technology, magnetocaloric refrigeration, and medical diagnostics [12–14]. Manganese ferrospinel MnFe2O4 is a well-known magnetic material. It may be high efficient potentially for metal ion binding due to the active functional groups on the surface allowing chemical interactions, which usually produce a large adsorption capacity. The sorption abilities of MnFe2O4 for removing Cr(VI) [15], As(III), and As(VI) [16] from aqueous solutions have been reported, but the adsorption efficiency was not satisfactory. Furthermore, conventional porous materials provide effective sites for many adsorption processes, and their irregular and wide pore size distribution will be benefit to the adsorption. In our study, we used porous manganese ferrospinel MnFe2O4 synthesized by a sol–gel process with egg white to remove Pb(II) and Cu(II) ions from aqueous solutions. The pH effect, adsorption kinetics, equilibrium, and

416

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421

the reuse were presented. The SEM, EDX, FTIR, and XPS analyses were investigated to make a full understanding of adsorption mechanism.

3. Results and discussion

2. Experimental

As shown in Fig. 1, the uptakes of Pb(II) and Cu(II) on MnFe2O4 were both strongly dependent on pH. Clearly, the adsorption capacities were found to be increased with the increase in pH. At 2.0 < pH < 4.0, Pb(II) and Cu(II) were hardly adsorbed because the sorbent surface would acquire a positive charge when pH was less than 7.7 (pHpzc) [19]:

2.1. Materials Mn(NO3)2, Fe(NO3)39H2O, Pb(NO3)2, CuCl23H2O, HCl, and NaOH were acquired from Kermel Factory (Tianjin, China). The fresh eggs were purchased from local market. All stock solutions were prepared by reagent grade chemicals using distilled water. Magnetic porous MnFe2O4 was prepared by a sol–gel process with egg white in our laboratory. The details can be found in the previous work [17,18]. 2.2. Characteristic analysis Magnetic porous MnFe2O4 before and after sorption was characterized by the following techniques. The SEM examination and the EDX microanalysis were performed using a ZEISS DSM-960 microscope equipped with an EDX unit. The FTIR spectra were recorded in KBr pellets using an FTIR-Biorad 135 instrument in 400– 4000 cm1 range. The XPS measurements were taken using a PHI 5700 ESCA spectrometer with a monochromated Al Ka radiation (hm = 1486.6 eV). The pHpzc was measured using a powder addition method. 2.3. Adsorption and desorption experiments Single batch adsorption experiments were performed in 250mL flasks, each containing 25 mL the known concentration of Pb(II) or Cu(II) solution. After the addition of 0.025 g MnFe2O4 each, the flasks were kept for shaking at 200 rpm in a controlled shaker (Julabo SW-21C) at 25 ± 2 °C for 3.0 h (a kinetic study showing that the adsorption equilibrium was achieved). Effects of pH (2.0–9.0), kinetic experiments (t: 0–240 min, pH: 6.0), and adsorption isotherm (C0Pb(II): 48.26–1206 lmol g1, C0Cu(II) : 157.5–3937 lmol g1, pH: 6.0) were studied. In order to test the reproducibility, the experiments were carried in duplicate and the relative error was found to be within ±2%. The pH values of the experimental solutions were adjusted by HCl or NaOH solutions. Afterward, the adsorbent was magnetically separated from the aqueous solution, and the residual concentrations of metal ions were determined by inductively coupled plasma-optical emission spectrometer (ICPS-750, Perkin Elmer). The quantity of ions adsorbed per unit mass of used adsorbent and the removal efficiency were calculated according to the following equations:

ðC 0  C e ÞV qt ¼ m

3.1. Effect of pH

ROHðsurfÞ þ HþðaqÞ ¡  ROHþ2ðsurfÞ

ð3Þ

where R denoted the surface of MnFe2O4. At pH > 4.0, the concentration of hydrogen ion in the solution decreased, which might lead to the leaving of covered H+ on MnFe2O4 surface. The increasing deprotonation resulted in more negatively charged sites available to the metal ions [20], causing the adsorption capacity increased dramatically. When referred to pH > 7.7, it was easy for the positively charged metal ions to be bound due to the fact that a negative charge had been acquired on the surface of MnFe2O4:

ROHðsurfÞ þ OH1ðaqÞ ¡  ROðsurfÞ þ H2 O

ð4Þ

Pb(II) and Cu(II) ions would present in the species of M2+, M(OH)+, and M(OH)2 (M: Pb(II) or Cu(II)) at different pH values in the aqueous solution [21]. M2+ was the dominant species at low pH region, and it might hydrolyze to form M(OH)+ and M(OH)2 with the increase in pH. The precipitation constants of Pb(OH)2 and Cu(OH)2 are 1.6  1017 and 5.6  1020 [22], corresponding to Pb(II) and Cu(II) precipitations at pH 6.5 and pH 7.4, respectively, which mean that the precipitation might play a main role at pH 6.5–9.0 and pH 7.4–9.0 for Pb(II) and Cu(II) adsorption, respectively. Therefore, the most feasible pH range was selected as 6.0 for both Pb(II) and Cu(II) adsorption, at which there were no hydroxide precipitations formed and the adsorption capacities were high enough. 3.2. Adsorption kinetics Two common adsorption kinetic equations were employed to simulate the procedure as follows: pseudo-first-order equation (Eq. (5)) and pseudo-second-order equation (Eq. (6)):

logðqe  qt Þ ¼ log qe 

k1 t 2:303

ð5Þ

800 Cu(II) Pb(II)

700

ð1Þ

7.4

ðC 0  C e Þ  100% C0 1

ð2Þ

where qt (lmol g ) is the amount of metal adsorption on MnFe2O4 at contact time t (min), E (%) is removal efficiency, C0 and Ce (lmol L1) are the initial and equilibrium metal ion concentrations in the solution, respectively, V (L) is the volume of the solution, and m (g) is the amount of adsorbent added to the solution. Five cycles were carried out to validate the reusability; for a desorption test, 0.025 g MnFe2O4 of Cu(II) or Pb(II) ion saturation adsorption was dipped into 25 mL 0.2 M HCl solution, the mixture was stirred continuously in a shaker at 200 rpm for 12 h at 25 °C, and then concentrations of metal ions in the solution were analyzed. MnFe2O4 was used in the next adsorption cycle after washing and drying.

q t (µmol g

E¼

-1 )

600 500

Cu(OH)2 precipitation

400 300

6.5

200 100 0

Pb(OH)2 precipitation

1

2

3

4

5

6

7

8

9

10

pH Fig. 1. Effect of pH on the adsorption of Pb(II) and Cu(II) on magnetic porous ferrospinel MnFe2O4. Dose: 1 g L1, CPb(II): 231.6 lmol L1, CCu(II): 656.4 lmol L1, t: 4.0 h, T: 25 ± 2 °C.

417

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421

800

400 qe ( µmol g-1)

-1 )

1000

q t (µmo l g

600

Pb(II)-experimental Cu(II)-experimental Pb(II)-Pseudo-first-order Cu(II)-Pseudo-first-order Pb(II)-Pseudo-second-order Cu(II)-Pseudo-second-order

1200

600 400 200

Pb(II)Experimental Langmuir equation Freundlich equation

200 0 0

200

400

600

800

1000

1200

0 0

50

100

150

200

800

250

t(min)

Cu(II)Experimental Langmuir equation Freundlich equation

400

Fig. 2. Pseudo-first-order and pseudo-second-order model of Pb(II) and Cu(II) adsorption on magnetic porous ferrospinel MnFe2O4. Dose: 1 g L1, CPb(II): 231.6 lmol L1, CCu(II): 656.4 lmol L1, pH: 6.0, T: 25 ± 2 °C.

0 0

500 1000 1500 2000 2500 3000 3500 Ce (µmolL-1)

Table 1 Kinetic parameters for the adsorption of Pb(II) and Cu(II) on magnetic porous ferrospinel MnFe2O4. Pseudo-first-order

Parameters

Pb(II)

Cu(II)

qe (lmol g ) K1 (min1) R2

29.56 0.01710 0.5433

172.8 0.02540 0.8615

Parameters qe (lmol g1) K2 (g lmol1 min1) R2

Pb(II) 222.2 0.01066 0.9999

Cu(II) 588.2 0.007606 0.9998

1

Pseudo-second-order

Fig. 3. Adsorption isotherms of Pb(II) and Cu(II) on magnetic porous ferrospinel MnFe2O4. Dose: 1 g L1, pH: 6.0, T: 25 ± 2 °C, t: 3.0 h.

Table 2 Langmuir and Freundlich isotherm equation constants and correlation coefficients for Pb(II) and Cu(II) sorption on magnetic porous ferrospinel MnFe2O4. Isotherm

Parameters

Pb(II)

Cu(II)

Langmuir

qm (lmol g1) KL (L lmol1) R2 KF (L g1) n R2

333.3 0.1744 0.9998 25.32 4.078 0.8200

952.4 0.00945 0.9986 10.88 2.719 0.9070

Freudlich

t 1 t ¼ þ qt k2 q2e qe

ð6Þ

where qt (lmol g1) is the same as in Eq. (1), qe (lmol g1) is the equilibrium adsorption capacity, k1 (min1) and k2 (g lmol min1) are the rate constants of pseudo-first-order and pseudo-second-order kinetic equations, respectively, and k2 q2e (lmol g1 min1) can be regarded as the initial adsorption rate as t ? 0. As shown in Fig. 2, the adsorption of Pb(II) and Cu(II) on MnFe2O4 was quite quick in the first 30 min, and 3.0 h was enough to achieve the equilibrium under our experimental condition. As shown in Table 1, the pseudo-first-order equation might not be sufficient to describe the adsorption process with quite low correlation coefficients. Whereas the pseudo-second-order equation, based on the assumption that the rate limiting step might be chemical adsorption between sorbent and sorbate, showed well fit to the experimental data with higher squared correlation coefficients. The initial adsorption rates (k2 q2e ) were 526.3 lmol g1 min1 for Pb(II) and 2631.5 lmol g1 min1 for Cu(II), indicating more rapid uptake of Cu(II) than that of Pb(II) on the surface of MnFe2O4 before coverage being appreciable. 3.3. Adsorption isotherms In this work, the equilibrium data for Pb(II) and Cu(II) adsorption on MnFe2O4 were modeled using the Langmuir equation (Eq. (7)) and Freundlich equation (Eq. (8)):

Ce Ce 1 ¼ þ qe qm qm K L log qe ¼

1 log C e þ log K F n

ð7Þ

ð8Þ

where Ce (lmol L1) and qe (lmol g1) are the same as in Eqs. (1) and (5), qm (lmol g1) and KL (L lmol1) are Langmuir constants related to the maximum adsorption capacity and energy of adsorption, respectively, and KF (L g1) and n are the isotherm constants calculated from the intercepts and slopes of the Freundlich plots of log qe against log Ce. The results drawn from Fig. 3 and Table 2 indicated that Langmuir model had perfect application for Pb(II) and Cu(II) sorption with a regression coefficient (R2) of 0.9998 and 0.9986, respectively. In other words, this adsorption process acted as monolayer adsorption on the surface of MnFe2O4 under the applied experimental condition. The maximum adsorption capacity (qm) from the Langmuir isotherm of Cu(II) (qmCu(II) = 952.4 lmol g1) on MnFe2O4 was 2.9 times than that of Pb(II) (qmPb(II) = 333.3 lmol g1). MnFe2O4 had a strong magnetism and could be fixed from the solution easily, which made it convenient to be separated and reclaimed after adsorption. A comparison of the maximum capacity of Pb(II) or Cu(II) on MnFe2O4 with those of other sorbents in the literatures was given in Table 3. It could be obviously seen that MnFe2O4 synthesized in our work had the highest adsorption capacity among these adsorbents.

3.4. Adsorption mechanism 3.4.1. SEM and EDX analyses As shown in Fig. 4, SEM and the corresponding EDX spectra of MnFe2O4 before and after adsorption were taken in order to investigate the changes that occurred during the sorption. In Fig. 4a1, it

418

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421

Table 3 Comparison of Pb(II) and Cu(II) adsorption on magnetic porous ferrospinel MnFe2O4 with other absorbents. Pb(II)

Cu(II) 1

Adsorbent

q (lmol g

Oxidized MWCNTs Goethite Manganese oxide–coated sand Iron oxide nanoparticles Al2O3 MnFe2O4

9.940 55.02 6.470 173.8 84.46 333.3

)

References

Adsorbent

q (lmol g1)

References

[30] [31] [32] [20] [33] This work

Hydroxyapatite nanoparticles Expanded perlite Monodisperse chitosan-bound Fe3O4 nanoparticles Functionalized SBA-16 mesoporous silica Acid-activated palygorskite MnFe2O4

478.8 135.8 338.6 572.9 507.7 952.4

[34] [35] [36] [37] [38] This work

(a)

20µm

(b)

20µm

20µm

b1

20µm c1

20µm b3

b2

Pb

O

20µm b4

20µm b5

Cu

O

Mn

Fe

20µm

20µm

c2

20µm a4

Fe

Mn

20µm

O

a3

a2

a1

20µm

(c)

Fe

Mn

20µm c4

c3

20µm c5

Fig. 4. SEM and EDX of (a) MnFe2O4, (b) MnFe2O4–Pb, and (c) MnFe2O4–Cu.

3.4.2. FTIR analysis The FTIR spectra of MnFe2O4, MnFe2O4–Pb, and MnFe2O4–Cu samples are shown in Fig. 5. The band at 1407 cm1 was corresponding to the characteristic stretching frequencies of the carboxyl [23], which might be a result of residual egg white remained on MnFe2O4 surface during a preparation process. However, such bands obviously shifted to a lower frequency approximately at 1388 and 1395 cm1 for MnFe2O4–Pb and MnFe2O4–Cu, respectively. The broad absorption peaks around 1123 and 992 cm1 represented the bonded hydroxyl groups on the metal of the oxide surface [8]; nevertheless, two peaks

MnFe2O4

585 1407

Transmittence

was clearly found that primordial MnFe2O4 had widely distributed pore size from 10 to 20 lm, which might give significant contributions in the adsorption process. There was no obvious change in MnFe2O4 morphology after adsorption (Fig. 4b1 and c1). The existence of manganese, ferrum, and oxygen of original MnFe2O4 was revealed in Fig. 4a2–a4. While after adsorption, the image of lead (Fig. 4b5) and copper (Fig. 4c5) was detected in addition, it further confirmed the adsorption of Pb(II) and Cu(II) ions on ferrospinel MnFe2O4 surface.

992

552

1123 MnFe2O4-Pb

593

1388

982

1105 MnFe2O4-Cu

1395

545

591

985 545

1108

4000 3500 3000 2500 2000 1500 1000 -1 Frenquency Wavenumber (cm )

500

Fig. 5. FTIR spectra of MnFe2O4, MnFe2O4–Pb, and MnFe2O4–Cu.

evidently changed to 1105 and 982 cm1 for MnFe2O4–Pb while to 1108 and 985 cm1 for MnFe2O4–Cu. The band at 585 cm1 could be assigned to the presence of Fe–O bond [19], but it moved

419

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421

to 593 cm1 after Pb(II) adsorption and to 591 cm1 after Cu(II) adsorption. A characteristic peak at 552 cm1 attributed to Mn–O bond [24], and it had a distinct variety to 545 cm1 for both MnFe2O4–Pb and MnFe2O4–Cu. All shifts could confirm the participation of carboxyl groups, hydroxyl groups, Fe–O groups, and Mn–O groups in the adsorption process.

26.24% to 24.26% and 22.5% for MnFe2O4–Pb and MnFe2O4–Cu, respectively. The prominent decrease in S–OH and –COOH groups suggested their sure participation in the adsorption process. As well as part disappearance of S–OH and –COOH might attribute to the formation of complexes with the metal ions on MnFe2O4 surface. That is, the binding metal ions substituted the hydrogen atoms in hydroxy and carboxyl groups bonded on MnFe2O4, forming new complex with oxygen in the homologous functional groups during the sorption process. Furthermore, it was noticeable that the variation of S–O for MnFe2O4–Cu was more obvious than that of MnFe2O4–Pb, which could explain the former result that the maximum adsorption capacity of Cu(II) is higher than that of Pb(II). As shown in Fig. 6b, two peaks appeared in the Pb4f spectrum. The Pb4f7/2 peak at 138.5 eV was ascribed to (MnFe2O4–O)2Pb formed between Pb(II) and the hydroxyl groups bonded on MnFe2O4 surface. With a higher binding energy, the peak at 143.2 eV was attributed to (MnFe2O4–COO)2Pb, which was formed through the reaction of Pb(II) and the carboxyl groups [26]. For MnFe2O4–Cu sample (Fig. 6c), one peak was observed at 934.7 eV in the Cu2p spectrum. The broad peak could be assigned to two peaks at binding energies of 933.6 and 935.9 eV, respectively.

3.4.3. XPS analysis XPS analysis had also been performed to gain further mechanism of Pb(II) and Cu(II) adsorption on MnFe2O4. High-resolution spectra of O1s, Pb4f, Cu2p, Mn2p, and Fe2p regions are shown in Fig. 6. As shown in Fig. 6a, O1s spectrum was divided into three peaks positioned at 530.1, 531.4, and 532.4 eV, which could be assigned to metal oxide bond (S–O) (S:Mn or Fe), hydroxyl bonded on metal (S–OH), and carboxyl (–COOH), respectively [25]. After adsorption, the contents of metal oxide (S–O) increased from 37.29% to 42.02% and 43.0% for MnFe2O4–Pb and MnFe2O4–Cu, respectively, which might be owing to the formation of new metal oxide bond such as Pb–O or Cu–O bonds. In contrast, the relative contents of S–OH dropped from 36.47% to 33.71% and 34.5% after Pb (II) and Cu (II) sorption, respectively. Changes also happened to the carboxyl (–COOH) groups with a lessening of area ratio from (a) O 1s

(b) Pb4f

(c) Cu2p

(MnFe 2O 4-O)2Pb

S-O(37.29%) S-OH(36.47%)

(MnFe2O4-O)2Cu

(MnFe 2O4-COO)2Pb

(MnFe 2O4-COO )2Cu

-COOH(26.24%)

MnFe2O4

135 S-O(42.02%)

140

Mn2p3/2 Mn 2p1/2

(d) Mn2p

S-OH(33.71%) Intensity (cps)

930 933 936 939

145

MnFe2O4-Cu

-COOH(24.26%)

MnFe O -Pb 2 4

MnFe2O4-Pb MnFe2O4 615 S-O(43.0%) S-OH(34.5%) -COOH(22.5%)

630

645

660

675

Fe2p3/2 Fe2p 1/2

(e) Fe2p MnFe2O4-Cu

MnFe O -Cu 2 4

MnFe2O4-Pb MnFe2O4 528

531

534

690

705

720

735

750

Binding energy (eV) Fig. 6. (a) O1s XPS spectra of MnFe2O4, MnFe2O4–Pb, and MnFe2O4–Cu, (b) Pb4f XPS spectra of MnFe2O4–Pb, (c) Cu2p XPS spectra of MnFe2O4–Cu, (d) Mn2p XPS spectra of MnFe2O4, MnFe2O4–Pb, and MnFe2O4–Cu, (e) Fe2p XPS spectra of MnFe2O4, MnFe2O4–Pb, and MnFe2O4–Cu.

420

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421

4. Conclusion Pb(II) Cu(II)

Removal efficiency/%

100

95

90

85 1

2

3

4

5

C ycle numbers Fig. 7. Removal efficiency of Pb(II) and Cu(II) after five adsorption–desorption cycles for MnFe2O4.

The present study indicates that magnetic porous ferrospinel MnFe2O4 is an effective adsorbent for the removal of Pb(II) and Cu(II) ions from aqueous solution, which can be separated or fixed by an additional magnetic field easily, and it may have a promising application in metal ion wastewater treatment. The optimal adsorption performance is obtained at about pH 6.0. The adsorption follows pseudo-second-order kinetics with initial sorption rates were 526.3 and 2631.5 lmol g1 min1 for Pb(II) and Cu(II) sorption. The equilibrium data can be well fitted with Langmuir adsorption isotherm with maximum adsorption capacities of 333.3 and 952.4 lmol g1 for Pb(II) and Cu(II) sorption. The adsorption mechanism is related to the carboxyl and hydroxyl groups binding to the metal surface. The adsorbed metal ions displace the hydrogen atoms and form new complex with oxygen in the homologous functional groups. The sorbent could be reused for five times effectively desorbed by 0.2 M HCl.

Acknowledgments The peak at 933.6 eV corresponds to (MnFe2O4–O)2Cu, and another peak at 935.9 eV could be assigned to (MnFe2O4–COO)2Cu [26]. The Mn2p and Fe2p spectra are shown in Fig. 6d–e. For MnFe2O4 adsorbent, the peak of Mn2p3/2 and Mn2p1/2 centered at 641.3 and 654.1 eV [27], respectively. A slight shift in Mn2p1/2 line toward lower binding energies could be noted in case of both MnFe2O4–Pb and MnFe2O4–Cu samples, suggesting the possibility that the hydroxyl and carboxyl groups bonded to manganese were involved in Pb(II) and Cu(II) sorption, and Mn–O groups at lower binding energies were formed. As Fig. 6e illustrated Fe2p spectra before and after sorption, the binding energies of 711.0 and 724.8 eV for Fe2p could be attributed to Fe2p3/2 and Fe2p1/2 [27], respectively. A slight shift of Fe2p3/2 peak could also be observed in both samples after adsorption, indicating that the functional groups bonded to ferrum also participated in the adsorption. This was consistent with the FTIR results perfectly. The possible complexation reactions in the adsorption process are the following:

2MnFe2 O4  COOH þ M2þ ! ðMnFe2 O4  COOÞ2 M þ 2Hþ 2MnFe2 O4  OH þ M2þ ! ðMnFe2 O4  OÞ2 M þ 2Hþ

ð9Þ ð10Þ

3.5. Regeneration test The consecutive regeneration for MnFe2O4 at eluent solution of 0.2 M HCl is shown in Fig. 7. The removal efficiency only decreased from 96.7% to 95.2% for Pb(II) sorption and from 97.8% to 95.7% for Cu(II) sorption in the third cycle; however, it increased again in the last two cycles, and it could be found that the efficiency remained as high as 99% after five cycles. This might be due to the fact that the large bulks break into small ones, leading more active sites exposed during the repeated cycles. Desorption by HCl solution might occur easily as there were abundant H+ ions in the solution, existing a dominant protonation between H+ ions and metal ions on the active sites at low pH. The complexation between the sorbent and metal ions was destroyed, and Pb(II) and Cu(II) ions could be desorbed from MnFe2O4 [28]. Then, the main possible desorption reaction could be summarized as [29]:

ðMnFe2 O4  COOÞ2 M þ 2H ! 2MnFe2 O4  COOH þ M2þ

ð11Þ

ðMnFe2 O4  OÞ2 M þ 2Hþ ! 2MnFe2 O4  OH þ M2þ

ð12Þ

We appreciate the financial support of the National Natural Science Foundation of China (Nos. 51108111, 51178134), China Postdoctoral Science Especial Fund (No. 200902396), and Fundamental Research funds for the Central Universities (HEUCF 101015).

References [1] L. Wang, J. Zhang, R. Zhao, Y. Li, C. Li, C.L. Zhang, Bioresour. Technol. 101 (2010) 5808–5814. [2] M.W. Wan, C.C. Kan, B.D. Rogel, M.L.P. Dalida, Carbohyd. Polym. 80 (2010) 891–899. [3] T.H. Shi, S.G. Jia, Y. Chen, Y.H. Wen, C.M. Du, H.L. Guo, Z.C. Wang, J. Hazard. Mater. 169 (2009) 838–846. [4] L.J. Dong, Z.L. Zhu, Y.L. Qiu, J.F. Zhao, Chem. Eng. J. 165 (2010) 827–834. [5] R.X. Mu, Z.Y. Xu, L.Y. Li, Y. Shao, H.Q. Wan, S.R. Zheng, J. Hazard. Mater. 176 (2010) 495–502. [6] E. Eren, J. Hazard. Mater. 159 (2008) 235–244. [7] L.J. Dong, Z.L. Zhu, H.M. Ma, Y.L. Qiu, J.F. Zhao, J. Environ. Sci. 22 (2) (2010) 225– 229. [8] E. Álvarez–Ayuso, A. García–Sánchez, X. querol, J. Hazard. Mater. 142 (2007) 191–198. [9] J.H. Wang, S.R. Zheng, Y. Shao, J.L. Liu, Z.Y. Xu, D.Q. Zhu, J. Colloid. Interface. Sci. 349 (2010) 293–299. [10] N.N. Mallikarjuna, A. Venkataraman, Talanta 60 (2003) 139–147. [11] Y. Zhang, M. Yang, X.M. Dou, H. He, D.S. Wang, Environ. Sci. Technol. 39 (2005) 7246–7253. [12] Y.I. Kim, D. Kim, C.S. Lee, Physica B 337 (2003) 42–51. [13] D.S. Mathew, R.S. Juang, Chem. Eng. J. 129 (2007) 51–65. [14] C. Liu, B.S. Zou, A.J. Rondinone, Z.J. Zhang, J. Phys. Chem. B 104 (6) (2000) 1141–1145. [15] J. Hu, I.M.C. Lo, G.H. Chen, Langmuir 21 (2005) 11173–11179. [16] J.G. Parsons, M.L. Lopez, Microchem. J. 91 (2009) 100–106. [17] X.Y. Hou, J. Feng, Y.M. Ren, Z.J. Fan, M.L. Zhang, Colloids Surf. A. 363 (2010) 1–7. [18] S. Maensiri, C. Masingboon, B. Boonchomb, S. Seraphin, Scripta Mater. 56 (2007) 797–800. [19] E. McCafferty, Electrochim. Acta 55 (2010) 1630–1637. [20] N.N. Nassar, J. Hazard. Mater. 184 (2010) 538–546. [21] N. Li, L.D. Zhang, Y.Z. Chen, Y. Tian, H.M. Wang, J. Hazard. Mater. 189 (2011) 265–272. [22] A. Naeem, J.R. Woertz, J.B. Fein, Environ. Sci. Technol. 40 (2006) 5724–5729. [23] C.R. Vestal, Z.J. Zhang, J. Am. Chem. Soc. 125 (2003) 9828–9833. [24] E. Eren, B. Afsin, Y. Onal, J. Hazard. Mater. 161 (2009) 677–685. [25] Z.J. Li, S.B. Deng, G. Yu, J. Huang, V.C. Lim, Chem. Eng. J. 161 (2010) 106–113. [26] H.J. Wang, A.L. Zhou, F. Peng, H. Yu, J. Yang, J. Colloid. Interface Sci. 316 (2007) 277–283. [27] G. Leem, S.S. Zhang, A.C. Jamison, E. Galstyan, I. Rusakova, B. Lorenz, D. Litvinov, T.R. Lee, Appl. Mater. Inter. 2 (2010) 2789–2796. [28] M.R. Huang, H.J. Lu, X.G. Li, J. Colloid. Interface Sci. 313 (2007) 72–79. [29] S. Debnath, D. Nandi, U.C. Ghosh, J. Chem. Eng. Data. 56 (2011) 3021–3028. [30] D. Xu, X.L. Tan, C.L. Chen, X.K. Wang, J. Hazard. Mater. 154 (2008) 407–416. [31] Z.G. Wu, Z.M. Gu, X.R. Wang, L. Evans, H.Y. Guo, Environ. Pollut. 121 (2003) 469–475. [32] R.P. Han, Z. Lu, W.H. Zou, D.T. Wang, J. Shi, J.J. Yan, J. Hazard. Mater. B 137 (2006) 480–488. [33] J. Yin, Z.C. Jiang, G. Chang, B. Hu, Anal. Chim. Acta. 540 (2005) 333–339.

Y. Ren et al. / Journal of Colloid and Interface Science 367 (2012) 415–421 [34] Y.J. Wang, J.H. Chen, Y.X. Cui, S.Q. Wang, D.M. Zhou, J. Hazard. Mater. 162 (2009) 1135–1140. [35] A. Sarı, M. Tuzen, D. Cıtak, M. Soylak, J. Hazard. Mater. 148 (2007) 387–394.

[36] Y.C. Chang, D.H. Chen, J. Colloid. Interface Sci. 283 (2005) 446–451. [37] X.M. Xue, F.T. Li, Microporous Mesoporous Mater. 116 (2008) 116–122. [38] H. Chen, Y.G. Zhao, A.Q. Wang, J. Hazard. Mater. 149 (2007) 346–354.

421