Polyethylenimine modified magnetic graphene oxide nanocomposites for Cu²⁺ removal

Abstract

Fe₃O₄ nanoparticles were synthesized on graphene oxide (GO), then mixed with polyethylenimine to obtain GO/Fe₃O₄/PEI nanocomposites. Due to the high surface area of GO, superparamagnetism of Fe₃O₄, and excellent complex ability of PEI, the nanocomposites showed extremely high Cu²⁺ removal efficiency. The Cu²⁺ removal capacity was 157 mg g⁻¹, which is higher than most reported results. The adsorption kinetics could be best described by a pseudo-second-order model. Moreover, GO/Fe₃O₄/PEI nanocomposites could be easily recycled by magnetic separation. After five cycles, the removal efficiency remained 84%.

---

1. Introduction

The increasing worldwide heavy metal contamination of freshwater systems has recently become one of the key environmental problems to human society.^1,2 Heavy metals are highly toxic even at very low concentration and can accumulate in living organisms, causing severe disorders and diseases.³ For example, copper is extensively used in many industrial processes, and it is also a necessary element for human life, but excess copper can cause haemolytic anaemia, neurological abnormalities, and even cancer.^4,5 Many techniques have been developed for heavy metal removal from water, including ion exchange,⁶ ultrafiltration,⁷ adsorption,⁸ and electrochemical treatment.⁹ However, those methods have some common problems such as high cost, secondary waste and low removal efficiency. Hence, it is urgent to develop new materials with high removal efficiency and easy separation for wastewater treatment.

Graphene oxide (GO) processes excellent physical and chemical properties, and has been extensively studied.^10–18 It can serve as an ideal adsorbent due to its sufficient quantity of oxygenous functional groups, large specific surface area, and high water solubility. Numbers of GO based composites have been developed as new adsorbents for water purification.^19–25 However, most of GO based composites are difficult to collect from water. GO/magnetic nanoparticles are the composites that can provide a rapid and effective way for recycling by applying an external magnet. Such composites combine the merits of large surface area of GO for adsorption and magnetic property for magnetic separation.^20,23,25

However, GO/magnetic nanocomposites usually suffer from low adsorption capacity, because the magnetic interactions between these nanosheets usually induce aggregation and poor dispersibility, which further reduce the effective specific surface area for adsorption. The functionalization of a chelator “coating” on materials provides an efficient way to enhance the performance of adsorbents in water treatment. For example, Chen and co-workers reported that the adsorption capacity of Cu²⁺ was enhanced (from 17.6 mg g⁻¹ to 38.5 mg g⁻¹) by modification of gum arabic on magnetic adsorbent.²⁶ Liu and co-workers reported an about 6-fold enhancement of removal efficiency (from 15% to 94.5%) by coating chitosan on magnetic nanoparticles.²⁷ These polymers play the role of a chelator coating on absorbents, they contain the functional groups (e.g., amines, imines, and thiols), which show strong affinity to heavy metal ions. The bulky polymers also prevent the aggregation induced by the magnetic nanoparticles. Polyethylenimine (PEI) is one of an effective chelators attributed to its amine-rich structure. The high number amino nitrogens can be protonated and lead to high charge density potential, then dispersibility of adsorbents can be improved. Moreover, cupric ions in particular have been shown to coordinate favorably with PEI.^28,29

In this work, GO/Fe₃O₄ nanoparticles were successfully synthesized by in situ growth of Fe₃O₄ nanoparticles on GO sheets, then PEI was used as chelator to modify the surface of GO/Fe₃O₄. The resulting GO/Fe₃O₄/PEI nanocomposites showed excellent removal capability and fast adsorption rates for Cu²⁺ due to the high surface area of graphene and extraordinary complex ability of PEI. Moreover, GO/Fe₃O₄/PEI could be easily separated due to the superparamagnetism of Fe₃O₄ and they maintained high removal efficiency after multiple recycles.

2. Experimental section

2.1 Materials

Ferric chloride hexahydrate (FeCl₃·6H₂O) and ferrous chloride tetrahydrate (FeCl₂·4H₂O) were purchased from Sinopharm Chemical Reagent Co., Ltd. Polyethyleneimine (PEI, branched, M_w 3–3.5 kDa) was purchased from Wuhan Hong Rui Kang Reagent Co., Ltd. Sodium hydroxide (NaOH) of analytical grade was purchased from Shanghai Chemical Reagent Company. Other reagents were of analytical grade purity and were used as received (Beijing Chemicals Co.). Deionized water was used throughout the experiments.

2.2 Preparation of GO, GO/Fe₃O₄, and GO/Fe₃O₄/PEI

GO was synthesized from natural graphite powder by the reported methods.^30,31 Typically, 10 g of graphite powder and 5 g of NaNO₃ were first mixed and stirred in 230 mL of concentrated sulfuric acid at 0 °C for 1 h. 30 g of KMnO₄ was added gradually to above solution with vigorous stirring. The ice-bath was then removed and the temperature of above suspension was brought to 35 °C, where it was maintained for 30 min. Then 460 mL of distilled water was added slowly into the paste and the temperature was raised to 98 °C. After 1 h, the suspension was further diluted to 1.4 L with warm water and treated with 3% H₂O₂ aqueous solution. Finally, the resulting suspension was filtered and washed with 5% HCl aqueous solution. The GO/Fe₃O₄ was synthesized by adding a 7 mL of solution containing 0.1156 g of FeCl₃·6H₂O and 0.0426 g of FeCl₂·7H₂O into a 52 mL solution containing 0.1 g of GO. The mixture was heated to 90 °C, and stirred for 15 min; then 30 mL of concentrated ammonia solution was quickly added. After being rapidly stirred for 30 min under nitrogen atmosphere, the GO/Fe₃O₄ composites were achieved.

GO/Fe₃O₄/PEI was prepared by adding 1.5 g of PEI to the above solution and stirred for additional 30 min. Then, the dark black solution was applied a lab magnet and the attracted solid powder was washed with Milli-Q water/ethanol and finally dispersed in water for further use.

2.3 Adsorption

All batch adsorption experiments were performed on a ZHWY-103B thermostat shaker with a shaking speed of 300 rpm. 0.1 g of GO/Fe₃O₄/PEI was added to 100 mL water with different concentrations of Cu(NO₃)₂. The pH of the aqueous solution was adjusted with 0.1 mol L⁻¹ HCl or 0.1 mol L⁻¹ NaOH solutions and measured by a pH meter (pHS-25). After adsorption equilibration, the adsorbents were separated by a magnet and the supernatant was analyzed by a flame/graphite furnace atomic absorption spectrophotometer (AA-7000). For the kinetic study, the GO/Fe₃O₄/PEI was kept at 1.0 g L⁻¹, and the concentration of Cu(NO₃)₂ was 50 mg L⁻¹.

The adsorption amount was calculated based on the difference of Cu²⁺ concentration in the aqueous solution before and after adsorption, according to the following equation:

Q = (C₀ − C_e)V/W

where C₀ and C_e are the initial and equilibrium concentrations of Cu²⁺, respectively (mg L⁻¹). V is the volume of the Cu²⁺ solution (L), and W is the weight of the GO/Fe₃O₄/PEI used (g).

2.4 Desorption and regeneration

The magnetical adsorbents loaded with Cu²⁺ were dispersed in 0.1 mol L⁻¹ HCl and shaken for 5 h at room temperature. Then, the adsorbents were separated by a magnet, washed with distilled water and 0.1 mol L⁻¹ NaOH repeatedly for 5 times, and dispersed in distilled water for further use. To test the reusability of the adsorbents, the above process was repeated for five times.

2.5 Characterizations

Infrared spectra were recorded on a Nicolet 510P FT-IR spectrometer (America, Nicolet) after the samples were mixed with KBr in agate mortar and pressed to a thin plate. JEOL JEM-2010 transmission electron microscope and Hitachi S-2500 scanning electron microscope were employed to observe the morphology of GO/Fe₃O₄ hybrids. The magnetic property of the GO/Fe₃O₄/PEI at room temperature was measured by a 9 T physical property measurement system (PPMS) by Quantum Design. N₂ adsorption–desorption isotherms were performed at 77 K, using a QuadraSorb SI apparatus utilizing Barrett–Emmett–Teller (BET) calculations for specific surface area.

3. Results and discussion

3.1 Characterization of GO/Fe₃O₄

Fig. 1a shows the synthesis of GO/Fe₃O₄/PEI nanoparticles. SEM and TEM images of GO/Fe₃O₄ are shown in Fig. 1b and c. In SEM images, the Fe₃O₄ nanoparticles (NPs) appear as bright dots, and GO flakes can be observed clearly. The product was also confirmed by TEM; it can be seen that GO are densely covered by narrow size distribution of Fe₃O₄ NPs with an average size about 10 nm. The distribution of Fe₃O₄ NPs on GO sheets is nearly even, no big conglomeration of Fe₃O₄ NPs is observed. The in situ growth of Fe₃O₄ NPs on GO prevents the conglomeration of Fe₃O₄ NPs, even after a long time of sonication during the preparation of the TEM specimen, the Fe₃O₄ NPs were still strongly anchored on the surface of GO with high density. The lattice fringe spacing is 0.25 nm, which agrees well with the lattice spacing of (311) planes of cubic magnetite.¹³

Fig. 1 (a) Synthesis of GO/Fe₃O₄/PEI nanocomposites; (b) SEM image of GO/Fe₃O₄; (c) TEM image of GO/Fe₃O₄ (inset is high-resolution TEM image); (d) room temperature hysteresis loops of GO/Fe₃O₄ and GO/Fe₃O₄/PEI; (e) FTIR spectra of GO, GO/Fe₃O₄ and GO/Fe₃O₄/PEI.

The magnetization property of GO/Fe₃O₄ and GO/Fe₃O₄/PEI composites were investigated at room temperature, the saturation magnetization decreased from 12 emu g⁻¹ to 9 emu g⁻¹ after addition of PEI (shown in Fig. 1d), because the large number of amino groups on PEI leads to positive charged surface, and higher repulsion force occurs between composites.

The FTIR patterns of GO, GO/Fe₃O₄, and GO/Fe₃O₄/PEI are shown in Fig. 1e. The spectrum of GO is in good agreement with previous work.³² The peaks at 1071, 1335, 1636 cm⁻¹ correspond to C–O–C stretching vibrations, C–OH stretching, and C–C stretching mode of the sp² carbon skeletal network, respectively. The peak at 584 cm⁻¹ refers to Fe–O deformation in octahedral and tetrahedral sites.³³ The broad and strong band ranging from 3000 to 3800 cm⁻¹ maybe due to the overlap of OH and NH stretching, while peaks of 2923, 2851, and 1628 cm⁻¹ are assigned to PEI.^34,35

3.2 The effect of pH

The pH of solution plays a unique role on metal-chelate formation and subsequent extraction. The adsorption of metal ions can thus be influenced by pH value;³⁶ protons in acid solution can protonate binding sites of the chelating molecules, while hydroxide in basic solution can precipitate metal ions, thus pH of solution should be optimized firstly. Fig. 2a shows the effect of solution pH on the Cu²⁺ adsorption by the prepared absorbent. The removal efficiency of Cu²⁺ was low at pH < 3 (concentration: Cu²⁺ was 50 mg L⁻¹, adsorbent dose was 1 g L⁻¹); this was attributed to the repulsive forces between the positive charged amino groups and positive metal ions. The adsorption efficiency increased from 54% to 94% with an increase of pH value from 2.0 to 5.0. This is due to the ability of amino group to be protonated was weakened, which caused by the rapid decrease of concentration of H⁺. The absorption efficiency decreased to 84% at pH 6.0, because the concentration of OH⁻ increased, the Cu²⁺ existed in the forms of Cu(OH)⁺ and Cu(OH)₂, which resulting in the decrease of adsorption efficiency.³⁷ Therefore, pH 5.0 was chosen as the optimum pH for further studies.

Fig. 2 (a) Effect of pH value on removal efficiency (Cu²⁺ concentration, 50 mg L⁻¹; adsorbent dose, 1 g L⁻¹; contact time, 10 h; temperature, 298 K). (b) Effect of adsorbent dose on adsorption capacity and removal percentage (pH 5.0; Cu²⁺ concentration, 50 mg L⁻¹; contact time, 10 h; temperature, 298 K).

3.3 The effect of adsorbent dose

The dependence of Cu²⁺ adsorption on adsorbent dose was studied. Different amounts of adsorbent (0.1–2.0 g L⁻¹) were added by keeping the pH (5.0) and temperature (298 K) constant. As shown in Fig. 2b, the removal percentage increased rapidly with the adsorbent dose increasing from 0.1 to 1.0 g L⁻¹, because more binding sites present on adsorbent surface. However, further increased the dose (from 1.0 to 2.0 g L⁻¹) did not exhibit significantly change the removal percentage. The reason may be that the equilibrium was established between Cu²⁺ on adsorbent and in the solution. Therefore, 1.0 g L⁻¹ adsorbent dose was chosen as the optimum adsorbent dose for further study.

3.4 Adsorption kinetics

Kinetics of adsorption was determined by analyzing adsorption capacity at different time intervals, pseudo-first-order kinetic model (eqn (1)) and pseudo-second-order kinetic model (eqn (2)) were used to describe adsorption kinetics, and parameters of the adsorption kinetic models were calculated.

ln(q_m − q_t) = ln q_m − k₁t (1)

where q_t (mg g⁻¹) is the adsorption capacity at time of t, q_m (mg g⁻¹) is the maximum adsorption capacity; k₁ (min⁻¹) and k₂ (g mg⁻¹ min⁻¹) denote the adsorption rate constants, all of which can be calculated according to the slope and intercept of fitting lines. The pseudo-second-order model fits the experimental data much better than pseudo-first-order model (Fig. 3a and Table 1). These results indicate that the overall rate of the Cu²⁺ adsorption process appears to be controlled by the chemical sorption.^38,39

Fig. 3 (a) Effect of time on the adsorption capacity. Insets are the fitting results of the experimental adsorption kinetic data in terms of pseudo-first-order and pseudo-second-order models. (pH 5.0; Cu²⁺ concentration, 50 mg L⁻¹; adsorbent dose, 1.0 g L⁻¹; temperature, 298 K). (b) Equilibrium isotherm for the adsorption of Cu²⁺. Insets are the fitting results of the experimental adsorption isotherm data in terms of Langmuir and Freundlich models (pH 5.0; contact time, 10 h; adsorbent dose, 1 g L⁻¹; temperature, 298 K).

Table 1 Adsorption kinetic parameters of Cu²⁺ onto GO/Fe₃O₄/PEI

Pseudo-first-order			Pseudo-second-order
qm (mg g^−1)	KL (min^−1)	R^2	k2 (g mg^−1 min^−1)	qm (mg g^−1)	R^2
5.83	0.069	0.8785	0.073	47.62	0.9998

---

3.5 Adsorption isotherms

The adsorption isotherms of Cu²⁺ at pH 5.0 were presented in Fig. 3b. The adsorption data were fitted with both Langmuir and Freundlich isotherm models, expressed by eqn (3) and (4), respectively.

Langmuir equation:

where q_e is the equilibrium adsorption capacity of ions on the adsorbent (mg g⁻¹); C_e is the equilibrium ions concentration in solution (mg L⁻¹); q_m is the maximum capacity of the adsorbent (mg g⁻¹); and K_L is the Langmuir adsorption constant (L mg⁻¹).

Freundlich equation:

where equilibrium capacity (q_e) and C_e are defined as above, K_F is the Freundlich constant (L mg⁻¹), and n is the heterogeneity factor.

The parameters derived by fitting the two isotherm models were presented in Table 2. Among these two isotherms, Langmuir model fitted much better (R² = 0.9962) than Freundlich model (R² = 0.7158), suggesting that Cu²⁺ sorption on absorbent is of a monolayer coverage. This may be attributed to the homogenous distribution of surface amino groups on GO/Fe₃O₄. In addition, the q_m of the adsorbent for Cu²⁺ is 157.48 mg g⁻¹, which is higher than most of the adsorbents reported (Table 3).

Table 2 Adsorption isotherm parameters of Cu²⁺ onto GO/Fe₃O₄/PEI

Langmuir isotherm			Freundlich isotherm
qm (mg g^−1)	KL (g mg^−1)	R^2	n	KL (mg g^−1)	R^2
157.48	0.069	0.9962	2.67	22.053	0.7158

---

Table 3 Comparison of adsorption capacity of various adsorbents for Cu²⁺

Type of adsorbent	Adsorption capacities of Cu^2+ (mg g^−1)	Reference
Amino functionalized Fe3O4@SiO2	30.8	40
Sugar beet pulp	21.3	41
Mesoporous adsorbent	184.2	42
Magnetic chitosan	103.16	39
Fe3O4@APS@AA-co-CA NPs	126.9	38
Magnetic cellulose–chitosan composite microspheres	65.8	34
Layered double hydroxide nanocrystals@carbon nanosphere	19.93	43
GO/Fe3O4	18.26	25
GO/Fe3O4/PEI	157.48	This work

---

3.6 The effect of common coexisting ions

Heavy metal pollutants often exist together with other common ions, such as Ca²⁺, K⁺, Na⁺, NO₃⁻, Cl⁻, HCO₃⁻. Fig. 4a shows the effect of coexisting ions on Cu²⁺ removal was rather insignificant. To investigate whether our absorbent is applicable to natural water and whether the naturally existed substances may interfere the Cu²⁺ removal, the Cu²⁺ adsorption performance of the adsorbent was evaluated by using sea water and Beijiushui (a river in Qingdao city). Piror to testing, the sea and river water were filtered through a 0.22 μm membrane. Similarly, there was no influence from the naturally existed substances in sea water and river water. Thus, the common coexisting ions basically do not affect Cu²⁺ adsorption of Go/Fe₃O₄/PEI, demonstrating the high possibility of practical application.

Fig. 4 (a) Effect of common ions (KNO₃, Ca(NO₃)₂, NaNO₃, NaCl, NaHCO₃, 150 mg L⁻¹) on the adsorption of Cu²⁺ (pH 5.0; Cu²⁺ concentration, 50 mg L⁻¹; adsorbent dose, 1 g L⁻¹; contact time, 10 h; temperature, 298 K). (b) Regeneration studies of GO/Fe₃O₄/PEI with five cycles. The concentration of GO/Fe₃O₄/PEI was 1 g L⁻¹. The initial Cu²⁺ concentration was 50 mg L⁻¹. The contact time was 10 h for each cycle.

3.7 Regeneration of Go/Fe₃O₄/PEI

For practical application, recycling and regeneration of adsorbent is indispensable, therefore, the adsorbents need to have higher adsorption capability as well as good desorption property. The Cu²⁺ adsorbent was treated with NaOH solution and analyzed the concentration of Cu²⁺ desorbed into the solution. The removal efficiency was found to reach 88% after first regeneration, indicating the adsorbed Cu²⁺ could be effectively washed away from the adsorbent. After five cycles of regeneration, the removal efficiency was still as high as 84%, as shown in Fig. 4b. Therefore, the adsorbent can be repeatedly used in Cu²⁺ removal.

3.8 Removal mechanism

The surface area of adsorbent plays an important role in adsorption.^20,24 In this work, the surface areas of GO, GO/Fe₃O₄, and GO/Fe₃O₄/PEI are 363.7, 189.3 and 323.5 m² g⁻¹, respectively, and the adsorption capacities are 89.32, 18.20, and 157.48 mg g⁻¹, respectively, as shown in Fig. 5. It can be seen that the modification of PEI not only improves the dispersibility of GO/Fe₃O₄, but also enhances the adsorption capacity.

Fig. 5 Surface areas and adsorption capacities of GO, GO/Fe₃O₄, and GO/Fe₃O₄/PEI.

PEI is a polymer with long chains, which favors its absorption onto GO and Fe₃O₄.¹⁰ Moreover, there are plenty of amino groups on the chains, which have high affinity to Cu²⁺.^28,29 Thus, PEI plays an important role in this work for Cu²⁺ sorption. To confirm this, different amount of PEI was added to GO/Fe₃O₄ solution, after the attachment of PEI on GO/Fe₃O₄, the GO/Fe₃O₄/PEI was separated from solution, the PEI did not attached in aqueous solution was analyzed by total organic carbon analysis. Fig. 6a shows the attached PEI on GO/Fe₃O₄ increases as the increase of PEI addition, then a plateau reaches, this indicates a saturation of the GO/Fe₃O₄ surface with PEI.⁴⁴ The removal efficiency of Cu²⁺ increases gradually from 18.3% to 94% as the increases of PEI addition and shows the same tendency as PEI attached. The Fe and PEI recoveries after five cycles of regeneration were investigated. Fig. 6b shows almost no loss for Fe, while PEI recovery decreased from 97.1% to 91.9% after five cycles of regeneration, as the decrease tendency of GO/Fe₃O₄/PEI removal efficiency after every time of regeneration. These indicate that it is PEI that plays the dominant role for Cu²⁺ sorption.

Fig. 6 (a) Cu²⁺ removal efficiency and PEI attached of GO/Fe₃O₄ coated with different amounts of PEI (pH 5.0; initial concentration, 50 mg L⁻¹; adsorbent dose, 1 g L⁻¹; contact time, 10 h), (b) Fe and PEI loss after five cycles of regeneration.

Fig. 7 illustrates the mechanism for Cu²⁺ adsorption and desorption. The lone pair electrons on nitrogen can be supplied to the empty atomic orbital of Cu²⁺, which forms amino group–metal complexes form on the surface of adsorbent. However, under acidic condition, H⁺ can cause the protonation of amino groups, which means that part of sites occupied by Cu²⁺ are replaced by H⁺ and Cu²⁺ returns to the solution (desorption). After being stirred in NaOH solution, the adsorption capacity of GO/Fe₃O₄/PEI can be reconditioned. Thus, the magnetic adsorbent keeps high capacity for Cu²⁺ removal in each cycle.

Fig. 7 The adsorption mechanism of Cu²⁺ on GO/Fe₃O₄/PEI.

4. Conclusion

In this study, PEI-modified magnetic graphene oxide could significantly increase the adsorption capacity of Cu²⁺ and the maximum adsorption capacity was 157.48 mg g⁻¹ under optimized adsorption conditions. Kinetics study indicated that the adsorption mechanism followed a pseudo-second-order model (R² > 0.9998) and Langmuir model was in good agreement with the adsorption isotherm study (R² > 0.9962). GO/Fe₃O₄/PEI nanocomposites could be easily separated from water by external magnetic field. The adsorbent exhibited very good adsorption performance after regeneration for five times. The results suggest that GO/Fe₃O₄/PEI can serve as a good nanoadsorbent for Cu²⁺ removal from wastewater.

Acknowledgements

This work was supported by the NSFC (Grant no. 21301103, 21077062, 61225018, 51272084, and 21105053), the 47th Scientific Research Foundation for the Returned Overseas Chinese Scholars, the Taishan Scholarship, the Shandong Natural Science Foundation (ZR2012FZ007, ZR2012EML08), the Shandong Province High Education Research and Development Program (J13LA08), Qingdao Municipal Science and Technology Commission (12-1-4-3-(27)-jch), NSF (CHE-1338346) and 3M Faculty Award.

References

1. M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis, B. J. Marinas and A. M. Mayes, Nature, 2008, 452, 301–310.
2. Y.-J. An, Y.-M. Kim, T.-I. Kwon and S.-W. Jeong, Sci. Total Environ., 2004, 326, 85–93.
3. L. Zhou, Y. Wang, Z. Liu and Q. Huang, J. Hazard. Mater., 2009, 161, 995–1002.
4. T. S. Anirudhan and S. Rijith, Colloids Surf., A, 2009, 351, 52–59.
5. Y. S. Ho, C. T. Huang and H. W. Huang, Process Biochem., 2002, 37, 1421–1430.
6. A. Shafaei, F. Z. Ashtiani and T. Kaghazchi, Chem. Eng. J., 2007, 133, 311–316.
7. E. Samper, M. Rodríguez, M. A. De la Rubia and D. Prats, Sep. Purif. Technol., 2009, 65, 337–342.
8. J. Dai, H. Yan, H. Yang and R. Cheng, Chem. Eng. J., 2010, 165, 240–249.
9. X. Feng, Z. Wu and X. Chen, Sep. Purif. Technol., 2007, 57, 257–263.
10. V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril and K. S. Kim, Chem. Rev., 2012, 112, 6156–6214.
11. H. Jang, Y.-K. Kim, H.-M. Kwon, W.-S. Yeo, D.-E. Kim and D.-H. Min, Angew. Chem., Int. Ed., 2010, 49, 5703–5707.
12. X. Sun, Z. Liu, K. Welsher, J. Robinson, A. Goodwin, S. Zaric and H. Dai, Nano Res., 2008, 1, 203–212.
13. H. He and C. Gao, ACS Appl. Mater. Interfaces, 2010, 2, 3201–3210.
14. J. Bian, M. Xiao, S. J. Wang, Y. X. Lu and Y. Z. Meng, Catal. Commun., 2009, 10, 1529–1533.
15. W. Li, F. Wang, S. Feng, J. Wang, Z. Sun, B. Li, Y. Li, J. Yang, A. A. Elzatahry, Y. Xia and D. Zhao, J. Am. Chem. Soc., 2013, 135, 18300–18303.
16. Y. H. Lee, L. Polavarapu, N. Gao, P. Yuan and Q.-H. Xu, Langmuir, 2011, 28, 321–326.
17. Y. Bu, Z. Chen, W. Li and B. Hou, ACS Appl. Mater. Interfaces, 2013, 5, 12361–12368.
18. J. Jing, Y. Zhang, W. Li and W. W. Yu, J. Catal., 2014, 316, 174–181.
19. V. Chandra, J. Park, Y. Chun, J. W. Lee, I.-C. Hwang and K. S. Kim, ACS Nano, 2010, 4, 3979–3986.
20. L. Fan, C. Luo, M. Sun and H. Qiu, J. Mater. Chem., 2012, 22, 24577–24583.
21. C. Gao, X.-Y. Yu, R.-X. Xu, J.-H. Liu and X.-J. Huang, ACS Appl. Mater. Interfaces, 2012, 4, 4672–4682.
22. S. Huang, G.-N. Zhu, C. Zhang, W. W. Tjiu, Y.-Y. Xia and T. Liu, ACS Applied Materials & Interfaces, 2012, 4, 2242–2249.
23. H. Jabeen, V. Chandra, S. Jung, J. W. Lee, K. S. Kim and S. B. Kim, Nanoscale, 2011, 3, 3583–3585.
24. X.-L. Wu, L. Wang, C.-L. Chen, A.-W. Xu and X.-K. Wang, J. Mater. Chem., 2011, 21, 17353–17359.
25. J. Li, S. Zhang, C. Chen, G. Zhao, X. Yang, J. Li and X. Wang, ACS Applied Materials & Interfaces, 2012, 4, 4991–5000.
26. S. S. Banerjee and D.-H. Chen, J. Hazard. Mater., 2007, 147, 792–799.
27. X. Liu, Q. Hu, Z. Fang, X. Zhang and B. Zhang, Langmuir, 2008, 25, 3–8.
28. W. Maketon, C. Z. Zenner and K. L. Ogden, Environ. Sci. Technol., 2008, 42, 2124–2129.
29. L. Steinmann, J. Porath, P. Hashemi and Å. Olin, Talanta, 1994, 41, 1707–1713.
30. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
31. Y. Xu, H. Bai, G. Lu, C. Li and G. Shi, J. Am. Chem. Soc., 2008, 130, 5856–5857.
32. L. Ai, C. Zhang and Z. Chen, J. Hazard. Mater., 2011, 192, 1515–1524.
33. L. Wang and L. Gao, J. Phys. Chem. C, 2009, 113, 15914–15920.
34. S. Peng, H. Meng, Y. Ouyang and J. Chang, Ind. Eng. Chem. Res., 2014, 53, 2106–2113.
35. G. Xie, P. Xi, H. Liu, F. Chen, L. Huang, Y. Shi, F. Hou, Z. Zeng, C. Shao and J. Wang, J. Mater. Chem., 2012, 22, 1033–1039.
36. I. Ali, Chem. Rev., 2012, 112, 5073–5091.
37. G. Yuvaraja, M. V. Subbaiah and A. Krishnaiah, Ind. Eng. Chem. Res., 2012, 51, 11218–11225.
38. F. Ge, M.-M. Li, H. Ye and B.-X. Zhao, J. Hazard. Mater., 2012, 211–212, 366–372.
39. M. Monier, D. M. Ayad, Y. Wei and A. A. Sarhan, J. Hazard. Mater., 2010, 177, 962–970.
40. J. Wang, S. Zheng, Y. Shao, J. Liu, Z. Xu and D. Zhu, J. Colloid Interface Sci., 2010, 349, 293–299.
41. Z. Reddad, C. Gerente, Y. Andres and P. Le Cloirec, Environ. Sci. Technol., 2002, 36, 2067–2073.
42. M. R. Awual, I. M. M. Rahman, T. Yaita, M. A. Khaleque and M. Ferdows, Chem. Eng. J., 2014, 236, 100–109.
43. J. Gong, T. Liu, X. Wang, X. Hu and L. Zhang, Environ. Sci. Technol., 2011, 45, 6181–6187.
44. I. Y. Goon, C. Zhang, M. Lim, J. J. Gooding and R. Amal, Langmuir, 2010, 26, 12247–12252.

---