Acrylic acid grafted and acrylic acid/sodium humate grafted bamboo cellulose nanofibers for Cu²⁺ adsorption

Abstract

Bamboo cellulose nanofibers-graft-poly (acrylic acid) (BCN-g-PAA) and bamboo cellulose nanofibers-graft-poly (acrylic acid)/sodium humate (BCN-g-PAA/SH) were synthesized for the first time and sequentially utilized as biosorbents for removal of Cu²⁺ from aqueous solutions. The chemical structure and morphology of both modified nanofibers were characterized by Fourier transform infrared spectroscopy and scanning electron microscopy, respectively. Batch adsorption experiments were conducted to elucidate their adsorption behaviors on Cu²⁺. The influencing factors, such as pH, contact time and initial Cu²⁺ concentration, on Cu²⁺ adsorption were investigated in detail. It was discovered that pH strongly influenced the Cu²⁺ adsorption. When pH increased from 2.0 to 4.5, the adsorption capacities of both modified nanofibers were improved significantly. Adsorption isotherm studies indicated that the Cu²⁺ adsorption could be described well by the Freundlich equation. Meanwhile, their adsorption kinetics was more likely to follow the pseudo-second-order model. These nanocellulose-based adsorbents exhibited very fast adsorption rates. The calculated adsorption capacities at equilibrium (q^cal_e) for BCN-g-PAA and BCN-g-PAA/SH were 0.727 and 0.709 mmol g⁻¹, significantly higher than that of BCN (0.286 mmol g⁻¹). Adsorption/desorption cycling tests suggested that the introduced SH segments allowed for improved reusability of BCN-g-PAA/SH.

---

1 Introduction

Copper is well known as one of the most common heavy metals, generated from industries like electroplating, iron and steel production, mining, the non-ferrous metal industry and metalworking.^1–3 However, improper disposal of copper waste without efficient treatment may cause serious environmental pollution.⁴ The high concentration of Cu²⁺ in the environment would cause disorders and various diseases such as insomnia, liver damage, hemolytic anemia and corneal opacity.^5–8 Therefore, it is essential to control the concentration of Cu²⁺ in drinking water as well as, from the economic point of view, recover as much copper as possible.⁹

Many adsorbent materials, including hydrous manganese dioxide,¹⁰ anatase titanium dioxide,¹¹ mesoporous silica,¹² kraft lignin¹³ and modified starch,¹⁴ have been developed to address Cu²⁺ pollution issues. Among these adsorbents, nanomaterials have received considerable attention due to the large specific surface area and excellent adsorbent capacities.^15,16 In recent years, there is growing interest to produce adsorbent materials based on biopolymers, not only because of the low cost of the raw materials, but also because of their less negative post-service impact on the environment since they are biodegradable in nature. Cellulose is the most abundant biopolymer on earth. It bears plenty of hydroxyls, which can be easily converted into different functionalities with high metal ion affinity. Chemically modified cellulose fibers for the adsorption of various heavy metals have been well investigated.¹⁷ However, the development of biosorbents from nanosized cellulose fibers is a relatively new research subject. Nanofibers offer a very large specific surface area which allows substantial improvement of adsorption capability, since the adsorption phenomenon mainly happens at materials' surface.¹⁸ Recently, cellulose nanofibers have been electrospun and chemically modified for heavy metals adsorption.¹⁹ However, as cellulose cannot be dissolved in common solvents due to its high crystallinity and complex hydrogen bonding, the cellulose nanofibers were regenerated from electrospun cellulose acetate through deacetylation. This process is not economical and environmentally friendly owing to the multiple preparation steps with toxic organic solvents, which produce additional environmental problem during the evaporation of solvent. In fact, cellulose nanofibers readily exist in plant. They are often embedded in the cell walls of plants with matrix substances such as hemicellulose and lignin. They can be isolated from plant fibers by simple mechanical shearing processes.^20–22

The objective of this study is to produce highly efficient biosorbents from bamboo cellulose nanofibers (BCN) and analyze their Cu²⁺ adsorption behaviors. Bamboo is an abundant natural resource in Asia and South America, because it takes only several months to grow up. In addition, it contains a great deal of cellulose with many distinguishing features, such as high flexibility, high strength, low density.^23,24 Hereby, BCN were extracted from bamboo pulp fibers by high shear homogenization coupled with ultrasonication. Thereafter, acrylic acid and sodium humate (SH) were grafted onto BCN through free radical polymerization to produce BCN-graft-poly (acrylic acid) (BCN-g-PAA) and BCN-graft-poly (acrylic acid)/sodium humate (BCN-g-PAA/SH) for Cu²⁺ adsorption from its aqueous solution. SH is composed of multifunctional aliphatic components and aromatic constituents, containing a large number of functional groups, such as carboxylates and phenolic hydroxyls.²⁵ Previous studies have indicated that the presence of SH favors the adsorption of heavy metals, promoting the adsorption capacities, improving the reusability of the adsorbent materials.²⁶ The influencing parameters on Cu²⁺ adsorption, such as pH, contact time and initial Cu²⁺ concentration were comprehensively investigated. Besides, the reusability of these biosorbents was also evaluated.

2 Materials and method

2.1 Materials

Never-dried moso bamboo pulp was supplied by Yongfeng Paper Co., Ltd, Muchuan, China. Acrylic acid (AR, Kelong Chemical Reagent Factory, Chengdu, China) was purified by distillation under reduced pressure prior to use. Chemically pure sodium humate was produced by Tianjin Guangfu Institute of Fine Chemicals. Other reagents, including ammonium ceric nitrate (AR), HNO₃ (AR), H₂SO₄ (AR), NaOH (AR), CuSO₄·5H₂O (AR) and ammonia water (AR), were all purchased from Kelong Chemical Reagent Factory, Chengdu, China. Deionized water was used throughout the experiment.

2.2 Preparation of BCN

Bamboo pulp fibers aqueous suspension (0.22 wt%, 700 mL) was treated by a high shear homogenizer (T18, IKA, Germany) at a rotation speed of 22 000 rpm for 60 min. After that, the suspension was further treated ultrasonically to obtain BCN. The ultrasonication was conducted using a horn type ultrasonic generator (JY99-IIDN, Scientz, China) at 1100 W for 60 min.

2.3 Graft copolymerization for BCN-g-PAA and BCN-g-PAA/SH

For BCN-g-PAA, 50 g aqueous suspension containing 0.5 g BCN was well dispersed under constant stirring. This suspension was then added in a three-necked flask equipped with water bath, a magnetic stirrer, a condenser, a nitrogen line and a thermometer. The system was kept at 45 °C for 30 min under nitrogen atmosphere. Afterwards, 3.975 g acrylic acid, 0.145 g ceric ammonium nitrate and 1.325 g HNO₃ were added into the three-necked flask. The reaction lasted for 120 min. The product was vacuum-filtrated and then Soxhlet-extracted with deionized water to remove any unreacted chemicals and homopolymer of PAA. Finally, it was subjected to lyophilization using a freeze dryer (FD-1A-50, Boyikang, Beijing). BCN-g-PAA/SH was prepared in the same manner except for an addition of 2.5 g SH.²⁷

2.4 Scanning electron microscopy (SEM)

The suspensions of the bamboo cellulose fibers before and after nanofibrillation and the modified BCN were air-dried to form sheets. The thin sheets were coated with gold using a vacuum sputter coater and then observed with SEM (Inspect F 50, FEI, USA) at 20 kV. All the samples were observed under the same conditions.

2.5 Fourier transform infrared spectroscopy (FTIR)

The FTIR spectra of nanofibers with and without modification were recorded on a Fourier transform infrared spectrometer (Nicolet 6700, USA). A small quantity (2 mg) of sample was blended with KBr powder (300 mg) and compressed to form a testing pellet. The spectrum for each sample was recorded as an average of 100 scans in the range from 4000 to 400 cm⁻¹ with a resolution of 2 cm⁻¹. All the samples and KBr were oven-dried before testing.

2.6 Cu²⁺ adsorption studies

2.6.1 Adsorption procedures. Adsorption experiments were carried out by adding 40 mg BCN-g-PAA or BCN-g-PAA/SH to 20 mL CuSO₄ aqueous solution, which was made by dissolving a certain amount of CuSO₄·5H₂O in deionized water. During the adsorption, the solutions were constantly stirred using a shaking table (KS, Aohua, China) at 20 °C/180 rpm. The concentration of Cu²⁺ after adsorption was detected by means of a UV-vis spectrophotometer (UV-1800, Mapada, Shanghai). Ammonia water was added to form a complex with Cu²⁺, which exhibits a characteristic adsorption peak at 608 nm.²⁸ All the samples were tested three times and the average adsorption intensity was used to estimate Cu²⁺ concentrations.

The effects of pH on Cu²⁺ adsorption was studied at an initial Cu²⁺ concentration of 40.0 mmol L⁻¹ for 120 min. The pH was adjusted with 0.1 mol L⁻¹ H₂SO₄ solutions. Kinetics studies were conducted using 20 mL 40.0 mmol L⁻¹ Cu²⁺ solutions to contact with 40 mg adsorbents. At different predetermined time (5–240 min), samples were analyzed for residual Cu²⁺ concentrations. The adsorption isotherm experiments were performed using different Cu²⁺ concentrations in the range 0.2–20.0 mmol L⁻¹.

2.6.2 Analysis of adsorption capacity. The adsorption capacities of BCN-g-PAA and BCN-g-PAA/SH were calculated using the following equation.²⁹

q = (C₂ − C₁)V/m (1)

Where q is the amount (mmol g⁻¹) of Cu²⁺ adsorbed, C₁ and C₂ are the initial and equilibrium Cu²⁺ concentrations (mmol L⁻¹) of Cu²⁺ solutions, respectively; V is the volume (L) of Cu²⁺ solution and m is the mass (g) of BCN-g-PAA or BCN-g-PAA/SH used.

2.7 Experiment of recycling

The desorption of Cu²⁺-loaded nanofibers was carried out using 0.1 mol L⁻¹ H₂SO₄ as the desorbing agent. 40 mg of each adsorbent was contacted with 20 mL 0.1 mol L⁻¹ H₂SO₄ solution in a conical flask, which was placed in the shaking table (180 rpm) for 120 min. Afterwards, the supernatant was discarded and the adsorbents were washed with deionized water for several times. The treated sample was then regenerated with 20 mL 0.1 mol L⁻¹ NaOH for 60 min and washed with deionized water to reach a neutral pH. The regenerated sample was used for another adsorption test.

3 Results and discussion

3.1 SEM of the modified nanofibers

SEM images of the bamboo cellulose fibers before and after nanofibrillation and the chemically modified BCN were shown in Fig. 1. Fig. 1a and b exhibited the morphological development from bamboo pulp fibers to BCN after mechanical nanofibrillation. Significant difference in length and diameter could be observed. Bamboo cellulose fibers were comprised of two different types of cells, fibrous cells and rectangular cells with several tens of microns in diameter. They were regarded mostly as fibers and parenchyma cells, respectively.³⁰ The rectangular cells almost disappeared and continuous nanofibers with diameters well below 100 nm were obtained after nanofibrillation. The specific surface area of fibers was determined by nitrogen adsorption using the Brunauer–Emmett–Teller (BET) method (see ESI†). As expected, BCN had a large specific surface area of 79 m² g⁻¹, much higher than that of precursor bamboo fibers (0.80 m² g⁻¹). Fig. 1c and d showed the morphologies of BCN-g-PAA and BCN-g-PAA/SH, respectively. The surface polymerization did not change the fibril morphology of BCN remarkably. However, the surface texture of BCN became rougher after chemical modification, especially BCN-g-PAA/SH, indicating the occurrence of surface polymerization.

Fig. 1 SEM images of bamboo cellulose fibers before and after mechanical nanofibrillation and surface modified bamboo cellulose nanofibers. (a) Bamboo pulp fibers; (b) BCN; (c) BCN-g-PAA; (d) BCN-g-PAA/SH.

3.2 FTIR studies of the modified nanofibers

The grafting polymerization of BCN-g-PAA and BCN-g-PAA/SH was verified by FTIR spectra shown in Fig. 2. Compared to the spectrum of BCN, a strong band appeared at 1718.50 cm⁻¹ in the IR spectrum of BCN-g-PAA, which was attributed to the acrylic acid C=O vibrations.³¹ The carboxyl content of BCN-g-PAA was titrationally determined to be 8.01% when the graft yield of PAA was 10.73% (see ESI†). Meanwhile, for BCN-g-PAA/SH, the band at 1703.22 cm⁻¹ could be assigned to C=O stretching of carboxylic groups of SH and the C=O vibrations of acrylic acid. The band at 1609.44 cm⁻¹ was due to the absorption of benzene ring in SH.^32,33 Additionally, compared to BCN-g-PAA, the absorption band at 1059.8 cm⁻¹ shifted to 1033.56 cm⁻¹ in the spectrum of BCN-g-PAA/SH, indicating certain chemical reactions between SH and BCN-g-PAA happened during the polymerization process.³³

Fig. 2 FTIR spectra of BCN, BCN-g-PAA and BCN-g-PAA/SH.

3.3 Effects of pH on adsorption

The influence of initial pH on Cu²⁺ adsorption was investigated over the pH range from 2.0 to 4.5. It is noteworthy that when pH was higher than 4.5, precipitation of Cu(OH)₂ would occur.^34–36 Fig. 3 compared the adsorption capacities of BCN, BCN-g-PAA and BCN-g-PAA/SH for Cu²⁺ at different pH. The unmodified nanofibers exhibited poor adsorption capacity over the entire pH range. The hydroxyl groups on cellulose nanofibers seemed to have only limited affinity towards Cu²⁺. Nevertheless, the adsorption performance of BCN could be largely enhanced when grafted with PAA or PAA/SH, introducing many carboxyl groups to BCN's surface. The adsorption behaviors of BCN-g-PAA and BCN-g-PAA/SH were highly pH-dependent. The Cu²⁺ adsorption capacity reduced with the lowering of pH and reached the minimum when the initial pH was at 2.0. When pH was low, –COOH was the primary form of functionalities, leading to poor Cu²⁺ adsorption as a result of weak electrostatic interaction between them. However, when pH was increased, the carboxyl groups could be partially ionized into –COO⁻. Consequently, the interaction between the –COO⁻ and Cu²⁺ would be improved.^37–39 The highest Cu²⁺ adsorption was achieved at pH = 4.5. This result well reproduced the early discovery on other biosorbents.² It is important to note that most industrial wastewater containing Cu²⁺ is weakly acidic.⁴⁰ Therefore, the present materials with maximum adsorption point of Cu²⁺ at pH = 4.5 are particularly suitable for purification of industrial waste water, representing a great potential for real-world applications.

Fig. 3 Effects of pH on the Cu²⁺ adsorption (initial Cu²⁺ concentration, 40.0 mmol L⁻¹; contact time, 120 min; oscillation frequency, 180 rpm; and adsorbent dose, 40 mg/20 mL).

The Cu²⁺ adsorption mechanism of modified nanofibers could be assumed to occur through chelation and ion-exchange between positively charged Cu²⁺ and ionized or nonionized carboxylic groups. This mechanism was schemed as the following equations.

R–COOH → R–COO⁻ + H⁺ (2)

R–COO⁻ + Cu²⁺ → R–COO⁻Cu²⁺ (3)

Based on the above analysis, pH = 4.5 was determined to be the optimal condition for Cu²⁺ adsorption and was hence used in the subsequent experiments.

3.4 Adsorption kinetics for Cu²⁺

The effects of contact time on the adsorption capacity of the modified nanofibers were studied. As shown in Fig. 4, the Cu²⁺ adsorption capacities of BCN-g-PAA and BCN-g-PAA/SH exhibited a sharp increase during the initial 10 min and reached an equilibrium state in 20 min, indicating there were plenty of readily accessible sites available for a rapid adsorption. It is conceivable that the huge specific surface area of nanocellulose promotes the adsorption rate to a large extent. This attribute is very advantageous for water treatment.

Fig. 4 Influences of contact time on the Cu²⁺ adsorption (initial Cu²⁺ concentration, 40.0 mmol L⁻¹; pH, 4.5; oscillation frequency, 180 rpm; and adsorbent dose, 40 mg/20 mL).

To examine the controlling mechanism of the adsorption process, two most commonly used kinetics models, the pseudo-first-order and pseudo-second-order models, were used to test the experimental data.⁴¹

Pseudo-first-order: ln(q_e − q_t) = −k₁t + ln q_e (4)

Pseudo-second-order: t/q_t = 1/(k₂q_e²) + t/q_e (5)

Where q_e (mmol g⁻¹) is the adsorption capacity at adsorption equilibrium, q_t (mmol g⁻¹) is the adsorption capacity at time t (min). k₁ (min⁻¹) and k₂ (g mmol⁻¹ min⁻¹) are the kinetics rate constants for the pseudo-first-order and pseudo-second-order models, respectively.

Table 1 showed the results obtained from two kinetics models. The correlation coefficients for the pseudo-first-order kinetics model were relatively low. In contrast, the pseudo-second-order equation agreed well with the data. The correlation coefficients were all high (>0.99), suggesting the pseudo-second-order kinetics model provided a good correlation in this study. The pseudo-second-order model implies that the adsorption is a rate controlling step, which involves valence forces through sharing electrons between metal ions and adsorbent.⁴² The calculated adsorption capacities at equilibrium (q^cal_e) of pseudo-second-order model were 0.727, 0.709 and 0.286 mmol g⁻¹ for BCN-g-PAA, BCN-g-PAA/SH and BCN, respectively. These results indicated a remarkable improvement of adsorption capability through chemical modification of BCN. Compared with other cellulose-based biosorbents, the BCN-g-PAA and BCN-g-PAA/SH in this study showed distinctly higher adsorption capacities for Cu²⁺. For example, Zhou et al. fabricated cellulose/chitin beads with a maximum Cu²⁺ adsorption capacity of 0.51 mmol g⁻¹.⁴³ Zhang et al. produced an efficient adsorbent from cellulose/alginic acid blend, which exhibited a maximum Cu²⁺ adsorption capacity of 0.52 mmol g⁻¹.⁴⁴

Table 1 Adsorption kinetic parameters for Cu²⁺ adsorption. q^cal_e, calculated q_e values; q^exp_e, experimental q_e values

	q^expe (mmol g^−1)	Pseudo-first-order			Pseudo-second-order
		q^cale (mmol g^−1)	k1	R^2	q^cale (mmol g^−1)	k2	R^2
BCN-g-PAA	0.716	0.689	0.303	0.9627	0.727	0.676	0.9989
BCN-g-PAA/SH	0.702	0.676	0.361	0.9787	0.709	0.723	0.9994
BCN	0.288	0.276	0.438	0.9851	0.286	1.816	0.9975

---

3.5 Adsorption isotherms of Cu²⁺

As shown in Fig. 5, the equilibrium adsorption capacity of BCN-g-PAA and BCN-g-PAA/SH increased with the increase of initial Cu²⁺ concentration from 0.2 to 20.0 mmol L⁻¹. When the initial Cu²⁺ concentration was low (≤1.0 mmol L⁻¹), Cu²⁺ in solutions could be almost completely removed by the modified BCN. Moreover, BCN-g-PAA/SH exhibited a higher adsorption capacity compared with BCN-g-PAA. Hence, the introduction of SH on BCN-g-PAA favored the adsorption when Cu²⁺ concentration was low. This conclusion also agreed well with previous report.²⁷

Fig. 5 Influences of initial Cu²⁺ concentration on the Cu²⁺ adsorption (pH, 4.5; oscillation frequency, 180 rpm; and adsorbent dose, 40 mg/20 mL).

To further explore the adsorption mechanism and describe how pollutants interact with adsorbents. Langmuir and Freundlich isotherm models were used to analyze the experiment data.³¹

Langmuir equation: c_e/q_e = 1/(bq_max) + c_e/q_max (6)

Freundlich equation: ln q_e = (1/n)ln c_e + ln K_f (7)

Where q_e (mmol g⁻¹) is the mass of Cu²⁺ adsorbed per mass unit of adsorbent at equilibrium, c_e (mmol L⁻¹) is the equilibrium concentration of the remaining Cu²⁺ in the solution, q_max (mmol g⁻¹) is the mass of adsorbent at complete monolayer coverage and b (L mmol⁻¹) is the Langmuir constant that related to the heat of adsorption. The values of q_max and b values were calculated from the slopes (1/q_max) and intercepts (1/(bq_max)). K_f (mmol g⁻¹) and n are the Freundlich constant and calculated from the slopes (n) and intercepts (ln K_f).⁴⁵

The Langmuir isotherm theory is based on the assumption that adsorption takes place at specific homogeneous sites within the adsorbent, the interaction among adsorbed substance can be negligible, and the adsorbent surface is saturated after monolayer adsorption. Meanwhile, Freundlich adsorption is used when more than monolayer coverage of the surface is assumed and the site is heterogeneous with different binding energy.⁴⁶ Table 2 listed the data obtained from the two isotherm models. The correlation coefficients of Langmuir isotherm model were rather low. In contrast, the adsorption of Cu²⁺ could be well fitted to the Freundlich isotherm model. The linear fitting of experimental data using Freundlich isotherm model was illustrated in Fig. 6. Freundlich parameters (K_f and n) indicate whether the nature of adsorption is favorable or not.^1,47 A high value of the intercept (K_f) is indicative of a high adsorption capacity.⁴⁸ Therefore, the adsorption capacities for Cu²⁺ could be interpreted in the order of: BCN-g-PAA/SH > BCN-g-PAA > BCN. The other Freundlich constants n is a measure of the deviation from linearity of the adsorption.⁴⁶ Isotherms with n > 1 are classified as L-type isotherms reflecting a high affinity between adsorbate and adsorbent and is indicative of chemisorption.⁴⁹

Table 2 Adsorption isotherm parameters for Cu²⁺ adsorption. q_m, the monolayer adsorption capacity

	Langmuir model			Freundlich model
	qm (mmol g^−1)	b	R^2	Kf	n	R^2
BCN-g-PAA	0.905	0.124	0.9718	0.104	1.389	0.9885
BCN-g-PAA/SH	1.016	0.175	0.9696	0.114	1.508	0.9915
BCN	0.390	0.327	0.9688	0.041	1.438	0.9925

---

Fig. 6 The linear fitting result for the Freundlich isotherm model.

The surface of BCN-g-PAA/SH after Cu²⁺ adsorption was further analyzed using SEM coupled with EDS (Fig. S1†). The acquired EDS spectrum (Fig. S1B†) and the estimated element composition (Table S1†) showed that the weight ratio of Cu in the selected area was as high as 68.17%, suggesting an excellent adsorption capacity of the materials.

3.6 Desorption and reusability

In the desorption and regeneration experiments, 0.1 mol L⁻¹ H₂SO₄ was used as the desorbing agent while 0.1 mol L⁻¹ NaOH was used as the regenerating agent to investigate the reusability of modified nanofibers.²⁷ Regeneration ratio was defined as the ratio of re-adsorbed Cu²⁺ amount to initially adsorbed one, and the results were presented in Fig. 7. The regeneration ratio of BCN-g-PAA/SH was much higher than that of BCN-g-PAA for all 4 cycles of adsorption/desorption. The adsorption capacity of BCN-g-PAA/SH was even remarkably improved after regeneration as its regeneration ratios were all higher than one. Therefore, it can be deduced that the introduction of SH is favorable for improving the adsorbent's reusability. During the regeneration process with NaOH solution, some carboxylic groups on the modified nanofibers could be transformed into carboxylate groups which exhibited stronger affinity to Cu²⁺. On the other hand, an amount of SH could be released from the BCN-g-PAA/SH in the system of alkaline solution. As a consequence, some vacancy sites could be formed on the surface of the adsorbent, which afforded the chance for Cu²⁺ entrance and increased the adsorption capacity.²⁵ As SH possesses many ionic functional groups, its capability to interact with metallic cations has long been recognized.⁵⁰ The present study further revealed the advantage of SH-modified adsorbents which allowed an excellent reusability. This feature is very promising for cost-effective water treatments.

Fig. 7 The regeneration ratios for the adsorbents (initial Cu²⁺ concentration, 40.0 mmol L⁻¹; pH, 4.5; oscillation frequency, 180 rpm; and adsorbent dose, 40 mg/20 mL).

4 Conclusions

In this study, BCN were extracted from bamboo pulp fibers through high shear homogenization coupled with ultrasonication. The BCN were chemically modified through surface graft polymerization with PAA and PAA/SH respectively for Cu²⁺ adsorption from aqueous solutions. Obtained data indicated the rapid removal of Cu²⁺ using such nanobiosorbents. The adsorption capacity of modified BCN was much higher than that of pristine BCN. When the Cu²⁺ concentration was low, BCN-g-PAA/SH had a higher adsorption capacity than BCN-g-PAA. In addition, the reusability could be remarkably improved after the introduction of SH segments. The adsorption kinetics data could be well described by the pseudo-second-order model, and the adsorption process followed the Freundlich isotherm model.

Acknowledgements

The authors would like to thank National Natural Science Foundation of China (51303112, 51473100 and 51433006) for financial support of this work.

References

1. B. Zhu, T. X. Fan and D. Zhang, J. Hazard. Mater., 2008, 153, 300–308.
2. M. Mukhopadhyay, S. B. Noronha and G. K. Suraishkumar, Bioresour. Technol., 2007, 98, 1781–1787.
3. D. Mohan and K. P. Singh, Water Res., 2002, 36, 2304–2318.
4. J. L. Wang, L. G. Wei, Y. C. Ma, K. L. Li, M. H. Li, Y. C. Yu, L. Wang and H. H. Qiu, Carbohydr. Polym., 2013, 98, 736–743.
5. T. A. Kurniawan, G. Y. S. Chan, W. Lo and S. Babel, Sci. Total Environ., 2006, 366, 409–426.
6. R. A. Goyer, Casarett and Doull's Toxicology, 1991, vol. 4, pp. 653–655.
7. M. Gavrilescu, Eng. Life Sci., 2004, 4, 219–232.
8. Z. Reddad, C. Gerente, Y. Andres and P. L. Cloirec, Environ. Sci. Technol., 2002, 36, 2067–2073.
9. I. Giannopoulou and D. Panias, Miner. Eng., 2007, 20, 753–760.
10. S. S. Tripathy and S. B. Kanungo, J. Colloid Interface Sci., 2005, 284, 30–38.
11. C. Ludwig and P. W. Schindler, J. Colloid Interface Sci., 1995, 169, 284–290.
12. Q. T. Meng, X. L. Zhang, C. He, G. J. He, P. Zhou and C. Y. Duan, Adv. Funct. Mater., 2010, 20, 1903–1909.
13. D. Mohan, J. C. U. Pittman and P. H. Steele, J. Colloid Interface Sci., 2006, 297, 489–504.
14. A. S. Singha and A. K. Rana, J. Polym. Environ., 2013, 21, 141–150.
15. B. Wang, H. B. Wu, L. Yu, R. Xu, T. T. Lim and X. W. Lou, Adv. Mater., 2012, 24, 1111–1116.
16. N. Li, L. D. Zhang, Y. Z. Chen, M. Fang, J. X. Zhang and H. M. Wang, Adv. Funct. Mater., 2012, 22, 835–841.
17. A. Demirbas, J. Hazard. Mater., 2008, 157, 220–229.
18. P. K. Neghlani, M. Rafizadeh and F. A. Taromi, J. Hazard. Mater., 2011, 186, 182–189.
19. M. Stephen, N. Catherine, M. Brenda, K. Andrew, P. Leslie and G. Corrine, J. Hazard. Mater., 2011, 192, 922–927.
20. J. Q. Zhao, W. Zhang, X. D. Zhang, X. X. Zhang, C. H. Lu and Y. L. Deng, Carbohydr. Polym., 2013, 97, 695–702.
21. J. Q. Zhao, X. He, Y. R. Wang, W. Zhang, X. X. Zhang, X. D. Zhang, Y. L. Deng and C. H. Lu, Carbohydr. Polym., 2014, 104, 143–150.
22. J. Q. Zhao, X. M. Zhang, R. Tu, C. H. Lu, X. He and W. Zhang, Cellulose, 2014, 21, 1859–1872.
23. Z. Wan, Z. Xiong, H. Ren, Y. Huang, H. Liu, H. Xiong, Y. Wu and J. Han, Carbohydr. Polym., 2011, 83, 264–269.
24. H. Liu, Y. Huang, L. Yuan, P. He, Z. Cai, Y. Shen, Y. Xu, Y. Yu and H. Xiong, Carbohydr. Polym., 2011, 83, 264–269.
25. J. Zhang and A. Wang, J. Chem. Eng. Data, 2010, 55, 2379–2384.
26. M. Arias, M. T. Barral and J. C. Mejuto, Chemosphere, 2002, 48, 1081–1088.
27. Y. S. Zheng, B. Hua and A. Q. Wang, Desalination, 2010, 263, 170–175.
28. J. N. Solanki, R. Sengupta and Z. V. P. Murthy, Solid State Sci., 2010, 12, 1560–1566.
29. N. C. Feng, X. Y. Guo and S. Liang, J. Hazard. Mater., 2009, 164, 1286–1292.
30. W. S. Chen, H. P. Li, Q. Yu, Y. X. Liu and J. Li, Soft Matter, 2011, 7, 10360–10368.
31. O. C. S. A. Hamouz and A. A. Shaikh, Ind. Eng. Chem. Res., 2012, 51, 14178–14187.
32. J. P. Zhang, A. Li and A. Q. Wang, React. Funct. Polym., 2006, 66, 747–756.
33. J. H. Liu, Q. Wang and A. Wang, Carbohydr. Polym., 2007, 70, 166–173.
34. I. Villaescusa, N. Fiol, M. Martínez, N. Miralles, J. Poch and J. Serarols, Water Res., 2004, 38, 992–1002.
35. H. A. Elliott and C. P. Huang, Water Res., 1981, 15, 849–855.
36. M. Asmal, A. H. Khan, S. Ahmad and A. Ahmad, Water Res., 1998, 32, 3085–3091.
37. H. Kaşgöz, S. Özgümüş and M. Orbay, Polymer, 2003, 44, 1785–1793.
38. G. Güçlü, A. Ebru, S. Emik, T. B. İyim, S. Özgümüş and M. Özyürek, Polym. Bull., 2010, 65, 333–346.
39. G. Güçlü, K. Güçlü and S. Keleş, J. Appl. Polym. Sci., 2007, 106, 1800–1805.
40. K. Kadirvelu, K. Thamaraiselvi and C. Namasivayam, Bioresour. Technol., 2001, 76, 63–65.
41. X. W. Peng, L. X. Zhong, J. L. Ren and R. C. Sun, J. Agric. Food Chem., 2012, 60, 3909–3916.
42. X. J. Hu, J. S. Wang, Y. G. Liu, X. Li, G. M. Zeng, Z. L. Bao, X. X. Zeng, A. W. Chen and F. Long, J. Hazard. Mater., 2011, 185, 306–314.
43. D. Zhou, L. N. Zhang, J. P. Zhou and S. L. Guo, Water Res., 2004, 38, 2643–2650.
44. L. N. Zhang, J. Cai, J. P. Zhou and Y. R. Tang, Sep. Sci. Technol., 2004, 39, 1203–1219.
45. L. C. Zheng, Z. Dang, C. F. Zhu, X. Y. Yi, H. Zhang and C. Q. Liu, Bioresour. Technol., 2010, 101, 5820–5826.
46. E. M. Ö. Kaya, A. S. Özcan, Ö. Gök and A. Özcan, Adsorption, 2013, 19, 879–888.
47. F. H. Frimmel and L. Huber, Environ. Int., 1996, 22, 507–517.
48. G. McKay, M. S. Otterburn and A. G. Sweeney, Water Res., 1980, 14, 21–27.
49. H. K. Boparai, M. Joseph and D. M. O'Carroll, J. Hazard. Mater., 2011, 186, 458–465.
50. J. Z. Yi and L. M. Zhang, Bioresour. Technol., 2008, 99, 2182.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra08307e

‡ These authors contributed equally to this work.

---