The stability of magnetic chitosan beads in the adsorption of Cu²⁺

Abstract

Magnetic chitosan beads (MCBs) synthesized through a common embedding method and crosslinked with glutaraldehyde were studied for the adsorption of Cu²⁺ in aqueous solution. MCBs were investigated by means of hysteresis curves, thermo-magnetization curves, TGA, FTIR and XRD. Kinetic, isothermal and thermodynamic studies were applied to investigate the adsorption of Cu²⁺ by the MCBs. The effects of the pH value and the coexisting ions on the adsorption of Cu²⁺ were explored. Subsequently, the stability of the MCBs including the adsorption capacity (Q_e) and the saturated magnetization (M_s) stability was investigated under various conditions. Simultaneously, FTIR and XRD were utilized to further analyze the change in the MCBs in the stability studies. The results suggested that the MCBs showed good stability without a significant loss in Q_e and M_s after regeneration from five sequential cycles. The MCBs presented excellent storage stability while stored in a drying vessel for ten months. For thermal stability, the Q_e and M_s of the MCBs presented similar variations after being heated in the atmosphere and in a vacuum at different temperatures for 30 h. The Q_e changed slightly after being heated at 50 and 100 °C, and decreased sharply at 150 and 200 °C. The magnetization was stable in the lower temperature range (50–150 °C) and increased slightly at 200 °C. Different magnetic fields had little effect on the stability of the MCBs.

---

1 Introduction

Heavy metal contamination in water bodies is derived from various industries such as metal plating, smelting processes, electric manufacturing, mining, oil refining, tanneries and textile and paper industries, etc.¹ It has brought serious harm to public health and the environment. Various techniques have been employed for heavy metal treatment, including adsorption, ion-exchange, membrane filtration, chemical precipitation, coagulation-flocculation, etc.^2–9 Among these techniques, adsorption, as an effective and simple technology, has been studied extensively due to its favorable properties and feasibility.

As a novel adsorbent, chitosan has been widely studied in the treatment of heavy metals due to its abundant source, easy modification, biodegradability, biocompatibility, favorable adsorption performance, and nontoxicity, etc. Moreover, various materials could blend with chitosan to form composites for the removal of heavy metals, including ceramic alumina,¹⁰ magnetite,⁴ perlite,¹¹ cellulose,¹² cotton fiber,¹³ clay,¹⁴ PVA,¹⁵ PVC,¹⁶ alginate,¹⁷ graphene,¹⁸ etc. Among these hybrid materials, magnetic chitosan materials have received much attention in heavy metal treatment and have obtained a desirable removal capacity. Jiang¹⁹ prepared magnetic chitosan beads to remove Cu²⁺ with the maximum capacity of 129.6 mg g⁻¹. The maximum capacity for triethylene-tetramine grafted magnetic chitosan to adsorb Pb²⁺ reached 370.63 mg g⁻¹.²⁰ Thiourea-modified magnetic CS microspheres exhibited a very high adsorption capacity for Hg²⁺ (625.2 mg g⁻¹).²¹ Furthermore, one more important characteristic is the magnetization of magnetic chitosan materials, which enables the magnetic chitosan materials to be separated from the liquid through a simple magnetic process to allow recycling after adsorbing heavy metals, saving cost without secondary pollution.

At present, numbers of researchers have focused on the preparation of novel magnetic chitosan materials, and the modification of the old materials for improving the adsorption capacity. Nonetheless, few research works were focused on studying the stability of magnetic chitosan materials and their adsorption capacity in various conditions, which were crucial for the practical application of magnetic chitosan materials. Favorable stability of magnetic chitosan materials was beneficial for practical application. Moreover, the stability of the adsorption capacity could maintain the ability to remove heavy metals. The stability of the magnetization could ensure the recoverability with the help of a magnetic field. The unstable performance of magnetic chitosan materials in various environmental conditions would hinder their application. The knowledge of the stability of magnetic chitosan materials in various environmental conditions might be helpful for a better understanding and planning of their applications and storage conditions. Therefore, it is essential to study the stability of magnetic chitosan materials for their practical application.

In the present paper, a common embedding method for preparing magnetic chitosan beads (MCBs) was studied to adsorb Cu²⁺. The uptake behavior of the MCBs was expounded via adsorption kinetics, isotherms and thermodynamics. The effects of the initial pH value and the coexisting ions on the adsorption of Cu²⁺ were explored, and the stability of the MCBs including the adsorption capacity (Q_e) and the saturation magnetization (M_s) stability was investigated. The regeneration stability was examined using a NaOH and ethylene diamine tetraacetic acid (EDTA) mixed solution, and the storage stability was studied by storing MCBs in desiccators for ten months. The thermal stability of MCBs was investigated after heating under atmospheric and vacuum conditions. Additionally, the stability of MCBs was also studied in different magnetic fields.

2 Materials and methods

Chitosan (deacetylation degree ≥ 95%) was purchased from Aladdin (China). Other reagents (analytical-reagent grade) were purchased from Sinopharm Chemical Reagent Co., Ltd (China) and distilled water was used to prepare all the solutions.

2.1. Preparation of Fe₃O₄

10 g of FeSO₄·7H₂O was dissolved into deionized water and the pH value was adjusted to 11.0 with 2 M NaOH solution. The solution was put in a water bath with a thermostat set at 90 °C under vigorous stirring for 4 h. The product was washed several times with deionized water and ethyl alcohol followed by drying in a vacuum oven at 50 °C. The obtained Fe₃O₄ was kept in a dryer for further use.

2.2. Preparation of MCBs

The procedure for preparing the MCBs was as in the former studies with some modifications.^22,23 0.4 g of chitosan was dissolved in 40 mL of acetic acid (1%, v/v) with magnetic stirring. 0.6 g of Fe₃O₄ was added into 300 mL of deionized water with ultrasound for 30 min to disperse, and then the solution was mixed uniformly with the prepared chitosan solution by magnetic stirring for 30 min. The Fe₃O₄–chitosan solution was sprayed into a NaOH solution with a peristaltic pump to form magnetic chitosan gel beads. The beads were thoroughly washed with deionized water to give a neutral pH, then glutaraldehyde solution (9.32 mL, 0.25%) was added to cross-link the beads for 12 h in an orbital water bath shaker at 150 rpm and room temperature. The beads were washed with ethyl alcohol and deionized water to remove the residual glutaraldehyde followed by freeze drying.

2.3. Characterization of MCBs

The crystal structure of Fe₃O₄ was analyzed using X-ray diffraction (XRD) measurements (Shimadzu XRD-6100, Japan) at room temperature. A PPMS-9T (EC-II) of Quantum Design (USA) was used at 300 K with an applied magnetic field up to 1.5 T to characterize the magnetic properties of Fe₃O₄ and the MCBs, and the temperature dependence of the magnetization was recorded in a magnetic field of 1.5 T with a temperature increase from 250 to 400 K. The MCBs were mixed with KBr and pressed to a pellet for Fourier-transform infrared spectroscopy (FTIR) on a Nicolet 6700 FT-IR spectrometer from Thermo Electron Corporation (USA). Thermogravimetric analysis (TGA) measurements of the materials were performed with a Mettler Toledo thermal analyzer (Netzsch, Germany) under N₂ and air atmospheres at a heating speed of 20 °C min⁻¹ from 30 to 900 °C.

2.4. Batch studies

Adsorption experiments were conducted for the adsorption of Cu²⁺ in a thermostated shaker by shaking 0.025 g of the MCBs into 25 mL of Cu²⁺ solution at the required initial Cu²⁺ concentration and pH value. MCBs were separated using a magnet (47 × 47 × 22 mm, 0.3 T surface magnetic field) at every sampling time and the supernatant was used to determine the concentration of Cu²⁺ with an atomic adsorption spectrometer (Control AA 700, Analytik Jena AG, Germany). The pH value of the Cu²⁺ solution was adjusted by adding 0.1 M NaOH or HCl and measured using a pH meter (HQ11d, HACH, USA). The final pH value after the adsorption process reached equilibrium was also measured to identify the variation of pH before and after the adsorption process. Cu²⁺ solutions (100 mg L⁻¹) with different concentrations of anions were prepared with various sodium salts (NaAc, NaCl, NaNO₃, Na₂SO₄, Na₂CO₃, Na₃PO₄). Chlorides such as NaCl, KCl, CaCl₂ and MgCl₂ were added into the Cu²⁺ solutions to explore the effect of common cations on the adsorption of Cu²⁺. Additionally, the concentration of the coexisting ions in the Cu²⁺ solution varied from 0 to 300 mg L⁻¹.

The stability of the MCBs was evaluated by measuring M_s and Q_e of Cu²⁺ with various treatments and comparing with the original ones. For regeneration stability studies, the adsorption and regeneration processes were performed in five consecutive cycles. 0.025 g of the MCBs was loaded with Cu²⁺ using 25 mL of Cu²⁺ solution (100 mg L⁻¹) at 30 °C and pH 4.0 and a contact time of 150 min. The Cu²⁺ loaded MCBs were collected with the help of the magnet and washed with distilled water several times to remove the unadsorbed Cu²⁺. The collected MCBs were then agitated with a mixed solution (25 mL of 0.1 M NaOH and 25 mL of 0.1 M EDTA) for 180 min at 30 °C for regeneration. The regenerated MCBs were washed thoroughly with deionized water and subsequently reconditioned for adsorption in the succeeding cycles.

The storage life stability was investigated by storing the MCBs in desiccators for ten months (from June, 2014 to March, 2015), and samples were taken every month to measure the Q_e and M_s. Thermal stability experiments were conducted by heating the MCBs in a vacuum oven and an air-blast oven at 50, 100, 150 and 200 °C for 30 h. MCBs were placed in 0.2, 0.3, and 0.4 T magnetic fields over a period of 3 months to study the effect of magnetic fields on the stability, and samples were collected every month. The adsorption experiment of every sample for the stability studies was conducted with 100 mg L⁻¹ Cu²⁺ and a contact time of 150 min at pH 4.0 and 30 °C.

3 Results and discussion

3.1. Characterization

3.1.1. Magnetic properties. The magnetic properties of the MCBs evaluated by measuring the saturation magnetization with a physical property measurement system are shown in Fig. 1A. The saturation magnetization of Fe₃O₄ and the MCBs was 82.58 and 35.42 emu g⁻¹, respectively. As expected, the M_s of the MCBs decreased due to the coating of chitosan on Fe₃O₄. The excellent magnetic properties of the MCBs indicated that they had a strong magnetic response and could be aggregated easily using a magnet.

Fig. 1 (A) Magnetic hysteresis loops of Fe₃O₄(I) and MCBs(II). (B) Thermo-magnetization curves.

The magnetization variation of the MCBs with increasing temperature from 250 to 400 K is presented in Fig. 1B. The magnetization of the MCBs decreased slightly from 36.12 to 33.04 emu g⁻¹ when the temperature increased from 250 to 400 K, suggesting that the MCBs could maintain favorable magnetization at higher temperatures. Hence, an appropriate operating temperature of the MCBs should be lower than 100 °C, and the slight reduction of saturation magnetization would have little impact on the recovery process with the help of a magnet.

3.1.2. Thermogravimetric analysis. TGA curves of the materials heated in N₂ (reducing condition) and air (oxidative condition) are reported in Fig. 2. The TGA curve of Fe₃O₄ in a N₂ atmosphere (curve II, Fig. 2) showed that weight loss from 30 to 900 °C was about 3.3%, probably attributed to the escape of the physically adsorbed water and structural water in the sample. Fe₃O₄ presented two steps in an air atmosphere (curve I, Fig. 2): the first one was due to the removal of the adsorbed water at a lower temperature (<100 °C), and the second one at about 250 °C was due to the oxidation of Fe₃O₄ by oxygen and the red residue was about 102.0 wt%.

Fig. 2 Thermogravimetric curves of the naked Fe₃O₄ (I, II), MCBs (IV, III), and chitosan (VI, V) heated in air or a N₂ atmosphere, respectively.

Chitosan presents two main degradation steps in a N₂ atmosphere (curve V, Fig. 2), the first one was because of water evaporation at a lower temperature (<100 °C) and the second one at about 300 °C was due to the polysaccharide degradation, and the residue was about 24.98 wt%. In an air atmosphere, chitosan (curve VI, Fig. 2) exhibited three obvious stages of weight loss, corresponding to the removal of adsorbed water, the thermal decomposition of chitosan, and the further decomposition and oxidation of the residue.²⁴ Finally only a small residue (about 6.5 wt%) was obtained at 900 °C. Compared with the chitosan curve (curve V, Fig. 2), the thermogravimetric curves of the MCBs in a N₂ atmosphere (curve III, Fig. 2) had one more stage at about 800 °C, probably related to the decomposition of the synthetic shell and the further decomposition of the crosslinked chitosan, and no significant weight loss was observed at a higher temperature (>800 °C). The thermogravimetric curves of the MCBs in an air atmosphere (curve IV, Fig. 2) displayed a similar thermal behavior to that of pure chitosan; the difference was that the third step occurred at about 550 °C compared with pure chitosan at about 700 °C. The results were similar to that of the magnetic chitosan films discussed by Cesano.²⁵

Therefore, the weight content of Fe₃O₄ in the MCBs was estimated to be about 41.2% and 38.5% from thermogravimetric results in N₂ and air atmospheres, respectively.

3.1.3. FTIR spectra analysis. FTIR spectra of the naked Fe₃O₄, chitosan and MCBs are illustrated in Fig. 3. In the spectrum of the naked Fe₃O₄, the peak at 560–600 cm⁻¹ was assigned to the Fe–O bond vibration of Fe₃O₄. The FTIR spectrum of the virgin chitosan indicated the presence of predominant peaks at 3433 cm⁻¹ (O–H and N–H stretching and inter-hydrogen bonding of polysaccharides), 2875 cm⁻¹ (–CH stretching vibration in –CH and –CH₂), 1601 cm⁻¹ (–NH bending vibration in –NH₂), 1420 and 1324 cm⁻¹ (vibrations of OH, CH in the ring), and 1159–896 cm⁻¹ (including the glycosidic bonds, C–O and C–O–C stretching from the polysaccharide skeleton).^26–28

Fig. 3 FTIR spectra of Fe₃O₄ (I), chitosan (II) and MCBs (III).

Compared with the spectrum of chitosan, a new peak at around 1648 cm⁻¹ appeared in the spectrum of the MCBs, probably due to the C=N of the Schiff’s base formed between the amino group of chitosan and the aldehyde groups of glutaraldehyde.¹⁷ Additionally, a new peak at 587 cm⁻¹ was assigned to the Fe–O group, suggesting that chitosan was coated to the Fe₃O₄ nanoparticles successfully. Furthermore, the surface of Fe₃O₄ with negative charges had an affinity toward chitosan, so protonated chitosan could coat the magnetite nanoparticles via an electrostatic interaction and a chemical reaction through glutaraldehyde cross-linking.²⁹

3.1.4. XRD analysis. Fig. 4 shows the XRD patterns of Fe₃O₄ and the MCBs. Five significant characteristic peaks for Fe₃O₄ (2θ = 30.1, 35.4, 43.1, 56.9 and 62.5°) were observed in the two samples, which were consistent with the database in the JCPDS file (PDF no. 19-1402). The results revealed that the crystal structure of Fe₃O₄ remained stable in the preparation process and the Fe₃O₄ was indeed incorporated into the MCBs. The intensity of the diffraction pattern decreased slightly with the chitosan coating on Fe₃O₄.

Fig. 4 XRD patterns of the naked Fe₃O₄ (I) and MCBs (II) (literature values for the peak positions and intensities for bulk magnetite samples are indicated by the vertical bars).

3.2. Adsorption studies of MCBs

The adsorption kinetics are illustrated in Fig. 5A. The adsorption amount increased sharply at first with the increase in contact time followed by slow rise and stability, and it reached saturation after 150 min. The higher correlation coefficient indicated that the pseudo-second-order model fitted the experimental data better than the pseudo-first-order model as shown in Table S1.† The results suggested that the rate-limiting step of the adsorption process was the chemical adsorption process.^30,31

Fig. 5 (A) Adsorption kinetics of Cu²⁺ with different initial concentrations (contact time 150 min, initial pH 4.0, 303 K). (B) Adsorption isotherm of Cu²⁺ on the MCBs. Insert is the fitting of the Langmuir and Freundlich models (initial Cu²⁺ concentration 0–1083 mg L⁻¹, contact time 150 min, initial pH 4.0, 303 K). (C) Plots of ln(C_s/C_e) vs. C_s at various temperatures (initial concentration of Cu²⁺ 20–750 mg L⁻¹, contact time 150 min, initial pH 4.0, 293, 303 and 313 K). (D) Effect of initial pH on Cu²⁺ adsorption. Insert illustrates the relationship between the initial and final pH of the reaction solution (initial Cu²⁺ concentration 100 mg L⁻¹, contact time 150 min, 303 K).

The effect of the initial Cu²⁺ concentration on the adsorption capacity is shown in Fig. 5B, and the adsorption capacity increased as the initial Cu²⁺ concentration was enhanced. The adsorption of Cu²⁺ on the MCBs showed a stronger correlation (coefficient R² = 0.9935) with the Langmuir isotherm model than the Freundlich isotherm model (coefficient R² = 0.9050). And the value of Q_m (167.22 mg g⁻¹) obtained from the Langmuir curve in Table S2† was almost consistent with that obtained in the experiment (162 mg g⁻¹), revealing the mainly monolayer adsorption process.

Thermodynamic parameters can be calculated from Fig. 5C and the obtained values of ΔH⁰ and ΔS⁰ are depicted in Table S3.† The negative value of ΔG⁰ implies that the adsorption processes are spontaneous. Additionally, lower values of ΔG⁰ were obtained at lower temperatures, meaning that the reaction was more favorable and became relatively easier at lower temperatures. The negative ΔH⁰ change suggests the exothermic interaction of Cu²⁺ adsorbed by the MCBs, and the positive ΔS⁰ indicates an increase in disorder and randomness at the adsorbent/adsorbate interface during the adsorption process.

The effect of pH (from 2.0 to 6.0) on the adsorption process is shown in Fig. 5D. A higher adsorption capacity was obtained at higher pH values. The lower adsorption capacity at lower pH values (2.0 and 3.0) was attributed to the protonation of the amino group.³² Based on the pK_a of chitosan (∼6.3),³³ when the pH decreased from 7.3 to 4.3, the extent of protonation increased from 9 to 99%.³⁴ A vigorous growth of Q_e was observed with the decrease in the concentration of H⁺ (pH 3.0–5.0). When the pH was higher than 5.5, the precipitation of Cu(OH)₂ occurred³⁵ to interfere with the adsorption of Cu²⁺. Scanty precipitation was observed at pH 6.0 during the experiments, and significant precipitation was seen at pH 7. Therefore, the effect of pH on the adsorption process was investigated in the range of 2.0–6.0.

3.3. Effect of coexisting ions on the adsorption of Cu²⁺

Generally, various anions are always coexisting with heavy metal ions such as Ac⁻, Cl⁻, NO₃⁻, SO₄²⁻, CO₃²⁻, and PO₄³⁻. Fig. 6A shows the effect of these anions on the uptake of Cu²⁺. Compared with the blank uptake capacity (64.5 mg g⁻¹), the uptake of Cu²⁺ decreased by 1.9, 4.5, 5.2, and 4.8 mg g⁻¹ in presence of 300 mg L⁻¹ NO₃⁻, Cl⁻, SO₄²⁻, and Ac⁻, respectively, and increased by 16.5 and 24.5 mg g⁻¹ in presence of 300 mg L⁻¹ CO₃²⁻ and PO₄³⁻, respectively. The effect of NO₃⁻ on the uptake of Cu²⁺ was almost negligible. Ac⁻, Cl⁻ and SO₄²⁻ had a slightly competitive influence on the adsorption of Cu²⁺, probably due to their slight interaction with the sorbent. CO₃²⁻ and PO₄³⁻ were favorable for the adsorption of Cu²⁺. The adsorption capacity increased as the concentration of CO₃²⁻ and PO₄³⁻ increased from 50 to 300 mg L⁻¹, probably attributed to the formation of CuCO₃ and Cu₃(PO₄)₂ precipitates during the adsorption process.

Fig. 6 Effect of (A) anions and (B) cations on the adsorption of Cu²⁺ (adsorption experimental conditions: initial Cu²⁺ concentration 100 mg L⁻¹, contact time 150 min, initial pH 4.0, 30 °C).

A number of alkali and alkaline earth metal ions such as Na⁺, K⁺, Ca²⁺ and Mg²⁺ are often in coexistence with heavy metal ions in wastewater. The impact of these metal ions on the adsorption of Cu²⁺ is displayed in Fig. 6B. Na⁺ and K⁺ played a minor role in the adsorption capacity of Cu²⁺ (reductions of only 4 and 4.7 mg g⁻¹ at 300 mg L⁻¹, respectively). However, the adsorption capacity of Cu²⁺ by MCBs decreased clearly as the concentration of the alkaline earth metal ions (Mg²⁺ and Ca²⁺) increased, resulting in reductions of 9.5 and 11.5 mg g⁻¹ at 300 mg L⁻¹, respectively. Therefore, it was concluded that Na⁺ and K⁺ had a slight effect on the uptake of Cu²⁺. Nevertheless, Ca²⁺ and Mg²⁺ had a suppressive effect on the adsorption of Cu²⁺, likely due to the stronger interactions occurring between the available sites and Ca²⁺/Mg²⁺.

3.4. Stability of MCBs

3.4.1. The regeneration stability of MCBs. The reusability stability is a potential economical parameter for the MCBs. MCBs that had adsorbed Cu²⁺ could be collected with a magnet, and the ability of MCBs to regenerate is crucial for practicability. HCl, HNO₃, NaOH, and chelating agents thiourea and EDTA are widely employed in the desorption process of chitosan materials, although Fe₃O₄ and chitosan without crosslinking would be unstable and dissolvable in acidic situations. Hence, a mixed solution of 0.1 M NaOH and 0.1 M EDTA were applied for the regeneration of MCBs as shown in Fig. 7.

Fig. 7 The stability of MCBs during the regeneration process.

The uptake capacity of Cu²⁺ on the MCBs decreased slowly with increasing cycle numbers, and the adsorption capacity was found to be 59 mg g⁻¹ by the fifth cycle. Additionally, the regenerated MCBs during the five cycles had almost the same M_s values as those of the fresh ones, suggesting that the MCBs presented favorable magnetization stability during the regeneration process.

EDTA, a strong chelating agent, is an effective eluent for Cu²⁺, and it could form a stable complex with Cu²⁺. In the basic solution, the positively charged amino groups were deprotonated. Therefore, NaOH was also employed for the regeneration of magnetic chitosan.^36,37 Hence, the mixture of NaOH and EDTA was effective for the regeneration of MCBs.

The FTIR analysis of the original, loaded and regenerated MCBs (ESI Fig. S1A†) was helpful to identify the possible functional groups involved in the binding of Cu²⁺ and the regeneration results. After adsorption of Cu²⁺, significant changes in the FTIR spectra were observed at wave numbers of 1159–896 cm⁻¹. The reduction of the intensities at 3433 cm⁻¹ (O–H and N–H stretching) and the changes at 1601 cm⁻¹ (NH group in amine) indicated that amino and hydroxyl groups were the adsorption sites for the Cu²⁺ adsorption on the MCBs. The main peaks of the functional groups on the MCBs after desorption were similar to those of the original ones, suggesting that the MCBs were effectively regenerated. The characteristic peaks of the XRD patterns for Fe₃O₄ appeared in the regenerated MCBs (ESI Fig. S1B†), demonstrating that the magnetic cores (Fe₃O₄) in the MCBs were stable after five regeneration cycles.

In short, the above results indicated that the Q_e and M_s values of the MCBs are stable after regeneration, and MCBs could be potentially applied after regeneration using a mixed solution of NaOH and EDTA.

3.4.2. The storage stability of MCBs. The MCBs were stored in desiccators for 10 months (from June, 2014 to March, 2015). The variations of Q_e and M_s with time are shown in Fig. 8, revealing that the adsorption capacity of Cu²⁺ changed little and the magnetization of the MCBs was almost constant.

Fig. 8 The stability of MCBs during the storage time.

The FTIR spectrum of the MCBs stored for ten months (ESI Fig. S2A†) was coincident with that of the original ones, indicating that the storage process did not destroy the MCBs. From the XRD patterns (ESI Fig. S2B†), it was apparent that the characteristic diffraction peaks were all observable in the MCBs stored for ten months, suggesting no phase change of Fe₃O₄ during the storage process. Based on this observation, it could be concluded that the MCBs retained a stable adsorption capacity of Cu²⁺ and magnetization after 10 months of storage in dry conditions.

3.4.3. The thermal stability of MCBs. Temperature plays an important role in the storage life of MCBs. The MCBs were heated in air-blast and vacuum ovens for 30 h to explore the thermal stability of the MCBs as shown in Fig. 9. The thermal stability of the adsorption capacity is presented in Fig. 9A. Under atmospheric conditions, the Q_e of the MCBs heated at 50 and 100 °C decreased slightly, and declined sharply after being heated at 150 and 200 °C. The reason for the decrease of Q_e might be that the heating process could destroy the chitosan structure and functional group. For the MCBs heated under vacuum, the downward trend in Q_e was similar to that under atmospheric conditions, and the extent of the variation was less than that in the atmosphere. It was likely that the MCBs heated under atmospheric conditions would accelerate the decomposition of chitosan coated on Fe₃O₄. From the thermogravimetric results, the MCBs only lost the weight of the water evaporated from 30 to 200 °C in N₂ or air atmospheres; the degradation and decomposition were little and only just began. However, the thermogravimetric curves obtained at a heating speed of 20 °C min⁻¹ could not fully explain the variation of Q_e after heating for 30 h.

Fig. 9 The thermal stability of the MCBs ((A) adsorption capacity and (B) magnetization stability) heated under atmospheric conditions and under vacuum for 30 h.

The change in Q_e after heating at different temperatures could be better explained by the FTIR spectra of the MCBs (ESI Fig. S3A†). The FTIR spectra of the MCBs after heating under atmospheric conditions at 50 and 100 °C had slight variations compared to the original ones, in consequence of little change of the Q_e. And the FTIR of the MCBs after heating under atmospheric conditions at 150 and 200 °C had significant changes compared to the original ones. The decrease of the intensities of the bands was clearly seen at 3433, 2875, 1601, 1324 cm⁻¹ and in the range 1159–896 cm⁻¹, which were attributed to dehydration, deacetylation and depolymerization reactions.²⁷ Schiff base and –NH₂ groups were broken down, resulting in the decrease of the bands at 1654 and 1601 cm⁻¹, but it was masked by the occurrence of the band at 1642 cm⁻¹, which was attributed to unsaturated structures formed during the degradation or oxidation of chitosan.^27,38 The reduction in Q_e of the MCBs after heating at 150 and 200 °C under atmospheric conditions should be due to the destruction of the chitosan structure.

The trends in the FTIR spectra of the MCBs heated under vacuum (ESI Fig. S3B†) were similar to those of the MCBs heated under atmospheric conditions except that the peaks at 3433 and 2875 cm⁻¹ existed after heating at 150 °C. Similarly, the Q_e of the MCBs after heating at 150 °C under vacuum was higher than that under atmosphere. The variations in the FTIR were basically in line with the change of Q_e.

From the FTIR spectra of the MCBs heated under atmosphere and vacuum, the peak at 560–600 cm⁻¹ (Fe–O bond) changed little, which indicated that the Fe₃O₄ in the MCBs was stable after the different heating processes. The XRD results of the MCBs heated under atmosphere and vacuum at 150 and 200 °C changed little (ESI Fig. S3C†), which demonstrated that the Fe₃O₄ in the MCBs was stable even after heating at 150 and 200 °C, which was consistent with the analysis of the FTIR for the Fe–O bond.

The magnetization variations of the MCBs after being heated in vacuum and air-blast ovens are given in Fig. 9B. The magnetization had a minor change after heating at 50, 100 and 150 °C under atmospheric and vacuum conditions. The value of M_s increased to 39.94 and 39.56 emu g⁻¹ for atmospheric and vacuum conditions respectively after heating at 200 °C for 30 h. It was likely due to the degradation and decomposition of the crosslinked chitosan coated on Fe₃O₄, resulting in the weight of non-magnetic material in the MCBs being reduced and the magnetic substance (Fe₃O₄) being changed little. Thus the quality proportion of Fe₃O₄ increased a little, which was reflected in the increase of magnetization. The magnetization variation trend of the MCBs heated under vacuum was similar to that under atmospheric conditions. Therefore, the MCBs showed excellent stability after being heated at 50 and 100 °C for 30 h and poor adsorption capacity stability and passable magnetization stability after being heated above 150 °C under vacuum and atmospheric conditions.

3.4.4. The stability of MCBs stored in magnetic fields. The MCBs were stored in different magnetic fields for three months to investigate the effect of magnetic fields on the stability of the MCBs as shown in Fig. 10. It was apparent that both the Q_e and M_s values of the MCBs showed no significant changes in the different magnetic fields. Therefore, the MCBs should not be sticking together after being stored in the magnetic fields and the storage process had little influence on their removal capacity. The MCBs had almost the same saturation magnetization value as the original ones, which could guarantee the feasible application for separation.

Fig. 10 The stability of MCBs stored in different magnetic fields.

The FTIR spectra of the MCBs (ESI Fig. S4A†) after being stored in different magnetic fields for three months had no significant change compared with the original ones, demonstrating little chemical change during the storage process in different magnetic fields. The characteristic peaks in the XRD data of the MCBs stored in different fields (ESI Fig. S4B†) had little alteration, which demonstrated that the magnetic substance was validated as pure Fe₃O₄. It was concluded that the MCBs exhibited excellent stability after being stored in different magnetic fields for three months.

4 Conclusions

MCBs prepared by a classic method proved to be efficient and convenient materials for the removal of Cu²⁺ from aqueous solution. The adsorption kinetics agreed well with the pseudo second order model. The equilibrium isotherm data followed the Langmuir adsorption isotherm with a maximum adsorption capacity of 167.22 mg g⁻¹. And the adsorption process was feasible, spontaneous and exothermic. Moreover, the regeneration of the MCBs was achieved using a mixture of NaOH and EDTA due to the reversible interaction between chitosan and Cu²⁺. The adsorption capacity was kept at 59 mg g⁻¹ at the fifth cycle and the saturation magnetization changed little during the regeneration process. There was negligible change in Q_e and M_s after being stored in desiccators for ten months. Under atmospheric and vacuum heating, the stability of the Q_e values of the MCBs was maintained after being heated at 50 and 100 °C for 30 h, however, it deteriorated when heated at 150 and 200 °C. The higher heating process (150 and 200 °C) would lead to the break of the functional groups of the MCBs and the decrease of the adsorption capacity. The magnetization of the MCBs was stable after being heated at 50, 100 and 150 °C for 30 h, and the M_s values of the MCBs increased a little when heated at 200 °C. The MCBs exhibited good stability after being stored in different magnetic fields for three months.

The present work predicts that MCBs have excellent regeneration stability. In addition, MCBs will be stable in a dry environment, at appropriate temperatures (≤100 °C) and in different magnetic fields (≤0.3 T). Therefore, MCBs are promising materials due to their highly-efficient performance and excellent stability, and have the potential to be applied for practical heavy metal wastewater treatment.

Acknowledgements

This study is supported by the Natural Science Foundation of China (Project No. 51278295).

References

1. J. Siemiatycki, L. Richardson, K. Straif, B. Latreille, R. Lakhani, S. Campbell, M. C. Rousseau and P. Boffetta, Environ. Health Perspect., 2004, 112, 1447–1459.
2. J. Wang and C. Chen, Biotechnol. Adv., 2009, 27, 195–226.
3. K. C. Kang, S. S. Kim, J. W. Choi and S. H. Kwon, J. Ind. Eng. Chem., 2008, 14, 131–135.
4. M. Monier, D. M. Ayad, Y. Wei and A. A. Sarhan, J. Hazard. Mater., 2010, 177, 962–970.
5. A. Dabrowski, Z. Hubicki, P. Podkoscielny and E. Robens, Chemosphere, 2004, 56, 91–106.
6. N. K. Srivastava and C. B. Majumder, J. Hazard. Mater., 2008, 151, 1–8.
7. Q. Chen, Z. Luo, C. Hills, G. Xue and M. Tyrer, Water Res., 2009, 43, 2605–2614.
8. A. G. El Samrani, B. S. Lartiges and F. Villieras, Water Res., 2008, 42, 951–960.
9. W. S. Wan Ngah, L. C. Teong and M. A. K. M. Hanafiah, Carbohydr. Polym., 2011, 83, 1446–1456.
10. V. M. Boddu, K. Abburi, J. L. Talbott, E. D. Smith and R. Haasch, Water Res., 2008, 42, 633–642.
11. K. Swayampakula, V. M. Boddu, S. K. Nadavala and K. Abburi, J. Hazard. Mater., 2009, 170, 680–689.
12. X. Sun, B. Peng, Y. Ji, J. Chen and D. Li, AIChE J., 2009, 55, 2062–2069.
13. R. Qu, C. Sun, F. Ma, Y. Zhang, C. Ji, Q. Xu, C. Wang and H. Chen, J. Hazard. Mater., 2009, 167, 717–727.
14. V. N. Tirtom, A. Dinçer, S. Becerik, T. Aydemir and A. Çelik, Chem. Eng. J., 2012, 197, 379–386.
15. W. S. Wan Ngah, A. Kamari and Y. J. Koay, Int. J. Biol. Macromol., 2004, 34, 155–161.
16. S. R. Popuri, Y. Vijaya, V. M. Boddu and K. Abburi, Bioresour. Technol., 2009, 100, 194–199.
17. W. S. Wan Ngah and S. Fatinathan, Chem. Eng. J., 2008, 143, 62–72.
18. H. Hadi Najafabadi, M. Irani, L. Roshanfekr Rad, A. Heydari Haratameh and I. Haririan, RSC Adv., 2015, 5, 16532–16539.
19. W. Jiang, W. Wang, B. Pan, Q. Zhang, W. Zhang and L. Lv, ACS Appl. Mater. Interfaces, 2014, 6, 3421–3426.
20. S. P. Kuang, Z. Z. Wang, J. Liu and Z. C. Wu, J. Hazard. Mater., 2013, 260, 210–219.
21. L. Zhou, Y. Wang, Z. Liu and Q. Huang, J. Hazard. Mater., 2009, 161, 995–1002.
22. G. L. Rorrer, T. Y. Hsien and J. D. Way, Ind. Eng. Chem. Res., 1993, 32, 2170–2178.
23. G. Z. Kyzas and E. A. Deliyanni, Molecules, 2013, 18, 6193–6214.
24. X. Xiong, Y. Wang, W. Zou, J. Duan and Y. Chen, J. Chem., 2013, 2013, 1–8.
25. F. Cesano, G. Fenoglio, L. Carlos and R. Nisticò, Appl. Surf. Sci., 2015, 345, 175–181.
26. W. S. Wan Ngah and S. Fatinathan, J. Environ. Sci., 2010, 22, 338–346.
27. D. de Britto and S. P. Campana-Filho, Thermochim. Acta, 2007, 465, 73–82.
28. M. M. A. Pawlak, Thermochim. Acta, 2003, 396, 153–166.
29. G. Y. Li, Y. R. Jiang, K. L. Huang, P. Ding and J. Chen, J. Alloys Compd., 2008, 466, 451–456.
30. D. Cho, B. Jeon, C. Chon, Y. Kim, F. W. Schwartz, E. Lee and H. Song, Chem. Eng. J., 2012, 200–202, 654–662.
31. M. V. Dinu and E. S. Dragan, Chem. Eng. J., 2010, 160, 157–163.
32. H. Li, S. Bi, L. Liu, W. Dong and X. Wang, Desalination, 2011, 278, 397–404.
33. L. D. Eric Guibal, C. Milot and J. Roussy, Polym. Int., 1999, 48, 671–680.
34. T. C. Coelho, R. Laus, A. S. Mangrich, V. T. de Fávere and M. C. M. Laranjeira, React. Funct. Polym., 2007, 67, 468–475.
35. H. Yan, L. Yang, Z. Yang, H. Yang, A. Li and R. Cheng, J. Hazard. Mater., 2012, 229–230, 371–380.
36. Z. Zhou, S. Lin, T. Yue and T. Lee, J. Food Eng., 2014, 126, 133–141.
37. Y. G. Abou El-Reash, M. Otto, I. M. Kenawy and A. M. Ouf, Int. J. Biol. Macromol., 2011, 49, 513–522.
38. J. Zawadzki and H. Kaczmarek, Carbohydr. Polym., 2010, 80, 394–400.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra20943a

---