Preparation of chitosan/MCM-41-PAA nanocomposites and the adsorption behaviour of Hg(II) ions

A novel functional hybrid mesoporous composite material (CMP) based on chitosan and MCM-41-PAA was reported and its application as an excellent adsorbent for Hg(II) ions was also investigated. Innovatively, MCM-41-PAA was prepared by using diatomite and polyacrylic acid (PAA) with integrated polymer–silica hybrid frameworks, and then CMP was fabricated by introducing MCM-41-PAA to chitosan using glutaraldehyde as a cross-linking agent. The structure and morphology of CMP were characterized by X-ray diffraction, Fourier transform infrared spectra, thermogravimetric analysis, scanning electron microscopy and Brunauer–Emmett–Teller measurements. The results showed that the CMP possessed multifunctional groups such as –OH, –COOH and –NH2 with large specific surface area. Adsorption behaviour of Hg(II) ions onto CMP was fitted better by the pseudo-second-order kinetic model and the Langmuir model when the initial Hg(II) concentration, pH, adsorption temperature and time were 200 mg l−1, 4, 298 K and 120 min, respectively, as the optimum conditions. The corresponding maximum adsorption capacity could reach 164 mg g−1. According to the thermodynamic parameters determined such as free energy, enthalpy and entropy, the adsorption process of Hg(II) ions was spontaneous endothermic adsorption.


Introduction
Hg(II) ion is regarded as one of the most toxic heavy-metal ions [1]. The wastewater from many industries involved in mercury lamps, paint and chloralkali production is a source of Hg(II) ion pollution with its release to the ecosystem [2,3]. Owing to its highly soluble properties, emission of Hg (

Characterization
Fourier transform infrared (FT-IR) spectra were obtained using a PerkinElmer Spectrum One (B) spectrometer (PerkinElmer, Foster City, CA) with the wavelength range between 500 cm −1 and 4000 cm −1 using a KBr pellet. The morphology of samples was examined by electron scanning microscopy (SEM, JEOL 6500F, Japan). The X-ray diffraction (XRD) data were obtained by an XRD powder diffractometer (D8 Advance, Bruker, Germany) using Cu-K α radiation (λ = 1.54 Å) between 1°and 10°(2θ) at 40 kV, 40 mA. Thermogravimetric analysis (TG) was performed with a TG instrument (Netzsch 209C, Hanau, Germany) at a heating rate of 20°C min −1 from 40 to 600°C under N 2 flow. The Brunauer-Emmett-Teller (BET) and Barret-Joyner-Halenda (BJH) methods were used to determine the surface area and pore size distribution of nanocomposites in N 2 adsorption-desorption (Quadrasorbsi, Quantachrome, USA) experiments at 77 K.

Preparation of MCM-41-PAA
Based on diatomite, CTAB and PAA as a silica source, a template and the MCM-41 hybrid framework, MCM-41-PAA was prepared according to the following procedure. Firstly, diatomite (2.70 g) was added to a solution of NaOH (1.02 g, 25 ml of distilled water). After heating at 150°C for 5 h, the precursor was transferred to solution of CTAB (3.06 g, 51 ml of distilled water) under stirring. A 2.7 g aliquot of a 30% PAA solution was added drop-wise to the mixed solution, and then several drops of ethanol were added drop-wise and vigorously stirred for 1 h. The pH of the mixture was adjusted to 10 with 2 mol l −1 H 2 SO 4 and diverted into a stainless steel autoclave at 100°C for 24 h. Then, the precipitate of the autoclave was centrifuged and dried. The product (1 g) was moved to a solution of NH 4 NO 4 (0.3 g) in ethanol (50 ml) as an extracting agent to remove the template; the mixture was heated to 78°C for reflux and extraction of 12 h, and centrifuged and dried. Finally, MCM-41-PAA was obtained with repetition of the above-mentioned operation two times.

Preparation of CS/MCM-41-PAA nanocomposites
The MCM-41-PAA (2 g) was dissolved in 1% acetic solution (50 ml), with formation of homogeneous suspension by stirring. CS powder (0.5 g) was added to the mixture and heated at 45°C under vigorous stirring for 30 min to increase the homogeneity, and then cross-linked with adding 1% of GA solution (0.39 ml) drop-wise under constant stirring for 4 h. The pH of the mixture was adjusted to 7 with 2 mol l −1 NaOH and then distilled water; then the precipitate was dried and ground into powder.

Adsorption experiments
To investigate the adsorption capacity and removal rate of CMP for adsorption of Hg(II) ions, stock solutions of standardized Hg(II) ions (1000 mg l −1 ) were configured from Hg(NO 3 ) 2 ; other initial concentrations of Hg(II) ions were obtained by further dilution. The pH of the solution was adjusted from 1 to 6 with 0.8 mol l −1 HNO 3 . The adsorption experiments were performed as follows: CMP (0.1 g) was added to a 200 mg l −1 Hg(II) ion solution (100 ml) in a 250 ml beaker placed on a magnetic stirrer at 600 r.p.m. for 3 h to ensure equilibrium. The concentration of Hg(II) ions after adsorption was calculated by using dithizone spectrophotometry. Under acidic conditions, Hg(II) ions formed an orange complex with dithizone, and the absorption of the complex was determined at λ max = 485 nm, so the adsorption capacity and removal rate of CMP were obtained by the following equations: where C 0 , C e, V, m, q e and η are the initial and equilibrium concentrations of Hg(II) ion (mg l −1 ), volume of solution (l), sorbent dosage (g), adsorption capacity (mg l −1 ) and removal rate (%), respectively.

Results and discussion
3.1. Characterization of CMP

Fourier transform infrared spectra analysis
The FT-IR spectra of MCM-41-PAA and CMP are shown in figure 2. Figure 2a represents the characteristic peaks of MCM-41-PAA. The peak at 3664 cm −1 was attributable to the stretching vibration of O-H (in -COOH group). In addition, the stretching vibration of v as CH (in -CH 2 group), v sy CH (in -CH 2 group) and C=O (in -COOH group) could also be observed at 2930, 2860 and 1650 cm −1 , respectively. The peaks at 1470 and 1330 cm −1 were assigned to CH of the bending mode and C-O of the stretching mode, respectively, which are characteristic peaks of PAA. The peaks at 1090 and 806 cm −1 showed the stretching vibration of Si-O, which correspond to characteristic peaks of SiO 2 . Figure 2b shows the FT-IR spectrum of CMP. Besides the SiO 2 characteristic peaks and PAA, two new peaks at 1635 and 1598 cm −1 could be attributed to the C-O stretching vibration of NHCO and the N-H bending of NH 2 ; the broad peak at 3429 cm −1 originated from overlapping vibration of O-H and N-H because the -NH 2 absorption band shifted to a lower value.

Scanning electron microscopy analysis
The SEM images of MCM-41-PAA and CMP are shown in figure 3. Figure 3a demonstrates an interconnection of the nearly spherical particles with size in the range of 50-100 nm for MCM-41-PAA. As can be clearly seen in figure 3b, the white striped CS was homogeneously dispersed with the mesoporous material, forming an interesting monolithic structure.  of nanocomposites was slightly changed due to change of the inherent order caused by the mixing of MCM-41-PAA and CS.

Nitrogen adsorption-desorption isotherms
Textural characteristics of MCM-41-PAA and CMP including the total pore volumes (V total , cm 3 g −1 ), the pore diameters (D BJH , nm) and the BET surface area (S BET , m 2 g −1 ) were measured using N 2 physisorption techniques at 77 K, and are summarized in       1-6, and the corresponding results are presented in figure 6. It could be clearly seen that the maximum adsorption was at pH = 4 for Hg(II) ions. There was a competition between Hg(II) ions and H + ions on the surface of CMP at lower pH (pH < 4), which leads to the protonation of surface functional groups, with loss of the binding sites to chelate Hg(II) ions with an increase in acidity, resulting in a sharp decline in the adsorption capacity [43]. At higher pH (pH > 4), the hydroxide precipitation of Hg(II) ions might be formed on the surface and mesoporous channels of CMP, which will block the pores, decrease the retention and prevent further adsorption [35].

The effect of contact time and temperature
The influence of contact time on the adsorption capacity of Hg(II) ions was investigated, and the corresponding results are shown in figure 7. The adsorption capacity increased rapidly with contact time from 0 to 60 min because numerous surface active groups such as -NH 2 , -OH and -COOH of the adsorbents played a very important role in the adsorption of Hg(II) ions. Then the adsorption capacity continued to increase slowly by only relying on adsorption sites inside the pore structure, due to the saturation of the available adsorption sites, decrease of Hg(II) ion concentration and the presence of diffusion resistance, and finally reached saturation at 120 min.

Sorption kinetics
To define the efficiency of sorption, the rate of adsorption can be described by studying the kinetics of adsorption. According to the effect of contact time at the temperatures of 298, 308 and 318 K, the data of Hg(II) ions adsorbed are fitted by pseudo-first order [44] and pseudo-second order models [45], and both models may be expressed by the following equations: and where q e (mg g −1 ) and q t (mg g −1 ) are the adsorption amounts at equilibrium and at contact time t, and k 1 and k 2 are the first-and second-order kinetic rate constants, respectively. The linear plots and the parameters of adsorption kinetics determined from the slope and intercepts are illustrated in figure 8 and table 2. It could be seen that the correlation coefficients (R 2 ) at different temperatures of the second-order kinetic model were higher than 0.99, while the corresponding values of the first-order kinetic model were less than 0.95. In addition, the values of calculated q e from the pseudo-second-order model were consistent with the experimental ones.
The obtained results indicate that the adsorption process of Hg(II) ions onto CMP follows the pseudosecond-order kinetic model, demonstrating that the adsorption of Hg(II) ions is chemical adsorption by valence forces via exchange or sharing of electrons between functional groups of CMP such as amine, carboxyl and hydroxyl and Hg(II) ions.

Sorption isotherm models
The interaction mechanism between adsorbate and adsorbent at equilibrium time is explored by Freundlich and Langmuir models [46,47]. In this work, the data of the equilibrium adsorption were    and q e = q m K L C e 1 + K L C e , (3.4) where q e (mg g −1 ) represents the adsorption capacity at equilibrium, q m (mg g −1 ) is the maximum adsorption capacity, C e (mg l −1 ) is the Hg(II) ion concentration at equilibrium, K F (mg g −1 ) and K L (l mg −1 ) express the Freundlich and Langmuir constants, respectively, and n reflects the adsorption intensity. Nonlinear regression analysis of the Langmuir and Freundlich isotherm models is illustrated in figure 9; the corresponding parameters of sorption isotherm models calculated are listed in table 3. In comparison with the Freundlich isotherm model, the Langmuir isotherm model was more suitable to describe the interaction mechanism of Hg(II) ion adsorption onto CMP, because the correlation coefficients (R 2 ) were higher than 0.99 and q m calculated by the Langmuir isotherm model was closer to the experimentally measured equilibrium adsorption capacity. It was thought that the adsorption of Hg(II) ions occurred at the identical limited number of monolayer adsorption sites on the surface of the adsorbent; firstly, Hg(II) ions could transfer from the solution to CMP by bulk diffusion and intraparticle diffusion, and then were adsorbed by chemical complexation at the active sites.

Sorption thermodynamics
The effect of temperature on the adsorption of Hg(II) ions onto CMP was investigated at 298, 308 and 318 K; related thermodynamic parameters including enthalpy ( H 0 ), entropy ( S 0 ) and Gibbs energy ( G 0 ) were obtained based on the following equations: ln where R (8.314 J mol −1 K −1 ) represents the ideal gas constant and T (K) is the absolute temperature. The straight line obtained by plotting ln(q e /C e ) versus 1/T is shown in figure 10. According to the slope and intercept of the line, H 0 (−32.97 kJ mol −1 ) and S 0 (−98.11 J mol −1 K −1 ) were determined; the negative value of H 0 suggests the exothermic nature of the whole adsorption process, which is consistent with the fact that the adsorption capacity of CMP decreases with increase in temperature. Based on equation (3.6), values of G were estimated to be −3.76, −2.70 and −1.80 kJ mol −1 for 298, 308 and 318 K, respectively, and the numerical value of G 0 decreased with the rise in temperature, indicating that the sorption process of the Hg(II) ions onto CMP was more favourable at lower temperatures. In addition, the negative value of G 0 reveals that the sorption process was exothermic.

Possible mechanism of Hg(II) ion adsorption onto CMP
The adsorption mechanism of Hg(II) ions can be described as follows. Firstly, Hg(II) ions are transferred from the solution to the surface of CMP by the active functional groups from CS. Those active sites, such as hydroxyl (-OH) and amino (−NH 2 ) groups, can form coordination bonds with metal ions via a chelation effect according to the following chelating mechanism [48][49][50]: (CMP) − OH + Hg 2+ + (CMP) − OH ↔ (CMP) − OH · · · Hg 2+ · · · HO − (CMP) (3.8) and (CMP) − NH 2 + Hg 2+ + (CMP) − OH ↔ (CMP) − NH 2 · · · Hg 2+ · · · HO − (CMP) .   Table 4 compares the best adsorption capacity of CMP with different common adsorbents for the removal of Hg(II) ions. It could be clearly seen that the adsorption capacity is higher than that of most adsorbents according to table 4. In addition, if we take into account the process characteristics of this work, for example, using cheap diatomite as the silica source can produce mesoporous materials, and, moreover, constructing organic-inorganic hybrid frameworks with polymer materials improves the properties of mesoporous materials, we can believe that CMP will be a good candidate for applications in heavy metal removal from wastewater.

Conclusion
In summary, the adsorbent (CMP) with a uniform adjustable pore structure, large specific surface area and rich in groups including -OH, -NH 2 and -COOH was successfully designed and characterized. These characteristics could facilitate the contact of Hg(II) ions and active adsorption sites and attainment of rapid equilibrium adsorption. The results demonstrated that the maximum adsorption capacities reached 164 mg g −1 with a pH value of 4 and a contact time of 120 min at 298 K; which demonstrated excellent adsorption ability for Hg(II) ions. The adsorption behaviour followed pseudo-second-order kinetics, and the equilibrium data were well fitted by the Langmuir isotherm. The negative H 0 and G 0 suggested that the adsorption was a spontaneous exothermic process. These interesting findings might provide some new inspiration that using cheap diatomite as the silica source is a versatile approach for designing other mesoporous materials, and that constructing organic-inorganic hybrid frameworks with polymer materials enhances the adsorption properties of mesoporous materials.
Data accessibility. This article does not contain any additional data. Authors' contributions. J.H. designed the study. Y.F. prepared all the samples for analysis. Y.F. and Y.H. collected and analysed the data. Y.F. interpreted the results and wrote the manuscript. All the authors gave their final approval for publication.