Preparation of Magnetic Carboxymethylchitosan Nanoparticles for Adsorption of Heavy Metal Ions.

The remediation of metal and heavy metal contaminants from water ecosystems is a long-standing problem in the field of water management. The development of efficient, cost effective, and environmentally friendly natural polymer-based adsorbents is reported here. Magnetic chitosan (CS) and carboxymethylchitosan (CMC) nanocomposites have been synthesized by a simple one-step chemical coprecipitation method. The nanoparticles were assessed for the removal of Pb2+, Cu2+, and Zn2+ ions from aqueous solution. Kinetic and thermodynamic models were used to describe and understand the adsorption process of the ions onto the nanomaterials. The interactions between the ions and the biopolymer-based composites are reversible, which means that the nanoparticles can be regenerated in weakly acidic or EDTA containing solution without losing their activity and stability for water cleanup applications.

Introduction

The waste products of manufacturing processes of over three centuries of industrialization have had unintentional consequences, of which the long-term issues of environmental pollution, in both terrestrial and water systems, are widely reported. One major issue is the contamination of water ecosystems by toxic heavy metal ions; this has been a concern for many decades. The ability of metal ions to accumulate in the tissue of plants, fish, and mammals poses a serious threat to human health. This is both through bioaccumulation in the food chain and also because of their ability to bind to proteins and metabolites in living organisms.[1−4] Several techniques are available for heavy metal ion removal from water; typical technologies include ion-exchange, chemical precipitation, membrane filtration, electrochemical methods, and adsorption, to name only a few.[5] Among these, the use of biopolymers as adsorbents is especially attractive because of their low operational cost and high efficiency. This has contributed to the development of novel adsorbents in recent decades.[6,7] The use of biopolymer-based adsorbents, such as polysaccharides, offers biocompatibility and biodegradability advantages. This has led to their wider application in the water treatment industry.[8] Chitosan (CS) is a type of polyaminosaccharide synthesized from the deacetylation of chitin, a polysaccharide consisting predominantly of unbranched chains of β-(1 → 4)-2-acetoamido-2-deoxy-d-glucose. Chitin is the second most abundant polymer in nature after cellulose.[9] The amino and hydroxyl functional groups of CS are effective binding sites for metal ions. Carboxymethylchitosan (CMC) is reported to have a higher metal ion binding capacity than CS because of the increased chain flexibility and higher concentration of chelating groups.[10−13] Compared with conventional adsorbents, magnetic adsorbents offer the attractive benefit of allowing their straightforward separation and removal from water, with the use of an external magnetic field. In the current study, CS and CMC magnetite nanoparticles were prepared by a simple, one-step chemical coprecipitation method. The resulting nanocomposites were used as a tool for the rapid removal of heavy metal ions, such as copper (Cu(II)), lead (Pb(II)), and zinc (Zn(II)). The relationship between the adsorption capacity and the adsorption mechanisms was studied, and equilibrium data were analyzed using Langmuir and Freundlich isotherms.

Experimental Section

Materials

CS (low molecular weight), iron(III) chloride hexahydrate (FeCl3·6H2O), ethylene glycol (EG), and ammonium acetate (NH4Ac) were purchased from Sigma Aldrich. CMC was purchased from Santa Cruz Biotechnology, Inc. Pure water (18 MΩ cm) was obtained from a Milli-Q system (Millipore, Milford, MA, USA). Other chemicals were of analytical grade and were purchased from Sigma Aldrich.

Synthesis of CS and CMC Magnetic Nanoparticles

Colloidal nanoparticles were synthesized via a versatile solvothermal reaction reported by Fang et al.[14] with a modification. Briefly, 2.5 g of FeCl3·6H2O, 4 g of NH4Ac, and 1.2 g of CS or CMC were dissolved in EG (75 mL) aided by sonication for 30 min to form a clear solution. The solution was then transferred to a Teflon-lined autoclaved (100 mL) and heated at 200 °C for 12 h. After the reaction, the solution was cooled to room temperature and isolated with a magnet. The nanoparticles were thoroughly washed with distilled water and ethanol and then also stored in ethanol.

Adsorption Kinetic Studies

Aqueous solutions of heavy metal ions were prepared by dissolving PbCl2, CuCl2, and ZnCl2 in distilled water to reach a final cation concentration of 100 mg·L–1. Sorption studies were then conducted by mixing 20 mg of CS and CMC magnetic nanoparticles into 20 mL of aqueous solutions of the desired heavy metal ion. The pH of the solution was then adjusted to 5.2 ± 0.5 below the pKa of CS (∼6.2). This ensures that the amine groups are protonated and the adsorption of metal ions onto (carboxymethyl) chitosan is optimal.[15] The resulting dispersions were shaken mechanically at 200 rpm at room temperature, and the effect of contact time was evaluated between 2 and 60 min. After 2, 5, 10, 30, and 60 min, 1 mL of solution was collected and the nanoparticles were removed by magnetic separation. The ion concentration was then analyzed by inductively coupled plasma atomic emission spectroscopy (ICP-OES). The adsorption capacities of CS and CMC magnetic nanoparticles q (mg/g) were calculated by eq where C0 and Ct denote the initial and final concentration of heavy metal ions in solution (mg·L–1), respectively. V denotes the volume of aqueous solution (L), and w is the mass of CS and CMC magnetic adsorbents (mg).

Adsorption Isotherm Studies

Adsorption isotherm studies describe the relationship between adsorbents and adsorbates at equilibrium.[16] Adsorption isotherms of Cu(II), Pb(II), and Zn(II) on CS and CMC magnetic nanoparticles were obtained by dispersing 0.1 g of magnetic adsorbent into 100 mL of heavy metal ion solution, with concentrations ranging from 100 to 700 mg·L–1. Adsorption isotherm experiments were conducted at pH 5.2 ± 0.5. Dispersions were shaken mechanically at 200 rpm at room temperature, and 1 mL of solution was withdrawn after 24 h. The colloidal nanoparticles were removed by magnetic separation, and the ion concentration was then established using ICP-OES analysis.

Regeneration Studies

For desorption experiments, 20 mg of CS and CMC nanoparticles previously loaded with Cu(II), Pb(II), and Zn(II) were collected and gently washed with distilled water to remove any unabsorbed metal ion, and then, 20 mL of EDTA solution (0.01 mol·L–1) was added. The samples were then shaken mechanically at 200 rpm at room temperature for 24 h for desorption to occur. Finally, colloidal nanoparticles were removed by magnetic separation, and the ion concentration was again analyzed using ICP-OES.

Results and Discussion

Characterization of CS and CMC Magnetic Nanoparticles

The scanning electron microscopy (SEM) images of the CMC magnetic colloidal particles are given in Figure . The particles exhibit a uniform size of approximately 500 nm. Interestingly, the high-resolution transmission electron microscopy (HRTEM) images of the CMC magnetic particles (Figure ) show fringes at the edges of the particles. The appearance of the colloidal particles is consistent with that of an agglomeration of smaller Fe3O4 precursors.

Figure 1

SEM micrographs of CMC magnetic nanoparticles at low (a) and high (b) resolution.

Figure 2

(a–c) HRTEM micrographs of CMC magnetic nanoparticles.

The structures of CS and CMC magnetite nanoparticles were characterized by X-ray diffraction (XRD) (Figure a). Six peaks that are characteristic of Fe3O4, namely, (220), (311), (400), (422), (511), and (440) were observed for all samples (JCPDS no. 75-1610). To confirm the presence of CS and CMC, FT-IR spectra of nanoparticle composites were examined and are shown in Figure b. The CS spectrum shows a strong peak at 3231 cm–1, which is assigned to the N–H extension vibration and O–H stretching vibration, as well as belonging to intermolecular hydrogen bonding, within the biopolymer. The signals at 2977 and 2882 cm–1 were assigned to the vibration of methyl and methylene C–H bonds, respectively. The peaks in 1089 and 1046 cm–1 corresponded to the bridge-O-stretch and C–O stretch, respectively.[17,18] The difference between CS and CMC spectra was the peaks obtained solely in the spectra of CMC at 1631 and 1396 cm–1; this IR region is specific for the COO– carboxylate group, and the peaks were attributed to the carboxylate C=O asymmetric stretching and the C=O symmetric stretching.[19]

Figure 3

(a) XRD patterns of CS and CMC magnetite composite nanoparticles and (b) IR spectra of CS and CMC composite nanoparticles.

Figure shows the magnetization of CS and CMC magnetite nanoparticles as a function of the magnetic field at 300 K. The measurements were obtained with a vibrating sample magnetometer (VSM) with an applied magnetic field ranging from −10 to 10 kOe. The absence of any significant remanence or coercivity in the magnetization curves suggests that the samples can be classified as paramagnetic. Moreover, the magnetic saturation of CS and CMC magnetite nanoparticles is 13 and 15 emu/g, respectively. Such magnetic properties mean that the nanoparticles are susceptible to an external magnetic field and can easily be separated from solution.

Figure 4

Magnetic hysteresis loops of CS and CMC magnetic nanoparticles at 300 K

Kinetic Adsorption Study of Metal Ions on CS and CMC Nanocomposites

The adsorption performance of CS and CMC magnetic nanoparticles with metal ions was investigated by a batch equilibrium technique for contact times varying between 2 and 60 min. The results for Cu, Pb, and Zn ions are shown in Figure . The spontaneously high rate of metal ion uptake at short contact times was associated with the significant number of active sites available on the nanoparticles. However, as the contact times increase, metal ions progressively bind to these sites. The rate of sorption then slows down before reaching equilibrium after 60 min, as illustrated by the plateau on the curves in Figure . The adsorption capacity (qe) of CS nanoparticles was in the following order: Pb(II) 88.5 mg·g–1 > Cu(II) 79.3 mg·g–1 > Zn(II) 61 mg·g–1. The performance of CMC nanocomposites was higher, with the following maximum ion uptakes: Pb(II) 107.5 mg·g–1 > Cu(II) 99 mg·g–1 > Zn(II) 75.8 mg·g–1. The increased sorption capacity of CMC was attributed to the availability of a large number of −COO– groups, which are able to coordinate to metal ions. To further investigate the adsorption mechanism and determine the sorption efficiency of the nanocomposites, the kinetic process was studied by fitting the experimental data to the kinetic models, presented in Table . According to the R-squared values of the regression analyses presented in Table , the pseudo-second-order model offers the best mathematical fit to describe the adsorption kinetics of Pb(II), Cu(II), and Zn(II) onto CS and CMC nanoparticles. Such kinetic models imply that chemisorption is likely to be the rate-limiting step of the adsorption process.[20,21]

Figure 5

Effect of time on the adsorption of metal ions onto CS and CMC magnetic nanoparticles: (a) Cu(II), (b) Pb(II), and (c) Zn(II). The error bars represent standard errors.

Table 1

Mathematical Equations Applied in the Kinetic Adsorption Study of Metal Ions onto CS and CMC Nanoparticles[22]

kinetic models	linear equations	plot	calculated coefficient
pseudo-first-ordera		ln(qe – qt) vs t	K1 = −slope × 2.303
			qe = eintercept
pseudo-second-orderb		t/qt vs t
			qe = slope–1

K1 (min–1) and K2 (g·mg–1·min–1) denote the order rate constants for the pseudo-first-order; q is the adsorption amount of metal ions at time t (mg·g–1), and qe is the equilibrium adsorption amount (mg·g–1), whereas t gives the contact time (min).

K1 (min–1) and K2 (g·mg–1·min–1) denote the order rate constants for the pseudo-second-order; q is the adsorption amount of metal ions at time t (mg·g–1), and qe is the equilibrium adsorption amount (mg·g–1), whereas t gives the contact time (min).

Table 2

Kinetic Parameters for Cu(II), Pb(II), and Zn(II) Ion Adsorption

		pseudo-first-order			pseudo-second-order
metal	material	qe	K1	R2	qe	K2	R2
Pb(II)	CS nanoparticles	28.7	9.8 × 10–2	0.969	88.5	6.3 × 10–3	0.999
	CMC nanoparticles	20.58	1.0 × 10–1	0.871	107.5	1.0 × 10–2	0.998
Cu(II)	CS nanoparticles	43.2	9.7 × 10–2	0.964	79.3	3.2 × 10–3	0.995
	CMC nanoparticles	32.2	1.1 × 10–1	0.947	99.0	5.6 × 10–3	0.998
Zn(II)	CS nanoparticles	37.4	7.9 × 10–2	0.816	61.0	2.6 × 10–2	0.996
	CMC nanoparticles	42.0	8.4 × 10–2	0.901	75.8	2.7 × 10–3	0.997

Isotherm Adsorption Study

Figure shows the adsorption equilibrium isotherm data obtained for Cu(II), Pb(II), and Zn(II) ions adsorbed onto CS and CMC nanocomposites. These adsorption isotherms allow for the amount of metal ions adsorbed (per unit weight of nanocomposites) to be correlated to the remaining concentration of metal ions in the aqueous phase at a given temperature under equilibria conditions. This type of data set is highly valuable to aid the understanding of the mechanism of adsorption, as well as to interpret the kinetics of the adsorption process. Figure shows that the uptake capacity of the nanocomposites was found to increase with the equilibrium concentrations of the metal ions in solution, progressively reaching saturation of the adsorbent. The experimental data were fitted to the Langmuir and Freundlich models; these equations are shown in Table and are commonly used for analyzing the nature of adsorbate–adsorbent interactions.[23,24]

Figure 6

Adsorption isotherms of CS and CMC magnetic nanoparticles in (a) Cu(II), (b) Pb(II), and (c) Zn(II) aqueous solutions. The error bars represent standard errors.

Table 3

Mathematical Equations Applied in Adsorption Isotherm Study of Metal Ions onto CS and CMC Nanoparticles[25]

models	linear equations	plot	calculated coefficient
Langmuira		Ce/qe vs Ce
			qm = slope–1
Freundlichb		log qe vs log Ce	KF = 10intercept
			n = slope–1

qe and Ce denotes the amount adsorbed (mg·g–1) and adsorbate concentration in solution (mg·L–1) at equilibrium, respectively. KL is the Langmuir constant (L·mg–1), and qm denotes the maximum adsorption capacity of the monolayer formed on the adsorbent.

Kf and n denote the Freundlich isotherm constants, indicating the adsorption capacity (mg·g–1) and adsorption intensity (dimensionless), respectively.

According to the R-squared values of the regression analyses presented in Table , both Langmuir and Freundlich models fitted the experimental isotherm equilibrium data reasonably well, with R2 values >0.927. According to these values, the Langmuir equation provided the best fit for the experimental data of the CS magnetic nanoparticles (for the three metal ions tested) in the entire range of concentrations studied. The Langmuir model assumes that the adsorbent surface has a finite number of binding sites of identical energy and that each adsorbate ion is located at a single site. The model suggests the formation of a monolayer of metal ions on the surface of CS nanoparticles. The maximum adsorption capacity (qm) of CS nanocomposites was Pb(II) 141 mg·g–1 > Cu(II) 123 mg·g–1 > Zn(II) 88 mg·g–1. For the CMC nanocomposites, the maximum adsorption capacity was significantly higher for the three metal ions in the order Pb(II) 243 mg·g–1 > Cu(II) 232 mg·g–1 > Zn(II) 131 mg·g–1. The Freundlich model was found to be more accurate when describing the adsorption of metal ions on CMC nanocomposites. Freundlich isotherms represent an empirical model employed to describe the equilibrium on a heterogeneous surface; they do not assume a monolayer capacity as in the case of the Langmuir model.[26] Freundlich isotherms suggest the presence of different binding sites on the adsorbent, which is in agreement with the chemical structure of CMC. Both carboxylate −COO– and amino −NH2 groups can participate in the metal-ion binding process. The value of 1/n (Freundlich model) can be used to predict the binding affinity between the sorbent and the sorbate[27] with small values, implying stronger interactions between sorbents and metal ions. From Table , the values of n for Pb(II) are larger than those for Cu(II) and Zn(II), suggesting stronger binding interactions between Pb(II) and the active sites of CMC nanocomposites.

Table 4

Isotherm Parameters of Metal Ion Adsorption on CS and CMC Nanocomposites

		Langmuir isotherm			Freundlich isotherm
metal	material	qm	KL	R2	1/n	Kf	R2
Pb(II)	CS nanoparticles	141	1.6 × 10–2	0.996	0.27	23.9	0.972
	CMC nanoparticles	243	1.9 × 10–2	0.983	0.26	45.1	0.995
Cu(II)	CS nanoparticles	123	9.0 × 10–3	0.990	0.39	9.3	0.939
	CMC nanoparticles	232	6.3 × 10–3	0.971	0.44	11.3	0.993
Zn(II)	CS nanoparticles	88	6.0 × 10–3	0.988	0.47	3.5	0.927
	CMC nanoparticles	131	2.3 × 10–3	0.977	0.63	1.4	0.990

The adsorption capacity of the synthesized CS and CMC magnetic nanoparticles toward heavy metal ions is compared with those reported, as shown in Table . The maximum capacity of CMC magnetic nanoparticles is certainly comparable and possibly even higher than that of other CS-based adsorbents for the three metal ions tested. This is with the exceptions of CS hydroxyapatite nanocomposite beads and CS-coated mesoporous calcium silicate hydrate microspheres because of the high specific surface area of the microspheres.

Table 5

Comparison of Adsorption Capacities of CS and CMC Magnetic Nanoparticles Reported Herein toward Heavy Metal Ions with Those Reported in the Literature

	adsorption capacity (mg·g–1)
adsorbent	Pb2+	Cu2+	Zn2+	references
CS mesoporous microspheres	796	425	400	(28)
CS magnetite beads	63			(29)
CS multifunctional nanocomposite	70	66		(30)
magnetic CS/cellulose microspheres	46	88		(31)
CS hydroxyapatite nanocomposite	1385			(32)
CS rectorite nanocomposite		21		(33)
CS and CMC magnetite nanoparticles	243	232	131	This work

Desorption experiments are of paramount importance for the recovery of the metal ions to help understand the adsorption process of the adsorbates. In the current study, the nanoparticles were regenerated by stripping the sorbed metal ions using EDTA chelation; such a procedure has previously been found to desorb metal ions from CS in a superior manner to using a solution of HCl.[33−35]Figure shows the regeneration of the nanocomposites over three successive cycles of adsorption and desorption. The results show that the magnetic nanocomposites release the adsorbed metal ions with regeneration efficiency above 60% based on the adsorption capacity in each cycle. Despite the overall adsorption capacity reducing slightly after each cycle, the adsorption capacity of the magnetic nanoparticles remains high, which suggests a limited number of irreversible sites on the surfaces of the synthesized nanocomposites that are not recovered when using the EDTA desorption solution. Interestingly, the capacity loss observed was larger for nanoparticles loaded with Pb(II) and Cu(II) than that for nanoparticles loaded with Zn(II). It is likely that the higher binding affinity between the sorbent and Pb(II) and Cu(II) highlighted in the isotherm adsorption studies somewhat limits the stripping effect of the EDTA solution, which tends to moderately reduce their loading ability after successive regeneration cycles.

Figure 7

Adsorption–desorption cycles of (a) Cu(II), (b) Pb(II), and (c) Zn(II) onto CS and CMC magnetite nanocomposites.

Conclusion

CS and CMC magnetic nanocomposites were synthesized using a simple one-step chemical coprecipitation method. The nanoparticles provide an efficient, fast, and environmentally friendly approach for the removal of the metal ions Cu(II), Pb(II), and Zn(II) from aqueous solutions. The current work shows that the binding capacity of the nanoparticles to metal ions is comparable or even higher than that of other CS-based adsorbents. In addition to the amine groups, the presence of a carboxylic moiety on CMC nanoparticles is likely to offer additional binding sites, which considerably enhance their loading capacity in comparison to the CS nanoparticles. This mechanism is in agreement with the good fitting of the experimental data with the Freundlich adsorption isotherm. Moreover, the magnetic properties of the nanoparticles when coupled to their regenerative performances make them a potential, efficient, and cost-effective tool for heavy metal ion removal for water treatment industries.