1. Introduction

Colored organic effluent is produced in different industries such as textile, rubber, paper, plastic and cosmetics. The colored effluents of wastewater from these industries can be mixed in surface and ground water systems, and may transfer to drinking water and bring a chief threat to the human health due to their toxic or mutagenic and carcinogenic effects [1, 2]. Therefore, colored wastewater cannot be discharged without adequate treatment and dye removal. Physico-chemical processes such as coagulation and flocculation [3], membrane separation [4], oxidation [5], electro-coagulation [6], and adsorption on activated carbon and clays [7] were employed for the treatment of dye-containing wastewater. Among these methods, adsorption process is considered to be promising technology. Activated carbon is the most widely used commercial adsorbent due to its excellent adsorption capacity [8]. However, the high cost and the poor regeneration are the main drawbacks for the use of activated carbon as adsorbent [9]. Many studies were undertaken to find low cost sorbents such as cotton, fly ash, guava leaf powder, sugarcane bagasse pith, rice husk, and coconut coir [10-16].

Recently, most researchers concentrated towards natural polymeric materials, because they are renewable, biodegradable, non-toxic, and hence they are considered environment friendly materials [17]. Among the natural polymeric materials, chitosan (Cs) is the most attractive polymer due to its low cost and availability. Cs is a linear copolymer composed of (1–4)-linked D-glucosamine and N-acetyl-D-glucosamine and it is obtained by alkaline deacetylation of chitin (the second abundant polymer in nature after cellulose) [18]. Cs was widely used for the removal of heavy and transition metals [8, 19, 20] due to its high number of functional amines and hydroxyl groups, thus enabling the adsorption of anionic dyes through electrostatic attraction. However, Cs is soluble in dilute organic acids and is difficult to be separated from the solution using traditional separation methods such as filtration and sedimentation. To overcome such a problem, various physical and chemical modifications were developed to improve the chemical stability of Cs in acidic media. Recently, many researchers were interested in the chemical modification of Cs by cross-linking method using cross-linking agents such as glutaraldehyde [21, 22], epichlorohydrin (ECH) [23, 24] and ethyleneglycol diglycidyl ether [25, 26]. Although the crosslinking methods were found to enhance the resistance of Cs against acids, it reduced the adsorption capacity of Cs. To overcome this problem, the crosslinked Cs can be modified with Polyoxyalkylene diamines to increase the adsorption capacity.

The aim of the present work is directed to prepare a crosslinked Cs followed by modification with polyoxypropylenediamine (D₂₀₀₀). Resultant D₂₀₀₀-Cs has the advantage of improving the chemical stability of Cs in acidic media. D₂₀₀₀-Cs was characterized using by X-ray diffraction (XRD), Fourier transform infrared (FT-IR), scanning electron microscopy (SEM) and thermal gravimetric analysis (TGA). Afterward, D₂₀₀₀-Cs was tested for the adsorption of anionic dye, acid green 25 (AG) from acidic aqueous solutions. The effect of physicochemical parameters, that influence the adsorption process such as the initial concentration of AG, pH and temperature, were also investigated.

2. Experimental

2.1. Materials

Cs with a molecular weight of 100,000-300,000 and 90% degree of deacetylation (Acros, USA), ECH 99% (Acros, USA), Polyoxypropylene-diamine (D₂₀₀₀) with molecular weight of 2000, primary amine content is 0.97 meq g^-1 (Huntsman corporation TX, USA) were used without further purification. AG (Aldrich, UK), acetic acid and methanol (Adwic, EGYPT) and extra pure NaOH pellets (Lobachemie, INDIA) were used as received.

2.2. Preparation of D₂₀₀₀-Modified Cs (D₂₀₀₀-Cs)

The preparation of D₂₀₀₀-modified Cs (D₂₀₀₀-Cs) was carried out by the following steps:

2.2.1. Preparation of Cs Gel

The Cs gel was prepared using the previously reported method by Huang et al. [27]. Accordingly, a solution of Cs was prepared by dissolving 1 g of Cs in 33.3 ml of 2% acetic acid followed by drop wise addition of Cs solution to an alkaline precipitation bath (50 mL of 0.5 M NaOH aqueous solution). Afterward, the gelatinous precipitate of Cs was collected and washed for several times by distilled water then by ethanol to remove residual NaOH. Finally, the precipitate was stored in distilled water for further use.

2.2.2. Preparation of Crosslinked Cs

The Cs gel obtained from the previous step was suspended in 100 mL of a mixture of water–methanol (v/v, 1:1). Afterward, an excess amount of ECH (2 mL) was added drop wise to the suspension solution then the mixture was stirred at 60 °C for 24 h and cooled. Finally, the product was filtered and washed three times by water and ethanol to give crosslinked Cs (ECH-Cs) with nitrogen content 1.24 wt%.

2.2.3. Preparation of D₂₀₀₀-Modified Cs (D₂₀₀₀-Cs)

D₂₀₀₀-Cs was prepared by suspending crosslinked Cs in 100 mL methanol/water mixture (1:1, v/v) followed by addition of 7.63 g D₂₀₀₀. The reaction mixture was kept at 60°C with stirring for 24 h. Then, the product was filtered, washed three times by water and ethanol to remove the unreacted D₂₀₀₀ and dried at 60°C. The product was finally grinded (Nitrogen content 4.31wt %).

2.3. Solubility Test

The solubility of D₂₀₀₀-Cs was examined in acidic medium by adding 5 mg of the sample into 10 mL HCl (pH 2.3) for 24 h then the sample was reweighed.

2.4. Swelling Test

The D₂₀₀₀-Cs sample was weighed in a sintered glass, Wₒ and then immersed in buffer solution (pH = 3, 5.4 and 7.8) at 25°C. After immersion, the sample was centrifuged and reweighed immediately, W_t. The process was repeated until W_t becomes constant. The swelling degree of the new material expressed as the amount of adsorbed buffer per 100 g of dry sample, the swellability percentage was calculated using the formula:

(1)

2.5. Adsorption Batch Experiments

A stock solution of AG dye (186.8 mg L^-1) was prepared in freshly distilled water, and the solutions of desired concentrations (18, 23, 28, 31 and 36 mg L^-1) were obtained by successive dilution. Calibration curve was constructed by measuring the absorbance of the AG solutions at λ_max = 642 nm.

Adsorption experiments were carried out using 0.02 g of D₂₀₀₀-Cs and ECH-Cs as adsorbents. For each experiment, 50 mL of AG dye solution (18 mg L^-1) along with the adsorbent were stirred magnetically (400 rpm) for 2, 5, 10, 15, 20, 25, 30, 35, 40, 50, 60 min at 25°C. The pH of the dye solution was adjusted to 3 for all the adsorption kinetic studies and equilibrium adsorption studies using 0.1 M HCl or 0.1 M NaOH. The samples of the dye solution, during the course of adsorption, were collected at regular intervals of time and centrifuged to remove the adsorbent.

The absorption spectra of dye solutions before and after adsorption were recorded in the range of 750 – 500 nm. The concentration of dye solutions were determined at a wavelength of λ_max = 642 nm. The quantity of adsorbed dye Q_e (mg g^-1) and the removal efficiency (R%) were calculated from the following equations:

(2)

(3)

where, C_o is the initial dye concentration in liquid phase (mg L^-1), C_e is the liquid phase dye concentration at equilibrium (mg L^-1), V is the volume of dye solution (L), and m is the mass of adsorbent (g).

Kinetic analysis is required to obtain an insight of the rate of adsorption and the rate limiting step of the transport mechanism, which are primarily used in the modeling. In order to evaluate the kinetic mechanism that controls the adsorption process, the pseudo first order [28], pseudo second order [29, 30], and intraparticle diffusion [31] models were tested. The pseudo first order rate equation:

(4)

The pseudo second order reaction:

(5)

The intraparticle diffusion model [32]:

(6)

where, Q_t (mg g^-1) is the amount of dye adsorbed at time t (min), k₁ is the rate constant for pseudo first order, k₂ is the rate constant of the pseudo second order mode (g mg^-1 min^-1), k_i is the intraparticle diffusion rate parameter (mg g^-1 min^-1/2) and C is the intercept, which is proportional to the boundary layer thickness (mg g^-1). Good correlation with the kinetic data explains the dye adsorption mechanism in the solid phase.

The isothermal kinetics of the adsorption were described by isotherm models of which Langmuir [33], Freundlich [34], and temkin [35] are the most widely used equations.

The Langmuir adsorption isotherm in the linear form is expressed as:

(7)

The Freundlich isotherm is expressed as:

(8)

The Temkin isotherm is given by:

(9)

where k_l and a_l are the Langmuir adsorption constants, k_f is the Freundlich adsorption constant, which roughly indicates the adsorption capacity of the adsorbent and n is the Freundlich constant, K_T is the equilibrium binding constant and B is a constant that is related to the heat of adsorption.

2.6. Instruments

The UV visible absorption spectra were measured using spectrometer, UVD-2960 (Labomed. INC). Fourier-transform infrared (FT-IR) spectra were recorded using Bruker, Tenson 27 FTIR spectrophotometer with frequency range of 4000 – 400 cm⁻¹ using KBr disk technique. X-ray diffraction (XRD) analysis was measured by GNR, APD 2000 PRO step scan X-ray diffractometer, Cu Kα radiation (l.54 Å), (40 KV and 30 mA) at a step per second of 0.02°. Thermal gravimetric analysis (TGA) was carried out using Perkin Elmer, STA 6000. The morphology of crosslinked Cs and D₂₀₀₀-Cs was studied by scanning electron microscope (SEM JEOL, JSM-5200LV) at an acceleration voltage of 25 KV. Quantitative elemental analysis for N in crosslinked Cs, D₂₀₀₀-Cs was performed on Heraeus (microanalysis center, Cairo University, Giza, Egypt).

3. Results and Discussion

3.1. Synthesis and Characterization of D₂₀₀₀-Cs

The aim of the present work is to produce a new material with high adsorption capacity and chemical stability in acid medium. To achieve this, Cs was dissolved in acetic acid to yield hydrogel in water. After that, Cs beads were crosslinked by using excess amount of ECH. Finally, the crosslinked Cs was modified by D₂₀₀₀. Similar modification was reported using ethylenediamine [27]. The prepared D₂₀₀₀-Cs showed satisfactory chemical stability in acid medium which was confirmed by its insolubility in strong acid media (pH = 2.3), however, Cs was completely soluble in weak acid medium.

The FT-IR spectrum of Cs (Figure. 1 (a)) shows a band at 3443 cm^-1 due to the axial stretching vibration of OH superimposed to NH₂ stretching and inter- and intramolecular hydrogen bonding of Cs. The bands at 2921 and 2878 cm^-1 are due to C–H asymmetric stretching vibration of CH₂ and C–H symmetric stretching vibration of CH₃ groups, respectively [36]. The band at 1646 cm^-1 is assigned to C=O stretching of amide group (NHCOCH₃) due to partial deacetylation of chitin. The bands at 1603, 1460 and 1426 cm^-1 are attributed to bending vibration of NH₂, CH₂, CH₃, respectively. The sharp peak at 1384 cm^-1 is related to C–N stretch of amide group and CH₃ bending. The band at 1325 cm^-1 refers to CH₂ scissoring and OH bending vibration ²⁹. The bands at 1259 and 1155cm^-1 are due to C–N and the bridge C–O–C stretching vibrations, respectively, while the peaks at 1082 and 1028 cm^-1 refer to C–O and CH₂OH stretching vibrations, respectively. The peak at 896 cm^-1 is attributed to wagging of saccharine structure of Cs [37].

The FT-IR spectrum of the ECH-Cs, Figure. 1 (b) shows all the significant characteristic bands of Cs with some shifts in the position of bands. The same functional groups of the ECH were present in the Cs. Therefore, the same vibrations were observed, but with different intensities. The peak at 1646 cm^-1 related to C=O stretch of amide group shifted to 1639 cm^-1. In addition, the band at 1603 cm^-1 of NH₂ group disappeared which is a spectroscopic evidence for the covalent bonds formation between ECH and NH₂ groups of Cs. The peak at 1259 cm^-1 related to C-N stretch became weak and shifted to 1255 cm^-1 proving the introduction reaction of ECH on NH₂ position of Cs [37].

Significant changes were observed in the spectrum of D₂₀₀₀-Cs presented in Figure. 1 (c). A band at 1603 cm^-1 that is attributed to NH bending in primary amine groups (NH₂) was observed. The band at 2921 cm^-1 related to asymmetric stretching vibration of C-H of CH₂ group shifted to 2973 cm^-1 due to the insertion of D₂₀₀₀ into Cs. In addition, the bands at 1460 and 1384 cm^-1 that are related to CH₂ bending and C–N stretching vibrations of amide group became more intense and shifted to 1456 and 1375 cm^-1, respectively. Furthermore, two new peaks appeared at 1301 and 1261 cm^-1 due to the primary and secondary amino groups, which verify the grafting of D₂₀₀₀.

Figure 1. FTIR spectra of Cs (a), ECH-Cs (b) and D2000-Cs (c)

The XRD pattern shown in Figure 2 (a) exhibits the characteristic crystalline peaks of Cs around 2θ of 10 and 20° [38-40]. After reaction with ECH, the peak at 2θ = 10° showed a pronounced disappearance and the peaks at 2θ = 10° and 20° became more broad due to the decrease in the crystallinity after crosslinking with ECH since amino groups were occupied [41]. In D₂₀₀₀-Cs, a broad peak at 2θ = 20° was observed concomitant with the complete disappearance of the peak at 2θ = 10° due to the decrease in crystallinity, which indicates that Cs loses its structure registry due to the modification.

Figure 2. XRD spectra of Cs (a), ECH- Cs (b) and D2000-Cs (c)

Figure. 3 shows the TGA curves of pure Cs, ECH-Cs and D₂₀₀₀-Cs in the temperature range from 41 to 800°C. Weight losses of pure Cs occurred at three stages [42, 43]. The first stage started from 41 to 100°C with weight loss of 7.7 wt%. The second stage started from 250°C with loss of 42 wt% while, the third stage occurred from 350°C with weight loss of 47 wt%. The second and third stages correspond to the degradation of Cs chains and decomposition of pyranose rings with ring opening reaction respectively. Compared with the TGA curve of pure Cs, there are obvious alterations in crosslinked Cs and D₂₀₀₀-Cs. The degradation temperature of crosslinked Cs shifted to lower temperature (200°C). This indicates that the reaction of Cs with ECH decreases the intermolecular and the intramolecular hydrogen bonding existed in Cs and hence decreases the thermal stability. However, in the D₂₀₀₀-Cs, much weaker action between Cs and water molecules adsorbed in the sample was observed leading to a more stable material in the range from 41-210°C small weight loss. The high weight loss at 300-600°C could be attributed to the degradation of D₂₀₀₀.

Figure 3. TGA of Cs (a), ECH- Cs (b) and D2000-Cs (c)

The surface morphology of crosslinked Cs and D₂₀₀₀-Cs were observed by SEM as shown in Figure. 4. The crosslinked Cs shows flaky structure however, D₂₀₀₀-Cs shows accumulation of spherical domains with irregular surface morphology. This spherical morphology provides large surface area for adsorption.

Figure 4. SEM of ECH- Cs (a) and D2000-Cs (b)

The modification of ECH-Cs with D₂₀₀₀ lead to increase in the N content from 1.24 wt% to 4.31 wt% as indicated from the elemental analysis. The increase in N content caused an increase in the adsorption capacity of the adsorbent. In addition, it would be expected that the elasticity of D₂₀₀₀ favors the adsorption capacity of modified Cs. The D₂₀₀₀-Cs was found to be insoluble in acid medium in comparison to the native Cs. The swelling of D₂₀₀₀-Cs was evaluated to be 248%, 137% and 124% at pH 3, 5.4 and 7.8, respectively. These values confirm its stability in buffer solution (no weight loss).

3.2. Adsorption of AG onto D₂₀₀₀-Cs

The D₂₀₀₀-Cs can be utilized as an ideal material for acidic wastewater treatment that includes anionic dyes. To investigate the effect of crosslinking and modification by D₂₀₀₀ on the adsorption of dyes, experiments were conducted for adsorption at pH = 3 from 50 mL of 18 mg L^-1AG with 20 mg of ECH-Cs and D₂₀₀₀-Cs.

Figure. 5 shows the adsorption capacity against time using D₂₀₀₀-Cs and ECH-Cs. It is clear that D₂₀₀₀-Cs exhibited much higher adsorption capacity as compared to that of crosslinked Cs. The amount of dye adsorbed per unit mass (Q_e) of D₂₀₀₀-Cs and crosslinked Cs were, 44.5 and 6.5 mg g^-1 with adsorbtivity (R%) equals 99.4 and 14.4%, respectively showed in table 1. This may be attributed to the high surface area of D₂₀₀₀-Cs and the large number of binding sites (NH₂ groups), which lead to high interaction between AG and the D₂₀₀₀-Cs. In addition, the loading of D₂₀₀₀ improves the adsorptivity. Amine groups can significantly uptake the AG in acid medium through electrostatic interaction between NH₃⁺ and O^- in AG.

Figure 5. Adsorption capacity of ECH- Cs (a) and D2000-Cs (b)

Table 1. Adsorption capacity (Qe), adsorptivity (R%) of D2000-Cs and ECH-Cs

3.3. Adsorption Kinetics

The kinetic analysis, in order to evaluate the kinetic mechanism that controls the adsorption process, were tested by applying the pseudo first order [28], pseudo second order [29, 30] and intraparticle diffusion [31]. Figure. 6 (a-c) shows the corresponding linear plots. The parameters, k₁, Q_e , K₂ and the correlation coefficient (R²) were determined and presented in Table 2. The results of the regression analysis showed that the adsorption of AG dye onto D₂₀₀₀-Cs was best described by pseudo-second order with the highest correlation coefficients.

Table 2. Kinetic parameters for the adsorption of AG onto D2000-Cs

Figure 6. The pseudo first order (a), pseudo-second-order (b) and Intraparticle diffusion kinetic plots for adsorption of AG onto D2000-Cs

3.4. Adsorption Isotherms

The adsorption isotherms of the adsorption of AG into D₂₀₀₀-Cs were studied. The most common sorption models used to fit the experimental data are Langmuir [33], Freundlich [34], and Temkin [35], are showed in Figure. 7 and the correlation coefficient (R²) was calculated and reported in Table 3. For the three studied systems, R² of Langmuir isotherm is more near to unity and is better than R² of Freundlich and Temkin isotherms. Hence, the Langmuir model is the most widely used for the sorption of AG onto D₂₀₀₀-Cs.

Table 3. Isotherm parameters of adsorption of AG on D2000-Cs

Figure. 7 (d) shows the variation of separation factor (R₁) with initial AG concentrations (C_o). The results exhibited that R_l values were in the range of 0–1, indicating that the adsorption AG onto D₂₀₀₀-Cs is favorable.

Figure 7. Langmuir isotherm model (a), Freundlich isotherm model (b), Tempkin isotherm model (c) and separation factor versus initial AG concentration (d)

3.5. The Effect of pH

The pH of the dye solution plays an important role in the whole adsorption process, particularly affecting the adsorption capacity. Figure. 8 shows that the amount of dye adsorbed onto D₂₀₀₀-Cs increased with decreasing pH of solution. At lower pH, more protons will be available for the protonation of NH₂ groups in D₂₀₀₀ or Cs leads to increasing the electrostatic attraction force between D₂₀₀₀-Cs and dye molecules. At high pH, the protonation of the amino groups decreases and leads to decrease adsorption [44]. The values of Q_e for the adsorption of AG dye solution onto D₂₀₀₀-Cs at pH = 3, 4.5, 7 and 9 were 40.5, 32, 28 and 16 mg g^-1, respectively.

Figure 8. Adsorption of AG onto D2000-Cs at different pH

3.6. The Effect of Initial Dye Concentration

Figure. 9 shows the effect of initial dye concentration (C_o) on the adsorption kinetics of the D₂₀₀₀-Cs. The initial pH of dye solution was adjusted to 3. The values of Q_e for the adsorption of AG dye solution (50 ml) with concentrations 18, 23, 31 and 36 mg L^-1 using D₂₀₀₀-Cs (0.02 mg) are 44.5, 55.4, 69 and 84 mg g^-1, respectively. Figure. 9 shows that the adsorption capacity increases with increasing C_o.

Figure 9. Effect of initial concentration on the adsorption of AG by D2000-Cs

3.7. The Effect of Temperature

The effect of temperature on the adsorption capacity of AG onto D₂₀₀₀-Cs was studied at pH = 3 and the results are shown in Figure. 10. The adsorption capacity increased with increasing temperature from 25 to 50°C, which indicates the endothermic nature of the adsorption process.

Figure 10. Effect of temperature on the adsorption of AG onto the D2000-Cs

4. Conclusions

Cs was modified by using polyoxypropylenediamine (D₂₀₀₀) to obtain (D₂₀₀₀-Cs). The product was tested for the adsorption of anionic dye, acid green 25 (AG), from acidic aqueous solutions and shows good stability. The adsorption ability of D₂₀₀₀-Cs for removal of anionic dyes in acid medium is efficient and overcomes the drawbacks of using Cs as adsorbent due to its solubility in acid medium.

The initial concentration of AG, pH and temperature significantly affect the adsorption process and the adsorption capacity of D₂₀₀₀-Cs. The Langmuir model agrees very well with the equilibrium isotherm and the pseudo-second order describe the equilibrium kinetics.