The FTIR spectra of CTS, TCCTS template before and after Fe(III) removal are presented in Figure 1. As compared with CTS, some new bands appeared in the spectrum of the TCCTS template. The new band near 1,556 cm⁻¹ is the vibration adsorption peak of the thiourea moiety C–N group. The characteristic adsorption peak of thiourea, which is assigned to the N–C=S group, appears at about 1,400 cm⁻¹ [18]. Moreover, obvious changes in the spectra of TCCTS template before and after Fe(III) removal were observed. The new band around 1,590 cm⁻¹ in the spectrum of TCCTS template before Fe(III) removal disappeared in the spectrum of TCCTS template after Fe(III) removal, at the same time, the intensity of the band at 1,358 cm⁻¹ was increased in the spectrum of TCCTS template after Fe(III) removal. This indicated that Fe(III) was removed in the final step of the synthesis process.

Figure 2 shows the SEM pictures of CTS and TCCTS template. From which we could see that the surface of CTS was relatively smooth and the structure of CTS was compact. The surface of TCCTS template was rough, there were a lot of cavities in its structure and the structure was reticular and incompact. This three dimensional structure of TCCTS template is favorable for metal cation adsorption [19].

As shown in Figure 3, the adsorption of Fe(II) and Fe(III) increased with increasing pH when pH < 5.0. Chitosan and its derivatives adsorb metal cations mainly by chelation with amine groups [20], at lower pH, the adsorption of Fe(II) and Fe(III) decreased because some amine groups were protonated to form –NH₃⁺, reducing the number of binding sites available for the adsorption. When the pH value increased, the adsorption of Fe(II) and Fe(III) increased due to the decreasing number of protonated amine groups and the resulting increase in the number of binding sites. The maximum adsorption for both Fe(II) and Fe(III) on the TCCTS template appear at pH 5.0. The adsorption capacity of Fe(III) is higher than that of Fe(II), mainly because the TCCTS template has a memory effect for Fe(III). In the higher pH range, Fe(II) and Fe(III) precipitation occurred, which resulted in the adsorption of Fe(II) and Fe(III) decreasing.

Figure 4 shows that the adsorption of Fe(II) and Fe(III) increased with increasing contact time and reached equilibrium at about 60 min on the TCCTS template. In order to investigate the adsorption process kinetics, the pseudo first-order and pseudo second-order kinetic models were applied in this study. The pseudo first-order model is expressed as [21]:

where q_e and q_t (mg/g) are the amounts of Fe(II) and Fe(III) adsorbed on TCCTS template at equilibrium and at time t, respectively, and k₁ is the pseudo first-order rate constant (min⁻¹) of adsorption. The rate constant, k₁ and correlation coefficient, R² were determined by plotting the log(q_e − q_t) versus t. The pseudo second-order model is expressed as [22]:

where k₂ is the pseudo second-order rate constant(g·mg⁻¹·min⁻¹) of adsorption. The rate constant, k₂ and correlation coefficient, R² were determined by plotting the t/q_t versus t. The kinetic models for Fe(II) and Fe(III) adsorption are shown in Figure 5 and Figure 6. The parameter values of the kinetic models are presented in Table 1. According to Figure 5 and Figure 6 and based on the correlation coefficient in Table 1, the pseudo second-order model could better describe the adsorption of Fe(II) and Fe(III) on the TCCTS template than the pseudo first-order model. This suggests that the rate-limiting step may be chemical adsorption [8]. In many cases, the pseudo second-order model correlates well to the adsorption of metal cation on chitosan and its derivatives [23].

Figure 7 shows the adsorption isotherms of Fe(II) and Fe(III) on the TCCTS template. The Langmuir and Freundlich equations were applied to experimental data in Figure 7 to examine the relation between sorption and metal ion concentration at equilibrium. The Langmuir model, which is widely used for monolayer sorption on a surface, is presented as:

where Q_e (mg/g) is the adsorption capacity of Fe(II) and Fe(III) at equilibrium concentration, Q (mg/g) is the maximum adsorption capacity, C_e (μg/mL) is the equilibrium concentration of Fe(II) and Fe(III), b (mL/μg) is the Langmuir constant. Q and b can be calculated by plotting C_e/Q_e versus C_e. For Langmuir model, it is estimated by a dimensionless separation factor whether the sorption is favorable or not. The separation factor, R_L is defined as:

where C₀ (ug/mL) is the initial concentration of Fe(II) and Fe(III), b (mL/μg) is the Langmuir constant. Values of 0 < R_L < 1 indicates that the sorption is favorable. The values of R_L in this study lie in the range of 0.15 and 0.42, 0.09 and 0.29 for Fe(II) and Fe(III), respectively, which shows that the adsorption of Fe(II) and Fe(III) on the TCCTS template are favorable.

The Freundlich model, which is widely used for sorption on a heterogeneous surface, is given by:

where K_F and n are Freundlich constants related to adsorption capacity and intensity, respectively. K_F and n can be determined from a linear plot of logQ_e versus logC_e. The constants of the two models along with correlation coefficient (R²) values are presented in Table 2. It is found that the Langmuir model fit the data better than the Freundlich model, which indicates that the adsorption of Fe(II) and Fe(III) on the TCCTS template is a type of monolayer sorption.

The adsorption capacity obtained from binary mixtures is presented in Table 3. The data showed that the TCCTS template was effective in selective adsorption of Fe(III) from solutions containing other metal ions.

Chitosan (deacetylation degree 90%) and thiourea were purchased from Shanghai National Reagent Company. The other reagents are all of analytical grade and provided by the chemistry laboratory of JiuJiang University. 0.1 mol/L HCl and 0.1 mol/L NaOH were used to control the pH values of the solutions, 1 g/L Fe(II) and Fe(III) stock solution were prepared by dissolving the appropriate amount of FeCl₂ and FeCl₃·6H₂O in doubly distilled water, which was used throughout the entire experiment. Iron was determined on a FAAS model AA6300C (Shimadzu, Kyoto, Japan) with a iron hollow cathde lamp and a deuterium background correction. Its operating conditions are given in Table 4. pH values was measured on a pH meter model PHS-3C(Shanghai Precision Instrument Company, Shanghai, China). IR spectrum of the product was performed on an infrared spectrometer model Vertex70 (Bruker, Germany) with KBr disc method. The SEM image was performed on a SEM model Vega ΙΙ (Tescan, Czech).

The process of preparation included the following steps (Scheme 1):

The effect of pH on adsorption of Fe(II) and Fe(III) was studied in pH range 1.0–7.0 at 25 °C by shaking dry TCCTS template (10 mg) with Fe(II) and Fe(III) solution (100 mL, 6 μg/mL) for 60 min at 300 rpm. The desired pH was adjusted using 0.1 mol/L HCl and 0.1 mol/L NaOH. After filtration, the concentration of Fe(II) and Fe(III) in solution was determined by FAAS.

Kinetic studies were conducted by placing TCCTS template (10 mg) in a 250 mL flask containing Fe(II) and Fe(III) solution (100 mL, 6 μg/mL) at pH 5.0 and 25 °C. The mixture was stirred by a magnetic stirrer at 300 rpm. Samples of solution (10 mL) were withdrawn at scheduled time intervals after filtration and analyzed for Fe(II) and Fe(III) concentration.

At 25 °C, a series of different concentrations of Fe(II) and Fe(III) standard solutions (100 mL) were prepared. The pH of the solution was adjusted to 5.0. TCCTS template (10 mg) was added into solution and the solution was stirred at 300 rpm for 60 min, then filtered. After filtration, the concentration of Fe(II) and Fe(III) was determined by FAAS. The adsorption capacity was calculated according to:

where Q_e (mg/g) is the adsorption capacity, C₀ (μg/mL) is the initial concentration of Fe(II) and Fe(III), C_e (μg/mL) is the equilibrium concentration of Fe(II) and Fe(III), V (mL) is the volume of the solution of Fe(II) and Fe(III), W (mg) is the weight of TCCTS template added.

The selective adsorption of Fe(III) from binary mixtures with Pb(II), Zn(II), Cd(II), Ni(II) (0.01 mol/L for each metal ion) was carried at pH 5.0. Dry TCCTS template (0.1 g) was added into the binary mixture solution and the other adsorption conditions is the same as described in Section 3.3. After adsorption equilibrium, the concentration of each metal ion was determined by FAAS.