Facile preparation of amidoxime-functionalized fiber by microwave-assisted method for the enhanced adsorption of chromium(VI) from aqueous solution

Abstract

In this study, a facile and highly efficient approach, the microwave-assisted (MW-aid) method, was applied for the synthesis of amidoxime-functionalized fibrous adsorbent (PAN_MW-AO fibers), which exhibited enhanced adsorption capacities for Cr(VI) in aqueous solution. The preparation condition was optimized under four independent variables including: MW power, mass of NH₂OH·HCl, time and bath ratio, based on the Box-Behnken design with response surface methodology (RSM). A period of 1 × 5 min was determined to be the optimum microwave (MW) ageing time for the synthesis of PAN_MW-AO, which is dramatically faster than using conventional heating methods. Effects of pH, contact time, initial concentration of Cr(VI) and temperature on adsorption were investigated systematically by bath adsorption experiments. Nonlinear solutions of a pseudo-second-order kinetic model and Langmuir isotherm model were found to provide the closest fit to the experimental data for the adsorption process of Cr(VI), which indicated that chemisorption is the controlling mechanism and monolayer adsorption is dominant. Thermodynamic parameters revealed the spontaneity of the adsorption process, and higher temperature favored adsorption. The formation of PAN_MW-AO and Cr(VI)-adsorbed PAN_MW-AO has been characterized by FTIR, SEM/EDX and XPS instrumentations, which demonstrate that Cr(VI) was adsorbed by the amidoxime group via a surface complexation mechanism. The adsorption of Cr(VI) ions was hardly affected by common coexisting ions such as SO₄²⁻, NO₃⁻ and Cl⁻. High desorption efficiency (>90%) of Cr(VI) was achieved using 0.1 M H₂SO₄ as effluent, by which the investigated adsorbent could be used repeatedly five times with a small decrease in sorption capacity. Rapidity of synthesis and low cost, coupled with highly efficient and rapid adsorption of Cr(VI) ions, make PAN_MW-AO fibers an attractive adsorbent in the potential application for wastewater treatment.

---

1 Introduction

Three Cr forms, Cr(0), Cr(III) and Cr(VI), are used commercially and are thermodynamically stable when present in aquatic environments. However, only Cr(VI) has been shown to cause cancers in laboratory animals and occupationally exposed workers as it can diffuse as HCrO₄⁻ or CrO₄²⁻ through cell membranes.¹ Sources for Cr(VI) contaminated water originate from industrial effluents such as electroplating, mining extraction, leather tanning and pigment production, causing various health effects ranging from skin irritation to lung cancer, as well as kidney, liver, and gastric damage in humans.^2,3 Because of its high toxicity, the discharge of Cr(VI) to surface water is regulated to below 0.5 mg L⁻¹ according to the Integrated Wastewater Discharge Standard of China (GB 8978-1996) while the total Cr [including Cr(III) and Cr(VI)] is regulated to below 1.5 mg L⁻¹.^4,5 Noteworthily, due to its oxyanion forms like Cr₂O₇⁻, HCrO₄⁻ and CrO₄²⁻ in aquatic environments, it is poorly adsorbed by negatively charged soil particles, resulting in high mobility in the water.⁶ Thus, it is mandatory to develop and implement an effective and reliable technology to remove Cr(VI) from the water on account of human health security.

The techniques being applied for the removal of Cr(VI) include precipitation,⁷ adsorption,⁸ ion exchange,⁹ reverse osmosis,¹⁰ electrodialysis¹¹ and others.¹² Among the various methods, adsorption is one of the simplest, economically beneficial methods that can remove the metal more effectively.^13,14 Heretofore, due to their high capacity and selectivity, adsorbents immobilized with amidoxime groups were considered to be the most promising materials for the removal of metal ions.¹⁵ Many types of adsorbents consisting of amidoxime groups have been prepared and their adsorption performances towards metal ions were reported. Chen et al.¹⁶ prepared silica gel supported amidoxime adsorbents with “heterogeneous” and “homogeneous” methods, which exhibited excellent adsorption selectivity for Hg(II). Shaaban et al.¹⁷ studied Cu(II), Ni(II) and Pb(II) adsorption by a new amidoxime chelating resin (PAO-AM) and the maximum adsorption capacities for these three metal ions were 146.17, 113.85 and 236.21 mg g⁻¹, respectively. The recovery of uranium(VI) from seawater by amidoxime functionalized adsorbents has attracted abundant attention during recent decades,^18–20 while few studies have covered research on Cr(VI) adsorption from wastewater. Ramazan Coskun et al.²¹ prepared a fibrous adsorbent especially including double amidoxime groups through free radical reaction for the removal of Cr(VI) from wastewater, and the maximum adsorption capacity reached 125 mg g⁻¹ at pH 2.

Nevertheless, the adsorption capacity as well as the amidoxime grafting rate have become the major challenges for the prepared adsorbents, and improvements are needed to be proposed. Currently, conventional heating is the universal method to fulfill the functionalization process, which has a lack of efficiency and wastes energy. The microwave-assisted (MW-aid) method as a highly efficient heating method has wide applications in chemosynthesis due to its characteristics of rapid volumetric heating, high reaction rate, short reaction time, enhanced reaction selectivity and energy saving.^22–24 As is well known, the effect of the MW-aid method is created by the interaction of the dipole moment of the molecules with the high-frequency electromagnetic radiation.^25,26 The water molecule is one of the best solvents for MW-aid owing to its large dipole moment and total nontoxicity.²⁷

Herein, we introduced a rapid and facile method to prepare amidoxime functionalized polyacrylonitrile (PAN) fiber by an environmentally friendly MW-aid method. This work is to serve as not only an expansion of the MW-aid method in applications of polymer synthesis, but also an expansion of the PAN-based fibrous material as a high-efficient adsorbent for Cr(VI) removal from aqueous solution. In order to find the optimum synthesis conditions, a comprehensive study in association with the factors of MW power, mass of NH₂OH·HCl, time and bath ratio was performed by using the response surface methodology (RSM). Kinetic, equilibrium isotherm models and thermodynamic parameters were explored to disclose the process of Cr(VI) adsorption. The regeneration of modified PAN fiber was evaluated in its application. Moreover, the adsorption mechanism was discussed.

2 Experimental

2.1 Materials and apparatus

The polyacrylonitrile fiber (PANF) made of 100% acrylonitrile with a diameter of 10 ± 0.5 μm was purchased from Beijing Rongnai Industry Material Co., Ltd. (Beijing, PR China). Hydroxylamine hydrochloride (NH₂OH·HCl) and sodium carbonate (Na₂CO₃) were both supplied by Aladdin Chemical Reagent Co., Ltd. (Shanghai, PR China). The Cr(VI) solution was prepared by dissolving an appropriate amount of potassium dichromate (K₂Cr₂O₇, Sinopharm Group Chemical Reagent Co., Ltd., Shanghai, PR China) in deionized water. All of the reagents were used without further purification.

2.2 Preparation of amidoxime functionalized PAN fiber

The polyacrylonitrile fiber was dried in the oven overnight before use and cut into 5 cm lengths, to prevent its becoming entwined while stirring. The preparation procedure is simple and described as follows: first, hydroxylamine hydrochloride and sodium carbonate were dissolved in deionized water, which was placed in a 250 mL three-neck flask. Subsequently the PANF was added into the solution and the mixture was subjected to microwave irradiation. A specialized commercial microwave reactor (COOLPEX-E) with a maximum power range of 1200 W was used for the reaction. Fig. S1 of ESI† shows the components of the MW reactor. After reaction, the modified fiber was washed several times with hot distilled water until neutral and dried in a vacuum at 343 K overnight.

The grafting rate percentage (GP%) was calculated by gravimetry through the following equation:

where m₀ and m₁ are the weights of raw polyacrylonitrile fiber and amidoxime-functionalized fiber, respectively. The modified PAN fiber was named as PAN_MW-AO.

2.3 Experimental response surface methodology (RSM)

RSM is an efficient technique for the optimization of a multivariable system.²⁸ In this study, the optimal GP% of PAN_MW-AO was obtained by RSM using Design Expert 7.0. The four independent parameters chosen in this study were MW power (X₁), mass of NH₂OH·HCl (X₂), time (X₃) and bath ratio (X₄), with the GP% of PAN_MW-AO fibers as the response. The low, center and high levels of each variable were designated as −1, 0 and +1, respectively, as illustrated in Table 1. Box-Behnken Design (BBD) is the most frequently used form of RSM,^29–31 which was employed to evaluate the influence of the four independent variables in 29 sets of experiments. The fundamental assumptions of RSM and more detailed information have been discussed elsewhere.

Table 1 Experimental range and levels of the independent variables

Variables	Symbol	−1	0	+1
MW power (W): A	X1	300	500	700
Mass of NH2OH·HCl (g): B	X2	2.0	3.0	4.0
Time (min): C	X3	1 × 3	1 × 5	1 × 7
Bath ratio (g g^−1): D	X4	20 : 1	40 : 1	60 : 1

---

A second-order polynomial equation was used to fit the experimental results of BBD as follows:

Y = b₀ + b₁X₁ + b₂X₂ + b₃X₃ + b₄X₄ + b₁₂X₁X₂ + b₁₃X₁X₃ + b₁₄X₁X₄ + b₂₃X₂X₃ + b₂₄X₂X₄ + b₃₄X₃X₄ + b₁₁X₁² + b₂₂X₂² + b₃₃X₃² + b₄₄X₄² (2)

where Y represents the response variable (GP%), b_i, b_ii and b_ij are the regression coefficients for linear and quadratic effects and the coefficients of the interaction parameters, respectively, and X_i are the independent variables.

2.4 Characterization and analysis

The concentration of metal ions was determined using an inductively coupled plasma optimal emission spectrometer (ICP-OES; PerkinElmer Optimal 8300). Zeta potential measurements were conducted with a Zeta voltmeter (Zetasizer Nano ZS90). Fourier transform infrared spectroscopy (FT-IR) experiments on PAN_MW-AO were performed via an FT-IR spectrometer (PerkinElmer spectrum 100), and the spectra were recorded at wave numbers ranging from 400 to 4000 cm⁻¹. The surface morphology of PAN_MW-AO was observed by scanning electron microscope-energy dispersive X-ray spectroscopy (SEM/EDX; FEI QUANTA 200). In order to observe the changes of surface morphology after adsorption, an SEM/EDX experiment on PAN_MW-AO after adsorbing Cr(VI) was performed. The X-ray photoelectron spectroscopy (XPS) measurements were conducted using an XPS spectrometer (Thermo Fisher Scientific, ESCALAB 250Xi), with monochromatized Al Kα X-rays.

2.5 Batch adsorption experiments

The prepared PAN_MW-AO fibers were used as adsorbents for the removal of Cr(VI) from aqueous systems. All batch adsorption experiments were performed on a model SHA-C shaker (Ronghua Instrumental Manufactory Co., Ltd., China) with a shaking speed of 100 rpm. The effect of pH value on adsorption was studied as pH ranged from 1 to 10 with the initial concentration of 300 mg L⁻¹ at room temperature (20 °C). Coexisting anion solutions were prepared by including different amounts of Na₂SO₄, NaNO₃ and NaCl in the Cr(VI) solution. The kinetic adsorption of Cr(VI) was studied by adding 0.05 g PAN_MW-AO into a 50 mL volume of Cr(VI) solution (300 mg L⁻¹), for a preset time at pH 2. Adsorption isotherms for each heavy metal were conducted with the pH value at 2 and initial concentrations ranging from 20 to 320 mg L⁻¹ at 293 K, 298 K, and 303 K respectively. The amount of adsorbed (q_e) on the modified fiber can be obtained by knowing the equilibrium concentration, C_e, remaining in the solution phase, the aqueous phase volume, V (50 mL), and the initial concentration of hexavalent chromium (C₀) using the relation:

2.6 Desorption and regeneration experiments

For regeneration of the prepared adsorbents, the PAN_MW-AO fibers were first placed into contact with 300 mg L⁻¹ Cr(VI) for 12 h at optimum pH and 293 K, and then rinsed with distilled water to remove any residual solution and dried at room temperature overnight. After that, the Cr(VI) binding fibers were immersed into 0.1 M H₂SO₄ and shaken at 293 K for 30 min. Subsequently, the desorbed Cr(VI) ions were determined by ICP-OES and the desorption efficiency (%) was calculated based on the percentage of the ratio between the desorbed and preadsorbed amounts of the ions. The adsorption/desorption cycle was repeated five times using the same PAN_MW-AO fibers to estimate the regeneration performance.

3 Results and discussion

3.1 Analysis of response surface design

The total set of experiments in this study based on a 3 level and 4 factor experimental design is elucidated in Table S1 of ESI.† In accordance with the experimental design, a second-order polynomial equation on the basis of actual factors is applied to elucidate the empirical relationships between the independent variables and the response:

GP% = 35.78 + 3.86X₁ + 3.00X₂ + 1.58X₃ − 0.21X₄ − 2.45X₁X₂ − 0.50X₁X₃ + 1.63X₁X₄ − 2.52X₂X₃ − 1.38X₂X₄ + 0.28X₃X₄ − 8.99X₁² − 0.67X₂² − 2.05X₃² + 0.76X₄²

Analysis of variance (ANOVA) was applied to evaluate the adequacy of the model. From the ANOVA of the empirical second-order polynomial model (Table 2), the F value and p value for the model are 20.18 and < 0.0001 respectively, indicating that the model is significant. There is only a 0.01% chance that the “model F value” could occur due to noise.³² Additionally, the “R-squared” of 0.9528 is in reasonable agreement with the “adj R-square” of 0.9056, confirming the good prediction capability of the model. Moreover, the values of p > F less than 0.05 indicate that the model terms are significant, while values greater than 0.05 indicate that the model terms are not significant. For this reason, many values are not significant at the 95% confidence level (a = 5%) and are dropped from the model to build a reduced model with better prediction capabilities.

Table 2 ANOVA for analysis of variance and adequacy of the quadratic model

Source	Sum of squares	Degree of freedom	Mean square	F-value	Prob > F
Model	958.38	14	68.46	20.18	<0.0001
X1	178.64	1	178.64	52.66	<0.0001
X1	108.00	1	108.00	31.84	<0.0001
X1	30.08	1	30.08	8.87	0.0100
X1	0.52	1	0.52	0.15	0.7011
X1X2	24.01	1	24.01	7.08	0.0186
X2^2	2.94	1	2.94	0.87	0.3676

---

Table 2 reports the sum of squares, together with their F value and corresponding p value. According to the monomial coefficient values of the regression model, p(X₁) < 0.001, p(X₂) < 0.001, p(X₃) = 0.0100, p(X₄) = 0.7011, the bath ratio is an insignificant factor, while time is a significant factor. By contrast, power of radiation (X₁) and mass of NH₂OH·HCl (X₂) play the most influential roles in the grafting percentage of modified fiber. The combined effect of MW power and mass of NH₂OH·HCl on the grafting percentage of PAN_MW-AO fibers is shown in Fig. 1 as response surface plots. The GP% increases up to 40.6% with the increase of both MW power and mass of NH₂OH·HCl.

Fig. 1 Response surface plot for GP% of PAN_MW-AO fibers showing interaction between power of radiation (W) and mass of NH₂OH·HCl (g) (time 1 × 5, bath ratio at 50 : 1).

The mathematic model generated during RSM implementation was validated by conducting experiments for a given optimal medium setting. To confirm the adequacy of the model for predicting the maximum grafting percentage of PAN_MW-AO fibers, a verification experiment was carried out using the optimum conditions (500 W, 4 g NH₂OH·HCl, 1 × 5 min and bath ratio at 40 : 1). The average maximum GP% of 39.8% that is obtained from three replicate experiments is close to the predicted value of 40.5%, which proves the validity of the model for simulating the GP% of PAN_MW-AO fibers because of the good agreement between the predicted value and the experimental value.

3.2 MW-aid mechanism of the preparation

In order to give an insight into the MW-aid mechanism for the preparation of PAN_MW-AO fibers, the products were prepared by conventional heating using a thermostatic magnetic heating agitator (CL-3, Gongyi City Yuhua Instrument Co., Ltd., China) and the prepared product was named PAN_CV-AO. The preparation parameters and the grafting results are shown in Table 3. It is obvious to observe that the consumed time is profoundly reduced and the obtained fibers from the MW-aid method show more than twice the GP% of those from conventional heating. Consequently, the adsorption capacity of the PAN_MW-AO fibers is considerably higher than PAN_CV-AO. Moreover, the mechanical strength of PAN_MW-AO is much more extensively maintained than PAN_CV-AO in terms of the remarkably reduced time of the functionalization process and thus is more suitable for practical use in water.

Table 3 Comparison of different heating types for preparation of PAN_MW-AO fiber

Products	Energy source	Parameters				GP%	Adsorption capacity (mg g^−1)
		A	B	C	D
PANMW-AO fibers	MW-aid method	500 W	4.0	5	40 : 1	40.6	219.8
PANCV-AO fibers	Conventional heating	373 K	4.0	720	40 : 1	19.8	92.6

---

MW irradiation offers a fast, inexpensive and convenient method of heating, which is an alternative way to provide energy to chemical reaction systems. Microwave effects in chemical reactions are related to short-range molecular friction, due to the continuous polarization of molecules caused by MW irradiation.³³ This process of ordered (with electromagnetic field) and disordered (without the field) dipoles enhances the molecular attrition, increasing the local temperature and reaction rates.^34,35 Moreover, it is believed that MW can provide a specific effect (non-thermal effect), which is generally connected to the selective adsorption of MW energy by polar molecules.³⁶ In our reactive system, the matrix of PAN fibers as well as the solvent (water) are both polar and thus could interact with MW, leading to the reduction of the activation energies of the reaction. These might cause the higher grafting rate of amidoxime groups, which would be favorable for the uptake of Cr(VI).

3.3 Adsorption behaviour

3.3.1 Effect of pH on adsorption capacity. Commonly, the change in the external pH value mainly affects the surface charges of the adsorbents and the existence form of the ions.³⁷ To explore the relationship between the external pH value and the adsorption capacity, the zeta potentials of the PAN_MW-AO and PAN fibers were determined over a range of pH values from 1 to 10. As shown in Fig. 2, the zeta potential of PAN_MW-AO fibers is positive at pH < 5.8, while that of the PAN fibers is positive at pH < 3.7, indicating that the zero point of the zeta potential for the PAN_MW-AO fibers is higher after the surface amidoxime reaction. The ascending zero point of the zeta potential may be attributed to the protonated –NH₃⁺ of the amidoxime groups on the fibers and thus can provide better sorption amounts for negative Cr(VI) ions via electrostatic attraction. Furthermore, the effect of pH on Cr(VI) removal was examined. Expectedly, as depicted in Fig. 2, PAN_MW-AO fibers show distinctly higher Cr(VI) adsorption capacity than PAN fibers throughout the pH range. For PAN_MW-AO fibers, the highest Cr(VI) adsorption amount is observed at pH 2. Under this condition, the capacity of PAN_MW-AO fibers is as high as 219.8 mg g⁻¹, more than 40 times that of PAN fibers. The Cr(VI) species are highly dependent on the pH value. H₂CrO₄ is the predominant species at pH below 1.0, which cannot be electrostatically attracted by the protonated amidoxime groups. HCrO₄⁻ and Cr₂O₇²⁻ are predominant in the pH range of 2.0–6.0, while CrO₄²⁻ is the major species when pH > 6.³⁸ At a lower pH (approximately 2), the Cr(VI) species are mostly in their monovalent forms (HCrO₄⁻), which require only one exchange site and are more likely to be adsorbed. Noteworthily, the wastewater effluent from chromium industries is usually acid, which gives an edge to the application of PAN_MW-AO fibers with high adsorption capacity in treating Cr(VI) under this condition.

Fig. 2 Effect of pH on the adsorption of Cr(VI) onto the PAN_MW-AO fibers.

3.3.2 Adsorption kinetics. Adsorption kinetics are important constants for the evaluation of a good adsorbent. As can be seen from Fig. 3, the Cr(VI) uptake on PAN_MW-AO fibers is rapid in the first 30 min, contributing to about 70% of the ultimate adsorption amount for Cr(VI), and then ascends gradually. In this study, the adsorption equilibrium is achieved within about 2 h.

Fig. 3 Adsorption kinetics curve of Cr(VI) on PAN_MW-AO at 298 K.

Kinetic models are a significant aspect of adsorption studies and define the efficiency of the adsorption process. In order to convincingly investigate the adsorption behaviors of Cr(VI) with the adsorbents, pseudo-first-order kinetics and pseudo-second-order kinetics have been employed to simulate the experimental data and the best fit model was selected based on both the nonlinear regression correlation coefficient (R²) and the calculated χ² values. The value of χ² represents the similarity between the data from the model and the data from the experiment test. A small χ² value indicates similarities, while a larger number represents variation from the experimental data.^39–41

The equations for these two kinetic models and χ² are presented respectively as follows:

q_t = q_e(1 − e^−k₁t) (4)

where q_e and q_t are the amounts of heavy metal ions adsorbed (mg g⁻¹) at equilibrium and at time t (min), respectively, while k₁, k₂ and k₃ are the rate constants of the pseudo-first-order model (min⁻¹) and pseudo-second-order model (mg g⁻¹ min⁻¹). The q_e,cal (q_e,calculated) equilibrium capacity was calculated from the model (mg g⁻¹) and q_e,exp. (q_{e,experimental}) was the equilibrium capacity obtained from the experimental data (mg g⁻¹). As seen from the results listed in Table 4, the pseudo-second-order provides better coefficients (R² = 0.9991) than the pseudo-first-order model (R² = 0.9758). Simultaneously, the good agreement between the q_e,cal (217.1 mg g⁻¹) and the q_e,exp (219.8 mg g⁻¹) suggests that the adsorption kinetics closely follows the pseudo-second-order rather than the pseudo-first-order model. Moreover, the χ² value of the pseudo-second-order model is smaller, which also proves its better fit with the experimental data. The pseudo-second-order model was developed on the assumption that the rate-determining step may be chemisorption promoted by covalent forces through electron sharing between adsorbent and adsorbate. Thus, the result suggests that the adsorption of Cr(VI) on the PAN_MW-AO fibers is mainly controlled by chemically reactive adsorption.

Table 4 Kinetic model constants for the adsorption of Cr(VI) onto PAN_MW-AO fibers

qe,exp. (mg g^−1)	Pseudo-first-order model				Pseudo-second-order model				Intra-particle diffusion model
	qe,cal. (mg g^−1)	k1 (min^−1)	R^2	χ^2	qe,cal. (mg g^−1)	k2 × 10^4 (mg g^−1 min^−1)	R^2	χ^2	k3 (mg g^−1 min^−1/2)	C (mg g^−1)	R^2
219.8	210.5	0.037	0.9758	10.5	217.1	1.76	0.9991	3.89	15.34	36.62	0.9152

---

Furthermore, an intra-particle diffusion model was also used to analyze the rate-limiting step in adsorption.⁴² The model can be expressed by the Weber–Morris equation as follows:

q_t = k₃t^0.5 + C (7)

where k₃ is the intra-particle diffusion rate constant and the intercept C (mg g⁻¹) is a constant related to the thickness of the boundary layer. It is postulated that the intra-particle diffusion is considered to be the rate-limiting step when the plot of q_t vs. t^0.5 yields a straight line that passes through the original data.

The plot of q_t vs. t^0.5 is given in Fig. 4. The adsorption plots of Cr(VI) do not pass through the origin, which suggests the intra-particle diffusion is not the only rate-limiting step. This behavior indicates that the Cr(VI) adsorption process by PAN_MW-AO fibers involves more than one single kinetic stage. The initial rapid adsorption may be governed by boundary layer diffusion and the subsequent slow uptake is attributed to the intra-particle diffusion effect. As listed in Table 3, the obtained high values of k₃ and C indicate that the PAN_MW-AO fibers exhibit a fast removal rate for Cr(VI) from aqueous solution.

Fig. 4 Intra-particle diffusion model for the adsorption of Cr(VI) onto PAN_MW-AO fibers.

3.3.3 Adsorption isotherms. The effect of initial concentration on the uptake amount of PAN_MW-AO fibers for Cr(VI) was studied at 293 K, 298 K and 303 K. As illustrated in Fig. 6, the amount of metal ions adsorbed onto the PAN_MW-AO fibers is increased with increasing Cr(VI) concentration, and then the plateau value is reached. The maximum adsorption capacity for Cr(VI) was found to be 218.6 mg g⁻¹ at 303 K.

The adsorption isotherms of PAN_MW-AO fibers were investigated by the Langmuir, Freundlich and Temkin models, respectively. The Langmuir model assumes that a monomolecular layer is formed when adsorption takes place with an interaction between the adsorbed molecules.

The Freundlich isotherm is an empirical equation assuming that the adsorption process takes place on heterogeneous surfaces and the adsorption capacity is related to the concentration of adsorbate at equilibrium. The Temkin isotherm assumes that the heat of adsorption of the molecules decreases linearly due to sorbent–sorbate interaction.⁴³ The Langmuir, Freundlich and Temkin models can be expressed as follows:

where q_e is the amount of metal ions adsorbed at equilibrium by the adsorbent (mg g⁻¹), C_e is the equilibrium concentration (mg L⁻¹), q_m is the theoretical saturation adsorption capacity (mg g⁻¹), K_d (L mg⁻¹) is the equilibrium Langmuir constant, and K_f (mg g⁻¹) and n are constants representing the adsorption capacity and intensity of adsorption. K_T (L g⁻¹) is the equilibrium binding constant corresponding to the maximum binding energy, B_T (kJ mol⁻¹) is the Temkin constant related to the heat of adsorption, R (8.314 J mol⁻¹ K⁻¹) is the universal gas constant and T (K) is the absolute temperature.

The fitting of the equilibrium data at different temperatures by the Langmuir, Freundlich and Temkin models was conducted and the results are summarized in Fig. 5 and Table 5. In the comparison of R² and χ² values at all temperatures, the adsorption isotherm data fit the Langmuir model better than both the Freundlich and Temkin models, which indicates the monolayer adsorption of Cr(VI) by PAN_MW-AO fibers. Additionally, the maximum adsorption capacities calculated from the Langmuir model are highly consistent with the actual saturated adsorption capacity for Cr(VI) at each temperature.

Fig. 5 Adsorption isotherm curves and the variation of R_L with initial concentration of Cr(VI) on PAN_MW-AO at 293 K, 298 K and 303 K.

Fig. 6 FT-IR spectrum of PANF, PAN_MW-AO and PAN_MW-AO–Cr(VI).

Table 5 Adsorption isotherm parameters for the adsorption of Cr(VI) on PAN_MW-AO at 293 K, 298 K and 303 K

T (K)	Langmuir parameters					Freundlich parameters			Temkin parameters
	qm (mg g^−1)	Kd (L mg^−1)	R^2	χ^2	n	Kf (mg g^−1)	R^2	χ^2	KT (L g^−1)	BT (kJ mol^−1)	R^2	χ^2
293	232.70	0.0536	0.9985	7.7	2.39	33.31	0.9555	238.5	0.588	48.94	0.9967	15.1
298	238.28	0.0700	0.9992	5.0	2.53	40.29	0.9459	320.9	0.783	48.92	0.9945	28.0
303	250.71	0.0841	0.9986	9.6	2.58	46.89	0.9429	384.8	1.04	48.09	0.9914	49.4

---

Moreover, the dimensionless separation factor R_L could be calculated by applying the Langmuir parameters in the following equation:

This factor can indicate the favorability of the adsorption process, which is indicated to be either unfavorable (R_L > 1), linear (R_L = 1), favorable (R_L < 1), or irreversible (R_L = 0) under each condition.⁴⁴ For the initial concentration studied at all temperatures, the R_L values of Cr(VI) fall between 0 and 0.5, confirming the favorable nature of adsorption by PAN_MW-AO fibers. Additionally, the values of R_L decrease with the increasing of the initial concentration of adsorbate, which indicates that the adsorption process is more favorable at higher concentration.

3.3.4 Thermodynamic analysis. In order to better understand the effect of temperature on the adsorption of Cr(VI) ions onto the PAN_MW-AO fibers, three basic thermodynamic parameters were studied: the Gibbs free energy change (ΔG), entropy (ΔS) and enthalpy (ΔH). The thermodynamic parameters ΔG, ΔH and ΔS for this adsorption process were determined by using the following equations:

ΔG = −RT ln K_d (12)

where K_d is the adsorption equilibrium constant obtained from the Langmuir model, R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), and T is the temperature (K). The values of ΔH and ΔS could be obtained as the slope and intercept from a linear plot between ln K_d versus 1/T.

The corresponding values of the thermodynamic parameters are presented in Table 6, which shows that the ΔH and ΔS are positive for all the experiments and ΔG is negative in all systems.

Table 6 Thermodynamic parameters for the adsorption of Cr(VI) on PAN_MW-AO

ΔH (kJ mol^−1)	ΔS (J mol^−1 K^−1)	ΔG (kJ mol^−1)
		293 K	298 K	303 K
41.57	17.45	−2.45	−3.20	−3.73

---

The negative values of ΔG revealed that the adsorption process was a chemical exothermic reaction in nature, and with the increase of the temperature, the adsorbed amount at equilibrium increased. The positive value of ΔS reveals the increased randomness and a rise in the degrees of freedom at the PAN_MW-AO fibers–solution interface during the binding of the Cr(VI) ions on the active sites of the adsorbent.³⁹ It is also observed that with an increase of temperature the value of ΔG decreases, indicating that the sorption process is spontaneous and thermodynamically favorable with an increase in temperature.

3.4 Characterization of mechanism study

The FT-IR spectra of the raw PANF and PAN_MW-AO fibers before and after Cr(VI) adsorption are shown in Fig. 6. A sharp and distinct adsorption band at 2243 cm⁻¹ is attributed to C≡N groups in the polyacrylonitrile fiber, which decreases remarkably after the conversion into amidoxime.¹⁸ Although the fiber was claimed to be made of 100% acrylonitrile, the adsorption at 1683 cm⁻¹ still confirms the existence of methyl acrylate or methyl methacrylate. From the PAN_MW-AO spectrum, it can be observed that new peaks appear at 3410 cm⁻¹ (broad, assigned to both N–H and O–H), 1645 cm⁻¹ (assigned to C=N), 1105 cm⁻¹ (assigned to C–N) and 921 cm⁻¹ (assigned to N–O), confirming the successful conversion of the nitrile group into the amidoxime group.¹⁶ After the adsorption of Cr(VI), the bands at 3410 cm⁻¹, 1645 cm⁻¹, 1105 cm⁻¹ and 921 cm⁻¹ are blue shifted into 3436 cm⁻¹, 1664 cm⁻¹, 1115 cm⁻¹ and 941 cm⁻¹, along with a new peak that occurs at 1037 cm⁻¹. The blue shift of the vibration frequency indicates the formation of coordination bonds between amidoxime groups and metal ions.⁴⁶ Consequently, these changes are all related to the chemical attachment of Cr(VI) onto the PAN_MW-AO fibers.

The morphologies and surface composition of raw PAN and PAN_MW-AO fibers before and after Cr(VI) adsorption were characterized by SEM/EDX and the results are shown in Fig. 8. Evidently, the smooth surface of the raw PAN fiber becomes creased, and the diameter of the modified fibers increases to a certain degree. Simultaneously, the EDX pattern provides the elemental information, which confirms the existence of Cr(VI) adsorbed on PAN_MW-AO fibers.

Fig. 7 XPS spectra of high resolution scan of Cr 2p: (a) PAN_MW-AO fibers before adsorption; (b) PAN_MW-AO fibers after adsorption.

Fig. 8 SEM images of (a) PAN, (b) PAN_MW-AO, (c) PAN_MW-AO–Cr and (d) EDX analysis of elemental composition of PAN_MW-AO–Cr(VI).

To further explore the possible mechanism of Cr(VI) adsorption by the PAN_MW-AO fiber, XPS experiments were performed to investigate the surface chemical composition of the PAN_MW-AO fibers before and after adsorption of Cr(VI). Normally, for the Cr 2p XPS spectrum, the significant binding energies at 577.0–578.0 eV and 586.0–588.0 eV correspond to Cr(III), while those for Cr(VI) are at 580.0–580.5 eV and 589.0–590.0 eV.⁴¹ As shown in the survey spectra in Fig. 7(b), the appearance of the Cr 2p spectra evidently confirmed the adsorption of Cr(VI) by the adsorbent. More specifically, two significant bands appear at binding energies of 586.5 eV and 576.6 eV, which are assigned to the Cr 2p_3/2 and Cr 2p_1/2 orbitals, respectively. The XPS data reveals that the chromium attached to the surface of the adsorbent is the trivalent form, which indicates that the Cr(VI) has been reduced into Cr(III) thoroughly during the adsorption onto the PAN_MW-AO fibers.

3.5 Effects of coexisting anions on Cr(VI) adsorption

In many cases, natural water sources contain various kinds of anions, which may compete with the Cr(VI) ions in the adsorption process. In the present study, the effect of different anions including NO₃⁻ (NaNO₃), SO₄²⁻ (NaSO₄), and Cl⁻ (NaCl) on Cr(VI) removal was studied, and the results are shown in Fig. 9. The capacity of Cr(VI) in the presence of 100 mg g⁻¹ SO₄²⁻ is 82.5% of that for solely Cr(VI) solution, whereas it is 89.4% and 92.3% for 100 mg g⁻¹ NO₃⁻ and Cl⁻. Comparatively, the effects of these coexisting anions on Cr(VI) adsorption by PAN_MW-AO fibers are smaller than or equal to those of other materials reported in the literature.^13,17,45 This could be explained by the different affinities of the ions toward PAN_MW-AO fibers, which mainly depend on the ion charge density. The Z/r (charge/radius) value of SO₄²⁻ is larger than those for NO₃⁻ and Cl⁻ and multivalent anions are adsorbed more readily than monovalent anions. Consequently, the decreasing trend of Cr(VI) adsorption capacity in the presence of competing anions is observed in the order: Cl⁻ < NO₃⁻ < SO₄²⁻.

Fig. 9 Effects of coexisting anions on the adsorption of Cr(VI).

3.6 Reusability test and comparison with literature

To assess their reusability, the used PAN_MW-AO fibers were separated by filtration after the first run and the extraction of adsorbed Cr(VI) ions, then 0.1 M H₂SO₄ was used to regenerate metal ions from the adsorbent. It is observed from Fig. 10 that the efficiency of the desorption process is generally above 90.5% and the adsorption capacity for Cr(VI) after five runs still maintains 80% of its original adsorption amount. On account of there being no significant loss of activity during the adsorption process, the PAN_MW-AO fibers are expected to be a reliable and powerful adsorbent to capture Cr(VI) from aqueous solution. Comparisons with other reported fibrous adsorbents, which were prepared through conventional heating, and with activated carbon, are listed in Table 7. The results reveal that MW irradiation is a facile and highly efficient way to improve the adsorption capacity of the adsorbent for Cr(VI). The consumed time and reagents are reduced while the amidoxime grafting rate is improved significantly. The combination of thermal and non-thermal effects of MW irradiation is believed to expedite the modification procedure and also result in the high grafting ratio of amidoxime groups. Commonly, MW irradiation is instantaneous, which leads to shorter start-up and rapid internal heating. Meanwhile, the non-thermal effects such as overheating, hot spots and selective heating may increase the probabilities of effective contacts between the reactants and thus improve the processes and obtain better yields.

Fig. 10 Adsorption capacity of the PAN_MW-AO fibers after five repeated regenerations.

Table 7 Comparison of PAN_MW-AO fiber adsorption capacities with other fibrous adsorbents and activated carbon

Adsorbents	Grafting time (h)	pH	Adsorption capacity (mg g^−1)	Reference
Hybrid carbon fibers	>10	2.0	190.0	47
EDTA-anchored PAN-based nanofiber	2	3.0	66.24	6
Strong alkaline anion exchange fiber	—	2.0	201.2	48
Thiol-modified cellulose nanofiber	>24	4.0	76.5 ± 2.0	49
Double amidoxime-containing PET fibers	10	2.0	22.72	21
PAN-NH2 nanofibers	4	2.0	137.6 ± 4.1	50
Fibrous adsorbent with amino and quaternary ammonium groups	6	4.3	98.5	46
Activated carbon	—	—	21.34	8
PANMW-AO fiber	0.17	2.0	219.8	Present study

---

4 Conclusions

The microwave-assisted method has been proven to be an efficient and rapid approach for the immobilization of amidoxime groups on PAN fibers. A significant reduction of the reaction time and energy consumption and a better adsorption capacity were obtained relative to conventional heating, and both thermal and non-thermal effects of MW irradiation were believed to make contributions to the grafting process. The preparation conditions have been investigated with a response surface methodology based on the Box-Behnken design, and the optimum conditions were found to be MW power at 500 W, 4 g NH₂OH·HCl, 1 × 5 min time and bath ratio at 40 : 1, with a maximum GP% gain of 40.6%. The obtained PAN_MW-AO fibers showed a very high affinity to Cr(VI) at pH 2 and the binding mechanism was proven to be surface complexation by the characterization results of FTIR, SEM/EDX and XPS. The pseudo second-order kinetic equation provides the best correlation for the process, suggesting a chemisorption process as the rate limiting step. According to the nonlinear regression correlation coefficient (R²) and the calculated χ² values of the adsorption isotherm equations, the Langmuir isotherm has been found to be better fitted than the Freundlich and Temkin isotherms. The adsorption thermodynamics demonstrated that the adsorption process is spontaneous and endothermic. Competition from common coexisting ions, such as SO₄²⁻, NO₃⁻ and Cl⁻, was negligible. PAN_MW-AO fibers could be reused repeatedly 5 times with more than 80% of initial adsorption capacity, suggesting that PAN_MW-AO fibers have promising potential as an adsorbent for the removal of Cr(VI) from aqueous solution.

Acknowledgements

The work was supported by State Key Laboratory of Urban Water Resource and Environment (Harbin Institute of Technology) (2015DX03), Fundamental Research Funds for the Central Universities (HIT. NSRIF. 201671) and National Science Foundation for Post-doctoral Scientists of China (2014M561356).

Notes and references

1. M. Owlad, M. K. Aroua, W. A. W. Daud and S. Baroutian, Water, Air, Soil Pollut., 2008, 200, 59–77.
2. D. W. O'Connell, C. Birkinshaw and T. F. O'Dwyer, Bioresour. Technol., 2008, 99, 6709–6724.
3. K. C. K. Lai and I. M. C. Lo, Environ. Sci. Technol., 2008, 42, 1238–1244.
4. M. Nuchter, B. Ondruschka, W. Bonrath and A. Gum, Green Chem., 2004, 6, 128–141.
5. X. Lv, J. Xu, G. Jiang, J. Tang and X. Xu, J. Colloid Interface Sci., 2012, 369, 460–469.
6. E. F. C. Chaúque, L. N. Dlamini, A. A. Adelodun, C. J. Greyling and J. Catherine Ngila, Appl. Surf. Sci., 2016, 369, 19–28.
7. R. C. Thomson and M. K. Miller, Acta Mater., 1998, 46, 2203–2213.
8. Z.-Y. Kong, J.-F. Wei, Y.-H. Li, N.-N. Liu, H. Zhang, Y. Zhang and L. Cui, Chem. Eng. J., 2014, 254, 365–373.
9. G. Wójcik, V. Neagu and I. Bunia, J. Hazard. Mater., 2011, 190, 544–552.
10. A. Bódalo, J. L. Gómez, E. Gómez, A. M. Hidalgo and A. Alemán, Desalination, 2005, 180, 277–284.
11. T. Ölmez, J. Hazard. Mater., 2009, 162, 1371–1378.
12. K. Y. Wang and T. S. Chung, J. Membr. Sci., 2006, 281, 307–315.
13. Q. Li, Y. Qian, H. Cui, Q. Zhang, R. Tang and J. Zhai, Chem. Eng. J., 2011, 173, 715–721.
14. C.-J. Li, Y.-J. Li, J.-N. Wang and J. Cheng, Chem. Eng. J., 2013, 220, 294–301.
15. C. Gunathilake, J. Gorka, S. Dai and M. Jaroniec, J. Mater. Chem. A, 2015, 3, 11650–11659.
16. J. Chen, R. Qu, Y. Zhang, C. Sun, C. Wang, C. Ji, P. Yin, H. Chen and Y. Niu, Chem. Eng. J., 2012, 209, 235–244.
17. A. F. Shaaban, D. A. Fadel, A. A. Mahmoud, M. A. Elkomy and S. M. Elbahy, J. Environ. Chem. Eng., 2014, 2, 632–641.
18. G. Bayramoglu and M. Y. Arica, Microporous Mesoporous Mater., 2016, 226, 117–124.
19. Y. Zhao, X. Wang, J. Li and X. Wang, Polym. Chem., 2015, 6, 5376–5384.
20. H. Zhao, X. Liu, M. Yu, Z. Wang, B. Zhang, H. Ma, M. Wang and J. Li, Ind. Eng. Chem. Res., 2015, 54, 3101–3106.
21. R. Coskun and Y. Dilci, J. Macromol. Sci., Part A: Pure Appl.Chem., 2014, 51, 767–782.
22. V. O. Njoku, K. Y. Foo and B. H. Hameed, Chem. Eng. J., 2013, 215–216, 383–388.
23. A. Roy and J. Bhattacharya, Chem. Eng. J., 2012, 211–212, 493–500.
24. X. J. Wang, Y. Wang, X. Wang, M. Liu, S. Q. Xia, D. Q. Yin, Y. L. Zhang and J. F. Zhao, Chem. Eng. J., 2011, 174, 326–332.
25. I. Bilecka and M. Niederberger, Nanoscale, 2010, 2, 1358–1374.
26. C. Oliver Kappe, Chem. Soc. Rev., 2008, 37, 1127–1139.
27. S. Deng, P. Wang, G. Zhang and Y. Dou, J. Hazard. Mater., 2016, 307, 64–72.
28. S. Chatterjee, A. Kumar, S. Basu and S. Dutta, Chem. Eng. J., 2012, 181–182, 289–299.
29. M. Khayet, A. Y. Zahrim and N. Hilal, Chem. Eng. J., 2011, 167, 77–83.
30. I. Dahlan, K. T. Lee, A. H. Kamaruddin and A. R. Mohamed, Environ. Sci. Technol., 2008, 42, 1499–1504.
31. J. Wu, H. Zhang, N. Oturan, Y. Wang, L. Chen and M. A. Oturan, Chemosphere, 2012, 87, 614–620.
32. M. J. K. Bashir, H. A. Aziz, M. S. Yusoff and M. N. Adlan, Desalination, 2010, 254, 154–161.
33. L. Perreux, A. Loupy and F. Volatron, Tetrahedron, 2002, 58, 2155–2162.
34. L. C. Lopes, M. T. M. Barreto, K. M. Gonçalves, H. M. Alvarez, M. F. Heredia, R. O. M. A. de Souza, Y. Cordeiro, C. Dariva and A. T. Fricks, Enzyme Microb. Technol., 2015, 69, 10–18.
35. M. Zieliński, M. Zielińska and M. Dębowski, Bioresour. Technol., 2013, 127, 223–230.
36. M. A. Herrero, J. M. Kremsner and C. O. Kappe, J. Org. Chem., 2008, 73, 36–47.
37. G. Ndayambaje, K. Laatikainen, M. Laatikainen, E. Beukes, O. Fatoba, N. van der Walt, L. Petrik and T. Sainio, Chem. Eng. J., 2016, 284, 1106–1116.
38. P. Miretzky and A. F. Cirelli, J. Hazard. Mater., 2010, 180, 1–19.
39. M. Sayın, M. Can, M. İmamoğlu and M. Arslan, React. Funct. Polym., 2015, 88, 31–38.
40. T. S. Anirudhan, J. R. Deepa and Binusreejayan, Chem. Eng. J., 2015, 273, 390–400.
41. H. Gu, S. B. Rapole, J. Sharma, Y. Huang, D. Cao, H. A. Colorado, Z. Luo, N. Haldolaarachchige, D. P. Young, B. Walters, S. Wei and Z. Guo, RSC Adv., 2012, 2, 11007–11018.
42. H. Gu, S. B. Rapole, Y. Huang, D. Cao, Z. Luo, S. Wei and Z. Guo, J. Mater. Chem. A, 2013, 1, 2011–2021.
43. F. Gao, H. Gu, H. Wang, X. Wang, B. Xiang and Z. Guo, RSC Adv., 2015, 5, 60208–60219.
44. C. Zhijiang, J. Jianru, Z. Qing and Y. Haizheng, RSC Adv., 2015, 5, 82310–82323.
45. N. Horzum, T. Shahwan, O. Parlak and M. M. Demir, Chem. Eng. J., 2012, 213, 41–49.
46. J. Wang, L. Zhao, W. Duan, L. Han and Y. Chen, Ind. Eng. Chem. Res., 2012, 51, 13655–13662.
47. W. Zheng, J. Hu, Z. Han, E. Diesel, Z. Wang, Z. Zheng, C. Ba, J. Langer and J. Economy, Chem. Eng. J., 2016, 287, 54–61.
48. W. Wang, M. Li and Q. Zeng, Sep. Purif. Technol., 2015, 149, 16–23.
49. R. Yang, K. B. Aubrecht, H. Ma, R. Wang, R. B. Grubbs, B. S. Hsiao and B. Chu, Polymer, 2014, 55, 1167–1176.
50. M. Avila, T. Burks, F. Akhtar, M. Göthelid, P. C. Lansåker, M. S. Toprak, M. Muhammed and A. Uheida, Chem. Eng. J., 2014, 245, 201–209.

---

Footnote

† Electronic supplementary information (ESI) available: Feature of the MW reactor, experimental design and response value for different conditions. See DOI: 10.1039/c6ra11727a

---