Abstract

In this study, a poly(acrylonitrile-co-N-vinyl pyrrolidone)/zeolite (poly(AN-co-VP)/zeolite) composite was synthesized by in situ free radical polymerization (FRP). The structural properties of the composite were analyzed by Fourier transform infrared spectroscopy (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM), thermogravimetric analysis (TGA), and differential scanning calorimetry (DSC). The characterization results indicated that the composite had a homogeneous and 3-dimensional (3D) structure. The decomposition temperature and glass transition temperature () were found as 410°C and 152°C, respectively. A poly(AN-co-VP)/zeolite composite was used to investigate the adsorption of brilliant green (BG) which is a water-soluble cationic dye. The kinetics, isotherms, and thermodynamics of adsorption were examined, and results showed that equilibrium data fitted the Langmuir isotherm model, and the adsorption kinetics of BG followed pseudo-second-order model. According to the thermodynamic properties, the adsorption process was endothermic and spontaneous. Response surface methodology (RSM), which was improved by the application of the quadratic model associated with the central composite design, was employed for the optimization of the study conditions such as adsorbent mass, time, and initial dye concentration. The RSM indicated that maximum BG removal (99.91%) was achieved at the adsorbent mass of 0.20 g/50 mL, an initial BG concentration of 40.20 mg/L, and a contact time of 121.60 minutes.

1. Introduction

Water pollution, which is one of the most unfavorable environmental problems in the world, stems from many reasons. One of these is synthetic dyes colored organic substances that cause irritation of the skin, mutagenicity, carcinogenicity, and damage to aquatic life [13]. Brilliant green (BG), commonly known as cationic dye, is one of the most important dyes and is extensively used as a dyeing agent in industries such as paper printing, wood, and textile. BG, which is from the triphenylmethane dye family, is very dangerous in case of contact with the skin and eyes and toxic to the lungs. It is very important to remove BG from aqueous solutions due to the formation of carbon dioxide, sulfur oxides, and nitrogen oxides during its decomposition [47]. It is a very difficult process to treat the dyes because they are not biodegradable and they have a highly complex aromatic molecular structure that makes them more stable [8].

There are many different physical, chemical, and biological techniques that can be used to remove dyes from water. Among them adsorption is a viable alternative method in the removal of contaminants from wastewaters due to its major advantages such as high efficiency, easy application, design simplicity, low cost, specific interactions between solid adsorbent molecules and dyes, and insensitivity to toxic pollutant [9, 10]. In addition, the main advantage of adsorption is that many natural and inexpensive materials such as natural materials, biosorbents, and solid wastes from industry and agriculture can be used as an adsorbent without any complicated equipment [1113].

In recent years, the preparation of organic–inorganic composites has become a focus of interest due to their low production cost and higher mechanical resistance, and thus they have been used for the adsorption of dyes. Different kinds of substances such as montmorillonite, activated clay, bentonite, zeolite, and kaolinite have been used to form composites. These clays incorporated into the polymer increase the adsorption capacity as well as the adsorption rate [14, 15]. Zeolites are natural components which are alternatively used to form the composite with polymers [1].

Poly 1-vinyl-2-pyrrolidone is used in many applications due to its numerous unique properties such as solubility in water, hydrophilic and biocompatible properties, dispersion, and adhesion. It is also interesting because poly(vinylpyrrolidone) (PVP) is nontoxic and biological inertness [1618].

Poly(AN-co-VP)/zeolite) composite has not been synthesized previously. However, there are some studies about synthesis of AN-co-VP copolymer. This copolymer was used for many purposes. For example, Staufenbiel et al. [19] synthesized a series of poly(acrylonitrile-co-N-vinyl-2-pyrrolidone) model nanoparticles to expand the set of model nanoparticles available for mechanistic studies of protein adsorption beyond polystyrene based lattices. Wan et al. [20] have used the poly(acrylonitrile-co-N-vinyl-2-pyrrolidone) for determining the total water content adsorbed on the PANCNVP film by swelling experiments performed at temperatures of interest. A series of rigid poly[acrylonitrile-co-(N-vinyl pyrrolidone)] model nanoparticles stably loaded with Nile Red or Rhodamine B, respectively, was comprehensively studied for biocompatibility and functionality by Zhang et al. [21]

In recent studies, a lot of adsorbents were prepared for brilliant green removal, but there are still some shortages on these adsorbents. For instance, polyaniline/silver nanocomposite was used as an adsorbent for removal of brilliant green (BG) from aqueous solutions [22]. However, the preparation of nanocomposites was probably complicated and silver is expensive. Ghaedi et al. [23] synthesized the zinc sulfide nanoparticle loaded on activated carbon and examined BG removal percentage toward various parameters including pH, adsorbent dosage, initial dye concentration, and contact time. However, this study is not very novel because activated carbon is widely used for adsorption studies due to its extremely porous and very large surface area. In our previous study, poly(AN-co-VP)/zeolite composite was prepared firstly and showed efficient adsorption ability for BG. Furthermore, this composite is generally considered to be nondegradable and thus is suitable as model systems that do not experience substantial alteration of properties in aqueous environment.

The intention of the present research is the synthesis and use of poly(AN-co-VP)/zeolite) composite to investigate the impact of three parameters: initial dye concentration, adsorbent mass, and contact time on the removal of the dye BG, using central composite design (CCD) which is a standard response surface methodology (RSM). RSM is the collection of statistical and mathematical tools and it was selected to explain the simultaneous effects of the parameters on the adsorption process [2426].

2. Materials and Methods

2.1. Materials

Monomers (AN, VP) and the dye (BG) were supplied from Sigma-Aldrich, and they were used as received commercially. Natural zeolite was provided by a local company in Turkey. The chemical composition of the zeolite, which was analyzed by X-ray fluorescence spectrometry, was given in Table 1.

The 2,2′-azobisisobutyronitrile (AIBN) was used as an initiator. All chemicals were of analytical purity. The chemical structure of BG and copolymers of AN and VP were given in Scheme 1.

2.2. Characterization

The FT-IR spectra were recorded on a Jasco-Spectrum FT-IR spectrophotometer for the region of 4000–400 cm−1. The TGA measurement was performed with a Shimadzu TGA-50 from room temperature to 500°C at a heating rate of 10°C/min. DSC analysis was recorded with a Shimadzu DSC-50 under nitrogen atmosphere from room temperature to 200°C at a heating rate of 20°C/min. Adsorption properties were examined by using a Shimadzu-1800 UV-VIS spectrophotometer. The morphology of the composite was evaluated by SEM and XRD. SEM was performed by Hitachi S-3500 with an accelerating voltage of 25 kV. The XRD spectrum was measured using an X-ray diffractometer (Rigaku miniflex-600) at voltage 40 kV, current 15 mA.

2.3. Synthesis of the Poly(AN-co-VP)/Zeolite Composite

The poly(AN-co-VP)/zeolite composite was prepared by using in situ polymerization method. For this purpose, 2 mL of AN, 2 mL of VP monomers, and 20% by mass of zeolite were added to 4 different polymerization tubes. The mixtures were mixed in an ultrasonic bath for 1 hour. AIBN (0.03 g) was added to these mixtures and the tubes that passed through an argon gas were submerged in an oil bath previously set at 60°C. After 40 minutes the polymerization was terminated and precipitated in hexane, dried under vacuum for 24 hours.

2.4. Adsorption Studies

To examine the adsorption properties of the poly(AN-co-VP)/zeolite composite, a series of adsorption experiments were carried out. 0.2 g solid poly(AN-co-VP)/zeolite composite was mixed with different concentrations (25 mg/L-100 mg/L) of dye solution in closed erlenmeyer flasks (50 mL) at natural pH (4.5). The flasks were kept at a constant speed (300 rpm) sealed shaker to provide temperature control for different periods at a temperature of 30°C. After predetermined time intervals, the absorbance values of the solutions were measured at 625 nm by UV-VIS spectrophotometry after being centrifuged at 5000 rpm for 10 minutes. The adsorbed amount (mg/g) of BG at time t () was calculated according to (1).In addition, the adsorbed amount of BG at the equilibrium () was calculated according to where , , and (mg/L) are the initial, equilibrium, and at any time BG concentrations, respectively. V (L) is the volume of solution, and (mg/g) is the mass per gram of the poly(AN-co-VP)/zeolite composite.

2.5. Experimental Design with CCD

The CCD is performed to test the combined and individual effects of three independent parameters of BG adsorption process on poly(AN-co-VP)/zeolite. Adsorbent mass (A), initial BG concentration (), and time () were chosen as independent parameters and the BG removal yield (%) was selected to be the response of the process (dependent parameter). A total of different experiments, including 6 axial points, 8 factorial points, and 6 replications of the central points, were performed in the five levels (-α, -1, 0, 1, +α) with CCD design matrix. Design Expert 7 (Stat-Ease, USA) was utilized for the evaluation of process parameters, graphical analysis, statistical analysis of the data found, and analysis of variance (ANOVA). The mathematical relationship between the process response and the three independent parameters was expressed by the following quadratic equation [33]:where is the predicted answer (BG removal yield), β0 is the constant coefficient, βi is linear effects of the input variable, βii is the quadratic effects, βij is the interaction effects between the input variables, and xi and xj are the coded values of the parameters.

3. Results and Discussion

3.1. Characterization of the Poly(AN-co-VP)/Zeolite Composite

Figure 1 shows the FT-IR spectra of the poly(AN-co-VP)/zeolite composite before adsorption and after adsorption. When the spectra were compared, it was seen that there was no important change in the chemical structure of the composite after the adsorption process. This can be explained by van der Waals interaction between the dye molecules and the composite molecules and it shows that the adsorption process was carried out on the surface of the adsorbent. The peak at 3444 cm−1 is due to OH groups of layered alumina silicates in the zeolite. The band at 1024 cm−1 corresponds to Si-O-Si and Si–O–Al stretching in the zeolite. Similar results have been found in the literature [34]. 1661 cm−1 is the characteristic peak of the C=O stretching of the tertiary amide of VP. Also, the peak at 2239 cm−1 characterizes the nitrile groups (CN) on acrylonitrile units.

SEM was used for composite surface area analysis before and after BG adsorption. In Figure 2(a) the polymer surface was very rough and porous with many cavities, and when filled with BG molecules the surface was very smooth (Figure 2(b)). As a result, the morphology of the polymeric composite filled with BG demonstrates a quite difference from the pure form and indicated that the BG adsorbs more efficiently on the composite.

The crystallinity of zeolite in the poly(AN-co-VP)/zeolite composite was evaluated on the basis of the XRD background in the range 2θ = 10–90°. On the other hand, XRD patterns of composite consist of many sharp peaks that reflect well the crystalline structure of the system. Zeolite is characterized by many sharp peaks by XRD and they are shown in Figure 3. These peaks are clearly seen in the composite structure when compared with the literature [24]. The peaks around 11.36°, 22.47°, 30.16°, and 56.25° reveal its cubic lattice structure assigned to the planes.

To investigate the thermal stability of the composite, TGA and DSC measurements were taken under the nitrogen atmosphere and the results are shown in Figure 4. According to Figure 4(a), the decomposition temperature is 410°C and the decomposition occurs in a single stage. This value indicates that the composite is quite stable. Also, residue of the composite at 500°C is around 25%. It is thought that the amount of this waste is derived from the zeolite. In DSC curve (Figure 4(b)), it can clearly be seen that the glass transition temperature () of the composite is 152°C.

3.2. Adsorption Studies
3.2.1. Adsorption Isotherms

The adsorption isotherms explain the relationship between the adsorbent and adsorbate and give an idea about the adsorption capacity at a certain temperature. They are important in terms of providing information about the surface properties and explaining the mechanism of adsorption. Furthermore, they play an important role in setting up the necessary physicochemical conditions for better understanding and designing the adsorption process. Apart from the available various isotherm models used to investigate adsorption equilibrium in the literature, two models, namely, the Langmuir and Freundlich isotherm models, are most commonly used to describe equilibrium adsorption [35, 36]. The Langmuir model successfully explains that the adsorption process takes place in a monolayer on a homogeneous surface. The linear Langmuir equation is expressed by the following relation [37]:where (mg/g) and (mg/L) are the amounts of adsorbed dye and unadsorbed dye concentrations in a solution at equilibrium, respectively. (L/g) is the adsorption isotherm constant and (L/mg) is the monolayer adsorption capacity related to the energy. The maximum adsorption capacity of the adsorbent () is numerically calculated from .

The Freundlich model is an exponential equation that exhibits adsorption on heterogeneous surfaces and it is also used for examining adsorption equilibrium in non-monolayer surfaces. The linear Freundlich equation is expressed by the following equation [38]:where (mg/L) is equilibrium concentration, (mg/g) is the Freundlich constants related to the amount of dye adsorbed on the adsorbent, and n is the applicability of adsorbent. The value of n between 1 and 10 shows suitable adsorption process. and were calculated from the slope and the intercept of the linear plot, respectively.

The constants and the correlation coefficient (R2) values calculated for the two equilibrium models are given in Table 2. The data were investigated for three different temperatures and compared. It was seen that the value of R2 for Langmuir isotherm was higher than the value obtained for the Freundlich isotherm. A higher value of R2 demonstrated that the adsorption isotherm for BG onto the poly(AN-co-VP)/zeolite composite is explained better by the Langmuir model [39]. The maximum adsorption capacity was obtained as 23.81 mg/g from Langmuir isotherm. This capacity was compared with many reported adsorbents and they are shown in Table 3. Furthermore, the value was above 1 (3.030, 2.632, 2.457), indicating that BG was favorably adsorbed onto the composite.

3.2.2. Adsorption Kinetics

The kinetics of the dye removal process were studied by using pseudo-first-order, pseudo-second-order, intraparticle diffusion, and external mass transfer kinetic models. Pseudo-first and pseudo-second-order kinetic studies were performed at different temperatures (20-40°C), constant composite dose (0.2 g/50 mL), and constant dye concentration (50 mg/L). Intraparticle diffusion and external mass transfer kinetic models were performed at different dye concentrations (25-100 mg/L), different temperatures (20-40°C), and constant composite dose (0.2 g/50 mL). The equations for the pseudo-first-order, pseudo-second-order, intraparticle diffusion, and external mass transfer kinetic models are as follows, respectively [40]:where and (mg/g) are the adsorption capacities of BG at equilibrium and at time t, respectively. k1 (1/min) is the pseudo-first-order rate constant, k2 (1/min) is the pseudo-second-order rate constant, (mg/gmin1/2) is the intraparticle diffusion rate constant, C is the BG concentration in time (mg/L), is the initial BG concentration (mg/L), and S is the specific surface area for mass transfer. The change of pseudo-first- and pseudo-second-order rate constants with temperature and the exchange of intraparticle diffusion rate constants and external mass transfer coefficients with temperature and different initial BG concentration are summarized in Tables 4 and 5, respectively.

In pseudo-first-order, plot versus t gives a straight line and the values of k1 and were determined using this straight line. For pseudo-second-order, the graph of t/ versus t was used to calculate and k2 from slope and intercept, respectively. It is clear in Table 4 that the calculated () value found from the graph is much lower than the experimental () value and low regression coefficient of BG. This demonstrates that pseudo-first-order cannot be applied for this adsorption experiment. Higher values of regression coefficient for pseudo-second-order illustrate greater harmony of BG adsorption than pseudo-first-order for all temperatures. Also in this kinetic model the qeq,cal and the values were very close to each other. For these reasons, pseudo-second-order model is better complied with the data [41, 42].

The external mass transfer model is related to the liquid-solid external mass transfer coefficient (cm/min) in the change of dye concentration over time. For the studied external mass transfer of BG onto the poly(VP-co-An)/zeolite, the plot versus t gives a straight line and the values were determined from the slope in all initial dye concentrations (25-100 mg/L) and at all temperatures (20, 30, 40°C). Findings showed that external mass transfer coefficients decreased with increasing initial BG concentration while they increased with increasing temperature. This increase in the coefficients can be attributed to the increase of kinetic energy and mobility of the BG molecules as the temperature increases. The external mass transfer is not negligible although this effect is important at the beginning of the adsorption [40].

curves against the t1/2 found at 20, 30, and 40°C and in all initial dye concentrations (25, 50, 75, 100 mg/L) consist of three steps. The first part is a sharply increasing section, and the second linear part shows the rate-limiting step of intraparticle diffusion. The third part is a slower internal diffusion part than other steps due to the decrease in the dye concentration in solution. For each concentration and temperature, values were calculated from these linear parts and displayed in Table 5. In the table, it is clear that values increased with increasing temperature and initial BG concentrations. For each temperature and concentration, linear curves of the second part did not pass through the origin. This demonstrates that the border layer control had some steps, and it also indicates that intraparticle diffusion was not the step solely controlling the rate [40].

An adsorption mechanism as shown in Figure 5 may be proposed for the adsorption of BG onto the composite, taking into account the chemical properties of the adsorbate and the adsorbent. The high adsorption capacity can be due to - interactions between the dye molecules containing aromatic ring and the poly(VP-co-An)/zeolite composite [43]. Meanwhile, van der Waals interactions have a significant effect on the amount of this adsorption.

3.2.3. Adsorption Thermodynamics

The thermodynamic parameters such as a changes in Gibbs-free energy (), enthalpy (), and entropy () for BG adsorption onto the poly(VP-co-An)/zeolite composite were calculated from the equations given below:where (8.314 J/mol K) is the universal gas constant, (K) is the absolute temperature, and is the equilibrium constant and it was calculated by using Langmuir constant () and molecular weight of BG (482.64 g/mol) [44]. Here, and were calculated from the slope and intercept of the graph of the versus 1/T.

The adsorption of solid materials is divided into physical and chemical adsorption. When compared to chemical adsorption the energy change for physical adsorption is lower than 40 kJ/mol while chemical adsorption changes from 40 kJ/mol to 120 kJ/mol [33]. As seen from Table 6 the adsorption process is physical according to the value. The negative value of shows that adsorption takes place spontaneously and exothermically at three different temperatures (20, 30, 40°C). Furthermore, the positive value of indicates that the solid-liquid interface is random during adsorption [45]. As a result, the adsorption of BG onto the poly(VP-co-An)/zeolite composite is more favorable.

3.3. CCD Optimization
3.3.1. ANOVA for Response Surface Quadratic Model

The experimental design including coded values of selected parameters and the observed and actual responses is shown in Table 7. Equation (13) indicates quadratic relationship between the response and independent parameters for the BG.

In (13), negative signs indicate antagonistic effect, whereas positive signs show the synergistic effect. As seen from (13), A (adsorbent mass) and B (time) are both positive. This shows that BG adsorption onto poly(VP-co-An)/zeolite will increase when these parameters are increased [46]. The significance of each term was established by p-value (Prob>F), which is shown in Table 8. According to this table, the model terms A, B, C, and A2 were significant, with a p-value being less than 0.05. The other model terms were insignificant (p> 0.05).

As shown in Table 8 (ANOVA), the model fit of the data is more significant. F-value (21.30) and p-value (<0.0001) of the model imply that it is quite important for BG adsorption onto poly(VP-co-AN)/zeolite. The high value of correlation coefficient (R2 = 0.95) indicated that only 5% of the total variation could not be described by the model. The higher value of R2 demonstrated the better fitness of the regression model in the studied experimental set [47]. Furthermore, the value of adjusted R2 (0.91) confirmed that the BG removal of the model is a significant parameter indicating good harmony between the experimental and predicted removal yield. Hence, the applied model is enough to make predictions at the interval of experimental factors. The adequate precision measures the signal to noise ratio. The ratio is required to be higher than 4 for a well model [46]. In this work, the adequate precision ratio of the model was found to be 16.487, which implied that there was an adequate signal. The coefficient of variation (CV) defines reliability of the model and is the ratio of the standard deviation as a percentage of the mean value of the observed response [48]. If CV of the model is lower than 10%, the model can be considered reproducible, and in the present study, the CV value was 5.45.

3.3.2. Response Surfaces and Contour Plots for BG Adsorption

The 3-dimensional response surface (3D) and 2-dimensional contour plots (2D) (Figure 6) were plotted based on the quadratic model in order to investigate the interactive effect of the tested parameters on the removal of BG. The response surface plots and corresponding contour plots were utilized to understand the interaction and individual effects of the chosen independent factors [49]. Response surface plots can be taken into account as an approach to estimating the removal yield (%) for different values of the studied parameters. Furthermore, 2D contour plots are beneficial to determine the type of interactions between the tested factors. Figure 6(a) illustrates the response surface and contour plots as a function of the initial BG concentration and adsorbent mass while the time was kept constant at 80 min. As shown, with the increasing of the initial BG concentration, the removal yield of dye was decreased. The decrease in dye removal percentage at high BG concentrations could be due to the saturation of dye-binding sites of the adsorbent. Increasing adsorbent mass ensured more surface area, thereby leading to more binding regions for the adsorption of dye onto poly(VP-co-An)/zeolite [50]. Figure 6(b) shows the interaction effect of the adsorbent mass and the time on the dye removal yield when the initial BG concentration was constant at 62.50 mg/L. It was observed that the BG removal efficiency increased with enhancing in adsorbent mass. It can be concluded that the availability of active sites increased with increasing the adsorbent mass [51, 52]. Figure 6(c) shows the effects of initial BG concentration and time on the removal yield of BG while the adsorbent mass was kept constant at 0.15 mg/50 mL. It was observed that at lower BG concentrations, the ratio of solute concentrations to free reactive adsorbent sites is lower and speeds dye removal which leads to an increment in dye adsorption, and at higher concentrations due to the saturation of adsorption areas the removal yield was reduced [52, 53].

The effect of selected adsorption parameters was optimized to reach optimal removal yield based on using the desirability function at the CCD [54, 55]. According to this desirability function, the optimum values were found to be an adsorbent mass of 0.20 g/50 mL, an initial BG concentration of 40.20 mg/L, and a time of 121.60 min, which will result in 99.91% removal of BG with 0.1000 desirability.

4. Conclusion

This study showed that the poly(VP-co-An)/zeolite composite prepared by FRP is an effective adsorbent for the removal of BG from aqueous solutions. A CCD was applied to determine the interactions among factors; the optimum values of three significant parameters, namely, time, adsorbent mass, and initial BG concentration, showed significant effect on adsorption process. The optimum values were found to be an adsorbent mass of 0.20 g/50 mL, an initial BG concentration of 40.20 mg/L, and a time of 121.60 min for maximum BG removal. A second-order quadratic model was developed to describe the behavior of adsorption process using Design Expert software. This mathematical model was used to develop contour plots for three factors effects. The significance of independent factors was tested by analysis of variance (ANOVA). According to ANOVA results, all of the variables have the linear effect on BG adsorption process. Both external mass transfer and internal particle diffusion played an important role in the adsorption kinetics of BG, and the adsorption process followed a pseudo-second-order kinetic model which has a higher correlation coefficient. Furthermore, the equilibrium isotherm was better explained by the Langmuir model. According to the results, this composite is of great importance since it has a higher maximum adsorption capacity than many of the adsorbents reported in the literature.

Data Availability

No data were used to support this study.

Conflicts of Interest

The authors declare that they have no conflicts of interest.