Abstract
Expanded polystyrene (EPS) waste was chemically recycled to a novel functional polystyrene-hydrazone (PSH) surface by acetylation of polystyrene (PS) and then condensation with phenyl hydrazine. The synthesized surface was characterized by the FT-IR and elemental analysis. Synthesized novel functional PSH surface was successfully applied for the treatment of phenol-contaminated industrial wastewater by solid-phase extraction. Multivariant sorption optimization was achieved by factorial design approach. 99.93% of phenol was removed from aqueous solution. FT-IR study showed the involvement of nitrogen of hydrazone moiety of synthesized surface for the uptake of phenol through the hydrogen bonding.
1. Introduction
Phenolic pollutants are found in wastewater as a result of various industrial processes such as wood preservation, coal conversion, petroleum refining, metal casting, and manufacturing of plastics, textiles, iron, steel, paper, and pulp [1–3]. The concentration of phenol in these wastewaters usually ranges from 100–1000 mgL−1. Removal of phenolic contaminants from wastewater is important before discharge into natural water because of their toxicity for human and aquatic bodies. Several methods have been reported for the removal of phenol from wastewater such as incineration, microbial degradation, bacterial and chemical oxidation, solvent extraction, electrochemical techniques, irradiation, and so forth [4]. These methods have serious shortcomings such as lack of purification, high costs, hazardous byproduct formation, and low efficiency. Therefore, attention has been paid to the development of attractive sorbents for removal of phenolic pollutants from wastewater [5]. Different polystyrene based materials such as benzoyl-PS-DVB [6], 2,4-dicarboxybenzoyl-PS-DVB [7], o-carboxybenzoyl-PS-DVB [8], acetyl-PS-DVB [9, 10], toluene-PS-DVB [11], diethylenetriamine-PS-DVB [12], sulfonated PS-DVB [13, 14], Amberlite XAD-4, and Amberlite IRA96C [15–17] have been reported for the removal of phenols and phenolic compounds. In this study, we focus on the novel route for chemical recycling of EPS waste to the new functional polystyrene-hydrazone (PSH) surface for the treatment of phenol contaminated industrial wastewater. Factorial design approach is applied for multivariant sorption optimization.
2. Materials and Methods
2.1. Adsorbent: Synthesis of Polystyrene-Hydrazone (PSH) Surface
Figure 1 shows the chemical route for conversion of EPS waste into functional polystyrene-hydrazone (PSH) surface. 3.0 g of EPS was dissolved into 100 mL of carbon tetrachloride (CCl4). Solution was filtered to remove any insoluble impurities by simple filtration. The filtrate was poured into round bottom flask contained 3.35 g of anhydrous aluminum chloride (AlCl3) and 1.98 mL of acetyl chloride (CH3COCl) was added drop wise with stirring and refluxed for 50 min at 60°C. The reaction mixture was settled at room temperature and worked up with 0.05 N HCl. Acetyl-PS (b) was filtered off, washed with 1.0 N sodium bicarbonate (NaHCO3) solution to remove excess of acid, washed with deionized water to remove base, and air-dried. 3.0 g acetyl-PS was taken into round bottom flask contained 50 mL of acidified distilled water and heated for 30 min at 50°C. 2.02 mL of phenyl hydrazine was added into the reaction mixture and refluxed for 60 min at 50°C. The final product, PSH (c), was filtered off, washed with deionized water, and air-dried.
2.2. Sorbate: Phenol Solution
A stock solution of phenol was prepared (1000 mgL−1) by dissolving appropriate amount of phenol (Merck) in 30% methanol. The stock solution was diluted with de-ionized water to obtained series of working solutions (5–55 mgL−1). pH of solutions was maintained at 9.0, 5.5, and 2.0 with 0.5 N NaOH, acetate buffer, and 1.0 N HCl, respectively. CH3COONa, CH3COOH, HCl, and NaOH were purchased from Merck (Germany).
2.3. Batch Experiments
All batch sorption experiments were performed in thermostated shaker at constant temperature of 25°C for a period of 10–180 min with shaking speed of 100 rpm using 250 mL stoppered conical flasks contained 10 mL of phenol solution of different concentrations (5–55 mgL−1) and pH (2–9). The adsorption of phenol onto PSH surface was conducted by shaking different weighted amounts (10–100 mg) of PSH surface with 10 mL phenol solutions (5–55 mgL−1). The concentration of phenol in the solution was analyzed via reverse phase HPLC (Hitachi model-655A-11) using Zorbax XDB-C18 column of dimensions; 150 × 4.6 mm, 5 μm. Phenol was eluted with methanol and water (75 : 25) at flow rate of 1.0 mL min−1 and was detected by UV-Visible detector (Hitachi 655A) at 254 nm. Equation (1) was used to calculate the percent of phenol adsorption: where and are the equilibrium and initial concentrations (mgL−1) phenol solutions, respectively.
2.4. Factorial Design
Classical sorption optimization requires large number of experiments to find optimum response of independent variables. Major drawback of this method is that it is based on variation of only one independent parameter at a time keeping constant the other parameters, so the combined effect of all the independent parameters cannot be studied simultaneously which could lead to unreliable results [18]. In factorial design approach interactions of two or more variables can be studied at a time, hence comparatively more reliable results with less number of experiments and minimum treatment time and cost can be obtained [19]. Factorial design approach is an experimental technique, designed to predict optimal response of variables. Central Composite Design (CCD) was chosen to study the effect of adsorbent amount (A, mg), pH of phenol solution (B), initial phenol concentration (C, mgL−1), and contact time (D, min) on the uptake of phenol. Each independent variable was studied at three different levels (high, medium, and low coded as +1, 0, and −1, resp.) as shown in Table 1. The CCD model consists of eighteen batch experiments; each experiment was performed twice to predict mean values for CCD analysis under Response Surface Methodology (RSM). Design of experiments was analyzed statistically by Stat Graphics plus for Windows 5.1 (Stat Point Technologies Inc. 2009) [20].
3. Results and Discussions
3.1. Characterization
Figure 2 represents the spectra for EPS, Acetyl-PS and PS-Hydrazone surface, the characteristic peak for C=O stretching at 1611.01 cm−1 in spectrum-A, supports the conversion of PS into acetyl-PS. The disappearance of peak for C=O at 1611.01 cm−1 in spectrum-B and appearance of additional peak for C=N and N–H stretching at 1411 cm−1 and 3292.5 cm−1, respectively, in spectrum-C supports the conversion of acetyl-PS into PS-hydrazone (PSH) surface.
The formation of acetyl-PS and PSH surface was confirmed by elemental analysis (CNHS Analyser, FLASH EA.1112). Acetyl-PS resulted as C, 82.43; H, 7.57.64; O, 10.01% and theoretically calculated values for C11H12O are C, 82.46; H, 7.55; O, 9.99%. Elemental analysis confirmed the successful introduction of acetyl group on PS. Elemental analysis of PSH surface resulted as C, 77.69; H, 6.51; N, 10.07; O, 5.73% and theoretically calculated values for C17H18N2 are C, 77.67; H, 6.52; N, 10.06; O, 5.75%, confirming the successful conversion of acetyl-PS into PSH surface.
3.2. Statistical Analysis
The fitting and accuracy of CCD model were estimated by analysis of variance (ANOVA) as given in Table 3. The ANOVA result indicated that lack of fit is not significant as (0.043 < 0.05), so the null hypothesis could not be rejected, as the CCD model would give poor or misleading results if it was an inadequate fit [23]. Residual and three-dimensional (3D) surface plots were examined to estimate the CCD model competency [18].
3.2.1. Interpretation of Residual Graphs
Figure 3(a) plots the residuals versus predicted values. The residual is the difference between the observed and the predicted values. All the residuals are scattered randomly about zero and all points were found in the range of +1.5 to −1.5, showing that the errors have a constant variance and confirmed the fitting of the model.
(a)
(b)
(c)
(d)
Figure 3(b) shows the plot for observed versus predicted values of percent removal of phenol on PSH surface. Actual values measure the percent removal data for a particular run and the predicted values were evaluated from the CCD model. Values of and were found to be 99.97% and 99.87%, respectively, indicating a close agreement between the predicted and observed values as shown in Table 2.
Figure 3(c) plot shows the normal probability versus residuals for the removal of phenol by PSH surface. Residuals show how well the model satisfies the assumptions of ANOVA whereas the residuals measure the number of standard deviations separating the actual and predicted values. Plot indicates that neither the response transformation was needed nor there was any apparent problem with normality.
Figure 3(d) shows the residual of each experimental run, plot shows the residual of each experiment are scattered randomly around the zero, and all points are found in the range of +1.5 to −1.5, showing that lack of fit is not significant for model.
3.2.2. Interpretation of 3D Response Surface Plots
The 3D response surface graph shows the combined effect of any two independent variables on the % adsorption of phenol keeping other parameters at their optimized conditions. Figure 4(a) shows combined effect of pH and agitation time on the removal of phenol by PSH surface at optimum initial concentration of phenol (5 mgL−1) and adsorbent dose (51 mg). The % adsorption of phenol increases with increase of agitation time and pH of phenol solution and becomes maximum at pH 7, with further increase of pH there is a slight decrease in % adsorption, which may be explained on the basis of decreasing the chances of hydrogen boding because of possible interaction of hydroxyl group of base with the acidic hydrogen of phenol and hydrazone moiety of surface.
(a)
(b)
(c)
Figure 4(b) shows combined effect of phenol concentrations and agitation time on the removal of phenol by PSH surface at optimum adsorbent dose (51 mg) and pH (4). The plot show the maximum % adsorption at initial concentration of 5 mgL−1 and agitation time of 67–80 min. The further increase in agitation time decreases the % adsorption which may be due to desorption of phenol from surface. Figure 4(c) shows combined effect of pH and adsorbent dose on % removal of phenol by PSH surface at optimum concentration of phenol (5 mgL−1) and agitation time (50 min). The phenol uptake increases with increase in pH and becomes maximum at pH 7; a slight increase is registered with increasing adsorbent dose. Optimum values obtained by CCD model are: adsorbent dosage, 51 mg; pH 7; phenol concentration, 5 mgL−1; agitation time, 50 min. Maximum adsorption of phenol on PSH surface was 99.93% achieved at optimum conditions.
3.2.3. Pareto Chart
Figure 5 shows the Pareto chart of the estimated effects in decreasing order of magnitude. The length of each bar is proportional to the standardized effect, which is the estimated effect divided by its standard error. This is equivalent to computing a -statistic for each effect. It was observed that for a 95% confidence level and eight degrees of freedom, the value is equal to 3.19. The vertical line can be used to judge which effects are statistically significant. Any bars which extend beyond the line correspond to effects which are statistically significant at 95.0% confidence level.
3.2.4. Main Effects Plot
Figure 6 shows the estimated % adsorption as a function of each experimental factor. In each plot, the factor of interest is varied from its low level to its high level, while all other factors are held constant at their central values. The % adsorption slightly decreases with an increase in adsorbent amount, slightly decreases and then become constant with increase of phenol concentration, potentially increases with increase in pH (7–9) and reaches maximum at pH 7 and decreases with further increase of pH due to interaction of acidic hydrogen of phenol and PSH surface with basic group, increases with increasing shaking time and decreases with further increase in shaking time due to desorption.
3.3. Isotherm Studies
Isotherm study describes sorption equilibrium. In this study isotherm study was performed by changing adsorbent concentration ranging 5–55 mgL−1 and keeping optimum other independent parameters (adsorbent dose, 51 mg; agitation time, 50 min; pH 7) at 25°C. Langmuir and Dubinin-Radushkevich (D-R) models were evaluated using (2) and (3), respectively. where is the adsorbed amount of phenol on PSH surface (mgg−1) and is the equilibrium concentration of phenol (mgL−1) while and are the Langmuir constants related to the monolayer sorption capacity (mgg−1) and affinity of the binding sites (L g−1), respectively, is related to the mean sorption free energy per mole of the sorbent when it is transferred from infinite distance in the solution to the surface of the solid, and is Polanyi potential and is equal to , where is temperature and is general gas constant (Jmol−1K−1). The isotherm constants and were calculate from the slope and intercept of plot between and . The isotherm showed good fit to the experimental data with good correlation coefficient (0.982). The characteristic separation factor of Langmuir isotherm, can be calculated by using (4) where is the initial phenol concentration and is the Langmuir constant. The numerical value of can be interpreted as ; irreversible, ; unfavorable, ; linear, and ; favorable [24]. The calculated values of were found in the range of 0.0057–0.1, indicated favorable nature of sorption.
D-R isotherm assumes no homogeneous surface of the sorbent material and a good linear relationship between and with correlation coefficient 0.973. The estimated value of mean sorption energy () was calculated, 7.93 k Jmol−1 from the slope of plot (). The magnitude of indicates the nature of sorption process; –16 k Jmol−1 (chemisorption) and k Jmol−1 (physisorption) [25]. On the basis of this observation it can be anticipated that sorption of phenol on PSH surface predominantly followed physisorption. The Langmuir and D-R parameters are summarized in Table 4.
3.4. Comparative Capacities for Phenol Adsorption
Different sorbents have been reported in literature for the removal phenol having different capacities. Table 5 shows the comparative capacities of PS based adsorbent for the adsorption of phenol from aqueous solutions. The capacity of PSH surface for the removal of phenol is comparable or better, which enable the synthesized surface to be more effective for the removal of phenol.
3.5. Possible Sorption Mechanism
Figure 7 shows the FT-IR spectrum for PSH surface plain (A) and phenol-loaded PSH surface (B) and the characteristic decrease in intensities of peaks at 3292.5 cm−1 and 1411 cm−1 indicated the involvement of nitrogen and hydrogen of PSH surface for the adsorption of phenol. The hydrazone moiety of PSH surface has participated for the uptake of phenol via hydrogen bonding as shown in Figure 8.
3.6. Method Validation
The optimum sorption conditions determined from mathematical model were validated by conducting sorption experiments at optimum conditions predicted by CCD model. The experimental and predicted removal values were found in good agreement as shown in Table 6.
3.7. Application of Method
The proposed method was succefuly applied for the treatment of industrial wastewater collected from Various industrial site areas in Pakistan. The % removal of phenol by PSH surface from each sample is given in Table 7.
4. Conclusion
The EPS waste was successfully chemically recycled to novel functional PSH surface. The synthesized PSH surface was applied for the treatment of phenol industrial wastewater. The multi-variant sorption optimization was achieved by factorial design approach. This study gives the solution of waste management problems caused by EPS waste along with phenol-contaminated water treatment technology. Adsorption capacities from Langmuir isotherm and D-R isotherm were calculated as 1.873 mmolg−1 and 1.962 mmolg−1, respectively.
Conflict of Interests
Authors do not have any conflict of interests with parties.