Abstract
Ferrioxalate is a visible light-responsive photocatalyst. The solar ferrioxalate/ process has high degradation efficiency because ferrioxalate is able to absorb light strongly at longer wavelength and generates hydroxyl radical with high quantum yield. Degradation of pesticide chlorothalonil in aqueous solution by ferrioxalate/ under solar irradiation was examined. The optimum operating conditions for treatment of a 300 mg/L chlorothalonil aqueous solution were obtained by using the central composite design of the response surface methodology. Under the optimum operating conditions (/COD molar ratio 2.75, /Fe3+ molar ratio 75, /C2H2O4 molar ratio 37.5, reaction time 90 min, and pH 3), COD, NH3-N, and TOC removal of 75.71, 47.11, and 54.33%, respectively, was achieved and the biodegradability (BOD5/COD ratio) improved from zero to 0.42. Model prediction and actual removal were in close agreement (<4% error). The solar ferrioxalate/H2O2 process is effective in pretreatment of the chlorothalonil aqueous solution for biological treatment.
1. Introduction
Chlorothalonil is a broad-spectrum of organochlorine pesticide (fungicide) used to prevent foliar diseases in vegetable and ornamental crops in agricultural fields [1] and belongs to class IV of the World Health Organization classification of pesticides [2]. It is classified as a probable carcinogen, and the 4-hydroxy chlorothalonil transformation product is more soluble, more stable, and, for some species, more toxic than its parent compound [3]. Chlorothalonil has the potential to contaminate water bodies adjacent to its point of use by spray drift, runoff, or sediment transport. Chlorothalonil has been detected in surface water [4, 5], inland waterways [6], natural water [7], rainfall [8], and air samples [9] generally adjacent to agricultural areas where it was applied. Chlorothalonil is sufficiently persistent to undergo long-range atmospheric transport on a regional scale [10] and can also exhibit chemical stability and resistance to biodegradation [11–13]. Reports have indicated that a minimum of 50 μg/L concentration of pesticide is toxic to guppy fish [14].
Advanced oxidation processes (AOPs) constitute a promising technology for the treatment of water and wastewater containing recalcitrant organic compounds with high toxicity and low biodegradability [15]. Oxidation technologies have shown that a partial oxidation of toxic water may increase its biodegradability [16, 17]. Oxidation with Fenton’s reagent is based on hydroxyl radical produced by catalytic decomposition of hydrogen peroxide in reaction with ferrous ion [18]. In the photo-Fenton process, the rate of radical formation is increased by photoreactions of and/or that produce radical directly or regenerate [19], thus increase the efficiency of the process. Ferrioxalate is a visible light-responsive photocatalyst. The solar ferrioxalate/ process has high degradation efficiency because ferrioxalate is able to absorb light strongly at longer wavelength and generates radical with high quantum yield [20]. The quantum yield of regeneration greatly increases when complexes with a carboxylic anion, such as oxalate [21]. The ferrioxalate complex, , is highly photosensitive, and reduction of to , through a photo-induced ligand to metal charge transfer, can occur over the ultraviolet and into the visible (out to ~ 550 nm):
The reactions can be collapsed into one reaction, since the short lifetime of the oxyl radical, , should preclude it from participation in other reactions, and its decarboxylation product, , is not involved in any other significant reactions:
There are no other significant photochemical reactions (e.g., photolysis) because the molar extinction coefficients of the reactants are such that ferrioxalate is the predominant absorber. The produced then generates radical via the Fenton reaction:
In the presence of a sufficient excess of oxalate, will coordinate with either two or three oxalate ligands. As with the photo-Fenton reaction, iron cycles between oxidation states and the production of hydroxyl radical is limited only by the availability of light, , and oxalate, the latter two of which are depleted during the reaction. UV-vis/ferrioxalate/ treatment of aniline wastewater [22], dyehouse waste [23], reactive dyes [24], orange II [25], and phenolic pollutants [26] have been reported. There is no report on degradation of pesticide chlorothalonil by solar ferrioxalate/ process.
The present study examined degradation of pesticide chlorothalonil in aqueous solution by solar ferrioxalate/ process in terms of chemical oxygen demand (COD), ammonia nitrogen () and total organic carbon (TOC) removal, and biodegradability (/COD ratio) improvement. The treatment was optimized by using the central composite design (CCD) of the response surface methodology (RSM).
2. Materials and Methods
2.1. Chemicals and Pesticide
Hydrogen peroxide (30%, w/w), oxalic acid , and ferric sulfate were purchased from R&M Marketing, Essex, UK. The pesticide chlorothalonil used to prepare aqueous solution was obtained from a commercial source and was used as received. Figure 1 shows the chemical structure of chlorothalonil.
2.2. Analytical Methods
Chemical oxygen demand (COD) was determined according to Method 5220D (closed reflux, colorimetric method) of the Standard Methods [27], where the sample contained hydrogen peroxide , to reduce interference in COD determination, pH was increased to above 10 so as to decompose hydrogen peroxide to oxygen and water [28, 29]. TOC analyzer (Model 1010, O & I Analytical) was used for determining total organic carbon (TOC). The pH was measured by a pH meter (HACH sension 4) and a pH electrode (HACH platinum series pH electrode model 51910, HACH Company, USA). Biodegradability was measured by 5-day biochemical oxygen demand (BOD5) test according to Method 5210B (seeding procedure) of the Standard Methods [27]. The treated pesticide aqueous solution was adjusted to pH 7 before the BOD5 test. Ammonia nitrogen was measured by the Nessler method [30]. DO was measured using YSI 5000 dissolved oxygen meter. The seed for the BOD5 test was obtained from a municipal wastewater treatment plant.
2.3. Chlorothalonil Aqueous Solution
Chlorothalonil aqueous solution was 300 mg/L of chlorothalonil in distilled water. It was prepared weekly and stored at 4°C. The characteristics of the aqueous solution were COD 350 mg/L, 1.58 mg/L, and TOC 94.49 mg/L. The high chlorothalonil concentration was chosen to reflect the concentration in pesticide-manufacturing wastewater and to explore the oxidation potential of the ferrioxalate/ process.
2.4. Experimental Procedure
Batch experiments were conducted with 200 mL of chlorothalonil aqueous solution in a 250 Pyrex beaker, placed in a SolSim solar simulator photoreactor (Luzchem Research Inc., Gloucester, ON, Canada), with solar intensity 0.85 kW/m2. The required amount of and was added to the aqueous solution and mixed by a magnetic stirrer to ensure complete homogeneity during reaction. Thereafter, necessary amount of was added to the mixture with simultaneous adjustment to pH 3 by using . The time at which was added to the mixture was considered as the beginning of the experiment. Samples were taken at preselected time intervals and filtered through 0.45 μm membrane filter for determination of COD, , and TOC, and when required BOD5.
2.5. Optimization and Response Surface Modeling
Design expert software Version 6.0.7 [31] was used for statistical design of experiment and data analysis. Central composite design (CCD) of the response surface methodology (RSM) was used to optimize the operating conditions (variables) of the treatment because it is well suited for fitting a quadratic surface, which usually works well for process optimization, and it is the experimental design mostly utilized for the development of analytical procedure as against three-level factorial design which is not frequently used and has been limited to the optimization of two variables [32]. The variables were simultaneously changed in a central composite circumscribed design. The values of the variables molar ratio, molar ratio, molar ratio, and reaction time were set at three levels: −1 (low), 0 (central) and +1 (high), and the total number of experiments with three factors was obtained as 30 , where is the number of factors (which equals 4 in this case). Twenty four experiments were augmented with six replications at the design center to evaluate the pure error and carried in randomized order as required in the circumscribed composite design. The variables molar ratio, molar ratio, molar ratio, and reaction time were studied in the range 1.5–4.0, 25–50, 50–100, and 60–120 min, respectively. Chosen response parameters for the process were removal of COD, , and TOC. Table 1 shows the experimental design and the predicted response (removal). Regression models and statistical analysis, contour plots normal probability and plots were made. Model terms were evaluated by the value (probability) with 95% confidence level. The quality of fit of the polynomial model was expressed by the coefficient of determination . The optimum operating conditions (variables) were identified from the contour plots and response equation simultaneously. The following response equation describing an empirical second-order polynomial model was used to assess the predicted results: where is the dependent response; is the constant coefficient; , , and are the coefficients for the linear, quadratic, and interaction effect; and are the factors (i.e., ,, and molar ratio and reaction time); signifies the number of independent variables, and is the random error [32]. The result was calculated as the sum of a constant , four first-order effects (A, B, C, and D), four second-order effects and four interaction effects (AB, AC, BD, and CD).
3. Results and Discussion
Based on the experimental design, predicted and actual removal (average of triplicate experimental results) are shown in Table 1.
3.1. Regression Models and Statistical Analysis
To ascertain the suitability of the regression model, assess the interaction between the independent variables (operating conditions) and the dependent variables (responses), and subsequently obtain the “goodness of fit”, analysis of variance (ANOVA) was performed. Fisher -test value, -value, coefficient of determination , and adequate precision (A.P) are shown in Table 2. -test value is a measure of variation of the data about the mean [26]. A -value less than 0.05 indicates the suitability of the proposed models for treatment as there is no lack-of-fit. The models for COD, , and TOC removal were significant by the -test at 95% confidence level employed as all responses had a -value < 0.05, and therefore the removal fits the data well. The coefficient of determination is the proportion of variability in a data set which indicates whether the empirical model is good enough for the quadratic fit to navigate the design space defined by the CCD [27]. The value gives the proportion of the total variation in the response predicted by the model to the actual data. The values were 0.8031 (COD), 0.7579 , and 0.8196 (TOC). Adequate precision (A.P.) ratio compares the range of the predicted value at the design points to the average prediction error. Ratios greater than 4 indicate adequate model discrimination and can be used to navigate the design space defined by the CCD [27]. The A.P. for all the responses was greater than 4. The ANOVA results indicate adequate agreement between the model prediction and actual removal. The following fitted regression models were obtained to quantitatively investigate the effects of A: molar ratio, B: molar ratio, C: molar ratio, and D: reaction time on COD, , and TOC removal, respectively. COD removal, removal, TOC removal,
In (7), (8), and (9), the values of the sum of a constant , (74.68, 43.88, and 51.19) represent the percentage removal of COD, , and TOC, respectively. The positive sign indicates that the variable is directly proportional to the response (COD, , and TOC removal), and the negative sign indicates that the variable is inversely proportional to the response.
3.2. Process Analysis
Visualization of the predicted model equation can be obtained from the contour plot [25]. A contour plot is a two-dimensional display of the surface plot, and, in the contour plot, lines of constant response are drawn in the plane of the variables. The contour plot helps to visualize the shape of a response surface. When the contour plot displays ellipses or circles, the center of the system refers to a point of maximum or minimum response. Sometimes, contour plot may display hyperbolic or parabolic system of the contours [28]. Figures 2, 3, and 4 show the contour plots for COD, , and TOC removal. Decreasing oxalate (increasing molar ratio) and increasing (increasing molar ratio) will reduce COD removal, increasing oxalate (decreasing molar ratio) and increasing (increasing molar ratio) will reduce removal, and increasing oxalate (decreasing molar ratio) and decreasing (decreasing molar ratio) simultaneously or one at a time will reduce TOC removal at molar ratio 75.0 and reaction time 90 min in all three cases. The adequacy of the models was also evaluated by the residuals that is, difference between the predicted and the actual response value. Plot of predicted versus actual removal (Figures 5, 6, and 7) indicates that there is no abnormalities in the model as all data were found around the line of “best fit”.
3.3. Optimization and Model Verification
Numerical optimization was used to determine the optimum operating conditions for COD, , and TOC removal. Based on the response surface and desirability functions (figure not shown), the optimum operating conditions were obtained. In this case, all responses were targeted to be in range and were goaled to be maximized. The optimum conditions were obtained for highest desirability at molar ratio 2.75, molar ratio 75, molar ratio 37.5, and reaction time 90 min at pH 3. Under the operating conditions, 72.71, 45.47, and 52.20% removal of COD, , and TOC, respectively, was predicted based on desirability function of 1.00 (Table 3). Actual removal under the optimum operating conditions is shown in Table 3. The model prediction and actual removal were in close agreement (<4% error).
3.4. Biodegradability
Under the optimum operating conditions ( molar ratio 2.75, molar ratio 75, molar ratio 37.5, reaction time 90 min, and pH 3), solar ferrioxalate/ treatment of the chlorothalonil aqueous solution improved the biodegradability (BOD5/COD ratio) from zero to 0.42, indicating that the treated pesticide aqueous solution was amenable to biological treatment [33].
3.5. Prospective Application
Complete depollution of pesticide (herbicide) aqueous solution by photoelectro-Fenton or electro-Fenton using boron-doped diamond electrode has been reported [34, 35]. However, solar ferrioxalate/ is a simple energy-efficient process and can be applied as pretreatment of pesticide wastewater to improve biodgradability for subsequent biological treatment.
4. Conclusions
Visible light-responsive photocatalyst ferrioxalate and under solar irradiation is effective in degradation of the pesticide chlorothalonil. The optimum operating conditions for solar ferrioxalate/ treatment of a 300 mg/L chlorothalonil aqueous solution obtained by using the central composite design of the response surface methodology were molar ratio 2.75, molar ratio 75, molar ratio 37.5, reaction time 90 min, and pH 3. Under optimum operating conditions, 75.71, 47.11, and 54.33% removal of COD, , and TOC, respectively was achieved and the biodegradability (BOD5/COD ratio) improved from zero to 0.42. Model prediction and actual removal were in close agreement. The solar ferrioxalate/ process is effective in pretreatment of chlorothalonil aqueous solution for biological treatment.
Acknowledgment
The authors are thankful to the management and authorities of the Universiti Teknologi PETRONAS for providing facilities for this research.