{Application of the eco-friendly subcritical water oxidation method in the degradation of epichlorohydrin} J. Serb. Chem. Soc. 84 (7) 757–767 (2019) UDC 547.583.1’422+628.16.094.3:66.092:574 JSCS–5224 Original scientific paper 757 Application of the eco-friendly subcritical water oxidation method in the degradation of epichlorohydrin ERDAL YABALAK*, İPEK TOPALOĞLU and AHMET MURAT GİZİR Department of Chemistry, Faculty of Arts and Science, Mersin University, Çiftlikköy Campus, TR-33343, Mersin, Turkey (Received 8 December 2018; revised and accepted 4 April 2019) Abstract: Degradation of epichlorohydrin was investigated using subcritical water oxidation method in the presence of hydrogen peroxide. Degradation rate was monitored by means of total organic carbon (TOC) analysis. The central composite design was used to determine optimal TOC removal conditions and modelling experimental process. The effects of all experimental variables (tem- perature, oxidant concentration of hydrogen peroxide and treatment time) on the TOC removal rates were evaluated and the theoretical prediction model was proposed. Reliability of the employed model was evaluated using ANOVA. F value and the p-value of the model were found to be 84.60 and <0.0001, respectively. 93.78 % of TOC removal was achieved in the degrad- ation of epichlorohydrin at 373 K of temperature and 75 min of treatment time using 90 mM of H2O2. Keywords: epichlorohydrin; degradation, ANOVA; eco-friendly method; sub- critical water. INTRODUCTION Epichlorohydrin (1-chloro-2-epoxypropane, EPC) is a raw compound used in the production of epoxy resins.1,2 It is widely used in the paper and pharma- ceutical industries, production of drinking water pipes and synthesis of cationic polyelectrolytes.3,4 Industrial wastewater and other contaminants are released during the EPC production process.5 EPC is listed by the EPA as being toxic to the aquatic environment.1 Central nervous system destruction, inflammation of lungs, local irrigation and nausea are some of the known harmful effects of EPC to human health.6,7 Moreover, EPC has been listed in group 2A by the Inter- national Agency for Research on Cancer as having the potential carcinogenic effect.5,6 Therefore, some effective methods for treating water containing EPC are needed. However, conventional methods are far from being a solution due to the * Corresponding author. E-mail: yabalakerdal@gmail.com https://doi.org/10.2298/JSC181208027Y 758 YABALAK, TOPALOĞLU and GİZİR hard-to-degrade structured contaminants. Further, these methods may cause the formation of even more harmful intermediates than the main pollutant.8 Subcritical water has been widely used, especially in oxidation, solubility, extraction and synthesis processes.8–11 In this study, subcritical water oxidation method (SWO), which is known as an environmentally friendly and effective method, was performed in the degradation of EPC. Subcritical water oxidation depends on the oxidation of organic compounds in the aqueous phase at a high temperature (373–647 K) and high enough pressure to keep water in the liquid state. Hazardous organic compounds and micro-pollutants can be degraded to harmless organic compounds such as CO2 and H2O by using this method. The oxidation process in which H2O2 is used as an oxidizer is effective in the applic- ation of wastewater containing medium and high concentration levels of organic carbon. In addition, H2O2 is a non-toxic and an ecological oxidant that does not cause any harmful by-product formation.11 The degradation of EPC was investigated using subcritical water oxidation and H2O2. Degradation rate was monitored by TOC analysis, which is the best- -known method for determining the organic content of an aqueous sample. The experimental parameters (temperature, time, oxidant concentration) and the opti- mum degradation percentage were assessed by the response surface method (RSM). RSM consists of a set of mathematical and statistical techniques and it is used to define the relationship between the response and the independent vari- ables of a system. Though several design methods of the response surface method have been studied, central composite design (CCD) arises as the most preferred one.8,11,12 EXPERIMENTAL Reagents and apparatus The analytical grade EPC and H2O2 were purchased from Sigma–Aldrich (St. Louis, MO, USA). N2 gas was provided from Linde gas (Turkey). Ultra-pure water (18 MΩ cm, 25 °C) was obtained using Millipore Milli-Q Advantage A10 apparatus (Darmstadt, Germany). Experiments were performed in the home-made stainless steel reactor which was given in the previous work in detail.8 TOC analyses of the stock solution (100 ppm) and treated samples were performed using TOC-L analyzer with an ASI-L autosampler (Shimadzu). Degradation method The experimental design of the independent variables was determined using CCD. The five levels of the three independent variables such as temperature, concentration of H2O2 and treatment time were assigned after preliminary experiments and established design model was applied in the experimental process (Table I). The degradation experiments were carried out according to the previously published methods which were briefly given below.8,11 150 mL of the stock solution of EPC was placed in the reactor followed by the specific amount of H2O2. The inner pressure of the reactor was fixed at 30 bar using N2 gas for providing the subcritical water medium. The reactor was heated to a certain temperature during a specific treatment DEGRADATION OF EPICHLOROHYDRIN USING SUBCRITICAL WATER OXIDATION 759 time. The mentioned amounts of H2O2, temperature and treatment time are given in Table II. 20 mL of treated sample was kept at 281 K after each run for further analyses. TABLE I. CCD model of the experimental variables along with their coded levels Factor Independent variable Coded levels –1.682 –1 0 1 1.682 x1 Temperature, K 352.55 373 403 433 453.45 x2 Concentration of H2O2, mM 7.96 25 50 75 92.05 x3 Treatment time, min 9.55 30 60 90 110.45 TABLE II. Experimental and predicted results of the TOC removal efficiency of EPC Run x1 x2 x3 TOC removal, % Exp. CCD pred. 1 403 (0) 50 (0) 60 (0) 88.05 87.56 2 403 (0) 50 (0) 60 (0) 87.29 87.56 3 373 (–1) 25 (–1) 90 (+1) 65.49 64.45 4 373 (–1) 75 (+1) 90 (+1) 93.78 92.43 5 403 (0) 50 (0) 60 (0) 85.94 87.56 6 433 (+1) 75 (+1) 30 (–1) 89.52 88.36 7 433(+1) 25 (–1) 30 (–1) 64.72 63.88 8 433(+1) 25 (–1) 90 (+1) 69.13 69.04 9 403 (0) 50 (0) 60 (0) 88.20 87.56 10 352.55 (–1.682) 50 (0) 60 (0) 56.39 60.25 11 373 (–1) 75 (+1) 30 (–1) 69.43 67.33 12 403 (0) 7.96 (–1.682) 60 (0) 47.60 49.96 13 403 (0) 50 (0) 60 (0) 88.38 87.56 14 403 (0) 50 (0) 60 (0) 88.05 87.56 15 433 (0) 75 (+1) 90 (+1) 89.68 91.26 16 403 (0) 50 (0) 110.45 (+1.682) 88.12 87.60 17 403 (0) 92.05 (+1.682) 60 (0) 93.32 94.07 18 403 (0) 50 (0) 9.55 (–1.682) 58.53 62.16 19 453.45 (+1.682) 50 (0) 60 (0) 82.54 81.79 20 373 (–1) 25 (–1) 30 (–1) 40.87 37.10 TOC method The TOC analysis was known as being a safe and practical method to measure the organic content of an aqueous sample.11,13 The TOC content of the stock solution and the treated samples was measured by TOC-L analyzer with an ASI-L autosampler (Shimadzu). The TOC removal percentages of the stock solution and the treated samples were calculated according to the equation given in the previous work. 8,11 CCD modeling RSM provides several advantages such as saving time, reagent, labour and etc. through representing efficient experimental designs and requiring a limited number of experiments.14 Not only does RSM reduce the number of experiments, but it also allows to determine the relationship between variables and the effect of the variables on the response.11 CCD, as one of the RSM models, provides an evaluation of interaction effects between the independent 760 YABALAK, TOPALOĞLU and GİZİR variables and the response and enables establishing the approximation equations for the pre- diction of the response.15 The CCD model was employed to establish the experimental design and the experimental parameters of each run, and they were demonstrated along with the experimental and The predicted TOC removal percentages in Table II. x1, x2 and x3 represent the temperature, the concentration of H2O2 and the treatment time, respectively. RESULTS AND DISCUSSION The experimental and the predicted results of the TOC removal efficiency of EPC were given in Table II along with the running parameters. The highest and the lowest experimental TOC removal rates were obtained to be 93.78 and 40.87 %, respectively at runs 4 and 20. Moreover, the predicted TOC removal of 92.43 and 37.10 %, which were obtained at run 4 and 20, respectively, show the accor- dance between the experimental and the predicted results. Also, these results proved the applicability of the employed CCD model. Statistical analysis of CCD modeling The significance of the model can be proved by means of statistical ana- lysis.8,11 Tables III and IV demonstrate the ANOVA results and the regression coefficients of the CCD model of the degradation of EPC, respectively. p-value and F value were obtained as 84.60 and <0.0001, respectively for the model. Both of the p-value and F value are at a satisfactory level.8,11 Thus, the employed CCD model can be used to navigate design, determine the combined effects of experimental factors on the response and achieve the approximation model. Moreover, x1, x2, x3, x1x3, x12, x22 and x32 are the other significant terms of the model. TABLE III. ANOVA results of the CCD model of the degradation of EPC Source Sum of squares df Mean square F value p-value prob > F Model 4976.32 9 552.92 84.60 < 0.0001 x1 560.09 1 560.09 85.60 < 0.0001 x2 2348.55 1 2348.55 359.34 < 0.0001 x3 781.42 1 781.42 119.56 < 0.0001 x1x2 16.53 1 16.53 2.53 0.1428 x1x3 246.42 1 246.42 37.70 0.0001 x2x3 2.55 1 2.55 0.39 0.5459 x1 2 493.16 1 493.16 75.46 < 0.0001 x2 2 435.63 1 435.63 66.65 < 0.0001 x3 2 289.90 1 289.90 44.36 < 0.0001 Residual 65.36 10 6.54 – – Lack of fit 61.15 5 12.23 14.53 0.0053 Pure error 4.21 5 0.84 Cor total 5041.68 19 – The reliability of the CCD model was also supported by regression and correlation analysis (Table IV). The 472.50 value of the predicted residual sum of squares (PRESS) indicates that the model fits each point in the design and differ- DEGRADATION OF EPICHLOROHYDRIN USING SUBCRITICAL WATER OXIDATION 761 ences between the actual and the predicted results are at an acceptable level.11 The R2 value of 0.9870 supports the above-mentioned findings. In addition, the obtained adjusted R2 (0.9754) and predicted R2 (0.9063) values are quite close to each other. This closeness demonstrates the high correlation level between the experimental and the predicted results of the employed model. TABLE IV. Regression coefficients of the CCD model Standard deviation 2.56 R2 0.9870 Mean 76.75 Adjusted R2 0.9754 Coefficient of variation value (C.V.), % 3.33 Predicted R2 0.9063 PRESS 472.50 Adequate precision 31.515 The accordance between the actual and the predicted values of the TOC removal of EPC was demonstrated in Fig. 1. This figure clearly shows the com- patibility within the points which represent the actual and predicted values. Almost all points (each point represent one run) are aligned on the line. Also, the closeness of adjusted R2 and predicted R2, which was mentioned above, supports this accordance: 1 2 3 1 2 1 3 2 2 2 2 3 1 2 3 7.56 1.44 5.55 0.56 5.50 4.49 6.40 13.11 5.858 87.56 Y x x x x x x x x x x x x + − − − +− −−− = + (1) Fig. 1. Correlation between actual and predicted values. The predicted values of EPC degradation percentages were obtained by sec- ond-order equation (Eq. (1)). The degradation rates can be predicted in the work- ing range of each system variables and the interaction of these variables and their contribution to the efficiency of the process can be analysed by this equation. Thus, the concentration of H2O2 was found to be the most effective variable on the TOC removal of EPC, following by the treatment time and the temperature. 762 YABALAK, TOPALOĞLU and GİZİR Evaluation of the combined effects of the experimental variables on the TOC removal of EPC Experimental TOC removal percentages of EPC were evaluated using three- -dimensional (3D) plots. These plots are useful to demonstrate the interactive effects of experimental variables on the response. Also, they allow easy evaluat- ion of the optimum conditions for the maximum theoretical TOC removal rate.8,11 The combined effects of the concentration of H2O2 and the temperature on the TOC removal of EPC at the fixed treatment time of 60 min were displayed in Fig. 2. The increase in the temperature favours the formation of hydroxyl and other radicalic species, thus allowing an increase in the TOC removal of target pollutant.11 It is clearly seen from Fig. 2 that the temperature values above 383 K and the concentration of H2O2 above 45 mM are adequate for the obtaining of the elevated TOC removal rates. Thus, the higher temperature and concentration of H2O2 should be seen as redundant. The broad red area of Fig. 2, which demon- strates the high yielded region, also supports this results. For instance, the TOC removal of EPC can be increased from 55.25 to 75.31% through doubling the 25 mM of concentration of H2O2 at 373 K and the fixed treatment time of 60 min. However, an extra 25 mM increase in the concentration of H2O2 at the same treatment time can only contribute 9% in the TOC removal. Moreover, increasing both of the temperature and the concentration of H2O2 to their highest levels (433 K and 75 mM, respectively) provides 94.30 % of the TOC removal at the fixed treatment time of 60 min. Fig. 2. Combined effects of the concentration of H2O2 and temperature on the TOC removal of EPC at the fixed treatment time of 60 min. Fig. 3 demonstrates the combined effects of the treatment time and the con- centration of H2O2 on the TOC removal of EPC at the fixed temperature of 400 K. It is clearly seen from this figure that high TOC removal rates can be achieved at DEGRADATION OF EPICHLOROHYDRIN USING SUBCRITICAL WATER OXIDATION 763 Fig. 3. Combined effects of treatment time and concentration of H2O2 on the TOC removal of EPC at the fixed temperature of 400 K. high concentration levels of H2O2 and in the short treatment time or moderate concentration level of H2O2 and in the moderate-long treatment time at the fixed temperature of 400 K. 87.59 % of TOC removal can be increased to 92.18 % by decreasing the concentration of H2O2 from 75 to 55 mM and increasing the treat- ment time from of and 40 to 75 min, at the fixed temperature of 400 K. However, it is hardly possible to increase the TOC removal from 97.28 to 97.69 % by pro- longing the treatment time from 75 to 90 min and adding 75 mM of the concen- tration of H2O2 at the fixed temperature of 400 K. Besides, the co-effect of the concentration of H2O2 and the treatment time was found to be crucial on the TOC removal of EPC. Though the concentration of H2O2 has a significant effect on the degradation of EPC, a specific time is required for efficient formation of free radicals from H2O2. Fig. 4 demonstrates the combined effects of the treatment time and the tem- perature on the TOC removal of EPC at the fixed concentration of H2O2 of 60 Fig. 4. Combined effects of treatment time and temperature on the TOC removal of EPC at the fixed concentration of H2O2 of 60 mM. 764 YABALAK, TOPALOĞLU and GİZİR mM. As can be seen from this figure, the red area is relatively wide and scattered. This means that 60 mM of the concentration of H2O2 offers an appropriate medium for obtaining the elevated TOC removal. Nevertheless, the low levels of the temperature and the treatment time are not enough to achieve high efficiency at the fixed concentration of H2O2 of 60 mM. Thus, the temperature and the treat- ment time should be increased. The TOC of 62.88 % removal can be achieved at 373 K and 30 min of the treatment time at the fixed concentration of H2O2 of 60 mM. Only 10 units of the raise of temperature and the treatment time provide 13.21 % of the increase in the TOC removal. Moreover, an extra 10 units of the increase of temperature and the treatment time provides an extra 9.69 % increase in TOC removal. Fig. 5 demonstrates the cube plot of the CCD model for TOC removal rates of EPC. This figure facilitates the evaluation of response over changing all expe- rimental variables. The possible TOC removal percentages can be predicted by adjusting the experimental variables from –1 and +1 levels. Furthermore, this plot enables further predictions and demonstrates the required conditions. For instance, the treatment time and the concentration of H2O2 must be adjusted to their highest level and the temperature must be adjusted to its lowest level to achieve 92.43 % of TOC removal. Fig. 5. Cube plot of the employed CCD model. Fig. 6 displays the perturbation plot of the TOC removal of EPC. The opti- mal conditions for the efficient TOC removal percentages of EPC can be seen in this figure. The slope and the direction of curves for each variable provide the determination of the TOC removal rates. Fig. 6 shows that a reasonable TOC removal percentage can be obtained at 422 K of the temperature, 53 mM of the concentration of H2O2 and 61 min of the treatment time. DEGRADATION OF EPICHLOROHYDRIN USING SUBCRITICAL WATER OXIDATION 765 Fig. 6. Perturbation plot of TOC removal of EPC Validation of the CCD model A series of experiments were performed to validate the precision of the employed method. The Table V demonstrates the validation experiments and the obtained results with standard deviation values. The experimental conditions of each variable were selected in the working range, but not the same as the ones given in Table II. It is clearly seen from Table V that the experimental and the theoretical results are in agreement with each other. The differences between the experimental and the theoretical results are at a reasonable level, considering the standard deviation value of the model (2.56). Thus, the reliability of the emp- loyed model was proved by the validation experiments beside being statistically evaluated. TABLE V. Validation of the CCD model Run x1 x2 x3 TOC removal, % Exp. Pred. 1 383 60 40 75.19±0.87 76.09 2 420 40 60 85.01±1.23 83.51 3 400 70 80 95.84±0.96 97.07 CONCLUSION The degradation of EPC was extensively investigated using the eco-friendly subcritical water degradation method and a green oxidising agent, H2O2. The rate of the degradation of EPC was monitored by the measuring of TOC removal. The maximum TOC removal was achieved as 93.78 %, at the temperature of 373 K, 75 min of the treatment time, using 90 mM of H2O2. The co-effects of the main parameters such as the temperature, the concentration of H2O2, and the treatment time on the TOC removal rates were determined using the CCD modeling. The reliability of the employed CCD model was evaluated by ANOVA. The approxi- 766 YABALAK, TOPALOĞLU and GİZİR mation model for the TOC removal percentage of EPC was proposed and the optimal conditions for efficient TOC removal were evaluated. It was obtained that the concentration of H2O2 is the most effective factor on the TOC removal rates of EPC. И З В О Д ПРИМЕНА ЕКОЛОШКИ ПРИХВАТЉИВЕ РАЗГРАДЊЕ ЕПИХЛОРОХИДРИНА МЕТОДОМ ОКСИДАЦИЈЕ У ПОДКРИТИЧНОЈ ВОДИ ERDAL YABALAK İPEK TOPALOĞLU и AHMET MURAT GİZİR Department of Chemistry, Faculty of Arts and Science, Mersin University, Çiftlikköy Campus, TR-33343, Mersin, Turkey Проучавана је разградња епихлорохидрина методом оксидације у подкритичној води у присуству водоник-пероксида. Брзина разградње праћена је анализирањем TOC. За одређивање оптималних услова за смањење TOC коришћено је моделовање експери- менталног процеса конструисањем централног композита (central composite design). Оце- њен је утицај свих експерименталних променљивих (температура, концентрација окси- данса водоник-пероксида и време третирања) на брзину смањивања TOC, те је пред- ложен теоријски модел. Поузданост овог модела је оцењена методом ANOVA. По моделу, вредности за F и p биле су 84,60 односно <0,0001. Постигнуто је 93,78 % сма- њења TOC у разградњи епихлорохидрина на температури од 373 K, 75 min третирања коришћењем 90 mM H2O2. (Примљено 8. децембра 2018, ревидирано и прихваћено 4. априла 2019) REFERENCES 1. C. Sarzanini, M.C. Bruzzoniti, E. Mentasti, J. Chromatogr., A 884 (2000) 251 (https://doi.org/10.1016/S0021-9673(00)00252-1) 2. H. Hindsø Landin, T. Grummt, C. Laurent, A. Tates, Mutat. Res. 381 (1997) 217 (https://doi.org/10.1016/S0027-5107(97)00171-1) 3. L. Lucentini, E. Ferretti, E. Veschetti, V. Sibio, G. Citti and M. Ottaviani, Microchem. J. 80 (2005) 89 (https://doi.org/10.1016/j.microc.2004.12.003) 4. M.C. Bruzzoniti, S. Andrensek, M. Novic, D. Perrachon, C. Sarzanini, J. Chromatogr., A 1034 (2004) 243 (https://doi.org/10.1016/j.chroma.2004.02.033) 5. J. Gaca and W.Grażyna, Talanta 70 (2006) 1044 (https://doi.org/10.1016/j.talanta.2006.02.017) 6. M. Lasa, R. Garcia, E. Millán, J. Chromatogr. Sci. 44 (2006) 438 (https://doi.org/10.1093/chromsci/44.7.438) 7. K. Chandrasekara Pillai, G. Muthuraman, I.-S. Moon, Electrochim. Acta 232 (2017) 570 (https://doi.org/10.1016/j.electacta.2017.02.147) 8. E. Yabalak, J. Environ. Chem. Eng. 6 (2018) 7132 (https://doi.org/10.1016/j.jece.2018.10.010) 9. E. Yabalak, A.M. Gizir, J. Serb. Chem. Soc. 78 (2013) 1013 (https://doi.org/10.2298/JSC120321123Y) 10. Y. Nural, M. Gemili, E. Yabalak, L. De Coen, M. Ulger, Arkivoc 2018 (5) (2018) 51 (https://doi.org/10.24820/ark.5550190.p010.573) 11. E. Yabalak, J. Environ. Sci. Health, A 53 (2018) 975 (https://doi.org/10.1080/10934529.2018.1471023) 12. J.L. Pilkington, C. Preston, R.L. Gomes, Ind. Crop. Prod. 58 (2014) 15 (https://doi.org/10.1016/j.indcrop.2014.03.016) DEGRADATION OF EPICHLOROHYDRIN USING SUBCRITICAL WATER OXIDATION 767 13. A. Llop, E. Pocurull, F. Borrull, Water Air Soil Pollut. 197 (2009) 349 (https://doi.org/10.1007/s11270-008-9816-7) 14. S. Karimifard, M.R.A. Moghaddam, Sci. Total Environ. 640–641 (2018) 772 (https://doi.org/10.1016/j.scitotenv.2018.05.355) 15. C. R. T. Tarley, G. Silveira, W. N. L. dos Santos, G. D. Matos, E. G. P. da Silva, M. A. Bezerra, M. Miró, S. L. C. Ferreira, Microchem. J. 92 (2009) 58 (https://doi.org/10.1016/j.microc.2009.02.002).