Application of quadratic regression model for Fenton treatment of municipal landfill leachate

• Published in: Waste management (Elmsford) 2012-10, Vol.32 (10), p.1895-1902
• Main Authors: Ghanbarzadeh Lak, Mehdi, Sabour, Mohammad Reza, Amiri, Allahyar, Rabbani, Omid
• Format: Article
• Language: English
• Subjects: Analysis of variance (ANOVA); Applied sciences; Exact sciences and technology; Fenton treatment; General treatment and storage processes; Hydrogen Peroxide - chemistry; Hydrogen-Ion Concentration; Iron - chemistry; Landfill leachate; Models, Theoretical; Optimization; Other wastes and particular components of wastes; Pollution; Quadratic regression model; Regression Analysis; Time Factors; Waste Disposal, Fluid; Wastes; Water Pollutants, Chemical - chemistry
• ISSN: 0956-053X; 1879-2456
• DOI: 10.1016/j.wasman.2012.05.020

► Developing a simple model indicating relationships among different variables and the response. ► Positive effects of the initial pH, [H2O2]/[Fe2+] and [Fe2+] on COD removal. ► No considerable effect of coagulation pH on COD removal. ► Significant interaction between reaction pH and [Fe2+] on COD removal. ► Maximum COD removal at initial pH=5.8, [H2O2]/[Fe2+]=8, [Fe2+]=22,500mg/L, and coagulation pH=8.7. The effectiveness of Fenton process in municipal landfill leachate treatment, as a pre- or post-treatment approach, has been demonstrated. However, no general recommendations of universal validity could be made in the term of optimized conditions affecting Fenton process. At the first stage of this study, collected leachate samples from Aradkooh site, Tehran, Iran, were investigated using one-factor-at-a-time method to find out optimum coagulation pH and flocculation time values. Subsequently, the obtained results in addition to data issued previously by the authors were employed to develop a predictive model of the true response surface, namely chemical oxygen demand (COD) removal efficiency. Finally, the main parameters of Fenton procedure, i.e. initial pH, [H2O2]/[Fe2+] molar ratio, Fe2+ dosage, and coagulation pH were optimized taking advantage of the above-mentioned quadratic model. The derived second-order model included both significant linear and quadratic terms and seemed to be adequate in predicting responses (R2=0.9896 and prediction R2=0.6954). It was found that the interaction between initial pH and Fe2+ dosage has a significant effect on COD removal. While, the optimal [H2O2]/[Fe2+] molar ratio was independent of ferrous ion dosage. The optimum conditions for the maximum COD removal of 50.76% for the parameters of initial pH, [H2O2]/[Fe2+] molar ratio, Fe2+ dosage, and coagulation pH were found to be 5.8, 8.0, 22,500mg/L, and 8.7 respectively.