Treatment of oily wastewater from mining industry using electrocoagulation: Fundamentals and Process Optimization

Summary The present work proposes a careful study of the effects of different operational factors on the electrocoagulation process of synthetic oil-water emulsions. The experimental conditions used were the pH of the solution, current density, electrolyte concentration, oil concentration, and distance between the electrodes. The results showed that the proposed process is efficient for de-emulsification and oil removal, as it reached 100 % removal in some conditions. Furthermore, the experiments showed that the first two variables were the most impactful ones in the efficiency of COD removal. To determine the optimal conditions in which the electrocoagulation should operate, two different models were developed: a polynomial one, using Genetic Algorithm to optimize its coefficients, and an Artificial Neural Network one, which used the input variables to predict the amount of oil removed. The modelling results showed good agreements, as the polynomial and ANN models had R2 values of 0.89 and 0.99, respectively, showing that the neural model is the most suitable one to predict the COD removal and optimize the operational conditions. The SSE values of 1979.08 (polynomial) and 4.92 (ANN) also indicate the neural model is the best one.