Optimization of Electrocoagulation for Natural Organic Matter Removal and Its Impact on Disinfection By-Products Formation

Abstract

A batch electrocoagulation (EC) process using aluminum electrodes was optimized for the removal of natural organic matter (NOM) and subsequent reduction in total trihalomethane (TTHM) formation potential from synthetic (SynW) and river water (RW) samples. Optimum operating conditions were found to be 45 min of electrolysis time, an initial NOM concentration of 10 mg/L, and an applied voltage of 15 V. Initial NOM concentrations tested were 3, 5, 10, 15, 20, 30, 40, and 50 mg/L and treatment efficiency increased with increasing initial concentration. The kinetic study demonstrated that NOM removal followed second-order kinetics. NOM removal resulted in the reduction of trihalomethanes (THMs) formation in treated water. THM concentrations were measured in untreated and EC-treated water. Bromodichloromethane had the highest removal (92.4%) compared to trichloromethane (89.5%) after 10 min. The TTHM concentration after a 10–30 min treatment was below the US Environmental Protection Agency (USEPA) maximum contaminant level of 80 μg/L. After 60 min, the TTHM concentration in the RW samples was reduced by 77.5%, which was considerably less than the reduction in the SynW samples (89.7%) within 10 min. The lower THMFP removal from RW may be due to interference from natural or anthropogenic constituents. The total amount of sludge generated in experiments was compared with the estimate derived using Faraday’s equation. A current efficiency of 1.4 was obtained, indicating super-faradaic behavior of the EC process at the optimum operating conditions. Fourier-transform infrared spectroscopy (FTIR) and X-ray diffraction (XRD) were used to characterize the generated sludge and entrapment of NOM in aluminum hydroxide [Al(OH)3] flocs as the most likely removal mechanism. At optimum operating conditions, the energy consumption was 2.391 kWh/m2. Thus, EC is an effective method for reducing NOM and THMFP from RW samples.