Evaluation and Prediction of Removal Efficiency of Pharmaceuticals in the Simulated O3 and UV/H2O2 Process for Drinking Water Treatment Process in the Downstream of Nakdong River

Abstract

Objectives : In the case of pharmaceuticals with high possibility of inflow into the large drinking water treatment plant (DWTP) located in the downstream of the Nakdong River, we tried to evaluate the removal efficiency of pharmaceuticals in the both ozone (O3) and the UV/H2O2 treatment as an alternative of post-O3 process. It was intended to be used as data for the advanced WTP project by prediction of the removal efficiency in the O3 and UV/H2O2 processes with varying water quality conditions.Methods : O3 and UV/H2O2 process were performed for 19 kinds of pharmaceuticals in the sand-filtered water of DWTP. In order to evaluate the removal efficiency in deionized water (DI) and sand-filtered water (SFW) matrices, 19 pharmaceuticals were spiked at a concentration of 100 ng/L, respectively. In the O3 process, the specific O3 dose was 0.1∼2.0 gO3/gDOC (0.25∼5.0 mgO3/L). In the UV/H2O2 process, H2O2 (5 and 10 mg/L) was added to the sample before UV was irradiated (0∼1,500 mJ/cm2).Results and Discussion : In the case of simulated post-O3 process, the removal efficiency of high-ozone reactive pharmaceuticals (kO3 6.5×102∼2.6×106 M-1 s-1) was up to 92% at the specific O3 dose of 0.2 gO3/gDOC. However, the removal efficiency of iopromide (IPM) and primidone (PRM) was only 36∼45% in the same O3 dose (0.2 gO3/gDOC) due to the low O3 reactivity (kO3 < 1 M-1 s-1). A specific O3 dose of 2.0 gO3/gDOC (=5 mgO3/L) was required to achieve a removal efficiency of over 90% for IPM and PRM, indicating that these O3-refractory pharmaceuticals are difficult to control by O3 process. In the case of simulated UV/H2O2 (10 mg/L H2O2) process, the removal efficiency of 19 pharmaceuticals at the UV fluence of 500 and 1,000 mJ/cm2 were 63∼99% and 87∼99%, respectively, and caffeine (CFN) had the lowest removal efficiency. For the O3-refractory pharmaceuticals (i.e., IPM and PRM), the removal efficiency was higher in the UV/H2O2 process than that in the O3 process due to the high reactivity with OH radical (kOH = 3.3×109 and 5.2×109 M-1 s-1). Prediction of removal efficiency for pharmaceuticals in the O3 and UV/H2O2 process was performed using chemical kinetics model to evaluate the change in removal efficiency with varying DOM concentration. As a result of prediction model for O3, when the DOM concentration increased from 1.5 to 3.0 mg/L, the removal efficiency of IPM and PRM decreased by 22∼24% and 15∼24%, respectively. In the case of UV/H2O2 process (10 mg/L H2O2 and UV fluence of 500~1,000 mJ/cm2), the removal efficiency of 16 kinds of pharmaceuticals was reduced by 0∼29%, and the degree of reduction in the removal efficiency of CFN was the highest.Conclusions : As a result of evaluation and prediction of the removal efficiency of pharmaceuticals in the O3 and UV/H2O2 treatment processes, it is confirmed that the possibility of applying the UV/H2O2 treatment as an alternative process to the O3 to abatement of pharmaceuticals.