Comparison of the new Cl2/O3/UV process with different ozone- and UV-based AOPs for wastewater treatment at pilot scale: Removal of pharmaceuticals and changes in fluorescing organic matter

This work critically compared the removal of fluorescing PARAFAC components and selected pharmaceuticals (carbamazepine, fluoxetine, gemfibrozil, primidone, sulfamethoxazole, trimethoprim) from a tertiary wastewater effluent by different UV- and ozone-based advanced oxidation processes (AOPs) operated at pilot-scale. Investigated AOPs included UV/H2O2, UV/Cl2, O3, O3/UV, H2O2/O3/UV, and the new Cl2/O3/UV. AOPs comparison was accomplished using various ozone doses (0–9 mg/L), UV fluences (191–981 mJ/cm2) and radical promoter concentrations of Cl2 = 0.04 mM and H2O2 = 0.29 mM. Chlorine-based AOPs produced radical species that reacted more selectively with pharmaceuticals than radical species and oxidants generated by other AOPs. Tryptophan-like substances and humic-like fluorescing compounds were the most degraded components by all AOPs, which were better removed than microbial products and fulvic-like fluorescing substances. Removal of UV absorbance at 254 (UV254) nm was always low. Overall, chlorine-based AOPs were more effective to reduce fluorescence intensities than similar H2O2-based AOPs. The Cl2/O3/UV process was the most effective AOP to degrade all target micro-pollutants except primidone. On the other hand, the oxidation performance of pharmaceuticals by other ozone-based AOPs followed the order H2O2/O3/UV > O3/UV > O3. UV/Cl2 process outcompeted UV/H2O2 only for the removal of trimethoprim and sulfamethoxazole. Correlations between the removal of pharmaceuticals and spectroscopic indexes (PARAFAC components and UV254) had unique regression parameters for each compound, surrogate parameter and oxidation process. Particularly, a diverse PARAFAC component for each investigated AOP resulted to be the most sensitive surrogate parameter able to monitor small changes of pharmaceuticals removal.