Ozonation in Tandem with Biosand Filtration to Remove Microcystin-LR

Abstract

A hybrid ozonation-biofiltration approach is evaluated to understand the necessity and concentration of ozone dose in removing the micropollutant microcystin-LR (MC-LR). To simulate real polluted water, three levels of natural organic matter—1, 2, and 5 mg/L—and cyanobloom intensity—low, medium, and high—under ozone exposure times—C1: 0.8 mg×min/L and C2: 1.6 mg×min/L—were studied (18 combinations in total). The feasibility of filter bioaugmentation (postozone treatment) using known MC-LR degraders Arthrobacter ramosus (Filter FA) and Bacillus sp. (Filter FB) is also discussed and compared with the feasibility of a noninoculated sand filter. Overall, the bioaugmented sand filters, FA and FB, enhanced filter performance by 19.5% and 10.5% for C1 samples and 6% and 2% for C2 samples, respectively, in terms of MC-LR removal. All three filters, including the control (FC), showed a negative correlation (FA: −0.987; FB: −0.973; FC: −0.977) between “residual ozone” and “MC-LR removal due to ozonation.” However, A. ramosus (Filter FA) showed strong resilience toward the residual ozone (0.1–0.4 mg/L) and did not affect MC-LR removal due to filtration as much as it affected Filters FB and FC. Only Filter FA showed a significant difference (p-value: 0.047) between bloom condition and MC-LR removal that showed less removal of the latter at higher bloom intensity and vice versa. Statistical analysis, too, suggested a strong influence of natural organic matter (NOM) on filter performance for MC-LR removal. Also, protein phosphatase inhibition assay (PPIA) toxicity showed less toxic by-product formation when native bacteria were co-cultured and inoculated with A. ramosus and Bacillus sp.) in a sand filter. Hence, combined ozonation-biofilter treatment using co-inoculation may simplify (eco)toxicological and biotransformation research. This will enable the study of diverse contaminants under other environmental parameters.