Treatment of Specific NDMA Precursors by Biofiltration

Removal of targeted N-nitrosodimethylamine (NDMA) precursors during non-optimized biofiltration of tertiary-filtered wastewater was investigated. The study evaluated removal for one spiked model chloramine-reactive precursor—i.e., ranitidine (RAN)—and three spiked model ozone-reactive precursors—i.e., daminozide (DMZ), 1,1,1′,1′-tetramethyl-4,4′-(methylene-di-p-phenylene)disemicarbazide (TMDS), and 2-furaldehyde dimethylhydrazone (2-F-DMH). Biofiltration was assessed using three parallel, anthracite-containing, 100 mL/min columns operated at three empty bed contact times (EBCTs) of 5, 10, and 20 min. Precursor removals for the 20 EBCT column were not significant for RAN and were up to 80.8 ± 16.2% for DMZ, 26.1 ± 20.6% for 2-F-DMH, and 24.3 ± 10.0% for TMDS. EBCT correlated only with TMDS removal, and dissolved oxygen concentration correlated with 2-F-DMH and TMDS removals, indicating the possible importance of aerobic conditions for these two compounds. Biotransformation was likely the main removal mechanism, and the order for precursor removal was DMZ >> 2-F-DMH ~ TMDS > RAN.