Removal of Pathogens and Chemicals of Emerging Concern by Pilot-Scale FO-RO Hybrid Units Treating RO Concentrate, Graywater, and Sewage for Centralized and Decentralized Potable Reuse

This study evaluated the water quality produced by forward osmosis-reverse osmosis (FO-RO) treatment at pilot scale of RO concentrate generated at centralized potable reuse facilities and of graywater and filtered sewage for decentralized potable reuse systems. Despite the concentrated nature of these challenging source waters, the concentrations of general water quality parameters in permeate produced by FO-RO treatment were comparable to those in conventional RO permeate produced at centralized facilities. However, the 4.3 mg of N/L ammonia in the permeate from FO-RO treatment of sewage would require high chlorine doses to leave a free chlorine residual. Graywater featured higher concentrations of 1,4-dioxane and chloroform compared to those of sewage. FO-RO treatment achieved >98% rejection of all of the 16-18 organic contaminants (non-disinfection byproducts) spiked into each of the three source waters, even though the rejection of benzotriazole was as low as 50% by the FO membrane alone. FO-RO treatment achieved ≥6.7-log removal of bacteriophage MS2 spiked into graywater and sewage, ≥5.4-log removal of native Escherichia coli (E. Coli) in graywater, and 7.9-log removal of native E. coli in sewage. However, the detection of MS2 and E. coli in the FO-RO permeates indicates that additional disinfection is needed. FO-RO rejection of chloroform and N-nitrosodimethylamine (NDMA) in FO feedwater was ∼75%. The trihalomethane concentrations exceeded regulatory limits when the FO-RO permeate from graywater treatment was chlorinated. While the NDMA concentration exceeded the 10 ng/L Notification Level during chlorination or chloramination of the FO-RO permeates from all three water sources, it also exceeded the Notification Level in the permeate of the primary RO treatment unit at the centralized reuse facility. In both cases, additional treatment (e.g., ultraviolet-based advanced oxidation) is needed to control NDMA. Overall, the results indicate that FO-RO treatment can be a potentially robust component of a treatment train producing high-quality water from treatment of RO concentrate at centralized facilities and graywater or filtered sewage for decentralized applications.