The evolving demands of drinking water treatment necessitate processes capable of removing a diverse suite of contaminants. Biofiltration can employ biotransformation and sorption to remove various classes of chemicals from water. Here, pilot-scale virgin anthracite-sand and previously used biological activated carbon (BAC)-sand dual media filters were operated for ∼250 days to assess removals of 0.4 mg/L ammonia as nitrogen, 50-140 μg/L manganese, and ∼100 ng/L each of trace organic compounds (TOrCs) spiked into pre-ozonated Colorado River water. Anthracite achieved complete nitrification within 200 days and started removing ibuprofen at 85 days. Limited manganese (10%) removal occurred. In contrast, BAC completely nitrified ammonia within 113 days, removed all manganese at 43 days, and exhibited steady state removal of most TOrCs by 140 days. However, during the first 140 days, removal of caffeine, DEET, gemfibrozil, naproxen, and trimethoprim decreased, suggesting a shift from sorption to biotransformation. Acetaminophen and sulfamethoxazole were removed at consistent levels, with complete removal of acetaminophen achieved throughout the study; ibuprofen removal increased with time. When subjected to elevated (1 μg/L) concentrations of TOrCs, BAC removed larger masses of chemicals; with a subsequent decrease and ultimate cease in the TOrCs spike, caffeine, DEET, gemfibrozil, and trimethoprim notably desorbed. By the end of operation, anthracite and BAC exhibited equivalent quantities of biomass measured as adenosine triphosphate, but BAC harbored greater microbial diversity (examined with 16S rRNA sequencing). Improved insight was gained regarding concurrent biotransformation, sorption, and desorption of multiple organic and inorganic contaminants in pilot-scale drinking water biofilters.

Pilot‐scale data were generated to improve operational guidance for fine bubble diffusion (FBD) and sidestream addition (SSA) systems with respect to unaccounted‐for ozone exposure—i.e., concentration times time (C × T), and bromate formation during ozone dissolution. In FBD systems, results showed significant ozone dissolution C × T (0.31–2.85 mg‐min/L) and bromate formation (0.8–9.1 µg/L) occurring in the bubble column, which is not included in regulatory compliance C × T calculations. Utilities may consider taking contactors out of service to minimize contact time and corresponding bromate formation in the dissolution zone. In SSA systems, ozone dosages (7.2–18.6 mg/L) were transferred to approximately 20% of the overall plant flow for ~5 s, which produced bromate (2.6–5.7 µg/L). As the contact time exceeded the design guidance in the sidestream flow (>30 s), bromate formation became far more significant (40–140 µg/L). When the sidestream flow is operated with short contact times (~5 s), bromate formation can be minimized.

This article discusses how dosing chlorine and chlorine dioxide concurrently provides excellent disinfection inactivation and limits chlorite formation. This was done when Aurora (Colorado) Water's 80‐mgd Griswold Water Treatment Facility was rebuilt in 1999. A 2002 needs assessment indicated that a similar strategy should be implemented during expansion of Aurora Water's 80‐mgd Wemlinger Water Treatment Plant (WTP). However, at the Wemlinger WTP, ClO2 formation rose higher than expected during startup, forming at up to 85 percent of the ClO2 dose. Chlorine dioxide also decayed more rapidly than expected. The article discusses how this situation was addressed along with other lessons learned at the Wemlinger WTP, the importance of monitoring and reporting, and the power of coapplication.

Ozone dissolution system design is important for meeting transfer efficiency (TE) goals. Large sidestream pump flow (L) and high venturi inlet pressure improves TE but increases operating cost. Ozone TE was examined at a 25 gpm (97-Lpm) pilot-scale sidestream system with (SS w-dg ) and without (SS wo-dg ) degas separation. Under constant ozone dose conditions, process operating parameters were varied including sidestream gas/liquid (G/L) ratio, venturi-inlet water pressure, venturi-outlet water pressure, feed gas pressure, and ozone gas concentration. Performance results included determination of TE, ozone exposure (CT HDT ), and hydraulic detention time (T HDT ). Several design aspects of sidestream ozone systems were examined to improve mass transfer by using remixing devices, protecting ozone gas piping from corrosion, calculating sidestream ozone residual, and driving force for mass transfer. Moisture contamination of ozone supply lines may cause corrosion and/or decomposition of ozone gas that releases heat and destroys ozone. Ozone gas piping design is critical to prevent trapping water that might enter gas pipe during power outage or when units are offline. During plant operation below design flow, multiple constant speed pumps or variable speed pumps were evaluated to reduce overall operating costs. ARTICLE HISTORY

The removal of biodegradable ozone by‐products was evaluated at pilot scale using a fixed‐bed biofilm reactor (FBBR) containing spherical plastic support media. Six FBBRs were operated in parallel with different media sizes (1‐, 1.25‐, or 2‐in. diameter) and empty bed contact times (6 or 12 min). Influent water was provided from a full‐scale water treatment plant after ozonation, coagulation, and flocculation processes. After seven months of operation, pseudosteady‐state conditions were achieved with up to 50% removal of assimilable organic carbon and up to 40% reduction in ultraviolet absorbance at 254 nm. Increases in FBBR effluent turbidity and head loss were also indicative of biomass development and sloughing. Process efficiency deteriorated because of the consumption of the biomass by snails and other invertebrates.