Search

This study examined the impact of ozone (O3) and hydrogen peroxide+ozone (H2O2+O3) on organic matter and subsequent DOC and DBP precursor uptake across biofilters. Organic matter fractionation, using liquid chromatography – organic carbon detection (LC-OCD), confirmed that both O3 and H2O2+O3 oxidized high molecular weight fractions (biopolymers and humics) creating smaller, more easily biodegraded compounds (low molecular weight acids). O3 decreased trihalomethane (THM) and haloacetic acid (HAA) formation potential (FP) by 23% and 15% respectively; H2O2+O3 resulted in an increase of 52% and 41%. Biological activated carbon (BAC) removed 6% of DOC; removals increased to 15% and 23% when preceded by O3 or H2O2+O3, respectively. BAC without pre-oxidation reduced THM FP by 12%, compared to 38% and 14% removal with O3 and H2O2+O3, respectively. HAA FP decreased by 36% for O3+BAC and 20% for H2O2+O3+BAC; better performance by O3 was likely due to increased HAA precursor formation by the AOP. BAC was found to perform better than anthracite-based biofilters for all parameters. This study highlights the impact of pre-oxidation on organic composition, the benefits of biofiltration when used in combination with pre-oxidants such as ozone, and the importance of pilot-testing to assess alternative treatment processes.

The objectives of this pilot-scale study were to evaluate biological acclimation of virgin granular activated carbon (GAC), quantify the impact of nutrient (phosphorus and nitrogen) enhancement and to compare the performance of parallel biologically active carbon (BAC) filters operated continuously or cyclically (12 h/day), with respect to removal of dissolved organic carbon (DOC) and disinfection by-product (DBP) precursors. Virgin GAC media outperformed biologically active carbon for an initial 4-month period in terms of DOC reduction (30%), as expected based on the superior performance associated with adsorption compared to biodegradation. Once the adsorptive capacity was exhausted and the media was biologically acclimated, the performance of the new GAC was statistically similar in terms of organic carbon and disinfection by-product precursor removal to a filter containing media harvested from a filter operating biologically for 12 years. Phosphorus addition to the filter influent (0.3 mg PO4-P/L; C:N:P = 400:1:30) had a small impact on DOC (3 ± 2%) and THM formation potential (5 ± 3%) reduction when compared to biofiltration without nutrient enhancement. Ammonia nitrogen added to the filter influent (0.8 mg NH4-N/L; C:N:P = 200:40:1) was completely consumed through the biofilter; however, no impact on measured performance parameters was observed. Cyclical operation of full-scale biofilters resulted in a modest, but significant improvement in DOC removal (3 ± 2%) when compared to continuously operated pilot filters. Genotyping of both cyclically and continuously operated biofilters (with varying GAC ages) showed similar community composition; however, differences in the phylogenetic diversity of the samples were evident.

The impact of source water humic substances (HS) on the rejection of selected pharmaceutically active compounds (PhACs) and endocrine disrupting compounds (EDCs) via nanofiltration was investigated using two source waters with significantly different HS concentrations (Lake Ontario and Otonabee River, Ontario, Canada). To remove the effects of colloidal particles, raw waters were pre-treated with ultrafiltration, which removed most (up to 93%) of the colloidal and suspended particles but allowed passage of indigenous HSs. The effect of cation concentrations was examined by spiking beyond ambient levels in UF-pretreated Otonabee River water. There were no significant differences in rejection of the ionic PhAC/EDCs from the raw and UF-pretreated water matrices examined. Rejection of neutral compounds in colloid-free natural waters increased in the presence of HSs. The negative impact of cations on rejection was more prominent in the HS-dominated River water when compared to the Lake water. Increased compound rejection was not linearly related to higher concentrations of humic substances in the source waters. The rejection of neutral compounds increased due to the presence of humic substances up to a certain concentration, above which additional HS had no additional effect on PhAC/EDCs removal via nanofiltration.

Coagulation as pre-treatment to ultrafiltration (UF) was optimized for the removal of biopolymers, i.e., a primary UF foulant, for three different natural water matrices. The impact of pre-coagulation on membrane reversible and irreversible fouling, as well as the retention of organic micropollutants, was investigated at bench scale. Jar test experiments indicated that the optimum alum dosage for removal of biopolymers, based on a point of diminishing returns analysis, was relatively low (0.5 mg/L as coagulant; 0.05 mg/L as Al 3+ ). This dose was effective at reducing membrane reversible and irreversible fouling (up to 48%) for waters with higher concentrations of organics (>4 mg/L as DOC) over 24 h of permeation and backwash cycles. Biopolymers were identified as contributing to both reversible and irreversible fouling. The retention of organic micropollutants was relatively low for UF alone ( K ow > 2). This study demonstrates that a low dose of coagulant, optimized for biopolymer removal, may reduce membrane fouling and may provide value added for the retention of some organic micropollutants; however dosages depend on the specific water being treated, as well as treatment-related water quality targets.

Biopolymers have been identified as the primary source of fouling for ultrafiltration (UF) membranes, however the relative contribution of its components (e.g., proteins and polysaccharides) to fouling are still unclear. Previous natural water studies have shown that applying low doses of coagulant (