| dc.description.abstract | The persistence and widespread occurrence of emerging contaminants (ECs) such as perchlorate (ClO4-), pharmaceutical compounds, artificial sweeteners, and perfluoroalkyl substances in the environment has received increasing attention due to the adverse effects on human health and aquatic ecosystems they may cause. Conventional wastewater treatment plants (WWTPs) collect household sewage which contains various types of ECs. Emerging contaminants are consequently released to the environment due to incomplete removal of the ECs in the WWTPs. These ECs have been widely found in treated wastewater, surface water, groundwater, and even drinking water at concentrations ranging from ng L-1 to µg L-1. This thesis describes laboratory experiments for evaluating the effectiveness of passive treatment systems and photocatalytic technologies for removing these ECs from water. In addition, a field investigation is described for tracking wastewater downstream of WWTPs using potential tracers in a receiving river.
Laboratory column experiments were performed to evaluate the effectiveness of reactive media zero-valent iron (ZVI), organic carbon (OC, wood chips), and a mixture of (ZVI + OC) for simultaneous removal of NO3-, SO42-, and ClO4- from contaminated water associated with mining and blasting sites. Input NO3- (~10.8 mg L-1 NO3-N) was effectively removed through NO3- reduction to NH4+ in Column ZVI, through denitrification in Column OC, and through the combination of NO3- reduction by ZVI and denitrification by OC in Column (ZVI + OC). Input SO42- (~24.5 mg L-1) was partially removed (up to 71%) in Column OC through biologically mediated SO42- reduction coupled to OC oxidation. Biological degradation of ClO4- (input concentration: ~860 μg L-1) to Cl- was observed in the columns containing OC, but not ZVI. Removals of NO3-, SO42-, and ClO4- within three treatment columns was enhanced as a result of a decrease in flow rate from 0.5 to 0.1 pore volume (PV) d-1. Addition of ZVI to OC reduced the inhibition of ClO4- removal by NO3- (NO3-N > 2 mg L-1), but sulfate did not inhibit ClO4- removal in any treatment column.
Environmentally relevant EC contamination is frequently derived from WWTP discharge. Two-year water sampling was conducted to identify potential tracers to track wastewater downstream from two WWTPs over a 31 km stretch of the Grand River in southwestern Ontario, Canada. The results indicate that elevated concentrations of Cl-, NH3-N, NO2-, and ECs including the artificial sweetener acesulfame-K (ACE-K), and pharmaceuticals carbamazepine (CBZ), caffeine (CAF), sulfamethoxazole (SMX), ibuprofen (IBU), gemfibrozil (GEM), and naproxen (NAP) were observed near the two WWTPs, and their concentrations decreased over distance downstream. A Spearman Rank correlation analysis shows strong correlation among ACE-K, CBZ, GEM, NAP, and Cl-, suggesting the potential use of these contaminants as co-tracers to track wastewater in the study area. However, Cl- was less reliable due to other sources of contamination such as road salt applications in winter.
Laboratory batch experiments were conducted to evaluate the ultraviolet light (UV) photocatalytic treatment of artificial sweetener ACE-K and pharmaceuticals CBZ, CAF, SMX, 3,4-methylenedioxyamphetamine (MDA), 3,4-methylenedioxymetham- phetamine (MDMA), IBU, GEM, and NAP using two types of catalyst, magnetically separable TiO2 (MST) recoverable nanoparticles (γ-Fe2O3@SiO2@TiO2 colloidal nanospheres) and graphene oxide supported TiO2 (GO TiO2) recoverable nanoparticles (GO/TiO2/CSA nanocomposites) compared to commercial P25 TiO2. The synthesized GO TiO2 exhibited comparable or greater photocatalytic ability compared to P25 TiO2 in terms of intrinsic reaction rate constants for removal of ACE-K (> 99% removed from 10 μg L-1) and eight pharmaceuticals (> 92% removed from 1 μg L-1) from water. Photocatalytic ability of MST nanoparticles was lower compared to GO and P25 TiO2 nanoparticles. The non-recoverable P25 nanoparticles have been reported to have adverse impact on human health and ecological systems when released to the environment after use. The GO TiO2 nanoparticles could potentially be used as a substitute for P25 nanoparticles in water treatment due to its competitive photocatalytic ability and high magnetic recovery.
Laboratory column experiments were conducted to evaluate the removal of pharmaceuticals, artificial sweeteners, and perfluoroalkyl substances using ZVI, biochar (BC), and a mixture of (ZVI + BC). The results show that input pharmaceuticals CBZ, CAF, SMX, MDA, MDMA, IBU, GEM, and NAP at ~9 µg L-1 were almost completely removed (> 97%) in Columns ZVI, BC, and (ZVI + BC). About 80 ̶ 99% of input artificial sweetener sucralose (SCL) (~110 µg L-1) was removed in three treatment columns. However, artificial sweeteners ACE-K and saccharin (SAC) were partially removed; cyclamate (CYC) was not removed in any column. About 60 ̶ 99% of input perfluorooctane sulfonic acid (PFOS) (24.0 ̶ 89.6 µg L-1) was removed in Columns BC and (ZVI + BC); less of input PFOS was removed in Column ZVI compared to the columns containing BC. Partial removal of input perfluorooctanoic acid (PFOA) (~45 µg L-1) was observed in Columns BC and (ZVI + BC), but less in Column ZVI. The removal rates of target contaminants within three treatment columns were not enhanced after column flow rates were decreased from 0.3 to 0.1 PV d-1, except for ACE-K.
Laboratory batch experiments were conducted to investigate the removal mechanisms of PFOA and PFOS by ZVI and a mixture of (ZVI + BC). The results show ~20% and ~60% of input PFOA (~20,000 µg L-1) were removed by ZVI and (ZVI + BC); ~90% and ~94% of input PFOS (~20,000 µg L-1) were removed by ZVI and (ZVI + BC). However, only ~17% of input short chain perfluoroalkyl carboxylic acid (PFCA) perfluoroheptanoic acid (PFHpA, C7-PFCA) (26 µg L-1) was removed by ZVI alone and (ZVI + BC); the input PFCA perfluorohexanoic acid (PFHxA, C6-PFCA) (0.8 µg L-1) was not removed by ZVI and (ZVI + BC). Similarly, the input short chain perfluoroalkyl sulfonic acids (PFSAs) including 330 µg L-1 of perfluoroheptane sulfonic acid (PFHpS, C7-PFSA), 13 µg L-1 of perfluorohexane sulfonic acid (PFHxS, C6-PFSA), and 6 µg L-1 perfluorobutane sulfonic acid (PFBS, C4-PFSA) were less effectively removed by ZVI and (ZVI + BC) compared to PFOS. About 57 ̶ 70% of input PFHpS, 30 ̶ 40% of input PFHxS, and 20% of input PFBS were removed by ZVI alone and (ZVI + BC). The removal efficiency of short chain PFCAs and PFSAs by ZVI and (ZVI + BC) decreased with a decrease in chain length. Sorption and reductive defluorination likely contributed to the removal of PFOA and PFOS by ZVI and (ZVI + BC). Fluoride (F-) is the indicative by-product of defluorination of PFOA and PFOS; increasing concentrations of F- were observed in the supernatants of (PFOA + ZVI) and (PFOS + ZVI) samples. The defluorination efficiencies of PFOA and PFOS were back-calculated based on the observed F- concentrations. About 10% of input PFOA and 5% of input PFOS (~20,000 µg L-1) were partially defluorinated (2F defluorinated from 15F of PFOA and 17F of PFOS) by ZVI alone, but not by the mixture of (ZVI + BC). The defluorination efficiency of PFOA and PFOS by (ZVI + BC) were likely underestimated due to sorption of F- by the reactive media.
This study demonstrates that ZVI, wood chips, and biochar are promising and cost-effective reactive media which can potentially be used in permeable reactive barriers or flow-through reactors for effective removal (> 97%) of perchlorate and pharmaceuticals (CBZ, CAF, SMX, MDA, MDMA, IBU, GEM, and NAP); for moderate removal of artificial sweeteners ACE-K and SCL and perfluoroalkyl substance PFOS; for some removal of artificial sweetener SAC and perfluoroalkyl substance PFOA; and for no removal of artificial sweetener CYC from contaminated water under ambient environmental conditions. | en |
