The efficacy of three different wastewater treatment configurations, conventional activated sludge (CAS), nitrifying activated sludge (NAS) and biological nutrient removal (BNR) for removal of selected micropollutants from authentic wastewater was investigated. The processes were also characterized based on their proficiency to reduce the estrogenic activity of the influent wastewater using the in vitro recombinant yeast assay. The removal efficiency of trimethoprim improved with the complexity of the three treatment process configurations. Ibuprofen, androstendione, sulfamethoxazole, nonyl-phenol, estrone and bisphenol-A had moderate to high removals (>65%) while carbamazepine and meprobamate remained recalcitrant in the three treatment process configurations. The removal of gemfibrozil was better in the NAS than in BNR and CAS treatment configurations. The yeast estrogen screen (YES) assay analyses showed an improvement in estrogenicity removal in the BNR and NAS treatment configurations as compared to the CAS treatment configuration. Comparing the estrogenic responses from the three treatment configurations, the removal efficiencies followed the order of BNR = NAS > CAS and all were greater than 81%.

INTRODUCTION

Wastewater treatment plant (WWTP) effluents have been identified as the primary route of entry for micropollutants (MPs) into the aquatic environment (Joss et al. 2004). The concern with respect to MPs in the environment results from the potential deleterious effects of MPs on the aquatic ecosystem. Effluents have been shown to cause developmental and reproductive abnormalities in various trophic levels of organisms (Fent et al. 2006; Parrott & Bennie 2009) and MPs have been measured in both effluent and surface water in North America (Kolpin et al. 2002; Lishman et al. 2006).

WWTPs are generally designed to remove conventional pollutants such as nitrogen, phosphorus, biochemical oxygen demand (BOD) and total suspended solids, and the limited removal of MPs from WWTPs is generally fortuitous. A multi-level approach with layers of different removal mechanisms (sorption, biodegradation, size exclusion, oxidation, etc.) has been suggested for improvement in the removal of MPs from wastewater (Koh et al. 2009; Qian et al. 2011). However, there is still uncertainty as to whether upgrading wastewater treatment for enhanced conventional pollutant removal (i.e., nitrification, denitrification, enhanced biological phosphorous removal) will yield improved removal of MPs.

Assessing the biological effects of MPs on the flora and fauna of the aquatic ecosystem is an indispensable tool required to conduct detailed and appropriate risk assessment of MPs in the environment. Unlike chemical analyses that can provide a quantitative measure of the compounds present in a sample, biological analyses such as in vitro bioassays can provide a qualitative and quantitative measure of the net endocrine-disrupting potential of all the endocrine-active substances that are present in an effluent (Leusch et al. 2010). Studies that have compared the removal of MPs and the potential for endocrine disruption in the effluents among different wastewater treatment technologies are few. Specifically, it has not been established whether improved removal of MPs will translate into a reduction in the biological effects on an aquatic ecosystem.

In this study, the removal of 10 MPs classified according to EU directive 93/67/EEC (CEC 1993) as either toxic (gemfibrozil (GEM), ibuprofen (IBU), meprobamate (MEP)), harmful (carbamazepine (CBZ)), non-toxic (sulfamethoxazole (SMX), trimethoprim (TMP)), estrogenic (estrone (E1), bisphenol-A (BPA) and nonyl-phenol (NP)) or androgenic (androstendione (ADR)) in three different wastewater treatment technologies (conventional activated sludge (CAS), nitrifying activated sludge (NAS) and biological nutrient removal (BNR)) was investigated. The assessment was conducted at steady state in pilot-scale processes treating authentic municipal wastewater as an influent. In addition, the potential biological effects of the effluents were assessed using a recombinant yeast estrogen screen (YES) assay to quantify their net estrogenicity. The reduction of estrogenic activity was estimated by calculating E2-equivalent concentration (E2-Eq) in the effluents of the treatment trains. It was hypothesized that as the sophistication of the treatment methods increased from CAS to BNR, there would be an improvement in MP removal and reduction in estrogenic activity in the effluents.

METHODOLOGY

Process description

Three pilot WWTPs were employed for this study. The pilot plants received raw municipal wastewater from Burlington Skyway WWTP, Ontario. A detailed description of the pilot plants was presented in Ogunlaja (2015). Supporting Information (SI) Figures S1 and S2 present the flow schematics of the pilot plants while SI Table S1 presents the design and operating conditions of the pilot plants (Figures S1 and S2 and Table S1 are available online at http://www.iwaponline.com/wst/072/213.pdf). The pilot plants were operated for over 365 days with monitoring for over 180 days to ascertain stable operation in terms of biomass and effluent characteristics so as to enable a comparison among the three treatment trains.

Sample collection

A total of five sampling campaigns were conducted over 3 months during the steady-state plant operation. Twenty-four hour composite samples of the influent (primary clarifier effluent) and final effluent were collected in 10 L stainless steel canisters using flow proportional (150 mL/30 minutes) refrigerated auto samplers (Hach Company, Loveland, USA) to determine the overall performance of each treatment train in terms of conventional parameters (carbonaceous BOD (cBOD5), total ammonia nitrogen (TAN), nitrate-nitrogen (NO3-N), nitrite-nitrogen (NO2-N), total Kjeldahl nitrogen (TKN) and ortho-phosphate (PO4-P)) and MPs while 8 hour composite samples were collected from the anaerobic, anoxic and aerobic zones of the BNR bioreactor to determine the concentration profile of conventional parameters and MPs along the redox zones of the bioreactor.

After sample collection, the samples were vigorously mixed, and 2.5 L was subsampled from the 7.2 L sample and a 500 mL subsample from the 4 L mixed liquor composite sample. The mixed liquor sample was centrifuged at 4,000 rpm for 5 minutes and the supernatant filtered through 1.5 μm glass microfiber filters (Whatman, Toronto, ON, Canada) for MP and conventional pollutants analysis. The influent and effluent samples were also filtered through 1.5 μm glass microfiber filters. For each sampling campaign, five 120 mL of the filtered subsamples (three mixed liquor (anaerobic/anoxic/aerobic), one influent and one effluent) were acidified to pH 3 to inhibit microbial biodegradation of the MPs and stored in amber glass bottles at 4 °C to prevent photolysis of the MPs until they were extracted and analyzed for MPs.

Sample extraction, chemical and biological analysis

One set of subsamples was extracted using a multi-residue solid phase extraction technique (SPE), previously described elsewhere (Li et al. 2010), before shipping to Trent University for chemical quantification using Liquid chromatography tandem mass spectrometry (LC-MS/MS) procedure previously described by Miao & Metcalfe (2003). The second set of samples was extracted using the same procedure as the first set but without spiking with deuterated compounds and was processed for the YES assay.

The samples were extracted using Oasis® MAX 6 mL (500 mg) SPE cartridges. After elution, the cartridges were aspirated to dryness under vacuum. The eluted fractions were dried under a gentle stream of nitrogen gas (N2) and reconstituted in 0.4 mL of methanol prior to YES assay or LC-MS/MS analysis.

Conventional parameters were analyzed according to Standard Methods (Eaton et al. 2005) while the YES assay procedure was conducted as described by Routledge & Sumpter (1996) with modifications as reported by Citulski & Farahbakhsh (2010). The method detection limit based on E2 standard was determined as 1 ng/L.

Target MPs

Ten MPs were investigated in this study and are presented along with their physico-chemical characteristics in SI Table S2 (available online at http://www.iwaponline.com/wst/072/213.pdf). The target compounds included a broad range of substances including acidic, basic and neutral drugs and estrogenic compounds and were selected on the basis of their detection frequency in WWTP influent and the ability to detect them at low concentrations (ng/L) using LC-MS/MS.

Statistical analysis

The MPs and the conventional parameters concentrations were analyzed for outliers using the Grubbs test. The regressions used to construct the YES assay response curves for the samples were compared to the response curves of estradiol (E2) standard using the F-test. To check the yeast growth absorbance (AB620 nm) for the presence of toxic effects resulting from the wastewater extracts, the yeast growth was compared to the average turbidity (± 3 standard deviations) of 12 ethanol-only negative-control wells that were incubated in the same plate as the samples. Samples that had yeast growth with turbidity values below the average minus 3 standard deviations of the negative control were removed from the dose–response curve analysis. The YES assay data were employed to generate the dose–response curve for the estimation of the EC50. A paired t-test was employed to compare the MPs' concentrations and the E2-Eq concentrations in the influent and pilots' effluent streams at the 95% confidence level using Microsoft Excel 2013.

RESULTS AND DISCUSSION

Conventional parameters

The pilot plants were monitored with respect to the removal of conventional wastewater pollutants such as BOD and nitrogen species. These data were employed to establish whether the treatment plants were operating within normally established ranges and to provide insight into the types of microbial metabolisms that were active in the bioreactors. The measured responses are presented in Table 1.

Table 1

Average influent and effluent concentrations of conventional parameters from each pilot plant (mg/L); n = 20

  TAN NO3-N NO2-N TKN cBOD5 PO4-P 
Influent 20.9 ± 6.6 0.62 ± 0.9 0.16 ± 0.23 27.1 ± 10.4 79.8 ± 56.4 5.15 ± 1.6 
CAS 21.1 ± 6.4 1.7 ± 4 0.4 ± 0.6 21.4 ± 5.7 13.3 ± 10.1 5.38 ± 1.4 
NAS 0.076 ± 0.039 22 ± 2.6 0.013 ± 0.018 1.2 ± 0.3 3.5 ± 1.7 5.24 ± 1.4 
BNR 1.3 ± 2.1 5.4 ± 2.2 0.3 ± 0.4 2.4 ± 2.3 6.4 ± 6 0.19 ± 0.4 
  TAN NO3-N NO2-N TKN cBOD5 PO4-P 
Influent 20.9 ± 6.6 0.62 ± 0.9 0.16 ± 0.23 27.1 ± 10.4 79.8 ± 56.4 5.15 ± 1.6 
CAS 21.1 ± 6.4 1.7 ± 4 0.4 ± 0.6 21.4 ± 5.7 13.3 ± 10.1 5.38 ± 1.4 
NAS 0.076 ± 0.039 22 ± 2.6 0.013 ± 0.018 1.2 ± 0.3 3.5 ± 1.7 5.24 ± 1.4 
BNR 1.3 ± 2.1 5.4 ± 2.2 0.3 ± 0.4 2.4 ± 2.3 6.4 ± 6 0.19 ± 0.4 

Considering the inherent variability of processes treating raw municipal wastewater, the effluents from the pilot plants, as indicated in Table 1, were relatively consistent with time. Carbonaceous BOD5 was consistently removed in all pilot plants and most final concentrations were less than 10 mg/L. This was considered indicative of good removal of biodegradable organic matter. The NAS process produced the lowest effluent cBOD5 concentrations while the effluent concentrations from the CAS process were consistently higher than NAS and BNR effluents.

Based upon the operating conditions that were employed in this study it was expected that the effluents from the CAS pilot would have higher concentrations of TKN and TAN and low concentrations of NO3-N and NO2-N as compared to NAS and BNR effluents. The data in Table 1 show that the CAS pilot effluent NO3-N concentrations were consistently low and the effluent TKN and TAN concentrations remained elevated. The NAS pilot plant also performed consistent with expectation with low concentrations of TKN and TAN and elevated concentrations of NO3-N in the effluent. Hence, it was concluded that the CAS pilot was not nitrifying while the NAS pilot was nitrifying effectively. Similarly, the BNR pilot plant had low effluent concentrations of PO4-P, TKN, TAN, NO3-N and NO2-N (Table 1), indicating that the pilot plant was effectively nitrifying, denitrifying and biologically removing phosphorus. Collectively, it was concluded that the pilot plants' operations were consistent with known efficiencies and operational characteristics of full-scale WWTPs (Tchobanoglous et al. 2003).

MP removal during treatment

All 10 compounds were detected in the influent wastewater. The observed average influent concentrations were generally lower than in previously reported studies (Lishman et al. 2006; Lajeunesse et al. 2012). The influent data were conditioned by using Grubbs' test to determine any outliers which were removed before conducting other statistical analysis. There was some overall variability associated with the influent concentrations of the selected compounds with a median relative standard deviation of 25% across all influent samples for the five sampling campaigns. The variability was observed to be random; thus it was assumed that there were no differences in the measured concentrations between replicate samples.

The results of the paired t-test showed a statistically significant difference between the mean concentrations in the influent and effluents from CAS, NAS and BNR for the compounds IBU, CBZ, GEM, ADR, E1, NP and BPA. However, there was no statistically significant difference between the mean concentrations in the influent and effluent from CAS, NAS and BNR for MEP. In addition, the result showed that the mean concentrations in the influent and the NAS effluent for GEM were statistically significant. The mean of the difference in the concentrations of TMP in the influent and in the BNR effluents was also statistically significant. In addition, there were no statistical differences between CAS and BNR effluents or between NAS and BNR effluents at the 95% confidence level for all the investigated compounds except for TMP. These results show that the majority of the removals of the MPs from the CAS, NAS and BNR treatment configurations were significant. Hence, the data were employed for further technical comparisons among the three treatment configurations.

Box plots and averages of the concentrations of the compounds in the influent and effluents from CAS, NAS and BNR are presented in SI Figures S3, S4, and S5 (available online at http://www.iwaponline.com/wst/072/213.pdf). These plots show that five MPs (IBU, ADR, E1, NP and BPA) had comparable effluent concentrations for CAS, NAS and BNR. Two MPs (CBZ, MEP) had effluent concentrations equal or higher than the influent concentrations for all the treatment processes. TMP, GEM and SMX effluent concentrations followed the trend of BNR < NAS < CAS, NAS < BNR < CAS and NAS < CAS = BNR, respectively.

The removal efficiencies of each compound across the treatment processes were calculated as the percent difference in concentration from influent to effluent. The term ‘removal’ used in this study described the loss of the target compound and did not necessarily imply mineralization. The removal of the target compound could have proceeded via different mechanisms including chemical and physical transformation, biodegradation or biotransformation, and sorption. The MP removal efficiencies across the three treatment processes are presented in Figure 1. It is apparent from Figure 1 that six MPs (IBU, ADR, SMX, E1, NP and BPA) were consistently removed with removal efficiency greater than 65% across the three treatment processes, while no removal was observed for two MPs (CBZ and MEP) regardless of the treatment process utilized.

Figure 1

MP removal efficiencies in pilot plants (deviation bar represents standard deviation of removal efficiencies (n = 5)). IBU – ibuprofen, MEP – meprobamate, GEM – gemfibrozil, CBZ – carbamazepine, TMP – trimethoprim, SMX – sulfamethoxazole, ADR – androstendione, E1 – estrone, NP – nonyl-phenol, BPA – bisphenol-A.

Figure 1

MP removal efficiencies in pilot plants (deviation bar represents standard deviation of removal efficiencies (n = 5)). IBU – ibuprofen, MEP – meprobamate, GEM – gemfibrozil, CBZ – carbamazepine, TMP – trimethoprim, SMX – sulfamethoxazole, ADR – androstendione, E1 – estrone, NP – nonyl-phenol, BPA – bisphenol-A.

The moderate to high removal efficiencies for the IBU, ADR, SMX, E1, NP and BPA were consistent with the removal efficiencies previously reported in the literature (USEPA 2010). Treatment processes with sludge retention times (SRTs) ranging from 2 to 68 days on a laboratory scale and 22 to 82 days at WWTP pilot scale were reported to have IBU removals ranging from 80 to 100% (Onesios et al. 2009). The average reported removal efficiency for BPA in a variety of treatment processes and SRTs was reported to be 83% (Melcer & Klecka 2011). Clara et al. (2005) reported BPA and IBU average removals of 82–98% at SRTs as low as 2 days in an activated sludge system. Reported values for ADR, NP and E1 removal efficiency in various aerobic activated sludge systems and SRTs have ranged from 98 to 100% with an average of 99%. These results suggest that the removals of IBU, ADR, SMX, E1, NP and BPA in the CAS, NAS and BNR were independent of the process configuration or system's SRT.

The apparent negative removals for CBZ and MEP (Figure 1) reflected the increased effluent concentrations of CBZ and MEP from the treatment processes as compared to the influent concentrations (SI Figures S3(d) and S4(e)). This suggests that CBZ and MEP were not removed by any of the treatment processes. This may have resulted from de-conjugation of conjugated versions of the parent compounds by the action of the micro-organisms during the treatment process. Gobel et al. (2007) suggested that the increase in effluent concentration of some MPs as compared to their influent concentration could be due to the encasing of the influent MPs in fecal particles, leading to an apparent increase in concentration during treatment as the fecal particles are degraded. Previous studies have shown that recalcitrant compounds like CBZ rarely sorb or degrade in a variety of wastewater treatment processes (Ternes et al. 2004; Clara et al. 2005). The recalcitrant behavior of CBZ and MEP could be a useful characteristic in applications where they might be employed as an anthropogenic tracer.

Figure 1 shows an increase in the removal of TMP as the treatment process complexity progressed from a simple non-nitrifying CAS to a complex BNR configuration. Longer SRTs accommodate slower growing bacteria like nitrifiers and can also support the proliferation of a wider range of bacteria species. The presence of a broad array of bacteria likely allows for a wider range for biotransformation of TMP, thereby leading to improved removal at longer SRTs (CAS 3 days, NAS 10 days and BNR 20 days). Correlations between treatment efficiency and SRT have previously been established (Clara et al. 2005) and an SRT longer than 10 days has been suggested as being more effective in removing MPs such as estrogens (Carballa et al. 2007). Gobel et al. (2007) investigated the removal of MPs by activated sludge and showed an increase in the removal of TMP from 50 to 90% when SRTs were increased from 16 ± 2 to 60 days. Therefore, the increased removal efficiency of TMP in the BNR could be due to microbial diversity and SRT of the process as compared to the CAS and NAS treatment configurations.

MP removal in BNR treatment configuration

The bioreactors of the CAS and NAS pilots were operated as single aerobic reactors but the BNR bioreactor was divided into three different redox conditions. Therefore, it was expected that the different zones would contribute differently to the biotransformation of the MPs in the BNR treatment configuration. This expectation was based on the fact that the amount of energy that is captured by the micro-organisms in aerobic conditions is usually higher than the energy captured in anoxic and anaerobic conditions (Tchobanoglous et al. 2003). This bioenergy could be instrumental to the biotransformation of the MPs in the BNR bioreactor. Hence, the contribution of the redox conditions to the removal of the MPs that were removed in the BNR treatment configuration was assessed.

The MP concentrations in the influent and the intermediate stages of the BNR bioreactor are presented in SI Figure S6 (online at http://www.iwaponline.com/wst/072/213.pdf) and Figure 2. It is apparent from SI Figure S6 and Figure 2 that the concentrations of the MPs decreased through each stage of the bioreactor, which suggested that each of the zones contributed to the overall MP removal. Recycle flows can have dilution effects on the BNR pilot plant. Therefore, a set of mass balances were employed to characterize the fate of the MPs that were removed in the pilot BNR treatment configuration.

Figure 2

MP concentrations in the influent and stages of BNR process (deviation bar represents standard deviation of measurements (n = 5)).

Figure 2

MP concentrations in the influent and stages of BNR process (deviation bar represents standard deviation of measurements (n = 5)).

In the mass balances, the biotransformation efficiencies of the MPs in each zone were calculated as the difference between the mass flow entering and leaving the zone, divided by the mass flow entering the zone. The difference between the mass inflow and outflow in the aqueous phase across each zone of the bioreactor was assumed to be due to microbial biotransformation within the zone of the bioreactor. SI Figure S7 (online at http://www.iwaponline.com/wst/072/213.pdf) shows a representative schematic of the mass flows employed for the MPs that were removed in the BNR treatment configuration. The biotransformation efficiencies of the MPs in the anaerobic, anoxic and aerobic sections of the BNR bioreactor are presented in Table 2. Table 2 shows that the biotransformation efficiencies in the bioreactor of the BNR process increased from the anaerobic to the anoxic and aerobic zone for IBU and TMP while only the aerobic zone degraded SMX, ADR, E1 and GEM. Hence, it was concluded that out of the six MPs that were removed in the BNR treatment configuration, only IBU and TMP were biotransformed in all the redox zones of the BNR bioreactor at different percentages while the other four MPs were biotransformed only in the aerobic zone.

Table 2

Biotransformation efficiency of MPs in BNR bioreactor

 Biodegradation efficiency (%) 
MP Anaerobic Anoxic Aerobic 
IBU 15 ± 1 45 ± 1 83 ± 5 
TMP 13 ± 12 17 ± 10 24 ± 4 
SMX nd nd 48 ± 11 
ADR nd nd 100 ± 14 
E1 nd nd 95 ± 0.2 
GEM nd nd 67 ± 23 
 Biodegradation efficiency (%) 
MP Anaerobic Anoxic Aerobic 
IBU 15 ± 1 45 ± 1 83 ± 5 
TMP 13 ± 12 17 ± 10 24 ± 4 
SMX nd nd 48 ± 11 
ADR nd nd 100 ± 14 
E1 nd nd 95 ± 0.2 
GEM nd nd 67 ± 23 

nd – not degraded.

Biological activity

Owing to constraints on chemical analysis, it is unlikely that the suite of chemicals selected for characterization would account for the entire range of endocrine-disrupting compounds present in the effluent. Differences in endocrine-active potency among different chemical compounds or possible antagonism or synergism of chemical mixtures cannot be accounted for by chemical analysis. Therefore, in vitro bioassays to determine whole-sample endocrine-disrupting effects, usually estrogenic (Routledge & Sumpter 1996), have been developed and were examined in this study.

The YES assay was employed to assess the performance of the treatment processes in terms of estrogenicity removal. The YES assay estrogenic equivalence (E2-Eq) values in the influent and from the effluents of the treatment processes are summarized in SI Figure S8 (online at http://www.iwaponline.com/wst/072/213.pdf). The influent E2-Eq average concentration was 37.6 ± 5 ng/L while the average effluent concentrations for the three processes were 6.3, 4.7 and 0.84 ng/L for the CAS, NAS and BNR processes respectively.

A paired t-test was used to compare the E2-Eq in influent to the E2-Eq in the three effluent streams. The results of the paired t-test showed a statistical difference between the influent and CAS, NAS and BNR effluents (P < 0.05). The BNR effluent values were significantly lower than the CAS (p = 0.006) but not significantly lower than the NAS effluent values. Further statistical analysis could not be employed to discriminate between the means of the data set because of the limited number of available data. The estrogenicity removal efficiencies were 84 ± 3%, 89 ± 10% and 98 ± 2% for CAS, NAS and BNR pilot plants, respectively. The results show that the BNR and the NAS performed better than the CAS in terms of estrogenicity removal. In general, the trend in the removal of estrogenicity by the process configurations was consistent with the compound removals of E1, NP and BPA that were previously presented. However, it is important to note that the similarity in the trends in MP removal and estrogenicity removal does not imply that the reduction in estrogenicity across the treatment trains was as a result of the MPs investigated.

CONCLUSIONS

  • The removal efficiency of TMP improved with the complexity of the three treatment processes configurations and SRTs.

  • IBU, ADR, SMX, NP, E1 and BPA had moderate to high removals (>65%) while CBZ and MEP remained recalcitrant in the three treatment process configurations.

  • The removal of GEM was better in the NAS than in the BNR and CAS treatment configurations.

  • The YES assay analyses showed an improvement in estrogenicity removal in the BNR and NAS as compared to the CAS treatment configuration.

  • A similar trend was observed among the treatment processes in terms of the removal of MPs and estrogenicity.

REFERENCES

REFERENCES
CEC
1993
Commission Directive 93/67/EEC of 20 July 1993 laying down the principles for assessment of risks to man and the environment of substances notified in accordance with Council Directive 67/548/EEC
.
Official Journal
L 227
,
08/09/1993
,
9
18
.
Citulski
J. A.
Farahbakhsh
K.
2010
Fate of endocrine-active compounds during municipal biosolids treatment: a review
.
Environ. Sci. Technol.
44
(
22
),
8367
8376
.
Eaton
A. D.
Franson
M. A. H.
,
American Public Health Association, American Water Works Association & Water Environment Federation
2005
Standard Methods for the Examination of Water and Wastewater
,
21st edn.
American Public Health Association
,
Washington, DC, USA
.
Gobel
A.
McArdell
C. S.
Joss
A.
Siegrist
H.
Giger
W.
2007
Fate of sulfonamides, macrolides, and trimethoprim in different wastewater treatment technologies
.
Sci. Total Environ.
372
(
2–3
),
361
371
.
Koh
Y. K. K.
Chiu
T. Y.
Boobis
A. L.
Scrimshaw
M. D.
Bagnall
J. P.
Soares
A.
Pollard
S.
Cartmell
E.
Lester
J. N.
2009
Influence of operating parameters on the biodegradation of steroid estrogens and nonylphenolic compounds during biological wastewater treatment processes
.
Environ. Sci. Technol.
43
(
17
),
6646
6654
.
Kolpin
D.
Furlong
E.
Meyer
M.
Thurman
E.
Zaugg
S.
Barber
L.
Buxton
H.
2002
Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000: a national reconnaissance
.
Environ. Sci. Technol.
36
(
6
),
1202
1211
.
Leusch
F. D.
de Jager
C.
Levi
Y.
Lim
R.
Puijker
L.
Sacher
F.
Tremblay
L. A.
Wilson
V. S.
Chapman
H. F.
2010
Comparison of five in vitro bioassays to measure estrogenic activity in environmental waters
.
Environ. Sci. Technol.
44
(
10
),
3853
3860
.
Lishman
L.
Smyth
S. A.
Sarafin
K.
Kleywegt
S.
Toito
J.
Peart
T.
Lee
B.
Servos
M.
Beland
M.
Seto
P.
2006
Occurrence and reductions of pharmaceuticals and personal care products and estrogens by municipal wastewater treatment plants in Ontario, Canada
.
Sci. Total Environ.
367
(
2–3
),
544
558
.
Melcer
H.
Klecka
G.
2011
Treatment of wastewater containing Bisphenol A: state of science review
.
Water Environ. Res.
49
(
1
),
23
31
.
Ogunlaja
O. O.
2015
Impact of Biological Nutrient Removal Process Operating and Design Conditions on the Removal of Micropollutants from Wastewater
.
PhD Dissertation
,
University of Waterloo
,
Waterloo, ON, Canada
.
Parrott
J. L.
Bennie
D. T.
2009
Life-cycle exposure of fathead minnows to a mixture of six common pharmaceuticals and triclosan
.
J. Toxicol. Environ. Health Part A
72
(
10
),
633
641
.
Tchobanoglous
G.
Burton
F. L.
Stensel
H. D.
& Metcalf and Eddy, Inc.
2003
Wastewater Engineering: Treatment and Reuse
.
McGraw-Hill
,
New York
.
Ternes
T. A.
Joss
A.
Siegrist
H.
2004
Scrutinizing pharmaceuticals and personal care products in wastewater treatment
.
Environ. Sci. Technol.
38
(
20
),
392
399
.
USEPA
2010
Treating Contaminants of Emerging Concerns; A Literature Review Database
.
United States Environmental Protection Agency
.
EPA-820-R-10-002
.

Supplementary data