Journal of Hazardous Materials 272 (2014) 155–164
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Seasonal variation of diclofenac concentration and its relation with wastewater characteristics at two municipal wastewater treatment plants in Turkey Sevgi Sari, Gamze Ozdemir, Cigdem Yangin-Gomec, Gulsum Emel Zengin, Emel Topuz, Egemen Aydin, Elif Pehlivanoglu-Mantas, Didem Okutman Tas ∗ Istanbul Technical University, Environmental Engineering Department, Maslak 34469, Istanbul, Turkey
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Influent diclofenac concentrations • • • •
in WWTPs were in the range of 295–1376 ng/L. The main removal mechanism of diclofenac in the WWTPs was biological (55%). Seasonal changes influence diclofenac occurrence in the influent. WWTP operational conditions, configuration, performance control diclofenac removal. Diclofenac removal was highly correlated with nitrogen removal mechanism.
a r t i c l e
i n f o
Article history: Received 31 October 2013 Received in revised form 27 February 2014 Accepted 7 March 2014 Available online 20 March 2014 Keywords: Conventional parameters Micropollutants Municipal wastewater treatment Pharmaceutical compound Priority pollutant
a b s t r a c t The pharmaceutically active compound diclofenac has been monitored during one year at separate treatment units of two municipal wastewater treatment plants (WWTPs) to evaluate its seasonal variation and the removal efficiency. Conventional wastewater characterization was also performed to assess the possible relationship between conventional parameters and diclofenac. Diclofenac concentrations in the influent and effluent of both WWTPs were detected in the range of 295–1376 and 119–1012 ng/L, respectively. Results indicated that the higher diclofenac removal efficiency was observed in summer season in both WWTPs. Although a consistency in diclofenac removal was observed in WWTP 1, significant fluctuation was observed at WWTP 2 based on seasonal evaluation. The main removal mechanism of diclofenac in the WWTPs was most often biological (55%), followed by UV disinfection (27%). When diclofenac removal was evaluated in terms of the treatment units in WWTPs, a significant increase was achieved at the treatment plant including UV disinfection unit. Based on the statistical analysis, higher correlation was observed between diclofenac and suspended solids concentrations among conventional parameters in the influent whereas the removal of diclofenac was highly correlated with nitrogen removal efficiency. © 2014 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding author. Tel.: +90 212 285 37 87; fax: +90 212 285 65 45. E-mail address: [email protected] (D. Okutman Tas). http://dx.doi.org/10.1016/j.jhazmat.2014.03.015 0304-3894/© 2014 Elsevier B.V. All rights reserved.
Pharmaceuticals are diverse group of chemical compounds consumed widely all over the world. The consumption of pharmaceuticals is expected to rise due to increasing population, life
156
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
expectancy and availability of more drugs in most parts of the world [1,2]. There are certain pharmaceuticals which are generally persistent to biodegradation. Pharmaceuticals are discharged to the aquatic environment in unmetabolized form or as metabolites [2,3]. Most of these compounds are considered as emerging contaminants due to the lack of legislations on them and their possible adverse effects to the environment. Pharmaceuticals are detected in the aquatic environment in the ng/L to g/L range [4]. They have potential to cause chronic toxicity to aquatic and terrestrial environments. Thus, the continuous discharge of these compounds into receiving waters and their potential effects on the ecosystem has received growing global attention [5]. The municipal wastewater treatment plants are not specifically designed to remove pharmaceuticals and have limited capacity to eliminate pharmaceuticals from wastewater. The removal efficiency depends on the compound’s nature, the process design, and operational conditions [6–10]. Non-steroidal anti-inflammatory drugs (NSAIDs) are an important group of pharmaceuticals and are reported as the most frequently found substances in the effluent of the wastewater treatment plants since in most countries these drugs are available over-the-counter [11]. The non-steroidal anti-inflammatory drug diclofenac 2-(2-(2,6-dichlorophenylamino)phenyl) acetic acid is widely used for pain and inflammation by humans. Diclofenac was reported as the most frequently detected drug in the water cycle of Europe [12] and frequently detected in groundwater and surface waters [13]. Diclofenac consumption rate in Turkey for the years of 2009 and 2013 were 2603 and 2712 g/capita/day, respectively [14]. Similar consumption rates (Germany, 2613 g/capita/day [15]; Spain, 2124 g/capita/day [16]; Austria, 2104 g/capita/day [17]) were also reported in the literature for other countries. Diclofenac has the potential to bioaccumulate in the tissues of organisms (log Kow = 4.51). Recent findings suggested diclofenac’s potential to result in the extinction of some organisms [5,18]. Diclofenac is not effectively removed by the conventional wastewater treatment systems and it is one of the least biodegradable anti-inflammatories. As a result it has been detected in the effluents of WWTPs reaching to receiving water bodies throughout the world [13,19–21]. Numerous investigations have been published about the occurrence of diclofenac in sewage treatment plants, surface and drinking waters [22–28]. However, little is known about its fate at each treatment unit, especially with regard to seasons. This study was conducted to assess the fate of diclofenac at each treatment unit of two wastewater treatment plants as well as the effect of seasonal variations on its occurrence and removal. Moreover, conventional parameters were also monitored to evaluate and interpret the possible correlation between the concentrations and removal efficiencies of conventional parameters and diclofenac in two municipal wastewater treatment plants (WWTPs) located in Istanbul, Turkey. 2. Materials and methods 2.1. Chemicals Diclofenac sodium salt (98%; CAS 15307-79-6) and diclofenacd4 were purchased from Sigma–Aldrich (Steinheim, Germany) and C/D/N Isotopes Inc. (Canada), respectively. Solvents (methanol (MeOH), acetonitrile (ACN), acetone) and chemicals were of gradient grade and purchased from Sigma–Aldrich (Sigma–Aldrich Corporation, Germany). Gradient grade water for HPLC (Merck, Germany) was used for all of the extraction and measurement procedures. Stock solutions of diclofenac salt and isotope labeled diclofenac-d4 (C/D/N Isotopes Inc., Canada), were prepared in methanol. All of the working solutions were prepared daily by diluting stock solutions with HPLC grade water.
2.2. Wastewater treatment plants and sampling Two WWTPs in Istanbul were selected to investigate the effect of seasonal changes on the treatment plant performance and the removal efficiency of diclofenac. Although some treatment units at the investigated WWTPs differ from each other; both contain tertiary treatment achieving carbon, nitrogen, and phosphorus removals. WWTP 1 contains screening, grit removal, anaerobicanoxic-aerobic units (designed as Johannesburg configuration), secondary sedimentation, sand filter, and UV disinfection; whereas WWTP 2 contains screening, grit removal, primary sedimentation, anaerobic-anoxic-aerobic units (designed as Bardenpho configuration), and secondary sedimentation. Detailed information related to the operational parameters of the investigated WWTPs’ is shown in Table 1. 24-hour composite samples from influent and effluent of WWTPs as well as grab samples from effluent of each treatment unit were collected during four sampling campaigns from each WWTP (July 2012–July 2013). Daily-composite samples were obtained by mixing 500 mL sample volumes collected every hour during 24 h. All samples were collected in 12 L fluorinated jerricans (Nalgene, USA), kept cool and transported to the laboratory. Immediately after delivery to the laboratory, 2 L of the samples were transferred to the glass bottles for the analyses of conventional parameters and the rest was centrifuged at 10,000 rpm for 15 min. Then the supernatant was filtered through 0.22 m membrane filters (Millipore, USA) and stored at 4 ◦ C in amber glass bottles. The samples for diclofenac quantification were extracted and analyzed by the method described below. Sludge samples were collected as grab samples in glass jars and immediately freeze-dried after being delivered to the laboratory. The sludge was ground and sieved to smaller than 0.5 mm, and then stored at −20 ◦ C until analysis. Quantification of diclofenac in solid phase was performed as previously described in Topuz et al. [29]. All non-volumetric glassware used in the experiments were soaked in methanol, rinsed with Milli-Q water and baked at 550 ◦ C for 4 h. 2.3. Analysis of the diclofenac Quantification of diclofenac was performed with LC–MS/MS (Thermo Accela UPLC coupled with Thermo Quantum Access tandem MS, USA). Liquid samples were purified and concentrated using OASIS HLB SPE cartridges (200 mg, 6 cc) (Waters, Millford, MA, USA). Cartridges were preconditioned with 10 mL acetonitrile, 5 mL methanol, and 5 mL water. Samples were filtered from cartridges at 5 mL/min flowrate under vacuum control and washed with 5 mL water. Cartridges were dried for 1.5 h under the air stream. Diclofenac adsorbed onto cartridges was eluted by using 2 mL acetonitrile and 2 mL methanol, respectively. Solvents were evaporated under N2 stream (TurboVap II, Caliper LifeSciences) to dryness. Diclofenac was re-dissolved in 1 mL methanol:water (10:90) and quantified with LC–MS/MS. Concentrations reported for the samples correspond to the average value of two aliquots for each sample. The method was validated in both influent and effluent wastewaters of the WWTPs. Limit of Detection (LOD) and Limit of Quantification (LOQ) were determined using signal to noise ratios of 3 and 10, respectively. 2.4. Determination of conventional characterization parameters The performance of the WWTPs was also assessed by monitoring total suspended solid (TSS), volatile suspended solids (VSS), total chemical oxygen demand (tCOD), soluble chemical oxygen demand (sCOD), total organic carbon (TOC), dissolved organic
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
157
Table 1 Characteristics of influent wastewater discharges and operational parameters of the WWTPs studied. Municipal WWTP
Equivalent population
WWTP 1 WWTP 2
2,500,000 2,400,000
Volume treated (m3 /day) 500,000 390,000
SRT (days) 20 10
HRT (hours) 11 7
SRT, solids retention time; HRT, hydraulic retention time.
carbon (DOC), pH, total Kjeldahl nitrogen (TKN), ammonia nitrogen (NH3 -N), nitrate nitrogen (NO3 − -N), nitrite nitrogen (NO2 − -N), total phosphorus (TP), phosphate (PO4 3− -P) and volatile fatty acids (VFAs) parameters. Measurements were carried out according to standard methods [30]. TOC and DOC were measured with a TOC analyzer (Shimadzu TOC-5000A, Kyoto, Japan). NO3 − , NO2 − , and PO4 3− were determined by ion chromatography (Dionex Corporation, Sunnyvale, CA, USA) equipped with a conductivity detector and an analytical column (AS14A). VFAs were measured using an Agilent 6890 Series GC equipped with a flame ionization detector and a 30 m HP-FFAP capillary, 0.53 mm i.d. column (Agilent, USA). 2.5. Removal efficiency in the wastewater treatment plants The removal efficiency (R%) of each conventional parameter and diclofenac in both of the evaluated WWTPs were calculated from Eq. (1), R(%) = [(Cinf − Ceff )/Cinf ] × 100
(1)
where Cinf and Ceff are the concentrations measured in the influent and the effluent, respectively. For the calculation of diclofenac removal efficiency (R%) in the treatment units; Cinf was considered as the exit of the previous unit for each treatment unit. 2.6. Correlation analyses Correlation analyses were performed to evaluate the possible relationship between the diclofenac concentration and the conventional parameters in influent wastewater, as well as between the removal efficiencies of diclofenac and the conventional parameters in WWTPs. Statistical analysis was performed using SigmaPlot version 11 software (Richmond, CA). 3. Results and discussion
the operational problem in UV disinfection unit during winter. NH3 N concentrations were reduced below 5 mg/L at the end of the sand filter unit for the samples representing summer and winter periods. TP removal efficiency was 89% and 72% for summer and winter periods, respectively. In WWTP 2, during the summer and winter period, tCOD and sCOD contents at the end of the secondary sedimentation tank were reduced to about 50 and ≤30 mg/L, respectively. It was observed that influent TKN concentration was about three times higher at the WWTP 2 compared to WWTP 1 in winter. This might be attributed to the fact that dilution occurred at the WWTP 1 during winter. TKN concentration was reduced below 5 mg/L at the end of the secondary sedimentation tank in the summer period with a removal efficiency of more than 94%, but decreased to 40% in the winter period with a high TKN concentration of 55 mg/L in the effluent. NH3 -N removal efficiencies for the summer and winter periods were calculated as ≥90% and 26%, respectively. Nitrification is the main process in the nitrogen removal from wastewater and failure can occur easily in wastewater treatment plants due to the slow-growing and sensitive characteristics of nitrifying bacteria. Nitrification can be inhibited by several environmental and operating factors [31]. The decrease in the NH3 -N removal efficiencies in winter can be attributed to the low temperature (≤5 ◦ C) during this season. TP measurements, on the contrary, indicated higher removal efficiency in the winter period with about 82% at the end of the secondary sedimentation tank. TP concentrations at the exits of grit removal and secondary sedimentation were found as 8 and 1.5 mg/L in the winter period; whereas TP concentrations at the exits of grit removal and secondary sedimentation were measured as 10 and 6 mg/L in the summer period. The VSS/SS ratio in biological units was in the range of 0.6–0.64 in the WWTP 1 operated at 20 days of solids retention time (SRT) during the monitoring period. VSS/SS ratio ranged between 0.56–0.57 and 0.67–0.68 in the WWTP 2 operated at 10 days of SRT during summer and winter, respectively (Tables 2 and 3).
3.1. Analytical method validation All samples for diclofenac measurement were injected three times and relative standard deviations (RSD) were below 20% indicating the high precision of the method. Diclofenac-d4 was used as an internal/surrogate standard for diclofenac quantification. The calibration curves provided very good correlation (r2 > 0.99) in the range of 1–500 ng/L both in influent and effluent matrices. The SPE efficiency for diclofenac was tested for different matrices where >80% and >90% efficiencies were achieved for influent and effluent wastewaters, respectively. 3.2. Conventional parameters 3.2.1. Grab samples The removal efficiencies (R%) of conventional parameters and diclofenac in the grab samples taken from the units of WWTP 1 and WWTP 2 were calculated considering the exit of grit removal unit as the influent concentration (Cinf ) and shown in Tables 2 and 3, respectively. In WWTP 1, for both summer and winter period, total and soluble COD contents at the end of the UV disinfection unit were both decreased below 30 mg/L. The removal efficiency in WWTP 1 was determined according to the effluent of the sand filter due to
3.2.2. Composite samples The characteristics of the 24-hour composite samples at both WWTPs were evaluated and compared with the results of the grab samples (Table 4). Results indicated that tCOD, TKN, and TP removal efficiencies ranged between 91% to 96%, 80% to 90%, and 73% to 85% in WWTP 1, respectively. On the other hand, NH3 -N removal was in the range of 69–92%. Results at the WWTP 2 indicated that tCOD content were removed in the range of 76–96%. TKN and TP at the WWTP 2 were not removed as effective as in WWTP 1. It was observed that TKN and TP removal efficiencies were also low during the winter period with 22% and 54%, respectively. Lower nutrient removals were also observed in the winter period at the WWTP 1 which might be due to the decrease in the temperature having an important effect on the performances of the biological nutrient removal systems. Results in the composite samples also indicated the lowest removal efficiencies in terms of conventional parameters at the WWTP 1 for the winter period. This finding was highly dependent on the mean concentration of each parameter indicating significant reduction in the influent flow (i.e. tCOD and TSS concentrations were measured as 350 and 95 mg/L and 762 and 327 mg/L in the samples representing winter and spring periods, respectively). It is obvious that
158
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
Table 2 Conventional characterization parameters, diclofenac concentration, and mean removal efficiencies in the effluent of each unit at the WWTP 1 during monitoring period (grab samples; mean value of the two replicates ± SD for conventional parameters, mean value of the three replicates ± SD for diclofenac). Parameter
Unit
Grit Removal
Anaerobic (Bio-P)
Anoxic
Aerobic
Secondary Sedimentation
Sand Filter
Disinfection (UV)
Summer Diclofenac Removal tCOD Removal sCOD Removal TOC Removal DOC Removal TSS Removal VSS Removal VFA pH TKN Removal NH3 -N Removal TP Removal PO4 -P NO3 -N NO2 -N
ng/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L – mg/L % mg/L % mg/L % mg/L mg/L mg/L
404 ± 37 – 385 ± 10
310 ± 48 23 2260 ± 390 – 38 ± 5 75 690 – 14 72 3720 ± 122 – 2370 ± 75 – – 7.08 273 ± 80 – 50 ± 5 – 64 ± 5 – 25 ± 0.06 0.31 ± 0.05 ≤0.05 ± 0.07
125 ± 20 69 5880 ± 770 – ≤30 ≥80 2278 ± 196 – 10 80 7760 ± 183 – 4627 ± 110 – – 6.94 400 ± 38 – 24 ± 2 50 165 ± 3 – 6.11 ± 0.06 ≤0.03 ± 0.05 ≤0.05 ± 0.07
177 ± 2 58 3285 ± 580 – 42 ± 15 72 1186 ± 8 – 10 80 6550 ± 481 – 4053 ± 342 – – 6.88 365 ± 110 – 39 ± 2 19 86 ± 13 – 3.31 ± 0.06 ≤0.03 ± 0.05 ≤0.05 ± 0.07
151 ± 31 63 ≤30 ≥92 ≤30 ≥80 10 90 ≤10 ≥80 ≤5 ≥97 8±3 95 – 7.78 34 51 ≤5 ≥90 0.4 ± 0.1 93 ≤0.12 ± 0.06 2.99 ± 0.05 0.32 ± 0.07
116 ± 19 71 ≤30 ≥92 ≤30 ≥80 ≤10 ≥90 ≤10 ≥80 ≤5 ≥97 ≤5 ≥97 – 7.37 5 92 ≤5 ≥90 0.8 ± 0.02 89 ≤0.12 ± 0.06 0.05 ± 0.05 ≤0.05 ± 0.07
39 ± 3 90 ≤30 ≥92 ≤30 ≥80 ≤10 ≥90 ≤10 ≥80 ≤5 ≥97 ≤5 ≥97 – 7.73 16 77 ≤5 ≥90 0.8 ± 0.02 89 0.19 ± 0.06 0.05 ± 0.05 ≤0.05 ± 0.07
Winter Diclofenac Removal tCOD Removal sCOD Removal TOC Removal DOC Removal TSS Removal VSS Removal VFA pH TKN Removal NH3 -N Removal TP Removal PO4 -P NO3 -N NO2 -N
ng/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L – mg/L % mg/L % mg/L % mg/L mg/L mg/L
303 ± 3 – 305 ± 40
190 ± 16 37 3000 ± 100 – ≤30 78 1315 ± 164 – 12 ± 0.8 43 5170 ±156 – 3180 ± 226 –
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Seasonal variation of diclofenac concentration and its relation with wastewater characteristics at two municipal wastewater treatment plants in Turkey Sevgi Sari, Gamze Ozdemir, Cigdem Yangin-Gomec, Gulsum Emel Zengin, Emel Topuz, Egemen Aydin, Elif Pehlivanoglu-Mantas, Didem Okutman Tas ∗ Istanbul Technical University, Environmental Engineering Department, Maslak 34469, Istanbul, Turkey
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Influent diclofenac concentrations • • • •
in WWTPs were in the range of 295–1376 ng/L. The main removal mechanism of diclofenac in the WWTPs was biological (55%). Seasonal changes influence diclofenac occurrence in the influent. WWTP operational conditions, configuration, performance control diclofenac removal. Diclofenac removal was highly correlated with nitrogen removal mechanism.
a r t i c l e
i n f o
Article history: Received 31 October 2013 Received in revised form 27 February 2014 Accepted 7 March 2014 Available online 20 March 2014 Keywords: Conventional parameters Micropollutants Municipal wastewater treatment Pharmaceutical compound Priority pollutant
a b s t r a c t The pharmaceutically active compound diclofenac has been monitored during one year at separate treatment units of two municipal wastewater treatment plants (WWTPs) to evaluate its seasonal variation and the removal efficiency. Conventional wastewater characterization was also performed to assess the possible relationship between conventional parameters and diclofenac. Diclofenac concentrations in the influent and effluent of both WWTPs were detected in the range of 295–1376 and 119–1012 ng/L, respectively. Results indicated that the higher diclofenac removal efficiency was observed in summer season in both WWTPs. Although a consistency in diclofenac removal was observed in WWTP 1, significant fluctuation was observed at WWTP 2 based on seasonal evaluation. The main removal mechanism of diclofenac in the WWTPs was most often biological (55%), followed by UV disinfection (27%). When diclofenac removal was evaluated in terms of the treatment units in WWTPs, a significant increase was achieved at the treatment plant including UV disinfection unit. Based on the statistical analysis, higher correlation was observed between diclofenac and suspended solids concentrations among conventional parameters in the influent whereas the removal of diclofenac was highly correlated with nitrogen removal efficiency. © 2014 Elsevier B.V. All rights reserved.
1. Introduction
∗ Corresponding author. Tel.: +90 212 285 37 87; fax: +90 212 285 65 45. E-mail address: [email protected] (D. Okutman Tas). http://dx.doi.org/10.1016/j.jhazmat.2014.03.015 0304-3894/© 2014 Elsevier B.V. All rights reserved.
Pharmaceuticals are diverse group of chemical compounds consumed widely all over the world. The consumption of pharmaceuticals is expected to rise due to increasing population, life
156
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
expectancy and availability of more drugs in most parts of the world [1,2]. There are certain pharmaceuticals which are generally persistent to biodegradation. Pharmaceuticals are discharged to the aquatic environment in unmetabolized form or as metabolites [2,3]. Most of these compounds are considered as emerging contaminants due to the lack of legislations on them and their possible adverse effects to the environment. Pharmaceuticals are detected in the aquatic environment in the ng/L to g/L range [4]. They have potential to cause chronic toxicity to aquatic and terrestrial environments. Thus, the continuous discharge of these compounds into receiving waters and their potential effects on the ecosystem has received growing global attention [5]. The municipal wastewater treatment plants are not specifically designed to remove pharmaceuticals and have limited capacity to eliminate pharmaceuticals from wastewater. The removal efficiency depends on the compound’s nature, the process design, and operational conditions [6–10]. Non-steroidal anti-inflammatory drugs (NSAIDs) are an important group of pharmaceuticals and are reported as the most frequently found substances in the effluent of the wastewater treatment plants since in most countries these drugs are available over-the-counter [11]. The non-steroidal anti-inflammatory drug diclofenac 2-(2-(2,6-dichlorophenylamino)phenyl) acetic acid is widely used for pain and inflammation by humans. Diclofenac was reported as the most frequently detected drug in the water cycle of Europe [12] and frequently detected in groundwater and surface waters [13]. Diclofenac consumption rate in Turkey for the years of 2009 and 2013 were 2603 and 2712 g/capita/day, respectively [14]. Similar consumption rates (Germany, 2613 g/capita/day [15]; Spain, 2124 g/capita/day [16]; Austria, 2104 g/capita/day [17]) were also reported in the literature for other countries. Diclofenac has the potential to bioaccumulate in the tissues of organisms (log Kow = 4.51). Recent findings suggested diclofenac’s potential to result in the extinction of some organisms [5,18]. Diclofenac is not effectively removed by the conventional wastewater treatment systems and it is one of the least biodegradable anti-inflammatories. As a result it has been detected in the effluents of WWTPs reaching to receiving water bodies throughout the world [13,19–21]. Numerous investigations have been published about the occurrence of diclofenac in sewage treatment plants, surface and drinking waters [22–28]. However, little is known about its fate at each treatment unit, especially with regard to seasons. This study was conducted to assess the fate of diclofenac at each treatment unit of two wastewater treatment plants as well as the effect of seasonal variations on its occurrence and removal. Moreover, conventional parameters were also monitored to evaluate and interpret the possible correlation between the concentrations and removal efficiencies of conventional parameters and diclofenac in two municipal wastewater treatment plants (WWTPs) located in Istanbul, Turkey. 2. Materials and methods 2.1. Chemicals Diclofenac sodium salt (98%; CAS 15307-79-6) and diclofenacd4 were purchased from Sigma–Aldrich (Steinheim, Germany) and C/D/N Isotopes Inc. (Canada), respectively. Solvents (methanol (MeOH), acetonitrile (ACN), acetone) and chemicals were of gradient grade and purchased from Sigma–Aldrich (Sigma–Aldrich Corporation, Germany). Gradient grade water for HPLC (Merck, Germany) was used for all of the extraction and measurement procedures. Stock solutions of diclofenac salt and isotope labeled diclofenac-d4 (C/D/N Isotopes Inc., Canada), were prepared in methanol. All of the working solutions were prepared daily by diluting stock solutions with HPLC grade water.
2.2. Wastewater treatment plants and sampling Two WWTPs in Istanbul were selected to investigate the effect of seasonal changes on the treatment plant performance and the removal efficiency of diclofenac. Although some treatment units at the investigated WWTPs differ from each other; both contain tertiary treatment achieving carbon, nitrogen, and phosphorus removals. WWTP 1 contains screening, grit removal, anaerobicanoxic-aerobic units (designed as Johannesburg configuration), secondary sedimentation, sand filter, and UV disinfection; whereas WWTP 2 contains screening, grit removal, primary sedimentation, anaerobic-anoxic-aerobic units (designed as Bardenpho configuration), and secondary sedimentation. Detailed information related to the operational parameters of the investigated WWTPs’ is shown in Table 1. 24-hour composite samples from influent and effluent of WWTPs as well as grab samples from effluent of each treatment unit were collected during four sampling campaigns from each WWTP (July 2012–July 2013). Daily-composite samples were obtained by mixing 500 mL sample volumes collected every hour during 24 h. All samples were collected in 12 L fluorinated jerricans (Nalgene, USA), kept cool and transported to the laboratory. Immediately after delivery to the laboratory, 2 L of the samples were transferred to the glass bottles for the analyses of conventional parameters and the rest was centrifuged at 10,000 rpm for 15 min. Then the supernatant was filtered through 0.22 m membrane filters (Millipore, USA) and stored at 4 ◦ C in amber glass bottles. The samples for diclofenac quantification were extracted and analyzed by the method described below. Sludge samples were collected as grab samples in glass jars and immediately freeze-dried after being delivered to the laboratory. The sludge was ground and sieved to smaller than 0.5 mm, and then stored at −20 ◦ C until analysis. Quantification of diclofenac in solid phase was performed as previously described in Topuz et al. [29]. All non-volumetric glassware used in the experiments were soaked in methanol, rinsed with Milli-Q water and baked at 550 ◦ C for 4 h. 2.3. Analysis of the diclofenac Quantification of diclofenac was performed with LC–MS/MS (Thermo Accela UPLC coupled with Thermo Quantum Access tandem MS, USA). Liquid samples were purified and concentrated using OASIS HLB SPE cartridges (200 mg, 6 cc) (Waters, Millford, MA, USA). Cartridges were preconditioned with 10 mL acetonitrile, 5 mL methanol, and 5 mL water. Samples were filtered from cartridges at 5 mL/min flowrate under vacuum control and washed with 5 mL water. Cartridges were dried for 1.5 h under the air stream. Diclofenac adsorbed onto cartridges was eluted by using 2 mL acetonitrile and 2 mL methanol, respectively. Solvents were evaporated under N2 stream (TurboVap II, Caliper LifeSciences) to dryness. Diclofenac was re-dissolved in 1 mL methanol:water (10:90) and quantified with LC–MS/MS. Concentrations reported for the samples correspond to the average value of two aliquots for each sample. The method was validated in both influent and effluent wastewaters of the WWTPs. Limit of Detection (LOD) and Limit of Quantification (LOQ) were determined using signal to noise ratios of 3 and 10, respectively. 2.4. Determination of conventional characterization parameters The performance of the WWTPs was also assessed by monitoring total suspended solid (TSS), volatile suspended solids (VSS), total chemical oxygen demand (tCOD), soluble chemical oxygen demand (sCOD), total organic carbon (TOC), dissolved organic
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
157
Table 1 Characteristics of influent wastewater discharges and operational parameters of the WWTPs studied. Municipal WWTP
Equivalent population
WWTP 1 WWTP 2
2,500,000 2,400,000
Volume treated (m3 /day) 500,000 390,000
SRT (days) 20 10
HRT (hours) 11 7
SRT, solids retention time; HRT, hydraulic retention time.
carbon (DOC), pH, total Kjeldahl nitrogen (TKN), ammonia nitrogen (NH3 -N), nitrate nitrogen (NO3 − -N), nitrite nitrogen (NO2 − -N), total phosphorus (TP), phosphate (PO4 3− -P) and volatile fatty acids (VFAs) parameters. Measurements were carried out according to standard methods [30]. TOC and DOC were measured with a TOC analyzer (Shimadzu TOC-5000A, Kyoto, Japan). NO3 − , NO2 − , and PO4 3− were determined by ion chromatography (Dionex Corporation, Sunnyvale, CA, USA) equipped with a conductivity detector and an analytical column (AS14A). VFAs were measured using an Agilent 6890 Series GC equipped with a flame ionization detector and a 30 m HP-FFAP capillary, 0.53 mm i.d. column (Agilent, USA). 2.5. Removal efficiency in the wastewater treatment plants The removal efficiency (R%) of each conventional parameter and diclofenac in both of the evaluated WWTPs were calculated from Eq. (1), R(%) = [(Cinf − Ceff )/Cinf ] × 100
(1)
where Cinf and Ceff are the concentrations measured in the influent and the effluent, respectively. For the calculation of diclofenac removal efficiency (R%) in the treatment units; Cinf was considered as the exit of the previous unit for each treatment unit. 2.6. Correlation analyses Correlation analyses were performed to evaluate the possible relationship between the diclofenac concentration and the conventional parameters in influent wastewater, as well as between the removal efficiencies of diclofenac and the conventional parameters in WWTPs. Statistical analysis was performed using SigmaPlot version 11 software (Richmond, CA). 3. Results and discussion
the operational problem in UV disinfection unit during winter. NH3 N concentrations were reduced below 5 mg/L at the end of the sand filter unit for the samples representing summer and winter periods. TP removal efficiency was 89% and 72% for summer and winter periods, respectively. In WWTP 2, during the summer and winter period, tCOD and sCOD contents at the end of the secondary sedimentation tank were reduced to about 50 and ≤30 mg/L, respectively. It was observed that influent TKN concentration was about three times higher at the WWTP 2 compared to WWTP 1 in winter. This might be attributed to the fact that dilution occurred at the WWTP 1 during winter. TKN concentration was reduced below 5 mg/L at the end of the secondary sedimentation tank in the summer period with a removal efficiency of more than 94%, but decreased to 40% in the winter period with a high TKN concentration of 55 mg/L in the effluent. NH3 -N removal efficiencies for the summer and winter periods were calculated as ≥90% and 26%, respectively. Nitrification is the main process in the nitrogen removal from wastewater and failure can occur easily in wastewater treatment plants due to the slow-growing and sensitive characteristics of nitrifying bacteria. Nitrification can be inhibited by several environmental and operating factors [31]. The decrease in the NH3 -N removal efficiencies in winter can be attributed to the low temperature (≤5 ◦ C) during this season. TP measurements, on the contrary, indicated higher removal efficiency in the winter period with about 82% at the end of the secondary sedimentation tank. TP concentrations at the exits of grit removal and secondary sedimentation were found as 8 and 1.5 mg/L in the winter period; whereas TP concentrations at the exits of grit removal and secondary sedimentation were measured as 10 and 6 mg/L in the summer period. The VSS/SS ratio in biological units was in the range of 0.6–0.64 in the WWTP 1 operated at 20 days of solids retention time (SRT) during the monitoring period. VSS/SS ratio ranged between 0.56–0.57 and 0.67–0.68 in the WWTP 2 operated at 10 days of SRT during summer and winter, respectively (Tables 2 and 3).
3.1. Analytical method validation All samples for diclofenac measurement were injected three times and relative standard deviations (RSD) were below 20% indicating the high precision of the method. Diclofenac-d4 was used as an internal/surrogate standard for diclofenac quantification. The calibration curves provided very good correlation (r2 > 0.99) in the range of 1–500 ng/L both in influent and effluent matrices. The SPE efficiency for diclofenac was tested for different matrices where >80% and >90% efficiencies were achieved for influent and effluent wastewaters, respectively. 3.2. Conventional parameters 3.2.1. Grab samples The removal efficiencies (R%) of conventional parameters and diclofenac in the grab samples taken from the units of WWTP 1 and WWTP 2 were calculated considering the exit of grit removal unit as the influent concentration (Cinf ) and shown in Tables 2 and 3, respectively. In WWTP 1, for both summer and winter period, total and soluble COD contents at the end of the UV disinfection unit were both decreased below 30 mg/L. The removal efficiency in WWTP 1 was determined according to the effluent of the sand filter due to
3.2.2. Composite samples The characteristics of the 24-hour composite samples at both WWTPs were evaluated and compared with the results of the grab samples (Table 4). Results indicated that tCOD, TKN, and TP removal efficiencies ranged between 91% to 96%, 80% to 90%, and 73% to 85% in WWTP 1, respectively. On the other hand, NH3 -N removal was in the range of 69–92%. Results at the WWTP 2 indicated that tCOD content were removed in the range of 76–96%. TKN and TP at the WWTP 2 were not removed as effective as in WWTP 1. It was observed that TKN and TP removal efficiencies were also low during the winter period with 22% and 54%, respectively. Lower nutrient removals were also observed in the winter period at the WWTP 1 which might be due to the decrease in the temperature having an important effect on the performances of the biological nutrient removal systems. Results in the composite samples also indicated the lowest removal efficiencies in terms of conventional parameters at the WWTP 1 for the winter period. This finding was highly dependent on the mean concentration of each parameter indicating significant reduction in the influent flow (i.e. tCOD and TSS concentrations were measured as 350 and 95 mg/L and 762 and 327 mg/L in the samples representing winter and spring periods, respectively). It is obvious that
158
S. Sari et al. / Journal of Hazardous Materials 272 (2014) 155–164
Table 2 Conventional characterization parameters, diclofenac concentration, and mean removal efficiencies in the effluent of each unit at the WWTP 1 during monitoring period (grab samples; mean value of the two replicates ± SD for conventional parameters, mean value of the three replicates ± SD for diclofenac). Parameter
Unit
Grit Removal
Anaerobic (Bio-P)
Anoxic
Aerobic
Secondary Sedimentation
Sand Filter
Disinfection (UV)
Summer Diclofenac Removal tCOD Removal sCOD Removal TOC Removal DOC Removal TSS Removal VSS Removal VFA pH TKN Removal NH3 -N Removal TP Removal PO4 -P NO3 -N NO2 -N
ng/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L – mg/L % mg/L % mg/L % mg/L mg/L mg/L
404 ± 37 – 385 ± 10
310 ± 48 23 2260 ± 390 – 38 ± 5 75 690 – 14 72 3720 ± 122 – 2370 ± 75 – – 7.08 273 ± 80 – 50 ± 5 – 64 ± 5 – 25 ± 0.06 0.31 ± 0.05 ≤0.05 ± 0.07
125 ± 20 69 5880 ± 770 – ≤30 ≥80 2278 ± 196 – 10 80 7760 ± 183 – 4627 ± 110 – – 6.94 400 ± 38 – 24 ± 2 50 165 ± 3 – 6.11 ± 0.06 ≤0.03 ± 0.05 ≤0.05 ± 0.07
177 ± 2 58 3285 ± 580 – 42 ± 15 72 1186 ± 8 – 10 80 6550 ± 481 – 4053 ± 342 – – 6.88 365 ± 110 – 39 ± 2 19 86 ± 13 – 3.31 ± 0.06 ≤0.03 ± 0.05 ≤0.05 ± 0.07
151 ± 31 63 ≤30 ≥92 ≤30 ≥80 10 90 ≤10 ≥80 ≤5 ≥97 8±3 95 – 7.78 34 51 ≤5 ≥90 0.4 ± 0.1 93 ≤0.12 ± 0.06 2.99 ± 0.05 0.32 ± 0.07
116 ± 19 71 ≤30 ≥92 ≤30 ≥80 ≤10 ≥90 ≤10 ≥80 ≤5 ≥97 ≤5 ≥97 – 7.37 5 92 ≤5 ≥90 0.8 ± 0.02 89 ≤0.12 ± 0.06 0.05 ± 0.05 ≤0.05 ± 0.07
39 ± 3 90 ≤30 ≥92 ≤30 ≥80 ≤10 ≥90 ≤10 ≥80 ≤5 ≥97 ≤5 ≥97 – 7.73 16 77 ≤5 ≥90 0.8 ± 0.02 89 0.19 ± 0.06 0.05 ± 0.05 ≤0.05 ± 0.07
Winter Diclofenac Removal tCOD Removal sCOD Removal TOC Removal DOC Removal TSS Removal VSS Removal VFA pH TKN Removal NH3 -N Removal TP Removal PO4 -P NO3 -N NO2 -N
ng/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L % mg/L – mg/L % mg/L % mg/L % mg/L mg/L mg/L
303 ± 3 – 305 ± 40
190 ± 16 37 3000 ± 100 – ≤30 78 1315 ± 164 – 12 ± 0.8 43 5170 ±156 – 3180 ± 226 –
