- Email: [email protected]
Pharmaceutical removal in tropical subsurface flow constructed wetlands at varying hydraulic loading rates
Chemosphere 87 (2012) 273–277
Contents lists available at SciVerse ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Technical Note
Pharmaceutical removal in tropical subsurface flow constructed wetlands at varying hydraulic loading rates Dong Qing Zhang a,⇑, Richard M. Gersberg b, Tao Hua a, Junfei Zhu a, Nguyen Anh Tuan a, Soon Keat Tan c a
DHI-NTU Centre, Nanyang Environment and Water Research Institute, Nanyang Technological University, N1.2-B1-02, 50 Nanyang Avenue, Singapore 639798, Singapore Graduate School of Public Health, San Diego State University, Hardy Tower 119, 5500 Campanile, San Diego, CA 92182-4162, USA c Maritime Research Centre, School of Civil and Environmental Engineering, Nanyang Technological University, N1.2-B1-02, 50 Nanyang Avenue, Singapore 639798, Singapore b
a r t i c l e
i n f o
Article history: Received 8 July 2011 Received in revised form 22 December 2011 Accepted 22 December 2011 Available online 21 January 2012 Keywords: Pharmaceutical compounds Subsurface flow Constructed wetlands Rhizosphere aeration Continuous-feeding regime
a b s t r a c t Determining the fate of emerging organic contaminants in an aquatic ecosystem is important for developing constructed wetlands (CWs) treatment technology. Experiments were carried out in subsurface flow CWs in Singapore to evaluate the fate and transport of eight pharmaceutical compounds. The CW system included three parallel horizontal subsurface flow CWs and three parallel unplanted beds fed continuously with synthetic wastewater at different hydraulic retention times (HRTs). The findings of the tests at 2–6 d HRTs showed that the pharmaceuticals could be categorized as (i) efficiently removed compounds with removal higher than 85% (ketoprofen and salicylic acid); (ii) moderately removed compounds with removal efficiencies between 50% and 85% (naproxen, ibuprofen and caffeine); and (iii) poorly removed compounds with efficiency rate lower than 50% (carbamazepine, diclofenac, and clofibric acid). Except for carbamazepine and salicylic acid, removal efficiencies of the selected pharmaceuticals showed significant (p < 0.05) enhancement in planted beds as compared to the unplanted beds. Removal of caffeine, ketoprofen and clofibric acid were found to follow first order decay kinetics with decay constants higher in the planted beds than the unplanted beds. Correlations between pharmaceutical removal efficiencies and log Kow were not significant (p > 0.05), implying that their removal is not well related to the compound’s hydrophobicity. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Many pharmaceutical compounds are the emerging contaminants in the modern society. Since these pharmaceuticals are typically present at trace levels (in the ng L1 range), and wastewater treatment plants (WWPTs) are not designed for their removal (Joss et al., 2006), many pharmaceuticals escape treatment and are released into the environment (Buser et al., 1998a; Heberer, 2002). Compared to conventional technical solutions for water treatment, constructed wetlands (CWs) provide a low-cost alternative which make them suitable for wastewater treatment where land availability is not a limiting factor (Kadlec and Knight, 1995). To date relatively little work has been conducted focusing on the removal of pharmaceuticals in CWs, and the most efficient configuration for their removal and the mechanisms involved are still unknown (Matamoros and Bayona, 2006; Park et al., 2009). The treatment efficiency of pollutants in a CW system is usually improved by decreasing the hydraulic loading (i.e., the longer the
⇑ Corresponding author. Tel.: +65 8165 6212; fax: +65 6790 6620. E-mail addresses: [email protected] (D.Q. Zhang), [email protected] (R.M. Gersberg), [email protected] (S.K. Tan). 0045-6535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2011.12.067
hydraulic retention time (HRT), the greater the pollutant removal efficiency). No specific guidelines have been established for designing a CW system to remove pharmaceutical compounds from water. Operating with a shorter HRT means smaller land-area requirements, which is a crucial factor for considering a CW. CWs in tropical regions (especially with their elevated temperatures year-round) are considerably more effective than systems in temperate climates, and may exhibit organic compound removal rates almost a factor 10 higher than standard CWs (Scheren, 2008). In this study, a variety of analgesic/anti-inflammatory drugs (naproxen, ibuprofen, diclofenac, ketoprofen and salicylic acid), a lipid regulator (clofibric acid), a stimulant (caffeine), and an antiepileptic drug (carbamazepine) were investigated. These products were selected because of their ubiquitous occurrence in sewage effluents and high frequency of detection reported in previous studies (Matamoros and Bayona, 2006; Hijosa-Valsero et al., 2010b). The specific goals of this study were to (i) investigate the removal efficiencies of selected pharmaceutical compounds in horizontal subsurface flow (HSSF) CWs planted with Typha angustifolia as compared to unplanted beds (sand filters) to evaluate the quantitative role of the aquatic plant on pharmaceutical removal under continuous-feeding regimes; and (ii) study the treatment kinetics of CW systems for operating at HRTs of shorter than 6 d.
274
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
2. Methodologies 2.1. Description of the mesocosm-scale CWs Six mesocosm-scale CWs were set up at the campus of Nanyang Technological University in Singapore. The location is a typical tropical environment and is generally hot and humid year round. The temperature ranges from 23 to 32 °C. All CWs in this study consisted of a fiberglass container (1.2 m 0.6 m 0.6 m, long wide deep), which contained the soil media and plants in the case of planted beds. The thickness of gravel bed was 0.30 m with 4–10 mm siliceous gravel (D60 = 3.5 mm) and the porosity was 0.45. Three CWs were left unplanted to act as controls (sand filters), while three CWs were planted with cattail (T. angustifolia) at a density of 14 plant m2 serving as HSSF CWs. In our case, T. angustifolia had been cultivated for one yr before purchase and then established in the tanks for 1 month. All the containers had a flat bottom and a horizontal drainage pipe (0.4 m long and 50 mm in diameter) located at the lower edge of the containers. 2.2. Sampling regime Two series of wetland studies were carried out: (i) the investigation of carbamazepine, naproxen, ibuprofen and diclofenac removal by the wetlands; and (ii) the investigation of the fate of caffeine, salicylic acid, ketoprofen and clofibric acid in the wetlands. For each series, the wetlands were operated in a continuous flow mode for four months. Plants were replaced after the first series was completed. Prior to beginning data collection for each series, all the wetland beds were allowed to adapt to the new steady state conditions for a period of 15 d. Additionally, different HRTs (2 and 4-d) were tested in both series and an additional 6-d HRT was included in the second series for investigating pharmaceutical removal kinetics. For logistical reasons, the operational modes were tested in sequence rather than in parallel. However, it is important to note here that the equatorial climate of Singapore offers a rather constant temperature and degree of solar radiation, without the seasonal variations of temperate countries. Therefore the environmental conditions of the plants and gravel beds were for the most part similar during the two periods of comparison. Continuous loading was achieved by using a six-channel peristaltic pump. A holding tank of 150 L was served as the reservoir for dispensing synthetic wastewater. In both series, the hydraulic application rates were 5.6 cm d1 (2-d HRT) and 2.8 cm d1 (4-d HRT), respectively. In addition, in the second series, a 6-d of HRT (1.9 cm d1 of hydraulic application rate) was tested. In both series of experiments, wastewater effluent samples were always collected from each of the 6 beds every other day. All the samples were collected on the same day and at the same time of day, i.e., during early morning. These samples were then transported refrigerated (4 °C) to the laboratory, where they were analyzed within 24 h. 2.3. Materials and chemicals Analytical grade glucose anhydrous (99% purity), ammonium sulfate (98% purity), magnesium sulfate heptahydrate (99% purity), calcium chloride dehydrate (99% purity), calcium chloride (74–78% purity), iron (III) chloride anhydrous (98% purity), sodium carbonate (98% purity), sodium hydrogen carbonate (99% purity), potassium dihydrogen phosphate (98% purity) obtained from Alfa Aesar, Germany were used for the preparation of synthetic wastewater. Matamoros et al. (2008b) fed SSF CWs a similar synthetic wastewater containing both readily assimilable (dissolved) glucose
and a particulate form of organic matter in the form of starch, and found that pharmaceutical removal efficiencies were independent of the organic matter type (dissolved versus particulate). Therefore we chose glucose as the carbon source for this study. HPLC-grade acetonitrile, methanol, hexane and ethyl acetate were obtained from Fisher (USA). Ultrapure water was obtained from Milli-Q water purification system (Millipore, Bedford, USA). Carbamazepine, diclofenac, ibuprofen and naproxen (97–100% purity) were obtained from Sigma–Aldrich (Singapore). The solid phase extraction (SPE) cartridges GracePureTM C18 (500 mg, 6 mL) were purchased from Grace Davison Discovery Sciences (Belgium). The HPLC column Inertsil ODS (35 lm, 4.6 150 mm) was obtained from Alpha Analytical (Singapore). The 0.45 lm glass fiber filters of 47 mm (Whatman) were purchased from Schleicher and Schuell (Germany). Stock solution of 50 mg L1 was prepared in methanol and stored at 4 °C. Working solutions were prepared by diluting the stock solution with methanol. During the experiments, all the beds were fed with synthetic wastewater with the same organic load. The composition of influent included (in mg L1) 300 COD, 27 NH4+, 7 PO4-P, 4 Fe3+, 7 Mg2+, and 6 Ca2+. With respect to the pharmaceutical injection, 50 L of synthetic wastewater were spiked with 1.25 mg of each pharmaceutical compound to obtain a final concentration of 25 lg L1. 2.4. Analytical procedures for pharmaceutical compounds Prior to extraction, 500 mL of effluent wastewater were filtered through a 0.45 lm glass fibre membrane filter (Millipore, USA) and then acidified to pH 2 with hydrochloric acid. The SPE cartridges were conditioned using 5 mL n-hexane, 5 mL ethyl acetate, 10 mL methanol and 10 mL of Milli-Q water at a flow-rate of approximate 3 mL min1. Samples were percolated to the SPE cartridges through a Teflon tube at a flow-rate of approximate 10 mL min1. The cartridge filter was then rinsed with the combination of methanol and Milli-Q (methanol: Milli-Q water = 10:90), followed by 20 mL of Milli-Q water. Thereafter the cartridges were allowed to dry for ca 30 min and then eluted with 5 mL ethyl acetate with elution solutions collected in 15 mL calibrated centrifuge tubes. The extracted solution was then concentrated to ca 400 lL under a gentle nitrogen stream and was then reconstituted to 500 lL with methanol. Chromatographic analysis was performed on a Shimadzu Ultra Fast Liquid Chromatograph (Shimadzu, Japan) equipped with a quaternary LC-20AD pump, a CTO-20A oven, and a SPD-M20A Diode Array Detector. The injector was SIL-20A HT fitted to a Shimadzu autosampler with a 20 lL sample loop. Chromatographic separations were carried out using an Inertsil ODS-3 (4.6 150 mm, 5 lm) HPLC column protected by a ODS-3 (C18) (4.6 50 mm, 5 lm) guard column (Alpha Analytical, Singapore). The system was controlled using an interface module and a personal computer. Chromatograms were processed using a ‘‘Shimadzu LCSolution program’’. 2.5. Recoveries, LOD and LOQ Effluent wastewater (1 L) was spiked with the pharmaceutical compounds to a concentration of 5 lg L1. The sample was extracted according to the procedure described earlier. The signals obtained from spiked samples were compared with signals from wastewater without standard addition and with those obtained by injecting standard solutions. In this study, the recoveries ranged from 70% to 105%. Limits of detection (LOD) and limits of quantification (LOQ) of the analytical procedure were established by using a water effluent from a microcosm plant effluent without pharmaceuticals. LOD and LOQ were determined based on a signal-to-noise ratio of 3
275
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
and 10, respectively (the ratio between peak intensity and intensity of the noise was used). The values of LOD and LOQ for 8 pharmaceutical compounds tested in our study ranged from 0.006 to 0.028 lg L1 and 0.019 to 0.083 lg L1, respectively. 2.6. Statistical test of experimental data The experimental data were tested for normal distribution. Tests to determine statistical differences between treatments were carried out by comparing the critical value through ANOVA. Comparisons were considered significantly different for p < 0.05. 3. Results and discussion 3.1. Removal efficiencies of pharmaceutical compounds Table 1 shows the mean effluent concentrations, the concentration ranges and the removal efficiencies of the pharmaceutical compounds tested in the SSF CWs under continuous-feeding regime at 2 and 4-d HRTs. In relation to removal efficiencies, the pharmaceutical compounds studied were classified into (i) very efficiently removed compounds with removal higher than 85% (ketoprofen and salicylic acid); (ii) moderately removed compounds with removal efficiencies between 50% and 85% (naproxen, ibuprofen, and caffeine); and (iii) poorly removed compounds with efficiency rate lower than 50% (carbamazepine, diclofenac and clofibric acid). Table 2 shows a comparison of the mean pharmaceutical removal efficiencies with other CW studies and conventional WWTPs indicating that CWs can offer similar or even better removal efficiencies for emerging organic pollutants. The very high removal efficiencies of salicylic acid are consistent with studies by Hijosa-Valsero et al. (2010a,b) and Matamoros and Bayona (2006) indicating that salicylic acid was an easily degradable substance in all CWs with removal efficiencies from 84% to 89%. For ketoprofen, when removal efficiencies were tested
in SSF CWs in a temperate climate (Barcelona, Spain) they were significantly lower (0–69%) (Matamoros and Bayona, 2006) than the values we observed. Since biodegradation is considered as a major elimination process of ketoprofen (Tixier et al., 2003), then the elevated temperatures in our tropical wetlands may play a role in the higher removal efficiencies that we observed. In addition, the relatively efficient removal of ibuprofen, naproxen, and caffeine observed in our planted CWs is also in good agreement with other studies (Matamoros and Bayona, 2006; Hijosa-Valsero et al., 2010a,b). In contrast, the relatively poor removal efficiencies of clofibric acid, diclofenac, and carbemazepine in our CWs have been also been found by others and attributed to the non-biodegradable and refractory nature of these compounds (Buser et al., 1998b; Matamoros and Bayona, 2006; Matamoros et al., 2009). In particular, carbamazepine is considered to be one of the most recalcitrant pharmaceuticals and the recalcitrant nature of this substance has been previously reported in other biological treatment systems (Bendz et al., 2005; Ternes et al., 2007). 3.2. Factors affecting pharmaceutical removal In our study, the removal efficiencies of all the pharmaceuticals (except for carbamazepine and salicylic acid) showed significant improvement (p < 0.05) in the planted beds compared to the unplanted beds clearly indicating that the presence of plants exerts a stimulatory effect on pharmaceutical elimination (Table 1). The precise mechanisms and pathway of pharmaceutical elimination, as well as the role that the higher aquatic plant plays are still unclear. However, it has been shown that the prevalence of aerobic conditions in the wetland subsurface may promote biochemical pathways through aerobic respiration, which are more efficient in removing emerging contaminants as compared to that of anaerobic pathways (Matamoros et al., 2008a; Hijosa-Valsero et al., 2010a). Furthermore, many authors have attributed this enhancement to the stimulatory effects of oxygen introduced into the subsurface
Table 1 Mean effluent concentrations (in lg L1), concentration ranges (in lg L1) and removal efficiencies (%) of the pharmaceutical compounds tested in the SSF CWs. Pharmaceutical compounds
HRT (2-d)
HRT (4-d)
Planted beds
Unplanted beds
Planted beds
Unplanted beds
Cabamazepine
18.3 ± 3.2 (10.1–24.9) 27
18.9 ± 3.8 (10.2–24.0) 24
18.1 ± 2.0 (10.4–21.6) 28
18.4 ± 2.3 (9.8–21.7) 26
Naproxen
4.8 ± 2.7a (0.5–10.7) 81
12.1 ± 5.9a (5.3–17.3) 52
1.7 ± 0.8a (0.3–5.2) 93
5.5 ± 3.2a (0.8–10.6) 78
Diclofenac
14.7 ± 4.5a (6.6–23.7) 41
17.1 ± 4.2a (10.5–21.4) 32
14.0 ± 3.5a (9.1–21.2) 44
19.1 ± 3.8a (11.6–24.4) 24
Ibuprofen
7.9 ± 5.0a (2.3–18.8) 68
9.8 ± 3.9a (4.2–12.1) 61
7.1 ± 2.4a (4.0–13.4) 72
12.1 ± 3.3 (6.3–18.4) 52
Caffeine
4.1 ± 0.7 (1.9–9.6) 84
4.6 ± 0.9 (2.3–10.7) 82
2.4 ± 0.5a (1.4–7.3) 90
4.0 ± 0.9a (1.9–10.1) 84
Salicylic acid
3.4 ± 1.4 (1.2–6.7) 86
2.9 ± 1.4 (1.4–8.0) 88
2.6 ± 1.6 (1.0–6.3) 90
3.0 ± 1.7 (0.9–8.4) 88
Ketoprofen
2.2 ± 0.9a (0.5–3.5) 91
2.5 ± 0.4a (0.6–3.9) 90
1.0 ± 0.4a (0.3–4.6) 96
1.5 ± 0.9a (0.5–4.0) 94
Clofibric acid
16.4 ± 5.7 (7.1–22.7) 34
16.7 ± 6.8 (8.3–22.9) 33
15.3 ± 5.6a (8.6–21.3) 39
16.5 ± 6.4a (9.0–24.3) 34
Note: mean (±standard deviation (SD)). a Statistically significant differences at a significance level of <0.05 (planted versus unplanted beds).
276
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
Table 2 Mean PPCP total removal efficiencies (%) and comparisons with other studies (SFs: sand filters; HSSF CWs: horizontal subsurface flow constructed wetlands; VSSF CWs: vertical subsurface flow constructed wetlands; FWS CWs: free water surface flow constructed wetlands).
Carbamazepine Naproxen Diclofenac Ibuprofen Caffeine Salicylic acid Ketoprofen Clofibric acid 1 2 3 4 5 6 7 8 9 10 11 12 13 14
CWs in this study
Other SFs
Other HSSF CWs
Other VSSF CWs
Other FWS CWs
Other hybrid CWs
Other WWTPs
24–28 52–93 24–44 52–72 82–90 86–90 94–96 33–39
8–111 66–801 39–761 39–761 75–981 77–981 – –
385; 58 0–472; 80–903; 905; 0–112; 0–453; 215; 17–522; 62–803; 625; 518; 85–942; 94–993; 77–972; 92–983; 0–693; 905 –
20–261 62–891 53–731 55–991 82–991 85–981 – –
3910; 3012 7210; 9210 8510; 9611 9610; 9611 – – 9810 97–9912 3410; 3611
– 73–859 65–879 42–999 83–969 93–979 77–819 –
77; 86; 3013; 667; 40–554 9313 55-9814 9–757; 176; 9-6014; 2213 907; 60–704 9613 79-10014; 9412 – 48–697 6513 51–100 14 2612
Matamoros et al. (2007). Matamoros and Bayona (2006) sand depth 0.5 m. Matamoros and Bayona (2006) sand depth 0.27 m. Carballa et al. (2004). Matamoros et al. (2009). Heberer (2002). Daughton and Ternes (1999). Matamoros et al. (2008a). Hijosa-Valsero et al. (2010a). Llorens et al. (2009). Matamoros et al. (2008b). Bernhard et al. (2006). Bendz et al. (2005). Lindqvist et al. (2005).
Fig. 1. Correlations between log Kow and removal efficiencies for the pharmaceutical compounds tested in the planted beds at 2-d (left) and 4-d HRT (right).
by the introduction of oxygen into the rhizosphere by the higher aquatic plant (Hijosa-Valsero et al., 2010a). Root exudates released by the plant in the rhizosphere, are known to result in intense microbial activity in the vicinity of roots (Brimecombe et al., 2001), and can enhance overall bioavailability of pharmaceuticals by acting as surfactants or transporters. In addition, research has indicated that active plant uptake can both directly and indirectly affect the fate of emerging organic pollutants in CWs (Dordio et al., 2009). Other research findings also indicated that passive plantassociated processes, specifically sorption also contributed to aqueous depletion of certain pharmaceuticals (Reinhold et al., 2010). In this manner, it might be expected that a compound’s removal in CWs might well be related to its hydrophobicity. Fig. 1 presents the correlation between micropollutants removal and log Kow. Although an inverse relation between log Kow and removal efficiency was apparent at both HRTs tested (Fig. 1), the correlations were not significant (p > 0.05), implying that the removal of these pharmaceuticals from CWs was not substantially related to the compound’s hydrophobicity. This finding is consistent with Park et al. (2009) who also found no distinct relationship between the removal of a variety of micropollutants from wetlands and their log Kow.
HRT is an important parameter for empirical design and operation of CWs. Significant (p < 0.05) correlations were observed for the removal rates of all 4 compounds (caffeine, salicylic acid, ketoprofen, and clofibric acid) with Pearson correlation values of 0.99 for all 4 compounds tested. This study showed that the removal rate of each of these four compounds tested was linearly proportional to the influent mass loading rate at HRTs between 2 to 6 d, making it possible to describe the removal of these selected pharmaceuticals in the wetlands by using first-order rate decay. Most wetland modelling utilizes first-order rate coefficients to describe the removal of organic compounds in treatment wetlands (Kadlec and Wallace, 2009). The first-order reaction can be modified by using certain reported kinetic data which are based on area (Kadlec and Knight, 1995; IWA, 2000):
Ce ¼ expðk=HLRÞ C0
ð1Þ
where Ce is the pharmaceutical concentration in the effluent (lg L1), C0 is the pharmaceutical concentration in the influent (lg L1), k is the area-based decay rate constant (m d1) and HLR is hydraulic loading rate (m d1). The values of k in all the planted
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
CW systems were higher than those in the unplanted systems for all compounds with the exception of salicylic acid where the values were nearly the same. The higher areal k values for caffeine, ketoprofen and clofibric acid (0.06, 0.08, and 0.01 m d1, respectively) in the planted beds as compared to the unplanted beds showed that macrophytes did indeed enhance pharmaceutical removal. There are very few reported studies in the literature with which we could compare our calculated reaction rate constants in CWs. Matamoros and Bayona (2006) calculated the k (based on the first order kinetics) for caffeine and reported a value of 0.14 m d1. By comparison, we calculated the areal k values of 0.06 m d1 for the continuous mode for caffeine. The k value found for caffeine in our study was significantly less than that determined by Matamoros and Bayona (2006). This might be attributed to the relatively low removal efficiencies (82–90%) we observed as compared to those (94–99%) in the study of Matamoros and Bayona (2006). 4. Conclusions The presence of plants in CWs significantly enhanced the removal of 6 of the 8 pharmaceutical compounds tested, including naproxen, ibuprofen, diclofenac, ketoprofen, caffeine and clofibric acid (the latter 2 compounds only at the 4-d HRT). Correlations between the octanol–water partition coefficient (log Kow) and removal efficiency at all HRTs tested were not significant (p > 0.05), implying that the removal of selected pharmaceuticals from these constructed wetlands was not substantially related to the compound’s hydrophobicity. Our findings on the specific also showed that the pharmaceutical removal efficiency rate was linearly proportional to the influent mass loading rate at HRTs between 2 to 6 d, and could be described by using a first order kinetics and decay constant (kv). The values of k in all the planted CW systems were higher than those in the unplanted systems for all the pharmaceutical compounds tested (with the exception of salicylic acid), showing that macrophytes did indeed enhance pharmaceutical removal. The rather efficient removal shown by the tropical wetlands tested here with HRTs of as little as 2–4 d indicates that such CW systems for removing conventional pollutants and selected pharmaceuticals from influent waters with minimal land requirement may be practical and viable in tropical regions. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2011.12.067. References Bendz, D., Paxéus, N.A., Ginn, T.T., Loge, F.J., 2005. Occurrence and fate of pharmaceutically active compounds in the environment: a case study: Höje River in Sweden. J. Hazard. Mater. 122, 195–204. Bernhard, M., Müller, J., Knepper, T.P., 2006. Biodegradation of persistent polar pollutants in wastewater: comparison of an optimised lab-scale membrane bioreactor and activated sludge treatment. Water Res. 40, 3149–3428. Brimecombe, M.J., Leij, F.A.A.M., Lynch, J.M., 2001. Rhizodeposition and microbial populations. In: Pinton, R., Varanini, Z., Nannipieri, P. (Eds.), The Rhizosphere: Biochemistry and Organic Substances at the Soil-plant Interface. Marcel Dekker, New York, pp. 74–98.
277
Buser, H., Poiger, T., Müller, M.D., 1998a. Occurrence and fate of the pharmaceutical drug diclofenac in surface waters: rapid photodegradation in a lake. Environ. Sci. Technol. 32, 3449–3456. Buser, H.R., Müller, M.D., Theobald, N., 1998b. Occurrence of the pharmaceutical drug clofibric acid and the herbicide mecoprop in various Swiss Lakes and in the North Sea. Environ. Sci. Technol. 32, 188–192. Carballa, M., Omil, F., Lema, J.M., Llompart, M., Carcía-Jares, C., Rodríguez, I., Gómez, M., Ternes, T., 2004. Behavior of pharmaceuticals, cosmetics and hormones in a sewage treatment plant. Water Res. 39, 2918–2926. Daughton, C.G., Ternes, T.A., 1999. Pharmaceuticals and personal care products in the environment: agents of subtle change. Environ. Health Persp. 107, 907–938. Dordio, A.V., Candeias, A.J.E., Pinto, A.P., Costa, C.T.D., Carvalho, A.J.P., 2009. Preliminary media screening for application in the removal of clofibric acid, carbamazepine and ibuprofen by SSF-constructed wetlands. Ecol. Eng. 35, 290– 302. Heberer, T., 2002. Occurrence, fate and removal of pharmaceutical residues in the aquatic environment: a review of recent research data. Toxicol. Lett. 131, 5–17. Hijosa-Valsero, M., Matamoros, V., Martín-Villacorta, J., Bécares, E., Bayona, J.M., 2010a. Assessment of full-scale natural systems for the removal of PPCPs from wastewater in small communities. Water Res. 44, 1429–1439. Hijosa-Valsero, M., Matamoros, V., Sidrach-Cardona, R., Martín-Villacorta, J., Bécares, E., Bayona, J.M., 2010b. Comprehensive assessment of the design configuration of constructed wetlands for the removal of pharmaceuticals and personal care products from urban wastewaters. Water Res. 44, 3669–3678. IWA, 2000. Constructed wetlands for pollution control. IWA Special Group on Use of Macrophytes in Water Pollution Control. Sci. and Tech. Rep. 8. IWA Publ., London. Joss, A., Zabczynski, S., Göbel, A., Hoffmann, B., Löffler, D., McArdell, C.S., Ternes, T.A., Siegrist, H., 2006. Biological degradation of pharmaceuticals in municipal wastewater treatment: proposing a classification scheme. Water Res. 40, 1686– 1696. Kadlec, R.H., Knight, R.L., 1995. Treatment Wetlands. Lewis Publ., Boca Raton, FL. Kadlec, R.H., Wallace, S.D., 2009. Treatment Wetlands, 2nd ed. CRC Press, Boca Raton, FL. Lindqvist, N., Tuhkanen, T., Kronberg, L., 2005. Occurrene of acidic pharmaceuticals in raw and treated sewages and in receiving waters. Water Res. 39, 2219–2228. Llorens, E., Matamoros, V., Domingo, V., Bayona, J.M., García, J., 2009. Water quality improvement in a full-scale tertiary constructed wetland: effects on conventional and specific organic contaminants. Sci. Total Environ. 407, 2517–2524. Matamoros, V., Arias, C., Brix, H., Bayona, J.M., 2007. Removal of pharmaceuticals and personal care products (PPCPs) from urban wastewater in a pilot vertical flow constructed wetland and a sand filter. Environ. Sci. Technol. 41, 8171– 8177. Matamoros, V., Arias, C.Brix, Bayona, J.M., 2009. Preliminary screening of smallscale domestic wastewater treatment systems for removal of pharmaceutical and personal care products. Water Res. 43, 55–62. Matamoros, V., Bayona, J.M., 2006. Elimination of pharmaceuticals and personal care products in subsurface flow constructed wetlands. Environ. Sci. Technol. 40, 5811–5816. Matamoros, V., García, J., Bayona, J.M., 2008a. Organic micropollutant removal in a full-scale surface flow constructed wetland fed with secondary effluent. Water Res. 42, 653–660. Matamoros, V., Osorio-Casellees, A., García, J., Bayona, J.M., 2008b. Behaviour of pharmaceutical products and biodegradation intermediates in horizontal subsurface flow constructed wetland. A microcosm experiment. Sci. Total Environ. 394, 171–176. Park, N., Vanderford, B.J., Snyder, S.A., Sarp, S., Kim, S.D., Cho, J., 2009. Effective controls of micropollutants in wastewater effluent using constructed wetlands under anoxic condition. Ecol. Eng. 35, 418–423. Reinhold, D., Vishwanathan, S., Park, J.J., Oh, D., Saunders, F.M., 2010. Assessment of plant-driven removal of emerging organic pollutants by duckweed. Chemosphere 80, 687–692. Scheren, P., 2008. Nature’s answers to the sanitation challenge: the application of constructed wetland technology for wastewater management. In: Addressing Land-based Activities in Western Indian Ocean, United Nations Environment Programme/Global Environment Facility (UNEP/GEF) WIO-Lab Project. Nairobi, Kenya. Ternes, T.A., Bounerz, M., Herrmann, N., Teiser, B., Andersen, H.R., 2007. Irrigation of treated wastewater in Braunschweig, Germany, an option to remove pharmaceuticals and musk fragrances. Chemosphere 66, 894–904. Tixier, C., Singer, H.P., Oellers, S., Müller, S.R., 2003. Occurrence and fate of carbamazepine, clofibric acid, diclofenac, ibuprofen, ketoprofen, and Naproxen in surface waters. Environ. Sci. Technol. 37, 1061–1068.
Contents lists available at SciVerse ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Technical Note
Pharmaceutical removal in tropical subsurface flow constructed wetlands at varying hydraulic loading rates Dong Qing Zhang a,⇑, Richard M. Gersberg b, Tao Hua a, Junfei Zhu a, Nguyen Anh Tuan a, Soon Keat Tan c a
DHI-NTU Centre, Nanyang Environment and Water Research Institute, Nanyang Technological University, N1.2-B1-02, 50 Nanyang Avenue, Singapore 639798, Singapore Graduate School of Public Health, San Diego State University, Hardy Tower 119, 5500 Campanile, San Diego, CA 92182-4162, USA c Maritime Research Centre, School of Civil and Environmental Engineering, Nanyang Technological University, N1.2-B1-02, 50 Nanyang Avenue, Singapore 639798, Singapore b
a r t i c l e
i n f o
Article history: Received 8 July 2011 Received in revised form 22 December 2011 Accepted 22 December 2011 Available online 21 January 2012 Keywords: Pharmaceutical compounds Subsurface flow Constructed wetlands Rhizosphere aeration Continuous-feeding regime
a b s t r a c t Determining the fate of emerging organic contaminants in an aquatic ecosystem is important for developing constructed wetlands (CWs) treatment technology. Experiments were carried out in subsurface flow CWs in Singapore to evaluate the fate and transport of eight pharmaceutical compounds. The CW system included three parallel horizontal subsurface flow CWs and three parallel unplanted beds fed continuously with synthetic wastewater at different hydraulic retention times (HRTs). The findings of the tests at 2–6 d HRTs showed that the pharmaceuticals could be categorized as (i) efficiently removed compounds with removal higher than 85% (ketoprofen and salicylic acid); (ii) moderately removed compounds with removal efficiencies between 50% and 85% (naproxen, ibuprofen and caffeine); and (iii) poorly removed compounds with efficiency rate lower than 50% (carbamazepine, diclofenac, and clofibric acid). Except for carbamazepine and salicylic acid, removal efficiencies of the selected pharmaceuticals showed significant (p < 0.05) enhancement in planted beds as compared to the unplanted beds. Removal of caffeine, ketoprofen and clofibric acid were found to follow first order decay kinetics with decay constants higher in the planted beds than the unplanted beds. Correlations between pharmaceutical removal efficiencies and log Kow were not significant (p > 0.05), implying that their removal is not well related to the compound’s hydrophobicity. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Many pharmaceutical compounds are the emerging contaminants in the modern society. Since these pharmaceuticals are typically present at trace levels (in the ng L1 range), and wastewater treatment plants (WWPTs) are not designed for their removal (Joss et al., 2006), many pharmaceuticals escape treatment and are released into the environment (Buser et al., 1998a; Heberer, 2002). Compared to conventional technical solutions for water treatment, constructed wetlands (CWs) provide a low-cost alternative which make them suitable for wastewater treatment where land availability is not a limiting factor (Kadlec and Knight, 1995). To date relatively little work has been conducted focusing on the removal of pharmaceuticals in CWs, and the most efficient configuration for their removal and the mechanisms involved are still unknown (Matamoros and Bayona, 2006; Park et al., 2009). The treatment efficiency of pollutants in a CW system is usually improved by decreasing the hydraulic loading (i.e., the longer the
⇑ Corresponding author. Tel.: +65 8165 6212; fax: +65 6790 6620. E-mail addresses: [email protected] (D.Q. Zhang), [email protected] (R.M. Gersberg), [email protected] (S.K. Tan). 0045-6535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2011.12.067
hydraulic retention time (HRT), the greater the pollutant removal efficiency). No specific guidelines have been established for designing a CW system to remove pharmaceutical compounds from water. Operating with a shorter HRT means smaller land-area requirements, which is a crucial factor for considering a CW. CWs in tropical regions (especially with their elevated temperatures year-round) are considerably more effective than systems in temperate climates, and may exhibit organic compound removal rates almost a factor 10 higher than standard CWs (Scheren, 2008). In this study, a variety of analgesic/anti-inflammatory drugs (naproxen, ibuprofen, diclofenac, ketoprofen and salicylic acid), a lipid regulator (clofibric acid), a stimulant (caffeine), and an antiepileptic drug (carbamazepine) were investigated. These products were selected because of their ubiquitous occurrence in sewage effluents and high frequency of detection reported in previous studies (Matamoros and Bayona, 2006; Hijosa-Valsero et al., 2010b). The specific goals of this study were to (i) investigate the removal efficiencies of selected pharmaceutical compounds in horizontal subsurface flow (HSSF) CWs planted with Typha angustifolia as compared to unplanted beds (sand filters) to evaluate the quantitative role of the aquatic plant on pharmaceutical removal under continuous-feeding regimes; and (ii) study the treatment kinetics of CW systems for operating at HRTs of shorter than 6 d.
274
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
2. Methodologies 2.1. Description of the mesocosm-scale CWs Six mesocosm-scale CWs were set up at the campus of Nanyang Technological University in Singapore. The location is a typical tropical environment and is generally hot and humid year round. The temperature ranges from 23 to 32 °C. All CWs in this study consisted of a fiberglass container (1.2 m 0.6 m 0.6 m, long wide deep), which contained the soil media and plants in the case of planted beds. The thickness of gravel bed was 0.30 m with 4–10 mm siliceous gravel (D60 = 3.5 mm) and the porosity was 0.45. Three CWs were left unplanted to act as controls (sand filters), while three CWs were planted with cattail (T. angustifolia) at a density of 14 plant m2 serving as HSSF CWs. In our case, T. angustifolia had been cultivated for one yr before purchase and then established in the tanks for 1 month. All the containers had a flat bottom and a horizontal drainage pipe (0.4 m long and 50 mm in diameter) located at the lower edge of the containers. 2.2. Sampling regime Two series of wetland studies were carried out: (i) the investigation of carbamazepine, naproxen, ibuprofen and diclofenac removal by the wetlands; and (ii) the investigation of the fate of caffeine, salicylic acid, ketoprofen and clofibric acid in the wetlands. For each series, the wetlands were operated in a continuous flow mode for four months. Plants were replaced after the first series was completed. Prior to beginning data collection for each series, all the wetland beds were allowed to adapt to the new steady state conditions for a period of 15 d. Additionally, different HRTs (2 and 4-d) were tested in both series and an additional 6-d HRT was included in the second series for investigating pharmaceutical removal kinetics. For logistical reasons, the operational modes were tested in sequence rather than in parallel. However, it is important to note here that the equatorial climate of Singapore offers a rather constant temperature and degree of solar radiation, without the seasonal variations of temperate countries. Therefore the environmental conditions of the plants and gravel beds were for the most part similar during the two periods of comparison. Continuous loading was achieved by using a six-channel peristaltic pump. A holding tank of 150 L was served as the reservoir for dispensing synthetic wastewater. In both series, the hydraulic application rates were 5.6 cm d1 (2-d HRT) and 2.8 cm d1 (4-d HRT), respectively. In addition, in the second series, a 6-d of HRT (1.9 cm d1 of hydraulic application rate) was tested. In both series of experiments, wastewater effluent samples were always collected from each of the 6 beds every other day. All the samples were collected on the same day and at the same time of day, i.e., during early morning. These samples were then transported refrigerated (4 °C) to the laboratory, where they were analyzed within 24 h. 2.3. Materials and chemicals Analytical grade glucose anhydrous (99% purity), ammonium sulfate (98% purity), magnesium sulfate heptahydrate (99% purity), calcium chloride dehydrate (99% purity), calcium chloride (74–78% purity), iron (III) chloride anhydrous (98% purity), sodium carbonate (98% purity), sodium hydrogen carbonate (99% purity), potassium dihydrogen phosphate (98% purity) obtained from Alfa Aesar, Germany were used for the preparation of synthetic wastewater. Matamoros et al. (2008b) fed SSF CWs a similar synthetic wastewater containing both readily assimilable (dissolved) glucose
and a particulate form of organic matter in the form of starch, and found that pharmaceutical removal efficiencies were independent of the organic matter type (dissolved versus particulate). Therefore we chose glucose as the carbon source for this study. HPLC-grade acetonitrile, methanol, hexane and ethyl acetate were obtained from Fisher (USA). Ultrapure water was obtained from Milli-Q water purification system (Millipore, Bedford, USA). Carbamazepine, diclofenac, ibuprofen and naproxen (97–100% purity) were obtained from Sigma–Aldrich (Singapore). The solid phase extraction (SPE) cartridges GracePureTM C18 (500 mg, 6 mL) were purchased from Grace Davison Discovery Sciences (Belgium). The HPLC column Inertsil ODS (35 lm, 4.6 150 mm) was obtained from Alpha Analytical (Singapore). The 0.45 lm glass fiber filters of 47 mm (Whatman) were purchased from Schleicher and Schuell (Germany). Stock solution of 50 mg L1 was prepared in methanol and stored at 4 °C. Working solutions were prepared by diluting the stock solution with methanol. During the experiments, all the beds were fed with synthetic wastewater with the same organic load. The composition of influent included (in mg L1) 300 COD, 27 NH4+, 7 PO4-P, 4 Fe3+, 7 Mg2+, and 6 Ca2+. With respect to the pharmaceutical injection, 50 L of synthetic wastewater were spiked with 1.25 mg of each pharmaceutical compound to obtain a final concentration of 25 lg L1. 2.4. Analytical procedures for pharmaceutical compounds Prior to extraction, 500 mL of effluent wastewater were filtered through a 0.45 lm glass fibre membrane filter (Millipore, USA) and then acidified to pH 2 with hydrochloric acid. The SPE cartridges were conditioned using 5 mL n-hexane, 5 mL ethyl acetate, 10 mL methanol and 10 mL of Milli-Q water at a flow-rate of approximate 3 mL min1. Samples were percolated to the SPE cartridges through a Teflon tube at a flow-rate of approximate 10 mL min1. The cartridge filter was then rinsed with the combination of methanol and Milli-Q (methanol: Milli-Q water = 10:90), followed by 20 mL of Milli-Q water. Thereafter the cartridges were allowed to dry for ca 30 min and then eluted with 5 mL ethyl acetate with elution solutions collected in 15 mL calibrated centrifuge tubes. The extracted solution was then concentrated to ca 400 lL under a gentle nitrogen stream and was then reconstituted to 500 lL with methanol. Chromatographic analysis was performed on a Shimadzu Ultra Fast Liquid Chromatograph (Shimadzu, Japan) equipped with a quaternary LC-20AD pump, a CTO-20A oven, and a SPD-M20A Diode Array Detector. The injector was SIL-20A HT fitted to a Shimadzu autosampler with a 20 lL sample loop. Chromatographic separations were carried out using an Inertsil ODS-3 (4.6 150 mm, 5 lm) HPLC column protected by a ODS-3 (C18) (4.6 50 mm, 5 lm) guard column (Alpha Analytical, Singapore). The system was controlled using an interface module and a personal computer. Chromatograms were processed using a ‘‘Shimadzu LCSolution program’’. 2.5. Recoveries, LOD and LOQ Effluent wastewater (1 L) was spiked with the pharmaceutical compounds to a concentration of 5 lg L1. The sample was extracted according to the procedure described earlier. The signals obtained from spiked samples were compared with signals from wastewater without standard addition and with those obtained by injecting standard solutions. In this study, the recoveries ranged from 70% to 105%. Limits of detection (LOD) and limits of quantification (LOQ) of the analytical procedure were established by using a water effluent from a microcosm plant effluent without pharmaceuticals. LOD and LOQ were determined based on a signal-to-noise ratio of 3
275
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
and 10, respectively (the ratio between peak intensity and intensity of the noise was used). The values of LOD and LOQ for 8 pharmaceutical compounds tested in our study ranged from 0.006 to 0.028 lg L1 and 0.019 to 0.083 lg L1, respectively. 2.6. Statistical test of experimental data The experimental data were tested for normal distribution. Tests to determine statistical differences between treatments were carried out by comparing the critical value through ANOVA. Comparisons were considered significantly different for p < 0.05. 3. Results and discussion 3.1. Removal efficiencies of pharmaceutical compounds Table 1 shows the mean effluent concentrations, the concentration ranges and the removal efficiencies of the pharmaceutical compounds tested in the SSF CWs under continuous-feeding regime at 2 and 4-d HRTs. In relation to removal efficiencies, the pharmaceutical compounds studied were classified into (i) very efficiently removed compounds with removal higher than 85% (ketoprofen and salicylic acid); (ii) moderately removed compounds with removal efficiencies between 50% and 85% (naproxen, ibuprofen, and caffeine); and (iii) poorly removed compounds with efficiency rate lower than 50% (carbamazepine, diclofenac and clofibric acid). Table 2 shows a comparison of the mean pharmaceutical removal efficiencies with other CW studies and conventional WWTPs indicating that CWs can offer similar or even better removal efficiencies for emerging organic pollutants. The very high removal efficiencies of salicylic acid are consistent with studies by Hijosa-Valsero et al. (2010a,b) and Matamoros and Bayona (2006) indicating that salicylic acid was an easily degradable substance in all CWs with removal efficiencies from 84% to 89%. For ketoprofen, when removal efficiencies were tested
in SSF CWs in a temperate climate (Barcelona, Spain) they were significantly lower (0–69%) (Matamoros and Bayona, 2006) than the values we observed. Since biodegradation is considered as a major elimination process of ketoprofen (Tixier et al., 2003), then the elevated temperatures in our tropical wetlands may play a role in the higher removal efficiencies that we observed. In addition, the relatively efficient removal of ibuprofen, naproxen, and caffeine observed in our planted CWs is also in good agreement with other studies (Matamoros and Bayona, 2006; Hijosa-Valsero et al., 2010a,b). In contrast, the relatively poor removal efficiencies of clofibric acid, diclofenac, and carbemazepine in our CWs have been also been found by others and attributed to the non-biodegradable and refractory nature of these compounds (Buser et al., 1998b; Matamoros and Bayona, 2006; Matamoros et al., 2009). In particular, carbamazepine is considered to be one of the most recalcitrant pharmaceuticals and the recalcitrant nature of this substance has been previously reported in other biological treatment systems (Bendz et al., 2005; Ternes et al., 2007). 3.2. Factors affecting pharmaceutical removal In our study, the removal efficiencies of all the pharmaceuticals (except for carbamazepine and salicylic acid) showed significant improvement (p < 0.05) in the planted beds compared to the unplanted beds clearly indicating that the presence of plants exerts a stimulatory effect on pharmaceutical elimination (Table 1). The precise mechanisms and pathway of pharmaceutical elimination, as well as the role that the higher aquatic plant plays are still unclear. However, it has been shown that the prevalence of aerobic conditions in the wetland subsurface may promote biochemical pathways through aerobic respiration, which are more efficient in removing emerging contaminants as compared to that of anaerobic pathways (Matamoros et al., 2008a; Hijosa-Valsero et al., 2010a). Furthermore, many authors have attributed this enhancement to the stimulatory effects of oxygen introduced into the subsurface
Table 1 Mean effluent concentrations (in lg L1), concentration ranges (in lg L1) and removal efficiencies (%) of the pharmaceutical compounds tested in the SSF CWs. Pharmaceutical compounds
HRT (2-d)
HRT (4-d)
Planted beds
Unplanted beds
Planted beds
Unplanted beds
Cabamazepine
18.3 ± 3.2 (10.1–24.9) 27
18.9 ± 3.8 (10.2–24.0) 24
18.1 ± 2.0 (10.4–21.6) 28
18.4 ± 2.3 (9.8–21.7) 26
Naproxen
4.8 ± 2.7a (0.5–10.7) 81
12.1 ± 5.9a (5.3–17.3) 52
1.7 ± 0.8a (0.3–5.2) 93
5.5 ± 3.2a (0.8–10.6) 78
Diclofenac
14.7 ± 4.5a (6.6–23.7) 41
17.1 ± 4.2a (10.5–21.4) 32
14.0 ± 3.5a (9.1–21.2) 44
19.1 ± 3.8a (11.6–24.4) 24
Ibuprofen
7.9 ± 5.0a (2.3–18.8) 68
9.8 ± 3.9a (4.2–12.1) 61
7.1 ± 2.4a (4.0–13.4) 72
12.1 ± 3.3 (6.3–18.4) 52
Caffeine
4.1 ± 0.7 (1.9–9.6) 84
4.6 ± 0.9 (2.3–10.7) 82
2.4 ± 0.5a (1.4–7.3) 90
4.0 ± 0.9a (1.9–10.1) 84
Salicylic acid
3.4 ± 1.4 (1.2–6.7) 86
2.9 ± 1.4 (1.4–8.0) 88
2.6 ± 1.6 (1.0–6.3) 90
3.0 ± 1.7 (0.9–8.4) 88
Ketoprofen
2.2 ± 0.9a (0.5–3.5) 91
2.5 ± 0.4a (0.6–3.9) 90
1.0 ± 0.4a (0.3–4.6) 96
1.5 ± 0.9a (0.5–4.0) 94
Clofibric acid
16.4 ± 5.7 (7.1–22.7) 34
16.7 ± 6.8 (8.3–22.9) 33
15.3 ± 5.6a (8.6–21.3) 39
16.5 ± 6.4a (9.0–24.3) 34
Note: mean (±standard deviation (SD)). a Statistically significant differences at a significance level of <0.05 (planted versus unplanted beds).
276
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
Table 2 Mean PPCP total removal efficiencies (%) and comparisons with other studies (SFs: sand filters; HSSF CWs: horizontal subsurface flow constructed wetlands; VSSF CWs: vertical subsurface flow constructed wetlands; FWS CWs: free water surface flow constructed wetlands).
Carbamazepine Naproxen Diclofenac Ibuprofen Caffeine Salicylic acid Ketoprofen Clofibric acid 1 2 3 4 5 6 7 8 9 10 11 12 13 14
CWs in this study
Other SFs
Other HSSF CWs
Other VSSF CWs
Other FWS CWs
Other hybrid CWs
Other WWTPs
24–28 52–93 24–44 52–72 82–90 86–90 94–96 33–39
8–111 66–801 39–761 39–761 75–981 77–981 – –
385; 58 0–472; 80–903; 905; 0–112; 0–453; 215; 17–522; 62–803; 625; 518; 85–942; 94–993; 77–972; 92–983; 0–693; 905 –
20–261 62–891 53–731 55–991 82–991 85–981 – –
3910; 3012 7210; 9210 8510; 9611 9610; 9611 – – 9810 97–9912 3410; 3611
– 73–859 65–879 42–999 83–969 93–979 77–819 –
77; 86; 3013; 667; 40–554 9313 55-9814 9–757; 176; 9-6014; 2213 907; 60–704 9613 79-10014; 9412 – 48–697 6513 51–100 14 2612
Matamoros et al. (2007). Matamoros and Bayona (2006) sand depth 0.5 m. Matamoros and Bayona (2006) sand depth 0.27 m. Carballa et al. (2004). Matamoros et al. (2009). Heberer (2002). Daughton and Ternes (1999). Matamoros et al. (2008a). Hijosa-Valsero et al. (2010a). Llorens et al. (2009). Matamoros et al. (2008b). Bernhard et al. (2006). Bendz et al. (2005). Lindqvist et al. (2005).
Fig. 1. Correlations between log Kow and removal efficiencies for the pharmaceutical compounds tested in the planted beds at 2-d (left) and 4-d HRT (right).
by the introduction of oxygen into the rhizosphere by the higher aquatic plant (Hijosa-Valsero et al., 2010a). Root exudates released by the plant in the rhizosphere, are known to result in intense microbial activity in the vicinity of roots (Brimecombe et al., 2001), and can enhance overall bioavailability of pharmaceuticals by acting as surfactants or transporters. In addition, research has indicated that active plant uptake can both directly and indirectly affect the fate of emerging organic pollutants in CWs (Dordio et al., 2009). Other research findings also indicated that passive plantassociated processes, specifically sorption also contributed to aqueous depletion of certain pharmaceuticals (Reinhold et al., 2010). In this manner, it might be expected that a compound’s removal in CWs might well be related to its hydrophobicity. Fig. 1 presents the correlation between micropollutants removal and log Kow. Although an inverse relation between log Kow and removal efficiency was apparent at both HRTs tested (Fig. 1), the correlations were not significant (p > 0.05), implying that the removal of these pharmaceuticals from CWs was not substantially related to the compound’s hydrophobicity. This finding is consistent with Park et al. (2009) who also found no distinct relationship between the removal of a variety of micropollutants from wetlands and their log Kow.
HRT is an important parameter for empirical design and operation of CWs. Significant (p < 0.05) correlations were observed for the removal rates of all 4 compounds (caffeine, salicylic acid, ketoprofen, and clofibric acid) with Pearson correlation values of 0.99 for all 4 compounds tested. This study showed that the removal rate of each of these four compounds tested was linearly proportional to the influent mass loading rate at HRTs between 2 to 6 d, making it possible to describe the removal of these selected pharmaceuticals in the wetlands by using first-order rate decay. Most wetland modelling utilizes first-order rate coefficients to describe the removal of organic compounds in treatment wetlands (Kadlec and Wallace, 2009). The first-order reaction can be modified by using certain reported kinetic data which are based on area (Kadlec and Knight, 1995; IWA, 2000):
Ce ¼ expðk=HLRÞ C0
ð1Þ
where Ce is the pharmaceutical concentration in the effluent (lg L1), C0 is the pharmaceutical concentration in the influent (lg L1), k is the area-based decay rate constant (m d1) and HLR is hydraulic loading rate (m d1). The values of k in all the planted
D.Q. Zhang et al. / Chemosphere 87 (2012) 273–277
CW systems were higher than those in the unplanted systems for all compounds with the exception of salicylic acid where the values were nearly the same. The higher areal k values for caffeine, ketoprofen and clofibric acid (0.06, 0.08, and 0.01 m d1, respectively) in the planted beds as compared to the unplanted beds showed that macrophytes did indeed enhance pharmaceutical removal. There are very few reported studies in the literature with which we could compare our calculated reaction rate constants in CWs. Matamoros and Bayona (2006) calculated the k (based on the first order kinetics) for caffeine and reported a value of 0.14 m d1. By comparison, we calculated the areal k values of 0.06 m d1 for the continuous mode for caffeine. The k value found for caffeine in our study was significantly less than that determined by Matamoros and Bayona (2006). This might be attributed to the relatively low removal efficiencies (82–90%) we observed as compared to those (94–99%) in the study of Matamoros and Bayona (2006). 4. Conclusions The presence of plants in CWs significantly enhanced the removal of 6 of the 8 pharmaceutical compounds tested, including naproxen, ibuprofen, diclofenac, ketoprofen, caffeine and clofibric acid (the latter 2 compounds only at the 4-d HRT). Correlations between the octanol–water partition coefficient (log Kow) and removal efficiency at all HRTs tested were not significant (p > 0.05), implying that the removal of selected pharmaceuticals from these constructed wetlands was not substantially related to the compound’s hydrophobicity. Our findings on the specific also showed that the pharmaceutical removal efficiency rate was linearly proportional to the influent mass loading rate at HRTs between 2 to 6 d, and could be described by using a first order kinetics and decay constant (kv). The values of k in all the planted CW systems were higher than those in the unplanted systems for all the pharmaceutical compounds tested (with the exception of salicylic acid), showing that macrophytes did indeed enhance pharmaceutical removal. The rather efficient removal shown by the tropical wetlands tested here with HRTs of as little as 2–4 d indicates that such CW systems for removing conventional pollutants and selected pharmaceuticals from influent waters with minimal land requirement may be practical and viable in tropical regions. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2011.12.067. References Bendz, D., Paxéus, N.A., Ginn, T.T., Loge, F.J., 2005. Occurrence and fate of pharmaceutically active compounds in the environment: a case study: Höje River in Sweden. J. Hazard. Mater. 122, 195–204. Bernhard, M., Müller, J., Knepper, T.P., 2006. Biodegradation of persistent polar pollutants in wastewater: comparison of an optimised lab-scale membrane bioreactor and activated sludge treatment. Water Res. 40, 3149–3428. Brimecombe, M.J., Leij, F.A.A.M., Lynch, J.M., 2001. Rhizodeposition and microbial populations. In: Pinton, R., Varanini, Z., Nannipieri, P. (Eds.), The Rhizosphere: Biochemistry and Organic Substances at the Soil-plant Interface. Marcel Dekker, New York, pp. 74–98.
277
Buser, H., Poiger, T., Müller, M.D., 1998a. Occurrence and fate of the pharmaceutical drug diclofenac in surface waters: rapid photodegradation in a lake. Environ. Sci. Technol. 32, 3449–3456. Buser, H.R., Müller, M.D., Theobald, N., 1998b. Occurrence of the pharmaceutical drug clofibric acid and the herbicide mecoprop in various Swiss Lakes and in the North Sea. Environ. Sci. Technol. 32, 188–192. Carballa, M., Omil, F., Lema, J.M., Llompart, M., Carcía-Jares, C., Rodríguez, I., Gómez, M., Ternes, T., 2004. Behavior of pharmaceuticals, cosmetics and hormones in a sewage treatment plant. Water Res. 39, 2918–2926. Daughton, C.G., Ternes, T.A., 1999. Pharmaceuticals and personal care products in the environment: agents of subtle change. Environ. Health Persp. 107, 907–938. Dordio, A.V., Candeias, A.J.E., Pinto, A.P., Costa, C.T.D., Carvalho, A.J.P., 2009. Preliminary media screening for application in the removal of clofibric acid, carbamazepine and ibuprofen by SSF-constructed wetlands. Ecol. Eng. 35, 290– 302. Heberer, T., 2002. Occurrence, fate and removal of pharmaceutical residues in the aquatic environment: a review of recent research data. Toxicol. Lett. 131, 5–17. Hijosa-Valsero, M., Matamoros, V., Martín-Villacorta, J., Bécares, E., Bayona, J.M., 2010a. Assessment of full-scale natural systems for the removal of PPCPs from wastewater in small communities. Water Res. 44, 1429–1439. Hijosa-Valsero, M., Matamoros, V., Sidrach-Cardona, R., Martín-Villacorta, J., Bécares, E., Bayona, J.M., 2010b. Comprehensive assessment of the design configuration of constructed wetlands for the removal of pharmaceuticals and personal care products from urban wastewaters. Water Res. 44, 3669–3678. IWA, 2000. Constructed wetlands for pollution control. IWA Special Group on Use of Macrophytes in Water Pollution Control. Sci. and Tech. Rep. 8. IWA Publ., London. Joss, A., Zabczynski, S., Göbel, A., Hoffmann, B., Löffler, D., McArdell, C.S., Ternes, T.A., Siegrist, H., 2006. Biological degradation of pharmaceuticals in municipal wastewater treatment: proposing a classification scheme. Water Res. 40, 1686– 1696. Kadlec, R.H., Knight, R.L., 1995. Treatment Wetlands. Lewis Publ., Boca Raton, FL. Kadlec, R.H., Wallace, S.D., 2009. Treatment Wetlands, 2nd ed. CRC Press, Boca Raton, FL. Lindqvist, N., Tuhkanen, T., Kronberg, L., 2005. Occurrene of acidic pharmaceuticals in raw and treated sewages and in receiving waters. Water Res. 39, 2219–2228. Llorens, E., Matamoros, V., Domingo, V., Bayona, J.M., García, J., 2009. Water quality improvement in a full-scale tertiary constructed wetland: effects on conventional and specific organic contaminants. Sci. Total Environ. 407, 2517–2524. Matamoros, V., Arias, C., Brix, H., Bayona, J.M., 2007. Removal of pharmaceuticals and personal care products (PPCPs) from urban wastewater in a pilot vertical flow constructed wetland and a sand filter. Environ. Sci. Technol. 41, 8171– 8177. Matamoros, V., Arias, C.Brix, Bayona, J.M., 2009. Preliminary screening of smallscale domestic wastewater treatment systems for removal of pharmaceutical and personal care products. Water Res. 43, 55–62. Matamoros, V., Bayona, J.M., 2006. Elimination of pharmaceuticals and personal care products in subsurface flow constructed wetlands. Environ. Sci. Technol. 40, 5811–5816. Matamoros, V., García, J., Bayona, J.M., 2008a. Organic micropollutant removal in a full-scale surface flow constructed wetland fed with secondary effluent. Water Res. 42, 653–660. Matamoros, V., Osorio-Casellees, A., García, J., Bayona, J.M., 2008b. Behaviour of pharmaceutical products and biodegradation intermediates in horizontal subsurface flow constructed wetland. A microcosm experiment. Sci. Total Environ. 394, 171–176. Park, N., Vanderford, B.J., Snyder, S.A., Sarp, S., Kim, S.D., Cho, J., 2009. Effective controls of micropollutants in wastewater effluent using constructed wetlands under anoxic condition. Ecol. Eng. 35, 418–423. Reinhold, D., Vishwanathan, S., Park, J.J., Oh, D., Saunders, F.M., 2010. Assessment of plant-driven removal of emerging organic pollutants by duckweed. Chemosphere 80, 687–692. Scheren, P., 2008. Nature’s answers to the sanitation challenge: the application of constructed wetland technology for wastewater management. In: Addressing Land-based Activities in Western Indian Ocean, United Nations Environment Programme/Global Environment Facility (UNEP/GEF) WIO-Lab Project. Nairobi, Kenya. Ternes, T.A., Bounerz, M., Herrmann, N., Teiser, B., Andersen, H.R., 2007. Irrigation of treated wastewater in Braunschweig, Germany, an option to remove pharmaceuticals and musk fragrances. Chemosphere 66, 894–904. Tixier, C., Singer, H.P., Oellers, S., Müller, S.R., 2003. Occurrence and fate of carbamazepine, clofibric acid, diclofenac, ibuprofen, ketoprofen, and Naproxen in surface waters. Environ. Sci. Technol. 37, 1061–1068.