Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transformation products as intrinsic redox-sensitive tracers

STOTEN-21686; No of Pages 9 Science of the Total Environment xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transformation products as intrinsic redox-sensitive tracers Meriam Muntau a, Manoj Schulz b, Kevin S. Jewell b, Nina Hermes b, Uwe Hübner a,⁎, Thomas Ternes b, Jörg E. Drewes a a b

Technical University of Munich, Chair of Urban Water Systems Engineering, Am Coulombwall 3, 85748 Garching, Germany Federal Institute of Hydrology, Am Mainzer Tor 1, 56068 Koblenz, Germany

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

• Two wet-dry cycles of a full-scale infiltration basin were monitored. • Known products from oxic transformation were identified to serve as intrinsic redox tracer. • The occurrence of intermediates confirmed limited availability of dissolved oxygen.

a r t i c l e

i n f o

Article history: Received 13 October 2016 Received in revised form 21 December 2016 Accepted 23 December 2016 Available online xxxx Editor: Jay Gan Keywords: Chemicals of emerging concern Dissolved organic carbon Iopromide Soil-aquifer treatment Transformation products

a b s t r a c t In this study, known products from oxic transformation of the X-ray contrast medium iopromide were introduced for the first time as intrinsic tracer for in situ characterization of the transition zone between oxic and suboxic conditions during the initial phase of soil-aquifer treatment (SAT). Two wet-dry cycles of a full-scale infiltration basin were monitored to characterize hydraulic retention times, redox conditions, removal of bulk organic parameters and the fate of chemicals of emerging concern (CECs). Tracer tests at the site showed an average hydraulic retention time of b 20 h before collection in drainage pipes located approximately 1.5 m below surface. Dissolved oxygen at different depth rapidly depleted and only increased towards the end of the flooding event. Transformation of iopromide and all known intermediates to persistent transformation products (TPs) usually occurring during oxic biodegradation was very limited in samples from suction cups immediately underneath the basin. But transformation was complete in samples collected from the drainage outlet indicating that dissolved oxygen had been introduced to the system before sample collection in the combined drainage outlet. Similar to iopromide and its TPs, removal of several CECs (diclofenac, bezafibrate, mecoprop, TCEP) was inefficient after 90 cm infiltration (b 35%) but significantly enhanced in the combined drainage outlet (N 80%). These results highlight that the analysis of iopromide along with its intermediates and persistent TPs can serve as a promising probing tool to determine overall efficiency of CEC biodegradation and to identify potential in situ oxygen limitations. © 2017 Elsevier B.V. All rights reserved.

⁎ Corresponding author. E-mail address: [email protected] (U. Hübner).

http://dx.doi.org/10.1016/j.scitotenv.2016.12.165 0048-9697/© 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

2

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

1. Introduction Water scarcity is a growing challenge in the world and freshwater resources increasingly run short to satisfy demand. Due to the uneven distribution of water resources and increasing degree of urbanization, water shortages are often a climate-related regional problem. Many communities depend on water resources that are partially replenished intentionally or unintentionally with treated wastewater (Pal et al. 2010). The presence of chemicals of anthropogenic origin in these water resources, such as pharmaceuticals, personal care products, household chemicals, endocrine disrupting compounds, flame retardants, plasticizers and pesticides including their respective metabolites and transformation products (TPs) has been recognized for a longer time (La Farre et al. 2008; Celiz et al. 2009; de Jongh et al. 2012). Even though most of these so-called ‘chemicals of emerging concern’ (CECs) are present at a low concentrations (nanogram per liter or even pikogram per liter, ng/L or pg/L range), there are concerns regarding potential adverse human and environmental health effects (Jones et al. 2004; Schwarzenbach et al. 2006; Du et al. 2014). Therefore, understanding the fate and transport of these chemicals, an assessment of potential adverse health effects of CECs, as well as more effective treatment systems for their mitigation are challenges that need to be addressed. Soil-aquifer treatment (SAT) is a rather inexpensive, sustainable treatment and reclamation technology to generate higher-quality water from treated domestic wastewater effluents for non-potable and potable uses (Regnery et al. 2013). SAT consists of a controlled passage of treated wastewater through porous media mainly for purification purposes, as well as for seasonal and multiannual storage. SAT includes a three-component treatment process consisting of the infiltration zone, vadose zone, and the underlying aquifer. Purification processes in SAT systems include filtration, biotransformation, mineralization, physical adsorption, chemical precipitation, and ion-exchange (Kanarek and Michail 1996; Maeng et al. 2011; Regnery et al. 2013). While this natural treatment system is highly attractive due to its low carbon footprint and the minimal generation of residuals, drawbacks of this technology are the rather large physical footprint requirements and a lack of understanding of system dynamics resulting in variable qualities of the recovered water (Regnery et al. 2013). SAT proved to be effective in the removal of turbidity, total nitrogen, phosphorus, pathogenic bacteria, protozoa, and viruses (Gerba et al. 1991, Idelovitch et al., 2003). A significant removal of dissolved organic carbon (DOC) after SAT has been reported (Quanrud et al. 2003; Drewes et al. 2006; Amy and Drewes 2007). Previous studies have also demonstrated that SAT is effective in reducing the concentrations of various CECs that are still present in wastewater treatment plant (WWTP) effluents (Quanrud et al. 2003; Amy and Drewes 2007; Ternes et al. 2007; Laws et al. 2011). The attenuation of trace organic chemicals during SAT strongly depends on structural moieties of the chemical, hydrogeological conditions, residence times, travel distances, redox conditions, temperature, availability of a primary substrate, microbial activity and functionality (Grünheid et al. 2008, Maeng et al. 2011, Li et al. 2014, Alidina et al. 2014). However, some residual trace organic chemicals, such as X-ray contrast media, antibiotics, chlorinated flame retardants and antiepileptic drugs (e.g., carbamazepine, primidone), persist during SAT and can occur in reclaimed water and recovered groundwater samples in the elevated ng/L-range (Snyder et al. 2003; Regnery et al. 2013). Several studies investigated the removal efficiency of organic matter and CECs during SAT. However, current knowledge is insufficient to properly reveal the transformation dynamics of CECs as a function of predominant redox conditions in particular during the initial phase of infiltration. The availability of dissolved oxygen plays a key role in determining the mobility, dissolution, transformation and toxicity of most CECs (Regnery et al. 2015a). Already at low dissolved oxygen

concentrations (defined as suboxic conditions with DO b 1 mg/L and little denitrification with ΔNO− 3 b 0.5 mg N/L) the microbial degradation of redox-sensitive CECs can be abated (Regnery et al. 2015a). While anoxic conditions can be distinguished by monitoring different redox pa− 2+ rameters (NH+ /Mn4 +; Fe2 +/Fe3 +; SO24 −/H2S), in situ 4 /NO3 ; Mn characterization of the transition between oxic and suboxic conditions is often hindered by inappropriate sampling techniques (e.g., suction cup samplers, redox probes). During this study, we utilized for the first time known transformation products (TPs) of the X-ray contrast medium iopromide as intrinsic tracers of oxygen limitations. This probing tool might be suitable to characterize the biotransformation of other CECs during short-term SAT. In order to test this hypothesis, this study investigated the performance of a full-scale SAT facility with short hydraulic retention time during travel through a 1.5 m deep vadose zone.

2. Materials and methods 2.1. Study site description In this study, an infiltration basin operated by the Abwasserverband Braunschweig, Germany and located in a wetland area of approximately 150 ha was selected for the field experiments. The WWTP Steinhof, with an annual treatment capacity of approximately 22 million m3, is employing full biological nutrient removal (i.e., nitrogen and phosphorus) by operating a three stage activated sludge system (with anaerobic, anoxic and oxic zones). About one third of the secondary effluent flow is fed into the wetland area. The entire wetland area is equipped with a drainage system that dates back to the end of the 20th century but still catches about 85% of the infiltrated water. The recovered water is subsequently drained via open ditches to the Aue-Oker-Canal and finally discharged into the river Oker. Within this area, infiltration basin S62 (Fig. S1, Supplemental information) with a surface area of approximately 1225 m2 was selected for full-scale SAT experiments. Water applied to the infiltration basin and percolating through the subsurface is collected in the historic drainage pipes located approximately 1.5 m below surface. The drainage pipes for this infiltration basin are combined into a common discharge point located approximately 150 m to the northwest of the basin. The combined drainage also collects water from adjacent infiltration basins, which were not operated during the entire period of the experiments.

2.2. Basin operation and sampling procedures Prior to the experiments, the infiltration basin was flooded on a regular basis every two to three weeks for a period of approximately 3 months. Two experiments were executed in consecutive trials (September and November). Before the field trials, the ground surface of the infiltration basin was mown and plowed to reduce potential uptake of water by plants growing in the basins. Precise flow measurement devices at the delivery point were not available, but the influent volume was estimated based on basin dimensions (length and width) and the impounding depth. The empty infiltration basin was flooded once to a water level of 36 cm above ground surface (resulting in a recharge volume of approximately 440 m3) in September and 40 cm (resulting in a recharge volume of approximately 490 m3) in November 2015. Samples were taken from secondary effluent (basin influent), from the flooded infiltration basin (2 samples/day taken at opposite side of delivery point), and from the combined drainage outlet (after vadose zone treatment). For the second experiment (November) additional suction cups (UMS; Germany) and oxygen optodes (PreSens; Germany) were installed at 30 cm, 60 cm and 90 cm depth beneath the infiltration basin.

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

2.3. Analytical methods All samples were analyzed on-site for pH, temperature, electrical conductivity, redox potential, and dissolved oxygen directly after sampling (Multi 3410; WTW; Germany) and subsequently filtered using 0.45 μm cellulose nitrate membrane filters. The electrical conductivity at the drainage outlet was continuously monitored during daytime (8 am–4 pm). Water samples for CEC analysis were stored at − 20 °C while the samples for bulk organic parameters, ammonia, alkalinity, phosphate/ + total phosphorus (TP), anions (Cl−, Br−, NO3−, SO− 4 ), cations (Na , K+, Mg2+, Ca2+), and UV254 measurement were stored at 4 °C prior to analysis. Total organic carbon (TOC) and dissolved organic carbon (DOC) were analyzed using a TOC/TNb analyzer (multi N/C 2100, Analytik Jena, Germany) with catalytic high temperature combustion and IR-detector according to DIN EN 1484-H3. The total nitrogen bound (TNb) was analyzed with the same instrument according to DIN ENV 12260H34. The chemical oxygen demand (COD) was quantified with COD cell test tubes (LCK 314, Hach-Lange, Germany) in the range of 15– 150 mg/L O2 and subsequent photometric measurement (DR 3900 photometer, Hach-Lange, Germany) according to DIN ISO 15705 H45. Inorganic ions, nitrate, sulfate and chloride were determined using a 930 Compact IC flex ion chromatograph (Metrohm, Germany) using the method DIN EN ISO 10304-2-1-D20. UV absorbance at 254 nm (UV254) was measured on a Sekol 1300 spectrophotometer (Analytik Jena, Germany). The specific UV absorbance (SUVA) was calculated from the ratio of UV254 and DOC as an indicator for aromaticity of dissolved organic carbon. The CEC analysis was performed by direct injection, i.e. without preconcentration, on an LC–tandem MS system (QTrap 5500, Sciex, Darmstadt, Germany) according to the method described by Rühmland et al. (2015). Standards and isotope-labelled internal standards were obtained from the vendors listed in Table S1 in the supplementary data. The TPs of iopromide were isolated by semi-preparative HPLC as described in Schulz et al. (2008). Before analysis, an internal standard mix containing isotope-labelled surrogates for all analytes was spiked to each sample, resulting in a final concentration of 0.2 μg/L, except for the internal standards of X-ray contrast media and acesulfame, which had a final concentration of 2 μg/L and 4 μg/L, respectively. This was to account for the higher calibration curve ranges for these compounds. A sample volume of 80 μL was injected into an Agilent 1260 Series liquid chromatography system (Agilent Technologies, Waldbronn, Germany). The chromatographic separation was achieved using a Zorbax Eclipse Plus C-18 column (2.1 × 150 mm, 3.5 μm, Agilent Technologies). Ultrapure water and methanol (both supplemented with 0.1% formic acid) served as mobile phase A and B, respectively. All target compounds were measured within one chromatographic run by scheduled multiple reaction monitoring (sMRM) using electrospray ionization (ESI) in both negative and positive mode. Two mass transitions were measured for quantification and confirmation. The limit of quantification (LOQ) was derived from a signal to noise (S/N) ratio in native samples. At the LOQ the S/ N ratio of the mass transitions was at least 10 for quantification and at least 3 for confirmation. An external calibration was used for quantification. The accuracy and precision was checked by recovery experiments within each measurement series. The results were only considered valid if the recovery was in the range of 75–125%.

2.4. Statistical analysis To test for differences in concentrations between feed water, pore water and drainage effluent samples of DOC, UV254 and SUVA multiple comparisons were conducted using one-way ANOVA with Tukey's post hoc test. Data were tested for normal distribution and homogeneity of variances using Shapiro Wilk's and Levene's test, respectively.

3

Significance was accepted at p b 0.05. Statistical analyses were conducted using the software PAST (Hammer et al. 2001). 3. Results and discussion 3.1. Hydrological site characterization The discharge volume of the drainage calculated from regular flow measurements during the entire infiltration periods (approximately 4 days) accounted for approximately 214 m3 in September and 255 m3 in November representing about 49% and 52% of the total flow delivered to the infiltration basin, respectively. Monitoring of groundwater in well WT-97 (data not shown, see Fig. S1 for location) revealed oxic subsurface conditions and a stable groundwater level of 3.7 m below surface, which was not affected by infiltration (only measured during September experiments). A hydraulic retention time (HRT) of b22 h on average from the infiltration basin to the combined drainage collection outlet was determined during the November experiment using lithium bromide as a conservative tracer (2.5 mg/L bromide, see Fig. S2). Bromide concentration in the combined drainage outlet only accounted for 58% of infiltrated water. However, the bromide concentrations measured in the pore water collected underneath the infiltration basin coincided with the basin water concentration, indicating that no dilution occurred in the infiltration zone. Although adjacent infiltration basins did not receive feed water during the experiments, the drainage pipes can still be collecting older groundwater residing underneath those infiltration areas, which might have resulted in some dilution of the conservative tracer applied to recharge basin S62. Results from chloride measurements also indicate some dilution during the September experiment. Little differences between secondary effluent and drainage outlet concentrations did not enable to validate the degree of dilution for November (Fig. S3). The conductivity was probably not a conservative parameter due to cation exchange and nitrification processes in the subsurface and was therefore not adequate to assess dilution and breakthrough (Fig. S4). 3.2. Redox conditions As illustrated in Fig. 2, the dissolved oxygen (DO) concentration in 30 cm depth dropped from 6 mg/L to 1.3 mg/L within the first 7 h after start of infiltration and was completely depleted the next day. This indicates a fast oxygen consumption by biodegradation processes within the upper soil layer. Depletion of dissolved oxygen in the deeper layers was delayed until day 3, presumably caused by dissolution of oxygen in soil pores from the previous drying period of the basin. After day 3, the DO concentration at 30 cm increased again, as a result of a partly dry basin area, allowing the penetration of atmospheric oxygen into the pore system of the top soil layers. During the November experiment, NH₄+-N concentrations declined from 9 mg/L in the basin influent to below the detection limit (b0.2 mg/L) in the combined drainage outlet after SAT (Fig. 2a). NO₃−N concentrations in the drainage outlet rapidly increased to a maximum concentration of 15.2 mg/L after 8 h and slowly decreased subsequently to 7.5 mg/L after 74 h (Fig. 2b). Results from September showed similar behavior with complete removal of 6.4 mg/L NH₄+-N in the basin influent and maximum NO₃−-N concentration of 26.7 mg/L after 8 h in the combined drainage outlet. Subsequently, the concentration decreased continuously during the following 74 h to levels of 7.9 mg/L NO₃−-N (data not shown). The observed dynamics for nitrogen species are likely the result of combined sorption and nitrification processes during periodic wet and dry cycles. During infiltration, ammonia is adsorbing to negatively charged soil particles and is subsequently oxidized to nitrate in the presence of atmospheric oxygen penetrating into the porous media during the dry period. These effects resulting in high NO₃−-N concentration in the pore water and subsequently the drainage water at the beginning

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

4

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

Table 1 Summary of bulk organic parameters in both experiments (feed water, basin samples and drainage effluent are shown as average with standard deviation; average drainage concentration was calculated from samples taken after 22 h of flooding; average TOC from 5 and 2 samples in feed and drainage, respectively). Parameter

September 2015

pH-value DOC [mg/L] UV₂₅₄ [1/m] SUVA [L/m ∗ mg] TOC [mg/L] COD [mg/L]

Feed(n = 9) 7.4 ± 0.2 11.0 ± 0.7 23.6 ± 1.0 2.2 ± 0.2 15.7 ± 4.1 28.1 ± 2.3

November 2015 Effluent(n = 8) 6.6 ± 0.1 7.0 ± 0.5 20.6 ± 0.5 3.0 ± 0.2 7.4 ± 0.7 18.5 ± 1.1

Removal [%] – 36 13 – 47 34

of the next flooding event have been previously described for other SAT sites (Idelovitch et al., 2003; Miller et al. 2006). The gradual, timeshifted NH₄ + -N breakthrough (in comparison to the conservative tracer bromide) in the suction cup samplers at different depths (Fig. 2a) supports the hypothesis of ammonia sorption to soil. NH₄+-N was not detected in the drainage water indicating efficient ammonia adsorption in the subjacent soil layers throughout the experiment. Under dynamic redox conditions during infiltration, also denitrification might have occurred. However, the concentration of nitrite, which might be an indicator for incomplete denitrification, was below the detection limit of

Feed(n = 8) 7.7 ± 0.5 12.8 ± 0.7 28.0 ± 0.4 2.2 ± 0.1 15.6 ± 0.6 29.3 ± 4.9

Effluent(n = 7) 6.8 ± 0.1 6.3 ± 0.9 18.3 ± 0.3 2.9 ± 0.4 6.9 ± 0.2 b15.0

Removal [%] – 51 35 – 56 N46

0.1 mg/L in all samples. Reduction and mobilization of manganese and iron was not observed. The decrease of NH₄+-N in the infiltration basin can be attributed to either nitrification or sorption at the watersediment interface. 3.3. Removal of bulk organic carbon Results from DOC analyses in September and November are summarized in Fig. 3. DOC showed significant concentration differences between feed, pore water and drainage effluent samples (p b 0.05).

Table 2 Influent concentration (n = 3) and removal of CECs in the infiltration basin after 54 h and after soil passage in average (drainage; n = 7) from experiments conducted in September and November 2015. (n.d.: not detected; n.c.: not calculated, n.a.: not analyzed). Substance

LOQ [ng/L]

LOQ [ng/L]

Average co [ng/L]

Sept

Removal basin [%]

Removal drainage [%]

Nov

Sept

Nov

Sept

Nov

Sept

Nov

Substances showing efficient removal during infiltration (N90%) 10,11-DiH-10-OH-CBZb 10 2-OH-CBZc 10 3-OH-CBZc 10 Atenolol 20 Bezafibrate 20 Benzophenone-4 (BZP-4) 100 Climbazol n.a. Codeine 20 Diclofenac (DCF) 20 Iopromide 50 Metoprolol 20 Sotalol 20 Trimethoprim 10

5 10 1 10 2 20 5 10 5 50 5 5 10

640 ± 20 190 ± 10 230 ± 20 200 ± 20 370 ± 30 2670 ± 70 n.a. 120 ± 10 3030 ± 90 9390 ± 1650 N5000 410 ± 8 160 ± 7

620 ± 40 210 ± 4 160 ± 4 210 ± 10 530 ± 3 2610 ± 20 90 ± 4 120 ± 10 3940 ± 150 4820 ± 1180 1730 ± 140 490 ± 9 220 ± 10

b10 32 34 29 24 13 n.a. 43 64 b10 b10 19 50

b10 32 31 16 b10 b10 19 18 19 b10 12 b10 11

n.c.d N95d N96d N90d N95d 89 n.a. N83d 96 96 N94 90 N94d

99 N95d 99 N95d 97 93 N95d N92d 96 97 98 92 N95d

Substances showing good removal during infiltration (70%–90%) 10,11-DiH-10,11-DiOH-CBZc 10 Sum of 4- and 5-methyl-benzotriazol 100 Acyclovir 50 Iomeprol 100 Mecoprop 20 Oxazepam 20 TCPPa 50 Venlafaxine 20

5 10 5 50 5 5 100 2

2140 ± 50 4650 ± 130 150 ± 20 480 ± 40 100 ± 6 130 ± 7 1360 ± 30 410 ± 10

2200 ± 30 3850 ± 100 210 ± 30 n.d. 30 ± 2 130 ± 6 910 ± 100 380 ± 10

b10 22 38 b10 20 19 b10 36

b10 b10 20 n.d. b10 b10 b10 b10

77 73 N66d N79d N81d N84d 85 86

76 70 N98d n.d. N83e 89 83 90

Substances showing moderate removal during infiltration (30%–70%) Benzotriazol 100 10 Iopamidol 50 20 Sulfamethoxazole (SMX) 20 5 Tramadol 20 2 a TBP 20 50 TCEPa 20 20

N5000 16,090 ± 3800 380 ± 60 670 ± 20 190 ± 30 120 ± 10

7900 ± 380 15,030 ± 4570 230 ± 30 780 ± 10 150 ± 10 120 ± 10

b10 b10 20 10 23 b10

b10 b10 b10 b10 33 22

n.c. n.c. 67 50 43 n.c.

n.c. 50 n.c. 63 N63e n.c.

Persistent substances (removal b 30%) Acesulfame Carbamazepine (CBZ) Diatrizoate Fluconazol Primidone

5110 ± 440 830 ± 8 6520 ± 780 130 ± 2 240 ± 20

5410 ± 3410 1110 ± 10 6170 ± 690 150 ± 5 270 ± 9

13 b10 b10 b10 b10

b10 b10 b10 b10 b10

n.c. b10 b10 b10 n.c.

n.c. 21 n.c. n.c. 13

a b c d e

100 10 50 10 50

1000 0 20 2 5

Organophosphate flame retardants, TCPP: Tris-(2-chloroisopropyl)phosphate; TBP: Tributylphosphate; TCEP: Tris-(2-chloroethyl)phosphate. Metabolite of oxcarbamazepine. Human metabolites of CBZ (Kaiser et al. 2014). Removal below LOQ in all drainage samples, effluent concentration set as LOQ for calculation. Several samples below LOQ, data shown as minimum removal.

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

Stable DOC concentrations were observed in the infiltration basin and in drainage samples after breakthrough (N 22 h). The initial DOC concentration of 12.8 mg/L in November was decreased by about 30% during percolation through the subsurface according to pore water samples taken at 30 cm (9.7 mg/L DOC), 60 cm (9.9 mg/L DOC), and 90 cm (9.1 mg/L DOC) depth. According to the conservative tracer test (Fig. S2), infiltrated water in suction cup samples was not diluted by native groundwater and thus the significant DOC removal of 30% can be attributed to biodegradation and adsorption occurring in the infiltration zone. Rapid biodegradation is confirmed by fast oxygen depletion within the first 30 cm of infiltration. The further decrease of DOC concentration by 21% observed in the drainage outlet might also partly be a consequence of dilution with local background water carrying lower DOC concentrations. Consistent with the results from DOC measurements, a statistically significant removal (p b 0.05) of UV absorbance at 254 nm (UV254) occurred during percolation through the first 30 cm, whereas no further decrease was observed at 90 cm depth (Fig. S5). SUVA remained constant within the first 90 cm of infiltration, but significantly increased (p b 0.05) in the drainage for both sampling periods (Sept: 36%, Nov: 32%, Fig. S6), indicating a preferred removal of aliphatic (non-humic) compounds compared to aromatic (humic) structures within the bulk of organic matter (Drewes et al. 2006). However, also dilution with background water having a higher degree of aromaticity might be an explanation for an increase in SUVA. Measured bulk organic matter parameters are summarized in Table 1. The results obtained for the COD removal are in line with those from DOC (see also Fig. S7). Despite winter season, the relative removal of organic matter during short-term SAT was more efficient in November. This might be explained by sporadically decreased efficiency in secondary treatment during November experiments, which resulted in significantly higher influent concentration in the basin. However, the residual DOC concentration after SAT only differed by 0.7 mg/L. 3.4. Removal of contaminants of emerging concern (CECs) In secondary effluent, 32 compounds were detected by LC–tandem MS analysis, including pharmaceuticals, industrial chemicals and some metabolites. The average secondary effluent concentrations of the analyzed CECs as well as their removal in the infiltration basin and during SAT are summarized in Table 2. Average removal at the drainage outlet was calculated from samples taken after breakthrough of the conservative tracer (N 22 h) only for compounds with consistent drainage effluent concentration (variance b10%). The 32 compounds were classified into 13 compounds showing efficient removal above 90%, 8 compounds

Fig. 1. Change of dissolved oxygen concentrations in three different depths below the infiltration basin during the flooding period in November. Data points represent single measurements at each time point (n = 1). Mark on the y-axis indicates average basin concentration: 9.5 ± 3 mg/L.

5

Fig. 2. Trend in ammonia (a) and nitrate (b) nitrogen concentrations in influent, basin, drainage and pore water (depths: 30 cm, 60 cm, 90 cm) samples. Data points represent single measurements at each time point (n = 1).

which are well removed up to 70%, 6 moderately removable CECs (with 30%–70% removal), and 5 persistent substances (b 30% removal). The percentage removal for samples taken in September and November were similar for the respective compounds. Calculated removal efficiencies determined in the combined drainage outlet do not account for dilution with local background water in the drainage. Breakthrough curves of the persistent substances carbamazepine, diatrizoate, fluconazole, and primidone from experiments in September and November are shown in Fig. 4 in comparison to the breakthrough of the conservative tracer bromide determined in

Fig. 3. DOC concentrations in feed water (influent and basin) and drainage effluent in September (left) and November (right) and in pore water samples in November (right). Boxplots show median, 25th and 75th percentile (box) and 5th to 95th percentile ranges (whiskers). Different letters indicate statistically significant differences (Tukey's pairwise comparison, p b 0.05).

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

6

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

Fig. 4. Breakthrough curves of three persistent compounds from experiments in a) September (influent concentrations: Carbamazepine c₀ = 834 ng/L; Primidone c₀ = 238 ng/L; Diatrizoate c₀ = 6517 ng/L; Fluconazol c₀ = 133 ng/L) and b) November (influent concentrations: Carbamazepine c₀ = 1114 ng/L; Primidone c0 = 271 ng/L; Diatrizoate c₀ = 6170 ng/L; Fluconazol c₀ = 151 ng/L). Breakthrough curve of the tracer bromide is also shown for the experiment conducted in November.

November. The persistent character of these substances under the described conditions was confirmed before in several studies (Ternes et al. 2007; Scheurer et al. 2009; Hoppe-Jones et al. 2010; Laws et al. 2011; Regnery et al. 2015b). The measured persistent CECs show a faster breakthrough than bromide, partly exceeding concentration of infiltrated water (diatrizoate in both experiments, carbamazepine and fluconazole in September). Based on these findings it is assumed that these CECs were already present in the uppermost groundwater from previous infiltration events. After breakthrough of infiltrated water, the relative residual concentration of persistent CECs is a function of dilution ratio as well as concentrations in infiltrated and background water. Different concentrations of CECs in the subsurface might be the reason for different breakthrough concentrations observed for primidone in comparison to other persistent substances. The presence of the legacy contamination of diatrizoate was reported before in infiltration systems of the Braunschweig recharge facility (Ternes et al. 2007). Although the artificial sweetener acesulfame had been suggested as a viable tracer for anthropogenic influences due to its high persistence (Buerge et al. 2009; Scheurer et al. 2009), several researchers observed significant removal during short-term SAT and even in biologically active filters (Zucker et al. 2015). In this study, the highest concentrations of acesulfame in the drainage outlet were observed at the beginning (after 2 h) and towards the end of each experiment (Fig. S8). Final concentration in the drainage outlet reached approximately 100% and 80%

of the average influent concentration during the September and November recharge events, respectively. After the initial peak, the concentration decreased to 17% of the influent concentration and slowly increased again. The reason for decreasing removal efficiency at the drainage outlet is not clear, but might be a result of a steadily decreasing redox potential, since acesulfame biodegradation is favored under oxic conditions (Regnery et al. 2015a). Fig. 5 illustrates the removal of selected compounds in the infiltration basin during the sampling periods in September and November 2015. Removal of most compounds including diclofenac, trimethoprim and atenolol was more efficient in September than in November (see also Table 2), which is likely due to the contribution of photolysis in particular for longer sunshine hours in September. Diclofenac degradation by direct photolysis in surface waters is well documented (Buser et al. 1998). Direct and indirect photolysis have also been reported for trimethoprim and atenolol (Ji et al. 2012; Luo et al. 2012). Indirect photolysis is initiated by light absorption of photosensitizers, such as nitrate or natural organic matter, which can generate reactive oxygen species and radicals (Ji et al. 2012; Dong et al. 2015). However, stable concentrations of non-photosensitive primidone demonstrate that contribution of indirect photolysis is not significant. On the other hand, also biodegradation and sorption to sediment might play a role for the decreasing concentration of trimethoprim and atenolol in the infiltration basin. Compound removal during SAT was monitored in the drainage effluent (Table 2) and in suction cup samples at different depths.

Fig. 5. Concentrations and removal in infiltration basin from experiments in a) September (Primidone c₀ = 238 ng/L; Atenolol c₀ = 201 ng/L; Trimethoprim c₀ = 154 ng/L and Diclofenac c₀ = 3032 ng/L) and b) November (Primidone c₀ = 271 ng/L; Atenolol c₀ = 207 ng/L; Trimethoprim c₀ = 215 ng/L and Diclofenac c₀ = 3941 ng/L).

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

7

Fig. 6. Attenuation patterns of selected CECs during soil passage from experiment 2. Samples were taken in 30 cm, 60 cm and 90 cm depth. Mark on the y-axis indicates average influent concentration (n = 3).

Biodegradation and sorption are the two main processes contributing to CEC attenuation in SAT systems (Maeng et al. 2011). Time resolved analyses from suction cup samples are depicted in Fig. 7 to show CECs representing different breakthrough behavior. Significant retardation was observed for atenolol breakthrough at different depths (Fig. 6a), which is not correlated with the availability of dissolved oxygen (see Fig. 1). Atenolol is a basic compound with a pKa value of 9.16 and the molecule is positively charged at neutral pH (Ternes et al. 2007; Yamamoto et al. 2009). The electrostatic interaction with the negatively charged surface layer and cation exchange mechanism in the subsurface are likely responsible for the retardation of atenolol. However, additional biotransformation cannot be ruled out as atenolol is known to be well biodegradable. Similar breakthrough curves have been observed for other betablockers, which are positively charged at neutral pH (i.e., sotalol and metoprolol) as well as for oxazepame and tramadol. As illustrated in Fig. 6b, a faster time-shifted breakthrough at the different depths was observed for diclofenac. As most of the non-steroidal anti-inflammatory drugs, diclofenac is a negatively charged acidic compound at neutral pH and sorption to negatively charged surfaces is unlikely (Maeng et al. 2011). The concentrations in suction cup samplers

remained below those of the secondary effluent and decreased after the initial breakthrough. This can be attributed to previous photolysis in the infiltration basin. Breakthrough behavior similar to diclofenac was also observed for other substances such as acyclovir (Fig. 6c) and bezafibrate and iopromide (data not shown). After 30 h, steady concentrations of these three CECs significantly below the influent concentrations were observed after SAT, likely caused by microbial degradation. An efficient removal of acyclovir (97%) has previously been reported by biological transformation during wastewater treatment (Prasse et al. 2010). Observed removal in suction cup samplers was in line with results from drainage effluent for many CECs. However, the compounds diclofenac, bezafibrate, iopromide, mecoprop and TCPP revealed removal of b 35% after 90 cm infiltration but they were well removed in the combined drainage outlet (N80%). Since several of these substances are better biodegradable under oxic conditions (Regnery et al. 2015b), it is hypothesized that an increased removal is caused by establishment of oxic conditions during later phases of infiltration. Re-aeration might appear through unsaturated zones of the infiltration basin or through the 150 m drainage pipes. However, more appropriate in situ analysis of redox conditions would be needed to confirm these assumptions.

Fig. 7. Proposed transformation pathway of iopromide during biological treatment under oxic conditions according to Schulz et al. (2008).

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

8

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

Fig. 8. Molar ratio of iopromide and its TPs for experiments in September (left) and November (right).

3.5. Fate and transport of CEC transformation products (TPs) Little is known about the environmental fate and potential toxicological effects of TPs of environmental contaminants, which are generated during biological and chemical transformations during wastewater treatment. Thus, the expression “removal by biodegradation” represents in most cases transformation to unknown transformation products. Due to their presence at very low concentrations in the environment, identification and detection in environmental matrices are challenging and only limited number of TPs have been identified so far. In this study, redox-dependent formation and fate of 5 major TPs of the triiodinated X-ray contrast medium iopromide is reported for the first time for short-term SAT based on the degradation pathways under oxic conditions reported by Schulz et al. (2008) (Fig. 7). Considering these pathways, analyzed compounds include one phase I product (TP 819), which is formed by oxidation of two terminal alcohols to carboxylic acids, and four phase III products formed by further oxidation and decarboxylation/deacetylation of TP 819. Fig. 8 summarizes results of average molar concentrations of iopromide and its TPs at different sampling locations during experiments in September and November. The cumulative influent concentration of iopromide and its TPs was similar in September and November experiments but the concentration of iopromide was higher in September, indicating varying transformation efficiencies during the preceding wastewater treatment. Analyses of combined drainage outlet samples during both experiments show almost complete transformation of iopromide and the phase I product TP-819 as well as an accumulation of phase III products. The total amount of iopromide and its TPs at the drainage outlet exceeds the accumulated molar concentration of the substance in the secondary effluent samples. This is most likely caused by phase I and II transformation products, which were present in secondary effluent but not analyzed in this study. While a rapid removal of iopromide under oxic subsurface conditions was reported before (Grünheid et al. 2005; Laws et al. 2011), complete transformation of all intermediates to phase III products was only known for longer travel times (Kormos et al. 2011) and is shown here for the first time to also occur during short-term SAT. In contrast to efficient transformation in the drainage outlet, results from suction cup samplers within the first 90 cm of infiltration only show minor changes of iopromide and its TPs. While concentrations of iopromide and TP-759 decreased by 50%, formation of phase III products TP-701-A and TP-643 slowly increased with depth. As the transformation pathway involves monooxygenases (e.g., phase I reactions), transformation efficiency during initial infiltration might be limited by the availability of dissolved oxygen. Advanced transformation and formation of phase III products during later phases of SAT was in line with results from several redox-sensitive CECs. The accumulation of phase III TPs in the drainage outlet confirms biodegradation as major removal

mechanism for parent compounds and further strengthens the assumption that dissolved oxygen is introduced into the system before sample collection in the combined drainage outlet. These results show that the analysis of iopromide along with its intermediates and persistent TPs can serve as a promising tool to determine overall efficiency of CECs biodegradation, to identify in situ oxygen limitations, and to distinguish microbial degradation (formation of TPs) from abiotic attenuation processes like sorption and dilution. This information is of special interest to systems, where oxygen-free sampling is impossible. 4. Conclusions In this study, two wet-dry cycles of a full-scale infiltration basin were monitored to characterize hydraulic retention times, redox conditions, removal of bulk organic parameters and the fate of chemicals of emerging concern during short-term SAT. This study utilized for the first time known products from oxic transformation of the X-ray contrast medium iopromide as intrinsic tracer for in situ characterization of the transition zone between oxic and suboxic conditions during the initial phase of SAT. The occurrence of these transition products confirmed that the transformation efficiency of redox-sensitive CECs during the initial phase of infiltration might be limited by the availability of dissolved oxygen. Further transformation into persistent products detected in the combined drainage outlet indicates that dissolved oxygen had been introduced to the system before sample collection. These results are inline with removal of redox-sensitive CECs diclofenac, bezafibrate, mecoprop and TCPP revealing removal of b35% after 90 cm infiltration but N 80% in the combined drainage outlet. They suggest that the analysis of iopromide along with its intermediates and persistent TPs can serve as a promising tool to identify in situ oxygen limitations and therefore better reveal opportunities to improve the overall efficiency of CECs biodegradation during SAT. Acknowledgments This work was performed within the research project FRAME with funding from JPI Water and the German Federal Ministry of Education and Research (BMBF) (02WU1345A and 02WU1345B). We gratefully thank Bernhard Teiser from Abwasserverband Braunschweig for his support as well as Jörg Walther and Birgit Fiebig from Stadtentwässerung Braunschweig GmbH for their tremendous help in conducting the field experiments. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2016.12.165.

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165

M. Muntau et al. / Science of the Total Environment xxx (2017) xxx–xxx

References Alidina, M., Li, D., Ouf, M., Drewes, J.E., 2014. Role of primary substrate composition and concentration on attenuation of trace organic chemicals in managed aquifer recharge systems. J. Environ. Manag. 144, 58–66. Amy, G., Drewes, J., 2007. Soil aquifer treatment (SAT) as a natural and sustainable wastewater reclamation/reuse technology: fate of wastewater effluent organic matter (EfOM) and trace organic compounds. Environ. Monit. Assess. 129 (1–3), 19–26. Buerge, I.J., Buser, H.R., Kahle, M., Muller, M.D., Poiger, T., 2009. Ubiquitous occurrence of the artificial sweetener acesulfame in the aquatic environment: an ideal chemical marker of domestic wastewater in groundwater. Environ. Sci. Technol. 43 (12), 4381–4385. Buser, H.R., Poiger, T., Müller, M.D., 1998. Occurrence and fate of the pharmaceutical drug diclofenac in surface waters: rapid photodegradation in a lake. Environ. Sci. Technol. 32 (22), 3449–3456. Celiz, M.D., Tso, J., Aga, D.S., 2009. Pharmaceutical metabolites in the environment: analytical challenges and ecological risks. Environ. Toxicol. Chem. 28 (12), 2473–2484. de Jongh, C.M., Kooij, P.J., de Voogt, P., ter Laak, T.L., 2012. Screening and human health risk assessment of pharmaceuticals and their transformation products in Dutch surface waters and drinking water. Sci. Total Environ. 427, 70–77. Dong, M.M., Trenholm, R., Rosario-Ortiz, F.L., 2015. Photochemical degradation of atenolol, carbamazepine, meprobamate, phenytoin and primidone in wastewater effluents. J. Hazard. Mater. 282, 216–223. Drewes, J.E., Quanrud, D.M., Amy, G.L., Westerhoff, P.K., 2006. Character of organic matter in soil-aquifer treatment systems. J. Environ. Eng. 132 (11), 1447–1458. Du, B., Price, A.E., Scott, W.C., Kristofco, L.A., Ramirez, A.J., Chambliss, C.K., Yelderman, J.C., Brooks, B.W., 2014. Comparison of contaminants of emerging concern removal, discharge, and water quality hazards among centralized and on-site wastewater treatment system effluents receiving common wastewater influent. Sci. Total Environ. 466, 976–984. Gerba, C.P., Powelson, D.K., Yahya, M.T., Wilson, L.G., Amy, G.L., 1991. Fate of viruses during soil aquifer treatment, designed for wastewater reclamation and reuse. Water Sci. Technol. 24, 95–102. Grünheid, S., Amy, G., Jekel, M., 2005. Removal of bulk dissolved organic carbon (DOC) and trace organic compounds by bank filtration and artificial recharge. Water Res. 39 (14), 3219–3228. Grünheid, S., Hübner, U., Jekel, M., 2008. Impact of temperature on biodegradation of bulk and trace organics during soil passage in an indirect reuse system. Water Sci. Technol. 57 (7), 987–994. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST-PAlaeontological STatistics, ver. 1.89. Palaeontol. Electron. 4 (1), 1–9. Hoppe-Jones, C., Oldham, G., Drewes, J.E., 2010. Attenuation of total organic carbon and unregulated trace organic chemicals in US riverbank filtration systems. Water Res. 44 (15), 4643–4659. Idelovitch, E., Icekson-Tal, N., Avraham, O., Michail, M., 2003. The long-term performance of soil aquifer treatment (SAT) for effluent reuse. Water Sci. Technol. Water Supply 3 (4), 239–246. Ji, Y., Zeng, C., Ferronato, C., Chovelon, J.M., Yang, X., 2012. Nitrate-induced photodegradation of atenolol in aqueous solution: kinetics, toxicity and degradation pathways. Chemosphere 88 (5), 644–649. Jones, O.A.H., Voulvoulis, N., Lester, J.N., 2004. Potential ecological and human health risks associated with the presence of pharmaceutically active compounds in the aquatic environment. Crit. Rev. Toxicol. 34 (4), 335–350. Kaiser, E., Prasse, C., Wagner, M., Bröder, K., Ternes, T.A., 2014. Transformation of oxcarbazepine and human metabolites of carbamazepine and oxcarbazepine in wastewater treatment and sand filters. Environ. Sci. Technol. 48 (17), 10208–10216. Kanarek, A., Michail, M., 1996. Groundwater recharge with municipal effluent: Dan region reclamation project, Israel. Water Sci. Technol. 34 (11), 227–233. Kormos, J.L., Schulz, M., Ternes, T.A., 2011. Occurrence of iodinated X-ray contrast media and their biotransformation products in the urban water cycle. Environ. Sci. Technol. 45, 8723–8732.

9

La Farre, M., Pérez, S., Kantiani, L., Barceló, D., 2008. Fate and toxicity of emerging pollutants, their metabolites and transformation products in the aquatic environment. Trends Anal. Chem. 27 (11), 991–1007. Laws, B.V., Dickenson, E.R., Johnson, T.A., Snyder, S.A., Drewes, J.E., 2011. Attenuation of contaminants of emerging concern during surface-spreading aquifer recharge. Sci. Total Environ. 409 (6), 1087–1094. Li, D., Alidina, M., Drewes, J.E., 2014. Role of primary substrate composition on microbial community structure and function and trace organic chemical attenuation in managed aquifer recharge systems. Appl. Microbiol. Biotechnol. 98 (10), 4347–4353. Luo, X., Zheng, Z., Greaves, J., Cooper, W.J., Song, W., 2012. Trimethoprim: kinetic and mechanistic considerations in photochemical environmental fate and AOP treatment. Water Res. 46 (4), 1327–1336. Maeng, S.K., Sharma, S.K., Lekkerkerker-Teunissen, K., Amy, G.L., 2011. Occurrence and fate of bulk organic matter and pharmaceutically active compounds in managed aquifer recharge: a review. Water Res. 45 (10), 3015–3033. Miller, J.H., Ela, W.P., Lansey, K.E., Chipello, P.L., Arnold, R.G., 2006. Nitrogen transformations during soil-aquifer treatment of wastewater effluent - oxygen effects in field studies. J. Environ. Eng. 132 (10), 1298–1306. Pal, A., Gin, K.Y.H., Lin, A.Y.C., Reinhard, M., 2010. Impacts of emerging organic contaminants on freshwater resources: review of recent occurrences, sources, fate and effects. Sci. Total Environ. 408 (24), 6062–6069. Prasse, C., Schlüsener, M.P., Schulz, R., Ternes, T.A., 2010. Antiviral drugs in wastewater and surface waters: a new pharmaceutical class of environmental relevance? Environ. Sci. Technol. 44 (5), 1728–1735. Quanrud, D.M., Hafer, J., Karpiscak, M.M., Zhang, J., Lansey, K.E., Arnold, R.G., 2003. Fate of organics during soil-aquifer treatment: sustainability of removals in the field. Water Res. 37 (14), 3401–3411. Regnery, J., Lee, J., Kitanidis, P., Illangasekare, T., Sharp, J.O., Drewes, J.E., 2013. Integration of managed aquifer recharge for impaired water sources in urban settings – overcoming current limitations and engineering challenges. Environ. Eng. Sci. 30 (8), 409–420. Regnery, J., Wing, A.D., Alidina, M., Drewes, J.E., 2015a. Biotransformation of trace organic chemicals during groundwater recharge: how useful are first-order rate constants? J. Contam. Hydrol. 179, 65–75. Regnery, J., Barringer, J., Wing, A.D., Hoppe-Jones, C., Teerlink, J., Drewes, J.E., 2015b. Startup performance of a full-scale riverbank filtration site regarding removal of DOC, nutrients, and trace organic chemicals. Chemosphere 127, 136–142. Rühmland, S., Wick, A., Ternes, T.A., Barjenbruch, M., 2015. Fate of pharmaceuticals in a subsurface flow constructed wetland and two ponds. Ecol. Eng. 80, 125–139. Scheurer, M., Brauch, H.J., Lange, F.T., 2009. Analysis and occurrence of seven artificial sweeteners in German wastewater and surface water and in soil aquifer treatment (SAT). Anal. Bioanal. Chem. 394 (6), 1585–1594. Schulz, M., Löffler, D., Wagner, M., Ternes, T.A., 2008. Transformation of the X-ray contrast medium iopromide in soil and biological wastewater treatment. Environ. Sci. Technol. 42 (19), 7207–7217. Schwarzenbach, R.P., Escher, B.I., Fenner, K., Hofstetter, T.B., Johnson, C.A., Von Gunten, U., Wehrli, B., 2006. The challenge of micropollutants in aquatic systems. Science 313 (5790), 1072–1077. Snyder, S.A., Westerhoff, P., Yoon, Y., Sedlak, D.L., 2003. Pharmaceuticals, personal care products, and endocrine disruptors in water: implications for the water industry. Environ. Eng. Sci. 20 (5), 449–469. Ternes, T.A., Bonerz, 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 (5), 894–904. Yamamoto, H., Nakamura, Y., Moriguchi, S., Nakamura, Y., Honda, Y., et al., Sekizawa, J., 2009. Persistence and partitioning of eight selected pharmaceuticals in the aquatic environment: laboratory photolysis, biodegradation, and sorption experiments. Water Res. 43 (2), 351–362. Zucker, I., Mamane, H., Cikurel, H., Jekel, M., Hübner, U., Avisar, D., 2015. A hybrid process of biofiltration of secondary effluent followed by ozonation and short soil aquifer treatment for water reuse. Water Res. 84, 315–322.

Please cite this article as: Muntau, M., et al., Evaluation of the short-term fate and transport of chemicals of emerging concern during soil-aquifer treatment using select transf..., Sci Total Environ (2017), http://dx.doi.org/10.1016/j.scitotenv.2016.12.165