Environ Sci Pollut Res

DOI 10.1007/s11356-017-0216-7

RESEARCH ARTICLE

Presence of active pharmaceutical ingredients in the continuum

of surface and ground water used in drinking water production
Heidi Ahkola 1 & Sirkku Tuominen 1 & Sanja Karlsson 1 & Noora Perkola 1 & Timo Huttula 1 &
Sami Saraperä 2 & Aki Artimo 2 & Taina Korpiharju 3 & Lauri Äystö 1 & Päivi Fjäder 1 &
Timo Assmuth 1 & Kirsi Rosendahl 1 & Taina Nysten 1

Received: 29 May 2017 / Accepted: 13 September 2017

# Springer-Verlag GmbH Germany 2017

Abstract Anthropogenic chemicals in surface water and suitability of the Chemcatcher passive samplers for detecting
groundwater cause concern especially when the water is used APIs in groundwater wells.
in drinking water production. Due to their continuous release
or spill-over at waste water treatment plants, active pharma- Keywords Passive sampling . Chemcatcher . Active
ceutical ingredients (APIs) are constantly present in aquatic pharmaceutical ingredients . Groundwater . Artificial
environment and despite their low concentrations, APIs can groundwater . Drinking water supply
still cause effects on the organisms. In the present study,
Chemcatcher passive sampling was applied in surface water,
surface water intake site, and groundwater observation wells Introduction
to estimate whether the selected APIs are able to end up in
drinking water supply through an artificial groundwater re- In Finland, more than 65% of municipal water intake is
charge system. The API concentrations measured in conven- groundwater or artificial groundwater and at sparsely populat-
tional wastewater, surface water, and groundwater grab sam- ed areas, groundwater wells are important sources of house-
ples were assessed with the results obtained with passive sam- hold water (Vienonen et al. 2012). Because of the uneven
plers. Out of the 25 APIs studied with passive sampling, four distribution of groundwater resources and population, the part
were observed in groundwater and 21 in surface water. This of artificial groundwater as water supply is estimated to in-
suggests that many anthropogenic APIs released to waste wa- crease. When surface water or groundwater is used for drink-
ter proceed downstream and can be detectable in groundwater ing water production, anthropogenic chemicals present in the
recharge. Chemcatcher passive samplers have previously been raw water can cause concerns (Focazio et al. 2008, Loos et al.
used in monitoring several harmful chemicals in surface and 2010). Therefore, knowledge on the sources and concentra-
wastewaters, but the path of chemicals to groundwater has not tions of the anthropogenic chemicals is important to maintain
been studied. This study provides novel information on the sustainable watershed management (Assmuth et al. 2016).
It is customary to measure APIs in groundwater using grab
sampling technique (e.g., Sui et al. 2015). The conventional
grab sample is taken after purging the stagnant water from the
Responsible editor: Philippe Garrigues observation well (Fetter 2001). This ensures that the sampled
water represents the conditions prevailing in the groundwater
* Heidi Ahkola area. However, a grab sample gives only an instant concentra-
heidi.ahkola@ymparisto.fi
tion of studied compounds at the moment of the sampling. So
far, the presence of APIs in Finnish groundwater has been
1
Finnish Environment Institute (SYKE), P.O.Box 140, studied with grab sampling in a pan-European study published
00251 Helsinki, Finland by Loos et al. (2010). In their study, the most frequently ob-
2
Turku Region Water Ltd., Maariankatu 1, 20100 Turku, Finland served APIs were carbamazepine (CBZ), sulfamethoxazole
3
The Water Protection Association of the River Kokemäenjoki (SMX), ketoprofen (KET), and ibuprofen (IBU) which were
(KVVY), P.O.Box 265, 33101 Tampere, Finland found in 42, 24, 10.4, and 6.7% of the samples, respectively.
 Environ Sci Pollut Res

In contrast to snapshot concentrations achieved through Materials and methods

grab sampling, by using passive samplers, the concentrations
can be studied for longer time periods of days to weeks. This Materials and chemicals
allows the determination of time-weighted average (TWA) con-
centrations (Kingston et al. 2000, Vrana et al. 2005a). Certain The purities of the solvents were at least HPLC-grade or equiv-
integrative passive samplers have been used for studying con- alent. The water used in sample handling and laboratory cali-
taminants in groundwater. Microporous, chemically inert bration trial was ultrahigh quality (UHQ) and had an internal
GORE Module passive samplers (AGI sampler at present, resistance ≥ 18 MΩ cm−1 at 25 °C. Native and mass-labeled
Parker et al. 2014) have been deployed in monitoring volatile pharmaceutical standards were purchased from Sigma-Aldrich
and semi-volatile organic compounds, water-soluble com- Corporation, LGC Standards, and Cambridge Isotope
pounds (such as 1,4 dioxane), and polyaromatic hydrocarbons Laboratories Inc. as neat crystals or ready solutions. The pu-
(PAHs) (ITRC 2007). The ceramic dosimeter passive sampler rities were ≥ 98%, except for the purity of simvastatin (SIM)
contains ceramic membrane as a sorbent, and it has been de- standard which was ≥ 97% and the purity of enalapril (ENA)
ployed to detect PAHs in groundwater (Bopp et al. 2005). Also, which was not assigned. Stock solutions of the neat standards
the SorbiCell passive sampler has been deployed in groundwa- (1 mg mL−1) were prepared in methanol and stored in ≤ 4 °C.
ter to determine nutrients and metals (Rozemeijer et al. 2010, In this study, we used a Chemcatcher passive sampler with a
Birch et al. 2011). To avoid the disturbance of the well hydrol- sulfonated styrenedivinylbenzene copolymer (SDB-RPS)
ogy, the stagnant water is not pumped before the deployment. Empore disk (47 mm diameter, 3 M Agilent Technologies
The Chemcatcher passive sampling technique has been Finland Oy) as a receiving phase (Vermeirssen et al. 2009,
used in monitoring of pollutants including APIs in waste wa- Lindholm-Lehto et al. 2016 a,b). The polycarbonate sampler
ters and surface waters (Kingston et al. 2000; Persson et al. housings that were used in surface water sampling were pur-
2001, Blom et al. 2002, Vrana et al. 2005b, Allan et al. 2007, chased from MP-Plast Inc. in Muurame, Finland. The glass
de la Cal et al. 2008, Gunold et al. 2008, Sánchez-Bayo et al. bottles (diameter 2.5 cm, height 7.5 cm), which were used in
2013, Vermeirssen et al. 2013, Ahkola et al. 2014, 2015, 2016, groundwater passive sampling, were purchased from VWR
Lindholm-Lehto et al. 2016a). Also, a polar organic chemical International (Helsinki, Finland). Grab samples and 24-h com-
integrative passive sampler (POCIS) with Oasis HLB sorbent posite waste water samples were collected in 1-l polypropylene
has been used for studying APIs in hospital sewage waters or HDPE bottles that were rinsed with HPLC-grade methanol.
(Bailly et al. 2013) as well as surface waters (MacLeod et al.
2007, Mills et al. 2007, Turja et al. 2015). The diameters of the Studied APIs
conventional Chemcatcher or POCIS samplers are over
70 mm. Thus, the samplers have to be modified for ground- We measured the occurrence of 25 APIs belonging to several
water sampling as the diameters of the observation wells are different pharmaceutical groups and used for a variety of pur-
typically small, e.g., 52 mm, in this study. poses, including, e.g., antibiotics, anti-inflammatories, psychi-
This study is part of the CONPAT project within the AKVA atric drugs, cardiovascular drugs, and an antifungal agent. The
program of the Academy of Finland, which focused on aquatic compounds and their average annual consumptions in Finland
contaminants, their pathways, health risks, and management are presented in Table 1. CBZ was developed to treat epilepsy,
specifically in a case of potable water production from river and nowadays, it is also used in suppressing neurogenic pain
water in a managed aquifer recharge system (CONPAT 2014). (Mcquay et al. 1995). Fluconazole (FLU) is an antifungal
In this study, the presence of APIs was determined in sur- medicine that is commonly used to suppress yeast infection.
face water, surface water intake site, and groundwater using Amlodipine (AML), bisoprolol (BIS), enalapril (ENA),
passive samplers. We screened several pharmaceuticals but losartan (LOS), metoprolol (MET), ramipril (RAM), and
focused more precisely on CBZ, SMX, IBU, and KET as telmisartan (TEL) are mainly used in treating hypertension
example compounds since they are the most commonly dis- but also coronary artery disease, cardiac ischemia, and diabet-
covered APIs in groundwater (Barnes et al., 2008; Loos et al. ic nephropathy among others. SMX has been widely used in
2010; Sui et al. 2015). Sampling rates (RS) for passive sam- combination with trimethoprim (TMP) to treat upper respira-
plers were determined for SMX, CBZ, IBU, and KET which tory tract infections. The combination is also used for veteri-
enabled the calculation of TWA concentration during the de- nary purposes, and traces of SMX and TMP have been found,
ployment. The TWA concentrations were compared to the e.g., in marine sediments in fish farming locations (Muziasari
concentrations obtained from surface and groundwater grab et al. 2014). Atorvastatin (ATO) and simvastatin (SIM) are
samples. In this study, the suitability of Chemcatcher passive used for decreasing the production of cholesterol. Quetiapine
sampling in monitoring APIs from groundwater is assessed. (QTP) and risperidone (RIS) are mainly used for treating
Likewise, the estimation if the APIs can end up from pharma- schizophrenia and bipolar disorder. Warfarin (WAR) was orig-
cy to the drinking water production will be conducted. inally introduced as a rodenticide, but nowadays, WAR and
Environ Sci Pollut Res

Table 1 Studied APIs (Fimea

2014) Pharmaceutical Abbreviation Type and use Average amount used in
Finland during
2012–2014 (kg a−1)

Amlodipine AML Cardiovascular 462

Atorvastatin ATO Lipid regulator 1203
Bisoprolol BIS Cardiovascular 719
Carbamazepine CBZ Antiepileptic 3442
Citalopram + escitalopram CIT + ESC Antidepressant 833
Dipyridamole DIP Antithrombotic 830
Enalapril ENA Cardiovascular 516
Fexofenadine FEX Antihistamine 467
Fluconazole FLU Antifungal agent 142
Ibuprofen IBU Anti-inflammatory 119,395
Ketoprofen KET Anti-inflammatory 415
Losartan LOS Cardiovascular 2793
Metoprolol MET Cardiovascular 4152
Naproxen NPX Anti-inflammatory 6068
Pantoprazole PAN Acid reflux 1927
Quetiapine QTP Antipsychotic 4112
Ramipril RAM Cardiovascular 340
Risperidone RIS Antipsychotic 24
Simvastatin SIM Lipid regulator 2956
Sulfamethoxazole SMX Antibiotic 255
Telmisartan TEL Cardiovascular 597
Tramadol TRA Opioid 1769
Trimethoprim TMP Antibiotic 1036a
Warfarin WAR Antithrombotic 245
Venlafaxine VEN Antidepressant 1864
a
Includes also trimethoprim in combination products with sulfamethoxazole or sulfadiazine

dipyridamole (DIP) are mainly used medicinally for the inhi- In the study area, there are seven wastewater treatment
bition of blood clot formation. Escitalopram (ESC) is the s- plants (WWTPs) which discharge their effluents to the surface
stereoisomer of citalopram (CIT) and both of them as well as waters, and two of them are shown in Fig. 1 (WWTPd 1 and 2).
venlafaxine (VEN) are used in treating depression and anxi- WWTP 1 has a population equivalent of 2600 and an average
ety. Pantoprazole (PAN) and fexofenadine (FEX) are used for discharge of 570 m3 day−1 (Environmental permit 2006). The
inhibiting the secretion of gastric acid and inhibiting allergy WWTP 2 has a population equivalent of 14,000 and an average
symptoms, respectively, while tramadol (TRA) is an opioid discharge of 4600 m3 day−1 (Environmental permit 2011).
used to relieve pain. IBU, KET, and naproxen (NPX) are The river water flows to the surface water intake site from
mainly used to treat pain, fever, and inflammation. which water is led to pre-treatment and thereafter to manage
recharge area of the Virttaankangas aquifer. The managed aquifer
Study area recharge (MAR, Dillon 2005, Niinikoski et al. 2016) site at
Virttaankangas consists of seven infiltration areas, each of which
The study area consists of Kokemäenjoki water course down- comprises of two to four infiltration ponds (Niinikoski et al.
stream from Tammerkoski, Tampere, and Virttaankangas 2016). After abstraction, the groundwater is pumped for water
aquifer in southwestern Finland (Fig. 1). The lake waters in supply to the city of Turku area, where 180,000 inhabitants use
the study area flow through Lake Pyhäjärvi near the city of artificial groundwater as household water. Also, several animal
Nokia and finally into the River Kokemäenjoki. In Karhiniemi farms and industries in the study area use the potable water
surface water intake site at Huittinen municipality, the river produced. The average water flow rate in surface water at study
waters are led to the Virttaankangas esker for producing arti- area is 15–30 cm s−1 while in groundwater wells, the flow rate is
ficial groundwater for Turku region (Artimo et al. 2008, approximately 0.002–0.006 cm s−1. The water flow in ground-
Assmuth et al. 2016). water observation wells is accelerated by infiltration.
 Environ Sci Pollut Res

Groundwater
recharge area

Fig. 1 Study area. a Finland, where locations of the cities are Tampere 61°29′53″N, 023°45′39″E, and Turku 60°27′05″N, 022°16′00″E. b Kokemäki
water course. c Groundwater sampling area

Experimental effluents of seven WWTPs located in the study area. The

24-h composite samples were taken of wastewater effluents
Wastewater sampling eight times during October 2012–August 2014, and of waste-
water influents once on 22 Oct and 12 Nov 2013. Upon sam-
To estimate the use and release of CBZ, IBU, KET, and SMX pling, the bottles were flushed with water to be sampled. The
to surface waters, they were determined in influents and bottles were stored cold in dark containers during transport,
Environ Sci Pollut Res

and samples were analyzed upon arrival to the laboratory or were placed approximately 1 m below the groundwater sur-
stored frozen until analysis, in order to minimize chemical face, and the length of the string was estimated so that despite
reactions of the analytes. the changes in water level, the samplers would remain below
the water surface. The inner diameter of the observation well
Preparation of the passive samplers was 52 mm, and holes were 0.3 mm in the 16–50 m sieving
part. The groundwater was not purged before deploying the
Before deployment, the receiving phase of the sampler (SDB- samplers to minimize the disturbance of hydrology in the well.
RPS disk) was conditioned with 20 mL methanol and 60 mL
UHQ-water and the sampler was assembled in a laboratory. Surface water sampling
The sampler housings were purified by soaking them in meth-
anol overnight. In surface water sampling, the SDB-RPS disk The grab samples were taken from three locations: upstream
was attached to polycarbonate sampler housing (Fig. 2, left) and downstream of the WWTP 2 and at the surface water
and in groundwater sampling, the SDB-RPS disk was fixed intake site from which water is pumped to the groundwater
around a glass bottle with three cable ties (Fig. 2, right). The recharge area. The samples were taken with Limnos water
bottle was filled with purified aquarium gravel (grain size 3– sampler on 4–8 occasions depending on the site between
4 mm) to prevent the bottle from floating during the sampling. 16th of Oct 2012 and 19th of Aug 2014. Grab samples taken
Each sampler was stored in a zip lock bag at + 4 °C before and during the whole project were used to estimate the concentra-
after the deployment. tion of APIs in surface water. The samples were taken and
treated as described above (2.4.1). The concentrations of
CBZ, IBU, KET, and SMX were analyzed from grab samples.
Groundwater sampling
Three replicate surface water Chemcatcher samplers were
deployed in upstream and downstream of the two WWTPs
This study involved three pairs of observation wells which are
(WWTP 1 and WWTP 2, Fig. 1) to study the variety and
labeled as 590 and 589, 585 and 591, and 25 and 52 (Fig. 1).
concentrations of selected APIs. In addition, three samplers
The passive samplers were deployed in three wells (590, 585,
were deployed downstream on the river at surface water intake
and 25, Fig.1), and the grab samples were taken from the
site where raw surface water is pumped to MAR-system. The
adjacent wells (589, 591, and 51, Fig. 1) with SS-Monsoon
samplers were kept at a depth of 1 m for 2 weeks during 16th–
pump. The grab samples were taken on 3–7 occasions depend-
30th of May 2014 (Ahkola et al. 2015).
ing on the observation well between 20th of Mar 2013 and 8th
of Sep 2014, and the groundwater was pumped for 20 min
Calibration of the passive samplers
before taking the grab sample. CBZ, IBU, KET, and SMX
were analyzed from grab samples. The sample bottles were
The sampling rates (RS) for samplers deployed in surface wa-
treated as described above for wastewater samples (2.4.1).
ter with polycarbonate sampler housing were determined for
Two replicate groundwater passive samplers were de-
example compounds CBZ, IBU, KET, and SMX in a labora-
ployed in each observation well at groundwater esker for
tory trial as described in Ahkola et al. (2015). A glass tank was
4 our weeks during 7th of May–4th of July 2014. The second
filled with 41 L of UHQ-water, and it was spiked with 2 mL of
sampler was fixed directly below the first one. The samplers
stock solution which contained concentrations of 5.5 μg ml−1
of CBZ, 5.4 μg ml−1 of both IBU and KET, and 2.3 μg ml−1 of
SMX in methanol. In the beginning of the trial, 10 samplers
were fixed to a stirrer rod with cable ties and the system was
continuously mixed to obtain flow rate of 33 cm s−1.The pas-
sive samplers were deployed for 2, 4, 7, and 9 days. As the
samplers were retrieved, the tank was emptied and refilled
with freshly spiked water. Grab samples were taken in every
renewal from discarded and freshly spiked tank water.

Sample treatment and analysis

After deployment, the SDB-RPS disk was removed from the

sampler housing and placed into a glass tube. The disk was
Fig. 2 Chemcatcher passive sampler deployed in surface water (left) and
spiked with mass-labeled surrogates and the solvent was
the modified version for groundwater sampling (right). The diameter of allowed to evaporate at room temperature for 10 min. The
the SDB-RPS disk is 47 mm disk was extracted with 5 mL methanol in an ultrasonic bath
 Environ Sci Pollut Res

for 15 min. The sample was centrifuged for 5 min (3000 rpm), results obtained in this study was quite high so the removal
and the eluent was transferred to a clean glass tube. The disk percents give only estimation. However, IBU seems to be
was then extracted with 5 mL of 2% formic acid in methanol removed effectively in both studies. SMX gave negative re-
in ultrasonic bath for 15 min. After centrifugation, the eluent moval percent but the variation cannot be estimated due to
was combined with the first extract and evaporated to dryness only one influent sample. The concentration of CBZ was only
with nitrogen stream. Finally, the dried sample was diluted to slightly higher in influent than in effluent, but the concentra-
50/50 (v/v) methanol/UHQ-water. tion has been commonly reported to increase during wastewa-
Aliquots of surface water and groundwater grab samples ter treatment (e.g., Jelic et al. 2011, Bahlmann et al. 2014,
(100 mL) and wastewater samples (50 mL) were spiked with Lindholm-Lehto et al. 2015).
mass-labeled surrogates. Prior to taking the aliquots, the sam-
ples were shaken vigorously for 1 min to avoid analyte loss to
the bottle walls. The pH was set to 2–3 with formic acid, and Groundwater sampling
the samples were extracted with 150 mg Oasis WAX solid
phase extraction cartridges (Waters, Milford, MA, USA). Passive samplers collected measurable amounts of 4 out of the
APIs were eluted from the cartridge with 10 mL of methanol. 25 studied APIs from groundwater observation wells: CBZ,
The spiked UHQ water samples from the passive sampler CIT + ESC, FLU, and SMX (Table 3). CBZ and SMX were
calibration trial (250 mL) were amended with 2 M ammonium observed in all of the groundwater wells with passive sam-
acetate and extracted with 200 mg Oasis HLB solid-phase pling. The highest observed API concentration was 23 ng
extraction cartridges (Waters). APIs were eluted from the car- sampler−1 for CBZ in the well 585. The amounts of SMX were
tridges with 6 mL of methanol and 6 ml of 2% formic acid in relatively low and near LOQ, ranging from 2.5 ng sampler−1
methanol. The extracts were evaporated to a smaller volume in well 25 to 4.3 ng sampler−1 in well 590. CIT + ESC were
and combined. All extracts were evaporated to dryness with discovered only in well 585 and FLU also in well 590 at near
centrifugal evaporator (EZ Envi, Genevac, Ipswich, UK). LOQ concentrations. Traces of AML, CIT + ESC, ENA, FLU,
Finally, the samples were dissolved in 1.0 mL of 50/50 (v/v) IBU, LOS, MET, NAP, RIS, TEL, and WAR were detected
methanol/UHQ-water. but they remained < LOQ. The lowest concentrations of CBZ,
The samples were analyzed with LC-ESI-MS/MS (Acquity FLU, and SMX were measured in well 25 located furthest
UPLC and Xevo TQ MS by Waters) in neutral, basic, or acidic downstream at the study area (Fig. 1). KET was not observed
conditions (i.e., 0.2 mM ammonium acetate, 0.2% ammonia, at all in groundwater samples, and its contents were low at
or 0.1% formic acid in UHQ-water and methanol as LC elu- surface water intake site.
ents) with gradient conditions. Chromatographic separation CBZ is persistent and frequently found in groundwater
was performed with Acquity UPLC BEH C18 (1.7 μm, (Clara et al. 2004, Sui et al. 2015), and as expected, it was
2.1 × 50.0 mm) analytical column. The limits of quantification discovered in all groundwater wells. Loos et al. (2010) report-
(LOQs) in passive samplers were as follows: TMP and DIP ed CBZ and SMX to be the two most relevant pharmaceutical
0.05 ng; BIS 0.07 ng; PAN, RIS, SIM, TEL, and VEN 0.1 ng; compounds for groundwater infiltration in their pan-European
WAR 0.2 ng; FEX, MET, QTP, and RAM 0.3 ng; ATO 0.4 ng; grab sampling survey on pollutants in groundwater. They re-
TRA 0.5 ng; LOS 0.6 ng; ∑CIT + ESC 0.7 ng; AML 0.8 ng; ported maximum concentrations of 390 and 38 ng L−1 for
SMX 1.8 ng; CBZ 2 ng; FLU 2.2 ng; IBU 3 ng; KET 5 ng; CBZ and SMX, respectively. The other two most frequently
NPX 6 ng; and ENA 9 ng per sampler. In surface and ground- observed APIs according to Loos et al. (2010) were IBU and
water samples, the LOQs of CBZ, IBU, KET, and SMX were KET with maximum concentrations of 395 and 2886 ng L−1.
2, 3, 30, and 5 ng L−1, respectively. In wastewater, the LOQ of As the removal of SMX was studied in laboratory bank filtra-
CBZ was 5 ng L−1, IBU and KET 50 ng L−1, and SMX tion simulation experiments, the residence time of several
10 ng L−1. weeks up to months was recommended (Baumgarten et al.
2011). Ending up of SMX to groundwater can be due to in-
sufficient residence time or unfavorable redox conditions.
Results and discussion CBZ was discovered in all wells also with grab sampling
but not on every occasion. The measured concentrations were
Wastewater sampling close to LOQ indicating that CBZ may be constantly present
at low concentrations. This also shows that when the sampling
Based on the influent and effluent concentrations (Table 2), is done with conventional grab sampling, APIs might stay
the average removal percent of IBU, KET, and CAR were 98, undiscovered although being frequently present in small quan-
34, and 9%, respectively. Lindqvist et al. (2005) studied tities. During passive sampling period, there can also be a
Finnish WWTPs and observed the removal percents of IBU concentration peak which is not necessarily detected with in-
and KET 92 and 78%, respectively. The variation of the stant grab sample.
Environ Sci Pollut Res

Table 2 Range of APIs at the WWTPs located in the study area same level upstream and downstream of the WWTPs and the
API −1
Influent (ng L ) −1
Effluent (ng L ) latter even at surface water intake site. Contents of CBZ, MET,
NPX, TEL, TRA, and VEN were higher at surface water in-
CBZ 87–570 20–530 take site than at other surface water sites. All the passive sam-
IBU 14,000–36,000 nd–6200 plers from five sampling sites gave measureable results for
KET 180–1100 90–700 SMX, but all of the grab samples had either amounts below
SMX < LOQ–310 nd–2000 LOQ or SMX was not detected at all (nd). This was observed
even though grab sampling results were from 4 to 8 different
sampling occasions conducted during 2012–2014.
Surface water sampling CBZ and IBU were discovered in all passive samplers.
KET was found only at one sampling site downstream of the
The sum of APIs was higher in samplers deployed down- WWTP 1 although it was measured in several grab samples.
stream than upstream from the WWTPs (Table 4) which clear- KET is known to metabolize in the human body mainly into a
ly proves that WWTPs are important sources of APIs to the glucuronide that may be reverted back into the original form
environment. However, FLU and SMX contents were at the (Jin et al. 1993). KET degrades rapidly when exposed to UV

Table 3 APIs measured in adjacent groundwater wells with passive and grab sampling

API Groundwater sampling with passive samplers (ng sampler−1) Grab sampling during 2013–2014a (ng L−1)

Observation Observation Observation Observation Observation Observation

well 590 well 585 well 25 well 589 well 591 well 51

AML < LOQ nd <LOQ na na na

ATO nd nd nd na na na
BIS nd nd nd na na na
CBZ 15 23 2.5 nd–3.1 (5/6) nd–2.9 (4/6) nd–2.4 (5/6)
CIT + ESC < LOQ 0.88 <LOQ na na na
DIP nd nd nd na na na
ENA nd nd <LOQ na na na
FEX nd nd nd na na na
FLU 2.4 3.0 < LOQ na na na
IBU nd nd < LOQ nd–4.5 (4/7) nd– < LOQ (3/7) nd–5.7 (3/7)
KET nd nd nd nd (0/3) nd (0/3) nd (0/3)
LOS < LOQ < LOQ < LOQ na na na
MET < LOQ < LOQ < LOQ na na na
NPX < LOQ < LOQ < LOQ na na na
PAN nd nd nd na na na
QTP nd nd nd na na na
RAM nd nd nd na na na
RIS < LOQ < LOQ < LOQ na na na
SIM nd nd nd na na na
SMX 4.3 3.7 2.5 nd– < LOQ (1/7) nd–7.7 (2/7) nd– < LOQ (3/7)
TEL nd nd nd na na na
TRA nd nd nd na na na
TMP nd nd nd na na na
WAR < LOQ < LOQ <LOQ na na na
VEN nd nd nd na na na
Number of observed APIs 3 4 2 2 2 2
Sum of API concentrations 22 31 5

In brackets: the number of grab samples in which API was discovered/the number of grab samples taken
na not analyzed, nd not detected
a
Results are recovery corrected
Table 4 APIs measured in surface water with passive and grab sampling

API Surface water sampling with passive samplers (ng sampler−1) Grab sampling during 2012–2014a (ng L−1)

Upstream of Downstream of Upstream of Down-stream Surface water Upstream of the Downstream Surface water
the WWTP I the WWTP I the WWTP II of the WWTP II intake site WWTP II of the WWTP II intake site

AML < LOQ < LOQ < LOQ nd nd na na na

ATO nd nd nd nd nd na na na
BIS 0.08 3.3 1.4 3.5 3.8 na na na
CBZ 7.2 15 18 19 27 < LOQ–3.3 (6/6) nd–3.6 (5/7) nd–3.2 (6/7)
CIT + ESC nd nd 1.0 nd <LOQ na na na
DIP < LOQ 0.78 < LOQ 0.21 nd na na na
ENA nd 160 < LOQ 33 13 na na na
FEX nd 5.9 1.6 2.6 2.0 na na na
FLU 3.7 3.2 3.4 3.4 <LOQ na na na
IBU 5.4 15 6.3 5.6 7.2 nd–3.7 (5/7) nd–4.8 (3/8) nd–6.0 (4/8)
KET nd 12.5 nd nd nd nd–110 (3/4) < LOQ–91(3/3) nd–160 (1/3)
LOS < LOQ 2.9 < LOQ < LOQ 0.6 na na na
MET 0.46 3.7 4.1 5.0 6.7 na na na
NPX < LOQ 24 < LOQ < LOQ 4.8 na na na
PAN nd nd nd nd nd na na na
QTP nd nd nd nd nd na na na
RAM < LOQ 1.1 0.65 0.78 0.76 na na na
RIS nd 0.20 nd nd 0.16 na na na
SIM nd 0.27 nd nd nd na na na
SMX 4.2 4.8 4.6 5.6 5.2 nd– < LOQ (1/7) nd– < LOQ (1/8) nd (0/8)
TEL nd 7.2 6.1 5.6 8.5 na na na
TRA nd 0.84 1.1 1.5 1.8 na na na
TMP nd 0.17 0.08 0.13 0.14 na na na
WAR 0.36 0.38 0.35 0.36 0.34 na na na
VEN 0.19 1.8 1.8 2.9 4.0 na na na
Number of observed APIs 8 20 14 15 16
Sum of API concentrations 22 262 51 89 86

In brackets: the number of grab samples in which API was discovered/the number of grab samples taken
na not analyzed, nd not detected
a
Results are recovery corrected
Environ Sci Pollut Res
Environ Sci Pollut Res

irradiation, forming, e.g., 3-ethylbenzophenone and 3- KET has the lowest sampling rate which implies the accumu-
acethylbenzophenone (Kotnik et al. 2014). The surface water lation to the sampler is slow.
samples with passive samplers were taken in May, and since
the samplers were placed near the water surface (1 m below Time-weighted average concentrations
the surface) where UV radiation can penetrate, it is likely that
photodegradation occurred (Quintana et al. 2005; Kotnik et al. The time-weighted average (TWA) concentrations of CBZ,
2014). The grab samples on the other hand gave measurable IBU, KET, and SMX in ground and surface waters are pre-
results as the samples were taken in January when there is less sented in Table 6 with the grab sampling results. As the grab
sunlight and ice cover. Lindholm-Lehto et al. (2016a) ob- samples taken during passive sampling trial were either
served higher KET contents in surface water during winter < LOQ or nd, we compared the passive sampling results with
than in summer at levels of 40–950 ng L−1, but in the study grab samples taken during the whole project (October 2012–
of Ternes (1998), KET was not detected in river water. The September 2014). CBZ and SMX were found at all surface
drastic variations in temperature and solar radiation between and groundwater sampling sites with passive sampling
summer and winter seasons in the North can have a major (Table 6). Despite that the CBZ concentrations remained at
effect on the degradation of certain APIs and their concentra- the same level in all grab samples, the passive samplers found
tions in the environment (Lindholm-Lehto et al. 2016b). higher contents in surface water sites. The same kind of trend
The highest API concentration measured with passive sam- was observed with SMX except SMX was measured at only
pling was 1160 ng sampler−1 of ENA downstream of the one groundwater grab sampling site. As the grab samples have
WWTP 1. Interestingly, it was not observed upstream of the been taken during 2 years, there are several snapshots which
WWTP 2 which is located further downstream in the river and should give an estimation of the overall picture of the water
the receiving watercourse from the WWTP 1. ENA is a quality. However, as the grab samples were taken only four
prodrug that is metabolized in human to its pharmacologically times a year, the concentration peaks were most likely missed.
active metabolite enalaprilat (Stankiewicz et al. 2015). In general, the concentrations were slightly higher in surface
Approximately 60% is excreted unmetabolized and 40% as waters, as expected. We used the same RS values for both
enalaprilat in urine. ENA and enalaprilat are both photode- surface and groundwater passive samplers to estimate the
gradable which can also affect their detection in surface waters TWA concentrations (Vrana et al. 2006; Allan et al. 2007;
exposed to sunlight (Perez et al. 2007). Single peak values Ahkola et al. 2015). At the study area, the flow rate in ground-
may be caused by changes in the discharges of the WWTPs water is much lower (0.002–0.006 cm s−1) than in surface
and occasional bypassing of the biological treatment. When water (15–30 cm s−1). With the lower flow rate, the sampling
untreated or only partly treated, wastewater is conveyed to the rate RS is decreased as well, leading to overestimation of the
watershed; it increases the load of compounds ending up in the groundwater concentrations. The use of performance refer-
environment. Bypassing may occur if the water loads are ence compounds (PRCs) to determine in situ sampling rate
exceeded due to leakage, exceptionally heavy rains, or power could give a more representative estimation of the sampling
cuts (Siintoharju 2016). rate in groundwater (Vrana et al. 2006). However, as PRC
compounds are deuterated compounds of interest, their use
Comparison of the results of surface and groundwater in groundwater wells used for drinking water supply may
passive sampling not be readily approved. However, according to Table 6, a
reasonable estimation about the TWA-concentration was
Calibration trial gained.
In general, the samplers detect lower concentrations of
In the calibration trial, the RS was determined, for example, APIs than grab samples because the samplers accumulated
compounds CBX, IBU, KET, and SMX (Table 5). The RS the APIs for 2 or 4 weeks and the trace concentrations were
value of CBZ was about three times higher than of SMX, concentrated to measurable levels. In addition to dissolved
which suggests accumulation of CBZ takes shorter time. fraction, the grab samples include also the particle-bound frac-
tion of the studied substance. Therefore, the correspondence
Table 5 The sampling of the results depends on the characteristic of compound. For
rates RS of selected APIs API RS example, CIT and BIS have been measured in settled particu-
(L day−1)
late matter below WWTPs (Lahti and Oikari 2011) and KET
CBZ 1.90 ± 0.52 in sediment particles (Lindholm-Lehto et al. 2015).
IBU 0.76 ± 0.13 The results indicate that several APIs are present in surface
KET 0.26 ± 0.12 waters but many of them are lost before they enter the ground-
SMX 0.61 ± 0.09 water either by degradation or by sorption to soil particles
during groundwater infiltration. Also, dilution affects the
 Environ Sci Pollut Res

Table 6 TWA concentrations (ng L−1) determined with passive sampling and the results of the grab samples

API Sampling Surface water (ng L−1) Groundwater (ng L−1)

WWTP 1 WWTP 2WWTP 2 Surface Observation wells

water
Upstream Downstream Upstream Downstream intake site 590 585 25

CBZ Passive 0.27 ± 0.01 0.57 ± 0.01 0.68 ± 0.03 0.73 ± 0.01 1.0 ± 0.08 0.29 ± 0.01 0.44 ± 0.04 0.05 ± 0.01
Grab na na <LOQ–3.3 (6/6) nd–3.6 (5/7) nd–3.2 (6/7) nd–3.1 (5/6) nd–2.9 (4/6) nd–2.4 (5/6)
IBU Passive 0.51 ± 0.01 1.4 ± 0.05 0.60 ± 0.03 0.53 ± 0.02 0.68 ± 0.03 nd nd nd
Grab na na nd–3.7 (5/7) nd–4.8 (3/8) nd–6.0 (4/8) nd–4.5 (4/7) nd– < LOQ (3/7) nd–5.7 (3/7)
KET Passive nd 3.4 ± 0.09 nd nd nd nd nd nd
Grab na na nd–110 (3/4) <LOQ–91 (3/3) nd–160 (1/3) nd (0/3) nd (0/3) nd (0/3)
SMX Passive 0.49 ± 0.01 0.57 ± 0.01 0.54 ± 0.01 0.66 ± 0.03 0.61 ± 0.02 0.25 ± 0.01 0.22 ± 0.01 0.15 ± 0.01
Grab na na nd– < LOQ (1/7) nd– < LOQ (1/8) nd (0/8) nd– < LOQ (1/7) nd–7.7 (2/7) nd– < LOQ (3/7)

Number in brackets indicates the number of taken grab samples/the number of samples in which API was detected
na not analyzed, nd not detected

concentrations of APIs. In this study, the artificial groundwa- is not readily sorbed to sludge. As excreted from human body,
ter is a mixture of infiltrated raw surface water and naturally APIs can form metabolites which are not detected when only
formed groundwater. Therefore, it is logical that although 16 parent compounds are measured. The metabolites are able to
different APIs were detected at the surface water intake site, cleave back to the parent compounds which are again detect-
only 4 of them were eventually found from groundwater wells ed, and it seems that the concentration of APIs increases. For
with passive sampling. In grab samples, the APIs may not be instance, CBZ is known to metabolize to N-glucuronide con-
detected due to their low concentrations but with passive sam- jugates which are at least partially cleaved back to their parent
pling, even the trace concentrations of the APIs can be compounds during wastewater treatment (Bahlmann et al.
detected. 2014). The amount of SMX varies also widely in effluents
of WWTPs depending on the applied treatment methods, ini-
Use and release of CBZ, IBU, KET, and SMX in the study area tial concentrations, and sludge retention times (Straub 2016;
Michael et al. 2013; Göbel et al. 2007). SMX is heavily me-
The amount of selected APIs sold in Finland in 2014 and their tabolized (80%), and its main human metabolite N 4 -
defined daily doses (DDDs) are presented in Table 7. Sales acetylsulfamethoxazole and its glucuronide conjugate can also
data for SMX is calculated from the DDD for combined prod- retransform back to the antibiotic itself during biological treat-
uct SMX-TMP which is the only type of SMX containing ment (Fimea 2016; Michael et al. 2013). In a critical review on
medicinal product sold in Finland (Fimea 2016). In Finland, aquatic environmental risk of SMX by Straub (2016), 190
IBU is one of the most and widely used non-steroidal anti- removal rates from 36 publications were assessed. The median
inflammatory drugs, and despite its DDD being eight times removal rate was 49%, average removal rate being 21% be-
higher than KET’s, the total mass of IBU usage is 316 times cause of a few extreme negative values. It is therefore likely
higher than the usage of KET. High concentrations of several that some of the SMX is removed at WWTPs, and the amount
APIs has previously been linked with mass events such as released in watershed could be less than the estimated amount
festivals (Daneshvar et al. 2012) and sports events in influents. To evaluate the true release of SMX in watershed,
(Lindholm-Lehto et al. 2015), during which unusually large measured concentrations from water samples should be ana-
crowds cause increased pharmaceutical load to local WWTPs lyzed and the main metabolite N4-acetylsulfamethoxazole and
or to the environment. Continuous release of APIs to the en- the glucuronide conjugate of SMX should also be taken into
vironment unnecessarily promotes antimicrobial resistance to account.
antibiotics (Khetan and Collins 2007), and also other effects In this study, the TWA concentrations of CBZ and SMX in
may result from chronic exposure even at low concentration surface and groundwater are in the same range although CBZ
level (Ferrari et al. 2003; Crane et al. 2006; Caracciolo et al. is used more than SMX based on the calculated annual sales
2015). data (Table 7). Despite that the usage statistics show that IBU
The elimination rates of SMX and CBZ during wastewater is used hundreds of kilos more than CBZ, KET, and SMX
treatment vary widely in numerous studies (Straub 2016; (Table 7), it was not detected in groundwater passive samplers
Bahlmann et al. 2014; Michael et al. 2013; Göbel et al. and only in few grab samples with near LOQ. One explanation
2007). It is likely that the level of CBZ in the effluents is could be that the characteristic of soil layers hinders the
higher than the estimated amount in the influents since CBZ leaching potential of IBU (Oppel et al. 2004). According to
Environ Sci Pollut Res

Table 7 Use, consumption, and

release of selected APIs at the API DDD (g) DDD (g)/1000 Sold in 2014 (kg) Amount in Amount in
study area (Finnish Medicines inhabitants/day influent (g/day) effluent (g/day)
Agency Fimea, 2014; WHOCC,
2016) CBZ 1 1.7 3300 21 15
CIT 0.02 13 520 nd nd
ENA 0.01 25 500 nd nd
ESC 0.01 14 270 nd nd
FLU 0.2 0.4 140 nd nd
IBU 1.2 49 120,000 2200 8.4
KET 0.15 1.3 370 63 22
SMX 1.6 0.1 290 7.2 10

na not analyzed

the estimated removal percentages of APIs at WWTPs, only used for drinking water production can be distracting as APIs
1% of IBU should reach the receiving watercourse. However, are brought back to this circle.
due to its wide consumption, the released amounts should still When studying harmful chemicals in groundwater, the con-
be ten times higher than CBZ, KET, and SMX. ventional grab sampling begins with pumping out several well
There are several studies on API concentrations in Finnish volumes which requires suitable equipment and energy. If the
rivers and lakes (Lindholm-Lehto et al. 2015, Äystö et al. purge water is suspected to be hazardous due to groundwater
2015, Brozinski et al. 2013, Vieno et al. 2006). Vieno (2007) contamination, it should be treated as wastewater and dis-
studied APIs at river Kokemäenjoki South-West Finland and posed accordingly. Passive sampling provides a technique to
found traces of IBU (2 ng L−1) but did not detect KET at all. In monitor the concentrations of chemicals without the genera-
Lake Päijänne, which is located in Central Finland and acts as tion of purge water and its possible need for purification.
a raw water source for capital area, have been reported SMX at When using passive sampling, the hydrology of the well is
concentrations of 14–26 ng L−1 (Ngumba et al. 2016), CBZ at not disturbed with pre-pumping and the possibly polluted
levels 1.0–2.1 ng L −1 , IBU at concentrations of 61– groundwater does not contaminate the soil. However, simul-
227 ng L−1, and KET at levels 59–97 ng L−1 (Lindholm- taneous grab sampling requires an adjacent well.
Lehto et al. 2016a). In river Vantaa in South-Finland, the mea- Several issues have to be considered when deploying
sured CBZ, IBU, and KET contents have been between 4.7– Chemcatcher passive samplers in groundwater well. For in-
65, 5.2–89, and 19–230 ng L−1, respectively (Lindholm-Lehto stance, the groundwater renewal is slower than in surface wa-
et al. 2016a). They observed sporadic concentration peaks ter where the water flow brings constantly more APIs for
during the sampling period (May to September). The reported samplers to adsorb so the samplers need longer time period
SMX, CBZ, and IBU contents were higher but KET can be in well. Also, the passive sampler developed for surface water
considered to be about the same level as in this study. has to be modified to fit into observation well. The RS is
determined in laboratory for surface water sampler, but as
the groundwater conditions differ (water velocity, turbulence,
temperature), the TWA-concentrations obtained with passive
Conclusions samplers are not as accurate as the instant grab sample. This
paper gives an estimation of TWA concentrations, but in fur-
As the waste water treatment process is not designed to re- ther studies, the Chemcatcher passive samplers should be cal-
move organic chemicals, many of them end up to surface ibrated in conditions representing the flow rates and sampler
waters. They can further infiltrate into aquifer especially in set-up in wells. Conventional grab sampling gives concentra-
areas where surface water is used to produce artificial ground- tion snapshot of total chemical content, while passive sam-
water. In this study, CBZ and SMX were transported a long plers collect the contaminants during a longer period of time,
path from releasing from WWTP to surface water and further usually several weeks which enables the detection of transient
to artificial groundwater. However, the APIs can form metab- concentration peaks as well. Therefore, passive sampling can
olites which can give misleading picture of the presence of be a suitable method especially for monitoring waters where
APIs if only the parent compounds are analyzed. This study API concentrations fluctuate or are presumed to be very low.
demonstrates that APIs can travel through the human body, In this study, many of the API concentrations determined in
waste water treatment process, surface water, and finally grab samples within 2 years remained below LOQ. Instead of
groundwater without degradation. The fact that APIs can occasional grab sampling, the presence of APIs could be de-
end up from the pharmacy to groundwater which is further termined with passive sampling in 2–4 weeks.
 Environ Sci Pollut Res

Acknowledgements The authors would like to thank Mr. Teemu Koski are found in the bile of wild fish caught downstream of a wastewater
from The Water Protection Association of the River Kokemäenjoki treatment plant. Environ Sci Technol 47:342–348
(KVVY) for valuable help with the surface water sampling campaign. Caracciolo AB, Topp E, Grenni P (2015) Pharmaceuticals in the environ-
This work was financially supported by the CONPAT project (Pathways, ment: biodegradation and effects on natural microbial communities.
risks and management of aquatic contaminants) financed by the AKVA A review. J Pharm Biomed Anal 106:25–36
program, Academy of Finland (SYKE 263499). Clara M, Strenn B, Kreuzinger N (2004) Carbamazepine as a possible
anthropogenic marker in the aquatic environment: investigations on
the behaviour of carbamazepine in wastewater treatment and during
groundwater infiltration. Water Res 38:947–954
References CONPAT (2014) Aquatic contaminants—pathways, health risks and man-
agement. Academy of Finland's Sustainable Governance of Aquatic
Resources (AKVA) program. http://en.opasnet.org/w/CONPAT.
Ahkola H, Herve S, Knuutinen J (2014) Study of different Chemcatcher Accessed 24 May 2017
configurations in the monitoring of nonylphenol ethoxylates and Crane M, Watts C, Boucard T (2006) Chronic aquatic environmental risks
nonylphenol in aquatic environment. Environ Sci Pollut Res Int from exposure to human pharmaceuticals. Sci Total Environ 367:23–41
21:9182–9192 Daneshvar A, Svanfelt J, Kronberg L, Weyhenmeyer GA (2012)
Ahkola H, Juntunen J, Laitinen M, Krogerus K, Huttula T, Herve S, Neglected sources of pharmaceuticals in river water—footprints of
Witick A (2015) Effect of the orientation and fluid flow on the a reggae festival. J Environ Monit 14:596–603
accumulation of organotin compounds to Chemcatcher passive sam- de la Cal A, Kuster M, de Alda ML, Eljarrat E, Barcelo D (2008) Evaluation
plers. Environ Sci-Processes Impacts 17:813–824 of the aquatic passive sampler Chemcatcher for the monitoring of high-
Ahkola H, Juntunen J, Krogerus K, Huttula T, Herve S, Witick A (2016) ly hydrophobic compounds in water. Talanta 76:327–332
Suitability of Chemcatcher (R) passive sampling in monitoring Dillon P (2005) Future management of aquifer recharge. Hydrogeol J 13:
organotin compounds at a wastewater treatment plant. Environ 313–316
Sci-Water Res Technol 2:769–778 Environmental permit (2006) Environmental permit concerning the dis-
Allan IJ, Knutsson J, Guigues N, Mills GA, Fouillac AM, Greenwood R charge and treatment of waste waters at WWTP 1. 1900Y0443–121
(2007) Evaluation of the Chemcatcher and DGT passive samplers (in Finnish)
for monitoring metals with highly fluctuating water concentrations. Environmental permit (2011) Environmental permit concerning the dis-
J Environ Monit 9:672–681 charge and treatment of waste waters at WWTP 2. 153/2011/1 (in
Artimo A, Sarapera S, Ylander I (2008) Methods for integrating an ex- Finnish)
tensive geodatabase with 3D modeling and data management tools Ferrari B, Paxeus N, Lo Giudice R, Pollio A, Garric J (2003)
for the Virttaankangas artificial recharge project, southwestern Ecotoxicological impact of pharmaceuticals found in treated waste-
Finland. Water Resour Manag 22:1723–1739 waters: study of carbamazepine, clofibric acid, and diclofenac (vol
Assmuth T, Simola A, Pitkanen T, Lyytimaki J, Huttula T (2016) 55, pg 359, 2003). Ecotoxicol Environ Saf 56:450–450
Integrated frameworks for assessing and managing health risks in Fetter CW (2001) Applied hydrogeology. Prentice Hall, New Jersey
the context of managed aquifer recharge with river water. Integr Fimea (2014) Finnish Medicines Agency, Finnish Statistics on Medicines
Environ Assess Manag 12:160–173 http://raportit.nam.fi/raportit/kulutus/laakekulutus.htm. Accessed 24
Bahlmann A, Brack W, Schneider RJ, Krauss M (2014) Carbamazepine May 2017 (in Finnish)
and its metabolites in wastewater: analytical pitfalls and occurrence Fimea (2016) Finnish Medicines Agency website http://www.fimea.fi/
in Germany and Portugal. Water Res 57:104–114 web/en/frontpage. Accessed 31 Aug 2016 (in Finnish)
Bailly E, Levi Y, Karolak S, (2013) Calibration and field evaluation of Focazio MJ, Kolpin DW, Barnes KK, Furlong ET, Meyer MT, Zaugg SD,
polar organic chemical integrative sampler (POCIS) for monitoring Barber LB, Thurman ME (2008) A national reconnaissance for
pharmaceuticals in hospital wastewater. Environmental Pollution pharmaceuticals and other organic wastewater contaminants in the
174:100–105 United States—II untreated drinking water sources. Sci Total
Environ 402:201–216
Barnes KK, Kolpin DW, Furlong ET, Zaugg SD, Meyer MT, Barber LB
Gobel A, McArdell CS, Joss A, Siegrist H, Giger W (2007) Fate of
(2008) A national reconnaissance of pharmaceuticals and other or-
sulfonamides, macrolides, and trimethoprim in different wastewater
ganic wastewater contaminants in the United States—I groundwater.
treatment technologies. Sci Total Environ 372:361–371
Sci Total Environ 402:192–200
Gunold R, Schafer RB, Paschke A, Schuurmann G, Liess M (2008)
Baumgarten B, Jahrig J, Reemtsma T, Jekel M (2011) Long term labora- Calibration of the Chemcatcher (R) passive sampler for monitoring
tory column experiments to simulate bank filtration: factors control- selected polar and semi-polar pesticides in surface water. Environ
ling removal of sulfamethoxazole. Water Res 45:211–220 Pollut 155:52–60
Birch H, Sharma AK, Vezzaro L, Mikkelsen PS, Holten Lützhøft HC ITRC (2007) Protocol for use of five passive samplers to sample for a variety of
(2011) Passive sampling and modeling of PAHs and heavy metals contaminants in groundwater; technical and regulatory guidance. The
in stormwater runoff. 12th international conference on urban drain- Interstate Technology & Regulatory Council Diffusion/Passive Sampler
age. Porto Alegre/Brazil. http://orbit.dtu.dk/en/publications/passive- Team. Washington DC. http://www.google.fi/url?sa=t&rct=j&q=&esrc=
sampling-and-modeling-of-pahs-and-heavy-metals-instormwater- s&source=web&cd=1&cad=rja&uact=8&ved=
runoff%283be6aad1-42fc-4d83-a289-8818b790b218%29.htm. 0ahUKEwjLqfqdrq7WAhXpQpoKHff0BV4QFggkMAA&url=http%
Accessed 18 Sept 2017 3A%2F%2Fwww.itrcweb.org%2FGuidanceDocuments%2FDSP-5.
Blom LB, Morrison GM, Kingston J, Mills GA, Greenwood R, Pettersson pdf&usg=AFQjCNEKxc_zLEN-zoQNZOicyMWl2Dfpuw Accessed
TJR, Rauch S (2002) Performance of an in situ passive sampling 19 Sept 2017
system for metals in stormwater. J Environ Monit 4:258–262 Jelic A, Gros M, Ginebreda A, Cespedes-Sanchez R, Ventura F, Petrovic
Bopp S, Weiss H, Schirmer K (2005) Time-integrated monitoring of poly- M, Barcelo D (2011) Occurrence, partition and removal of pharma-
cyclic aromatic hydrocarbons (PAHs) in groundwater using the ceramic ceuticals in sewage water and sludge during wastewater treatment.
dosimeter passive sampling device. J Chromatogr A 1072:137–147 Water Res 45:1165–1176
Brozinski JM, Lahti M, Meierjohann A, Oikari A, Kronberg L (2013) Jin CJ, Miners JO, Lillywhite KJ, Mackenzie PI (1993) Complementary
The anti-inflammatory drugs diclofenac, naproxen and ibuprofen deoxyribonucleic-acid cloning and expression of a human liver
Environ Sci Pollut Res

uridine diphosphate-glucuronosyltransferase glucuronidating car- (FKA the GORE (registered trademark) modules) for passive sam-
boxylic acid-containing drugs. J Pharmacol Exp Ther 264:475–479 pling of groundwater. Engineer Research and Development Center.
Khetan SK, Collins TJ (2007) Human pharmaceuticals in the aquatic envi- Arlington, Virginia
ronment: a challenge to green chemistry. Chem Rev 107:2319–2364 Perez S, Eichhorn P, Barcelo D (2007) Structural characterization of
Kingston JK, Greenwood R, Mills GA, Morrison GM, Persson LB photodegradation products of enalapril and its metabolite enalaprilat
(2000) Development of a novel passive sampling system for the obtained under simulated environmental conditions by hybrid
time-averaged measurement of a range of organic pollutants in quadrupole-linear ion trap-MS and quadrupole-time-of-flight-MS.
aquatic environments. J Environ Monit 2:487–495 Anal Chem 79:8293–8300
Kotnik K, Kosjek T, Krajnc U, Heath E (2014) Trace analysis of Persson LB, Morrison GM, Friemann JU, Kingston J, Mills G,
benzophenone-derived compounds in surface waters and sediments Greenwood R (2001) Diffusional behaviour of metals in a passive
using solid-phase extraction and microwave-assisted extraction sampling system for monitoring aquatic pollution. J Environ Monit
followed by gas chromatography-mass spectrometry. Anal Bioanal 3:639–645
Chem 406:3179–3190 Quintana JB, Weiss S, Reemtsma T (2005) Pathway’s and metabolites of
Lahti M, Oikari A (2011) Pharmaceuticals in settleable particulate mate- microbial degradation of selected acidic pharmaceutical and their
rial in urban and non-urban waters. Chemosphere 85:826–831 occurrence in municipal wastewater treated by a membrane biore-
Lindholm-Lehto PC, Ahkola HSJ, Knuutinen JS, Herve SH (2015) actor. Water Res 39:2654–2664
Occurrence of pharmaceuticals in municipal wastewater, in the re- Rozemeijer J, van der Velde Y, de Jonge H, van Geer F, Broers HP,
cipient water, and sedimented particles of northern Lake Paijanne. Bierkens M (2010) Application and evaluation of a new passive
Environ Sci Pollut Res 22:17209–17223 sampler for measuring average solute concentrations in a catchment
Lindholm-Lehto PC, Ahkola HSJ, Knuutinen JS, Herve SH (2016a) scale water quality monitoring study. Environmental Science &
Widespread occurrence and seasonal variation of pharmaceuticals Technology 44:1353–1359
in surface waters and municipal wastewater treatment plants in cen- Sanchez-Bayo F, Hyne RV, Kibria G, Doble P (2013) Calibration and
tral Finland. Environ Sci Pollut Res 23:7985–7997 field application of Chemcatcher(a (R)) passive samplers for detect-
Lindholm-Lehto PC, Ahkola HSJ, Knuutinen JS, Koistinen J, Lahti K, ing Amitrole residues in agricultural drain waters. Bull Environ
Vahtera H, Herve SH (2016b) Suitability of passive sampling for the Contam Toxicol 90:635–639
monitoring of pharmaceuticals in Finnish surface waters. Environ Siintoharju P (2016): Management and environmental risks of pump
Sci Pollut Res 23:18043–18054 station overflows and wastewater treatment plant bypasses in
Lindqvist N, Tuhkanen T, Kronberg L (2005) Occurrence of acidic phar- Pirkanmaa region. Centre for Economic Development, transport
maceuticals in raw and treated sewages and in receiving waters. and the environment for Pirkanmaa, reports 11/2016 (in Finnish)
Water Res 39:2219–2228 Stankiewicz A, Giebultowicz J, Stankiewicz U, Wroczynski P, Nalecz-
Loos R, Locoro G, Comero S, Contini S, Schwesig D, Werres F, Balsaa P, Jawecki G (2015) Determination of selected cardiovascular active
Gans O, Weiss S, Blaha L, Bolchi M, Gawlik BM (2010) Pan- compounds in environmental aquatic samples—methods and re-
European survey on the occurrence of selected polar organic persis- sults, a review of global publications from the last 10 years.
tent pollutants in ground water. Water Res 44:4115–4126 Chemosphere 138:642–656
Macleod SL, Mcclure EL, Wong CS (2007) Laboratory calibration and Straub JO (2016) Aquatic environmental risk assessment for human use
field deployment of the polar organic chemical integrative sampler of the old antibiotic sulfamethoxazole in Europe. Environ Toxicol
for pharmaceuticals and personal care products in wastewater and Chem 35:767–779
surface water. Environ Toxicol Chem 26:2517–2529 Sui Q, Cao X, Lu S, Zhao W, Qiu Z, Yu G (2015) Occurrence, sources
Mcquay H, Carroll D, Jadad AR, Wiffen P, Moore A (1995) and fate of pharmaceuticals and personal care products in the
Anticonvulsant drugs for Management of Pain—a systematic re- groundwater: a review. Emerging Contaminants 1:14–24
view. Br Med J 311:1047–1052 Ternes TA (1998) Occurrence of drugs in German sewage treatment
Michael I, Rizzo L, McArdell CS, Manaia CM, Merlin C, Schwartz T, plants and rivers. Water Res 32:3245–3260
Dagot C, Fatta-Kassinos D (2013) Urban wastewater treatment Turja R, Lehtonen KK, Meierjohann A, Brozinski JM, Vahtera E,
plants as hotspots for the release of antibiotics in the environment: Soirinsuo A, Sokolov A, Snoeijs P, Budzinski H, Devier MH,
a review. Water Res 47:957–995 Peluhet L, Paakkonen JP, Viitasalo M, Kronberg L (2015) The mus-
Mills GA, Vrana B, Allan I, Alvarez DA, Huckins JN, Greenwood R sel caging approach in assessing biological effects of wastewater
(2007) Trends in monitoring pharmaceuticals and personal-care treatment plant discharges in the Gulf of Finland (Baltic Sea). Mar
products in the aquatic environment by use of passive sampling Pollut Bull 97:135–149
devices. Anal Bioanal Chem 387:1153–1157 Vermeirssen ELM, Bramaz N, Hollender J, Singer H, Escher BI (2009)
Muziasari WI, Managaki S, Parnanen K, Karkman A, Lyra C, Tamminen Passive sampling combined with ecotoxicological and chemical
M, Suzuki S, Virta M (2014) Sulphonamide and trimethoprim resis- analysis of pharmaceuticals and biocides—evaluation of three
tance genes persist in sediments at baltic sea aquaculture farms but are Chemcatcher (TM) configurations. Water Res 43:903–914
not detected in the surrounding environment. Plos One 9(3):e92702 Vermeirssen ELM, Dietschweiler C, Escher BI, van der Voet J, Hollender
Ngumba E, Kosunen P, Gachanja A, Tuhkanen T (2016) A multiresidue J (2013) Uptake and release kinetics of 22 polar organic chemicals in
analytical method for trace level determination of antibiotics and an- the Chemcatcher passive sampler. Anal Bioanal Chem 405:5225–
tiretroviral drugs in wastewater and surface water using SPE-LC-MS/ 5236
MS and matrix-matched standards. Anal Methods 8:6720–6729 Vieno N (2007): Occurrence of pharmaceuticals in Finnish Sewage.
Niinikoski P, Sarapera S, Hendriksson N, Karhu JA (2016) Geochemical Treatment plants, surface waters, and their elimination.
and flow modelling as tools in monitoring managed aquifer re- Dissertation, Tampere University of Technology
charge. Appl Geochem 74:33–43 Vieno NM, Tuhkanen T, Kronberg L (2006) Analysis of neutral and basic
Oppel J, Broll G, Loffler D, Meller M, Rombke J, Ternes T (2004) pharmaceuticals in sewage treatment plants and in recipient rivers
Leaching behaviour of pharmaceuticals in soil-testing-systems: a using solid phase extraction and liquid chromatography-tandem
part of an environmental risk assessment for groundwater protection. mass spectrometry detection. J Chromatogr A 1134:101–111
Sci Total Environ 328:265–273 Vienonen S, Rintala J, Orvomaa M, Santala E, Maunula M (2012): Climate
Parker L, Willey R, McHale T, Major W, Hall T, Bailey R, Gagnon K, change impacts and the adaptation measures in the Finnish water
Gooch G (2014) Demonstration of the AGI universal samplers services The Finnish Environment 24/2012 (in Finnish)
 Environ Sci Pollut Res

Vrana B, Mills G, Greenwood R, Knutsson J, Svenssone K, Morrison G Vrana B, Mills GA, Dominiak E, Greenwood R (2006) Calibration of the
(2005b) Performance optimisation of a passive sampler for monitoring Chemcatcher passive sampler for the monitoring of priority organic
hydrophobic organic pollutants in water. J Environ Monit 7:612–620 pollutants in water. Environ Pollut 142:333–343
Vrana B, Mills GA, Allan IJ, Dominiak E, Svensson K, Knutsson J, WHOCC (2016): WHO Collaborating Centre for Drug Statistics
Morrison G, Greenwood R (2005a) Passive sampling techniques Methodology Website http://www.whocc.no/. Accessed 31
for monitoring pollutants in water. Trac-Trends in Analytical Aug 2016
Chemistry 24:845–868 Äystö L, Mehtonen J, Mannio J (2015) Lääkeaineet Etelä-Suomen
pintavesissä ja puhdistetussa jätevedessä. Ympäristö ja terveys 4:
40–46 (in Finnish)