Analysis of organic pollutants during sewage treatment processes

Analysis of organic pollutants during sewage treatment processes

A tracing of Sulfomethoxazole, N-Acetylsulfamethoxazole, Ranitidine and Mefenamic acid

Gattiker Etienne, Schälchli Jeremy & Zimmerli RogerInstitute for biogeochemistry and pollutant dynamics, ETH Zurich

Monday, May 21, 2012

Supervision:Aurea C. Chiaia-Hernandez, Rebekka Gulde, Juliane Hollender, Philipp Longree, Christoph Moschet, Heinz

Singer, Marc Suter, Sara Pati, Jürgen Van der Voet & Reto Wijker

Report for the Practical Course: Analysis of Organic Pollutants

Spring Semester 2012


AbstractCurrent STPs are designed to remove nutrients, oil and particles but not micropollutants.

Micropollutants such as pharmaceuticals, personal care products, and industrial chemicals are

released into the environment in a concentration range of ng·l-1 to μg·l-1 (Sturm et al., 1998).

Varying fractions of the different compounds are eliminated by sorption to the sewage sludge.

This study concerns about four organic pollutants, namely Sulfomethoxazole and its

metabolite N-Acetylsulfamethoazole, Ranitidin and Mefenamic acid. The STP Neugut in

Dübendorf was chosen as a sampling site.

Five different sampling points were chosen representing three stages of treatment in the STP

and two stages of interactions with the ecosystems. The primary effluent (PE) after

preliminary sedimentation of solids, the secondary effluent (SE) after biological treatment, the

final effluent (FE) after sand filtration, water of the river Glatt upstream (GU) and

downstream (GD) the STP discharge.

The passive samplers consisting SDB disc covered with a PES membrane gives about 60-70%

lower values than using the time-proportional (or flow-proportional) samplers. It was shown

that the LC-MS/MS is a good and sensitive tool to measure micropollutants with best recover-

ies in the FE following by the GU and then the PE samples. The present elimination rates of

the target micropollutants were determined while Mefemic Acid removal is below anticipated

value and Sulfamethoxazole above.

Table of content Abstract................................................................................................................................................21 Introduction.......................................................................................................................................3

1.1 Relevance and objectives.............................................................................................................31.2 STP Dübendorf............................................................................................................................31.3 Target compounds........................................................................................................................4

2 Material and Methods.......................................................................................................................42.1 Sampling collection.....................................................................................................................42.2 Sample preparation......................................................................................................................52.3 Sample extraction and enrichment...............................................................................................62.4 Passive Sampling.........................................................................................................................72.5 Quantitative Analysis by LC-MS/MS..........................................................................................72.6 Data evaluation............................................................................................................................7

3 Results ...............................................................................................................................................93.1 Quality of quantitative analysis ..................................................................................................9

3.1.1 LOD / LOQ.........................................................................................................................93.1.2 Relative and absolute recovery............................................................................................9

3.2 Quantitative analysis by LC-MS/MS.........................................................................................103.3 Qualitative Analysis – Detection of compounds in the sediment...............................................13

4 Discussion.........................................................................................................................................15

Page 2 ETH Zurich, 21. May. 2012


1 Introduction

1.1 Relevance and objectives

In Switzerland, 97% of the Swiss population was connected to a sewage treatment plant (STP)

by 2005 (FOEN, 2009). Current STPs are designed to remove nutrients, oil and particles but

not micropollutants. Micropollutants such as pharmaceuticals, personal care products, and

industrial chemicals are released into the environment by sewage in a concentration range of

ng·l-1 to μg·l-1. Studies show that those pollutants are often only partially removed in sewage

treatment plants (STPs) (Chiaia-Hernandez et al. 2012) and that their effects to the

environment are not well understood (Perazzolo et al. 2010). Varying fractions of the different

compounds are eliminated by sorption to the sewage sludge (Chiaia-Hernandez et al. 2011).

There are around 30'000 chemicals in daily use (Schwarzenbach et al, 2006) from which

2,000 are pharmaceutical (Stamm et al, 2006). Public and scientific concerns are particularly

high for hormone-active substances such as pharmaceuticals because of their potential long-

term effects on humans and environmental organisms (Perazzolo et al. 2010). The swiss edict

for protection of water (SR 814.201) contains “de lege ferenda” an obligation for about 100

STPs to upgrade by an additional treatment step, like «active carbon filtering» or

«ozonization», to remove organic pollutants with an efficency of 80 %. Therefore it is

important to have a exact tool to measure the concentrations of organic pollutants within the

STP's and in the environment. The measured concentrations can be used to define the

efficiency of STPs and their treatment steps and further to assess the concentration ranges for

ecotoxicological long-term studies.

1.2 STP Dübendorf

The STP Neugut in Dübendorf was chosen as a sampling site. It was built in 1964 and

upgraded in 1993 (Chiaia-Hernandez et al. 2012). It treats both domestic and industrial

sewage from several surrounding municipalities. The population equivalent value of the STP

Neugut is 100’000. This includes 40’000 inhabitants, 10’000 commuters and 50’000

population equivalent from industry. Because a populaiton equivalent requires to 170 L per

day, the STP Neugut treats 17 million L of water each day (Chiaia-Hernandez et al. 2012).

At first, a screen in the STP Neugut removes solid waste and large particles are separated

through gravitational settling. In a conventional activated sludge (CAS) system both

nitrification and denitrification take place (Chiaia-Hernandez et al. 2011) in separate



compartments, under oxic and anoxic conditions respectively. In contrast to most STP, in

Neugut phosphorus is also removed biologically (Chiaia-Hernandez et al. 2011) rather than

through chemical co-precipitation with iron oxides. The last step consists of a sand filter to

remove remaining particles (Chiaia-Hernandez et al. 2011) and to further mineralise organic

compounds. The treated water is eventually released into the river Glatt.

During all processes, particles of diverse composition settle down to form sewage sludge. The

sludge consists mainly of organic material and thus is a potentially suitable sorbent for apolar

and therefore hydrophobic organic pollutants. It is usually decomposed under anaerobic

conditions resulting in biogas and residuals, which are incinerated with garbage in

Switzerland.

1.3 Target compounds

This study concerns about four organic pollutants, namely Sulfomethoxazole and its

metabolite N-Acetylsulfamethoxazole, Ranitidine and Mefenamic acid (see Table 1).Table 1: Target compounds: Properties and Description

Substance [1] Properties [2] Description [1]

Sulfomethoxazole (C10H11N3O3S)

M = 253.28 g·mol-1

log Kow = 0.48

Sulfomethoxazole is a sulfonamide antibiotic. Annual consumption in Switzerland (7.2 mio inhabitants) makes up to 2'300 kg·a-1. The predicted elimination in a STP is about 40 %.

N-Acetylsulfamethoxazole (C12H13N3O4S)

M = 295.31 g·mol-1

log Kow = 1.21

N-Acetylsulfamethoxazole is a conjugate of sulfamethoxazole derived from acetylation in the liver. In human excretion it makes up to 50% of consumed Sulfomethoxazole. The predicted elimination in a STP is about 90 %.

Ranitidine (C13H22N4O3S)

M = 314.40 g·mol-1

log Kow = 0.29

Ranitidin is an alimentary tract drug used for treatment of peptic ulver and gastrooesophageal reflux disease. Annual consumption in Switzerland makes up to 1'150 kg·a-1. The predicted elimination in a STP is about 40 %.

Mefenamic acid (C15H15NO2)

M = 241.29 g·mol-1

log Kow = 5.28

Mefenamic acid is an anti-inflammatory drug used to treat pain. Annual consumption in Switzerland makes up to 22'000 kg·a-1. The predicted elimination in a STP is about 90 %.

Source: [1]: Chiaia-Hernandez et al. 2012; [2]: ChemSpider

2 Material and Methods

2.1 Sampling collection

Five different sampling points were chosen (Fig. 1) representing three stages of treatment in

the STP and two stages of interactions with the ecosystems (Chiaia-Hernandez et al. 2012).


http://www.chemspider.com/


The primary effluent (PE) after preliminary sedimentation of solids, the secondary effluent

(SE) after biological treatment, the final effluent (FE) after sand filtration, water of the river

Glatt upstream (GU) and downstream (GD) the STP discharge. Samples were taken by

permanently installed samplers in proportion to the flow over 24 hours for PE, SE and FE

(Chiaia-Hernandez et al. 2012). GU and GD samples were taken time proportional before and

after the outlet of the STP with an automatic sampler (Chiaia-Hernandez et al. 2012).

Altogether 24 samples were taken (Table 2) and additional four trip blanks (see chapter 2.2).Table 2: Overview of the different samples

Sample Shortcut Volume Replicates

Primary effluent PE 100 ml 6

Secondary effluent SE 100 ml 3

Final effluent FE 100 ml 6

Glatt upstream GU 500 ml 6

Glatt downstream GD 500 ml 3

Trip blank - 500 ml 4

2.2 Sample preparation

All samples were filtered with a glass-fiber filter to remove particulate matter. Three of these

six aliquots from PE, FE and GU were spiked with analyte standard solution (see below) that

contained all target compounds at a concentration of 10µg/ml. Every aliquot was spiked with


Fig. 1: Scheme of the sampling points (black dots with red circle) at the STP Neugut and the River Glatt. (Chiaia-Hernandez et. al, 2012)


an internal standard solution.

We therefore obtained three replicates of each sample and additionally three spiked replicates

from PE, FE and GU and four blank aliquots. The three replicates spiked with the analyte

standard were used to determinate accuracy and precision (see chapter 2.6 & 3.1). The

internal standard was used to over come losses of the analyte over the method (see Table 3).

A sludge sample was planned but was not available. Instead of the sludge we took a sediment

sample of lake Greifensee.

Every sample was stored at -20°C for one week until enrichment.

Internal standard solution are labelled compounds almost identical to the compounds of

interest. They are used as a baseline to track potential losses of target compounds during

analysis through signal drift, matrix effects or analyte loss (Chiaia-Hernandez et al. 2012).

Samples were spiked with 100 μl of a 1 mg·l-1 solution was used with the surrogates:

Sulfamethoxazole-D4, Acetyl-Sulfamethoxazole-D5, Mefenamic acid-D3 and Ranitidine-D6.

Analyte standard solutions were used to prepare calibration curves and to spike samples in

order to determine the recovery of the analysis. The solution contains all target compounds in

a defined concentration, which in our case was 10 mg l-1. Our calibration curves consisted of

nine data points with intended concentrations of 2, 5, 10, 20, 50, 100, 200, 500, and 1’000

μg·l-1 (Chiaia-Hernandez et al. 2012). Three triplicate samples were spiked with the standard

solution, the sample GU with 100 μg·l-1, PE and FE with 1000 μg·l-1 and 500 μg·l-1

respectively.

2.3 Sample extraction and enrichment

Before the enrichment step all the samples were adjusted to a pH of 7. The enrichment was

done via solid phase extraction with an Oasis™ HLB cartridge after moisturising this

cartridge with methanol (3 × 2 ml) and water (2 × 3 ml water). The retained analytes were

eluted with 8 ml methanol. This volume was further reduced to 0.1 ml via evaporation and

then increased to 1 ml with HPLC grade water (see Fig. 2).

The extraction and pre-concentration of the sediment sample was done by pressurized liquid

extraction (PLE) using a combination of ethyl acetate and acetone at a temperature of 80°C

followed by liquid-liquid-extraction (LLE) to clean the extract from sludge matrix. After ex-

traction all samples will be stored ad 4°C until GC-MS analysis.



2.4 Passive Sampling

The passive sampler consisting of a Styrenedivinylbenzene (SDB) disc covered with a poly-

ethersulfone (PES) membrane. They were placed on the two sampling sites GU and GD for

14 days.

Directly after deployment, the six SDB discs were inlaid in 7ml acetone, internal standard was

added and the discs were extracted in a shaker for 30min. After bringing the samples to the

lab, the SDB discs were extracted with methanol likewise. The two obtained fractions were

combined, filtered, reduced to a volume of 0.1 ml via evaporation accelerated by heating and

ventilation with nitrogen gas. Finally the sample volume was re-increased with HPLC grade

water to a final volume of 1 ml.

2.5 Quantitative Analysis by LC-MS/MS

Quantification measurements were done by first separating the analytes with liquid

chromatography (LC), using a Rheos 2000 LC pump and an Atlantis T3 column (3 µm, 150 x

3.0 mm; Waters Corp.). 20 µl of the samples were injected. The mobile phase consisted of

water and methanol and both mobile phases had 0.1% by volume formic acid. For detection

and quantification, a triple quadrupole TSQ Quantum MS/MS from Thermo with electrospray

ionization interface (ESI) was used (Chiaia-Hernandez et al. 2012).

2.6 Data evaluation

The Predicted environmental concentrations (PEC) of the PE, FE and GD samples are calcu-

lated with the values given in chapter 1.2 & 1.3 and Table 1. The river Glatt and STP Neugut

were assumed to have an averaged discharge of 5000 l·s-1and 253 l·s-1 respectively. The PEC


Fig. 2: Schematic overview of the extraction and enrichment steps. (Chiaia-Hernandez et. al, 2012)


are given in Table 7.

Additional the average concentration of Sulfamethoxazole was calculated by Eq. 1 using the

measured amount of Sulfamethoxazole within the SDB disc and a specific sampling rate (RS)

of 0.09 l·d-1 (Vermeirssen et al. 2009).

During mass spectrometry the two most prominent fragments of each compound were

analysed and quantified. The fragment with the highest signal-to-noise ratio was used as

quantifier. Peak integration was performed by Thermo Xcalibur™ software and checked for

correctness. Quantification of the received signals was done by comparing with internal

standard signals and an external standard calibration curve, to assure linear dependence

between signal and concentration.

To determine limit of detection (LOD) the Signal-to-noise ratio (S/N) was used. The limit of

quantification (LOQ) is about 3 times higher. This follows of the convention for the S/N for

LODs and LOQs of 3 and 10 respectively. The LOD of the samples was determined using

Eq. 2.

The matrix factor reflects the change of analyte peak intensity of different sample matrices

(i.e. PE to FE). It was calculated by using LC-MS/MS in data and Eq. 3:

The accuracy of the method was determined by relative recovery, which describes the fraction

of originally present analyte mass that can still be measured after (each of) the different steps

of sample preparation, separation, and detection (Chiaia-Hernandez et al. 2012). The relative

recovery was computed by following equation:


Eq. 1: Calculation of the amount of target compound sorbed on the SDB, where MS

[ng] is the amount of the substance sorbed, cw [ng/l] the water condenration, t [d] the time of the deployment and Rs [l/d] specific sampling rate. (Chiaia-Hernandez et al. 2012)

M s( t)=cw⋅Rs⋅t

Eq. 2: Determining LOD by S/N ratio, with:LODnanopure = calculated LOD of a low concentration calibration standardm = matrix factor

LOD=LODnanopure

m⋅dilution factor

Eq. 3: Calculation of the matrix factor for the different samples matrices of PE, SE, FE, GU and GD.

m=( Area spiked sample−Areaunspiked sample)

Areastandard (as spiked conc.)

Eq. 4: Calculation of the relative recovery

Rel.Recovery=(conc spiked sample−concunspiked sample)

concadditional spike


3 Results 3.1 Quality of quantitative analysis

3.1.1 LOD / LOQ

The values are in a reasonable order of magnitude (see Fig. 4). One can easily see the influ-

ence of the matrix, as LOQs usually decrease from PE to Glatt samples. This behaviour is in

line with the expectation.

3.1.2 Relative and absolute recovery

To assess the precision and accuracy of our method, standard deviations (see Fig. 3 & Table 4)

and relative and absolute recoveries were taken into account. As already mentioned, one of

the three SE measurements yielded concentrations that were much too high for all compounds

(see chapter 10). This was probably due to a wrong amount of spiked internal standard in that

aliquot. We therefore discarded that measurement. After doing so, relative standard deviations

of the measured concentrations were 10% or lower. This shows that overall precision of our

measurements are okay.

Table 3: Absolute and relative recoveries in per cent

Sulfamethoxazole N-Acetylsulfamethoxazole

Absolute Relative

Absolute Relative

PE 56 95 PE 67 90

FE 81 104 FE 88 101

GD 55 91 GD 76 91

Ranitidine Mefenamic acid

Absolute Relative

Absolute Relative

PE 61 86 PE 80 76

FE 75 98 FE 95 103

GD 31 88 GD 74 91

Recovery is an indicator for accuracy of the method. The values are calculated as described in

Chiaia-Hernandez et al. (2012) and shown in Table 3.

Especially relative recovery is of interest. It is affected by differences between each analyte

and its surrogate which in turn might pose a significant problem for accurate quantification.

However, for our measurements relative recoveries are all between 76 and 104 % which in-



dicates an acceptable accuracy. Linearity of the external calibration curve within the measured

concentration range is also important for proper quantification. In our case, this was given

since linear regression coefficients were very close to 1, namely 0.97 (usually 0.98 to 0.99).

3.2 Quantitative analysis by LC-MS/MS

Measured concentrations of the four compounds in flow-proportional samples are given in

Fig. 3. Additionally, they are listed in Table 4 with some remarks about their quality. The val-

ues can be used to calculate removal efficiency of the STP and dilution within the river Glatt.

They can also be compared with expected concentrations based on consumption information

and river discharge (see Table 7 for PE, FE & GD) and concentration values obtained with

passive sampling (see Table 5 & 6 for GU & GD).


Fig. 3: Measured concentrations of all four compounds in flow-proportional samples. All values in ng/l. Standard deviations are given as black bars.

0

50

100

150

200

250

300

350

400

450

500

PE SE FE GU GD

Sulfomethoxazole

0

50

100

150

200

250

300

350

400

450

500

PE SE FE GU GD

N-Acetylsulfamethazole

0

20

40

60

80

100

120

140

PE SE FE GU GD

Ranitidin

0

200

400

600

800

1000

1200

1400

1600

1800

PE SE FE GU GD

Mefemic Acid


Table 4: Measured concentrations of the four compounds from time-proportional sampling in ng/l

Sulfamethoxazole N-Acetylsulfamethoxazole

PE 352 ± 19 - PE 479 ± 47 -

SE 151 SE1 dissmissed SE 22 SE 1 dismissed

FE 154 ± 11 - FE 0 below LOQ

GU 53 ± 4 - GU 9 below LOQ

GD 67 - GD 9 only 1 value, below LOQ

Ranitidine Mefenamic acid

PE 121 ± 7 - PE 1550 ± 60 -

SE 111 SE1 dismissed SE 224 SE1 dismissed

FE 99 ± 11 - FE 98 ± 10 -

GU 16 ± 5 - GU 35 ± 3 -

GD 15 only 2 values GD 49 only 2 values

One measurement of SE (SE 1) yielded systematically too high concentrations for all four

compounds. This was probably due to a wrong amount of spiked internal standard. We there-

fore dismissed that value for all four compounds. The other missing values are due to concen-

trations below limit of detection. Please note that most values for N-Acetylsulfamethoazole

concentrations are below LOQ.

Table 5: Concentration of Sulfamethoxazole (ng/l) from the passive sampling in the Glatt (GU: upstream; GD: downstream). There were 6 measurements made. On the right side the average is presented. GU1 GU2 GU3 GU4 GU5 GU6 AV_GU

31 18 17 15 16 15 16

GD1 GD2 GD3 GD4 GD5 GD6 AV_GD

18 16 24 27 26 25 27

For all compounds, there is a significant decrease of the concentration from PE to FE and

river water (see Fig. 3 & Fig. 4). Removal efficiency of the STP is 56% for Sulfamethoxazole

based on concentration measurements only. If we account for N-Acetylsulfamethoazole which

is transformed to Sulfamethoxazole within the WWTP and therefore represents a source of the

latter, the removal efficiency is even 81%. Both values are clearly higher than the predicted

40% removal. N-Acetylsulfamethoazole decrease amounts to at least 86%, if LOQ is taken as

FE concentration. In fact N-Acetylsulfamethoazole concentration is below LOQ. Removal ef-

ficiency of Ranitidin and Mefemic Acid is 94% and 18%, respectively. Expected removal

would be 90% and 40%. We therefore find that STP removal efficiency of N-Acetyl-

sulfamethoazole and Ranitidin are in line with expected values, while Mefemic Acid removal

is below anticipated value and Sulfamethoxazole above.



Dilution within the river Glatt can be calculated with GU, GD and FE values and compared to

expected dilution (1/20) based on known discharge of the STP (253 l·s-1 in average) and the

river before STP (5000 l·s-1). These calculations yield a dilution of 1/11 for Sulfamethoxazole

and 1/6 for Mefemic Acid. While these values compare well with the expected dilution of

1/20, values for dilution of N-Acetylsulfamethoazole and Ranitidin are not computable be-

cause of concentrations below LOQ (N-Acetylsulfamethoazole) or no significant difference

between GU and GD concentrations (Ranitidin).

Passive Sampling measurements in the river Glatt (GU and GD) have been done for

Sulfamethoxazole (see Table 5). The results are in good accordance with the flow-proportion-

al samples. There is a factor of 0.3 (GU) and 0.4 (GD) between the concentrations measured

with the two sampling methods (see Table 6). We therefore conclude that passive sampling

might yield trustworthy results (at least for qualitative purposes), although it is susceptible to

many factors (flow rate, biofouling, compound specific sampling rate, etc.).

Table 6: Measured concentrations of Sulfamethoxazole (ng/l) in the river Glatt

Sulfamethoxazole Flow-proportional sampling Passive sampling Ratio

GU 53 16 0.3

GD 67 27 0.4

At last, measured concentrations can be compared with expected amounts of the compounds

in PE, FE and GD (see Table 7). This comparison shows that our measurements seem plaus-

ible, since obtained concentrations are in good agreement with expected ones. All associated

values are in the same order of magnitude. However, the not detectable amount of N-Acetyl-


Fig. 4: Limits of Quantification as determined by signal-to-noise ratio of peak height. All values in ng/l. Name of compounds and used fragments: Sulfam. = Sulfamethoxazole (156 Da fragment), N-Acetylsulfam. = N-Acetylsulfamethoazole (134 Da fragment), Ranitidin (176 Da fragment), Mefemic = Mefemic Acid (224 Da fragment).

0102030405060708090

100

Sulfam. N-Acetylsulfam. Ranitidin Mefemic

Limits of Quantification (ng/l)

PE

SE

FE

Glatt


sulfamethoazole in FE is a remarkable deviation, but in line with the fact that for all four ana-

lytes measured concentrations in PE and FE are lower than estimated. On the other hand, con-

centrations in the river Glatt are higher than expected. This might be due to an already con-

taminated river upstream the STP.

Table 7: Measured vs. expected concentrations of all four compounds in PE, FE and GD in ng/l

Sulfamethoxazole N-Acetylsulfamethoazole

Measured Expected Ratio


PE 352 515 0.68 PE 479 1287 0.37

FE 154 309 0.50 FE 0 129

GD 67 15 4.47 GD 9 6 1.5

Ranitidin Mefemic Acid



PE 121 386 0.31 PE 1550 2462 0.63

FE 99 232 0.43 FE 98 246 0.4

GD 15 12 1.25 GD 49 12 4.08

3.3 Qualitative Analysis – Detection of compounds in the sediment


Fig. 5: Print Screen from the GC Analysis. On the top our measurement on the bottom the proposition of the library for sulphur.


Because the sludge sample of the sewage treatment plant was unusable we used sediment core

from lake Greifensee to assess the qualitative analysis using GC-MS. Fig. 5 shows a screen-

shot of the measurement and the library data. The high consistency between the two screens

shows the high probability that the found compound is sulfur. As second compound we found

Palmitic acid also known as hexadecanoic acid. The mentioned is a common fatty acid found

in animals or plants. Fig. 6 show an isotopic composition of the found compound and a simu-

lated isotopic composition Fig. 7. Relative abundance is decreasing in both simulated and

measured figures from 256 to 257 to 258.


Fig. 7: Simulated composition of the isotope composition of palmitic acid

Fig. 6: Scheme of the isotopic composition of the palmitic acid.


4 Discussion

It could be shown that calculated elimination rates fit to the predicted ones. Half of the invest-

igated compounds (namely N-acetyl sulfomethoxazole and mefenamic acid) get eliminated

more than 80 % as submitted by the draft of the SR 814.201. But an additional elimination

step is still needed to remove more sulfomethoxazole and ranitidine.

The LC-MS/MS is good and sensitive tool to measure micropollutants in concentration ranges

of ng/l to ug/l. But passive samplers somehow gives about 60-70% lower values than using

the time-proportional (or flow-proportional) samplers (see Table 6).

BibliographyChiaia-Hernandez et al. 2012: Aurea C. Chiaia-Hernandez, Rebekka Gulde, Juliane

Hollender, Philipp Longree, Christoph Moschet, Heinz Singer, Marc Suter, Sara Pati, Jürgen

Van der Voet & Reto Wijker: Analysis of organic pollutants - Tracing polar organic

contaminants during sewage treatment processes. Courseware of a practical course at the

Eawag Dübendorf. 55 pages.

FOEN, 2009: Federal Office for the Environment: Indicator Connection to wastewater

treatment plants. Website: http://www.bafu.admin.ch/umwelt/, retrieved 20.04.2012.

Perazzolo et al. 2010: Perazzolo, C., Morasch B., Kohn T., Magnet A., Thonney D., Chèvre

N. : Occurrence and Fate of Micropollutants in the Vidy Bay of Lake Geneva, Switzerland.

Part I: Priority List for Environmental Risk Assessment of Pharmaceuticals. Environmental

Toxicology and Chemistry, volume 29 number 8, pages 1649-1657.

Schwarzenbach et al, 2006: Schwarzenbach R. P., Escher B. I., Fenner K., Hofstetter T. B.,

Johnson C. A ., Von Gunten U., Wehrli B.: The Challenge of Micropollutants in Aquatic

Systems. Review. Science, volume 313, pages 1072-1077.

Sturm et al., 1998: Sturm B., Knauth H.-D., Theobald N. & Wünsch G.: Hydrophobic

organic micropollutants in samples of coastal waters: efficiencies of solid-phase extraction in

the presence of humic substances. Fresenius' Journal of analytical chemistry, volume 361,

number 8. pages 803-810.

Vermeirssen et al., 2009: Vermeirssen E.L.M., Bramaz N., Hollender J., Singer H., Escher

B.I.: Passive sampling combined with ecotoxicological and chemical analysis of

pharmaceuticals and biocides – evaluation of three Chemcatcher™ configurations. Water

Researchvolume 43, issue 4, pages 903–914


http://www.bafu.admin.ch/umwelt/indikatoren/08605/08610/index.html?lang=en

Analysis of organic pollutants during sewage treatment processes

Documents