project acronym poseidon contract no. evk1-ct-2000-00047 · juan m. lema, marta carballa, maria...

POSEIDON, detailed report related to the overall duration (1.1.2001-30.6.2004) 09/05/06

Energy, Environment and Sustainable Development

Assessment of Technologies for the Removal of Pharmaceuticals and Personal Care Products in Sewage and Drinking Water Facilities to

Improve the Indirect Potable Water Reuse

Project acronym POSEIDON

Contract No. EVK1-CT-2000-00047 http//www.eu-poseidon.com

Project co-ordinator: Dr. Thomas Ternes

Detailed REPORT related to the overall project duration: January 1st, 2001 – June 30th, 2004

Workpackage leader: Thomas A. Ternes, Marie-Laure Janex-Habibi, Thomas Knacker, Norbert Kreuzinger, Hansruedi Siegrist

August, 2004


Official partners of the POSEIDON project Federal Institute of Hydrology (BfG), Koblenz; ESWE-Institute for Water Research and Water Technology, Wiesbaden, Germany Thomas Ternes, Matthias Bonerz, Derek McDowell, Guido Fink, Nadine Herrmann, Burkhard Hoffmann, Dirk Löffler, Jeanette Stüber

EAWAG, Dübendorf, Switzerland Hansruedi Siegrist, Alfredo Alder, Silvio Canonica, Adriano Joss, Anke Göbel, Marc Huber, Elvira Keller, Christa McArdell, Angela Thomsen, Urs von Gunten Tampere University of Technology, Tampere, Finland Tuula Tuhkanen, Susanna Korhonen, Niina Lindqvist, Sari Kuusisto, Jari Lehtiranta Vienna University of Technology, Vienna, Austria Norbert Kreuzinger, Manfred Clara, Oliver Gans, Birgit Strenn, Birgit Vogel University of Santiago de Compostela, Santiago de Compostela, Spain Juan M. Lema, Marta Carballa, Maria Llompart, Francisco Omil Silesian University of Technology, Gliwice, Poland Korneliusz Miksch, Joanna Surmacz-Górska, Sebastian Zabczynski ECT Oekotoxikologie GmbH Thomas Knacker, Markus Liebig, Johann F. Moltmann CIRSEE - Environment Marie-Laure Janex-Habibi, Auguste Bruchet

Associated end-user of the POSEIDON project Waterworks Biebesheim Lilo Weber

WEDECO Umwelttechnologie GmbH Achim Ried

Wastewater association Braunschweig Bernhard Teiser Kemira Jukka Jokela Scientific Officer of the EU for POSEIDON Kirsi Haavisto

Energy, Environment and Sustainable Development


Content 1 DETAILED REPORT, RELATED TO THE OVERALL PROJECT DURATION............................................. 1

1.1 BACKGROUND ................................................................................................................................. 1

1.2 SCIENTIFIC AND SOCIO-ECONOMIC OBJECTIVES ........................................................................... 2

1.3 APPLIED METHODOLOGY, SCIENTIFIC ACHIEVEMENTS AND MAIN DELIVERABLES ..................... 4

1.3 1ANALYSIS AND OCCURRENCE OF PPCPS ..................................................................................... 4

1.3.2 WASTEWATER TECHNOLOGY AND SOURCE CONTROL............................................................ 11

1.3.3 INDIRECT DISCHARGE OF TREATED WASTEWATER INTO UNSATURATED SOIL ........................ 23

1.3.4 DRINKING WATER TECHNOLOGY............................................................................................. 28

1.3.5 ENVIRONMENTAL RISK ASSESSMENT........................................................................................ 38

1.3.6 RECOMMENDATIONS FOR INDIRECT POTABLE WATER REUSE IN RELATION TO PPCP REMOVAL................................................................................................................................... 43

1.4 CONCLUSIONS INCLUDING SOCIO-ECONOMIC RELEVANCE, STRATEGIC ASPECTS AND POLICY IMPLICATIONS.......................................................................................................... 50

1.5 DISSEMINATION AND EXPLOITATION OF THE RESULTS ............................................................... 51

1.6 PUBLICATIONS .............................................................................................................................. 56


1. Detailed report, related to the overall project duration

1.1 Background

The limited quantity of unpolluted water available for future use as a resource for drinking water production is one of the major challenges faced around the world, including Europe. For instance in Mediterranean countries limited water resources and therefore water quality is an important economic factor [Water management Europe 1993/94]. Indirect reuse can increase the water supply in areas in which the growth of urbanized population has exceeded the quantity of available natural water sources [Crook et al., 1990, 1998, 1999; Cuthbert and Hajanosz, 1999; Harhoff and van Merve, 1996; Lauer and Roger, 1996; Lauer et al., 1991, Vazquez et al., 1996; Jagals and Lues, 1996]. Currently, many communities in Europe and world-wide use water resources for drinking water production that contain a significant portion of wastewater. So far, strategies for municipal wastewater treatment have hardly been focused on the elimination of organic trace pollutants, although for instance prescription and non-prescription pharmaceuticals and personal care products (PPCPs) are produced and used by humans in quantities that exceed thousands of metric tons annually. Approximately 3000 different pharmaceutical ingredients are used in the EU today, including painkillers, antibiotics, antidiabetics, beta-blockers, contraceptives, lipid regulators, antidepressants, antineoplastics, tranquilizers, impotence drugs and cytostatic agents. As these compounds are frequently transformed in the body, a combination of unchanged pharmaceuticals and metabolites are excreted by humans. Human-use pharmaceuticals enter raw sewage via urine and feces and by improper disposal. These pharmaceuticals are discharged from private households and from hospitals and eventually reach municipal wastewater treatment plants (WWTPs). If PPCPs are only partially eliminated, residual quantities enter ambient waters or groundwater. However, direct inputs into natural waters are also possible through storm water overflow and leaks in the sewer system. Personal care products include the ingredients of shampoos, liquid bath admixtures, skin care products, dental care products, soaps, sun screen agents, hair styling products etc., which are used in enormous quantities throughout the world. In the early 1990s their annual production exceeded 550,000 t for Germany alone [Daughton and Ternes, 1999]. Fragrances such as nitro and polycyclic musks as well as UV blockers (e.g. methylbenzyliden camphor) and preservatives (e.g. parabens and isothiazolin derivatives) are also included [Ternes et al., 2003]. In contrast to pharmaceuticals, personal care products do not have to pass through the human body. They enter the wastewater via their regular use during showering or bathing. Frequently they are used as components of cosmetics which mainly consist of lipids or oils (e.g. sun creams) so that a higher lipophilicity is crucial for them. Considerable persistence and bioaccumulation potential has been reported for them [Geyer et al., 2000]. The precautionary principle with regard to drinking water supply and wastewater treatment, however, implies an efficient removal of all potential harmful constituents. PPCPs are frequently polar and persistent organic compounds, and furthermore possess extremely high biological potency (i.e. estrogens). However, these chemicals recently detected in surface water and drinking water are not considered in the Drinking Water Directive 98/83/EC. The indirect potable water reuse (unplanned or planned) of municipal WWTP discharges leads to an exposure of the environment and ultimately of drinking water to these chemicals. The removal efficiency of existing wastewater treatment must be optimized and new technologies need to be developed.

1


1.2 Scientific and socio-economic objectives

So far, the behavior and removal of PPCPs as potential contaminants was neither appropriately investigated in wastewater nor in drinking water treatment processes. As a major innovation, POSEIDON tackled this problem adequately at source, by focusing on the development of alternative methods of wastewater collection (i.e. separate collection of urine) in addition to advanced, innovative wastewater treatment (i.e. membrane technology) and optimization of conventional techniques. A comprehensive scheme should be developed for the implementation of measures regarding the elimination of persistent domestic chemicals as contaminants of reclaimed WWTP discharges. Both wastewater and drinking water technologies optimized for the removal of domestic chemicals could be applied world-wide for indirect water reuse. Analysis and Fate Profound knowledge of the degradation and fate of PPCPs is important to evaluate the elimination processes in WWTPs and to assess environmental and health risks. Methods should be developed to enable the quantification of PPCPs in wastewater and sludge as well as the partitioning between the aqueous phase and sludge. Furthermore, degradation of PPCPs and information on their metabolites in wastewater and oxidative drinking water treatment are directly related to the efficiency of treatment technologies. Wastewater Technology The suitability of distinct wastewater treatment processes for the elimination of PPCPs has hardly been studied, so far. In order to determine their efficiency, as a major innovative step, advanced techniques such as membrane technology and effluent ozonation were tested to be combined with various conventional techniques such as activated sludge and biofilter systems (for nitrification and denitrification). In addition, POSEIDON addressed the efficiency of novel sustainable approaches such as urine separation with conventional and advanced purification techniques. Since many pharmaceuticals are excreted mainly by the urine, the implication of urine separation for reducing PPCP exposure was investigated. Indirect Discharge of Treated Wastewater into Unsaturated Soil

Irrigation or infiltration of treated wastewater or receiving water on urban and agricultural areas could be proposed as planned indirect reuse in arid areas with scarce water resources to tackle water quantity problems. POSEIDON elucidated whether the irrigation of treated wastewater can lead to groundwater contamination and which appropriate wastewater treatment techniques will ensure that groundwater contamination does not occur. Drinking Water Technology Due to the contamination of rivers and groundwater with PPCPs, some of these compounds are present in drinking water, though in relatively low concentrations. As an innovative step the efficiency of drinking water management and quality control is assessed. In particular, the efficiency of individual removal technologies such as flocculation, activated carbon, oxidation processes (ozonation, advanced oxidation processes) or membrane technologies applied to raw water are investigated. To achieve cost efficiency tailor-made combinations of technologies are needed according to the level of pollution in raw water.

2


Environmental Risk Assessment A scheme for a risk assessment for pharmaceuticals has been outlined by the European Agency for the Evaluation of Medicinal Products (EMEA) in a draft Note for Guidance ‘NfG’ (EMEA 2003). The innovative steps within POSEIDON was to study the effect of selected PPCPs to aquatic and sediment dwelling organisms, to adapt existing exposure assessment procedures to PPCPs and to perform ERAs specifically designed for PPCPs. A product-specific approach (eco-labeling) was developed to assess the impact of personal care products on the environment. A large number of these bio-active compounds enter wastewater and the receiving water bodies without being tested to determine their environmental impact. Since PPCPs are designed to exert biological activities, the evaluation of their fate and effects should be required, just like it is required for all other chemicals. Management Tool for Indirect Potable Reuse of Treated Wastewater A management tool for planned potable indirect reuse taking into consideration the removal and fate of PPCPs is essential for the sustainability of water distribution cycles by simultaneously minimizing hazards for environmental organisms and human beings. These combined aspects represent a major innovation in this field, as, in the past, the presence of PPCPs was not taken into consideration in potable water reuse. The main objective of POSEIDON can be summarized: 1. to develop a strategy to assess and improve the removal of PPCPs in wastewater and

drinking water by

• investigating conventional and advanced wastewater and drinking water treatment with respect to their efficiency in eliminating PPCPs.

• analyzing PPCPs in wastewater and drinking water treatment processes to elucidate fate and - if possible - degradation pathways.

• studying irrigation in agricultural areas to assess the pollution of groundwater by PPCPs. • investigating source control by urine separation, thus reducing the contamination caused by

pharmaceuticals in wastewater facilitating their treatment. 2. to develop a strategy for indirect potable water reuse of treated urban wastewater with

respect to the removal of PPCPs and thus combining wastewater and drinking water technologies.

3. to perform environmental risk assessments (ERAs) for selected PPCPs.

3


1.3. Applied methodology, scientific achievements and main deliverables

6.3.1 Analysis and occurrence of PPCPs In recent years several studies in Europe and North America were reported which exhibited the occurrence of pharmaceuticals and estrogens in wastewater and ambient waters [Kolpin et al., 2000; Daughton and Ternes, 1999; Heberer, 2002]. In general, the concentrations of PPCPs in WWTP effluents ranged from the ng L-1 to the low µg L-1 range. In surface waters the concentrations of these compounds ranged mainly between 10-500 ng L-1. Even in ground water and drinking water PPCP residues were detected up to the µg L-1 level. The question arises whether these residues pose risks for aquatic ecosystems or humans. In the POSEIDON project representatives of different medicinal classes and of personal care products have been selected as target PPCPs primarily based on elevated annual prescriptions (e.g. ibuprofen), specific mode of actions as indicators for environmental and/or human risks (e.g. estrogens, antibiotics) and/or different physico-chemical properties such as sorption (e.g. heavily sorbing polycyclic musk fragrances) and/or (bio-)chemical transformation (e.g. persistence: carbamazepine, biodegradability: natural estrogens). Furthermore, the availability of analytical methods for the determination of the selected PPCPs in water was another crucial criterion. The following PPCPs have been selected as target compounds, in order to limit the analytical method development and to limit the efforts for the number of analysis for the technological experiments:

• Acidic drugs: diclofenac, ibuprofen (both antiphlogistics), bezafibrate (lipid regulator) • Neutral drugs: diazepam (tranquilizer), carbamazepine (anti-epileptic) • Personal care products: tonalide (AHTN), galaxolide (HHCB) (musk fragrances) • Antibiotics: sulfamethoxazole, roxithromycin • Iodinated contrast media (ICM): iopromide • Estrogens: 17α-ethinylestradiol, 17β-estradiol, estrone

I. Analytical quality assurance (AQ)

When measuring PPCPs the elevated polarity and the expected low concentrations (ng L-1-, ng g-1-levels) in environmental matrices (e.g. wastewater, surface water, sludge, sediment, and biota), requires a comprehensive quality assurance. Especially, the potential influence of matrix effects on the recoveries has to be elucidated (i.e. by standard addition, spiking of a non polluted sample, or determining recoveries of the surrogate standards). It is strongly recommended that the LOQ lie within the linear range of the calibration. The signal/noise ratio should be at least 10 or higher.

AQ parameters used can be summarized as follows: • 5-10 point calibration over the total method: starting from extraction to the final detection • Surrogate standard (SS) for each group spiked at the beginning • Instrumental standard (IS) spiked in the final volume – if available • At least 3 recoveries with acceptable relative standard deviation (RSD) or 95% confidence

intervals confirming the quality of the method • One recovery and one blank sample in each sample series • Absolute and relative recoveries from spiked real water samples • Linearity of calibration curve over the range for quantitation • Limit of quantitation (LOQ) has to be within the linear calibration range; signal/noise ratio ≥ 10.

4


Interlaboratory comparison study

After an analytical training course, organized by ESWE/BfG, an interlaboratory comparison study (ILCS) was obligatory for all participants or subcontractors of POSEIDON analyzing water and wastewater samples for PPCPs. The ILCS includes spiked and non spiked treated wastewater (WW) and drinking water (DW). The quality of the analytical competence can be seen by the high rating in Table 1.1 for the 11 PPCPs selected. It has to be noted that not all participants had to analyze all PPCPs and all matrices selected within POSEIDON. Analytical methods used in the ILCS for water samples The participants slightly modified the analytical methods described below, but in general they were according to Ternes, 2001.

Antibiotics • The solid phase materials 0.1 g LiChrolute®-EN/0.25 g RP-C18 or 0.20 g HLB 200 were

filled into a glass cartridge. • Filtration over glass fiber (<1µm, pH 7-8.), spike of surrogate standards (SS) 500 ng

Oleandomycin and Sulfamethoxazole-D4. • The samples were sucked through the glass cartridges at a flow rate of ~20 mL min-1 and

dried completely by a nitrogen stream for 1 h • The analytes was eluted with methanol, evaporation to dryness by a gentle N2 stream and re-

dissolved. • Detection by LC electrospray Tandem MS Iodinated contrast media • The solid phase materials (200 mg of ENV+, 150 mg RP-C18ec) were manually packed in

glass cartridges. • Filtration over glass fiber (< 1µm), adjusted to pH 2.8 and then spiked with 500 ng DMI as

SS. • The samples were sucked through the glass cartridges: flow rate ~10 mL min-1, dried

completely by a nitrogen stream for 1h and the RP-C18ec material was removed • The analytes were eluted with methanol, evaporation to dryness by a gentle N2 stream and re-

dissolved • Detection by LC electrospray Tandem MS Non charged pharmaceuticals at neutral pH • The solid phase materials 0.5 g RP-C18ec were filled into a glass cartridge. • Filtration over glass fiber (< 1µm, pH 7-8), spike of SS 500 ng dihydrocarbamazepine. • The samples were sucked through the packed glass cartridges: flow rate ~20 mL min-1 dried

completely by a nitrogen stream for 1h. • The analytes were eluted with methanol, evaporation to dryness and re-dissolved. • Detection either by LC electrospray Tandem MS or GC/MS after silylation

De-protonated (acidic) pharmaceuticals at neutral pH • The solid phase materials 0.1 g LiChrolute®-EN and 0.25 g RP-C18 were successively filled

into a glass cartridge. PTFE-frits were placed on top, in between and below the SPE materials.

• Filtration over glass fiber (< 1µm), adjusted to pH 2 and then spiked with 500 ng Fenoprop as SS.

• The samples were sucked through the packed glass cartridges: flow rate ~20 mL min-1. • The cartridges were dried completely by a nitrogen stream for 1 h

5


• The analytes were eluted four times with 1 mL of methanol, evaporation to dryness and re-dissolved.

• The extracts were evaporated to dryness and derviatized by adding 0.2 mL diazomethane/diethylether-solution by a gentle nitrogen stream and re-dissolved in 200 µL n-hexane.

• Finally 500 ng of PCB-30 was added as an internal standard and the detection were performed by GC/MS.

Polycyclic musk fragrances • The solid phase materials 0.5 g RP-C18 were filled into a glass cartridge. • Filtration over glass fiber (< 1µm, pH 7-8), spike of SS 500 ng Tonalide-D3 (AHTN). • The samples were sucked through the packed glass cartridges: flow rate ~20 mL min-1 dried

completely by a nitrogen stream for 1h. • The analytes were eluted with methanol, evaporation to dryness and re-dissolved in 200 µL

hexane, and the internal standard 50 ng PCB-30 was added. • Detection by GC/MS in SIM mode.

Table 1.1: Results of the interlaboratory comparison study (ILCS) of POSEIDON.

lab. 1 lab. 2 lab. 3 lab. 4 tap w. sewage tap w. sewage tap w. sewage tap w.

Sulfamethoxazole ++ + + + ++ ++ Roxithromycin ++ + ++ ++ ++ + Tonalide + ++ + ++ + + ++ Galaxolide + ++ + + ++ ++ + Bezafibrate + ++ + ++ + Diclofenac ++ (+) ++ ++ ++ ++ ++ Ibuprofen + ++ + (+) ++ ++ (+) Carbamazepine ++ (+) ++ ++ (+) Diazepam ++ + + + ++ + + Iopromide - (+) ++ ++ ++ ++

lab. 5 lab. 6 lab. 7 tap w. sewage tap w. sewage tap w. sewage

Sulfamethoxazole Roxithromycin Tonalide + ++ ++ Galaxolide + ++ ++ Bezafibrate ++ ++ Diclofenac ++ + + - Ibuprofen ++ ++ ++ Carbamazepine ++ + + - Diazepam ++ (+) ++ ++ Iopromide

++: excellent (deviation between found and spiked conc.: < 10 %) +: good (deviation between found and spiked conc.: 10-30 %) (+): acceptable, but need of improvement (deviation between found and spiked conc.: 30-50 %) - : not acceptable (deviation between found and spiked conc.: >50 %)

6


II. Occurrence of PPCPs in European wastewaters and surface waters

After the analytical methods for the water matrices have been successfully established in the laboratories of the POSEIDON participants, monitoring data were obtained and collected from the countries participating.

Occurrence of Pharmaceuticals and musk fragrances in European wastewater and rivers

Table 1.2 lists occurrence data of PPCPs in different European countries. One has to be aware that the data sets of the six countries are based on a different number of samples and therefore a direct reliable comparison of the quantitative results is not possible. Data sets vary from samples of only one WWTP over a short period of time to samples of a multitude of WWTPs taken several times. Furthermore, for some countries the data may not be representative due to the limited number of sites sampled. Since the fraction of the treated wastewater discharged into the sampled rivers is varying due to different dilution factors, data for surface waters give a rough range of environmental concentrations occurring in European countries. Nevertheless, the data collected in Table 1.2 exhibits that the PPCPs selected are occurring in several European countries. A European wide relevance can therefore be expected. In tendency it can be concluded: • For some of the PPCPs such as carbamazepine and diclofenac the concentrations listed in Table

1.2 are comparable for most of the countries. For other PPCPs such as the macrolide antibiotic roxithromycin significant differences were found. The highest concentration levels of pharmaceuticals seem to occur in Germany.

• In the Polish WWTP, tonalide (AHTN) and roxithromycin were not detected at all, and in the Spanish WWTP diclofenac and roxithromycin were not found.

• For all seven countries diazepam was hardly detected.

• In France, the concentrations of roxithromycin and iopromide in the WWTP effluents and in the rivers were extremely low.

• For Finland the situation seems to be a little bit different, since the wastewater concentrations of diclofenac and bezafibrate are lower and those of ibuprofen are higher than in central Europe.

The diverting concentrations in the influents probably result from different consumption patterns in the countries. However, much more measurements on additional sites are needed to confirm those conclusions for the whole countries.

7


Table 1.2: Median (maximum) concentrations in Germany (GER), Austria (AUT), Poland (PL), Spain (ES), France (FR), Switzerland (CH). Concentrations in WWTP influents and effluents and in surface waters are given in ng L-1.

PPCP Location GER AUT PL ES FR CH FIN

Diclofenac influent 3500 (28000)

3100 (6000) 1750 (2000) n.d. n.a. 1400 (1900) 350 (480)

effluent 810 (2100) 1500 (2000) n.a. n.d. 295 (300) 950 (1140) 250 (350)

river 150 (1200) 20 (64) n.a. n.a. 18 (41) 20–150 15 (40)

Ibuprofen influent 5000 (14000)

1500 (7200) 2250 (2800) 2750 (5700) n.a. 1980 (3480) 13 000 (19 600)

effluent 370 (3400) 22 (2400) n.a. 970 (2100) 92 (110) < 50 (228) 1300 (3900)

river 70 (530) n.d. n.a. n.a. 23 (120) n.d.–150 10 (65)

Bezafibrate influent 4900 (7500) 2565 (8500) 780 (1000) n.d. n.a. n.a. 420 (970)

effluent 2200 (4600) 103 (611) n.a. n.d. 96 (190) n.a. 205 (840)

river 350 (3100) 20 (160) n.a. n.a. 102 (430) n.a. 5 (25)

Diazepam influent < LOQ n.d. n.a. n.d. n.a. n.d. n.d.

effluent < LOQ (40) n.d. n.a. n.d. n.d. n.d. n.d.

river n.d. n.d. n.a. n.a. n.d. n.d. n.d.

Carba- influent 2200 (3000) 912 (2640) 1150 (1600) n.a. n.a. 690 (1900) 750 (2000)

mazepine effluent 2100 (6300) 960 (1970) n.a. n.a. 1050 (1400) 480 (1600) 400 (600)

river 250 (1100) 75 (294) n.a. n.a. 78 (800) 30–150 70 (370)

SMX

influent 1370 (1700) n.d. (470) 1550 (2000) 600 n.a. 425 (570) 1670a)

(1900 a))

n.a.

effluent 400 (2000) 31 (234) n.a. 250 n.d. 290 (860) 400a) (880a))

n.a.

river 30 (480) n.d. n.a. n.a. 25 (133) n.a. n.a.

Roxithro- influent 830 (1000) 43 (350) n.d. n.d. n.a. 20 (35) n.a.

mycin effluent 100 (1000) 66 (290) n.a. n.d. n.d. 15 (30) n.a.

river <LOQ (560) n.d. n.a. n.a. 9 (37) n.a. n.a.

Iopromide influent 13000 (22000)

n.d. (3840) 1330 (2700) 6600 n.a. 810 (7700) n.a.

effluent 750 (11000) n.d. (5060) n.d. 9300 n.d. 790 (2000) n.a.

river 100 (910) 91 (211) n.a. n.a. 7 (17) n.a. n.a.

Tonalide influent 400 (450) 970 (1400) n.d. 1530 (1690) n.a. 545 (940) 200 (230)

(AHTN) effluent 90 (180) 140 (230) n.a. 160 (200) n.a. 410 (500) 40 (50)

Galaxolide influent 1500 (1800) 2800 (5800) 610 (1200) 3180 (3400) n.a. 1660 (2200) 750 (980)

(HHCB) effluent 450 (610) 470 (920) n.a. 500 (600) n.a. 1150 (1720) 120 (160)

n.d. non detectable (< detection limit); n.a. non available Influent concentrations in Germany are mean concentrations a) SMX including the human metabolite N4-acetyl-sulfamethoxazole

8


III. Development of analytical methods for sludge and urine

Analysis of PPCPs in sludge Analytical methods for the determination of PPCPs in sludge were developed for the following groups: estrogens, antibiotics, iodinated contrast media, acidic and neutral pharmaceuticals and musk fragrances. In general, the recoveries exceeded 70 %, whereas the quantification was improved by using surrogate standards for the respective groups. The LOQ were of 2 ng g-1 for the estrogens and ranged between 10-50 ng g-1 for the other PPCPs. The analytical methods have been used for analysing the sludge of batch-scale and pilot-scale experiments (determination of Kd values) with primary and secondary sludge as well as for the monitoring of whole WWTPs.

Enclosed a rough description of the methods for PPCP analysis in sludge:

• Estrogens: a) ultrasonic solvent extraction (methanol/acetone), b) GPC and silica gel clean up and c) silylation with MSTFA d) detection with GC/MS/MS. Limit of quantification (LOQ) was of 1.5 ng g-1. The method is appropriate for the determination of E1, E2 and EE2 in secondary activated sludge and digested sludge.

• Musk fragrances: a) ultrasonic solvent extraction (methanol/acetone) or accelerated solvent extraction (methanol, 100°C, 100 bar), b) SPE and silica gel clean up and c) detection with GC/MS. Limit of quantification (LOQ) was of 250 ng g-1. The method is appropriate for the determination of galaxolide and tonalide in secondary activated sludge and digested sludge.

• Acidic compounds: a) ultrasonic solvent extraction (methanol/acetone) b) SPE and c) detection by LC Tandem MS. Diclofenac, ibuprofen, hydroxy-ibuprofen, indomethacine, ketoprofen (antiphlogistics) and clofibric acid, bezafibrate (lipid regulators) can be analyzed. Limit of quantification (LOQ) was of 50 ng g-1.

• Neutral pharmaceuticals: a) ultrasonic solvent extraction (methanol/acetone) b) SPE and c) detection by LC Tandem MS. Carbamazepine, phenazone, propyphenazone, dimethylamino-phenazone, ifosfamide, cyclophosphamide, glibenclamide, pentoxifylline, caffeine and diazepam can be analyzed. Limit of quantification (LOQ) were achieved down to 20-50 ng g-1.

• Iodinated contrast media: a) ultrasonic solvent extraction (methanol/acetone) b) SPE and c) detection by LC Tandem MS. Iopromide, diatrizoate, iohexol, iopamidol, iothalamic acid, ioxithalamic acid and iomeprol can be determined. Limit of quantification (LOQ) was of 50 ng g-1.

• Antibiotics: a) ultrasonic solvent extraction (methanol/acetone) or accelerated solvent extraction (methanol-water mixture (50/50, v/v), 100°C, 100 bar), b) SPE and c) detection by LC Tandem MS. Several sulfonamides such as sulfamethoxazole and macrolides such as roxithromycin and clarythromycin can be analyzed in secondary and digested sludge. Limit of quantification (LOQ) was of 10 ng g-1.

Analysis of PPCPs in urine

A GC-MS method was adapted for the simultaneous determination of diclofenac, ibuprofen and carbamazepine in the matrix urine. The quantification range of the method was adapted to the concentrations of ecotoxicological studies (100-4000 µg L-1).

9


IV. Identification of oxidation products

A combination of freeze drying, semi-prep HPLC, GC-MS/MS and LC-MS/MS was used to assist in structural elucidation of ozonation by-products of carbamazepine and diclofenac. One product from each compound could be isolated in significant quantities for NMR analysis. Further details are reported in the chapter drinking water treatment.

• Ozonation of diclofenac has yielded an oxidation product 2-[2’,6’-dichlorophenyl)amino]-5-

hydroxyphenylacetic acid. This product was positively confirmed using GC-MS/MS as well as 1H-NMR, COSY and NOSY experiments. The NOSY experiment was essential to identify that the hydroxyl substitution was in the para position (from the amine). Ozone attacks diclofenac at the susceptible amine group but it’s known from literature that an aromatic hydroxylamine can undergo OH transfer to the para-position under acidic conditions.

• A scheme was developed to produce large quantities, > 10 mg, of a carbamazepine ozonation product. The chemical structure was extremely difficult to prove, even using mass spectrometry and NMR techniques. A 13C-NMR experiment as well as a 13C-spin echo gave the final indications to elucidate the chemical structure. The chemical shifts observed by 13C-NMR were compared to those of known components reported in literature. The oxidation product was identified as a quinoline type derivative; 1(2-benzaldehyde) (1H,3H)-quinazoline (BQD) (Figure. 1.1). As far as we know that chemical has never been reported in literature and thus, nothing is known about its toxicity or its behavior in drinking water treatment.

50 100 150 200 250m/z

0%

25%

50%

75%

100%

50 63 76 105 115

140

152

167

195

207 222

238

266281

Spect 152.582 min. Scan: 3155 Chan: 1 Ion: 747 us RIC: 1444455 BCBP 195 (312115=100%) dm_012f3.ms

NC

CN

O

O

H

CO

H

m/z =195A

B

m/z = 167

m/z = 238

C

50 100 150 200 250m/z

0%

25%

50%

75%

100%

50 63 76 105 115

140

152

167

195

207 222

238

266281


50 100 150 200 250m/z

0%

25%

50%

75%

100%

50 63 76 105 115

140

152

167

195

207 222

238

266281


50 100 150 200 250m/z

0%

25%

50%

75%

100%

50 63 76 105 115

140

152

167

195

207 222

238

266281


NC

CN

O

O

H

CO

H

m/z =195A

B

m/z = 167

m/z = 238

C

Figure 1.1: Schematic representation of three common fragmentation processes observed in the GC-MS spectrum of BQD.

Several oxidation products have been identified for 17α-ethinylestradiol and the natural estrogens E1 and E2 using LC Tandem MS and GC/MS/MS. Since all these products are rather polar (see Table 3.3) the LC/MS/MS technique was the crucial analytical method for identification. The confirmation of the products were done by using model substances such as 5,6,7,8-tetrahydro-2-naphthol (THN) and 1-ethinyl-1-cyclohexanol (ECH), which represent the reactive phenolic moiety and the ethinyl group of EE2. The oxidation products formed during the ozonation of the natural hormones Ε2 and E1 were investigated as well. Instead of an ethinyl group, E2 and E1 exhibit an alcohol and a keto group at the C-17 position. The experiments yielded the same two major products for E2 and E1. With LC-MS/MS they were identified as an adipic acid derivative and a di-(α-hydroxycarbolic acid) derivative which are also formed during the ozonation of EE2 (Table 3.3).

10


1.3.2 Wastewater Technology and Source Control I. Mechanisms relevant for the removal of PPCPs in municipal WWTPs The removal of PPCP during wastewater treatment can occur by means of the following mechanisms:

• Biological degradation: sludge age has shown to be a major factor influencing the palette of chemical structures being microbiologically transformed. The observed degradation rates of various compounds differ significantly without showing any evident correlation to specific molecular structure: currently no quantitative structure activity relationship (QSAR) can be identified. The observed removal rates vary form very fast (e.g. estradiol, paracetamol) to zero (e.g. carbamazepine, diatrizoate). Therefore the degradation of each compound has to be determined experimentally (i.e. the rate constant related to sludge concentration and sludge age).

• Sorption onto sludge issues in a removal of the sorbed share out of the water phase and into the sludge processing path. Sorption behavior can be estimated with the help of the sorption coefficient (Kd), a value depending mainly from characteristics of the compound as well as of the sludge. Currently no correlation of the observed Kd with literature value (e.g. octanol water partitioning KOW or partitioning to soil organic carbon KOC) could be found: besides hydrophobic also electrostatic interactions are relevant for sorption onto activated sludge. Nevertheless for the musk fragrances the high KOW correlated with a high Kd. Therefore the sorption coefficient has to be measured for each compound and for each sludge type (e.g. primary, secondary, digested). Concerning the elimination from the water phase of municipal wastewater, sorption can be neglected for compounds with a Kd ≤ 500 L kgSS-1.

• Stripping is not a relevant process for pharmaceuticals, since these exhibit a fairly good solubility and therefore a low gas-water-partitioning coefficient. WWTPs equipped with mechanical surface or coarse bubble aeration (e.g. membrane bioreactor) represent an exception, due to the higher amount of air getting in contact with the wastewater compared to fine bubble aeration: in this case volatile compounds (e.g. musk fragrances) can be stripped in significant amounts.

• Chemical oxidation: ozonation of the effluent has confirmed being a feasible polishing step for biologically treated wastewater with the potential of eliminating a wide variety of PPCPs.

Conclusion: Biological degradation and sorption are the main mechanisms for PPCP removal during municipal wastewater treatment. Ozonation is an interesting option for advanced treatment.

11


II. Biological degradation in the water line of WWTPs According to batch experiments, the degradation of pharmaceuticals can be described by pseudo first order degradation (observed degradation rates are found in Figure 1.2):

ibiolii CSSk

dtdC

⋅⋅= , (1.1)

Ci: soluble substance concentration of the compound i inside the reactor [µg L-1] ki,biol: kinetic constant for pseudo first order degradation [L gSS-1 d-1] SS: suspended solids concentration [gSS L-1]

Since the reaction rate constant kbiol is expressed per suspended solids concentration, it not only depends on the degradability of each specific compound, but also on the sludge composition. In this behalf the sludge age is assumed to take influence in three independent ways:

• Biodiversity of the active biomass: according to the specific growth rate, each species of microorganisms has its characteristic minimal sludge age (i.e. average residence time inside the reactor) required to allow the settlement of a stable population. For the elimination of a significant number of micropollutants ≥ 10d sludge age (nutrient removal) has shown to be crucial for biodegradation.

• Share of active biomass within the total suspended solids: the higher the sludge age, the more the sludge is being stabilized and correspondingly also the fraction of inert and inorganic matter increases. Therefore kbiol is expected to slightly decrease with increasing sludge age.

• Decrease of specific sludge production due to the increasing effect of sludge decay with increasing sludge age

Figure 1.2: Biological pseudo first order degradation rate constants kbiol observed in aerobic batch experiments run with activated sludge from plants with a sludge age ≥ 8d. In the case of several observations being available, error bars indicate variation range (minimum and maximum). The red line at kbiol 0.1 L gSS-1 d-1 indicates the limit below which no significant degradation can be expected in typical municipal wastewater treatment plants (Figure 1.3).

0.001

0.010

0.100

1.000

10.000

100.000

1'000.000

Azithro

mycin

Clarith

romyc

in

Erythro

mycin

N4-Ace

tyl-S

ulfam

ethox

azole

Roxith

romyc

in

Sulfam

ethox

azole

Diazep

am

Carbam

azep

ine

Acetyl

salic

ylic a

cid

Diclofe

nac

Fenop

rofen

Ibupro

fen

Naprox

en

Parace

tamol

ATHDAMI

Iohex

ol

Iomep

rol

Ioprom

ide

Ioxita

lamic

acid

Estrad

iol

Estron

e

Ethiny

lestra

diol

Bezafi

brate

Clofibr

ic ac

id

Fenofi

bric a

cid

Gemfib

rozil

Pirace

tam

Deg

rada

tion

cons

tant

kbi

ol [L

gSS

-1 d

-1] Antibiotic

Antid

epre

ssan

t

Antie

pile

ptic

Antiphlogistic Iodinated contrast agents Lipid regulator

Noo

tropi

cs

Estrogens

0.001

0.010

0.100

1.000

10.000

100.000

1'000.000

Azithro

mycin

Clarith

romyc

in

Erythro

mycin

N4-Ace

tyl-S

ulfam

ethox

azole

Roxith

romyc

in

Sulfam

ethox

azole

Diazep

am

Carbam

azep

ine

Acetyl

salic

ylic a

cid

Diclofe

nac

Fenop

rofen

Ibupro

fen

Naprox

en

Parace

tamol

ATHDAMI

Iohex

ol

Iomep

rol

Ioprom

ide

Ioxita

lamic

acid

Estrad

iol

Estron

e

Ethiny

lestra

diol

Bezafi

brate

Clofibr

ic ac

id

Fenofi

bric a

cid

Gemfib

rozil

Pirace

tam

Deg

rada

tion

cons

tant

kbi

ol [L

gSS

-1 d

-1] Antibiotic

Antid

epre

ssan

t

Antie

pile

ptic

Antiphlogistic Iodinated contrast agents Lipid regulator

Noo

tropi

cs

Estrogens

12


Figure 1.3 Relative amount degraded in reactor configurations typical for municipal wastewater treatment: 8 h hydraulic retention time, reactor divided in one two or three cascaded completely stirred compartments (CSTR; eq. 1.2 and 1.4). Suspended solids concentration of 3.5 gSS L-1 is typical for conventional activated sludge systems (CAS, SA=10-15d), while membrane bioreactors (MBR, SA=25-30d) often run with up to 10 gSS L-1. A high degree of removal can be expected for compounds with kbiol values ≥ 1 L gSS-1 d-1.

0.01 0.1 1 10 500

10

20

30

40

50

60

70

80

90

100

Degradation constant kbiol [L gSS-1 d-1]

Rel

ativ

e am

ount

deg

rade

d [%

]

MBRSA 25-30dSS 8g L-1

CASSA 10-15dSS 3.5g L-1

1 CSTR2 CSTR3 CSTRPlug flow

In a plug flow or a batch reactor the relative amount degraded is (see also eq. 1.3):

(1.2) SASPkHRTSSk

ini

outi biolibioli eeCC ⋅⋅−⋅⋅− == ,,

,

,

Ci,in influent substance concentration of the compound i [µg L-1] Ci,out final substance concentration of the compound i [µg L-1] HRT hydraulic retention time of the whole reactor or duration of the batch [d] SP specific sludge production per amount of wastewater treated [gSS m-3

wastewater] SA sludge age [d] Equation 1.2 and 1.4 imply that SS·HRT = SP·SA, what is comprehensible according to the sludge mass balance:

QSPSASSV ⋅⋅=⋅ (1.3) V reactor volume [m3] Q flow rate of the treated wastewater [m3 d-1]

In the case of a cascade of completely stirred reactors of equal volume, the relative concentration in the effluent is calculated as follows:

n

bioli

n

bioliini

outi

nSASPk

nHRTSSkC

C

⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜

⎝

⎛

⋅⋅+

=⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜

⎝

⎛

⋅⋅+

=

,,,

,

1

1

1

1 (1.4)

n number of cascaded compartments [-]

13


In Figure 1.3 the calculated removal (eq. 1.2 and 1.4) is plotted in function of the kinetic degradation constant for typical reactor configurations:

• kbiol < 0.1 [L gSS-1 d-1]: no substantial removal due to biological degradation • 0.1 < kbiol < 10: degree of removal strongly dependent on reactor configuration • kbiol > 10: more than 95% removal by biological degradation

Conclusion: Biological degradation can be described by pseudo first order kinetic. Results of full scale plant

The sampling of the full scale plants generally is in good agreement with the measured kinetic values, which implicitly means that the conjugated fraction in the (full scale) influent has no major impact on the apparent removal (except for sulfamethoxazole and estrogens): otherwise significant differences are to be expected between lab scale experiments (only non-conjugates species spiked) as compared to full scale plants (containing the naturally occurring conjugated share). The following results originate from studies of full scale plants. Unless stated otherwise, no degradation in the anaerobic compartment was observed.

• Bezafibrate: > 95% removal by degradation at sludge age ≥ 5d, corresponding to a kbiol of 4 to 10 L gSS-1 d-1; no removal at 1d sludge age

• Carbamazepine is not removed in wastewater treatment, corresponding to a kbiol <0.1 L gSS-1 d-1.

• Diatrizoate: according to various lab experiments no aerobic removal is occurring (kbiol <0.1 L gSS-1 d-1), but clear anaerobic removal could be shown (estimated kbiol 0.1 – 1 L gSS-1 d-1).

• Diazepam: no significant removal in lab scale, corresponding to a kbiol <0.1 L gSS-1 d-1; compound not found in full scale plants

• Diclofenac: removal occurred mainly in the aerated compartment and ranged between 15-40% for nutrient removing plants, corresponding to a kbiol of 0.25±0.2 L gSS-1 d-1. No significant removal at SRT < 2d.

• Estradiol and Estrone: >95% removal by degradation at nutrient removing sludge age. The corresponding kbiol value cannot be estimated for the full scale sampling, since the data points where below quantification limit. At sludge age <8d significantly reduced removal was seen for estrone.

• 17α-Ethinylestradiol: no removal at sludge age <4d; biological degradation >90% is achieved with nutrient removing sludge age corresponding to aerobic removal rates kbiol of 5 to 10 L gSS-1 d-1; the amount removed by sorption onto sludge was <5% of the influent load

• Ibuprofen and hydroxy-ibuprofen: >90% removal by biological degradation at sludge age ≥ 5d corresponding to a kbiol of 23±10 d assuming that the removal is occurring mainly in the aerobic reactor volume; at 1d sludge age no significant removal was observed

• Roxithromycin: Removal of 0 to 60% was seen at sludge ages ≥ 5d, corresponding to a kbiol of <0.7 L gSS-1 d-1. No significant removal at SRT < 2d.

14


For further compounds contradictory results have been obtained which can only be solved by further experiments:

• Galaxolide and Tonalide: different removal rates were seen for these strongly sorbing compounds: according to some campaign a removal of 70% to 90% by degradation is seen, while other campaigns show no substantial degradation, but 40% to 65% removal by sorption onto the sludge. It is not clear whether the strong hydrophobicity of the compounds cause some analytical problems (possible artifacts during sampling or analytic) or if the sorption characteristic of the compound has not been accounted for correctly (e.g. dependence of the Kd on the sludge type, leading to errors in the mass balance).

• Iopromide: on the Polish and Swiss full scale plants (plant size of 55’000 to 100’000 population equivalents) a great removal variation was observed, possibly due to the fact that according to the therapeutic dosage, the influent load was only due to very few persons excreting Iopromide in the catchment area. The correspondingly high fluctuations in the influent load may therefore be sampled inaccurately. A clear removal in the anaerobic compartment could be shown in lab experiments (estimated kbiol of 0.5 to 2 L gSS-1 d-1)

• Sulfamethoxazole: a mass flux of this compound can be set up only if the human excretion metabolite (N4-Acetyl-Sulfamethoxazole) is included; the removal in batch experiments and also on laboratory and full scale plants have shown high variation between 0 and 90%, corresponding to kbiol of 0.1 to 10 L gSS-1 d-1.

Conclusion: The kinetic degradation rates found in batch experiments can be applied to full scale plants; for partially degraded compounds some accuracy limitations for modeling are probably due to influencing factors not accounted for (e.g. sludge loading, specific sludge composition).

III. Fate of single compounds in anaerobic sludge digestion Two pilot-scale (10 L) anaerobic reactors, mesophilic (37ºC) and thermophilic (55ºC), have been operated as conventional anaerobic digesters. The feeding for both reactors consists of a mixture of primary and secondary sludge (70:30 v/v; suspended solids concentrations between 20 – 30 gSS L-1). After three months of operation, the digesters started to be fed with slugs previously spiked with PPCPs. Since then, three stages of operation of each digester have been performed. The mesophilic digester was operated with a HRT of 30, 20 and 10 days; and the thermophilic one was operated with a HRT of 20, 10 and 6 days. Different liquid and sludge samples were taken from each reactor to determine the removal of PPCPs at each stage of operation. Significant anaerobic biological degradation (>80%) is ascertained only for naproxen and sulfamethoxazole. Further conclusions drawn from the anaerobic digestion pilot plant may be biased by significant quantitative uncertainty:

• Galaxolide: Spiking experiments showed adsorption (65-85%) according to Kd. Therefore, no

significant removal can be expected for raw sludge (soluble and sorbed phase in equilibrium). • Tonalide: Same behavior as galaxolide. • Carbamazepine: Data accuracy does not allow to discriminate between no or a partial removal

(0 to 60%). • Diazepam: See Carbamazepine. • Ibuprofen: Medium elimination in both digesters (20-45%). • Naproxen: Very high removal by degradation in both mesophilic (80%-85%) and thermophilic

(80-95%) range, independently of the HRT.

15


• Diclofenac: no clear results were obtained for this compound since in some cases the removal could not be quantified due to the high deviation of the data. In those cases where it was possible, the efficiencies ranged between 25% and 75%.

• Iopromide: better removal in thermophilic (30-80%) than in mesophilic range (10-40%). • Sulfamethoxazole: Very high removal by degradation in both mesophilic (>95%) and

thermophilic (85-95%) range, independently of the HRT. • Roxithromycin: contradictory results obtained

Sludge adaptation to PPCPs (several sludge retention times), chemical pre-treatment (pH 12 for 24h) or thermal pre-treatment (heating to 130°C for 1h) had only minor impact on the observed removal.

Conclusion: Some PPCPs are degraded to a significant extent during anaerobic sludge digestion.

IV. Sorption coefficients of PPCPs onto sludge For municipal wastewater treatment (typical sludge production between 200 and 400 gSS m-3

wastewater corresponding to 400 – 700 g COD m-3) only substances with Kd values >500 L kgSS-1 partition significantly (>10%) onto the sludge (Figure 1.4). Although a quantitative prediction of the Kd value based on the molecular characteristics of a compound is currently not possible, according to the available data only hydrophobic compounds (musk fragrances, estrogens, diclofenac) and positively charged ionic substances (norfloxacin) have shown Kd values high enough to make sorption a relevant process for compound elimination. Accordingly the following two mechanisms are assumed to be relevant for sorption onto particulate matter in municipal wastewater treatment:

• Absorption: Hydrophobic interactions of the aliphatic and aromatic groups of a compound with the lipophilic cell membrane of the micro-organisms and the lipid fractions of the sludge.

• Adsorption: Electrostatic interactions of positively charged groups of chemicals with the negatively charged surfaces of the micro-organisms

For a compound i in equilibrium conditions, the concentration sorbed onto sludge (Ci,sorbed) is assumed to be proportional to the concentration in solution (Ci,soluble):

solubleiidsorbedi CSSKC ,,, ⋅⋅= Ci,sorbed concentration of the compound i sorbed onto sludge [µg L-1] Kd,i sorption coefficient of the compound i [L kg-1] SS suspended solids concentration in raw wastewater or production of suspended

solids in primary of secondary treatment, per L of treated wastewater [kg L-1wastewater]

Ci,soluble soluble concentration of the compound i [µg L-1]

It is important to note, that for estimating the sorbed amount in municipal wastewater treatment, the relevant SS value is not the suspended solid concentration measured in the mixed liquor, but the amount of sludge generated per unit of wastewater treated. This issues from the fact of having a sludge retention time significantly higher than the hydraulic retention time; therefore the sludge amount being re-circulated during wastewater treatment be assumed to be in equilibrium with the compound concentration in solution; only the newly generated sludge being available for sorption.

(1.5)

16


According to eq. 1.5, the proportion of sorbed quantities in a fully mixed tank can be predicted using the following formula:

Currently no correlation could be found between the observed Kd values for sludge of municipal wastewater treatment with literature values for the specific compounds (e.g. KOW or KOC values). Nevertheless for the musk fragrances the high KOW correlated with a high Kd. Since the KOW describes the hydrophobicity of compounds, it is not well suited to discriminate between the sorption characteristics caused by electrostatic interactions of various polar compounds in aquatic solution. Before new modeling approaches of the sorption constants of pharmaceuticals (considering the properties and the polarity of the functional groups of a molecule) are available, the Kd value has to be assessed for each compound and each sludge type by experimental measurement (Table 1.3).

Figure 1.4: Share of a PPCP sorbed onto sludge as a function of the sorption coefficient Kd and the specific sludge production (expressed per m3 of treated wastewater). Compound names indicate typical Kd values for secondary sludge.

(1.6) SSK

SSKCC

C

id

id

solubleisorbedi

sorbedi

⋅+⋅

=+ ,

,

,,

,

1

17


Table 1.3: Sorption coefficients for sludge of municipal wastewater treatment, measured in batch experiments and octanol-water partitioning coefficients (according to Ternes et al. 2004).

Primary sludgeKd [L kgSS

-1] Secondary sludge

Kd [L kgSS-1]

log KOW

Acidic pharmaceuticals I. II. III.

Diclofenac 459±32 16.0±3.1 4.6 Ibuprofen -- (< 20 ) 7.1±2.0 3.5 Clofibric acid -- (< 30 ) 4.8±2.5 2.57 Neutral pharmaceuticals Ifosfamide 21.8±13.8 1.4±0.4 0.86 Cyclophosphamide 55.4±19.6 2.4±0.5 0.63 Carbamazepine -- (< 20 ) 1.2±0.5 2.45 Diazepam 43.9±26.1 21.1±7.6 2.82 Musk fragrances Galaxolide (HHCB) 4919±2073 1807±534 5.9 Tonalide (AHTN) 5299±1905 2372±958 5.7 Iodinated contrast media Iopromide -- (<5) 11±1 -2.33 Estrogens 17α−Ethinylestradiol 278±3 349±37 3.9

Conclusion: Removal by sorption onto suspended solids is an important mechanism for hydrophobic and positively charged compounds. V. Stripping into air The amount of a compound being stripped from the water phase into the gas phase during an aeration process, depends mainly on the gas-water exchange (i.e. amount of air getting in contact with water, type of aeration) and from the liquid-gas partitioning coefficient or Henry coefficient in the case of air-water partitioning.

(1.7)

TRCp

Ksolublei

iHi ⋅⋅=

,,

Ki,H Henry or air water partitioning coefficient of the compound i [-] pi partial pressure in the gas phase [Pa] R universal gas constant; 8.314 [J Mol-1 K-1] T Temperature [K] Ci,soluble soluble concentration of the compound i [µg L-1]

A Henry coefficient >3·10-3 [-] is required for significant stripping in a bioreactor with fine bubble aeration. Pharmaceuticals normally have values <10-5, since these are mostly compounds meant to take effect in a aqueous environment (e.g. blood), and are therefore rather hydrophilic. With Henry coefficients in the range of 5·10-3 the musk fragrances may be stripped to some extent in modern

18


wastewater treatment plants (Figure 1.5). In formerly used mechanical surface or coarse bubble aerators significantly higher stripping is expected to occur. In modern municipal wastewater treatment plants these are not used anymore, due to the low oxygen input efficiency, the high energy consumption and also due to the higher aerosol formation. Also membrane bioreactor use higher amount of air (up to 25 m3

air m-3wastewater) as compared to conventional activated sludge treatment (6

– 10 m3air m-3

wastewater). This is due to coarse bubble aeration used for scouring the membrane surface, and also due to the higher air requirement caused by the high sludge concentration.

The relative share stripped can be calculated with the Henry coefficient value and the specific

aeration rate. For a fully mixed reactor this is: airi

airistripping QH

QH⋅+⋅

=1

η

ηStripping relative amount stripped [-] Hi Henry coefficient of the compound i [-] Qair amount of air required [m3

air m-3wastewater]

10-3

10-2

10-1

100

0

10

20

30

40

50

60

70

80

90

100

Henry coefficient KHi [-]

Stri

pped

sha

re o

f i [%

]

5

10 15 25 MBR

Aeration m3air m

-3w w

Chl

orob

enze

ne

Dim

ethy

leth

er

Bip

heny

l

Gal

axol

ide,

Ton

alid

e, D

ioxi

n (D

D)

Non

ylph

enol

Figure 1.5: Share of a PPCP stripped during aeration as a function of the gas/water partitioning coefficient KH,i and the amount of air required (expressed per m3 of treated wastewater). Compound names indicate some typical Henry coefficients for chemicals. Pharmaceuticals show values <10-5.

Conclusion: Stripping is not relevant for PPCP removal in state of the art municipal wastewater treatment.

19


VI. Reactor configuration • Sludge age has an impact on the specific degradation activity by three independent ways: by

influencing a) the biodiversity, b) the content of inert matter in the sludge and c) the sludge production. Biodiversity (influence on kbiol): For several compounds it was found that a minimal sludge age exists, beyond which a partial or total removal by degradation occurs. Therefore, the palette of chemical structures being microbiologically transformed broadens with increasing sludge age. A significant increase is seen between COD-removing plants (sludge age ≤ 4d) and nutrient removing plants (sludge age of 10d – 15d). Highly loaded wastewater treatment plants (sludge age of 1d – 4d) show none or only slight removals of the investigated PPCPs. Nevertheless, such plants would meat the required effluent quality for non sensitive areas within the EU. At sludge ages beyond 15d the number of compounds being biologically degraded increases further but the increase is less prominent; some compounds are not removed even at sludge ages in the range of 100d. Inert sludge content (influence on kbiol): An increase in sludge age is paralleled by a rise of the inert fraction in the sludge, due to the accumulation of decay products. Since kbiol is expressed per unit of sludge dry mass concentration, its value decreases with increasing content of inert matter (organic and inorganic). Sludge production Due to the increased importance of sludge decay and mineralization with increasing sludge age the observed sludge production per volume of wastewater (SP) decreases.

• Hydraulic retention time and wastewater dilution: The formula 1.2 and 1.4 show that the hydraulic retention time (HRT) respectively the specific sludge production (SP) have a significant impact on the removal (i.e. on the term SS·HRT or SA·SP): this explains the observed lower removal during rain events. The consequence hereof is that for maximizing micropollutant removal the biological treatment is to be done at the highest possible PPCP concentration. Accordingly reducing rain and extraneous water input (i.e. infiltration) into the sewer (e.g. groundwater infiltration) as well as locating biological treatment as close as possible to PPCP sources (source separation and source treatment) bear significant efficiency advantages.

• Cascades: Figure 1.3 shows that the number of compartments in series significantly increases the biological removal for all compounds with kbiol >0.5 L gSS-1 d-1.

• Membrane bioreactor (MBR): For most PPCPs the removal achieved with a micro- or ultrafiltration membrane bioreactor is comparable to a conventional activated sludge treatment plant run at a comparable sludge age. For those compounds requiring an increased sludge age for being removed, MBRs typically perform better since they are often run at 20 to 50 days of sludge age, while conventional plants are mostly run at 10 to 15 days. The smaller sludge particle size found in MBRs (median diameter >50µm as compared to 300-500µm) may issue in higher kinetic degradation constants for those compounds where diffusion is a limiting factor (i.e. kbiol >100 L gSS-1 d-1); nevertheless for practical reasons this fact is of minor importance, since the main focus is on increasing the removal efficiency of recalcitrant compounds and in this case mass transfer by diffusion is not a factor limiting degradation. Figure 1.3 is not meant to suggest that MBR are expected to perform a better PPCP removal compared to conventional activated sludge systems, since for several compounds significantly lower kbiol for MBR sludge were found, issuing in the comparable removal observed during the full scale plant sampling (in spite of the higher sludge age).

• Biofilter (fixed bed reactor): Nitrogen removal reactor with biofilm growth performed comparably to the conventional activated sludge reactor run in parallel (the monitoring was done on two lanes of a single full scale plant operated in parallel but equipped with different reactor

20


types). Since in this type of reactor the HRT is significantly lower (in this specific case 35 min on average, as compared to typically 8 – 12 hours in conventional activated sludge plants) this result indicates that for biofilters the lower HRT is approximately compensated by the higher bioactive sludge concentration per m3 of reactor (HRT·SS = constant). In other words, for a given influent composition the PPCP removal efficiency is not directly interrelated with the hydraulic retention time of the different treatment system if they reach the same efficiency in nutrient removal.

Conclusion: Sludge age, number of cascaded compartments and dilution of the wastewater are factors with major influence on biological degradation efficiency.

VII. Ozonation of pure water and treated wastewater

One of the first crucial outcomes of the EU-project POSEIDON in preventing PPCP contamination of receiving waters is the establishment of ozonation for treated wastewater (see also section 1.3.3/case study 1). A kinetic data base on the oxidation of PPCPs with ozone and OH• radicals was established in lab-scale experiments (see also section 1.3.4). It exhibited that ozone-based oxidation processes have a high potential for the elimination of many crucial target PPCPs. These predictions were confirmed by pilot scale experiments done with the effluent of a wastewater treatment plant (performed in co-operation with WEDECO).

An important highlight of the ozonation was the effective oxidation/degradation of three major endocrine disrupters (17α-ethinylestradiol, 17β-estradiol and estrone), which probably lose most of their estrogenic potency. Thus, ozonation will drastically reduce estrogenic effects on fish caused by discharging treated municipal wastewater into rivers and streams. Furthermore, it can be predicted that the potential for the formation of resistant bacterial strains is lowered significantly because antibiotics were no longer detected in the ozonated wastewater.

Ozonation is compared with conventional activated sludge treatment a rather cheap but energy consuming process: a total requirement <0.04€ and <0.3kWh per m3

wastewater is estimated (for activated sludge wastewater treatment typically 0.5€ and 0.3kWh are required). It might therefore be a potential measure to improve PPCP removal in urban regions with a high portion of treated wastewater in receiving surface water or with direct infiltration.

The results with Fenton's reagent in laboratory scale experiments showed a very low extent of PPCP removal even under drinking water treatment conditions with low DOC and low alkalinity (scavenging the oxidizing agents; see section 1.3.4). Therefore, Fenton's reagent is not suitable for PPCP removal which leads to the decision that the corresponding full-scale experiments were cancelled. A good removal can be seen at acidic conditions (pH 3), but this is of little interest for practical application in municipal wastewater treatment.

Since for a cost effective removal of PPCPs only a partial oxidation of the compounds is achieved, the biological degradability (biological oxygen demand) of the effluent is increased by the proposed post-ozonation. Therefore, an additional filter is recommended to degrade transformation products if ozone treated wastewater is not passing a percolation zone during infiltration.

Conclusion: Ozonation of treated effluent substantially reduces the PPCP content at feasible cost.

21


VIII. Estimation of PPCP occurrence in municipal wastewater

• The concentrations of pharmaceuticals estimated according to medicine sales and wastewater production of the analyzed Polish plants are comparable with the real measurements done in wastewater.

• In a plant near Vienna two of the selected PPCPs (sulfamethoxazole and diazepam) could not be detected. Diazepam, diclofenac, roxithromycin and 17α-ethinylestradiol were not found in the wastewater of Santiago de Compostela, Spain. Diazepam was not found in Austrian and Swiss wastewater treatment plants, what can be explained by very low consumed quantities (about 125 kg/year in Austria).

Conclusion: A rough prediction of the PPCP concentration in raw wastewater can be done according to the sold PPCP quantity.

IX. Modeling of PPCPs in WWTPs

For modeling the PPCP removal the following processes have to be considered:

• Biologic degradation: for those compounds that proved to be biodegradable pseudo first order

kinetics suited the removal in batch experiments. Accordingly the removal can be modeled with equations 1.2 or 1.4, depending on the reactor configuration. Biological degradation in municipal wastewater treatment is relevant only for compounds with a kbiol >0.1 L gSS-1 d-1. The reaction constant kbiol varies over more than five orders of magnitude, mainly due to differences in molecular structure of the compounds. Further a certain variability of the kinetic constant kbiol (up to a factor of 4; e.g. gemfibrozil or fenoprofen) on reactor configuration (e.g. reactor cascade and floc size), sludge age and loading rate was seen. Therefore, mainly for compounds being only partially degraded (0.1<kbiol<50 L gSS-1 d-1) an accurate modeling requires the determination of kbiol on site, or an increased modeling uncertainty must be accepted.

• Sorption onto sludge is assumed to be fast compared to biologic degradation or hydraulic retention time. Therefore equilibrium conditions according to equation 1.5 are assumed. Accordingly also for dynamic modeling, the sorption process can be considered as a static equilibrium. For compounds with Kd value <100 L kgSS-1 sorption is not a relevant process, and therefore omitted for modeling the PPCP removal from the water phase (due to the significantly higher suspended solids concentration, for the sludge digestion the limit of relevance is around Kd <1 L kgSS-1).

For several compounds (e.g. sulfamethoxazole or estrogens) modeling is only possible, if conjugated species and parent compounds are considered as well. Conclusion: Main aspects for modeling PPCPs are available (pseudo first order degradation and sorption).

22


1.3.3 Indirect discharge of treated wastewater into unsaturated soil

I. Irrigation onto agricultural fields

In the current study the irrigation of treated municipal wastewater on arable land was investigated in Braunschweig, Germany. In Braunschweig an irrigation area was selected where treated wastewater is irrigated the year around for 50 years (Figure 1.5). In the winter time only the treated wastewater of the sewage treatment plant of Braunschweig is irrigated, while in the summer time additionally digested sludge is mixed with the treated wastewater prior to irrigation. The mechanical treatment of the current Braunschweig WWTP consists of a screen, an aerated grit-removal tank and a primary clarifier. The primary sludge collected in the primary clarifier is pre-thickened and then pumped into the digester (first process: thermophilic, 55°C, 4 d retention time; second process: mesophilic, 37°C, 16 d retention time). The primary effluent is directed via mixing basins to the activated sludge system for biological phosphate removal, denitrification and nitrification. After settling in the secondary clarifier, the activated sludge is returned via mixing basins to the inlet of the activated sludge tank. The secondary effluent is irrigated on agricultural fields. The activated sludge system is operated with a solids retention time of 11-13d, which is typical for a nitrifying plant with simultaneous denitrification.

In the case study several lysimeters in the agricultural fields were monitored at three different depths and groundwater probes with regard to the occurrence of 52 PPCPs (e.g. betablockers, antibiotics, lipid regulators, antiphlogistics, carbamazepine, diazepam, psychiatric drugs, musk fragrances, ICM and estrogens). Four intensified sampling campaigns were performed from spring to the fall within one continuous vegetation period to investigate the infiltration of PPCPs into the groundwater and to elucidate differences in pollution due to irrigation of treated wastewater with and without stabilized digested sludge. The following samples were measured for PPCPs: raw water, treated wastewater used for irrigation, groundwater from six wells within the area of a high irrigation intensity, four lysimeters at different depths (0.40, 0.80, 1.2 m) within the irrigation fields (low and medium influenced by irrigation). The analytical methods used are described above.

Figure 1.5: Scheme for the irrigation filed in Braunschweig with the

included sampling sites.

Irrigation (high intensity)Irrigation (low intensity)

Groun

dwat

er

Flow

dire

ctio

n

lysimeter Dephts of the wells ≈ 12-15 m, saturated zone: 1.5-2 m

Road B 214

path

path

wind shielding hedge

wind shielding hedge

lysimeterAlysimeter C

lysimeter D

Sam

plin

gro

w II

Sam

plin

gro

w I

Well 1

Well 2

Well 3

well 4

Well 5

Well 6lysimeter B

Field

23


Results Most of the selected PPCPs were not detected in any of the lysimeter and groundwater samples, although they were present in the treated wastewater irrigated onto the fields. In the original WWTP effluent, typically 5 antibiotics (0.34-0.63 µg L-1), 5 betablockers (0.18-1.7 µg L-1), 4 antiphlogistics (0.10-1.3 µg L-1), 2 lipid regulator metabolites (0.12-0.13 µg L-1), the antiepileptic drug carbamazepine (2.1 µg L-1), 4 ICM (1.1-5.2 µg L-1), the natural estrogen estrone (0.015 µg L-1) and 2 musk fragrances (0.1-0.73 µg L-1) were detected by LC-electrospray tandem MS and GC/MS/MS. ICM, derived from radiological examinations, were present with the highest concentrations (diatrizoate: 4.6 µg L-1, iopromide: 5.2 µg L-1). In the respective groundwater and lysimeter samples primarily ICM (diatrizoate, iopamidol), carbamazepine and the antibiotic sulfamethoxazole were detected up to several µg L-1, while the acid pharmaceuticals, musk fragrances, estrogens and the betablockers were sorbed or transformed by passing the top soil layers. Due to the absence of estrogens, estrogenic effects should totally disappear after irrigation. Conclusion: During irrigation and soil passage most of the PPCPs (>80%) are sorbed or degraded. However, the irrigation can lead to a pollution of groundwater with ICM and selected pharmaceuticals such as carbamazepine and sulfamethoxazole independent whether digested sludge were mixed to the wastewater irrigated or not. Oxidative Treatment to avoid a groundwater contamination By applying 10-15 mg L-1 ozone (contact time: 9 min), all the pharmaceuticals investigated as well as musk fragrances (HHCB, AHTN) and estrone were no longer detected. However, ICM (diatrizoate, iopamidol, iopromide and iomeprol) were still present in appreciable concentrations. Even with a 15 mg L-1 ozone dose, the ionic diatrizoate only exhibited removal efficiencies of no higher than 14 %, while the non-ionic ICM were oxidized to a degree of higher than 80 %. Advanced oxidation processes (O3/UV-low pressure mercury arc, O3/H2O2) did not lead to significantly higher oxidation efficiency for the ICM than ozone alone. A highlight of the ozonation was the effective oxidation/degradation of the three major endocrine disrupters (EDCs) 17α-ethinylestradiol (EE2), 17β-estradiol (E2) and estrone (E1), which lose most of their estrogenic potency (see below). Conclusion: Ozonation or AOPs are able to substantially reducing the contamination of groundwater prior to irrigation of treated wastewater with reasonable costs.

II. Infiltration of treated wastewater into unsaturated soil

During a period between April 2000 and June 2001 samples were taken from the investigated area with an interval of app. 2 month resulting in 7 samples for all sampling sites. Sampling implements 5 sites associated to different treatment steps of the waste water treatment plant and 7 groundwater sites (Table 1.4) including two sites with uninfluenced groundwater. The inflow of the WWTP and the effluent after secondary clarification were collected as 24 h composite samples. Groundwater was sampled from groundwater probes exchanging the volume of the probes three times before sampling.

24


Table 1.4: Sites sampled in this study and their characteristics

Sampling site Abbreviation Typ* Flow time**(d)

Content of treated wastewater(%)

Reference site withuninfluenced ground water uninf gw -50 0

Inflow W W TP Inflow wwtp (-) 100Effluent W W TP Effluent wwtp (10) 100Polishing Lagoon Pol.Lag. wwtp (10) 100Garvel Filter Gr.Filter wwtp (2) 100Sicker Biotope Sicker wwtp (5) 100Groundwater probe hp1 gw 18 100Groundwater probe hsr4 gw 25 100Groundwater probe hp3 gw 75 91Groundwater probe hp2 gw 105 72Groundwater probe hp4 gw 140 76

*gw = groundwater; WWTP = wastewater treatment plant; ** mean values related to infiltration in sicker biotop as t = 0; mean hydraulic retention time in treatment steps within WWTP in brackets

The investigation area is located in the eastern part of Austria. The region’s settlements are characterized by rural structures concentrating on agriculture and in some regional parts on viniculture. Regarding the geo-morphological situation, the region is characterized by glacial sediment deposits resulting from an intensively braided tributary system of the former River Danube basin. The irregular but frequent rearrangements of the tributaries caused a very heterogeneous sediment structure with gravel as its predominant deposit. The high amount of sand components is characteristic. The predominant soil type according to OECD TG 106 is soil type 1. The extent of the aquifer within the investigation region varies between 4 m to 14.5 m. The mean depths to groundwater within the region can be classified between 0.5 m and 6 m. Mean annual precipitation is about 550 mm. Evaporation often exceeds the precipitation rate during the summer months which causes a very low regeneration rate of groundwater. The connected WWTP is designed for 7,000 population equivalent (p.e.) and provides activated sludge treatment including intermittent nitrification/denitrification and phosphorus precipitation with FeCl3. The WWTP consists of screen and grit chamber, two aeration tanks (V = 2 x 1,546 m3) and two secondary clarifier (V = 2 x 949 m3, A = 2 x 304 m2) for final sedimentation. Post treatment steps consist of a polishing lagoon (V = 3,000 m3), a gravel filter (A = 500 m2, h = 1.05 m) and three infiltration ponds (total A = 1,500 m2, h = 1.5 m) where the treated wastewater is infiltrated into the soil-aquifer system. The actual mean load of the facility is about 2,450 p.e. (1 p.e. = 120 g COD d-1), the maximum loading as an average value of two weeks about 4,350 p.e. The excess sludge is removed very infrequently what effects a very high sludge retention time (SRT20°C) > 25 d, depending on the seasonal loading.

A groundwater flow model was calibrated for use in the frame of the project allowing estimating hydraulic flow times in the underground for the sampled groundwater probes. Obtained flow times are summarized in Table 1.4. In addition the dilution of the infiltrated treated wastewater by uninfluenced ground water was calculated using boron as a conservative tracer (Figure 1.6). The boron concentrations in the course of the WWTP exhibited no change, while the measured concentrations in the ground water were reduced due to dilution by non polluted ground water. Postulating a persistent behavior of boron in the underground, dilution factors for the ground water sampling sites were calculated (Table 1.4). The calculated dilution factors were used to normalize the data measured (conventional parameters as well as PPCPs) in order to consider only the reduction of concentrations due to adsorption or degradation processes in the saturated zone but to exclude a decrease due to dilution. Both considerations (regarding and neglecting dilution) are necessary for different concerns.

25


Figure 1.6: Groundwater flow model based on boron as tracer (portion of treated wastewater/dilution by uninfluenced groundwater; Grey area: borders of model uncertainties)

Results Diclofenac is a substance removed to a very limited extent in WWTPs (section 1.3.1): low removal rates of about 14 % are observed. Within the next treatment step, the polishing lagoon, a tremendous decrease occurs, that cannot be explained only by adsorption and biodegradation: photo degradation may be a possible reason. A further decrease of the concentration can be observed in the infiltration pond, where also UV irradiation is possible. During the first 25 days of subsurface flow no additional removal is observed. Nevertheless, after additional 50 days the concentrations were below limit of detection (LOD), indicating a complete removal within the saturated zone.

Bezafibrate shows a very good removal during wastewater treatment with 99 % of the inflow concentration being removed in the aeration tank. During post treatment steps as the polishing lagoon additional removal down to limit of quantification (LOQ) was observed, so finally a total decrease in concentration of 99.5 % was observed in our study. During groundwater flow no significant additional removal could be observed and concentrations were around the LOQ. Only after 140 days of flow time all measured concentrations were below LOD. Due to low concentration already in the water infiltrated, no removal rates in the saturated zone are given.

Ibuprofen shows a similar behavior as bezafibrate. Over 99% of the inflow concentration is removed during wastewater treatment. Post treatment steps show additional removal beneath LOQ. As the concentration in the infiltrated water already is below the LOD in most cases, no conclusion on the subsurface behavior of Ibuprofen can be given.

Diazepam: For diazepam the inflow concentrations of the investigated treatment plant already are in the range of LOQ, the effluent concentrations between LOQ and LOD. Even in the post treatment steps no significant change in concentrations was found. In the groundwater no change during the first 25 days of subsurface flow can be observed, but all data were below LOD after additional 50 days.

Carbamazepine passes the WWTP without any significant change in concentration, since only up to 10 % were the difference between inflow and effluent. Post treatment steps led to an additional 5 % loss. During subsurface flow a very slow process could be found resulting in about 30 % removal compared to the inflow after 100 days. It is not clear whether degradation, sorption or both are responsible for the dissipation of the compound.

26


Tonalide and Galaxolide show a very similar behavior during wastewater treatment as well as during groundwater flow; whereas galaxolide is measured in double the concentrations levels than for tonalide in all sampling sites. Both show removal rates of approximately 80 % during wastewater treatment mainly due to adsorption due to their high adsorption coefficient. Compared to the inflow concentrations additional 10 % are removed in the post treatment steps, whereas no additional removal in the saturated zone is observed even after 140 day of flow time.

For Iopromide only low influent concentrations were observed, because there are no hospital and radiological practices providing X-ray examinations in this rural catchment area. Nevertheless, concentrations up to 120 ng L-1 iopromide can be found. In all samples of the effluent investigated concentrations were below LOD as in all post treatment steps. In groundwater concentrations in the range of LOQ were observed for all sites and flow times. This may indicate a removal of iopromide during wastewater treatment but hardly any additional removal in the groundwater.

Roxithromycin: For the antibiotic roxithromycin a removal rate of 46 % was calculated within the WWTP. Post treatment steps show no consistent picture, but regarding further concentrations in the groundwater after a flow time up to 25 days, additional removal for the post treatment steps can be stated with additional 25 %. Similar to diazepam concentrations remained unchanged until 25 days of subsurface flow but then decreased below LOQ.

Sulfamethoxazole: The inflow concentrations of the investigated WWTP are already in the range of the LOQ, the effluent concentrations between LOQ and LOD, so no information on the behavior in the saturated zone can be given.

Conclusion During subsurface flow in the saturated zone the PPCPs monitored showed similar behavior with regard to removal from the aqueous phase. Acidic drugs such as diclofenac, bezafibrate and ibuprofen that are removed easily during wastewater treatment are subject to additional removal during post treatment steps like polishing lagoon, gravel filter or infiltration pond. On the other hand, neutral substances such as diazepam and carbamazepine that hardly show any removal during wastewater treatment remain stable during post treatment steps as well as in the groundwater. The polycyclic musks tonalide and galaxolide exhibited a significant removal during wastewater treatment and post treatment steps, but no significant further reduction during groundwater flow.

Regarding the general behavior of investigated substances it can be observed, that after a flow time of 75 days in the groundwater no additional removal for another 70 days could be observed. Concentrations reached after 75 days remained stable for the rest of the flow time investigated. It cannot be excluded that a situation with stable concentrations was reached before 75 days, because there was no sampling site between a flow time of 25 and 75 days respectively.

27


1.3.4 Drinking Water Technology

I. Coagulation/Flocculation Many waterworks employ coagulation/flocculation processes as one of the main techniques for removal of natural organic matter (NOM) and particles. The process was studied in lab scale experiments to determine removal efficiencies with respect to PPCPs. Results from lab-scale experiments (jar test) exhibited no significant removal of PPCPs by coagulation/flocculation with the exception of the musk fragrances HHCB (galaxolide) and AHTN (tonalide). Removal efficiencies up to 40-50 % were attained with ferric sulfate/ferric chloride in the lab-scale experiments for the musk fragrances HHCB and AHTN, which bound to NOM. However, for all the other polar PPCPs no significant removal was observed considering the analytical statistical errors of about 20 %. These results could be confirmed in full-scale systems (see below) with the same raw waters. Conclusion: With few exceptions coagulation and flocculation is inappropriate to remove PPCPs.

II. Ozonation

Ozone is a strong oxidant used as a disinfectant and oxidant in drinking water treatment. During ozonation processes, many micropollutants are susceptible to oxidation by ozone and OH radicals. Therefore, to predict how efficiently PPCPs will be oxidized under given treatment conditions, rate constants for the reactions with the two oxidants ozone and OH radicals were determined. Furthermore, the oxidation products of selected PPCPs were identified. Rate constants for the reactions of ozone and OH radicals Second-order rate constants (kO3) for the reaction of ozone with selected PPCPs were determined in bench-scale experiments. The rate constants are listed in Table 1.5. For five out of nine PPCPs, half-lives at 1 mg L-1 ozone and pH 7 were < 1 s (apparent kO3 > 5×104 M-1s-1), indicating that the respective compounds are very efficiently transformed during ozonation processes. Besides the reaction with ozone, reactions with OH radicals also contribute to the oxidation of micropollutants during ozonation processes. Therefore, second-order rate constants for the reaction of OH radicals with selected PPCPs were determined as well. OH radical rate constants (Table 1.6) ranged from 3 to 10×109 M-1s-1. Compared to other drinking water micropollutants (e.g. atrazine, methyl tert.-butyl ether, perchloroethylene) the selected PPCPs reacted approximately two to three times faster with OH radicals. This indicates that advanced oxidation processes (AOPs), even if not ozone-based, are able to oxidize most of the selected PPCPs more efficiently than many other relevant micropollutants. Therefore, PPCPs that do not react with ozone directly will be oxidized during conventional ozonation through reactions with OH radicals. Ozonation of Natural Waters The determined rate constants could be confirmed during experiments in various natural waters. For fast-reacting PPCPs (apparent kO3 (pH 7) > 5 × 104 M-1s-1), an ozone dose of 0.5 mg L-1 was sufficient to achieve transformations > 97% in bank filtrate from the River Seine (Paris) and in water from Lake Roine, Finland. The oxidation of diazepam, iopromide and ibuprofen (slow-reacting PPCPs) in natural waters ranged from 25-80 % depending on water quality parameters and rate constants. In advanced oxidation processes (AOP) such as ozone/hydrogen peroxide, the oxidation efficiencies were significantly higher (> 75 % for iodinated contrast media, 80-90 % for ibuprofen).

28


Estrogenic Activity of Oxidation Products of 17α-Ethinylestradiol (EE2) Oxidation products formed during ozonation were identified for 17α-ethinylestradiol (EE2) (Table 1.7). The identification was carried out with GC ion trap MS and LC-MS/MS analysis by performing experiments with model compounds. The phenolic moiety of all identified degradation products was destroyed and the oxidation products were generally more polar. Since the phenolic moiety of estrogens is crucial for the binding to the estrogen receptor, it can be assumed that the oxidation products of EE2 are less estrogenic. A recombinant yeast estrogen screen (YES) was conducted to verify this assumption. The results of the YES exhibited that already low ozone doses are sufficient to reduce the estrogenic activity of EE2-containing solutions by at least a factor of 200. At higher ozone doses which result in substantial ozone exposures the estrogenicity is further reduced (Figure 1.7). However, the reduction for high ozone doses is not as high as expected based on kinetic considerations. A small fraction (< 0.5%) of EE2 react with ozone in a side reaction hindering the cleavage of the phenolic ring. Therefore, for practical applications, it can be predicted that EE2-based estrogenicity is reduced at least by a factor of 200, if an ozonation process achieves a 2-log inactivation of Giardia muris cysts or by a factor of 1000 if a 2-log inactivation of Cryptosporidium parvum oocysts is attained. Table 1.5. Second-order rate constants for the reaction of ozone (O3) with PPCPs

Compound pKa kO3 (T=20°C)(a)

(M-1s-1) Reactive Species

Bezafibrate 3.6 590±50 dissociated Carbamazepine - ~3×105 neutral Diazepam - 0.75±0.15 neutral Diclofenac 4.2 ~1×106 dissociated 17α-Ethinylestradiol 10.4 ~7×109 dissociated Ibuprofen 4.9 9.6±1 dissociated Iopromide - <0.8 neutral Sulfamethoxazole 5.7 ~2.5×106 dissociated Roxithromycin 8.8 (4.5±0.5)×106 neutral

(a) Rate constants for the most reactive species given in the last column

29


Table 1.6. Rate Constants for the Reaction of OH Radicals with PPCPs Compound OH Generation kOH (109 M-1s-1)(a)

Bezafibrate UV/H2O2 7.4±1.2 Carbamazepine UV/H2O2 8.8±1.2 Diazepam UV/H2O2, γ-radiolysis 7.2±1 Diclofenac γ-radiolysis 7.5±1.5 17α-Ethinylestradiol UV/H2O2 9.8±1.2 Ibuprofen UV/H2O2 7.4±1.2 Iopromide O3/H2O2, γ-radiolysis 3.3±0.6 Sulfamethoxazole UV/H2O2 5.5±0.7

(a)experimental conditions pH=7, T=25°C, errors = 95% confidence intervals

-4

-3

-2

-1

0

0 5 10 15 20 25O3 exposure [mg/L*min]

log

c/c 0

EEQsEE2

Figure 1.7. Reduction of estrogenicity (EEQ) and the EE2 concentration as a function of ozone exposure. Experimental conditions: [EE2]0=1 μM, [O3]0=1 mg L-1, pH=8, T=10 °C, [t-BuOH]=5 mM.

Table 1.7. Examples of oxidation products formed during the ozonation of EE2. O

HO

O

HO O

OH

O

HOHO

O

OH

O

O

OH

O

OH

OH

O

OH

O

HO

OHO MW=268 (1) MW=298 (2) MW=292 (3) MW=326 (4)

Ozonation products of carbamazepine Ozone reacts very quickly with carbamazepine (Table 1.8). The primary oxidation products formed were quite stable with regard to further ozone attack. However, they can be oxidized by OH radicals. All these oxidation products have never been described in literature. Therefore, no toxicological data is available. To assess the toxicity of one of these oxidation products (BQD), bacterial tests, namely, the AMES test (testing mutagenicity) and the UMU test (testing genotoxicity) were applied. Salmonella bacteria, cultured with and without rat liver enzyme and serial dilutions of BQD (concentrations from 1.88 to 60.0 µM), showed no mutagenic induction. The UMU test, performed with BQD concentrations from 7.23 nM to 14.8 µM, also showed no major effect. Therefore, it can

30


be concluded that the ozonation product BQD, is neither mutagenic, genotoxic or cytotoxic to the salmonella bacteria test system at concentrations far above at expected levels for drinking water. Table 1.8. Carbamazepine (CBZ): Structures of ozonation products and kinetic data for the oxidation by ozone and OH radicals

5H-Dibenz[b,f]azepine-5-carboxamide

(Carbamazepine)

M.W. = 250.2 g/molkozone = ~ 6.9 M-1s-1

k.OH = ~ 6.7x109 M-1s-1

M.W. = 266.2 g/molkozone = ~ 1.1 M-1s-1

k.OH = ~ 5.1x109 M-1s-1

M.W. = 282.2 g/mol

NC

CN

O

O

H

CO

OHNC

CN

O

O

H

CO

H

BQM BQD BaQD

NC

CN

O

H

CO

HN

CNH2O

CBZ

1-(2-benzaldehyde)-4-hydro-(1H,3H)-

quinazoline-2-one

1-(2-benzaldehyde)-(1H,3H)-quinazoline-

2,4-dione

1-(2-benzoic acid)-(1H,3H)-quinazoline-

2,4-dione

M.W. = 236.3 g/molkozone = ~ 3x105 M-1s-1

k.OH =8.8± 1.2x109 M-1s-1

5H-Dibenz[b,f]azepine-5-carboxamide

(Carbamazepine)

M.W. = 250.2 g/molkozone = ~ 6.9 M-1s-1

k.OH = ~ 6.7x109 M-1s-1

M.W. = 266.2 g/molkozone = ~ 1.1 M-1s-1

k.OH = ~ 5.1x109 M-1s-1

M.W. = 282.2 g/mol

NC

CN

O

O

H

CO

OHNC

CN

O

O

H

CO

H

BQM BQD BaQD

NC

CN

O

H

CO

HN

CNH2O

CBZ

1-(2-benzaldehyde)-4-hydro-(1H,3H)-

quinazoline-2-one

1-(2-benzaldehyde)-(1H,3H)-quinazoline-

2,4-dione

1-(2-benzoic acid)-(1H,3H)-quinazoline-

2,4-dione

M.W. = 236.3 g/molkozone = ~ 3x105 M-1s-1

k.OH =8.8± 1.2x109 M-1s-1

Overall, the results exhibited that most of the selected PPCPs can be efficiently oxidized during conventional ozonation processes. Due to the high reactivity towards hydroxyl radicals, most PPCPs and their ozonation products can also be efficiently oxidized by OH radicals based AOPs. A combined process of ozone followed by ozone/hydrogen peroxide has the highest potential for the oxidation of PPCPs. In the case of EE2, it could be shown that ozone doses applied for disinfection of drinking water can efficiently remove the estrogenicity associated with the presence of this compound in the water. Conclusion: Ozonation is a very effective treatment process to oxidize PPCPs. Currently, there is

no indication that the formed oxidation products are toxic.

III. Chlorine Dioxide Chlorine dioxide (ClO2) is an alternative to chlorine for disinfection of drinking water due to its lower potential of forming disinfection by-products such as trihalomethanes. To predict whether PPCPs will be oxidized under given conditions, second-order rate constants for the reaction of chlorine dioxide with selected PPCPs were determined in bench-scale experiments. The rate constants are summarized in Table 1.9. The deprotonated species of diclofenac, EE2 and sulfamethoxazole as well as the neutral species of roxithromycin exhibit a high reactivity to chlorine dioxide. Taking into account the pKa values of these PPCPs, it can be concluded that diclofenac, EE2 and sulfamethoxazole are quickly oxidized at pH 7 whereas roxithromycin shows only an intermediate reactivity at this pH. The reactivity of the pyrazole derivatives (phenazone, propyphenazone, and dimethylaminophenazone) is estimated to be intermediate or high. Most of the remaining compounds do not react at an appreciable rate with chlorine dioxide. Compared to ozone, fewer compounds react with chlorine dioxide and the rate constants are considerably lower. As a consequence, only a limited number of pharmaceuticals will be efficiently removed during water treatment with chlorine dioxide.

31


Table 1.9. Second-order rate constants for the reaction of chlorine dioxide with PPCPs Compound (species) pKa kClO2 [M-1s-1] Error(a) Studied pH range Bezafibrate 3.6 < 0.01 7 Caffeine < 1(b) 7.4 Carbamazepine <1, >13 < 0.015 7 Clofibric acid 3.0 < 1(b) 7.4 Cyclophosphamide <1(b) 7.4 Diazepam < 0.025 7.4 Diclofenac (anion) 4.2 1.05 × 104 ± 0.1 × 104 5 to 10 Dimethylaminophenazone > 100(b) 7.4 EE2 (neutral) 10.4 < 200 2 to 3 EE2 (anion) 4.6 × 108 ± 0.8 × 108 4 to 6 Fenoprofen < 1(b) 7.4 Gemfibrozil < 10(b) 7.4 Glibenclamide < 10(b) 7.4 Ibuprofen 4.9 < 0.01 7 Ifosfamide < 18(b) 7.4 Iopromide < 0.01 7 Ketoprofen < 1(b) 7.4 Naproxen 10-100(b) 7.4 Pentoxifylline < 1(b) 7.4 Phenazone > 100(b) 7.4 Propyphenazone > 100(b) 7.4 Roxithromycin (neutral) 8.8 1.4 × 104 ± 0.3 × 104 6 to 7 Sulfamethoxazole (neutral) 5.7 < 100 4 Sulfamethoxazole (anion) 7.9 × 103 ± 0.9 × 103 5 to 10

(a) 95% confidence intervals (b) (b) Estimated rate constants based on experiments with drinking water that was spiked with selected pharmaceuticals (Co= 1 µg L-1) and subsequently treated for 30 min with 0.95 and 11.5 mg L-1 ClO2. With the determined rate constants, the oxidation of selected PPCPs by chlorine dioxide could be accurately predicted in lake water and groundwater. This exhibited that rate constants determined in pure waters at high PPCP concentrations can be applied to predict the oxidation of PPCPs present in natural waters in the low concentration range of ng L-1 to μg L-1-range. The fact that chlorine dioxide reacts fast with certain compounds, shows that the specific classes of PPCPs (e.g. sulfonamides, and estrogens) can be efficiently removed during chlorine dioxide disinfection, which might be the only treatment step in a small water utility. However, an overall assessment of this process is not possible because very little is known about the formation of oxidation products and their biological activity.

32


IV. Chlorination Chlorine is a strong oxidant used primarily as a disinfectant in drinking water treatment. To predict whether PPCPs will be oxidized under given conditions, the oxidation of PPCPs by chlorination was investigated. Lab scale tests were performed with sodium hypochlorite (NaOCl). Treated drinking water samples from a full-scale water treatment plant were spiked with selected pharmaceuticals up to a concentration of 1 μg L-1. The experiments were performed as batch processes with chlorine doses 0.5 and 6.0 mg L-1 Cl2 and a reaction time of 30 min. Chlorine was found most effective towards pyrazole derivatives phenazone, dimethylaminophenazone and propyphenazone. A complete oxidation of these compounds was achieved by a chlorine dose of 0.5 mg L-1 Cl2. Diclofenac, indomethacine and naproxen were also removed efficiently. Their average removal for a chlorine dose of 0.5 mg L-1 Cl2 were about 90%, 60% and 40%, respectively. The other selected pharmaceuticals, including caffeine, carbamazepine, diazepam as neutral compounds, bezafibrate, clofibric acid, ibuprofen as acidic compounds, were refractory to oxidation by chlorine.

Conclusion: Disinfection with chlorine and chlorine dioxide does not lead to a general

oxidation/removal of PPCPs. Only a few PPCPs tend to be transformed. The transformation products are currently unknown.

V. Adsorption on activated carbon

Activated carbon is a common process applied in waterworks to eliminate many organic micro pollutants such as pesticides by adsorption. Therefore, its efficiency for PPCP removal was investigated in lab scale experiments.

Adsorption isotherm coefficients in Milli-Q water The bottle point isotherm technique was used to determine the equilibrium capacity of PAC (Powdered Activated Carbon) for the different compounds. This was done in comparison with atrazine, taken as reference compound because activated carbon adsorption is recommended as BAT (Best Available Technology) for this pesticide. Adsorption coefficients K and 1/n were determined according to the Freundlich adsorption model, which enabled calculation of the PAC dose required for a given removal target. The results are reported in Figure 1.7, for removal objectives of 50 %, 90 % and 99 % in Milli-Q water. The doses of PAC are low, which shows that adsorption on activated carbon is a good solution for PPCP removal. PPCPs can therefore be classified in 2 categories:

1st category: carbamazepine, diazepam, tonalide and galaxolide are very easily adsorbable (<0.2 mg/L AC for 99% removal): Neutral compounds and the musks, with rather high Kow values (log Kow between 2.4 and 6.4);

33


2nd category: ibuprofen, roxythromycin, sulfamethoxazole and iopromide are medium adsorbable and showed a behavior similar to that of atrazine. All are either charged compounds, or compounds with lower Kow values (log Kow between –2.3 and 2.7).

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

50% 90% 99%

% Removal for C0=1µg/l

PAC

dos

e (m

g/l)

IBPCBZ

DZPAHTNHHCBSMX

ROXAtrazine

Figure 1.8. PAC doses calculated for removal objectives of 50, 90 and 99% for individual PPCPs

in Milli-Q water. Initial concentration of PPCPs, 1 μg L-1.

Adsorption on PAC in real waters The impact of natural organic matter on the efficiency of PPCP adsorption was investigated by measuring adsorption isotherms in four waters, characterized by a DOC up to 2.5 mg L-1. As expected, the increase in DOC leads to an increase of PAC dose required for achieving a certain removal efficiency. As a summary, the following conclusions can be made for waters with DOC between 1 and 2.5 mg L-1:

• 90% removal of the above mentioned PPCPs could be achieved with 5 mg L-1 PAC, 12 mg L-1 PAC enabled the target of 99 % removal for all selected PPCPs;

• For a removal target < 90 %, only carbamazepine was adsorbed more easily than atrazine. The different compounds could be classified as follows: (decreasing order of adsorption): carbamazepine > atrazine, ibuprofen, roxithromycin, iopromide > sulfamethoxazole.

• For a removal target higher than 95 %, the different compounds could be classified as follows: (decreasing order of adsorption): carbamazepine ≈ roxithromycin ≈ iopromide > ibuprofen ≈ atrazine > sulfamethoxazole.

• In all cases, the antibiotic sulfamethoxazole was found as the least adsorbable compound. This finding is consistent with its low polarity since sulfamethoxazole is negatively charged at pH>6.

Conclusion: Activated carbon is a powerful process to remove PPCPs. Only a limited number of

PPCPs such as iodinated contrast media and the antibiotic sulfamethoxazole show insufficient affinity to activated carbon.

34


VI. Membranes

Membrane filtration is an emerging technology in drinking water treatment. It can be applied with and without addition of PAC. Its efficiency to remove PPCPs was investigated in pilot-scale experiments. Ultrafiltration/PAC adsorption The Cristal® process, which has been developed by SUEZ-ENVIRONNEMENT in the early 1980’s, consists of combining ultrafiltration with PAC adsorption. It is typically used for the removal of contaminants such as pesticides. The flexibility of this process, which enables the dose of PAC to be adapted to the quality of the influent at any time during the operation, makes this process very attractive. Pilot-scale experiments were performed with the following process: tap water was spiked continuously with PPCPs at different concentrations (a few µg L-1), and with varying PAC doses (5 to 10 mg L-1). The filtration cycle was 40 minutes, 3 permeate samples were taken over a cycle duration. The results exhibited that the use of 10 mg L-1 PAC enabled removal efficiencies between 93 and 99 %. The classification between compounds was not very different from that obtained during the isotherm tests, except for iopromide, which appeared less adsorbable, and ibuprofen, which appeared more adsorbable. The order of adsorption is (degreasing order): diazepam ≈ carbamazepine ≈ roxithromycin ≈ ibuprofen > atrazine > iopromide, sulfamethoxazole. For the two musks tonalide and galaxolide, ultrafiltration alone (without adding PAC) did provide an efficient removal, probably due to adsorption phenomena on the membrane itself. For the other compounds, ultrafiltration had no effect on their removal, as expected from the cut-off of such membranes (about 10 nm). Nanofiltration Three different types of nanofiltration and reverse osmosis membranes were tested at pilot-scale in parallel (NF90, XLE and BW30, from Filmtec, all spiral wound elements). The pre-treatment was performed by ultrafiltration. Different recovery rates were compared, corresponding to different qualities of permeate. PPCPs were spiked at 1 to 2 µg L-1, which is about 3 order of magnitude higher than what could be expected under realistic conditions. Samples were taken after 3 hours to reach hydraulic equilibrium in the system. Even at the highest recovery rate (85 %), which corresponds to the lowest quality of permeate (worst-case scenario), all the PPCP investigated (ibuprofen, carbamazepine, diazepam, sulfamethoxazole, roxithromycin and iopromide), which were spiked at a few µg L-1, were removed by more than 98 % (concentrations were lower than quantification limit for permeate). As a comparison, the removal of sodium ion was 18 %, 66 % and 78 % for NF90, XLE and BW30 membranes, respectively.

Conclusion: Nanofiltration and Ultrafiltration/PAC are powerful processes to remove PPCPs.

35


VII. Fate of PPCPs in full-scale waterworks The results obtained in lab-scale and pilot scale experiments were tested in full-scale waterworks in France, Germany and Finland.

Fate of PPCPs in French waterworks of the Paris area Several campaigns performed on full-scale plants treating water from the river Seine, with multi-barrier treatment processes, showed that most of PPCPs investigated were not found above the LOQ in the treated water. The treatment trains included pre-ozonation, flocculation, sand filtration, main ozonation, GAC filtration and final chlorination for one plant; artificial groundwater recharge, flocculation with PAC addition, double filtration, main ozonation and final chlorination for the second plant. The initial contamination levels in the river water were rather low, with maximum values of 200 ng L-1 for sulfamethoxazole and HHCB (Galaxolide) in the period of the lowest flow rate. None of the 15 antibiotics monitored was detected in the treated water, although 7 antibiotics were present in the river Seine. However, 4 of 8 ICM were detected in the treated water, with the highest concentration for iopamidol (60 ng L-1). The benefit of artificial groundwater recharge was demonstrated on one site for sulfamethoxazole and galaxolide, whereas carbamazepine was hardly removed. Sulfamethoxazole was detected after artificial groundwater recharge however, its concentration was decreased by more than 60% relative to the original level and HHCB was totally removed, presumably mainly by sorption. This is in agreement with results a) from the biological wastewater treatment, where sulfamethoxazole was biodegradable to some extent, and Carbamazepine was totally persistent, and b) from the soil passage after irrigation of treated wastewater leading to a total removal of HHCB, whereas Carbamazepine was leaching into groundwater. Already after pre-ozonation at the waterworks that treats directly the water from the river Seine, most PPCPs were below the LOQ. In one treatment plant, the clarification step contributed to a significant reduction of carbamazepine, galaxolide and tonalide, which was mainly attributed to the addition of 2 to 3 mg L-1 PAC in the settling tank. Fate of PPCPs in German and Finnish Waterworks To assess the removal efficiency of PPCPs, individual drinking water unit processes were investigated. The treatment train of waterworks I includes particle removal, pre-ozonation, flocculation, main ozonation, coagulation, filtration and granular activated carbon filtration (GAC). The treatment train of waterworks II and III includes flocculation, coagulation filtration and GAC. The treated raw water withdrawn from the river exhibits a DOC of 2.3 to 3.5 mg L-1, a turbidity of 10 to 20 FNU and a pH of 8-8.2. In raw water of waterworks I and II the concentrations of diclofenac, bezafibrate, ibuprofen, diazepam were about 10 to 40 ng L-1, the concentrations of the antibiotics sulfamethoxazole, clarithromycin, erythromycin, roxithromycin and sulfadoxin ranged between 20-60 ng L-1 and carbamazepine 60-160 ng L-1. Higher concentrations were found for the X-ray contrast media iopromide, iomeprol, iopamidol, iohexol and diatrizoate (60 to 370 ng L-1) with maximum levels for diatrizoate. The quantified musk fragrances varied between <20- 90 ng L-1) for AHTN and <20-240 ng L-1 for HHCB. Pre-ozonation of river water showed a partial oxidation of most of the pharmaceuticals, only a few compounds such as carbamazepine were totally oxidized. With flocculation (ferric chloride, recycled sludge and polyacrylamide as coagulation aid; ferric chloride alone) no significant decrease of pharmaceuticals and musk fragrances was observed. However, the efficiency of ozonation is improved when an efficient flocculation step is located prior to the ozonation due to the DOC removal and the corresponding low turbidity (0.2-0.3 FNU) of flocculated water. Main ozonation leads to a complete oxidation of the pharmaceuticals carbamazepine, ibuprofen, diclofenac, bezafibrate, diazepam, sulfamethoxazole, clarithromycin, erythromycin, roxithromycin,

36


sulfadoxin and the musk fragrances AHTN and HHCB. Thus, all these PPCPs were not further detected after ozonation. The ozone dosage was 1 to 1.2 g m-3 (pre-ozonation) and 1.2-1.5 g m-3 (main ozonation). However, removal rates of only 30-60 % were attained for the X-ray contrast media iopamidol, iohexol, iomeprol, iopamidol and diatrizoate with pre-ozonation – flocculation – main-ozonation in series. GAC filtration was found to be an very effective removal process for the pharmaceuticals which are still present after ozonation in waterworks I. Even X-ray contrast media were adsorbed to GAC for longer operation times. The removal rates for non-ionic X-ray contrast media were 80 to > 90 % for conventional operating times of the GAC-filters. Only diatrizoate was found to break through the filters at a noticeable reduced operating time (specific throughput of about 30 m³ kg-1 GAC). A distinct deviation in removal efficiency of diatrizoate was observed using two different types of GAC.

Using GAC directly after flocculation of raw water II/III leads to an excellent removal of most PPCPs. GAC filtration removed 40 % of ibuprofen and 95 % of naproxen. Total removal rate was 60 % for ibuprofen and 95 % for naproxen. However, due to the elevated DOC in comparison to waterworks I, ICM break through to a higher extend.

VIII. Final conclusions for the removal of PPCPs in waterworks

The results obtained in the lab-scale and pilot-scale experiments for the removal of PPCPs were well transferable to the conditions of full-scale waterworks. In most cases the lab-scale results could be confirmed by the elimination of PPCPs in French, Finish and German waterworks. In the investigated waterworks polishing treatment processes provide an efficient barrier against a contamination of drinking water with PPCPs: • Ozonation was shown to be very effective in oxidizing PPCPs, either via direct ozone attack or via

excitation with OH radicals. Major oxidation products were identified for the fast reacting PPCPs EE2, carbamazepine and diclofenac. Based on biological tests performed with EE2, it was shown that low doses of ozone were sufficient to reduce the estrogenicity by at least a factor of 200. BQD, one of the major degradation products of Carbamazepine did not show any genotoxicity, mutagenicity and cytotoxicity.

• Only a few PPCPs were transformed by chlorine or chlorine dioxide. However, for those PPCPs containing amino or phenolic moieties a complete transformation can be expected.

• Activated carbon adsorption was found to be a very good option for PPCP removal, either via GAC filtration, or the combined process PAC/ultrafiltration (Cristal® process).

• Nanofiltration and reverses osmosis membranes were found to remove PPCP completely.

Based on the overall results, it appears that a significant PPCP contamination of drinking water produced from surface water is quite unlikely in most European facilities, which have complete treatment trains including activated carbon, ozonation or membrane filtration. Only X-ray contrast media are capable to pass through these processes, with poor adsorption to GAC and low reactivity with ozone. However, it has to be noted that small waterworks without these more advanced technologies and unexpected pharmaceuticals residues in their raw water do not provide a barrier for the removal of these polar compounds. As a consequence, frequently a contamination of drinking water is likely if raw water is contaminated with PPCPs.

37


1.3.5 Environmental Risk assessment I. Hazard Assessment For the substances under investigation, physico-chemical data, as well as data on the fate and behavior in the environment were collected. Criteria for the selection were the attempt to cover several pharmaceutical or therapeutic groups, the availability of relevant information about effects and exposure in the scientific literature, and the availability of analytical methods. Selected PPCPs for preliminary environmental risk assessments were the antiepileptic carbamazepine (CBZ), the antibiotic sulfamethoxazole (SMX), the oral contraceptive17α-ethinylestradiol (EE2) and the synthetic musk tonalide (AHTN).

II. Exposure Assessment Environmental concentrations for the use in the risk characterization for the selected PPCPs were based on predicted environmental concentrations (PECs) derived by model calculations and analytically measured environmental concentrations (MECs) as published in literature. For the estimation of PECs, two exposure models provided by the EMEA (EMEA 2001, 2003) recommended for the use with pharmaceuticals and the assessment program EUSES (EC 1997) were applied. The publications and their data were subjected to quality criteria provided by the 'Technical Guidance Document' TGD (EC 2003). Averaged MECs for German surface waters derived from recently published data of measured environmental concentrations were 517 ng L-1 for carbamazepine, 0.6 ng L-1 for 17α-ethinylestradiol, 140 ng L-1 for sulfamethoxazole and 311 ng L-1 for tonalide (Liebig et al., in preparation).

In most cases the MECs were slightly lower than the calculated PECs. However, for use in initial exposure assessment of a risk assessment estimation of PECs should always result in higher values than the real environmental concentration in order to have a safety margin. In this investigation we showed that for the selected PPCPs the exposure model provided by EMEA in 2001 seems to be the most adequate for initial exposure assessment of PPCPs compared to the other models considered in this study.

III. Effects assessment A literature search was conducted for effect data for the selected PPCPs, which served as a basis for decision making on which complementary effect studies had to be conducted for the performance of preliminary environmental risk assessments. Additionally to the standard tests (algae, daphnia, fish), fish embryo toxicity tests were performed with SMX and CBZ according to an OECD draft guideline (OECD 1998a). Furthermore, a daphnia reproduction study was performed with CBZ according to the OECD guideline 211 (OECD 1998b). For the lipophilic compounds EE2 and AHTN with octanol-water partition coefficients log Pow of 4.2 and 5.7, respectively, sediment toxicity studies were performed using the oligochaete Lumbriculus variegates as test organism. Due to its high endocrine potential for fish, its lipophilicity and its potential for secondary poisoning, a bioaccumulation study was performed with radiolabeled 14C-EE2 using L. variegatus following the method described by Egeler et al. (1999, 2000).

The results of the studies conducted and the most sensitive data found in literature were used to derive the Predicted No Effect Concentrations (PNECs) for the selected PPCPs to be used in the risk characterization by applying respective assessment factors.

38


SMX was determined not to be toxic for fish and daphnids (EC50/LC50 > 100 mg L-1), whereas the green alga showed higher sensitivity in the growth inhibition test with an EC50 = 3.2 mg L-1. The most sensitive organism was duck weed in the Lemna test with a NOEC = 10 µg L-1 (Brain et al. 2004). The most sensitive organisms for CBZ were the crustaceans in the reproduction tests with NOECs of 0.4 and 0.025 mg L-1 for D. magna (own study) and C. dubia (Ferrari et al. 2003), respectively. The toxicity studies with sediment dwelling organisms resulted in low toxicity of AHTN and EE2 for L. variegatus (own studies). For CBZ a NOEC < 0.625 mg kg-1 sediment dry weight was determined for C. riparius by Oetken et al. (in preparation).

In the bioaccumulation study L. variegatus showed a high potency to accumulate 14C-EE2. After 35 days of exposure an accumulation factor (AF) of 75 based on total radioactivity and normalized to the worm’s lipid content and sediment TOC was determined. Thin layer chromatography (TLC) of liquid extracts of worms by allowed to determine the amount of conjugated and free 14C-EE2 accumulated in L. variegatus. It was shown that after treatment of the worm extracts with β-glucuronidase, 84 ± 5% of the extracted total radioactivity corresponded to the 14C-EE2-standard in the TLC analysis. The high potency of L. variegatus to accumulate EE2 and the fact that the major part of the accumulated form of EE2 can be enzymatically transferred into the biologically active parent compound, shows the possibility for EE2 to be transported within the food web and to cause secondary poisoning of predators which are not directly exposed to EE2.

IV. Risk Characterization and Environmental Risk Assessment The preliminary environmental risk assessment of the selected PPCPs was performed according to the two-phased tiered ERA-scheme proposed by the EMEA (EMEA 2003). Phase I consists of a crude initial PEC assessment of the substance in surface water based on the maximum daily dose of the active ingredient. If the estimated PEC exceeds the "action limit" of 10 ng L-1, a quantitative risk assessment is performed in Phase II with refined PECs and investigations on fate and effect of the compound. The effect assessment in Phase II is divided in Tier A and Tier B. In Phase II, Tier A studies on physico-chemical properties are conducted which may serve as information for further PEC-refinement in Tier B. Furthermore, acute toxicity towards aquatic organisms is assessed in Tier A. If the quantitative risk characterization based on acute toxicity indicates a risk for the aquatic environment or if trigger values referring to lipophilicity, adsorption potential or transformation in water-sediment systems are exceeded, the ERA will be continued in Phase II, Tier B with another PEC-refinement and assessment of chronic toxicity. Otherwise, the substance is not considered to pose a risk to the environment and the ERA may be finalized.

According to the EMEA model, crude PECs were estimated as > 10 ng L-1 for all investigated PPCPs except for EE2, where PEC-estimation resulted in < 1 ng L-1. For EE2, the high endocrine potency is well known, and adverse ecotoxicologic effects may occur at concentrations lower than the action limit. Therefore, the ERA ought to be continued in Phase II irrespective of the initially estimated PEC.

For the antiepileptic CBZ the initial risk assessment based on acute effect data as well as the refined risk assessment did not indicate a risk for the aquatic compartment. In contrast to this, the assessment of sediment toxicity based on an estimated PEC and the effect data measured by Oetken et al. (in preparation) revealed a risk for benthic organisms. However, this result should be used as a crude initial risk characterization for the sediment compartment, since PEC is roughly estimated and effect data are based on nominal test item concentrations.

According to the initial risk assessment of the antibiotic SMX based on acute toxicity data, the compound is unlikely to represent a risk to the aquatic environment. However, the refinement of exposure and effect assessment based on measured data resulted in a PEC/PNEC ratio > 1 indicating a risk for the environment. For further refinement of the risk assessment of SMX, long-term testing

39


using test organisms of other trophic levels and specific tests using micro-organisms are recommended.

The risk characterization for EE2 based on acute effect data did not indicate a risk. Refined exposure and effect data based on its specific endocrine activity, however, led to the conclusion that the estrogen indeed represents a risk to the aquatic environment. Furthermore, bioaccumulation of EE2 in benthic organisms might pose a risk via secondary poisoning for worm eating predators. Although the synthetic estrogen was unlikely to represent a risk for the aquatic environment according to initial risk assessment, further testing resulted in the reverse conclusion, which makes further investigations necessary.

The initial risk characterization for AHTN based on acute effects data indicated a risk for the aquatic environment. The refined exposure and effects assessment revealed a lower PEC and a higher PNEC, which led to a an acceptable risk. However, the PEC/PNEC ratio considering the compartment sediment is only slightly lower than 1. Therefore, due to the limited availability of exposure and effects data a risk for benthic organisms caused by AHTN cannot be excluded.

Considerations on the EMEA (2003) Risk Assessment Scheme The principle assessment scheme is based on the requirements given in the TGD (EC 2003) for chemicals in general, while adaptations on a specific assessment of human pharmaceuticals are restricted to some features only:

• Action limit of 10 ng L-1 in the primary exposure assessment (Phase I);

• Overriding the action limit in case of any unusual potential for adverse effects (e.g. endocrine disruption);

• Consideration of metabolites if to be expected in relevant amounts;

• Possibility of the integration of human excretion profile into the PEC refinement in Phase II, Tier B;

• Recommendation of specific precautionary and safety measures for risk management if a probable risk was identified.

The example of endocrine disrupting chemicals has shown, that the application of an action limit, which actually terminates any further risk assessment for substances not exceeding this limit (i.e. PEC < 10 ng L-1), may lead to disregarding of long-term, low-level effects. Experience with ecotoxicologic effects of human pharmaceuticals is still very limited, so "expert evaluation" for overriding the action limit and progressing with the risk assessment might be rather vague.

Likewise, desired therapeutic effects might not be the main environmental mode of action. Presently, testing is done on 'standard representative organisms', but further experience is needed to demonstrate, that these species are particularly sensitive to human pharmaceuticals and, as the case may be, to personal care products. The specificity of antibiotics is another example for the need of tailored testing. It also demonstrates the need for introduction of additional endpoints, such as 'build-up of resistance', into the risk assessment. In general, it appears that chronic effect tests are more suitable for testing of pharmaceuticals than acute tests, and chronic tests should be made obligatory.

Although monitoring data on the occurrence of human pharmaceuticals in natural sediments are limited, the contamination of sediments is realistic due to the intrinsic properties and elevated persistence of several pharmaceuticals. Therefore, the consideration of the sediment compartment in the revised ERA scheme of the 2003 is a considerable advance in so far, as the assessment of sediment toxicity is required for compounds for which the water/sediment transformation study (OECD 2002)

40


demonstrated an extensive partition of the drug to the sediment. However, the term ‘extensive shifting’ is not further defined in the draft guidance.

V. Eco Labeling The number of chemicals used in personal care products (PCPs) is higher than 8,000 which makes a single-substance-specific approach time consuming and difficult. To evaluate environmental effects of PCPs we have focussed on shampoos, shower gels and foam baths (SSBs). The ingredients of a random sample of 40 commercially available SSBs were compiled. The literature search for ecotoxicologic data of a selected number of ingredients revealed that in most cases no data were available. For few ingredients the data indicated high ecotoxicologic potential. A preliminary risk assessment performed according to the rules of the TGD (EC 2003) indicated potential risk for the aquatic environment for three out of six compounds assessed. In support of a comprehensive quantitative environmental risk assessment which might be required for ingredients of PCPs after revision of the EU Cosmetic Directive an eco-labeling for SSBs as an effective and immediate policy instrument was proposed. The basic criteria to grant an eco-label should include information on ecotoxicity and degradation potential of all ingredients in a SSB, and on saving of resources and packaging material. SSBs that comply with the proposed criteria will contribute to a lower chemical burden of waste water and surface water. The developed basic criteria for an eco-label award for SSBs were presented to the German Federal Environmental Agency and to producer and consumer associations.

VI. Conclusions The target of environmental risk assessment section was to collect or generate data on fate of effects of PPCPs and to perform an Environmental Risk Assessment according to current guidance.

Hazard Assessment

It has been proven to be difficult to derive some of the data which are essential for exposure assessment, in particular data on adsorption. Many of the pharmaceuticals are water soluble with a relative low log POW. Therefore, except of musk fragrances and estrogens non specific sorption which is mainly considered to be responsible for the partitioning to sludge and/or sediment, will often not be relevant for the selected PPCPs. This has been demonstrated when considering the Kd-values for sewage sludge shown in previous chapters. Rivers and sediments are the main environmental compartments at risk, while exposure to soil, which may receive sewage sludge deposits, is of minor relevance.

41


Exposure Assessment

Predicted environmental concentrations (PECs) were derived for PPCPs by applying rather crude partitioning models (EMEA 2001 and 2003) and a more sophisticated Mackay level II-type multi-media steady state fate model, EUSES. In parallel, measured concentrations (MECs) were retrieved from the literature. MECs have to be evaluated carefully with respect to adequacy and accuracy of the analytical methodology applied and the representativeness of sampling. To achieve a safety margin in risk assessment, estimated concentrations (PECs) should be higher than measured concentrations. Nevertheless, the PECs calculated for four selected PPCPs (all < 1 µg L-1) were found to be acceptable for risk characterization.

Effects Assessment

The battery of tests necessary for an environmental risk assessment of PPCPs complies with regulations for industrial chemicals. In addition to data from literature, own experiments on acute and chronic toxicity of the four selected PPCPs to several species were conducted. As expected, PPCPs did not show a significant acute toxicity (L(E)C50 mostly > 500 µg L-1). Nevertheless, a chronic toxicity is observed with some of the substances, and this may be attributed to specific modes of action (e.g. endocrine disruption). The application of an assessment factor (safety margin) on the results of the effects assessment leads to the predicted no effect concentration (PNEC) at which effects in real ecosystems are expected not to occur.

Risk Characterization

Risk characterization is the quotient of exposure to toxicity (PEC/PNEC) followed by an evaluation where the calculated quotient is compared to a trigger value (usually 1). An unacceptable risk (PEC or MEC/PNEC > 1) was indicated for the musk fragrance tonalide (AHTN), but not for the other three investigated pharmaceuticals. The latter would normally lead to a termination of the risk assessment and an acquittal for the substance. However, further chronic risk assessment of these substances indicated unacceptable risk in connection with endpoints such as reproduction or endocrine effects. This demonstrates that the present decision scheme within the proposed risk assessment for PPCPs did not turn out completely satisfactory. In view of long-term, low-level exposure to PPCPs, it is recommended to make chronic ecotoxicity testing compulsory.

Improvement of Environmental Risk Assessment for PPCPs

The current risk assessment scheme for PPCPs by EMEA (2003) is based on the requirements given in the TGD (EC 2003) for chemicals in general, while adaptations on a specific assessment of human pharmaceuticals are restricted to some features only. More emphasis should be given to chronic ecotoxicity testing and the choice of endpoints specific for known modes of action of PPCPs.

42


1.3.6 Recommendations for indirect potable water reuse in relation to PPCP removal

POSEIDON was focused on the assessment and improvement of technologies for the removal of ingredients of pharmaceuticals and personal care products (PPCPs) in wastewater and drinking water facilities in order to prevent the contamination of receiving waters, groundwater and drinking water by indirect potable water reuse of treated municipal wastewater. Potable indirect reuse is currently a viable option in areas where severe water scarcity occurs [Water Management Europe 1993/94, Crook et al., 1998].

In order to avoid any human health problems caused by ingredients of wastewater a so called “multi barrier system” has to be introduced. This approach implements two basic ideas: If a single natural or technical removal process becomes ineffective, a contamination of drinking water with pathogens or chemicals will be avoided by other processes following afterwards. Additionally, the specific removal potencies of the particulate steps in the multi-barrier system support the efficiency and stability of the whole system. In order to enhance efficient and unpolluted water supply, to minimize environmental impacts from wastewater and to avoid potential human health hazards, POSEIDON comprises the following main objective:

Completion of a strategy for indirect potable water reuse of treated urban wastewater with the emphasis on distinct PPCP contamination in the wastewater, and the aim to optimize the combination of wastewater and drinking water technologies in order to enable or enhance indirect potable water reuse in water scarcity areas. Within the EU project POSEIDON many relevant techniques and processes involved in the urban water cycle have been assessed regarding their removal efficiency for PPCPs. Furthermore, the irrigation and the direct injection of treated wastewater have been investigated within two case studies in Germany and Austria considering the removal of PPCPs during soil and aquifer passages. In any case technical solutions for indirect reuse have to prevent a break through of pathogens and viruses into drinking water. Furthermore, the quality of groundwater should not be decreased by any groundwater management activities, according to the groundwater directive (see “Proposal for a Directive of the European Parliament and the Council on the Protection of Groundwater Against Pollution, draft 2003) which is still under revision. Furthermore, the Water Framework Directive of the EU (2000) requires the identification and reversion of any significant and sustained upward trend in the concentration of any pollutant in groundwater. Therefore, using groundwater as a natural reservoir for indirect reuse purposes, a relevant contamination with pathogens, viruses as well as inorganic and organic pollutants should be avoided. This is especially of concern, since persistent chemical pollutants entering the groundwater may stay there for a long time without substantial degradation and therefore create a reservoir that may be a problem for future generations. Although, for many bioactive PPCPs the environmental toxicity is unknown, a groundwater contamination has to be prevented due to their persistence in groundwater and due to the pre-cautionary principles. The soil properties (e.g. permeability described by Kf values) in the respective areas of water scarcity are determining the feasibility of irrigation and the efficiency of PPCPs elimination via degradation and sorption. When lakes are used as an alternative reservoir of treated wastewater within the indirect potable water reuse process, degradation by solar irradiation may play an important role for the elimination of organic pollutants in addition to microbial degradation and sorption. However, the quality of the treated wastewater used for augmenting lakes should fulfil the same quality criteria as for groundwater recharge, if the lake water infiltrates in appreciable quantities into an aquifer.

43


I. Strategy for planned indirect reuse of wastewater with the emphasis on PPCP removal The site specific local conditions determine the feasibility of a ground water recharge or the augmentation of surface water. The parameters which quantify these conditions have to be evaluated in detail prior to any realization of planned indirect reuse of wastewater. These parameters have been described in common guidelines for reuse [Crook et al., 1998]. For the irrigation of treated wastewater crucial criteria are: the depths of the ground water layer (i.e. the thickness of the vadose zone), the thickness of the microbial active upper layer of the soil (e.g. Ah horizon, humus layer), the time the reclaimed water has to be retained in the aquifer before withdrawal (e.g. 1 year), cation and anion exchange capacity of the soil, percolation rate (Kf values) of the underground and maximum allowable percentage of reclaimed water in the extracted well water. Furthermore, reliable monitoring wells should be available or disposed in the vicinity to detect the influence of the recharge operations on the groundwater level and its quality. Therefore, feasibility studies for the irrigation of treated wastewater or augmenting surface water are very complex. They have to integrate all on-site aspects from the suitability of the hydrogeological and hydrological conditions, the environmental and human health perspectives as well as the social economic acceptances of the local population. The evaluation and integration of all theses aspects were not part of the POSEIDON project. In POSEIDON the removal for PPCPs was investigated for several common and advanced treatment technologies in wastewater and drinking water facilities. A comprehensive study regarding the fate of PPCPs within the indirect potable water reuse cycle was not planned. A first approach was reported by Drewes and Shore [2001], who described that PPCPs are efficiently removed by artificial groundwater recharge due to the contact with soil. An overview of the removal efficiencies obtained in POSEIDON is given in Table 1.10. For the selection of the treatment processes it was taken into account that the costs for the implementation of individual treatment processes are reasonable, otherwise the public will not accept the technical solutions. For instance it is known that nanofiltration in wastewater purification or reverse osmosis in drinking water treatment are removing most of the PPCPs, but considering the costs the acceptance of these techniques will be low in the public as long as other processes lead to similar results. The detailed results are described in the chapters 1.3.1, 1.3.2 and 1.3.3.

Based on the results of POSEIDON the following suggestions can be made with regard to the removal of PPCPs supporting simultaneously the safe removal of pathogens and viruses.

Municipal wastewater treatment Removal of organic pollutants by biological processes is the central step of wastewater treatment. Removal of hormones and PPCPs from aqueous phase as most important aspect in POSEIDON with regard to indirect reuse mainly occurs due to mineralization or modification and adsorption to the matrix of the biosolids in the system. The efficiency of theses processes correlates to the solid retention time (SRT) of the microorganisms in the wastewater treatment. The age depends on the sludge loading rate. With increasing SRT the microbial population becomes more diverse, probably increasing the occurrence of specific biological pathways. For the design of a biological wastewater treatment plant (WWTP) an aerobic SRT of 8-10 days, allowing safe nitrification all year proved to be essential for the removal of hormones and PPCPs. For energy saving reasons denitrification should be implemented additionally, although this further increase of the SRT for denitrification or even aerobic sludge stabilization does not lead to significantly better removal of PPCPs. The requirement for nitrification and denitrification is independent from the technical process applied for separation of the biosolids (secondary clarifier or membranes), from the size of the plant and whether it discharges to a sensitive or non sensitive area. Nitrification and denitrification is crucial for all WWTPs as long as the water is reused for groundwater recharge and the augmentation of lakes/reservoirs.

Although the membrane bioreactor does not substantially increase the removal of PPCPs, it is recommended for indirect reuse since it produces a microbial pure effluent. If no separation by

44


membranes is possible, the secondary clarifier should be designed with generous safety to avoid loss of sludge in case of high hydraulic flow (e.g. storm water). Post treatment steps (polishing lagoon, gravel filter), that are not necessary in the case of membrane bioreactors, yield an additional decrease in hormones and PPCPs content. But it has to be kept in mind that a polishing lagoon usually increases again the DOC which is a major disadvantage for indirect potable reuse (possible formation of THM in the downstream drinking water treatment).

Ozonation as post treatment step for both, membrane bioreactors and systems with secondary clarifier facilitate the further biodegradation of recalcitrant PPCPs/micropollutants. A complete mineralization of the PPCPs and background DOC is not strived for because of the high energy consumption for ozone. However, it can be expected that a partial oxidation of PPCPs/micropollutants will lead to a better biodegradability during further soil passage. In addition, the adverse environmental effects of many PPCPs are reduced by partial oxidation. Therefore, ozonation is recommended prior to irrigation or augmenting lake water. The ozonation has two major benefits: a) disinfection of the treated wastewater to a high degree (reduction of microorganisms up to 3-4 orders of magnitude, including strains with multiple resistances to antibiotics) takes place and b) many PPCPs are oxidized very efficiently. Further biodegradation of the formed oxidation products is expected in contact with soil during irrigation conditions, but to a smaller extent in lakes.

Reservoirs

A) Groundwater Irrigation should be used as long as the hydrogeological conditions are feasible for groundwater recharge, since the soil passage with a microbiologic active upper layer is a very effective treatment process due to sorption and biodegradation as well as a barrier for microorganisms and viruses. Since direct injection of treated wastewater without a soil passage is used, it should only be considered if no other options are available. In this case the wastewater has to be additionally treated by more advanced technologies such as activated carbon, nanofiltration or reverse osmosis. The replenishment of groundwater has to be in compliance with the demands of the EU directives.

B) Lake The augmentation of lake water is an alternative to field irrigation. Crucial criteria will be the dilution factor and the residence time when discharging treated sewage into a lake (to guarantee further degradation of the organic PPCPs) and the absence of pathogens. In case of low dilution factors the pre-treatment of the sewage should be more advanced, in order to avoid the enrichment of pollutants in lake water. Concerning the released pollutants the infiltration of lake water into groundwater has to be in compliance with the demands of the EU directives.

Drinking water treatment Whenever indirect potable water reuse is taken into account, drinking water treatment processes are an important barrier with the prospect of supplying safe drinking water. In this case it is essential to guarantee the removal of PPCPs as well as viruses and pathogens during drinking water treatment, in the case of a breakthrough into the water resources. Based on the Poseidon results and the current knowledge in addition to common drinking water treatment processes (e.g. bank filtration, flocculation, filtration, disinfection by ClO2) the following technologies and/or appropriate combinations are suggested:

Ozonation, activated carbon filtration advanced oxidation based on UV or ozone.

45


46

To remove PPCPs, viruses and pathogens a combination of common treatment processes with advanced treatment technologies has to be established in the form of a multiple barrier system. Crucial criteria for the design of a drinking water treatment scheme are the raw water quality aspects (e.g. quantity of humic acids, concentration and character of pollutants, variability of raw water quality), possible expected hazards and the individual site situation. If appropriate treatment is applied, most of the PPCPs are removed. However, some iodinated contrast media are only removed with relatively new activated carbon.

Measures at the source in order to prevent a PPCP release into wastewater The best approach to prevent a contamination of the environment and the drinking water is to substantially reduce the quantities of PPCPs entering the raw sewage. Therefore, any measures at the source (source control, e.g. ban or eco label for PPCPs, and source separation, e.g. urine separation) will facilitate the removal in the treatment processes afterwards. A detailed discussion for source control can be found above. Further since the removal efficiency in the wastewater treatment increases with decreasing dilution, measures taken to reduce intrusion of rain or parasitic water into the sewer contribute in reducing the PPCP load in treated wastewater.

Implementation of water safety plans A focus of the new WHO drinking water guideline is the implementation of a process controlled quality management system. This approach based on the successfully applied HACCP-principle (Hazard Analysis and Critical Control Points) in food-processing industry is described as water safety plan.

The hazard analysis and critical control point system is a systematic approach to control safety hazards in a process by first identifying hazards, their severity and likelihood of occurrence. Then critical control points and their monitoring criteria are identified to establish controls that will reduce, prevent or eliminate the identified hazards.

Ensuring that the use of treated wastewater in an indirect water reuse cycle does not result in adverse health effects a quality management system as used in the HACCP approach is needed. A water safety plan should be implemented with the objective of a well-adjusted evaluation of hazard for each individual indirect portable water reuse system starting at the source up to the consumer to adjust a quality management system based on monitoring plans, control systems and corrective action plans.

SEIDON, detailed report related to the overall duration (1.1.2001-30.6.2004) 09/05/06

47

Ibu: Ibuprofen, Dicl: Diclofenac, Bezf: Bezafibrate, Clof: Clofibric acid, E1: Estrone, E2: 17β-Estradiol, EE2: 17α-Ethinylestradiol, SMX: Sulfamethoxazole,

Table 1.10: Semi quantitative removal to the influent load of each specific treatment process

--: < 10%, +: 10 to 50%, ++: 50 to 90 %, +++: >90%, n.d.: no data, (brackets): predicted, based on the expert knowledge * AOP: only an OH radical based process is considered (like UV/H2O2) ** UV dose about 100 times higher than that used for DW disinfection (typically 40000 compared to 400 mJ/cm²)

Rox: Roxithromycin, Carb: Carbamazepine, Diaz: Diazepam, Iopr: Iopromide, Diatr: Diatrizoate, Iopam: Iopamidol

1: Drewes J, Heberer T, Rauch T, Reddersen K Ground Water Monitoring & Remediation 23, 3, 2003, 64-72. 2: Verstraeten IM et al., ASCE Oct. 2003, 253-263, 3: Ternes, Th.A, Meisenheimer, M., McDowell, D., Sacher, F., Brauch, H.-J., Haist-Gulde, B., Gudrun Preuss, G., Wilme, U., Zulei-Seibert, N. Environ. Sci. Techn. 36, 3855-3863 (2002) 4: Liu B, Liu X, Sci of Total Env, 320 (2004) 269-274) 5: personal communication by Laurence Meunier, EAWAG (dose of 40000 J/m² with LP lamps)

PO


References

Angelikas A.N., Marecos Do Monte M.H.F., Bontoux L, Asano T (1999) The status of wastewater reuse practise in the mediterranean basin: need for guidelines. Wat. Res. 33, 2201-2217.

Brain, R. A., Johnson, D. J., Richards, S. M., Sanderson, H., Sibley, P. K., & Solomon, K. R. 2004: Effects of 25 pharmaceutical compounds to Lemna gibba using a seven-day static-renewal test. Environmental Toxicology and Chemistry 23, 371-382.

Crook J., Asano T., Nellor M.H. (1990) Groundwater recharge with reclaimed water in California. Water Environ. Techn. 2, 42-49.

Crook J., Engelbrecht R.S., Benjamin M.M., Bull R.J., Fowler B.A., Griffin H., Haas C.L., Rose J.B., Trussell R.R. (Committee to evaluate the viability of augmenting potable water supplies with reclaiming water). In Potable Reuse, the viability of augmenting drinking water supplies with reclaimed water. Editor: Dobbs D.A., National Academic Press, Washington DC, 1998.

Crook J., MacDonalds J.A., Trussell R.R. (1999) Potable use of reclaimed water. J. AWWA 91, 40-49. Cuthbert R.W., Hajnosz A.M. (1999) Setting reclaimed water rates. J. AWWA 91, 50-57. Daughton Ch.G., Ternes Th. A. (1999) Pharmaceuticals and Personal Care Products in the Environment: Agents of

Subtle Change? Environ. Health Perspect. 107, 907-938. Drewes J.E., Shore L.S. Concerns about Pharmaceuticals in Water Reuse, Groundwater recharge, and Animal wastes. In

Pharmaceuticals and Personal Care Products in the Environment Editors: Daughton Ch.G., Jones-Lepp T.L. ACS-Symposium Series 791, 206-228 (2001).

EC (European Commission) 1997: EUSES - European Union system for the evaluation of substances. [Version 1.00]. European Chemicals Bureau.

EC (European Commission) 2003: Technical Guidance Document on Risk Assessment in support of Commission Directive 93/67/EEC on Risk Assessment for new notified substances, Commission Regulation (EC) No 1488/94 on Risk Assessment for existing substances and Directive 98/8/EC of the European Parliament and of the Council concerning the placing of biocidal products on the market, Parts I, II and IV. European Communities, 2003. EUR 20418 EN/1.

Egeler, P., Knacker, T., Römbke, J., Schallnaß, H.-J., Meller, M., Studinger, G., Franke, C., & Nagel, R. 2000: Proposal for a new guideline - bioaccumulation: sediment test using benthic oligochaetes. submitted to OECD.

Egeler, P., Römbke, J., Meller, M., Knacker, T., & Nagel, R. 1999: Bioaccumulation test with tubificid sludgeworms in artificial media - development of a standardisable method. Hydrobiologia 406, 271-280.

EMEA 2001: Discussion paper on environmental risk assessment of non-genetically modified organism (non-GMO) containing medicinal products for human use. CPMP/SWP/4447/00 draft corr. The European Agency for the Evaluation of Medicinal Products - Evaluation of Medicines for Human Use (CPMP Committee for Proprietary Medicinal Products).

EMEA 2003: Note for guidance on environmental risk assessment of medicinal products for human use - Draft. CPMP/SWP/4447/00 draft. The European Agency for the Evaluation of Medicinal Products - Evaluation of Medicines for Human Use (CPMP Committee for Proprietary Medicinal Products).

EMEA 2003: Note for guidance on environmental risk assessment of medicinal products for human use - Draft. CPMP/SWP/4447/00 draft. The European Agency for the Evaluation of Medicinal Products - Evaluation of Medicines for Human Use (CPMP Committee for Proprietary Medicinal Products).

Ferrari, B., Paxéus, N., Lo Guidice, R., Pollio, A., & Garric, J. 2003: Ecotoxicological impact of pharmaceuticals found in treated wastewaters: study of carbamazepine, clofibric acid and diclofenac. Ecotoxicology and Environmental Safety 55, 359-370.

Geyer, H. J.; Rimkus, G.; Scheunert, I.; Kaune, A.; Schramm K.-W.; Kettrup A.; Zeeman, M.; Muir, D. C. G.; Hansen, L. G.; Mackay, D. Bioaccumulation and occurrence of endocrine disrupting chemicals (EDCs), persistent organic pollutants (POPs), and other organic compounds in fish and other organisms including humans. In Beek B (ed.): The handbook of environmental chemistry Vol. 2, Part Bioaccumulation. Berlin, Heidelberg, Springer Verlag. 2000.

Van Houtte E., Verbauwhede J. Re-use of wastewater effluent for indirect production of drinking water in Flanders (Belgium). American Waterworks Association Membrane Technology Conference, 2003.

Heberer T., (2002a) Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment, a review. Toxicol. Lett. 131, 5.

Jagals P., Lues J.F.R. (1996) The efficiency of a combined waste stabilisation pond/maturation pond system to sanitise wastewater intended for recreational re-use. Wat. Sci. Tech. 33, 117-124.

Kolpin DW, Furlong ET, Meyer M, Thurman EM, Zaugg SD, Barber LB, Buxton HAT (2002) Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci. Technol. 36, 1202-1211.

Lauer W.C., Rogers S.E. (1996) The demonstration of direct potable reuse: Denver`s pioneer project. In AWWA/WEF Water Reuse Conference Proceedings. American Water Works Association, Denver, p. 269-289, 1996.

Lauer W.C., Rogers S.E., La Chance A.M., Nealy M.K. (1991) Process selection for potable reuse health effect studies. J. AWWA 83, 52-63.

48


OECD 1998a: Fish embryo toxicity test with the zebrafish Danio rerio. OECD draft. In: UBA-Texte 58/98 – Ökotoxikologische Testverfahren mit aquatischen Organismen. 80-93. Umweltbundesamt, Berlin, Germany.

OECD 1998b: OECD guidelines for testing of chemicals - 211. Daphnia magna reproduction test. Paris. OECD 2002: OECD guideline for testing of chemicals - 308. Aerobic and anaerobic transformation in aquatic sediment

systems. Ternes, Th. A.; Knacker, Th.; Oehlmann, J. Körperpflegemittel in der aquatischen Umwelt –Eine bisher vernachlässigte

Stoffgruppe, Z. Umweltchem. Ökotox. 15, 169-180 (2003). Ternes, T.A., Stüber, J., Herrmann, N., McDowell, D., Ried, A., Kampmann, M., Teiser, B.: Ozonation: A tool for

removal of pharmaceuticals, contrast media and musk fragrances from wastewater? Water Res., 37, 1976-1982 (2003).

Ternes, T.A. Analytical methods for the determination of pharmaceuticals in aqueous environmental samples. Trends Anal. Chem. 20, 419-434 (2001).

Vazquez Montiel O., Horan N.J., Mara D.D. (1996) Management of domestic wastewater for reuse in irrigation. Water Sci. Technol. 33, 355-362.

Water Management Europe 1993/94: An annual review of the European water and wastewater industry. Editor: Wiseman R., Sterling Publication Ltd., London, 1993.

49


1.4. Conclusions including socio-economic relevance, strategic aspects and policy implications POSEIDON established a basic knowledge on the removal of PPCPs in wastewater and drinking water treatment. Both wastewater technology and drinking water technology optimized for the removal of domestic chemicals can be applied world-wide for indirect water reuse. Based on the POSEIDON outcome, a form of BAT (Best Available Technology) for water treatment concerning PPCP removal was proposed and will provide a basis to implement administrative measures with regard to persistent domestic chemicals as contaminants of reclaimed WWTP discharges. Moreover, the exchange with competent European authorities as well as the exchange with water and wastewater industry involved will be the basis for quality improvement of water supply. Thus, POSEIDON contributes to the environmental policy of the Community, in particular to the Community water policy as outlined in Directive 91/271/EEC on urban waste water treatment, Directive 80/68/EEC on the protection of groundwater against pollution caused by certain dangerous substances and especially in the Water Framework Directive COM (99) which intends to promote sustainable water use and prevent further deterioration of aquatic ecosystems.

As stated by the Community water policy (§130r of the Treaty), POSEIDON improves the level of environmental protection, promotes aspects of the Precautionary Principles, and propose preventive actions and rectification of damage at source.

POSEIDON has demonstrated that the proposed scheme for the environmental risk assessment of single human pharmaceuticals needs further improvement to reduce the likelihood of releasing hazardous trace pollutants to the environment.

Municipal wastewater is highly contaminated by PPCPs and (subtle) effects such as feminization of fish have already been found in the receiving waters of wastewater treatment plants. Due to the wide-spread of PPCPs in rivers and groundwater also a contamination of drinking water is known in some cases. Several technological solutions have been elaborated in POSEIDON which avoid a contamination of surface water and drinking water with estrogenic compounds and other PPCPs. For instance estrogens, the most potent PPCPs known, can be eliminated in common nutrient WWTPs (sludge retention times ≥ 15 days), and with more advanced technologies such as effluent ozonation, membrane filtration and activated carbon. For most of the PPCPs such as antibiotics only the advanced technologies lead to an efficient removal. However for wastewater, ozonation is the most promising treatment process, due to its cost effectiveness, while for drinking water production also the other advanced techniques are an option. It is crucial to note that a removal of PPCPs in wastewater treatment and measures at the source such as urine separation will guaranty that surface water, groundwater and drinking water will not be further contaminated.

Nevertheless, a contamination of drinking water is quite unlikely in most European facilities, with complete treatment trains including activated carbon, ozonation or membrane filtration. Only X-ray contrast media are capable to pass through these processes. However, it is crucial to mention that small waterworks without these more advanced technologies are not appropriate to remove PPCPs. As a consequence, contamination of drinking water might occur as long as the raw water is contaminated with PPCPs.

Although, the socio-economic relevance of a PPCP contamination is difficult to quantify since the ecological and toxicological relevance of micropollutants and their mixtures are not yet clearly understood, the importance of water resources for human and nonhuman life is so central that the preservation of its quality deserves great attention. Since a wide variety of new PPCPs with specific chemical and biological properties but often unknown (eco)toxicological effects are produced and day by day released in unmatched and increasing quantities, to seek for their impact and for feasible methods of remediation is a mandatory minimal variant of the precautionary principle for preservation of primary life supporting fundaments.

50


1.5. Dissemination and exploitation of the results

A number of meetings and initiatives have been run to assist in the exchange of information between project participants and other interested parties. These include:

End-user symposium of POSEIDON in Braunschweig, 4th-5th November, 2003. An end-user

conference has been organized to present the major results of POSEIDON. The scientific program was set up in parallel sessions in English and German, in order to make the results available for a broader auditorium. In total 21 lectures were given by the POSEIDON participants in German and 22 lectures in English. In total 19 posters were presented. Number of participants: in total 146, from Germany: 110, Belgium: 2, Denmark 4, England 1, Finland 3, France 3, The Netherlands: 1, Poland 3, Switzerland: 14, Sweden: 1, Spain 2, Austria: 2

A press briefing of the EU of the “Pharma cluster” took place in Goeteburg. Poseidon presented

two posters and gave a talk about the removal of PPCPs in wastewater treatment. In the frame of the press briefing more than 20 contacts with journalist took place, including a radio interview with BBC 4 from London and a press article published in the journal NATURE.

Participation in the organization of the Envirpharma conference in Lyon as part of the

Pharma-Cluster with about 250 participants. Several Poseidon presentations were held between April 14-16, 2003 in Lyon.

Press interviews has been given by POSEIDON participants (see enclosed press articles)

Cooperation with end-user Cooperation with WEDECO, Herford, Germany (company producing ozone devices):

Although the cooperation with WEDECO was originally not foreseen, the company provided POSEIDON with a pilot device and respective operators for ozonation and AOPs of wastewater. Therefore, pilot scale experiments in Braunschweig, Germany and in Dübendorf, Switzerland were performed. In most of the POSEIDON committee meeting WEDECO participated without getting any funding from the EU.

Cooperation with the company Zenon, Germany providing POSEIDON with membranes for the pilot plant studies.

Cooperation with the company Kemira Chemicals Oy, Finland providing POSEIDON with chemicals.

Cooperation with companies operating WWTPs. Several wastewater associations were involved in pilot plant experiments and full scale sampling campaigns. They analyzed wastewater samples by themselves and provided the POSEIDON members with infrastructure needed for the pilot plant experiments. Furthermore, the wastewater association of Braunschweig organized the technical and registration part of the Braunschweig symposium with their own employees. In most of the POSEIDON committee meeting they participated without getting any funding from the EU.

Cooperation with waterworks companies such as Wasserverband Hessisches Ried, Biebesheim participating in the monitoring programs and the ILCS. They analyzed samples by themselves and provided an individual training course regarding PPCP analysis for one of the Poseidon partners. In most of the POSEIDON committee meeting they participated without getting any funding from the EU.

51


Training Course for Wastewater Professionals organized by the Swiss Water Pollution Control Association; Emmetten, Switzerland.

Training courses within the Triton project (EVK1-2002-00577) the accompany measures of POSEIDON coordinated by Tampere University, Finland.

• Environmental risk assessment of PPCPs in Gliwice, Poland • Water and wastewater technologies to prevent/minimize the risks caused by PPCPs • Analysis and Fate of PPCPs in the Environment

POSEIDON was invited by the German Federal Environmental Agency (Umweltbundesamt, Berlin) to comment on the Draft Note for Guidance, (EMEA 2003), to the European Commission. Contacts have taken place with several project partners and the Umweltbundesamt in Berlin (end user) for the achievement of data on occurrence of the substances in the environment and on the market.

An eco-labeling for shampoos, shower gels and foam baths (SSBs) as an effective and immediate policy instrument was proposed. The developed basic criteria for an eco-label award for SSBs were presented to the German Federal Environmental Agency and to producer and consumer associations.

Home page (http://www.eu-poseidon.com) was further improved and extended

POSEIDON was dedicated to participate with two partners at the Pelsten workshop (Environmental Risk assessment of pharmaceuticals) organized from SETAC in Snowbird.

POSEIDON was dedicated to participate with two partners at the Gordon Conference (Environment Water) in Plymouth, NH, USA.

A national symposium in Austria was held with the title “Pharmaceuticals in the aquatic environment” in September 2002. This meeting (~85 participants from Austrian governmental and private organizations) was co-organized by IWT together with the ÖWAV (Austrian Association for Water- and Waste Management).

Publications: A number of papers have been published in peer-reviewed and to more public journals that describe the work performed during the project (see next section). Additional papers have been submitted or are in preparation.

Oral Presentations

POSEIDON results were presented in a number of international and national conferences world-wide: e.g. IWA Conferences and meeting in Sydney, Mexico, Marrakech, Aachen etc., IWA Global Conference on Leading Edge Water and Wastewater Treatment Technologies; Jahrestagung Gesellschaft Deutscher Chemiker, München, Germany; Dechema Workshop on “Pharmaceutical micropollutants in aquatic environments – paths, risks and remediation” (Frankfurt, Germany); Unesco-IHE International Masters in Water Management, Duebendorf, Switzerland; International Conference on Pharmaceuticals and Endocrine Disrupting Chemicals in Water, Minneapolis, USA; several SETAC conferences in Vienna, Prague, Orlando; ACS meeting in Philadelphia etc.

A selection of presentations are listed below (not enlisted are contributions to the Pharma-Cluster conference Envirpharm Lyon 14 – 16 April ’03 and to the Poseidon Symposium in Braunschweig 4 – 5 November ‘03): A. Bruchet, C. Hochereau, C. Picard, V. Decottignies, J.M. Rodrigues and M.L. Janex-Habibi (2004) “Analysis of Drugs

and Personal Care Products in French Source and Drinking Waters : The Analytical Challenge and Examples of Application”. IWA Conference, Marrakech, Marocco, Sept 19-24, 2004. Paper + Oral presentation (CIRSEE).

Adriano Joss 11th June 2004, oral presentation and workshop participation:; “Wastewater purification: are we ready for reuse?”; Unesco-IHE International Masters in Water Management, Duebendorf, Switzerland

Adriano Joss 15th September 2003, oral presentation:, “Assessment of Technologies for the removal of Pharmaceuticals”, Ecohazard, IWA Conference, Aachen Germany

52

http://www.eu-poseidon.com/


Adriano Joss 1st – 4th June 2004, poster presentation:, “Removal of estrogens in municipal wastewater treatment under aerobic, anoxic and anaerobic conditions” 2nd IWA Leading-Edge Conference on Water and Wastewater Treatment Technologies (Prague, Czech Republic)

Adriano Joss 1st June 2004, oral presentation:, “Fate of pharmaceuticals, personal care products and hormones in municipal wastewater treatment – a new challenge for wastewater management?” 2nd IWA Leading-Edge Conference on Water and Wastewater Treatment Technologies (Prague, Czech Republic)

Adriano Joss 23rd April 2004, oral presentation and workshop participation:, “Abbau von Mikroverunreinigungen in der dommunalen Abwasserreinigung”; 57th Course for Wastewater Professionals organised by the Swiss Water Pollution Control Association; Emmetten, Switzerland,

Adriano Joss 30th March 2004, oral presentation and workshop participation:; „Elimination von Pharmaka-Verunreinigungen bei der kommunalen Abwasserbehandlung: Relevante Prozesse und Resultate aus dem EU-Projekt Poseidon“; Kooperationsforum Innovation der Wasserwirtschaft Nordrhein-Westfalen (Mühlheim an der Ruhr, Germany)

Adriano Joss 4th March 2004, oral presentation and workshop participation:; “Pharmazeutische Spurenstoffe in Wassersystemen – Transferpfade, Risikobewertung und – verminderung”; Dechema Workshop on “Pharmaceutical micropollutants in aquatic environments – paths, risks and remediation” (Frankfurt, Germany)

Adriano Joss 8th October 2003, oral presentation:, “Arzneistoffe in allen Wässern: Notwendiges Übel oder vermeidbar?“, Jahrestagung Gesellschaft Deutscher Chemiker, München, Germany

Adriano Joss, 28th April 2003, oral presentation: „Micropollutant removal by MBR processes”, Symposium on Applications and perspectives of MBRs in wastewater treatment and reuse (Cremona, Italy)

Christa S. McArdell 2nd April 2004, oral presentation:; “Pharmazeutika – Von der Ausscheidung bis ins Gewässer“; Annual assembly of Swiss Water Pollution Control Association (Olten, Switzerland,)

Clara, M., Fimml, C, Strenn, B., Martinez, E., Kreuzinger, N. (2004). Case study on the behaviour of selected micropollutants during anaerobic sludge digestion. IWA Anaerobic Digestion 10, 29th August – 02nd September 2004, Montreal, Canada.

Clara, M., Strenn, B., Ausserleitner, M., Kreuzinger, N. (2003). Comparison of the behaviour of selected micro pollutants in a membrane bioreactor and a conventional wastewater treatment plant. 4th IWA Specialised Conference on the Assessment and Control of Hazardous Substances in Water – ECOHAZARD 2003, 14th – 17th September 2003, Aachen, Germany, 72/1-72/7

Clara, M., Strenn, B., Fimml, C., Martinez, E., Kreuzinger, N., A case study on the behaviour of selected micropollutants during anaerobic digestion, 29th-2nd Sep 04, IWA 10th World Congress Anaerobic Digestion 2004

Clara, M., Strenn, B., Gans, O., Kreuzinger, N., Carbamazepine, Diclofenac, Ibuprofen and Bezafibrate - Investigations on the behaviour of selected Pharmaceuticals during Wastewater Treatment., 1st –4th Sep 03, 9th IWA Specialised Conference on Design, Operation and Economics of Large Wastewater Treatment Plants, Praha, Czech Republic

Clara, M., Strenn, B., Gans, O., Kreuzinger, N., Investigations on the Behaviour of selected Pharmaceuticals in wastewater treatment, 19th-21st Mar 03, 3rd International Conference on Pharmaceuticals and Endocrine Disrupting Substances, Minneapolis, Minnesota, USA

Clara, M., Strenn, B., Gans, O., Kreuzinger, N., The elimination of selected pharmaceuticals in wastewater treatment – lab scale experiments with different sludge retention times, 30th Apr –4th May 03, 2nd International Conference on Water Resources Management, Las Palmas, Spain

Clara, M., Strenn, B., Kreuzinger, N. (2002). Die Entfernung von hormonell aktiven Substanzen und Pharmazeutika in Kläranlagen: Eine neue Herausforderung an die Abwasserreinigung. Fachtagung zur Eröffnung der 1. Österreichischen kommunalen Membrankläranlage, 01.10.2002, St. Peter ob Judenburg, Austria.

Clara, M., Strenn, B., Kreuzinger, N., Verhalten ausgewählter Pharmazeutika in der Abwasserreinigung, ÖWAV Seminar „Arzneimittel in der aquatischen Umwelt, Wiener Mitteilungen 178, 113-138, Sep 02

Oral presentation at the 87th annual CSC (Canadian Society of Chemistry) Conference, London, Canada, May 29-June 1st 2004. (BfG)

Hansruedi Siegrist 17th September 2003, oral presentation:, “Mikroverunreinigungen – Abwasserentsorgung vor neuen Anforderungen“; EAWAG Information Symposium (Zürich, Switzerland)

Hansruedi Siegrist 23rd April 2004, oral presentation and workshop participation:, “Verhalten von Spurenstoffen in der Siedlungswasserwirschaft”; 57th Course for Wastewater Professionals organised by the Swiss Water Pollution Control Association; Emmetten, Switzerland,

Hansruedi Siegrist 2nd April 2004, oral presentation:; “Mikroverunreinigungen – Abwasserentsorgung vor neuen Anforderungen“; Annual assembly of Swiss Water Pollution Control Association (Olten, Switzerland,)

Huber M.M., Canonica S., and von Gunten U., Oxidative Treatment of Pharmaceuticals in Drinking Waters., 18-20 Apr 2002, AWWA-symposium : Endocrine Disruptors and the Water Industries, Cincinnati, OH (USA)

Janex ML, Bruchet A., Ternes, TA., Bonerz, M., Mc Dowell D., von Gunten, U., Huber, M., Tuhkanen T. “Removal of Pharmaceuticals in the water cycle – POSEIDON European Project”. NGWA 3rd Conference on Pharmaceuticals and Endocrine Disruptors, Minneapolis, March 19-21 2003. Oral presentation (CIRSEE).

53


Kreuzinger, Behaviour and Fate of selected organic micropollutants during transport in the saturated zone of a groundwater layer, International Conference on Bioremediation of soil and groundwater, Cracow, Poland, 5.9. – 8.9.2004

Kreuzinger, N., Clara, M., Strenn, B., Kroiss, H. (2003). Relevance of the sludge retention time (SRT) as design criteria for waste water treatment plants for the removal of endocrine disruptors and pharmaceuticals from waste water. 4th IWA Specialised Conference on the Assessment and Control of Hazardous Substances in Water – ECOHAZARD 2003, 14th – 17th September 2003, Aachen, Germany, 6/1-6/8

Kreuzinger, N., Clara, M., Strenn, B., Vogel, B. (2003). Investigations on the Behaviour of Selected Pharmaceuticals in the Groundwater after Infiltration of treated wastewater. IWA 4th International Symposium on Wastewater Reclamation and Reuse, 12th – 14th. November 2003, Mexico City, Mexico.

Kreuzinger, N., Occurrence of highly discussed Pollutants in the Stretch of the Austrian Danube related to the Catchment Area, SETAC Europe, 12th annual Meeting, Vienna, Mai 2002

Kreuzinger, N., Removal of Bisphenol-A, 17-�-Estradiol and 17-�-Ethinylestradiol during wastewater treatment. Degradation and adsorption to activated sludge, SETAC Europe, 12th annual Meeting, Vienna, Mai 2002

Kroiss, H., Clara, M., Strenn, B., Kreuzinger, N. (2003). Vergleichende Untersuchungen zum Verhalten ausgewählter Spurenschadstoffe in einer konventionellen Kläranlage und einem Membranbioreaktor. In T. Melin und M. Dohmann (eds) Membrantechnik in der Wasseraufbereitung und Abwasserbehandlung – Perspektiven, Neuentwicklungen und Betriebserfahrungen im In- und Ausland, Poster, 5. Aachener Tagung Siedlungswssserwirtschaft und Verfahrenstechnik, 30.09.-01.10.2003, Aachen, Germany.

Lindqvist N, Korhonen S, Jokela, J and Tuhkanen T. 2003. Occurrence of Pharmaceuticals and Personal Care Products in Finnish Surface Waters and Their Removal by Chemical Coagulation. NGWA 3rd Conference on Pharmaceuticals and Endocrine Disrupting Chemicals in Water, March 19-21, 2003, Minneapolis, Minnesota, USA. Oral presentation (TUT)

Lindqvist N., Tuhkanen, T. and Kronberg, L. 2004. Occurrence of acidic pharmaceuticals in Finnish sewage, surface and drinking waters. The annual seminar of Finnish Water and Wastewater Association, June 9-10, 2004, Espoo, Finland. Oral presentation (TUT).

M.Carballa, F.Omil, and J.M.Lema, Removal of Pharmaceuticals and Personal Care Products (PPCPs) from municipal wastewaters by physico-chemical processes, 6th -10th May 03, Environment 2010

M.Carballa, F.Omil, T. Ternes and J.M.Lema, Pharmaceuticals and Personal Care Products (PPCPs) in waters and wastewaters: Poseidon project, 6th -10th May 03, Environment 2010

M.L. Janex-Habibi « Anticiper les évolutions réglementaires : Exemples des Composés Pharmaceutiques », 5ème Carrefour des Gestions Locales de l’Eau, Rennes, France, Jan 28-29, 2004. Oral presentation (CIRSEE).

Miksch K., Rychta U., Woźniak E., Occurrence of the Pharmaceuticals in the Environment, 7th Polish Symposium “Environmental Biotechnology”, 5 –7. December 2001 , Wisła Jarzębata (Poland)

Miksch K., Rychta U., Woźniak E.: Presence of pharmaceuticals in wastewater in Poland. SETAC Europe 12th Annual Meeting “Challenges in environmental Risk Assessment and Modelling: Linking Basic and Applied Research” 12-16 May 2002 Vienna, Austria. Book of Abstracts, p.161.

ML. Janex-Habibi, A. Bruchet (2004) “Pharmaceuticals and Personal-Care Products : occurrence and fate in drinking water treatment”. AWWA Annual Conference, Orlando, June 12-16, 2004. Paper + Oral presentation (CIRSEE).

ML. Janex-Habibi, A. Bruchet, T. Ternes (2004) « Effet des traitements d’eau potable et d’épuration des eaux usées sur les résidus médicamenteux – Résultats du projet Poseidon», 83ème Congrès de l’ASTEE, Aix-Les-Bains, May 24-28, 2004 (CIRSEE). Paper + Oral presentation (CIRSEE).

Ried, A., Mielcke, J., Kampmann, M., Ozone and UV processes for additional wastewater treatment to remove pharmaceuticals and EDC’s, 1st Jun 04, 2nd IWA Leading Edge Conference on Water and Wastewater Treatment Technologies, Prague, Czech Republic

Strenn, B., Clara, M., Gans, O., Kreuzinger, N., The Comportment of Selected Pharmaceuticals in Sewage Treatment Plants., 18th–20th Jun 03, 7th International Conference on Water Pollution, Cadiz, Spain

Strenn, B.; Clara, M.; Gans, O.; Kreuzinger, N.; (2003). Carbamazepine, Diclofenac, Ibuprofen and Bezafibrate - Investigations on the behaviour of selected Pharmaceuticals during Wastewater Treatment. 4th IWA Specialised Conference on the Assessment and Control of Hazardous Substances in Water – ECOHAZARD 2003, 14th – 17th September 2003, Aachen, Germany, 18/1-18/7.

Thomas Ternes. 2 Vorträge: A) Beurteilung pharmazeutische Spurenstoffe in Gewässern aus Sicht der Trinkwasseraufbereitung, B) Verminderung des Eintrages in die Umwelt “Pharmazeutische Spurenstoffe in Wassersystemen – Transferpfade, Risikobewertung und – verminderung”; Dechema Workshop, März 2004

Thomas Ternes. Behaviour of steroid hormones in wastewater treatment SACH CONFERENCE 2003 „Analytical Chemistry of Organic Contaminants in the Environment“ Zurich, Switzerland, September 3 - 5, 2003

Thomas Ternes. Die Wiederverwendung von Abwasser: Wo bleiben Arzneimittelrückstände und Mikroorganismen? Dresdner Wasserseminar, Juli 2002

Thomas Ternes. Elimination von Arzneimittelwirkstoffen bei der Abwasserreinigung. ATV-DVWK Seminar “Weitergehende Abwasserreinigung” Kassel, April 2004.

Thomas Ternes. Entfernung von Pharmaka und endokrinen Substanzen in der Abwasserreinigung, 41. Tutzinger-Symposium 2003 “Pharmazeutische Reststoffe und endokrinwirksame Verbindungen in Gewässern” 2003

54


Thomas Ternes. Fate and removal of pharmaceuticals and endocrine disrupters by the irrigation of treated wastewater on agricultural fields. European Conference on Human and Veterinary Pharmaceuticals in the environment. 14-16. April 2003, Lyon.

Thomas Ternes. Fate of estrogens in a municipal sewage treatment plant. 10. Workshop des DFG-Graduierten Kollegs AGEESA “Occurrence and elimination of trace contaminants in waste water treatment and environmental compartments” Aachen, April, 2004.

Thomas Ternes. Micropollutant removal a challenge for the cooperation between analytical chemists and environmental engineers, Gordon Conference: Environmental Sciences, Plymouth, New Hampshire from June 27-July 2, 2004

Thomas Ternes. Pharmaceuticals and Personal care products: Are there consequences for European regulations? Conference on science, environment and development: accepting research challenges to reduce societal risks, Murcia, 29-31 May 2002 (eingeladen zu Vortrag und Platform-Diskussion).

Thomas Ternes. Pharmaceuticals and related compounds in aqueous samples. 100 Years of Chromatography Vol. 1000- Journal of Chromatography, Ermlo, The Netherlands, Juni 2003

Thomas Ternes. Pharmaceuticals in the environment: Presence, Fate and Removal (PHARMA Cluster), International Conference "Science in Support of European Water Policies - Sustainability of Aquatic Ecosystems" (AQUAECO), Nov. 2002

Thomas Ternes. Pharmaceuticals in wastewater and surface water, Wasser Berlin, April 2003 invited. Thomas Ternes. Pharmaceuticals, musk fragrances and estrogens: Removal in wastewater and drinking water treatment.

4th International Conference on Pharmaceuticals and Endocrine Disrupting Chemicals in Water, Minneapolis/USA, Oct., 2004

Thomas Ternes. Sorption onto sludge from municipal STPs: a relevant process for the removal of pharmaceuticals and musk fragrances? 228th ACS National Meeting Philadelphia/USA, August 2004

Urs von Gunten 2nd April 2004, oral presentation:; “Ozonung des ARA-Ablaufs – Eine sinnvolle Sofortmassnahme?”; Annual assembly of Swiss Water Pollution Control Association (Olten, Switzerland,)

Zessner, M., Vogel, B., Clara, M., Kavka, G., Kroiss, H. Monitoring of Influences on Groundwater Caused by Infiltration of Treated Waste Water IWA 4th International Symposium on Wastewater Reclamation and Reuse, Mexico City, Mexico, 12.11.-14.11.2003

55


1.6 Publications Peer Reviewed Articles (39) of the POSEIDON project Andersen H., Siegrist H., Halling-Sørensen B., Ternes T.A.: The fate of estrogens in a municipal sewage treatment plant.

Environ. Sci. Technol. 37, 4021-4026 (2003). Bruchet A., Hochereau C., Picard, C., Decottignies V., Rodrigues J.M., Janex-Habibi M.L.: Analysis of drugs and

personal care products in French source and drinking waters: the analytical challenge and examples of application. Water Science and Technology 52, 53-61 (2005)

Calmano W., Bilitewski U., Flemming H.-C., Hofmann T., Peiffer S., Ternes T.A., Wilken R.-D.: The German water chemical society: actual trends and fields of research in the principle committee „Basic Research“. Acta Hydrochim. Hydrobiol. 29, 419-427 (2002).

Carballa, M., Omil, F., Alder, A.C. and Lema, J.M. Comparison between the conventional anaerobic digestion (CAD) of sewage sludge and its combination with a chemical or thermal pre-treatment concerning the removal of Pharmaceuticals and Personal Care Products (PPCPs). Water Science and Technology, in press (2006).

Carballa, M., Omil, F., Lema, J.M., Llompart, M., García, C., Rodríguez, I., Gómez, M. and Ternes, T. (2005). Behaviour of pharmaceuticals and personal care products in a sewage treatment plant of northwest Spain. Water Science and Technology, 52 (8): 29-35.

Carballa M., Omil F., Lema J.M.: Removal of cosmetic ingredients and pharmaceuticals in sewage by coagulation-flocculation and flotation processes. Water Res. 39,4790-4796 (2004).

Carballa M., Omil F., Lema J., Llombart M., Fracia-Jares C., Rodrigues I., Gomez M., Ternes T.A.: Behavior of pharmaceuticals, cosmetics and hormones in sewage treatment plants. Water Res. 38, 2918-2926 (2004).

Clara M., Strenn B., Gans O., Martinez E., Kreuzinger N., Kroiss H.: Removal of selected pharmaceuticals, fragrances and endocrine disrupting compounds in a membrane bioreactor and conventional wastewater treatment plants. Water Research 39, 4797-4807 (2005)

Clara M., Strenn B., Ausserleitner M., Kreuzinger N.: Comparison of the behaviour of selected micro pollutants in a membrane bioreactor and a conventional wastewater treatment plant. Water Science and Technology 50(5), 29-36 (2004).

Clara M., Kreuzinger N., Strenn B., Gans O., Kroiss H.: The solids retention time—a suitable design parameter to evaluate the capacity of wastewater treatment plants to remove micropollutants. Water Res. 39, 97-106 (2004),

Clara M., Strenn B., Kreuzinger N.: 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(4), 947-954 (2004).

Fellis, E., Miksch, K., Surmacz-Gorska, J. Ternes, T.A., Presence of pharmaceutics in wastewater from waste water treatment plant "Zabrze-Śrόdmieście" in Poland, Archives of Environmental Protection, 31, No. 3, 49 – 58 (2005)

Göbel A., Thomsen A., McArdell C.S., Joss A., Giger W.: Occurrence and Sorption Behavior of Sulfonamides, Macrolides and Trimethoprim in Activated Sludge Treatment. Environ. Sci. Technol. 39, 3961-3989 (2005a).

Göbel A., Thomson A., McArdell C.S., Alder A.C., Giger W., Theiβ N., Loeffler D., Ternes T.A.: Extraction of Sulfonamide and Macrolide Antimicrobials from Sewage Sludge. J. Chromatogr. A 1085, 179-189 (2005).

Göbel A., McArdell C.S., Suter M.J.-F., Giger W.: Trace determination of macrolide and sulfonamide antimicrobials, a human sulfonamide metabolite, and trimethoprim in wastewater using liquid chromatography coupled to electrospray tandem mass spectrometry. Anal. Chem. 76(16), 4756-4764 (2004).

Huber, M. M., Korhonen, S., Ternes, T., von Gunten, U.: Oxidation of pharmaceuticals during water treatment with chlorine dioxide. Water Res. 39, 4290-4299 (2005)

Huber, M.M., Goebel, A., Joss, A., Hermann, N., Kampmann, M., Löffler, D., McArdell, M.S., Ried, A., Ternes, T.A. and von Gunten, U., Oxidation of Pharmaceuticals during Ozonation of Municipal Wastewater Effluents: A Pilot Study. Environ. Sci. Technol. 39, 4290-4299 (2005).

Huber M.C., Ternes T.A., Von Gunten U.: Removal of estrogenic activity and formation of oxidation products during ozonation of 17α-Ethinylestradiol. Environ. Sci. Technol. (2004) 38, 5177-5186.

Huber M.M., Canonica S., Park G.-Y., von Gunten U.: Oxidation of Pharmaceuticals during Ozonation and Advanced Oxidation Processes (AOPs). Environ. Sci. Technol. 37(5), 1016-1024 (2003).

Joss A., Zabczynski, S., Göbel, A., Hoffmann, B., Löffler, D., McArdell, C.S., Ternes, T.A., Thomsen, A., Siegrist, H., , Biological degradation of pharmaceuticals in municipal wastewater treatment: Proposing a classification scheme, Wat. Res. 40, 1686-1696 (2006)

Joss A., Ternes T.A., Alder A., McArdell C.S., Göbel A., Keller E., Siegrist H.: Removal of pharmaceuticals and fragrances in biological wastewater treatment. Water Res. 39, 3139-3152 (2005).

Joss A., Andersen H., Ternes T.A., Richle P.R., Siegrist H.: Removal of estrogens in municipal wastewater treatment under aerobic and anaerobic conditions: consequences for plant optimization. Environ. Sci. Technol. 38, 3047-3055 (2004).

Klaschka U., Liebig M., Knacker T.: Ausgezeichnete Produkte – weniger Umweltbelastung durch Umweltzeichen. Chemie in unserer Zeit, 1-23 (2003).

56


Kreuzinger N., Clara M., Strenn B., Vogel B.: Investigations on the Behaviour of Selected Pharmaceuticals in the Groundwater after Infiltration of treated wastewater. Water Sci. Technol. 50 (2), 221-228 (2004).

Kreuzinger N., Clara M., Strenn B., Kroiss H.: Relevance of the sludge retention time (SRT) as design criteria for waste water treatment plants for the removal of endocrine disruptors and pharmaceuticals from waste water. Water Sci. Technol, 50, 149-156 (2004).

Larsen, T.A., Lienert, J., Joss, A., Siegrist, H., How to avoid pharmaceuticals in the aquatic environment. Journal of Biotechnology 113, no. 1-3, 295-304 (2004).

Liebig, M., Egeler, P., Oehlmann, J. and Knacker, T. Bioaccumulation of 14C-17α-ethinylestradiol by the aquatic oligochaete Lumbriculus variegatus in spiked artificial sediment. Chemosphere 59(2), 271-280 (2005a).

Liebig, M., Moltmann, J.F. and Knacker, T. Evaluation of measured and predicted environmental concentrations of selected human pharmaceuticals and personal care products. Environ. Sci. Poll. Res. online first. DOI: http://dx.doi.org/10.1065/espr2005.08.276 (2005b).

Lindqvist N., Tuhkanen T., Kronberg L.: Occurrence of acidic pharmaceuticals in raw and treated sewages and in receiving waters. Water Res. 39, 2219-2228 (2005).

McDowell D.C., Huber M.M., Wagner M., von Gunten U. and Ternes T.A.: Ozonation of carbamazepine in drinking water: Identification and kinetic study of major oxidation products. Environ. Sci. Technol. 39, 8014-8022 (2005).

Strenn B., Clara M., Gans O., Kreuzinger N.: Carbamazepine, Diclofenac, Ibuprofen and Bezafibrate - Investigations on the behaviour of selected Pharmaceuticals during Wastewater Treatment. Water Sci. Technol. 50, 269-276 (2004).

Ternes T.A., Bonerz M., Herrmann N., Keller E., Bagó Lacida B., Alder, A.C.: Determination of pharmaceuticals, iodinated contrast media and musk fragrances in sludge by LC tandem MS and GC/MS. J. Chrom. A 1067, 213-223 (2005).

Ternes, T.A., Joss, A. Siegrist, S. Pharmaceuticals and personal care products: wastewater practice under close scrutiny. Environ. Sci. Technol. October 15, 393A-399A (2004).

Ternes T.A., Herrmann N., Bonerz M., Knacker T., Siegrist H., Joss A.: A rapid method to measure the solid-water distribution coefficient (Kd) for pharmaceuticals and musk fragrances in sewage sludge. Water Res. 38, 4075-4084 (2004).

Ternes T.A., Joss A. Siegrist S.: Pharmaceuticals and personal care products: wastewater practice under close scrutiny. Environ. Sci. Technol. 15, 393A-399A (2004).

Ternes, T.A., Joss, A., Siegrist, H.: Scrutinizing pharmaceuticals and personal care products in wastewater treatment. Environmental Science & Technology 38, no. 20, 392A-399A (2004).

Ternes T.A., Stüber J., Herrmann N., McDowell D., Ried A., Kampmann M., Teiser B.: Ozonation: A tool for removal of pharmaceuticals, contrast media and musk fragrances from wastewater? Water Res. 37, 1976-1982 (2003).

Ternes T.A., Meisenheimer M., McDowell D., Sacher F., Brauch H.-J., Haist-Gulde B., Gudrun Preuss G., Wilme U., Zulei-Seibert N.: Removal of pharmaceuticals during drinking water treatment. Environ. Sci. Technol. 36, 3855-3863 (2002).

Ternes T.A., Andersen H. R., Gilberg D., Bonerz M.: Determination of estrogens in sludge and sediments by liquid extraction and GC/MS/MS. Anal. Chem. 74, 3498-3504 (2002).

Non-Peer Reviewed Articles (2) of the POSEIDON project Hansruedi Siegrist, Adriano Joss, Thomas Ternes, Jörg Oehlmann. Fate of EDCS in Wastewater Treatment and EU

Perspective on EDC Regulation, Surface Water Quality & Ecology: EDCs in Wastewater: Implications for the Water Quality Industry Session 38, 78th Annual Technical Exhibition and Conference, Water Environment Federation WEFTEC, 29th Oct.- 2nd Nov. 2005, Washington DC.

Ternes, T., Joss, A., Kreuzinger, N., Miksch, K., Lema, J.M., von Gunten, U., McArdell, C.A., Siegrist, H. (2005),

Removal of pharmaceuticals and personal care products - Results of the Poseidon Project, Industrial Issues & Technical Treatment Session 2, 78th Annual Technical Exhibition and Conference, Water Environment Federation WEFTEC, 29th Oct.- 2nd Nov. 2005, Washington DC.

Submitted (5)

Carballa, M., Omil, F. and Lema, J.M.: Mass balance of Pharmaceutical and Personal Care Products (PPCPs) in a Sewage Treatment Plant. Environ. Sci. and Technol., submitted (2006).

Carballa, M., Manterola, G., Larrea, L., Ternes, T.A., Omil, F. and Lema, J.M. (2006). Influence of ozone pre-treatment on anaerobic digestion operation and digested sludge characteristics: removal of pharmaceutical and personal care products. Chemosphere, submitted (2006)

Carballa, M., Omil, F., Ternes, T.A. and Lema, J.M.: Fate of Pharmaceutical and Personal Care Products (PPCPs) during anaerobic digestion of sewage sludge. Water Res., submitted (2006).

Göbel, A., McArdell, C.S., Joss, A., Siegrist, H., Giger, W.: Behavior of Sulfonamides, Macrolides and Trimethoprim in Different Wastewater Treatment Technologies. Chemosphere, submitted (2005).

57


Ternes T.A., Bonerz M., Herrmann N., Teiser B., Andersen H.: Irrigation of treated wastewater in Braunschweig, Germany: An option to remove pharmaceuticals and musk fragrances. Water Res., submitted (2006).

In preparation (6)

Carballa, M., Fink, G., Lema, J.M. and Ternes, T.A. (2006). Determination of the solid-water distribution coefficient (Kd) of Pharmaceutical and Personal Care Products (PPCPs) in digested sludge. Water Res., in preparation.

Carballa, M., Manterola, G., Larrea, L., Ternes, T.A., Omil, F. and Lema, J.M. (2006). Influence of ozone pre-treatment on anaerobic digestion operation and digested sludge characteristics: removal of pharmaceutical and personal care products. Chemosphere, in preparation

McDowell D.C., Wagner M. and Ternes T.A.: Removal of diclofenac during drinking water production and identification of its oxidation products after ozonation. Water Res., in preparation.

Knacker T., Moltmann J.F., Liebig M., Egeler P.: Environmental Risk Assessment (ERA) for pharmaceuticals and personal care products. Case studies and reflections on an increased reliability of the ERA scheme. Environ. Sci. Technol., in preparation.

Liebig M., Egeler P., Moltmann J.F., Knacker T.: Ecotoxicological data for four pharmaceuticals (Iopromid, 17α-Ethinylestradiol, Sulfamethoxazole, Carbamazepine) and one personal care product (Tonalide) for an Environmental Risk Assessment. Environ. Sci. Technol., in preparation..

Zabczynski S., Surmacz-Gorska J., Miksch K.: Fate of PPCPs in sequencing batch reactor, Water Res. In preparation Books and other publications (11)

Andersen H., Ternes T.A., Halling-Soerensen B.: Steroidoestrogner i renset spildevand kann vaere arsag til effekter pa danske fisk Dansk keni. Dansk kemi 83, nr.2 (2002).

Clara M., Strenn B., Gans O., Kreuzinger N. (2003): The elimination of selected pharmaceuticals in wastewater treatment – lab scale experiments with different sludge retention times. In C.A. Brebbia (ed.), Water Resources Management II, WIT Press, Southampton, UK, ISBN 1-85312-967-4, 227 - 236.

Clara M., Strenn B., Kreuzinger N. (2002): Zum Verhalten ausgewählter Pharmazeutika bei der Abwasserreinigung. In H. Kroiss (ed.), Arzneimittel in der aquatischen Umwelt, Wiener Mitteilungen 178, Institute for Water Quality and Waste Management, Vienna, Austria, ISBN 3-58234-069-1, 113-138.

Clara M., Strenn B., Fimml C., Martinéz E., Kreuzinger N. A case study on the behaviour of selected micropollutants during anaerobic sludge digestion, 10th World Congress - Anaerobic Digestion 2004, 29. Aug.–2. Sept. 2004, Montréal – Canada.

Janex M.L., Bruchet A., Lévi Y., Ternes T. : Composes Pharmaceutiques : Presence Dans L’Environnement Et Devenir En Traitement D’Eau Potable In Journées Information Eau, Vol., Poitiers, 2002.

Janex M.L., Lévi Y., Ternes T.A.: Pharmaceutical compounds: Occurrence in the environment and fate in drinking water treatment. American Waterworks association – Water Quality Technology Conference, 9-13 Nov 2002, Seattle/USA.

Klaschka U., Liebig M., Moltmann J.F., Knacker T. (2004): Potential environmental risks by cleaning hair and skin. Eco-Label – A possibility to reduce exposure to personal care products. In: Kümmerer K. (ed.): Pharmaceuticals in the environment. Springer, Berlin.

Knacker T., Moltmann J.F., Duis K., Liebig, M. (2003): Unterschiede in der Umweltrisikobewertung von Pharmaka und Industriechemikalien – Konsequenzen für endokrin wirksame Substanzen. 'Differences in the environmental risk assessment of pharmaceuticals and industrial chemicals – consequences for endocrine disrupting substances. In: Track, Th. & Kreysa, G. (eds.): Spurenstoffe in Gewässern – Pharmazeutische Reststoffe und endokrin wirksame Substanzen. Wiley-VCH, Weinheim.

Kroiss H., Clara M., Strenn B., Kreuzinger N. (2003): Vergleichende Untersuchungen zum Verhalten ausgewählter Spurenschadstoffe in einer konventionellen Kläranlage und einem Membranbioreaktor. In T. Melin und M. Dohmann (eds) Membrantechnik in der Wasseraufbereitung und Abwasserbehandlung – Perspektiven, Neuentwicklungen und Betriebserfahrungen im In- und Ausland, Department of Environmental Engineering, Aachen, Germany, ISBN 3-921955-28-9, P19.

Siegrist H., Joss A., Alder A., McArdell-Bürgisser Ch., Göbel A., Keller E., and Ternes T.A.: Micropollutants – New Challenge in Wastewater Disposal? EAWAG News, 57e, 7-10 (2003).

Strenn B., Clara M., Gans O., Kreuzinger N. (2003): The comportment of selected pharmaceuticals in sewage treatment plants. In C.A. Brebbia, D. Almorza, D. Sales (eds.), Water Pollution VII, Modelling, Measuring and Prediction, WIT Press, Southampton, UK, ISBN 1-85312-976-3, 273 - 282.

58

project acronym poseidon contract no. evk1-ct-2000-00047 · juan m. lema, marta carballa, maria...

Documents