Integrated stochastic modeling of pharmaceuticals in sewage networks

Pharmaceuticals and particularly antibiotics can harm sensitive aquatic species. Their occurrence in urban wastewater systems is the consequence of five successive processes: (i) ingestion of the substance, (ii) accumulation in the urine, (iii) excretion, (iv) degradation in the sewer system and (v) transport to the wastewater treatment plant (WTP). These processes were included in an integrated model that can be used to assess the dynamics of pharmaceuticals at a WTP inlet. First, information on sales data, posology, pharmacokinetics and toilet flushing frequency were combined to create a source model of pharmaceuticals entering a sewer system. This production function was then coupled with a transport/degradation model to simulate concentrations of pharmaceuticals at a WTP inlet. In an example application, the full model was applied to simulate the concentration of the antibiotic ciprofloxacin on an hourly time scale. In this application, the model was calibrated and validated for a case study at a WTP in Lausanne, Switzerland. Validation of the integrated approach was successful despite the high variability evident in the model results. This modeling approach has potential use in pollution management and epidemiology related to wastewater.

A strategy for xenobiotic removal using photocatalytic treatment, microbial degradation or integrated photocatalytic-biological process

Miléna Lapertot

According to the limited natural resources and due to the risks of anthropogenic pollution, it appears necessary to react efficiently in order to remove existing contaminations and avoid the creation of new ones. Therefore, the purpose of this thesis is to propose a sustainable strategy for treating problematic pollutants with the most adequate process. First, an overview of the different treatment processes has been given. In particular, biological, photocatalytic and integrated biological-photocatalytic treatments have been described as efficient methods to remove xenobiotics. Biological treatment of contaminated environments and industrial effluents has been presented as the most attractive method on a cost-benefit extent. But for biorecalcitrant substances, the Advanced Oxidation Processes (AOP) have remained the most efficient methods although expensive. Therefore, the coupling of biological-photocatalytic has emerged as the best compromise for removal of recalcitrant xenobiotics. The first step of the treatment strategy has consisted of evaluating the biotraitability of 19 chemicals: pesticides, pharmaceuticals and volatile organic compounds (VOCs) have been assessed according to three biodegradability methods (Zahn-Wellens, Manometric Respirometry, Closed-Bottle tests), whose official protocols had to be adapted to the compound specificities (volatility, hydrophobicity). Structure-Activity Relationship (SAR) estimation models have also been used to compare and validate the experimental results. For the compounds which have been assessed as biotraitable (VOCs), further biodegradation studies have been led in order to improve the treatment (batch with pulses of substrate): automated monitoring of the biodegradation process, increased degradation yields and biomass productivity have been successfully completed for Toluene, Ethylbenzene, Xylenes, Chlorobenzenze and Dichlorobenzene removal. Conversely, the compounds assessed as biorecalcitrant have been oriented towards the TiO2 and photo-Fenton photocatalytic treatments. Thus, the degradation of four p-halophenols by TiO2 photocatalysis has been investigated and has demonstrated the halogen effect on removal rates, aromatic intermediates and toxicity variations. Finally, the treatment of five bioreacalcitrant pesticides (Alachlor, Atrazine, Chlorfenvinphos, Diuron, Pentachlorophenol) has been studied in order to develop a coupled photocatalytic-biological system: first, enhanced biotreatability has been evaluated using the Zahn-Wellens procedure and then fixed-bed bioreactors have been operated and permitted to find out the best moment for coupling.

EPFL2006

Sources, Spreading and Fate of Antibiotic Resistance Genes and Resistant Bacteria in Vidy Bay, Lake Geneva, Switzerland

Nadine Czekalski

Increasing numbers of multiresistant bacterial pathogens cause severe health problems on a global scale. Large quantities of bacteria and their resistance genes are discharged into natural environments via human and animal waste. Not much is known so far about the fate and further spread of resistance determinants in the environment. There is evidence that aquatic environments may function as important compartments for preserving and exchanging antibiotic resistance genes (ARG) from sewage and drinking water. Lake Geneva is the largest fresh water reservoir of Western Europe and supplies the surrounding cities. At the same time these cities discharge sewage into the lake. Lausanne is the largest city in the lake’s catchment and discharges the highest amount of treated, as well as in case of heavy rain events, partly untreated waste waters. Also two rivers, Chambering and Flown, transport untreated sewage (bypass), in case of storm events. Only 3.4 km from the sewage discharge point lake water is pumped for drinking water preparation for Lausanne. Many studies have reported on contamination of the bay with heavy metals, inorganic nutrients, organic carbon and fecal indicator bacteria due to the sewage discharge. The present thesis aimed on investigating the sources, spreading and fate of antibiotic resistant bacteria and antibiotic resistance genes in the Vida bay, Lake Geneva. In the first part of this dissertation, a screening for antibiotic- and specifically multi-resistant bacteria was carried out as one of the first studies in a Swiss waste water stream and the receiving fresh water lake. For this purpose, a combined set of culture based (viable bacterial counts on nutrient media supplemented with antibiotics) and molecular tools (detection and quantitation of antibiotic resistance genes from environmental DNA-extracts using quantitative real-time PCR) was applied. Raw hospital and communal sewage, as well as partly treated/mixed sewage, which is discharged into the lake via an effluent pipe 700 m off shore, were taken into account. Furthermore lake water samples nearby the end of the waste water discharge pipe and the drinking water pump as well as close to the two river mouths were analyzed. Finally, the role of Lausannes WTP was evaluated for its potential to remove resistant and multiresistant bacteria, as well ARG. Results show highest levels of multiresistant bacteria and ARG in both hospital and communal sewage. Incidence was found that the WTP is a selective environment for multiresistant bacteria and resistance genes. Moreover, lake water and sediments nearby the end of the WTP discharge pipe and the two river mouths exhibited significantly higher levels of resistant determinants as compared to samples from the drinking water uptake. If the level of antibiotic resistance close to the drinking water uptake represents the natural background level of the lake or whether it is impacted by the WTP plume remained open during this first survey. The spatial distribution and persistence of ARG discharged via waste water into lake systems has been evaluated for the first time in the second part of this thesis. 22 sediment cores were sampled along two major transects (leading either into the deep lake or along the south-western shore line to the drinking water uptake) and in the vicinity of the WTP discharge pipe in the Vidy bay. A subset of samples was also analyzed for contamination with a “conventional” pollutant of sewage (mercury), for which a similar transport behavior (particle bound) as for bacteria and their resistance genes is assumed. The microbial community composition in all sediment as well as in waste water samples was analyzed using ARISA. Strong pollution with both ARG and mercury was detected in close proximity of the WTP discharge pipe and both contaminants significantly decreased along the transect into the deep lake. In general however, the spatial contamination of mercury and ARG was only moderately correlated. While the mercury contaminated sediment surface area tended to have a pollution tail in south eastern direction, ARG showed only a slow decrease along the south western transect. These findings imply differences in transport and fate of mercury and ARG in the bay, or additional sources of ARG contamination. Analysis of the microbial diversity in the samples support these hypotheses. This study revealed feasibility of qPCR as a tool to semi-quantitatively track spatial ARG contamination in fresh water lake sediments. Not much is known yet on the antibiotic resistance background level of fresh water lakes and how this may be influenced by anthropogenic activities in the surrounding catchment. Water samples from 21 Swiss lakes have been collected and were analyzed for the prevalence of different ARG (2 sulfonamide resistance genes sul1 & sul2, 3 tetracycline resistance genes tetB, tetM, tetW and one plasmid mediated fluoroquinolone resistance gene qnrA). In order to document the abundance of these genes in the bacterial populations, they were normalized to bacterial 16S rRNA gene copy numbers. The lakes differ in size, geographical position, land use pattern, and trophical status. The composition of the microbial community of all lakes was analyzed using ARISA. Sulfonamide resistance genes were detected in all investigated lakes (0.2-4%, sul1 and 0.01-0.3%, sul2), with highest levels in Lake Baldegger (21%). Lake Baldegger also contained all of the three tetracycline resistance genes, which were found in only 4 other lakes. None of the lakes was positive for qnrA. Important anthropogenic catchment activities linked to sul gene prevalence could not be identified based on our data and may require further analysis and an additional variables to complete our data set. However, our study demonstrates contamination of Swiss lakes with sulfonamide resistance genes and the potential of fresh water lakes preserve these genes, even remote from direct sources of pollution.

EPFL2013

Origines et flux de biocides et de filtres UV dans les stations d'épuration des eaux usées

Cécile Plagellat

Adverse effects on the environment have become a matter of increasing concern considering the increase of industrialization and of the amount of commercialized chemicals. At present, approximately 100'000 substances are on the market, 3000 are sold at high volumes. In Europe, 200 to 300 new substances are authorized every year (in Switzerland, approximately 50). Contamination is induced by consumption of numerous products frequently applied by private or professional users. Persistence and ecotoxicity of certain organic compounds contained in these products might have impacts on the quality of our environment. Knowledge on potential sources of these substances is the first step in order to limit their release. The project developed within the present PhD thesis aims at determining the sources of certain compounds by applying material flux analysis on the basis of sewage sludge. Presently, a large number of anthropogenic organic compounds originating from private households, craft industry, industry and stormwater is disposed of into the collector system and ends up in sewage sludge. Carbendazim, diuron, Irgarol 1051®, octhilinone, permethrin, tributyltin (TBT) and triphenyltin (TPT) comprise numerous applications as biocides and pesticides whilst methylbenzylidene camphor (4-MBC), octyl methoxy cinnamate (OMC), octocrylene (OC) and octyl triazone (OT) are used as UV-screens (type UVB) in a large number of cosmetic products. Their toxic effects on non-target organisms and their potential for bioaccumulation give rise to environmental concern. These substances are used as ingredients in products of daily use and are disposed of into wastewater. Due to their physico-chemical properties they are more or less lipophilic and are likely to persist in sewage sludge. On the basis of the analysis of sewage sludge samples from well characterized sites (catchments and wastewater treatment plants, WWTPs) it is possible to determine the sources (private households, craft industry, industry and stormwater). The first step of this study was the development of analytical methods which allow for quantification of these compounds in sewage sludge. Due to their different physico-chemical properties, four different methods were required. These methods have recovery rates ranging between 75 and 106 % and limits of detection in the order of μg/kg dry matter. The analysis of sludge from 12 sites obtained within 2 sampling periods show that most of the substances are present in all samples. Mean concentrations were 6,8 μg/kg dm (dry matter) for carbendazim, 9,5 μg/ kg d.m. for diuron, 98,1 μg/kg d.m. for permethrin, 148 μg/kg d.m. for TBT, 1777 μg/kg d.m.for 4-MBC, 110 μg/kg d.m. for OMC, 4834 μg/kg d.m. for OC and 5517 μg/kg d.m. for OT. Octhilinone was not detected whilst Irgarol 1051® and TPT were found in 7 and 11 samples, respectively. The transfer of these compounds from wastewater into sludge is an important parameter for material flux analysis on the basis of sewage sludge. A study on the fate of carbendazime and permethrin during wastewater treatment has been carried out on two WWTPs. Whilst removal of permethrin was high (>94%) up to 70% of carbendazime present in sewage were found in treatment plant effluents. Adsorption is an important process for removal of permethrin. However, only 4 to 15 % of the load in wastewater was transferred to sewage sludge. Carbendazim is mainly associated to the aqueous phase and thus, a very small quantity ends up in sludge (approximately 1%). Therefore, the loads found in sewage sludge are more or less representative for consumption of the compounds in the catchments according to their physico-chemical properties. Additionally, it has been shown that the part of permethrine ending up in sludge depends on the treatment techniques of the WWTPs. Thus, for comparison of the pollutant loads in sewage sludge between different sites their fate in the WWTPs has to be known. The loads per inhabitant of the three well characterized types of sites have been calculated in order to determine the sources of the substances. Type A comprises sites with a separate sewer system and a catchment where mainly private households are present. Type B has a combined sewer system collecting sewage from private households, craft industry and stormwater whilst type C comprises a combined sewer system which includes effluents from private households and industry as well as stormwater. The results are as follows: Stromwater is considered as main source of carbendazim in wastewater whilst diuron originates mainly from stormwater and from industrial effluents. Private households contribute significantly to the loads of permethrine and TBT in sewage. For UV-screens the loads from private households are also important but stormwater and industrial effluents are likely to be potential sources as well. A difference in consumption between rural (type A) and more urban sites (type B and C) is possible which might hamper a proper estimate of the sources of UV-screens.

EPFL2004