Search

The occurrence of contaminants in environmental matrices is a major concern and particular attention should be paid to analytical methods currently available. Therefore, this report offers a short bibliography note on the application of highresolution mass spectrometry (HRMS) for the non-targeted analysis of organic contaminants in sediments and biological samples. A bibliometric survey shows that HRMS gains progressively in importance over the period 1990-2010 and became even more significant over the last decade. However, the distribution of publications reveals a wider HRMS application on water samples, and particularly freshwater samples, compared to sediments or biological samples. The scientific literature indicates that several techniques (sonication, microwave, pressurized liquid…) are used for sample preparation before HRMS analysis. It seems that sample preparation protocols are those usually applied for the conventional targeted analysis without further optimization for non-targeted analysis. Scientific literature also shows that for the optimal detection of organic contaminants in a sample, it would be necessary to combine the HRMS analysis with both liquid chromatography (LC-HRMS) and gas chromatography (GC-HRMS) interface. Moreover, for each type of chromatography, it would also be necessary to apply different ionization methods. However, such a holistic detection is not performed and, in practice, research is usually carried out using a single chromatographic separation and a single ionization technique. Indeed, a single HRMS analysis generates already a very large amount of data that must be processed using automated workflows that differ according to the objective of the study. The development of relevant quality controls is also a challenge with HRMS analysis since there is no analytical standard allowing correcting the signals of unknown chemicals that are only identified a posteriori according to different approaches, more or less complex. Nonetheless, initial applications of LC-HRMS and GC-HRMS for the analysis of organic contaminants in sediments and biological samples proved efficient since several studies have demonstrated the occurrence of multiple classes of contaminants with various properties as well as new substances not detected during the conventional targeted analysis., A l’heure ou l’occurrence des contaminants dans l’environnement est une préoccupation sociétale majeure, il faut s’interroger sur les méthodes analytiques disponibles. Ce rapport se propose donc une note de synthèse bibliographique portant sur l’application de la spectrométrie de masse à haute résolution (HRMS) pour l’analyse dite non-ciblée des contaminants organiques dans des échantillons environnementaux sédimentaires et biologiques. Une étude bibliométrique montre que la HRMS a connu un premier essor entre les années 1990 et 2010, puis un autre encore plus important au cours de la dernière décennie. Cependant la distribution des publications révèle une application nettement plus étendue de la HRMS sur la matrice eau, et en particulier eau douce, que sur la matrice sédiment ou biote. La littérature scientifique montre que plusieurs techniques (ultrasons, microondes, liquide pressurisé…) sont employées pour la préparation d’échantillons avant l’analyse instrumentale. Cependant, il semble que les protocoles d’extractions proviennent de développements réalisés pour une analyse ciblée puis qu’ils soient directement transposés à l’analyse HRMS sans plus d’optimisation. La littérature scientifique montre également que pour une détection optimale des contaminants organiques dans un échantillon, il serait indispensable de combiner une analyse HRMS avec une séparation chromatographique préalable en phase liquide (LCHRMS) d’une part et en phase gazeuse (GC-HRMS) d’autre part. De plus, pour chaque interface il faudrait également tester différents types d’ionisation. Cependant, bien que ces analyses avec différentes interfaces chromatographiques et différentes techniques d’ionisation soient complémentaires, les études réalisant des analyses HRMS se focalisent généralement sur une seule séparation chromatographique et un seul type d’ionisation. En effet, une seule analyse HRMS génère déjà un très large volume de données dont le traitement doit encore être automatisé en suivant une succession de taches dont la nature et l’ordre varient selon l’objectif de l’étude. Enfin, le développement et la mise en oeuvre de contrôles qualité est également un challenge lors de l’analyse HRMS car il n’y a pas d’étalons analytiques permettant de corriger le signal de composés inconnus et qui ne sont identifiés qu’a posteriori selon diverses approches plus ou moins complexes à mettre en oeuvre. Néanmoins, les premières applications de la LC-HRMS et GC-HRMS pour l’analyse des contaminants organiques dans les sédiments et les échantillons biologiques se sont révélées positives car plusieurs études ont mis en évidence la présence de multiples classes de contaminants aux propriétés diverses ainsi que de nouvelles substances non détectées lors de l’analyse ciblée traditionnelle.

International audience; The increasing need on energy and fluvial transport due to industrialization during the last century resulted in the construction of dams along rivers. To improve the production of electricity and avoid problems due to sedimentation, dam flushing operations are periodically organized. The Rhône Sediment Observatory investigated the impacts of such operations on suspended particulate matter (SPM) dynamics (resuspension and fluxes) and quality (physicochemical characteristics and contamination) during a flushing operation performed in June 2012 on 3 major dams on the Rhône River downstream the Lake Geneva (Verbois, Chancy-Pougny and Génissiat dams). The concentrations of several contaminants (hydrophobic organic contaminants, trace metal elements, radionuclides) as well as particulate organic carbon (POC) and particle size distribution were measured on SPM samples collected during this event at a permanent monitoring station (150 km downstream of the last flushed dam). Data obtained at this station from 2011 to 2016 were also used to compare the SPM and contaminant concentrations and fluxes during the 2012 dam flushing operations with those during flood events and baseflow regime. For the same water discharge, mean SPM concentrations during flushing were on average 6–8 times higher than during flood events recorded from 2011 to 2016. While of short duration (19 days), the flushing operations led to the massive resuspension of SPM and contributed to a third of the mean annual SPM flux. The SPM contamination was generally lower during flushing than during baseflow or flood, probably due to the fact that SPM during flushing operations is only issued from resuspended sediment, with no autochtonous particles nor recently eroded soil. The only exception are polycyclic aromatic hydrocarbons and bis(2-ethylhexyl)phthalate with higher concentrations during flushing, which must be issued from the resuspension of legacy-contaminated sediments stored behind the dams before the implementation of emission regulations. The observed increase of organically bound tritium is related to the resuspension of fine tritiated particles from watchmaking activities of the past century. During flushing, the variations of POC and contaminant concentrations are mostly driven by particle size.

International audience; Land-based micropollutants are the largest pollution source of the marine environment acting as the major large-scale chemical sink. Despite this, there are few comprehensive datasets for estimating micropollutant fluxes released to the sea from river mouths. Hence, their dynamics and drivers remain poorly understood. Here, we address this issue by continuous measurements throughout the Rhône River basin (∼100,000 km2) of 1) particulate micropollutant concentrations (persistant organic micropollutants: polychlorobiphenyls [PCBi] and polycyclic aromatic hydrocarbons [PAHs]; emerging compounds: glyphosate and aminomethylphosphonic acid [AMPA]; and trace metal elements [TME]), 2) suspended particulate matter [SPM], and 3) water discharge. From these data, we computed daily fluxes for a wide range of micropollutants (n = 29) over a long-term period (2008-2018). We argue that almost two-thirds of annual micropollutant fluxes are released to the Mediterranean Sea during three short-term periods over the year. The watershed hydro-climatic heterogeneity determines this dynamic by triggering seasonal floods. Unexpectedly, the large deficit of the inter-annual monthly micropollutant fluxes inputs (tributaries and the Upper Rhône River) compared to the output (Beaucaire station) claims for the presence of highly contaminated missing sources of micropollutants in the Rhône River watershed. Based on a SPM-flux-averaged micropollutant concentrations mass balance of the system and the estimates of the relative uncertainty of the missing sources concentration, we assessed their location within the Rhône River catchment. We assume that the potential missing sources of PAHs, PCBi and TME would be, respectively, the metropolitan areas, the alluvial margins of the Rhône River valley, and the unmonitored Cevenol tributaries.

The implementation of the Water Framework Directive (WFD) requires the monitoring of micropollutants in the aquatic environments to prevent any damages to both human health and ecosystems. In this context, a number of organic and inorganic compounds have been selected as priority pollutants and their measurement is necessary to ensure that water-quality standards are maintained (EC, 2008; EC, 2013; EC, 2015). For two decades, passive sampling methods have been developed for the monitoring of organic and inorganic compounds. Passive samplers allow measuring these compounds at trace levels by accumulation and concentration over long-term exposure. Moreover, the use of integrative passive samplers (IPS) allows a better representativeness of measurements because it takes into account the episodic pollution (Miège et al., 2015). Such passive sampling techniques have been recommended in the European Commission Guidance Document on surface water chemical monitoring, as complementary methods to improve the level of confidence in water monitoring data in comparison with conventional spot sampling (EC, 2009). In the context of the WFD monitoring programs, French government has mobilized the expertise of AQUAREF for two main objectives: - Demonstrate, in-situ, the interest to deploy IPS for monitoring of priority substances in aquatic environments; - Diffuse protocols and guidelines on the use and analyses of IPS, and initiate the formation of future stakeholders. We present the in situ campaigns to demonstrate theIPS relevance. Conventional IPS will be deployed: POCIS (Polar Organic Chemical Sampler), silicone membrane and DGT (Diffusive Gradients in Thin films). In this poster, we will focus on micropollutants needed to be controlled in the water matrix (the more hydrophilic ones), not in biota1. In a first round, the contamination by around 50 micropollutants on 3 sites will be measured every 2weeks, continuously during one year. In a second round, the contamination by around 100 micropollutants on 20 sites spread throughout France will be studied during a 2 weeks long campaign. Results obtained with IPS will be compared with spot water samples. This large scale study should highlight the interest to use IPS for WFD monitoring program by increasing the frequency of quantification and obtaining a better temporal representativeness of the contamination. In this context, the formation of stakeholders is also a strategic challenge; we present the current gaps and our actions on this aspect at a national scale.

The objective of this poster is to present specific results obtained during this interlaboratory exercise for pesticide sampling. Two trials were conducted during 14 days in the Charente River (at Beillant, France) and in a marine site, the Thau Lagoon (Hérault, France) with the participation of 11 laboratories at Beillant and 8 laboratories at Thau. Nine pesticides (acetochlor, S-metolachlor, 5 WFD priority substances, i.e., alachlor, atrazine, diuron, isoproturon, simazine and two atrazine metabolites, i.e. desethylatrazine, deisopropylatrazine) were considered. Among the tested PSs, we found Polar Organic Chemical Integrative Sampler (POCIS, pharmaceutical & pesticide configuration), polar Chemcatcher (with SDB-XC, SDB-RPS or C18 phases), silicon rod and silicon sheet, speedisk hydrophylic DVB06, ... We will present here and discuss results on i) Comparison of pesticide quantities accumulated per surface of passive sampler exposed ii) Mean and variability of Time Weighted Average (TWA) water concentrations of pesticides obtained with passive samplers and comparison with water concentration from grab samples iii) Pesticide fingerprints in the various passive samplers compared with those in grab samples

communication par affiche; International audience

Various integrative samplers, at different stages of development, are now available to estimate time-weighted average (TWA) concentrations of freely dissolved hydrophobic contaminants in aquatic environments. The ECLIPSE project (2009-2011, coord. C Miège, Irstea) involves 5 laboratories and aims to calibrate and compare 5 different integrative samplers for the monitoring of indicator and dioxin-like PCB in water: semi-permeable membrane device (SPMD, studied by Irstea), low-density polyethylene strip (LDPE, Ifremer), silicone rubber (SRPS, Deltares), Chemcatcher® (BRGM) and ontinuous-flow integrative sampler (CFIS, LABAQUA). The first step of the project was the simultaneous calibration of these samplers under laboratory conditions. During summer 2009, samplers were exposed under constant agitation and temperature in a stainless steel tank filled with 200 L of PCB contaminated water. A constant PCB concentration of about 1 ng/L was achieved by immersing a large amount of silicone rubber sheets [Rusina 2010] spiked with indicator and dioxin-like PCB. Prior to deployment, samplers were spiked with Performance Reference Compounds (PRC) and exposure durations ranged from 1 day to 3 months. After exposure, processing and analysis, samplers were compared in terms of sampling rate, linear uptake phase duration, repeatability and accuracy of calculated TWA concentration. The second step of the ECLIPSE project was the deployment and comparison of these samplers in real aquatic environments. In summer and autumn 2010, and in spring 2011, three field campaigns were carried out along the Rhône River and near the lake Le Bourget (France), known to be contaminated by PCB. Using appropriate cages, canisters and holders, studied samplers were exposed simultaneously at an approximate depth of 1 m during 1, 2, 3 and 4 weeks with duplicates and/or triplicates. These field campaigns allowed to compare in-situ ease of use and robustness of the five integrative samplers in addition to repeatability of sampling and TWA concentration estimate.

communication par affiche; International audience; POCIS (Polar Organic Chemical Integrative Sampler) is a new emerging tool for sampling polar organic micropollutants in water. There is still a need of research concerning its domain of validity (e.g., molecules sampled, type of water studied, optimal exposure duration) and its performances (molecules sampling rates, repeatability, accuracy of the evaluation of time-weighted average or TWA concentrations). In order to obtain in situ TWA concentrations, POCIS needs to be calibrated in laboratory for each molecule of interest. We performed 3 different laboratory calibration experiments using a flow-through calibration system in order to evaluate the optimal exposure duration and to calculate the sampling rates for 56 polar organic contaminants, and to identify and validate performance reference compounds (PRCs). The calibration system was composed of 2 aquaria (50 L) containing i) spiked tap water (circa 3 μg/L) and non-spiked POCIS for the determination of accumulation kinetics, ii) non-spiked tap water and spiked POCIS for the determination of desorption kinetics and iii) spiked tap water and spiked POCIS for the validation of the potential PRCs. In all experiments, water temperature, pH, conductivity and dissolved organic carbon were controlled. Agitation in the aquaria was ensured via a submerged pump. The resulting flow velocity was 10±5 cm/s and was directed perpendicularly towards POCIS surface. Triplicate POCIS were analyzed at 0, 1, 3, 6 and 12 hours and at 1, 3, 7, 11, 14, 21 and 28 days. We present results for the 56 studied molecules, including the duration of linear accumulation phase, sampling rates and possible performance reference compounds (PRCs). The accumulation phase was curvilinear for almost all molecules; so, we obtained the duration of optimal linear accumulation phase using t1/2 criteria. Concerning the desorption experiments, 5 compounds showed a good potential as PRC; they enable to decrease the effect of variable environmental conditions, so they can be used to obtain more reliable in situ TWA concentrations. We also checked isotropic exchange for these molecules comparing the exchange constant ke in accumulation and desorption experiments. At last, we validated the PRC strategy by calculating TWA concentrations from the third laboratory experiment mentioned above.

International audience

International audience; In recent years, metal oxides such as titania have been commercially available as chromatographic beds that can potentially be used to achieve novel separations of polar compounds. For example beta blockers, which are more often encountered in environmental sciences, have a wide range of polarity, and their basic character leads to difficult sample treatment and separation on conventional silica-based sorbents. The contribution of titania to the selective analysis of nine beta blockers was evaluated in terms of retention mechanisms observed in hydrophilic interaction LC using acetonitrile/water mobile phases with various additives. The mobile phase additives enabled to control the beta blocker charge as well as the titania surface charge. Depending on their respective ionic state, various retention mechanisms were identified at low water contents (

International audience; organic micropollutants in water (Alvarez et al., 2004). It is still in development concerning its domain of validity (e.g. molecules sampled, type of water studied, optimal exposure duration) and its performances, including the definition of molecules sampling rates, repeatability, accuracy of the evaluation of time-weighted average (TWA) concentrations. Performance reference compounds (PRCs) enable to decrease the effect of variable environmental conditions, so they can be used to obtain more reliable TWA concentrations in situ. To our knowledge, for POCIS, only one PRC has been identified to date: deisopropylatrazine-d5, that has been used and proved efficient for polar herbicides monitoring in freshwater (Mazzella et al., 2007, 2010). In order to obtain in situ TWA contaminant concentrations, POCIS needs to be calibrated in laboratory for each molecule of interest. Our work was based on laboratory calibration of POCIS for 5 alkylphenols, 9 hormones, 12 pesticides, 29 pharmaceuticals, 3 phenols and 1 UV filter in a flow-through system. This system was composed of 2 aquaria (50 L) containing i) spiked tap water (circa 3 μg/L) and non-spiked POCIS for the determination of accumulation kinetics or ii) non-spiked tap water and spiked POCIS for the determination of desorption kinetics. In both experiments, water temperature, pH, conductivity and dissolved organic carbon were controlled. Agitation in the aquaria was ensured via a submerged pump. The resulting flow velocity was 10±5 cm/s and was directed perpendicularly towards POCIS surface. Triplicate POCIS were analyzed at 0, 1, 3, 6 and 12 hours and at 1, 3, 7, 11, 14, 21 and 28 days. We will present various results depending of molecules such as the duration of linear accumulation phase and sampling rates, sampling repeatability and possible PRCs. As for example, linear accumulation phase for estrogenic hormones seems to be 21 days, since at 28 days the equilibrium regime begins to be reached. For betablockers, linear accumulation phase is shorter: less than 14 days. Sampling rates for this 2 families of compound vary from 0.058 L/d for sotalol to 0.537 L/d for 17-a-estradiol. Concerning the desorption experiments, atenolol-d7 and sotalol-d6 for example show a good potential as PRCs. Indeed, a desorption of 70% for atenolol-d7 and 50% for sotalol-d6 was observed in 24 days.

Regarding to the growth of interest concerning the presence and the identification of human made pollutants and xenobiotics in the environment, multi-residue analysis techniques and representative water sampling methods need to be developed. Some of these micro pollutants of interest are known as Endocrine Disruptor Compounds (EDCs) due to their effects on the endocrine causing behavior disorders, decreased reproduction or birth malformations for example. As those molecules display large range of physico-chemical properties, most of the current studies are focused on one or two EDCs families. This project is focused on the evaluation of the current pollution of surface water media threw the development of a multi-residue and multi-family of EDCs analysis method coupled to passive sampling devices. 28 EDCs of interest have been selected for this project among emerging contaminants listed by the European Union. This combination of different EDCs families representating agricultural, pharmaceutical and human pollution, has been chosen to be found in surface water and analysed by liquid chromatography tandem mass spectrometry (LC-MS/MS). Two passive sampling devices have been selected to monitor those compounds: the Polar Organic Chemical Integrative Samplers (POCIS in Pharm configuration) and the Chemcatcher (in polar configuration). Both are still in development but already showed good efficiency to monitor some of the compounds of interest. Calibration of those systems have been realised with the Cemagref Lyon in 50L aquariums with continuous renewal of doped solutions for 28 days. Both showed good linearity for most of the compounds of interest up to 21 or 28 days but seem to have reached equilibrium after 14 days for some of the molecules. Calibration experiment has been validated by Cemagref Lyon with controled and stable temperature, conductivity and flow. Both systems showed same behaviour but calculated sampling rates were higher for the POCIS than for the Chemcatcher. The two systems have been exposed in the field for 4 weeks between june and july 2010. Experiments confirmed linearity for somme of the molecules and enable us to calculate sampling rates for some that hasn't been tested yet. Those two passive samplers combined to an optimised multi-residue analytical method allow us to monitor a representative mixture of endocrine disruptors in surface waters.

Les matières organiques naturelles (MO) issues des sols interviennent dans les interactions entre les écosystèmes terrestres et aquatiques. Facteurs actifs du transfert / stockage des polluants et de nutriments, les MO extractibles constituent l'un des traits caractéristiques des sols. Le rôle de ces MO est fort variable selon leur structure moléculaire (caractère aromatique, aliphatique ...) or cette structure évolue dans le temps sous le contrôle de paramètres biotiques mais également abiotiques. Afin de décrire l'évolution conformationnelle des matières organiques naturelles après une extraction à l'eau, nous avons analyses ces MO par chromatographie d'exclusion stérique couplant en série un détecteur de fluorescence, d'absorption UV-vis et de COD. L'échantillon de sol provient de l'horizon organique d'un cambisol situé dans une clairière en moyenne montagne calcaire. Le sol a été sèché puis broyé légèrement avant d'être tamisé à 2 mm. La solution obtenue après agitation orbitale (4h) d'un mélange sol:solution de 1:10 dans de l'eau osmosée a été filtrée à 0,2 μm puis conservée à 6°C à l'abri de la lumière. Les analyses HPSEC de cet échantillon ont été réalisées de 1 à 10 jours après l'extraction. La première analyse réalisée (premier jour) sur l'extrait révèle principalement que cet extrait est composé de molécules de petite masse moléculaire (mode aux alentours de 500 Da). Très peu de molécules présentent des masses supérieures à 100 000 Da. Progressivement (au cours des 10 jours de mesures), la quantité de ces petites molécules diminue au profit de composés de plus grosse taille moléculaire. On voit alors augmenter la proportion relative de modes qui étaient présents dès la première analyse. De même, la proportion d'agrégats de masse moléculaires supérieurs à 100 000 Da augmente au cours du temps. À 10 jours, un précipité organique apparaît dans la solution ce qui conduit à la diminution relative de la part de ces agrégats. Ces premiers résultats montrent que la conformation des matières organiques extraites à l'eau est très variable dans le temps. Cette dynamique induit une grande variabilité temporelle dans le rôle que peuvent jouer les MO depuis leur mobilisation jusqu'à leur stockage au sein des sols ou dans d'autres compartiments des écosystèmes. Enfin, cette étude implique des questionnements sur le devenir des MO en milieu naturel ou de tels phénomènes, reversibles ou non, sont envisageables après des modifications des équilibres de la solution (agitation, dilution, concentration).

National audience

Edytem equipe 1 2003-2010; National audience

Edytem equipe 1 2003-2010; National audience