Comparison of two experimental systems for the calibration of passive SBSE (Stir Bar Sorptive Extraction) for 16 pesticides in water | Comparaison de deux systèmes expérimentaux pour la calibration de passive-SBSE pour 16 pesticides dans l'eau

[Departement_IRSTEA]Eaux [TR1_IRSTEA]BELCA [TR2_IRSTEA]ARCEAU

International audience

英语. Several types of passive samplers have been developed for the time-integrative monitoring of various organic contaminants in the aquatic environment. The passive SBSE (p-SBSE) has been recently demonstrated to be a suitable passive sampler for pesticides monitoring in agricultural watersheds. Nevertheless, for the determination of time-weighted average (TWA) concentrations, a calibration of the passive samplers under controlled conditions has to be performed prior to field exposure in order to determine for each pesticides accumulation kinetics and sampling rates (Rs). Many calibration systems have been described in the literature, but we could not find any comparison of these experimental systems for a same passive sampler and tested compounds. The aim of this study was to test two different laboratory experimental systems (continuous flow and static mode) for the calibration of p-SBSE for 16 pesticides with a wide range of physico-chemical properties (log Kow 2.2 – 5.1). We evaluated the performances of both systems in terms of consumption of water and pesticides, and stability of flow rate, temperature and pesticides concentration during the calibration experiments. The continuous flow calibration was performed in 15 glass channels (3 L each and a triplicate of p-SBSE per channel) and was based on a constant supply of water contaminated with the studied pesticides. The nominal concentrations ranged from 0.2 to 40 µg/L depending on the pesticides. The static mode calibration consisted in placing the passive samplers (up to 24 p-SBSE) in a 30 L-stainless steel tank containing spiked silicone sheets (8 sheets used), as a diffuse source of pesticides in order to generate a constant water contamination during the calibration period. The nominal concentrations of pesticides ranged from 0.1 to 150 µg/L. We performed the calibration at 20°C during 7 days for both experimental systems. The flow velocity was 2.5 cm.s-1 in the continuous flow calibration, and 10 cm.s-1 in the static mode calibration. For both systems, the temperature and the water flow velocity were stable during the calibration (20°C ± 1 for temperature, and RSD lower than 14% (n=240) for flow velocity measurements). The main advantage of the continuous flow calibration system is that the hydrodynamic conditions are close to the in situ conditions. However, it requires huge volumes of water (2000 L) and a large amount of pesticides (3.5 g) for spiked water renewal in order to maintain hydrophobic pesticide concentrations stable. Moreover, the system is relatively complex to operate due to the large number of pumps that need to be accurately calibrated. On the contrary, the static mode calibration system allows a limited consumption of water (30 L) and pesticides (0.1 g) as water renewal was not needed. This system is easy to use and stable contaminant concentrations in water can simply be maintained. However, it requires a previous step for spiking the silicone sheets which implies to know the exchange kinetics between silicone sheet, water and passive samplers before calibration. Both systems enabled to obtain curvilinear sorption profiles for 16 pesticides that were used for the determination of Rs. For hydrophilic compounds, the calculated Rs are similar with both systems, whereas for hydrophobic pesticides, the Rs obtained with the batch mode calibration system are generally greater, which can be relied to the difference of flow velocity between the two systems.