Modern analytical methods for the determination of pesticide and pharmaceutical residues in natural waters and sediments

Among the vast array of contaminants of anthropogenic origin reaching the environment, pesticides and pharmaceuticals constitute two emerging classes raising concerns about long-term consequences on human health. Widespread contamination has been observed in the different environmental compartments such as surface, marine and ground waters, soils and sediments as well as in invertebrates and vertebrates, at concentrations from few ng L-1 to several mg L-1. Continuous discharge into aquatic systems make the residues of those compounds ubiquitous in the environment. Several of them have been recently added to the ‘priority’ and ‘watch list’ of substances, underlining the significance to study their environmental occurrence. Pesticides, which reach the environment through agricultural run-off, improper use, spray drift etc, are extensively studied and well-documented. Being considered as a vital component of modern farming, they play a major role in maintaining high agricultural productivity. Hence, in high-input intensive agricultural production systems, the widespread use of pesticides to manage pests has emerged as a dominant feature. However, reliance on pesticides is difficult to sustain because of unintended long-term adverse effects on the environment and human health. On the other hand, pharmaceuticals constitute a more ‘recent’ class of contaminants. They enter the environment through defective septic systems, leaking underground sewage, surface run-off, but mainly inadequate degradation or removal through wastewater treatment plants (WWTPs). The rate of their usage and consumption has been increasing rapidly and so will be, due to the expanding population, inverting age structure, new target age groups, and expansion of potential markets aggravates the problem. Over the last 20 years increasing attention has been paid to understanding the presence and impacts of pesticides and pharmaceuticals in freshwater ecosystems. However, a lack of a systematic approach to the detection of such compounds into coastal environments has led to fragmented literature of serendipitous approaches. This is a critical knowledge gap, given the significant increase in coastal human populations around the globe and the growth of coastal megacities, together with the importance of coastal aquaculture around the world. There is now widespread recognition of the need for a cradle-to-grave stewardship of pesticides and medicines for minimizing environmental exposure while improving crops or promoting human and animal health respectively. What is more, the majority of monitoring programs focus on the aqueous compartment. Studies about the presence of these compounds in sediments -let alone marine ones- are scarce. However, since sediments are the result of the integration of various processes, they tend to be a sink for many contaminants, providing not only a global estimation about the quality status of water bodies but also reliable historical contamination information. Sensitive analytical methods to evaluate the extent of the risk for the plethora of pesticides and pharmaceuticals are required. Multi-residue methods are powerful tools, as they may provide greater knowledge about the overall contamination. Recently, the evolution of high-resolution mass spectrometry has initiated a new trend in analytical data processing. High resolving power and high mass accuracy are the ‘gold’ parameters, enabling the distinction between any possible interfering matrix components and target analytes, within a decimal difference, avoiding false positives. Moreover, the identification and confirmation are more selective and reliable thanks to the accurate mass measurements for each molecular ion and subsequent fragments.The instrumentation of ion trap/Orbitrap (LTQ–Orbitrap) is a combination of two different mass spectrometer types, with excellent detection and identification ability for compounds of a range of molecular weights, even in the most complex matrices. It outweighs against the traditionally-used low-resolution methods, enabling the quantification and confirmation of a theoretically unlimited number of analytes within a single analytical run, as well as isobaric compounds that differ in the elemental composition, can be distinguished.To keep up with this evolution in analytical strategies, generic sample preparation methodologies employing different clean-up techniques, offering reliable analysis of a large number of compounds, are needed. However, given the different classes and physicochemical properties of the analytes and the complexity of the matrices, multi-residue analysis in such matrices is a challenging issue. Stepping towards a more comprehensive assessment of environmental exposure to contaminants, the present dissertation looks at the development and validation of analytical methods that combine efficient extraction combined with the cutting of-the-edge LC–Orbitrap/MS mass spectrometry. Following the current trend to apply cheaper, miniaturized and more ‘green’ extraction techniques with sufficiently cleaned-up extracts amendable to LC-MS analysis, the so-called QuEChERS approach followed by dispersive solid phase extraction (dSPE) clean-up was selected and modified accordingly for sediments analysis. For water analysis, the common, extensively studied and effective SPE was the method of choice. The LC-MS crucial parameters were optimized and finally, the analysis, operated in positive ionization mode, was carried out in a short overall analysis time. Identification was based both on accurate mass and retention time. Matrix-matched calibration approach along with isotopically-labeled internal standards was employed for quantification and correction of matrix effect. Apart from mass accuracy provided by Orbitrap FT-MS technology, the confirmation of positive findings was also based on MS/MS fragmentation and isotope abundance, employed to reduce the number of possible formulas for a given mass. Afterward, the extraction methods were also optimized and validated in accordance with the European Commission requirements, while further confirmation was achieved by MS fragmentation. Regarding QuEChERS, to obtain the optimum conditions, pH, extraction salts, sediment amount and clean-up sorbents were studied for pesticides and pharmaceuticals separately. In the optimized method, recoveries for pesticides were calculated between 70.8% and 106.2% and for pharmaceuticals between 64-101 % in the intermediate spiking level, with a relative standard deviation below 15% in almost all cases. Method quantification limits were below 10 ng g-1 for the vast majority of pesticides, while for pharmaceuticals ranged from 1.3 to 47 ng g−1. The linearity, expressed as a correlation coefficient, was over 0.992 in all cases, within a range from each compound’s method quantification limit to 500 ng g−1. Matrix effect study revealed slight signal suppression that was decreased largely the clean-up step for the majority of compounds. For the SPE optimization, different analytical protocols were examined for each mix of compounds, testing different cartridges, sample pH, washing and elution solvent. For pesticides, the recovery rates ranged from 61% -96% in the river water for medium spiking level, while they were slightly higher (60% -111%) for seawater due to increased ionic strength but also substrate effect. The recoveries for pharmaceutical compounds in river waters ranged from 61% to 105.3% and in seawater from 55.9% to 110.3%. In all cases, the repeatability and intra-laboratory reproducibility was less than 15%. Method quantification limits ranged from 1.7 ng L-1 to 90 ng L-1 for pesticides and from 0.8 ng L-1 to 25 ng L-1 for pharmaceuticals. Linearity was excellent in all cases, with the correlation coefficient being greater than 0.997. Matrix effect study in seawater revealed strong signal suppression only for dimethoate, fluoxetine, paroxetine, and sertraline. The developed analytical methodologies were implemented in an EU-wide case study, including samples from coastal areas where aquaculture is active in Greece, Albania, Italy, Spain, Portugal, the United Kingdom, and Norway. Concerning the pesticides, the study revealed the presence of 67% of the selected pesticides in 75% of the water samples, while only 50% of these compounds were detected at a concentration above the quantification limit. Metalaxyl, diuron, S-metolachlor, and fenpyroximate were the most ubiquitous pesticides. Italy was the country with the most positive findings while Norway with the fewest. In sediments, the frequency of detection was quite low and at concentrations below the quantification limit. With regard to the detection of the pharmaceutical compounds in waters, 36% of them were detected at least once. The most frequently detected compounds both in water and sediment were carbamazepine and venlafaxine. Moreover, those exhibited the highest concentration in water, while in sediments the highest concentration was observed for risperidone. Equally to pesticides, Italy was in the first place of positive findings. Environmental risk assessment was performed implementing the risk quotient method for three trophic levels (algae, daphnids, and fish). In almost all cases for pesticides, RQ values were much 0,997). Η επικύρωση των πολύ-υπολειμματικών μεθόδων που αναπτύχθηκαν πραγματοποιήθηκε με βάση την ισχύουσα νομοθεσία της ΕΕ, όπως αυτή ορίζεται στις οδηγίες 2002/657/ΕΚ και SANTE 11813/2017. Η εφαρμογή των μεθόδων σε δείγματα νερών και ιζημάτων από υδάτινα σώματα πλησίον υδατοκαλλιεργειών στην Ελλάδα, την Αλβανία, την Ιταλία, την Ισπανία, την Πορτογαλία, την τη Νορβηγία και το Ηνωμένο Βασίλειο επιβεβαίωσε την παρουσία πολλών φυτοπροστατευτικών και φαρμακευτικών ενώσεων κυρίως στα νερά, και λιγότερο στα ιζήματα. Συγκεκριμένα, το 67% των επιλεγμένων φυτοπροστατευτικών ενώσεων ανιχνεύθηκε στο 75% των δειγμάτων νερού, ενώ μόνο το 50% αυτών των ενώσεων ανιχνεύθηκαν σε συγκέντρωση πάνω από το όριο ποσοτικοποίησης. Στα ιζήματα, η συχνότητα ανίχνευσης ήταν αρκετά χαμηλή και σε συγκεντρώσεις κάτω από το όριο ποσοτικοποίησης. Όσον αφορά την ανίχνευση των φαρμακευτικών ενώσεων στα ύδατα, το 36% αυτών ανιχνεύθηκε τουλάχιστον μία φορά. Οι ενώσεις που ανιχνεύθηκαν συχνότερα τόσο στο νερό όσο και στο ίζημα ήταν το carbamazepine και το venlafaxine. Επιπλέον, εμφάνισαν την υψηλότερη συγκέντρωση σε νερό, ενώ στα ιζήματα την υψηλότερη συγκέντρωση εμφάνισε το risperidone. Η Ιταλία ήταν η χώρα με τις περισσότερες θετικές ανιχνεύσεις. Η εκτίμηση περιβαλλοντικού κινδύνου έγινε με την εφαρμογή της μεθόδου του πηλίκου (risk quotient) για τρία τροφικά επίπεδα (φύκη, ασπόνδυλα και ψάρια). Σχεδόν σε όλες τις περιπτώσεις οι τιμές RQ ήταν μικρότερες από 0,01, που αντιστοιχεί σε "χαμηλό κίνδυνο", με εξαίρεση μια φυτοπροστατευτική και μια φαρμακευτική ένωση. Η μέθοδος αυτή υποδεικνύει την ανάγκη για περαιτέρω διερεύνηση της υπολειμματικότητας τέτοιων ρύπων και της έκθεσης οργανισμών σε αυτούς, για την ολοκληρωμένη αποτίμηση της υποβάθμισης του υδάτινου περιβάλλοντος