Abstract:
Reports of the contamination of natural water bodies with micropollutants have increased in the last decades. Most importantly, persistent and polar micropollutants are of major concern as they may leach into groundwater, the main source of drinking water in many countries within the European Union. Consequently, for environmental authorities and researchers, it is important to investigate the environmental fate of such micropollutants. Conventional assessment approaches rely on changes in concentrations of the contaminant and its metabolite. This, however, is often inconclusive as the simultaneous formation and transformation of the metabolite, differences in the mobility between parent compound and metabolite, or repeated mobilization may lead to erroneous interpretations. Compound- specific stable isotope analysis (CSIA) is a complementary approach to identify transformation processes based on the analysis of natural isotope abundances of an element (e.g. carbon, nitrogen). As CSIA has so far been limited to the analysis of pollutants in the sub-μg/L range, this thesis aims to broaden the application of CSIA by developing analytical methods for polar micropollutants in environmental samples, applying these for systematic field studies and testing the limits of CSIA in concentrations in the low ng/L range. To this end, the frequently detected micropollutants desphenylchloridazon (DPC), 2,6- dichlorobenzamide (BAM), atrazine (ATZ) and desethylatrazine (DEA) were used as model compounds.
In the second chapter of this thesis, methods for carbon- and nitrogen-isotope analysis (δ13C and δ15N) of polar micropollutants were developed using liquid chromatography-isotope-ratio mass spectrometry (LC-IRMS) and derivatization gas chromatography-IRMS (GC-IRMS). DPC was used as a representative compound for polar contaminants during method development. Both methods resulted in reproducible and accurate δ13C and δ15N analysis of DPC with a limit of precise isotope analysis in the μg/L concentration range. For carbon isotope analysis 996 ng of DPC on-column were sufficient. Nitrogen isotope analysis was achieved by derivatization of DPC with a 160-fold excess of (trimethylsilyl)diazomethane. To enable the application of CSIA to environmental samples, where micropollutants are present in a concentration range of ng/L to sub-μg/L, more sensitive methods were required. Thus, the nitrogen isotope analysis was optimized using on-column injection, which resulted in accurate δ15N analysis for amounts greater than 100 ng DPC on-column. The feasibility of both methods was proven by measuring the isotopic composition of DPC in DPC-containing environmental-seepage water spiked with chloridazon (CLZ). The analysis indicated that it is possible to distinguish DPC containing different isotopic signatures.
After the feasibility of CSIA for polar micropollutants such as DPC was shown, a systematic field study of the DPC and its parent compound CLZ was carried out as detailed in Chapter 3. This study gave new insights into DPC degradation pinpointing the influence of simultaneous formation and transformation of the metabolite using two analytical approaches—the well- established metabolite-to-parent compound ratio and the recently available carbon and nitrogen CSIA. We found that (i) DPC was transformed in all lysimeters, showing a significant enrichment in 13C and 15N by approximately +4 ‰ and +3 ‰, respectively. (ii) Formed DPC, which had not been subject to further transformation yet, showed the same nitrogen isotope value as its precursor CLZ. As further transformation took place, significant carbon and nitrogen isotope fractionation was observed that was partially attenuated when mixing with freshly mobilized DPC from the vadose zone took place. The extent of isotope fractionation varied depending on the method of application of the parent herbicide and metabolite, implying the influence of plants, and the preferential flow on the formation and degradation of DPC. Additionally, we demonstrated that (iii) when DPC was further transformed, the isotopic signature, as an integrated signal of DPC degradation, was more reliable as an indicator of degradation than the metabolite-to-parent-compound ratio. Hence, this study enables the application of CSIA as an indication of DPC transformation, provided that there is no simultaneous formation and transformation of DPC. On the other hand, when DPC formation dominated and evidence from CSIA was not conclusive because changes in isotope values were reduced by the fresh input, metabolite-to-parent-ratios reached a maximum and could provide evidence of DPC formation. This leads to the conclusion that both methods are complementary, in particular when only partial degradation of the herbicide is occurring. While the metabolite-to-parent ratio provides information about the re-mobilization of a compound, CSIA shows the evolution of a compound’s degradation.
In Chapter 4 of this thesis, challenges in CSIA of polar and persistent micropollutants in groundwater were identified and critically discussed by evaluating a large volume extraction method of up to 100 L groundwater using ATZ, DEA and BAM as model compounds in low ng/L concentration ranges. It was found that, in contrast to previous laboratory experiments, where tap water was used, extracts from environmental groundwater resulted in low and non- reproducible recoveries. Since groundwater contains organic matter such as humic and fulvic acids, and as acidification was part of the extraction procedure, it is assumed that the change in pH prior to solid-phase extraction (SPE) may favor the formation of analyte-fulvic acid complexes leading to the low recoveries observed. In addition to unsatisfactory recoveries, the extensive sample enrichment also resulted in an extensive isotope fractionation as the isotopic signature of the organic matter interfered with the carbon isotope ratio of the target analytes. Such an interference would lead to an overestimation in the quantification of degradation, if unidentified. Thus, it is essential for future analytical method developments to critically evaluate each method and to include the investigation about a possible influence of sample matrix on the analysis already in the pre-tests. As this study has shown the limitations of CSIA of polar micropollutants in complex sample matrices, future studies may use this as a starting point towards more sensitive isotope analysis by methodological and instrumental advances.