| dc.description.abstract | The release of oil and petroleum products into the environment has long been a cause for concern due to its toxic effects on marine species. In particular, the Atlantic haddock has shown enhanced sensitivity towards the effects of crude oil. Much of the crude oil toxicity has been attributed to metabolites of polycyclic aromatic hydrocarbons (PAHs). While many PAHs undergo bioactivation, including phenanthrene, recent research has revealed that PAHs in isolation are unable to act with the same potency as PAHs in crude oil mixtures. This suggests that the contribution of PAHs in crude oil toxicity is more complex than what was initially believed. This thesis aims to validate and apply a quantitative liquid chromatography tandem mass spectrometry method developed for the analysis of phenanthrene metabolites in Atlantic haddock embryos. Attention is also given towards measures for method optimization. The work performed throughout this thesis emphasises the use of rigorous statistical and chemometric approaches to obtain reliable and justifiable results. Method validation was conducted by spiking blank haddock eggs with standard solutions. The validation parameters included a response function, linearity, trueness, precision, accuracy, limit of detection, limit of quantitation and selectivity. The results demonstrated that the method had good quantification abilities for most of its analytes within the validated range. The validated method was used to analyse samples from two exposure studies. In the first study, results showed that haddock embryos exposed to crude oil mixtures possess an enhanced rate of metabolite formation compared to PAH-only containing fractions. The second study revealed relative abundances of metabolites formed in haddock embryos after phenanthrene exposure. Quality control samples showed that the method provides accurate results within its validated range. However, the sample application study also revealed that the validated method failed to capture the relevant concentration range. Thus, while trends were visible, no quantification could be performed. To optimize the method, experimental designs were employed on the electrospray ionization source. Fractional factorial designs were used to screen for relevant parameters. Screening revealed that ion source voltages could be improved. After voltage adjustments, a response surface was generated. The surface revealed that the method was seemingly still unoptimized, and that method response could be further enhanced by changing the gas temperature and flow in tandem. The optimized method was found to provide a considerably improved method response when compared to the initial results. Both the method validation and sample application revealed that the method works well within its validated range, suggesting that the statistical approaches were fit for their purpose. Additionally, after method optimization, instrumental response was significantly improved. Therefore, although some further optimization may be necessary and a lower concentration level must be validated, the work performed in this thesis demonstrated that the proposed method has good potential for the analysis of phenanthrene metabolites in haddock embryos. | |
