Determination of organochlorine priority substances in fish tissue: Optimisation of the clean-up step balancing removal of lipids with analytes' recovery.

Quality-assessed analytical methods are required to determine organic priority substances (PS) in biota for the monitoring of the water status according to the EU Water Framework Directive. Although the literature describes several analytical methods to determine these substances in fish, discussion about the efficiency of the clean-up procedures to remove the lipids in the final organic extract (and decrease the disturbance of co-extractives at the detection step) is scarce. This work highlights the results of the development of an analytical method for organochlorines in fish tissue focused on the optimisation of the clean-up step in order to obtain a final extract with the lowest amount of lipids. The efficiency of the purification of the final extract was assessed by quantifying the removal of co-extractives gravimetrically, by considering the sensitivity of the gas chromatography-mass spectrometry (GC-MS) method for the analytes and by assessing the analytes' recovery. An analytical method based on accelerated solvent extraction followed by gel permeation chromatography and dual solid phase extraction (as clean-up steps) and quantification via isotope dilution GC-MS was applied to the analysis of seven PS [α-, β-, γ- and δ-hexachlorocyclohexane (HCH) isomers, pentachlorobenzene (PeCB), hexachlorobenzene (HCB) and hexachlorobutadiene (HCBD)]. A preliminary validation of the method was carried out with satisfactory results for all analytes in terms of intermediate precision (2.9-9.4%, except 11.8% for β-HCH). Repeatability values were satisfactory for α-, γ-HCH, PeCB and HCB (4.3-6.9%), while the obtained results for β-, δ-HCH and HBCD showed the need for further optimisation. Trueness was within the target performance (recovery range: 96.0-107.5%) for all analytes except β- and δ-HCH. Limits of quantification between 0.5 and 3.3 ng/g wet weight were achieved, depending on the analytes. The proposed method can be employed to determine the mentioned PS in fish tissue with up to approximately 6% lipid content. The presented results show the challenges in establishing an analytical method which aims at balancing the required accuracy with the routine applicability (and a minimised impact on the detection system) as needed in the context of environmental monitoring.