A comprehensive evaluation of two sample treatment procedures for the determination of emerging and historical halogenated flame retardants in biota

Two different sample preparation protocols for the determination of 37 emerging and historical halogenated flame retardants (HFRs) in marine tissues were compared with regards to extraction recovery, lipid removal efficiency, repeatability, reproducibility, and ability to measure sub-ng g⁻¹ (dry weight) concentrations in marine biota. One method involved a purification step using gel permeation chromatography (GPC) followed by a HPLC fractionation step on a Partisil amino-cyano normal phase (GPC-Partisil procedure) and the other more traditional method was based on sulphuric acid treatment followed by silica column fractionation (H₂SO₄-silica procedure). The samples were analysed by gas chromatography (GC) and liquid chromatography (LC) tandem mass spectrometry (MS/MS). Sample fractionation in both methods enabled unique sample preparation procedures to isolate the GC from the LC amenable compounds. Both methods could remove > 99% of the lipids which was necessary prior to GC- and LC-MS/MS analyses. The majority of the target compounds (70%) had acceptable recoveries between 60–120% for both methods. However, the sulphuric acid treatment resulted in the degradation of the TBP-AE and the silica column fractionation resulted in the loss of BEH-TEBP and the elution of PBB-Acr and TBBPA-BME in the unsuitable fraction. High recoveries of DBE-DBCH (α+β), EHTBB, BTBPE, BEH-TEBP, and PBB-Acr were attributed to matrix effects, suggesting the need to use isotope-labelled surrogate standards of the target compounds. The optimisation of the silica column chromatography, GPC, and Partisil fractionation is described and discussed to afford easy implementation of the method. The method using GPC followed by Partisil fractionation is more efficient and more reproducible than the sulphuric acid-silica procedure. The application of this method to marine biota reference materials revealed the presence of relatively high concentrations of DBE-DBCH isomers and BDE-47 in fish samples. The method detection limits comply with the recommendations of the European Commission.