Use of experimental design in the optimisation of stir bar sorptive extraction followed by thermal desorption for the determination of brominated flame retardants in water samples

A method for the determination of polybrominated diphenyl ethers (PBDEs) and polybrominated biphenyls (PBBs) in water samples is proposed. The method involving stir bar sorptive extraction (SBSE) and thermal desorption followed by gas chromatography coupled with mass spectrometry was optimised using statistical design of experiments. In the first place, the influence of different polydimethylsiloxane stir bars was studied. A Plackett-Burman design was chosen to estimate the influence of five factors on the efficiency of the SBSE process: desorption time (5-10 min), desorption temperature (250-300 degrees C), desorption flow (50-100 mL min(-1)), cryofocusing temperature (-130 to 40 degrees C) and vent pressure (0-12.8 psi). Afterwards, two central composite designs were used to find the optimal process settings that were applied to the optimisation of both desorption and extraction efficiency. In the case of the desorption parameters, long desorption times (10 min) and desorption flows lower than 70 mL min(-1) yielded the best signals for the majority of compounds. However, different behaviour among the analytes was observed for the vent pressure and we decided to fix it at an intermediate value (7 psi). In the case of extraction parameters, the sample volume and the addition of NaCl did not have a significant effect, while the addition of methanol yielded better extraction responses. Remarkable recovery (82-106%) and repeatability (less than 18%) were attained. Furthermore, excellent regression coefficients (r(2)=0.991-0.999) and low detection limits (1.1-6.0 ng L-1) were also achieved for the congeners studied. The proposed method was applied to the analyses of PBDEs and PBBs in waters from the Basque Country, Spain.