Fig. 1. Scheme of the analytical procedure.

Fig. 2. GC-ECD chromatograms for the BP–MS mixture: (A) split injection (split ratio 300:1) in the narrow-bore column; (B) thermal desorption (splitless mode for 2 min) of a PDMS fibre in the narrow-bore column; (C) splitless injection, 0.32 mm I.D. column; and (*) groups of peaks which could be partially separated only with the 30 m column.

Fig. 3. GC-ECD chromatograms for a spiked sediment sample (40 ng/g). HS SPME at 100 °C for 10 min: (A) 0.5 g of sediment plus 5 ml of water (B) 0.5 g of sediment plus 5 ml of KMnO₄ (0.2 M) and 0.5 ml of sulphuric acid 6 M; and (*) sulphur interferences.

Fig. 4. Pareto charts showing the influence of main effects on the yield of the SPME for CBs 28, 52, 118 and 180. The slope of each line is proportional to the effect of each factor. The curvature observed in the plot of KMnO₄ for CBs 52 and 118 indicates that the best yield corresponded to intermediate concentrations.

Fig. 5. Time course of the HS SPME for CBs 28, 52 and 180. Extractions at 100 °C using 0.5 g sediment samples (spiked level 40 ng/g) with magnetic stirring, n = 2 replicates.

Fig. 6. Overlay of chromatograms for the reference sediment CRM 536. Solid line, non-spiked sample. Dotted line, sample spiked with CBs 28, 52, 101, 118, 153, 138 and 180 at the 60 ng/g level. A and B chromatograms obtained using the narrow-bore column, C. Detail of chromatograms obtained using the 30 m × 0.32 mm I.D. column.

Effect of the type of sample treatment on the efficiency of the extraction

HS SPME at 100 °C for 10 min over 0.5 g of spiked sediment. Normalised responses for each congener. Average responses for two replicates.

Experimental domain and standardized values of main effects associated to factors considered in the optimisation of the HS SPME process

Linearity, repeatability (n = 4 replicates) and quantification limits (S/N = 10) of the method. Data obtained for the spiked river sediment

Concentrations of PCBs (ng/g) in reference sediment CRM 536, n = 4 replicates