Prior to the extraction, samples were spiked with per-deuterated PAHs standards (phenanthrene-d10, crysene-d12 and perylene-d12), which were used as surrogate standards. The extracts were reduced in a rotary evaporator to 1 mL and then solvent-exchanged into isooctane. All samples were then fractionated on a 3% deactivated alumina column (3 g) with a top layer of anhydrous sodium sulphate. Each column was eluted with 12 mL of dichloromethane/hexane NVP-BEZ235 mw (2 : 1 v/v). The PAH fraction was concentrated in a rotary evaporator and solvent-exchanged to isooctane under a gentle stream

of nitrogen. After concentration, the samples were transferred to injection vials and 25 μL of anthracene-D10 and benzo(a)anthracene-D12 find more were added as injection standards. All the samples were analysed by GC-MS using a Thermo Electron (San Jose, CA; model Trace 2000 operating in selected ion monitoring

(SIM) mode. Details of temperature programs and monitored ions are given elsewhere (Cabrerizo et al., 2009, 2011). All analytical procedures were monitored using strict quality assurance and control measures. One field and laboratory blanks were introduced every three soil samples. Phenanthrene, fluoranthene and pyrene were detected in blanks, but they accounted for < 3% of the total sample concentrations. Samples, therefore, were not blank corrected. The surrogate per cent recoveries from the soil samples reported here were (mean ± SD) 70% ± 11 for phenanthrene-d10, 105% ± 17 for crysene-d12 and 90% ± 13 for perylene-d12. Catabolic degradation of 14C-phenanthrene was determined in 250-mL screwcap Erlenmeyer flasks (Reid et al., 2001). The respirometers contained 10 g of soil rehydrated to 40–60% water-holding capacity and spiked with unlabelled and 14C-phenanthrene (80 Bq 14C-phenanthrene g−1 soil) using toluene as a carrier solvent. A 7-mL scintillation vial containing 1 M NaOH was attached to the screwcap to serve Florfenicol as a CO2 trap. Respirometers were stored in the dark at 4, 12 and 22 °C. A slurry system was also set-up containing 30 mL minimal basal salts (MBS) medium and securely placed on a SANYO®

Gallenkamp orbital incubator set at 100 r.p.m. and 22 °C to agitate and ensure adequate mixing over the period of the incubation. NaOH traps were replaced every 24 h, after which 6 mL of Ultima Gold scintillation cocktail was added to each spent trap and the contents analysed on a Packard Canberra Tri-Carb 2250CA liquid scintillation counter. The incubation lasted for 35 days. Lag phases were measured as the time before 14C-phenanthrene mineralization reached 5%. Analytical blanks containing no 14C-phenanthrene was used for the determination of levels of background radioactivity. Colony-forming units (CFUs) of heterotrophic bacteria were enumerated on plate count agar (PCA) using a viable plate counts technique (Lorch et al., 1995).