Dissolved polycyclic aromatic hydrocarbons (PAHs) and dissolved organic carbon (DOC) in the River Wyre (northwest England)

Water samples were taken at five sites along the River Wyre, a typical river system in the northwest of England, draining upland and agricultural areas with soils rich in organic matter. Clean amber glass bottles with Teflon-line lids were used to collect 5 L water samples for the PAH analysis from each site during each of the four sampling events (19th August 2010, 6th December 2010, 6th March 2011 and 6th June 2011). Additionally, 100 mL samples were collected for DOC analysis. Samples were kept at 5°C and processed within 48 hours of collection. Samples were filtered through a 0.45µm filter and DOC concentrations were determined on the basis of UV absorbance at 270 and 350nm using an algorithm developed by Tipping et al. 2009. Prior to processing, samples were divided into two equal sub-samples, for the analysis of total dissolved and freely dissolved PAHs respectively. To isolate freely dissolved PAHs in this second sub-sample, the DOC and DOC-associated PAHs were precipitated by adding 0.4g of Al2(SO4)3 (dissolved in 5mL of Milli-Q water) and adjusting the pH to 6, the optimal flocculation pH for Al2(SO4)317, using NaOH or HCl. The flocculated DOC was then removed by passing the sample through a GF-F using a Millipore vacuum filtration unit. Al2(SO4)3 was found to remove DOC efficiently, particularly those substances PAHs tend to partition to strongest, and does not precipitate PAHs (Laor and Rebhun, 1997). Total dissolved (the sum of freely dissolved and DOC-associated) PAHs were measured in a raw sample filtered through a GFF filter. The concentrations of PAHs associated with DOC were determined indirectly by subtracting the concentration of freely dissolved PAHs from the sum of freely dissolved and DOC-associated PAHs. Laboratory blanks were generated by treating Milli-Q water in exactly the same way as the samples. After filtration, half of each sample, was transferred to a 1.5 L separating funnel, spiked with a mixture of deuterated PAHs to monitor recovery of the extraction and cleanup method, and liquid-liquid extracted with 80 mL of dichloromethane (DCM) three times . This procedure was repeated with the remainder of each sample. The extracts of both portions were pooled and anhydrous sodium sulphate (baked at 550°C) was added to remove any remaining water. These were then reduced to 1 mL on a Buchi Syncore evaporation system and cleaned on a column packed with 0.8 g of alumina (activated at 550 °C) and a small amount of anhydrous sodium sulphate. The target compounds were eluted with 10 mL of DCM. After the samples and blanks were blown down under a gentle stream of nitrogen they were transferred to small amber vials, further reduced to ca. 0.5 mL, spiked with a solution containing d10-acenaphthene and d12-benz(a)anthracene as internal standards, and analysed for all 28 target PAHs and the recovery compounds. The GC-MS analysis was carried out on an Agilent GC 6890N coupled to an Agilent MSD 5973N. 20µL of the extracts were injected in solvent vent mode and separated on a HT8 column (SEG, 50m, 0.22mm I.D., 0.25µm film thickness) with helium as the mobile phase at a constant flow of 2 mL min-1. The programmable temperature vaporization (PTV) inlet was kept at 20°C for 0.51 min, then heated to 350°C at a rate of 700°C min-1 and kept at 350°C for 5 min. Then the temperature was reduced to 300°C min-1 at a rate of 10°C min-1. The oven temperature programme was: isothermal at 50°C for 2.5 min, 15°C min-1 to 200°C, 5°C min-1 to 250°C, 8°C min-1 to 330°C and was held at 330°C for 25.5 min. The transfer line was heated to 350°C. The MS detector was operated in EI-mode, using selected ion monitoring, the quadrupole temperature was set to 150°C and the ion source temperature to 230°C.