4.1 Selection of Sampling Locations
The principal aim of the study was to determine levels of benzene in and around petrol service stations. As benzene levels are expected to be higher within urban areas as opposed to rural locations, monitoring was focused within the Greater London area, where the background concentration will be one of the highest in the UK.
A total of 12 service stations were selected for the study, with the assistance of two major superstore retailers and the UK Petroleum Industries Association (UK PIA) who initially contacted the individual retailers and requested stations to be nominated for the study. The stations were selected from an initial pool of 25 stations provided by PIA and eight supplied by the superstore retailers, and were chosen to give a representative cross-section of retailer, station size, proximity to busy roads, and VOC controls. As traffic figures were not readily available at the time of site selection, a subjective assessment was made of the adjacent road, based upon two categories 'busy' and 'non-busy'. There were no other petrol stations within 100 metres of the selected sites. A summary of the selected stations is given in Table 6, with a more detailed description provided in Appendix 1.
Table 6: Summary of Service Stations Selected for the Study
No Controls
Stage I Controls
Stage II Controls
Non-Busy Road
Station A Station B
Station E Station F
Busy Road
Station C Station D
Station G Station H Station I Station J
Superstore
Station K
Station L
4.2 Sampling and Analysis Techniques A total of six sampling locations were identified at each service station. These included four sites within the service station forecourt and two sites at the adjacent roadside at a distance of approximately 100 metres to either side of the station. Roadside sites were included in order to provide some indication of the contribution of benzene emissions from road traffic to measured forecourt levels. Of the four sites located within the service station forecourt one was located at the pumps in order to provide an indication of peak forecourt levels. The three remaining sites were located at the forecourt perimeter in order to provide an indication of levels which may be expected at residential properties adjacent to petrol stations. Two sites were located at either side of the forecourt and one site was located at the rear of the forecourt.
Monitoring was carried out at each location using Perkin Elmer passive diffusion tubes filled with Chromosorb 106 and fitted with diffusive heads. The tubes were positioned at each sampling location within a rainproof shelter, at a height of approximately 2 metres. Sampling was carried out over a period of six consecutive two-week periods from 2 April to 3 July 1997. Following exposure, the tubes were analysed by gas chromatography/mass spectrophotometry (GCMS) following thermal desorption. A description of the analytical procedures used is provided in Appendix 2.
In addition, a period of continuous monitoring was carried out at Station H for a period of 4 weeks from 13 March to 12 April 1996, using the Chrompack BTX Analyser. This analyser has previously been validated against the Chrompack VOCAIR system used in the UK Hydrocarbon Network.
