| Summary: | Former manufactured gas plants (FMGPs) are a ubiquitous source of environmental con- tamination. The process of gas production created a number of by-products, including coal tar and ammoniacal liquor. Coal tar contains a complex mixture of organic and inorganic compounds, many of which are toxic and carcinogenic. It is estimated that over 3000±1000 FMGPs exist in the United Kingdom alone, yet there are few recent publications detailing the analysis of coal tars. The complex composition of coal tar is known to vary due to a number of factors, including production method, temperature and coal type, making the analysis and interpretation of such samples extremely challenging. Environmental forensics is concerned with the source, fate and transport of contaminants. The introduction of recent legislation such as the EU Environmental Liabilities Directive 2004/35/EC, which promotes the "polluters pay" policy, has encouraged the development of accurate and robust scientific methods for the identification of contaminants. Analytical instrumentation is constantly evolving, thus new protocols to trace the origin of contamination must also be developed to utilise these technological advances. Two-dimensional gas chromatography (GCxGC) and compound specific isotope analysis (CSIA) are two exam- ples of advanced analytical instruments which have the potential to aid source identification. GCxGC provides enhanced separation of complex mixtures compared to conventional gas chromatographic techniques, while CSIA allows chemically identical contaminants to be compared based on their isotopic composition. In this study, preliminary research investigated the isotopic composition of coal tars, as this technique is currently a major tool for source apportionment in environmental forensics. However, the results demonstrated .that the similar nature of British coals used for gas production at the investigated sites produced similar isotopic values in the resultant tars, making the technique redundant in this case. Therefore, the potential of GCxGC was investigated for ultra resolution chemical fingerprinting of coal tars. Traditionally, chemical fingerprinting of complex mixtures, such as coal tar, is performed using a tiered approach including rigorous sample preparation steps and analysis by multiple instruments. In this work, a new, single-step analytical procedure was developed for the analysis of coal tars by GCxGC. Automated sample extraction techniques combined with GCxGC analyses were employed to provide detailed chemical fingerprinting in a fast, yet accurate, manner. This research represents a major advance in knowledge of compositional variation within coal tars. The enhanced separation of GCxGC provides vast quantities of chemical data which can be difficult to interpret without statistical methods. A multivariate statistical model was developed to provide process-specific classification of coal tars. The statistical model was validated through use of a blind study, indicating that process-specific apportionment of coal tars was achievable. Based on these results, the application of the single-step procedure for environmental forensics on a commercial basis was evaluated.
|
|---|
