of ancient gaswork soils. Effect of polycyclic aromatic hydrocarbons (PAHs) on plant germination. Organic Geochemistry, Elsevier, 1999, 30 (8) ABSTRACTThe phytotoxicity of various contaminated soils was assessed by plant inventories on ancient industrial fields, and by phytotoxicity tests. Industrial fields are well colonised by numerous weedy plants. Phytotoxicity was tested with pure PAHs, ancient industrial soils, soil leaches, liquid tar and tar volatile compounds. Both field studies and toxicity tests show that contaminated samples can be classified into two categories: first, a recently excavated soil/liquid tar that was foul-smelling and phytotoxic, and second, an 'aged', surface soil that was weathered and nonphytotxoic. Plant germination and growth are strongly inhibited by the presence of volatile, water-soluble low molecular-weight hydrocarbons (< 3 rings) such as benzene, toluene, xylene (BTX), styrene, indene, naphthalene and other possibly toxic substances. On the other hand, high molecular weight PAH (3-5 rings) did not show any phytoxicity under the conditions studied. These findings suggest that once low molecular weight aromatic hydrocarbons are removed e.g. by volatilization, biodegradation, weathering, tillage and fertilising, plants should be able to grow.

Detailed analytical characterization of the organic matter (OM) of aged polluted soils from five former manufactured gas plants (MGP) and of two coal tars was completed. It was aimed at obtaining information relevant to the physicochemical state of the polycyclic aromatic hydrocarbon (PAH) pollutants and to their in-situ evolution in time. Overall characterization of total OM (essentially polluting OM) was carried out directly on soil samples with or without prior extraction with solvent. It involved a technique of pyrolysis/oxidation coupled to flame ionization/thermal conductivity detection (Rock Eval). Extracts in solvent were fractionated by liquid chromatography into saturated hydrocarbons, PAH, and resins, the first two fractions being further characterized by gas chromatography and mass spectrometry. The compositions of OM of soils were found to be very similar. A total of 28% of organic carbon, including all PAH, was extractable by solvent. The compositions of coal tars were qualitatively similar to those of OM of MGP soils but with a higher proportion (48%) of total extractable OM and of PAH, in particular lower PAH. Contamination of MGP soils appeared essentially as coal tar having undergone natural attenuation. The constant association of PAH with heavy OM in MGP soils is important with respect to the mobility and bioaccessibility of these pollutants.

Analytically well-characterized soils from four different former manufactured gas plants (MGP) sites contaminated by coal tars were used in tests of extensive biodegradation of polycyclic aromatic hydrocarbons (PAHs) in stirred reactors. In all cases, the extent of biodegradation was limited to 80-100% for 2-and 3-ring PAHs, 40-70% for 4-ring PAHs, and below 20% for 5-and 6-ring PAHs. The capacities to transfer pollutants to water were compared for leachates from soils that had or had not undergone biological treatment. Leachate analysis involved determination of PAHs and bacterial tests of acute toxicity (Microtox) and genotoxicity (SOS Chromotest). For some untreated soils, PAH leaching was observed, and positive responses to the Microtox test were well correlated to the concentrations of naphthalene and phenanthrene. Biologically treated soils had lost all capacities for leaching as concluded from PAH determinations and responses to the Microtox test. All soil leachates were devoid of genotoxic effect, in accordance with the low concentrations observed of mutagenic PAHs. The results of this risk-based approach for assessment of MGP soils showed that pollutants remaining after biological treatment were unavailable for further biodegradation and that the extent of leaching had been reduced to the level that it did not represent a significant threat to groundwater.