Estimating the retardation coefficient of trichloroethene for a sand aquifer low in sediment organic carbon - a comparison of methods

Significant groundwater contamination by trichloroethene (TCE) has been identified in the Spearwood Sands aquifer, Perth Western Australia. The organic carbon content of the aquifer sediment is very low and reported retardation coefficients for TCE in low organic carbon subsurface environments an rare. Carbon-based correlation equations were reported to underestimate the retardation of nonionic hydrophobic organic compounds in aquifers characterised by low organic carbon contents. Estimating the retardation coefficient for TCE in the Spearwood Sands aquifer, therefore, is difficult. Various methods to estimate the retardation coefficient for TCE in the Spearwood Sands aquifer were investigated. Estimates obtained by commonly used correlation equations were compared to retardation coefficients calculated based on laboratory batch and column studies. An in situ transport experiment using deuterium labelled TCE (TCEd1) within a TCE contaminated zone was carried out to assess the reliability of the various estimation techniques. The field transport experiment showed that TCEd1 was not retarded in the Spearwood Sands aquifer. Carbon-based correlations overestimated the retardation coefficient for TCE. The equations are shown to be limited by the lack of reliable estimates of the organic carbon content for the Spearwood Sands. Laboratory batch studies resulted in more reliable estimates of TCE retardation. Retardation coefficients from the batch studies, however, were still overestimated since Teflon-coated silicone liners used in the experiments show considerably higher sorption capacity for TCE than does the aquifer sediment. Laboratory column studies were accurate in assessing the sorption behaviour of TCE in the Spearwood Sands. Column studies eliminated the additional uncertainty inherent in determining the input concentration for TCE, by assessing the retardation coefficient in relation to an inorganic tracer. Column studies also avoided the experimental difficulties experienced with the batch studies and proved to be more cost effective. It is shown, however, that also with batch experiments an accurate prediction of the field retardation for TCE can be obtained, if naphthalene is used in the batch studies to determine the sorptive properties of the sediment. (C) 1998 Elsevier Science B.V.