Sorption of atrazine and dicamba in Delaware coastal plain soils: a comparison of soil thin layer and batch equilibrium results

Abstract

The mobility and retention of atrazine and dicamba in six Atlantic Coastal Plain soils were estimated by soil thin-layer chromatography (soil-TLC). The soils studied were representative of the major agricultural regions in Delaware and were sampled, by horizon, to the water table. Four horizons from each profile were leached simultaneously with distilled water on one soil-TLC plate. Two values were obtained from each plate: the ratio of the distance traveled by the herbicide center of mass over that traveled by the solvent front (Rm), and a sorption distribution coefficient (Kd). The Rm values ranged from 0·06 to 0·94 for atrazine and from 0·80 to 0·94 for dicamba. Herbicide mobility was found to be greatest in coarse-textured soil horizons that contained low levels of organic matter, clay, and Fe and Al oxides. Correlation analysis indicated that effective cation exchange capacity, exchangeable acidity, exchangeable aluminum, and clay were useful predictive variables or both atrazine mobility and sorption. Organic matter was not useful for predicting soil-TLC derived sorption estimates; however, it was correlated to Kd-batch estimates. Distribution coefficients calculated from soil-TLC data were found to be in general agreement with Kd values obtained for the same soils by batch equilibrium techniques. The average Kd-soil-TLC values for atrazine and dicamba were 2·09(±2·24) and 0·03(±0·02), respectively. The ratio of the batch Kd to the soil-TLC Kd ranged from 0·1 to 19 (x̄=1·6, SD=3·8) for atrazine and from 2·9 to 38 (x̄=12·6, SD=8·7) for dicamba. Thus, although for some horizons agreement between the two methods was good, for other horizons significant discrepancies existed. It is suggested that the soil-TLC gives results under non-equilibrium conditions, whereas the batch procedure is, by definition, at quasi-equilibrium. These fundamental differences may account for the observed differences between the two methods. It is also suggested that, due to this difference, the soil-TLC procedure can provide additional information relevant to herbicide partitioning in the field environment that is not provided by traditional batch equilibrium techniques. © 1998 Society of Chemical Industry