CRITICAL CONCEPTS FOR COLUMN TESTING

Concepts critical for the evaluation of the miscible transport parameters from laboratory soil columns are presented. Failure to recognize a distinction between solute concentrations within the soil and solute concentrations collected in an effluent reservoir may result in errors in the evaluation of the column Peclet number, P(L), and the retardation factor, R(d). An evaluation of three analytical models commonly used to describe miscible transport through soil columns indicates that only one model is capable of predicting the diffusive dominated transport observed in laboratory and field studies for transport at low flow rates. The common assumption that the retardation factor (R(d)) is equal to the pore volumes of flow (T) at which the relative effluent concentration [(c(e) - c(i))/(c(o) - c(i))] equals 0.5 is shown to be invalid at low column Peclet numbers (P(L) less-than-or-equal-to 10) based on mass balance constraints for finite soil columns. A more correct assessment of R(d) is obtained by considering the area above the effluent breakthrough curve at equilibrium conditions [i.e., at (c(e) - c(i))/(c(o) - c(i)) = 1]. Finally, when effluent is allowed to collect in a reservoir over regular sampling intervals, the measured concentrations represent average, incremental concentrations. Regression analyses indicate that errors in regressed values of both R(d) and P(L) in terms of an analytical model based on instantaneous concentrations generally are less-than-or-equal-to 10% when the sampling interval, DELTAT, less-than-or-equal-to 0.25 and the average, incremental concentrations are plotted at the middle of the sample interval, and the errors tend to decrease as P(L), R(d), and DELTAT decrease.