Phenanthrene Breakthrough and Elution in "Aged" Macropore Columns: Effects of Exposure Time

The overall goal of this work is to evaluate, quantify, and model the effect of long-term "aging" (exposure time) of contaminants with some selected (and perhaps exemplary) geosorbent materials. For many systems, a combination of pore diffusion and nonlinear sorption can lead to especially slow rates of desorption. When sorption equilibrium is not attained by the time desorption is begun, observed desorption rates can be unusually slow and appear as "hysteretic", despite being controlled by a single set of equilibrium and rate relations. In this work, we test the hypothesis that these phenomena are relevant to phenanthrene transport in a porous medium bounded by an impermeable layer of sorbing fine-grained material, as relevant to a field site at Dover AFB, DE. To test this hypothesis, experiments were conducted for various aging periods using the natural clay and silt geosolids obtained from the Dover site. The fine-grained material was loaded into a laboratory column and penetrated with a sand macropore in a manner previously described elsewhere (Young and Ball, Environ. Sci. and Tech., vol. 32, pp. 2578-2584, 1998). This column was subsequently subjected to breakthrough, aging and elution experiments. To evaluate the extent of system "predictability" during transport, breakthrough of phenanthrene was modeled on the basis of batch-derived sorption parameters and with hydrodynamic properties of the column as determined using dirac injections of tritiated water. To evaluate the applicability of the same conceptual model and parameters during desorption, column elution was subsequently conducted under varied conditions of "aging" and using model simulations that fully accounted for the non-equilibrium nature of sorptive uptake. Macropore columns were loaded with steady influent phenanthrene concentrations for either 10, 230, or 665 days prior to the onset of elution. At the end of the longer breakthrough periods, effluent concentrations were observed to be essentially equal to the influent concentration (within our ability to discern), although in fact the modeling suggested that none of the systems was actually at sorptive equilibrium. Subsequent elution and sampling was continued until the concentrations in the effluent were below detection using liquid scintillation methods. Finally, deconstruction and extraction of the experimental columns was used to determine the mass of the contaminant that was remaining after elution, as an additional point of comparison with modeled values. Both the modeling and measurements show that concentrations within the fine-grained region were still very high at the end of the experiment, despite the very low final effluent concentrations and fluxes. Residual sorbed phase concentrations increased with aging time, as expected. For the system that was subjected to the longest aging time and an elution period of 180 days, average residual phenanthrene concentrations were approximately 42% of the equilibrium capacity - i.e., 42% of the sorbed mass that would represent equilibrium with the initial influent. For these columns, tailing during desorption was severe and beyond that which would be expected without consideration of the initial (non-equilibrium) conditions. Although sorption nonlinearity has not yet been incorporated into our modeling (at the time of this writing), initial diffusion modeling suggests that non-equilibrium attainment during sorption can explain most of the observed effects.