Technical Comments

Response to Comment on “Salt-Pump Mechanism for Contaminant Intrusion into Coastal Aquifers”

Science  31 Oct 2003:
Vol. 302, Issue 5646, pp. 784
DOI: 10.1126/science.1089464

We appreciate the opportunity to elaborate on aspects of our analysis (1) that were condensed largely because of space limitations. In our paper, we showed enhanced transport of organic compounds (OCs) from a saltwater source to a freshwater collector. The transport of organic molecules as a result of the “salt pump mechanism” is clearly not a purely molecular diffusive process, and, indeed, density differences between the saltwater and freshwater reservoirs influence the transfer of salt and OC across the sand interface. These effects were recognized implicitly in our paper. The main points we chose to stress were (i) the existence of a high “carrying capacity of the aqueous solution” (CCAS), well above OC solubilities, and (ii) the potential for transfer of OCs into coastal aquifers.

Density-driven transport leads to significant transport of salt and OCs, presumably both in dissolved and in droplet form. As noted by Lloyd (2), figure 1B in Dror et al. (1) can be used to examine contributions of density-dependent and diffusive transport of salt across the interface (the appellation “diffusive transfer” should not have been used therein). Clearly, comparison of the curves to a solution of the standard diffusion equation shows the significant effect of density-dependent transport, which is as much as an order of magnitude or more faster than purely diffusive transport. This rate of transport can be corroborated by a density-driven flow balance analysis within the flow cell (3), which provides a similar estimate for the density-driven transport rate of OC and salt.

However, even diffusive transport is significant over a matter of hours to days, and it is crucial to recognize that in contrast to NaCl, OC concentrations of only a few ppb render freshwater unpotable. We argue that the “salting-out effect” (which refers explicitly to the lower solubility of OCs in saltwater relative to freshwater, and the resulting effect of enhanced diffusive transport of OCs from saltwater to freshwater) indeed exists and contributes explicitly to the transport of OCs across the interface. Also, we have shown that the CCAS is well above solubility limits in saltwater, and much higher than that in freshwater. As such, “diffusive” transport of organic droplets must be considered. To further demonstrate these effects, and to permit separation of OC transport effects by salting out and density-driven transport, we performed additional experiments using a setup similar to that described previously (1), but with vertically oriented flow cells: the source cell (with a gentle mixer) sits below a horizontal sand interface, with the (freshwater) collector cell above it. This orientation minimizes effects of density-driven transport, and OCs appearing in the collector cells can be attributed to “diffusive” transport. These experiments (3) showed conclusively that OC concentrations in the collector cell can increase by a factor of two or more when the source cell contains saltwater rather than freshwater. For example, OC concentrations as high as 900 to 1600 ppb toluene were found in the collector cells—for freshwater and saltwater source cells, respectively—in as little as 30 min.

Related analyses of coupled OC salting out and diffusion in aqueous solutions have shown cotransport of several moles of dioxane for each mole of diffusing electrolyte, even if the electrolyte is infinitely diluted. The driving force of the salting out is the increasing gradient in the dioxane chemical potential with increasing electrolyte concentration (4). Additional work dealing with an aqueous–aqueous OC extractive membrane process (5) demonstrated that the ionic strengths of aqueous solutions on both sides of a membrane directly influence OC transport, because of increased activity gradient and equilibrium properties.

As a consequence, we conclude that the presence of salt, together with a chemical gradient of OCs, leads to enhanced transport of OCs. Whether the diffusive-transport contributions to dissolved and droplet-form OC migration are by Fickian diffusion, by other types of diffusion [for example, “turbulent diffusion” (6)], or by both—and how to parameterize this diffusive transport and its interactions with density-driven transport— remains a subject of research. Considering all of these results and those reported in (1), it is suggested that salting out advances droplets and dissolved OC to the saltwater-freshwater interface, and that density effects (together with diffusion) force the OC further ahead into the collector cell.

We agree entirely that many questions remain unanswered, at the laboratory scale and, most notably, with regard to the magnitude of OC transport from actual (open system) contaminated marine environments to coastal aquifers.


Related Content

Navigate This Article