9418172 Elimelech The transport of radionuclides, metals, and non-polar organic compounds in groundwater is severely restricted by adsorption to immobile aquifer sediments. In the presence of colloids, however, these low-solubility contaminants migrate over distances much greater than those predicted by models that consider only the distribution of the contaminant between the dissolved and the adsorbed, immobile phase. The presence of colloids requires inclusion of an adsorbed, mobile phase of the contaminant in transport models. This requirement has shifted our attention to the mobilization, transport, and deposition colloids in aquifers. Colloid formation may occur by in situ precipitation or mobilization caused by chemical or physical perturbations in the aquifer. When a chemical perturbation increases the repulsive forces between the colloid and grain surfaces, colloids are mobilized and transported with the groundwater. If the transport of the solute causing that perturbation is retarded relative to the groundwater, the mobilized colloid will eventually pass the solute front and encounter sediments that have not yet been affected by the solute, or colloid-mobilizing agent. Under these conditions, we expect that the colloids will be re-deposited on grains and will remain there until the solute "catches up." We hypothesize that the transport of colloids mobilized by a chemical perturbation will never exceed the transport of the colloid-mobilizing agent. To test this hypothesis, we propose to (1) develop a model that will simultaneously account for the transport of the colloids and the colloid-mobilizing agent and (2) conduct a series of small-scale, intermediate-scale, and field experiments simulating and testing colloid mobilization and transport. The model will be developed and rigorously tested by the UCLA researchers. The model will account for colloid deposition and release, microscopic and macroscopic charge heterogeneity of colloid and grain surfaces, the effect of retained colloids on the deposition and release of colloids, solute adsorption and desorption, and "megascopic" heterogeneities (i.e., laying in aquifer sediments). It will be formulated to model colloid and solute transport in one and two dimensions for the laboratory experiments and it will be extended to three dimensions for the field experiment. The small-and intermediate-scale experiments will be conducted at the University of Colorado's Water Resources laboratory. The materials used in the experiments will include hematite and kaolinite colloids, quartz and ferric oxyhydroxide-coated quartz porous media, and phosphate dodecanoic acid (a surfactant), and isolate NOM from the field site as colloid-mobilizing agents. Small-scale column experiments will be conducted to identify parameters for the intermediate-scale and field experiments. The intermediate-scale experiments will be conducted in a two-dimensional tank of 10 m length 2 m height, and 5 cm width filled with homogeneous and heterogeneous (layered) porous media. The tank experiments will directly test the hypothesis relating colloid transport to the transport of the colloid-mobilizing agent. The field experiment will be conducted at the Barouch Forest Science Institute site in Georgetown, S.C. The surficial aquifer at the BFSI site is composed primarily of quartz sand, ferric oxyhydroxides, and layered heterogeneity. A field experiment is proposed that will examine the deposition and mobilization of synthesied kaolinite collids labeled with a stable isotope (deuterium or 18O) or titanium as an isomorphous substitute for silicon. Separate injections will test the effects of dodecanoic acid and NOM-rich water from a nearby pond as the colloid-mobilizing agent in both oxic and suboxic portions of the aquifer.
