Environ. Sci. Technol. 2002, 36, 314-322
Mobilization of Natural Colloids from an Iron Oxide-Coated Sand Aquifer: Effect of pH and Ionic Strength REBECCA A. BUNN,† ROBIN D. MAGELKY,† J O S E P H N . R Y A N , * ,† A N D MENACHEM ELIMELECH‡ Department of Civil, Environmental, and Architectural Engineering, University of Colorado, Boulder, Colorado 80309-0428, and Department of Chemical Engineering, Environmental Engineering Program, Yale University, New Haven, Connecticut 06520-8286
Field and laboratory column experiments were performed to assess the effect of elevated pH and reduced ionic strength on the mobilization of natural colloids in a ferric oxyhydroxide-coated aquifer sediment. The field experiments were conducted as natural gradient injections of groundwater amended by sodium hydroxide additions. The laboratory experiments were conducted in columns of undisturbed, oriented sediments and disturbed, disoriented sediments. In the field, the breakthrough of released colloids coincided with the pH pulse breakthrough and lagged the bromide tracer breakthrough. The breakthrough behavior suggested that the progress of the elevated pH front controlled the transport of the mobilized colloids. In the laboratory, about twice as much colloid release occurred in the disturbed sediments as in the undisturbed sediments. The field and laboratory experiments both showed that the total mass of colloid release increased with increasing pH until the concurrent increase in ionic strength limited release. A decrease in ionic strength did not mobilize significant amounts of colloids in the field. The amount of colloids released normalized to the mass of the sediments was similar for the field and the undisturbed laboratory experiments.
Introduction Concern over colloid-facilitated transport of contaminants in the subsurface environment (1, 2) has spurred interest in the genesis of colloids. In most cases, the source of colloids is mobilization of fine particles from the porous medium itself (3, 4). Mobilization of an amount of colloids sufficient to allow colloid-facilitated transport usually requires a perturbation to the existing chemical or physical conditions of the subsurface environment. Chemical perturbations leading to colloid mobilization in natural porous media include a decrease in ionic strength (5-10); a change in ionic composition from bivalent to monovalent cations (11); the introduction of a surfactant, complexing ligand, or reductant (12-19); or an increase in pH (13, 15, 16, 20, 21). While some of these perturbations can cause dissolution of mineral phases * Corresponding author phone: (303)492-0772; fax: (801)327-7112; e-mail:
[email protected]. † University of Colorado. ‡ Yale University. 314
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 3, 2002
resulting in colloid mobilization (18, 22), the common link between these perturbations is an increase in the electrostatic repulsion between colloids and grains. Once mobilized, colloids must bind contaminants and migrate through the subsurface environment for colloidfacilitated transport to be important. Normally, colloid transport occurs at the rate of groundwater flow (23), but in a chemically perturbed environment, we hypothesize that colloid transport is limited by the rate of transport of the chemical perturbation. If the transport of the chemical perturbation were retarded, then the colloid transport would also be retarded by re-deposition ahead of the chemical perturbation. If this “colloid mobilization front” scenario is accurate, the transport of mobilized colloids may be tracked by simply monitoring the migration of the chemical perturbation. Grolimund and Borkovec (24) theoretically explored a similar scenario based on travel velocities of the chemical perturbation and the colloids and a single rate coefficient for colloid release. Testing the colloid mobilization front scenario would require (i) a colloid-mobilizing agent for which transport was retarded and (ii) a sufficient length scale to observe the retarded transport. For retarded transport, a decrease in ionic strength alone would not suffice, nor would many of the surfactants, complexing ligands, and reductants used in previous colloid mobilization experiments because their rates of transport in porous media could not be easily predicted. An increase in pH by addition of sodium hydroxide was the best choice for a retarded chemical perturbation because of the high buffering capacity of most aquifer sediments. To clearly observe the retarded advance of an elevated pH “plume”, transport distances greater than those typically used in laboratory column (5-10 cm) would be needed. Groundwaters of elevated pH have been found near marl alteration zones (25) and cement liners at waste disposal sites (26). To test the colloid mobilization front concept, we conducted field experiments in the ferric oxyhydroxide-coated sand and gravel aquifer at the U.S. Geological Survey’s Cape Cod research site, where we monitored the natural-gradient transport of injected sodium hydroxide plumes over distances up to 6 m in both uncontaminated and sewage effluentcontaminated zones of the aquifer. The field experiments provided the added advantage of observing colloid mobilization in situ. We also conducted short-length scale column experiments with Cape Cod aquifer sediments to better understand the mechanisms of colloid mobilization by elevated pH. These laboratory experiments were conducted with undisturbed, oriented sediments and disturbed, disoriented sediments to compare the amount of colloids mobilized from the sediments under field and laboratory column conditions.
Materials and Methods Field Site. The field experiments were conducted at the U.S. Geological Survey’s Cape Cod Toxic Substances Hydrology research site near the Massachusetts Military Reservation on Cape Cod. About 50 yr of secondary sewage effluent disposal onto infiltration beds created a groundwater plume characterized by low dissolved oxygen and elevated pH, specific conductance, and organic carbon (Table 1). Above this plume, the uncontaminated groundwater is nearly saturated with dissolved oxygen and low in specific conductance and organic carbon. The aquifer sediments consist of Pleistocene glacial outwash with interbedded ferric oxyhydroxide-coated sand and gravel layers (27, 34). 10.1021/es0109141 CCC: $22.00
2002 American Chemical Society Published on Web 01/04/2002
Mc ) vA
TABLE 1. Characteristics of Groundwater and Sediments at U.S. Geological Survey Cape Cod Research Site parameter
uncontaminated zone
units
contaminated zone
ref
27
Groundwater pH alkalinity specific conductance temperature dissolved oxygen dissolved organic carbon Ca2+ + Mg2+ Fe(II) PO43-
µM µS cm-1
5.5-5.7 28 55-65
5.8-6.2 640 170-180
°C mg L-1 mg L-1
15-16 4.5-6.5 0.5-1.5
14-15 0-0.5 4-5
µM µM µM
65
