Urea hydrolysis and calcium carbonate reaction fronts

The mobility of toxic or radioactive metal contaminants in subsurface environments can be reduced by the formation of mineral precipitates that form co-precipitates with the contaminants or that isolate them from the mobile fluid phase. An engineering challenge is to control the spatial distribution of precipitation reactions with respect to: 1) the location of a contaminant, and 2) where reactants are introduced into the subsurface. One strategy being explored for immobilizing contaminants, such as Sr-90, involves stimulating mineral precipitation by forming carbonate ions and hydroxide via the in situ, microbially mediated hydrolysis of urea. A series of column experiments have been conducted to explore how the construction or design of such an in situ reactant production strategy can affect the temporal and spatial distribution of calcium carbonate precipitation, and how the distribution is coupled to changes in permeability. The columns were constructed with silica gel as the porous media. An interval midway through the column contained an adsorbed urease enzyme in order to simulate a biologically active zone. A series of influent solutions were injected to characterize hydraulic properties of the column (e.g., bromide tracer), profiles of chemical conditions and reaction products as the enzyme catalyzes urea hydrolysis (e.g., pH, ammonia, urea), and changes that occur due to CaCO3 precipitation with the introduction of a calcium+urea solutions. In one experiment, hydraulic conductivity was reduced as precipitate accumulated in a layer within the column that had a higher fraction of fine grained silica gel. Subsequent reduction of permeability and flow (for a constant head condition) resulted in displacement of the hydrolysis and precipitation reaction profiles upstream. In another experiment, which lacked the physical heterogeneity (fine grained layer), the precipitation reaction did not result in loss of permeability or flow velocity and the reaction profile, characterized by the pH profile and hydrolysis reaction species, was extended downstream of the enzyme zone. Downstream extension of the reaction profile was due partially to the partial mobility of the enzyme in the column. The experiments are helping to illustrate the complexity of transient reaction fronts as well as the needs and challenges for advanced modeling approaches. A modeling platform developed at the Idaho National Laboratory, which is capable of simulating tightly coupled physical-chemical processes (the Reactive Transport simulator), is being applied to pre-experimental simulations and post-experimental interpretation of results.