Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering

First Advisor

W. David Constant


Bioremediation and phytoremediation are in their infancy and this research addresses concerns for application of these technologies. This laboratory and mathematical modeling effort predicts the pollutant flux from soils and sediments to the overlying waters. Little work has been done on transport in surface waters such as the slow moving and stagnant areas of the southern U.S. Mass transfer is a proposed rate limiting factor for biological transformation when using phytoremediation or natural attenuation. The research quantified mass transport of two dfferent contaminants, TNT and chlorinated benzenes, under different advective conditions. The TNT study was part of a "riffle-bed reactor" of the TNT leaching unit and the plant catalyzed (hydroponic) reactor. The contaminant flux rates to the biological reactor were quantified for successful use of the riffle bed system. Experimental flux rates were calculated based on overlying bulk water concentrations. Static diffusion flux rates were determined for the TNT system. Different flow regimes were tested to increase leaching efficiency for TNT. The experimental flux rates were compared to published diffusion models for predictability. The chlorinated benzene site underwent remedial activities including natural attenuation. Initial experiments were performed with a single individual contaminant, chlorobenzene or 1,3-dichlorobenzene. Experiments were also performed utilizing a mixture of chlorobenzene, 1,2-dichlorobenzene, 1,3-dichlorobenzene, and 1,2,4-trichlorobenzene because natural systems rarely contain one pollutant. The impact of multiple contaminants on desorption and dissolution were quantified via flux rates in dynamic stream systems. The continuous flow regimes followed characteristic exponential decay while the cyclic flow regime demonstrated spikes of higher flux when the flow rate changed, yielding a higher overall average flux rate. Thus demonstrating the pulsed flow regimes prevented equilibrium conditions thereby increasing leaching efficiency. The core results support the removal of contaminants from sediment due to leaching. The experimental flux was used to calibrate published diffusion models. Once calibrated, tortuosity and effective diffusivity were determined to be system descriptors. The two descriptors were then validated by comparing flux rates from the slow flow and cyclic flow to model predicted flux rates. Models adequately described and predicted chemical transport based on root mean squared error and correlation coefficient.