Doctor of Philosophy (PhD)


Chemical Engineering

Document Type



This study investigated the correlation of desorption resistance of hydrophobic organic contaminants (HOCs) in sediment and the heterogeneity of sediment organic matter (SOM). The amorphous and condensed phase organic carbon contents were defined by the method of Gustafsson et al. (1997) using thermal oxidation of sediment under 375°C. The desorption-resistant fraction of contaminants in sediments was operationally defined as the fraction that could not be removed by the adsorbent XAD-2. The correlation of the desorption-resistant fraction with sediment organic carbon characteristics was evaluated for both laboratory-inoculated and field-contaminated sediments. No strong correlation between the desorption resistant fraction in laboratory-inoculated sediments with their condensed phase organic carbon contents was observed, apparently due to the relatively short period of sorption and desorption. The desorption-resistant fractions of contaminants in field-contaminated sediments, however, were well correlated with the amount of condensed phase organic carbon. Detailed size and density separation of sediment did not assist in understanding the desorption-resistant phenomena. Large particles could be visually separated, however, and differentiation into woody, charcoal, coal-like and coal-cinder particles provided insight into desorption characteristics and equilibrium partitioning. A mechanistic model of sorption and desorption including both kinetic and equilibrium effects was developed assuming that sorption and desorption to the amorphous carbon is fast, reversible and characterized by relatively low partition coefficients while sorption and desorption to the condensed phase carbon is slow and characterized by relatively high partition coefficients. The model assumes that sorption to the amorphous carbon is associated with pore diffusion while sorption to the condensed phase carbon is associated with solid phase diffusion. Model simulations of sorption and desorption in field-contaminated sediments were more successful than those in laboratory-inoculated sediments. The diffusivity of contaminant and volume to surface area ratio in the condensed phase organic matter were two fitting parameters used in the model. The results were more sensitive to the diffusivity of the condensed phase organic carbon fraction. Changes in diffusivity of an order of magnitude or more were required to significantly impact the correlation between the model and observations.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Danny D. Reible