Date of Award

Spring 1998

Document Type


Degree Name

Doctor of Philosophy in Chemical Engineering - (Ph.D.)


Chemical Engineering, Chemistry and Environmental Science

First Advisor

Gordon Lewandowski

Second Advisor

Piero M. Armenante

Third Advisor

Basil Baltzis

Fourth Advisor

Norman W. Loney

Fifth Advisor

Edward J. Bouwer

Sixth Advisor

David Kafkewitz


This thesis was motivated by the need for better engineering tools to predict the extent of contaminant plume migration in the saturated zone. The principal result was a mathematical model analogous to a catalytic packed-bed reactor, in which the subsurface was considered to be composed (conceptually) of soil aggregates (the "catalyst" particles, in which biodegradation, diffusion, and sorption take place), and a mobile phase (groundwater) passing around the aggregates, in which convection, axial diffusion, and mass transfer from the aggregates take place.

The modelling emphasis was on a more detailed exposition of the biokinetics of the system (including an inhibitory expression for biomass growth, and oxygen limitation) than is the case in most prior models (which place greater emphasis on physical effects such as sorption, diffusion, and hydrodynamics).

Several coupled partial differential equations were used to describe the processes taking place, and these were solved numerically in dimensionless form by the Method of Lines. Model parameters were determined by a combination of laboratory experiments, existing empirical correlations, and estimates based on the literature. Sensitivity analyses of these parameters showed that the biokinetic constants indeed dominated the system response, which justified the emphasis placed on those factors in the model development.

A laboratory soil column was used to test the model, and the results showed good agreement with model simulations. However, the simulations were particularly sensitive to three interrelated parameters (one of the inhibitory biokinetic constants, another related to tie rate of loss of active microorganisms from the system, and the initial biomass concentration), two of which (bacterial loss rate, and initial biomass concentration) were not known. The loss rate was presumably a function of both the rate of cell death as well as transport out of the system. Thus, some estimation and adjustment of these parameters was necessary in order to obtain a fit of the data.

The effect of oxygen limitation was simulated only. Confirmatory experiments proved difficult to conduct. The simulations indicated the dominant effect that oxygen has on the performance of the system. Even with a saturated feed, oxygen can be rapidly depleted in the direction of groundwater flow, leading to a substantial decrease in the rate of biodegradation.

Field tests of the model are currently being planned.