Development and Application of One Dimensional Multi-component Reactive Transport Models
Type of Degreedissertation
MetadataShow full item record
The overarching goal of this dissertation is to develop reactive transport models and explore their applications to groundwater remediation problems. The primary focus of this dissertation is aimed at developing models that can support laboratory studies investigating remediation strategies, as this is an important intermediate step before the remediation methods can be scaled up to apply at field sites. As a part of this research effort, a comprehensive, one-dimensional, multi-component reactive transport model, RT1D, which can be used for simulating biochemical and geochemical reactive transport problems, has been developed. The code can be run within the standard Microsoft EXCEL Visual Basic platform and it does not require any additional software tools. The capabilities of the tool were illustrated by solving several benchmark problems taken from the literature that have varying levels of reaction complexity. These literature-derived benchmarks were used to highlight the versatility of the code for solving a variety of practical reactive transport problems. This model was subsequently applied to a published experimental dataset that described bioaugmentation processes to remediate PCE-DNAPL trapped in a fracture system. A mathematical framework was first formulated to model the bioremediation processes in a PCE contaminated single fracture system augmented with Dehalococcoides Sp. (DHC). The mathematical framework describes multi-species bioreactive transport processes that include bacterial growth and detachment dynamics, biodegradation of chlorinated species, competitive inhibition of various reactive species, and the loss of daughter products due to back-partitioning effects. Two sets of experimental data, available in Schaefer et al. (2010b), were used to calibrate and test the model. The simulation results indicate that the yield coefficient and the DHC maximum utilization rate coefficient were the two important process parameters. A detailed sensitivity study was completed to quantify the sensitivity of the model to variations in these two parameter values. The proposed model provides a rational mathematical framework for simulating remediation systems that employ DHC bioaugmentation for restoring chlorinated solvent contaminated groundwater aquifers. While calibrating the DHC bioaugumentation model, several inefficiencies related to the use of trial and error methods for parameter estimation were identified. In order to improve the efficiency of the parameter estimation process, a parallel genetic algorithm (PGA) was developed to automate the parameter estimation process. The performance of the PGA was tested by solving four benchmark problems that have published experimental data or analytical/numerical solutions. Benchmarking results indicate that the PGA estimated parameters are close to the true parameters. A shared memory parallel computing platform that utilized OpenMP FORTRAN was used to demonstrate the speedup of the code on a four processor desktop Pentium computer. The parallelized code showed linear speedup with increasing number of processors. The PGA routines used in this study are generic and can be easily adapted to solve parameter estimation problems in other environmental modeling applications.