Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Environmental Engineering and Earth Sciences

Committee Member

Dr. David L. Freedman, Committee Chair

Committee Member

Dr. Ronald W. Falta

Committee Member

Dr. Lawrence C. Murdoch

Committee Member

Dr. Kevin T. Finneran


Chlorinated ethenes are common industrial chemicals and among the most frequently detected groundwater contaminants. Behavior and remediation of chlorinated ethenes in unconsolidated aquifers composed of granular materials (e.g., sand, silt and gravel) has been extensively studied for several decades. Nevertheless, it was not until the end of twentieth century that the role of matrix diffusion in plume persistence gained widespread acceptance. Matrix diffusion commonly occurs in complicated hydrogeological settings such as fractured sedimentary bedrock aquifers, where the permeable fractures act as the major conduit for groundwater flow and the less permeable but high capacity matrix acts as the primary storage place for contaminants. A fractured sandstone aquifer at an industrial site in southern California is contaminated with trichloroethene (TCE) to depths in excess of 244 m. Field monitoring data and previous microcosm studies suggest that TCE is undergoing reductive dechlorination to mainly cis-1,2-dichloroethene (cDCE) and additional attenuation through slow abiotic transformation that generates acetylene, CO2 and soluble compounds (referred to as non-strippable residue, or NSR). Biostimulation has been identified as a promising technology to treat this site by enhancing both biological and abiotic degradation. The objectives of this study were to determine the effect of biostimulation on reductive dechlorination of TCE to cDCE and other transformation pathways using crushed rock microcosms; to develop and operate a novel type of microcosm composed of intact rock cores to evaluate natural attenuation and biostimulation in a fracture-matrix system; and to develop and validate a numerical model for reactive transport of chlorinated ethenes in intact rock core microcosms, and then use the model to determine rate constants for natural attenuation. To achieve the first objective, over 500 crushed rock microcosms were constructed, using TCE and cDCE, and eleven treatments covering various types of amendments. In addition to the conventional headspace and liquid phase analyses, 14C-labeled TCE and cDCE were used to quantify the rate and extent of product formation; enrichment in δ13C was measured in microcosms without 14C added. Lactate, hydrogen release compound® (HRC), and emulsified vegetable oil (EVO) significantly enhanced the rate of TCE reduction to cDCE. Lactate also stimulated reductive dechlorination of cDCE to vinyl chloride (VC) and ethene, suggesting the presence of indigenous Dehalococcoides that are not active in situ due to donor-limited conditions. Illumina sequencing and qPCR analysis demonstrated that Geobacter spp. are responsible for reductive dechlorination of TCE to cDCE and Dehalococcoides spp. for reduction of cDCE to VC and ethene. The rate of TCE reduction to cDCE and cDCE to VC was faster than for VC to ethene, suggesting that Dehalococcoides perform the final dechlorination step co-metabolically. This was subsequently confirmed in enrichment cultures fed with VC where no activity was observed, while TCE and cDCE were readily reduced to ethene. Abiotic transformation of TCE and cDCE was observed based on accumulation of 14C daughter products and δ13C enrichment in the absence of reductive dechlorination. Electron donor and sulfate amendments did not enhance abiotic transformation, in spite of repeated sulfate consumption. Accumulation of 14CO2 plus 14C-NSR in unamended microcosms was used to determine pseudo-first order abiotic transformation rates of 0.038 yr-1 for TCE and 0.044 yr-1 for cDCE, corresponding to half-lives of 18 and 16 yr, respectively. Since crushing disturbs the surface area of the rock, it was unclear the extent to which crushed rock microcosms deviate from the behavior in undisturbed rock. The second objective was to learn about the processes in undisturbed rock using intact core microcosms. A novel microcosm design was developed. Each microcosm consisted of a sandstone core inserted between stainless steel end caps, and sealed inside layers of Teflon tape, a Teflon sleeve and an outer stainless steel case. Site groundwater amended with TCE, bromide and resazurin was forced through the rock under pressure to contaminate the core. One end cap was hollowed out to create a groundwater reservoir and was connected with two valves for sampling. Paired cores with similar characteristics were set up, one serving as an unamended control and the other as a treatment biostimulated with lactate. Lactate was chosen because it was the most effective electron donor for enhancing reductive dechlorination in crushed rock microcosms. Weekly sampling was conducted that also served the purpose of lactate delivery and to generate a groundwater flow over the simulated fracture. Samples were analyzed for TCE and volatile daughter products, anions, organic acids, and pH. Evaluation of δ13C was carried out every 3-4 months. Lactate addition created low redox conditions and stimulated sulfate reduction as well as reductive dechlorination. However, only TCE to cDCE degradation occurred, indicating a low population or absence of indigenous Dehalococcoides, potentially caused by their heterogeneous distribution at the site. Biostimulation significantly enhanced the contaminant removal rate by increasing the concentration gradient of TCE between the matrix and fracture and converting TCE to its more mobile daughter product, cDCE. Enrichment in δ13C was observed for TCE in rock core microcosms that did not undergo a discernible level of reductive dechlorination, and for cDCE formed via reductive dechlorination of TCE. This outcome indicated that an alternative transformation pathway for TCE and cDCE occurred, as observed in the crushed rock microcosms. To achieve the third objective, a numerical model was developed in a 2D radial symmetrical system using COMSOL Multiphysics. The model simulates diffusion, biotic/abiotic reactions, sampling and isotopic fractionation within dual porosity media (i.e., rock-fracture). The model was successfully calibrated with three sets of experimental data (TCE/cDCE, bromide and δ13C) from the intact core microcosms, and generated site-specific parameters including rock diffusivity, Monod kinetic constants, and abiotic transformation rates. This greatly elevated the relevancy and applicability of intact core microcosms to evaluation of transformation processes that occur in the field. Sensitivity analyses indicated that parameters such as matrix diffusivity, maximum specific growth rates, and decay coefficients play key roles in controlling the TCE and cDCE concentration. Also, abiotic enrichment factors have a significant impact on predicting the rates of TCE and cDCE transformation. Model simulations indicated that abiotic transformation is governed by reactions in the core, while reductive dechlorination occurred in both the chamber and the core. Abiotic transformation rates predicted by the model based on intact rock core microcosms correspond to half-lives of 37 to 88 yr for TCE and 37 to 63 yr for cDCE. These rates were longer than those determined with 14C-labeled compounds in the crushed rock microcosms, likely due to an increase in surface area during crushing.