Date of Award

Fall 9-2015

Embargo Period

10-6-2016

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil and Environmental Engineering

Advisor(s)

Athanasios Karamalidis

Second Advisor

Greg Lowry

Abstract

Carbon capture, utilization, and storage (CCUS), including enhanced oil recovery (EOR), is one of the most promising mitigation strategies for climate change. CCUS involves the capture of CO2 from point sources and the subsequent injection of that CO2 into geologic storage formations. Depleted oil reservoirs will be the first targets of CCUS due to the economic benefits associated with EOR. EOR operations are expected to produce large volumes of wastewater brines with the crude oil. Brines, which can have high concentrations of salts and dissolved organic compounds; and CO2, which can have dissolved organic compounds, have potential to leak into shallower aquifers. Therefore, fundamental research is needed on the levels of organic compounds in both the reservoir brines and CO2 in case of leakage. This thesis was divided into two parts. Part I was concerned with the aqueous solubility of organic compounds in brines. The presence of dissolved salts typically results in a decrease in organic compound aqueous solubility, this is called the salting-out effect, and it is typically modeled by the Setschenow Equation. Setschenow constants, which are empirical salting-out parameters, are assumed to be additive, meaning that they are applicable in mixed electrolyte solutions. However, this has not been verified by extensive experimental work. For accurate risk assessment modeling, Setschenow constants are needed for NaCl and CaCl2 for hundreds of organic compounds relevant to oil and gas reservoirs. However, there are only ~190 reported NaCl Setschenow constants and ~19 reported CaCl2 Setschenow constants. For the majority of these compounds, the validity of the Setschenow Equation has only been proven up to 0.5 – 1 M NaCl/CaCl2; and has not been extended up to salt concentrations relevant to oil and gas reservoir brines. The first objective of this study was to determine the validity of the Setschenow Equation for selected hydrophobic compounds in the range of 2 – 5 M NaCl, 1.5 – 2 M CaCl2, and in mixed electrolyte brines. The salting-out effect was measured in NaCl, CaCl2, and mixed Na-Ca brines for naphthalene, fluorene, phenanthrene, thiophene, benzothiophene, and dibenzothiophene. In this study, the Setschenow Equation was proven to be valid up to 2 – 5 M NaCl and 1.5 – 2 M CaCl2 for the organic compounds studied here. Setschenow constants were additive for fluorene and thiophene from moderate to high ionic strengths. Results demonstrated that previously determined Setschenow constants measured at low salt concentrations do not need to be re-measured at high salt concentrations. Objective 2 was to determine the validity of the Setschenow Equation for selected hydrophilic compounds up to 5 M NaCl, 2 M CaCl2, and in mixed electrolyte brines. The Setschenow Equation was proven to be valid in predicting the salting-out effect up to those high salt concentrations for three phenol, p-cresol, hydroquinone, 9-hydroxyfluorene, pyrrole, and hexanoic acid. Setschenow constants were additive for p-cresol and 9-hydroxyfluorene up to high ionic strengths. In addition to demonstrating the validity of the Setschenow Equation for these selected organic compounds, both Objective 1 and 2 added to a sparse database of NaCl and CaCl2 Setschenow constants. In Objective 3 models were evaluated, updated, and developed for prediction of Setschenow constants. Two models, a poly-parameter linear free energy relationship (pp-LFER) and a single parameter (sp) LFER, for prediction of NaCl Setschenow constants were evaluated and updated with new NaCl Setschenow constant data from both this study and the literature. The pp-LFER uses the Abraham solvation parameters of the organic compound of interest as inputs and the sp-LFER uses the octanol-water partitioning coefficient of the organic compound to predict NaCl Setschenow constants. Both models produced predictions of Setschenow constants that had good agreement with the experimental NaCl Setschenow constants in this study. The update of these models increased the breadth of organic compounds, and therefore confidence, in these models. In addition, four new models were developed to predict Setschenow constants of four other salts, which include CaCl2, KCl, LiCl, and NaBr. Extensions of this study include determining whether the Setschenow Equation is valid in predicting the salting-out effect for additional organic organics, different salts, and in additional mixed electrolyte systems. Finally, Part II of this study explored the data gaps related to the partitioning of organic compounds from water to sc-CO2. Objective four was to develop new linear partitioning models based on experimental water-sc-CO2 data of selected nitrogen, sulfur, and oxygen containing organic compounds and literature data. There are only ~37 organic compounds that have reported water-sc-CO2 partitioning coefficients; however thousands of partitioning coefficients are needed over a range of temperature and pressure conditions. Therefore, models are needed to accurately predict these partitioning coefficients. Partitioning coefficients over a range of temperatures and pressures were measured for thiophene, pyrrole, and anisole. Those measured partitioning coefficients followed trends based on vapor pressure and aqueous solubility. These partitioning coefficients, along with literature values were used to update a pp-LFER. Five new models based on inputs of vapor pressure, aqueous solubility, and CO2 density were developed to predict water-sc-CO2 partitioning coefficients. Those models were developed using data from this study and literature data. Four of those models are specific to organic compound classes, which include monopolar substituted benzenes, polar substituted benzenes, chlorinated phenols, and nitrogen containing compounds, and the other model is available to any organic compound that has vapor pressure, aqueous solubility, and CO2 density inputs that fall within the specified training range. Possible extensions of this study include further research testing of different groups of organic compounds in water-sc-CO2 systems, co-solvency effects, determining the effects of the salting-out effect in water-sc-CO2 partitioning, and using sc-CO2 for water treatment.

Share

COinS