Geochemical reaction modeling of carbon dioxide

Geochemical reaction modeling of carbon dioxide

##### Files

##### Publication or External Link

##### Date

2014

##### Authors

Alizadeh Nomeli, Mohammad

##### Advisor

Riaz, Amir

##### Citation

##### DRUM DOI

##### Abstract

Recently, the need to decrease CO<sub>2</sub> concentration in the atmosphere has been recognized because of the role of CO<sub>2</sub> as a greenhouse gas that contributes to global warming. Carbon Capture and Sequestration CO<sub>2</sub> is one of the most promising long term solutions for the reduction of CO<sub>2</sub> in the atmosphere. To this end, injection of CO<sub>2</sub> into saline aquifers has been proposed and investigated theoretically and experimentally in the last years. Modeling the storage of CO<sub>2</sub> in saline aquifers on a reservoir scale is very demanding with respect to computational cost. Long-term subsurface storage of CO<sub>2</sub> in saline aquifers may induce a range of chemical processes in response to disturbances in existing chemical equilibria that include, but are not limited to, dissolution of primary minerals and precipitation of secondary carbonates. CCS projects also can be done above the ground by injecting CO<sub>2</sub> into designated fractures. The work presented in this thesis focuses on developing a fundamental understanding and modeling approach for 3 basic aspects of sequestration: 1) the effect of CO<sub>2</sub> solubility on rates of geochemical reactions, 2) density of brine after dissolution of CO<sub>2</sub>, and 3) time dependent porosity variation of a single fracture due to precipitation of carbonates.
A new method is developed to determine the reaction rates of minerals, on the basis of transition state theory, in saline aquifers containing brine and a supercritical CO<sub>2</sub> phase. A general Arrhenius-type equation that depends explicitly on the pH of brine is employed to determine the reaction rates. The dependence of pH on the amount of aqueous CO<sub>2</sub> dissolved in brine is modeled in this study. An accurate pressure-volume-temperature-composition (PVTx) model is employed to determine the dissolution of supercritical CO<sub>2</sub> in brine for the temperature range of 50-100<super>°</super>C and pressures up to 600 bar. Solubility of CO<sub>2</sub> and dissolution rate of calcite predicted by the model are validated with experimental data available in the literature for specific thermodynamic and salinity conditions. The effects of CO<sub>2</sub> activity on pH and reaction rates are evaluated by means of four different models for the activity coefficient of dissolved CO<sub>2</sub>. The results indicate that dissolution of CO<sub>2</sub> decreases the pH of the system but an increase in the temperature and salinity values limits the pH reduction. The rates of reactions are found to increase with pressure and temperature. The results suggest that among the available mineral compositions in deep saline aquifers, the dissolution rate of anorthite is the rate limiting factor. The significance of the pre-exponential factor and the reaction order associated with the modified Arrhenius equation is evaluated to determine the sensitivity of the reaction rates as a function of the system pH. We find that the transition state theory can reasonably reproduce experimental data with new parameters that we are obtained in this study on the basis of sensitivity analysis. Finally, we develop a long-term geochemical modeling of CO<sub>2</sub> storage on a designated fracture above the ground to investigate the impacts of temperature, pressure, and salinity on the reaction rates and, subsequently, the critical time of blockage due to precipitation of carbonates.
With regards to the second focus of this thesis, the effect of CO<sub>2</sub> solubility on the density of binary H<sub>2</sub>O-CO<sub>2</sub> and ternary H<sub>2</sub>O-CO<sub>2</sub>-NaCl solutions in saline aquifers is investigated. These solution densities as dispensable properties play pivotal roles in estimating the dynamic evolution of plume containing aqueous CO<sub>2</sub> and brine and also affect fracture dynamics in geologic media. An improved model is proposed to predict the density of saturated binary and ternary solutions as a function of pressure, temperature, and salinity. The model is based on an extended form of the Redlich-Kwong Equation of State that yields more representative values of the molar volumes of liquid CO<sub>2</sub>. The extension involves finding new coefficients for the cubic Redlich-Kwong equation to match one of the solutions with the liquid molar volume deduced from experimentally measured density of binary solutions. The new coefficients are constrained to ensure that the solution branch corresponding to the molar volume in the gas phase remains unchanged. Due to variability in the experimentally measured density of binary solutions, an effective liquid molar volume is obtained with the help of a multi-parameter non-linear regression with respect to pressure and temperature. The proposed model is validated by comparing it with experimental data and is used to study the density-pressure-temperature-salinity relationship. It is found that the density of the saturated solution is proportional to the mole fraction of dissolved CO<sub>2</sub>. In general, solution density is found to increase with pressure but decreases with an increase in temperature within the range of 50-100<super>°</super>C. It is also found that the solution density decreases monotonically with an increase in salinity.
Finally, a new model is presented to simulate a reactive fluid within a fracture. Permeability of a fracture controls the path of aqueous CO<sub>2</sub> migration, therefore aperture width of a fracture has a pivotal effect on solubility and mineral trapping of injected CO<sub>2</sub>. This study investigates the impact of the formation of precipitates within fractures on CO<sub>2</sub> transport and storage capacity. The problem involves the flow of CO<sub>2</sub> between finite walls that represents a single fracture. Fluid convection, diffusion, and chemical reactions inside a finite space are solved as a simplified representation of natural mineral trapping. The model is composed of direct numerical simulation of incompressible flow and transport combined with the kinetics of corresponding chemical reactions. For each time step, transport and reactions are solved by means of a finite difference method using a sequential non-iterative approach. The purpose of the current study is to show the time evolution of the aperture shrinkage caused by precipitation of calcite. The current model predicts the actual efficiency of the mineral trapping mechanism by considering the physical properties of the fluid such as its density, pH, and the characteristics of the mineral compositions. It is found that for low Peclet numbers (Pe ≤ 100), the critical time for a fracture blockage is insensitive to the Reynolds number and increases with the Peclet number. At a high Peclet number (Pe=1000), however, the critical time generally decreases with increasing Reynolds number.