Publication Date

8-2012

Date of Final Oral Examination (Defense)

6-11-2012

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Geophysics

Department

Geosciences

Major Advisor

Ludmila Adam, Ph.D.

Advisor

Kasper Van Wijk, Ph.D.

Advisor

Lee Liberty, M.S.

Advisor

Craig White, Ph.D.

Abstract

The continued burning of fossil fuels as a source of energy is contributing to greater concentrations of carbon dioxide (CO2) in the atmosphere. Increased levels of CO2 in the atmosphere have been linked to an increase of global mean surface temperature. To mitigate the continued release of CO2, projects to capture this gas at large point sources and sequester it in geologic formations are in place. Carbon dioxide sequestration in basalts promises permanent trapping of the fluid as these rocks react with carbonic acid and precipitate carbonate minerals. It is important to monitor the injection of CO2 to assure it is not leaking into freshwater aquifers or towards the surface. Seismic methods are a geophysical tool that can be useful in monitoring physical changes in a reservoir.

To study the feasibility of using seismic methods to monitor rock property changes in this type of sequestration, I perform elastic wave laboratory experiments on basalt core at reservoir conditions. This thesis quantifies the elastic and physical property variations for basalt rocks exposed to CO2 and water. When CO2 is injected into a water-saturated basalt, two elastic wave propagation effects are expected: 1) the substitution of a more compressible fluid such as CO2 for water decreases the P-wave velocity and 2) stiffening of the rock, resulting from the dissolution of basalt and forming minerals under acidic conditions, P- and S- wave velocities increase as carbonates precipitate in pores and cracks. Although theoretically these are the expected changes in wave velocity, there have not been any previous studies on elastic rock properties in basalts with CO2 and on real time mineralization on whole basalt core.

At ultrasonic frequencies and at a differential pressure of 17.2 MPa (depth of 1Km), I measure a 10% decrease in velocity due to CO2 substituting water, but at seismic frequencies (2-100 Hz) I observe a velocity decrease between 3% and 10%. The amount of change results from the frequency dependent velocity when basalt cores are saturated with water. Larger elastic changes due to fluid substitution are measured for the bulk modulus, with an average change of 30%. The water and CO2 saturated rock bulk modulus is modeled with two rock physics theories. For the three measured samples, Gassmann's theory predicts the measured data at frequencies lower than 20 Hz, but underpredicts ultrasonic modulus measurements. Kuster-Toksöz, an elastic theory developed for high frequencies, predicts the ultrasonic measured bulk modulus when rock analysis on pore distribution and shapes is incorporated into the computations. The differences in measured elastic properties with frequency and pressure are directly related to the amount of open cracks and compliant pores in these rocks.

To estimate the effects of carbonate precipitation on elastic properties, the basalt samples are placed in a reactor vessel at reservoir conditions to instigate mineralization. Wave velocity on dry basalt samples is measured with a laser ultrasonic system before reactions occur and at two reaction time intervals: 15 and 30 weeks. P-wave velocity estimated from direct arrivals increases on average by 6.7% for the first 15 week measurement and an average change of 7.8% from 15 to 30 weeks. When analyzing the data with coda wave interferometry, the basalt shear wave velocity increases on average by 1.8% from 0-15 weeks and 1.9% from 15-30 weeks. Rock microstructure and composition analysis before and after rock alterations support these velocity observations. Porosity and permeability decrease by 1.7% and 20%, respectively and the mineral precipitate is observed in a variety of pore shapes and sizes from 3D CT-scan images. I observe carbonate precipitation (possibly with iron and magnesium composition) from analyzing vesicles and crack walls in thin sections. From petrography, I interpret that the glassy groundmass is the source of dissolution, as crystals are not observed to be altered after 30 weeks of reaction.

Although basalt flows are difficult to image with seismic methods, previously modeled time-lapse changes from coda waves in layered basalt were able to resolve a 5% velocity change. Because the velocity data reported in this work for fluid substitution and mineral precipitation are equal or greater to 5%, I conclude that field-based seismic methods can potentially monitor fluid and rock changes in a basalt reservoir. This work also contributes information for developing the use of elastic waves to monitor rock alterations in CO2-water environments present in other geologic settings such as active volcanoes, mid-oceanic ridges, and geothermal reservoirs.

Download

Share

COinS