Show simple item record

dc.contributor.advisorHolt, Rune
dc.contributor.advisorBauer, Andreas
dc.contributor.authorSzewczyk, Dawid
dc.date.accessioned2017-07-13T08:21:07Z
dc.date.available2017-07-13T08:21:07Z
dc.date.issued2017
dc.identifier.isbn978-82-326-2375-4
dc.identifier.issn1503-8181
dc.identifier.urihttp://hdl.handle.net/11250/2448628
dc.description.abstractShales are the most common constituent of the overburden of conventional hydrocarbon reservoirs. As such, they have a strong influence on seismic waves and knowledge about overburden mechanical properties is required for interpretation of seismic data. With the development of timelapse seismic, it has become possible to detect changes in overburden seismic travel times that can be linked to pore pressure changes in the reservoir (e.g. during petroleum production or sequestration of CO2). Shale characterisation is further required for addressing cap rock integrity and costly operation problems such as borehole instability during drilling operations. With the recent developments of oil and gas shales, shale data is also required as part of reservoir characterisation. Many groups have investigated static and ultrasonic elastic properties of shales. However, theoretical predictions and some experimental investigations have shown that shales have a dispersive nature. Although laboratory methods allowing to investigate rock properties at seismic, sonic and ultrasonic frequencies are available (e.g. forced oscillation techniques, resonant bar techniques, and pulse transmission techniques), seismic and sonic methods entail many practical issues, which strongly limits the number of publications on seismic-dispersion measurements. As a consequence, theoretical descriptions of seismic dispersion often remain unverified, and seismic dispersion is often neglected in rock physics models developed based on ultrasonic data and applied to the interpretation of seismic data. Therefore, systematic laboratory studies of seismic dispersion under well-defined conditions are of great value. This thesis focuses on measurements of seismic dispersion of partially saturated Mancos shale and Pierre shale I exposed to different stresses and stress paths. Experiments were performed on samples conditioned under different relative humidities, which allowed control of their saturation states. In addition, tests were also performed on as-received and oven-dried samples. The experiments were conducted in a custom-built triaxial compaction cell (developed partially as a part of this PhD), allowing for quasi-static deformation and ultrasonic measurements, as well as dynamic elastic-stiffness measurements at seismic frequencies under controlled deviatoric stresses and pore pressures. The experimental setup utilizes the forced oscillation technique to determine Young's modulus and Poisson's ratio at seismic frequencies (1 Hz – 155 Hz), pulse transmission method to determine P- and S-wave velocities at ultrasonic frequencies and static compaction measurements to determine quasistatic Young's modulus and Poisson's ratio. The importance of seismic dispersion for the interpretation of overburden seismic data was investigated by exposing differently saturated core plugs to an incremental isotropic loading up to 20 MPa, and to a constant mean-stress path loading that is assumed to be representative for overburden iiin- situ stress changes during injection (e.g. CO2 for storage or enhanced oil recovery) into a reservoir. Measurements of static and dynamic rock properties at different frequency ranges allowed us to check some of the available theories describing the relationship between static and dynamic rock moduli. The observed stress sensitivity of acoustic waves of the two studied shales was found to be frequency dependent, which may be attributed to a stress-induced change in dispersion, and in general was higher at seismic frequencies than at ultrasonic frequencies. This could lead to inappropriate stress change predictions if the rock physics models used for interpretation of seismic data would be based on the ultrasonic measurements and not account for dispersion effects. However, increasing water content was found to decrease the differences between stress sensitivities at seismic and ultrasonic regimes, indicating a need for investigations of saturation effects on dispersive properties of shales. At the same time, experiments performed along two different stress paths has revealed that the stress sensitivity of shales is affected not only by saturation and frequency but also by the applied stress path. The saturation sensitivity of dispersive properties of shales was further tested by loading differently saturated core plugs (saturation ranging between ̴ 0% to 100%) to the same stress. In order to preserve the initial saturation of the core plugs, techniques allowing for fast handling and subsequent sealing of samples (within about 1 hour) have been developed. We have observed qualitatively similar frequency and saturation dependencies of the dynamic rock stiffness of the two shales, which might be surprising considering the large differences in porosity and mineralogy. Changes of water saturation was found to have a large impact on dynamic stiffness, acoustic velocities, and seismic dispersion. Increased saturation caused strong softening of the samples at seismic frequencies and increased dispersion leading to a complex behaviour of the shale stiffness at ultrasonic frequencies. Poisson's ratio was found to strongly increase with increasing saturation. Acoustic velocities exhibit non-monotonous saturation dependencies that are different at seismic and ultrasonic frequencies, resulting in saturation dependent dispersion, which is high for partially saturated samples and lower for dry samples or highly saturated samples. We have interpreted the stiffness changes in terms of adsorption and capillary effects, whereas the dispersion was linked to local-flow type of mechanisms at grain contacts or micro-cracks involving adsorbed (bound) water. It was also shown that the experimental data can be fitted by the anisotropic Gassmann model with saturation dependent effective frame stiffness and fluid modulus. The obtained results show that shales exhibit complex stress, frequency and saturation dependencies that need to be further explored in order to fully understand the rock physics of shales, and presented here elastic properties of shales measured over a wide frequency spectrum can provide a better understanding of mechanisms underlying those dependencies.nb_NO
dc.language.isoengnb_NO
dc.publisherNTNUnb_NO
dc.relation.ispartofseriesDoctoral theses at NTNU;2017:151
dc.titleFrequency dependent elastic properties of shales – Impact of partial saturation and stress changesnb_NO
dc.typeDoctoral thesisnb_NO
dc.subject.nsiVDP::Technology: 500::Rock and petroleum disciplines: 510::Petroleum engineering: 512nb_NO


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record