<p> Understanding the elastic properties of unconsolidated granular media is crucial for interpreting seismic and sonic log data in soils and unconsolidated petroleum reservoirs. Rock and soil deformations are often estimated indirectly using rock physics models that relate changes in elastic properties to pore compliance. The complex microstructure of geological materials in represented by simple geometries in most rock physics models. One such model, the Hertz-Mindlin model, uses a pack of identical spheres to calculate elastic properties of unconsolidated sediments. The input parameters required for this model, porosity, grain radius, coordination number (number of contact points per grain) and grain to grain contact radius, are often unknown parameters and adjusted to fit the data. Direct observations of deformation can show the limitations in applicability of rock physics models. This requires 3D images obtained under in-situ pressure and temperature conditions. </p><p> I imaged changes in dry, unconsolidated quartz sand with micro X-ray computed tomography (microCT) together with ultrasonic P-wave velocities at pressure from atmospheric pressure (0.08 MPa) to 27.6 MPa. In addition to an overall compaction of the sediment leading to a 30% reduction in porosity; the microCT images show a 60% reduction in grain size due to grain crushing, a 26% increase in coordination number, and 50% to 100% increase in contact radius. </p><p> I used the image-derived porosity, grain radius, coordination number and contact radius as input data for the Hertz-Mindlin contact-radius model to compute P-wave velocities as functions of pressure. The microCT images show that numerous assumptions of the Hertz-Mindlin model are violated in sands and consequently, the model drastically overpredicts velocities. Although the velocity mismatch can be eliminated for undamaged sediments by assigning a reduced shear modulus to the contact zones, this adjusted model still overpredicts velocities of the sediment once grain crushing occurs. Thus, the Hertzian contact model should be applied with caution to angular, unconsolidated sediments. </p><p> Understanding gas hydrate morphology and the relationship between hydrate saturation and elastic properties is crucial to characterize natural occurring hydrate resources and assess their potential for production. Gas hydrates in unconsolidated sediment are often represented by effective medium models of the sediment frame and hydrate inclusions in the pore space with different morphologies which allow us to estimate gas hydrate saturation from sonic log or seismic velocities. Most effective medium models assume microstructural parameters to predict acoustic velocities. Without the constraints of direct observation, for example, pressure- and temperature-dependent variations of the sediment frame or packing rearrangements during hydrate formation, such predictions lead to discrepancies in hydrate saturation calculated from velocities. </p><p> My results verify hydrate pore-scale distributions by direct, visual observations which were previously implied by indirect, elastic property measurements. I used laboratory measurements on THF-hydrate bearing glass beads as proxy for naturally occurring gas hydrate in unconsolidated, coarse-grained sediment. Both, microCT images and ultrasonic velocity measurements, indicate that THF hydrate forms in the pore space with a part of the hydrate bridging the grains and becoming load-bearing at higher hydrate saturations. These hydrate-bearing sediments appear to follow a pore-filling model with a portion of the hydrate becoming a load-bearing part of the sediment frame at higher hydrate saturations. </p><p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10788570 |
| Date | 05 June 2018 |
| Creators | Schindler, Mandy |
| Publisher | Colorado School of Mines |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0016 seconds