Linking microstructural, transport, and elastic property evolution in low porosity rocks upon thermal shocking

Abstract

Geophysical observations -- those made of a rock's bulk physical properties -- are in many ways dictated by a rock's underpinning microstructure, which includes the composition and arrangement of mineral grains and pores. For porous media, relationships between bulk properties (e.g., permeability or compressional and shear wave velocities) and microstructural properties (e.g., porosity) are relatively well established. For example, generally, as porosity increases, compressional (P-) and shear (S-) wave velocities are expected to decrease, while permeability increases. However, these relatively straightforward relationships are often difficult to establish for low-porosity, low-permeability "tight rocks" where porosity primarily exists as low-pore-volume, pressure-sensitive cracks among tightly mated mineral crystals. For instance, should a crack-containing rock experience external pressures, cracks may close, causing wave velocities to substantially increase, all without much change in porosity. Therefore, a feature of a rock's microstructure, other than its amount of porosity, may be controlling bulk elastic and transport properties. Moreover, a lithology's microstructure is dynamic and can be altered through time. One example is when a rock is exposed to a sudden temperature change, it may undergo thermal cracking, which can alter permeability and P- and S-wave velocities. Therefore, in order to properly link microstructural controls to changes in bulk rock properties, a time-lapse approach is needed for making observations of both the microstructure and measured bulk properties. In my thesis, I address the challenge of linking dynamic, microstructural features to changes in bulk elastic properties. In doing so, I developed a set of experiments using three different tight lithologies, all of which have similar and low values of porosity but distinctly different microstructures. In my experiments, I introduced cracks into my samples through a process of thermal shocking (slow heating to a target temperature followed by fast cooling in ambient temperature water). After thermal shocking, I monitored changes in each lithology's microstructure, porosity, permeability, and P- and S-wave velocities. My approach of incorporating time-lapse imaging offers a key improvement over other thermal shocking experiments in the literature. In doing so, I provide observations throughout a dynamic process in which the microstructure -- through thermal cracking -- alters the bulk geophysical properties. My results show that mineral composition and associated thermal properties influence the extent of thermal crack initiation. In turn, extensive thermal cracking increases permeability in lithologies, like granodiorite, which is composed of minerals with heterogeneous values of thermal expansion coefficients and thermal conductivities. Minimal thermal cracking does not enhance permeability in lithologies, like trachybasalt, composed of minerals with more uniform and low thermal expansion and conductivity values. Surprisingly, even a monomineralic lithology, like carbonate, will experience thermal cracking if it is composed of thermally anisotropic minerals, like calcite. My results also show that the extent of thermal cracking is directly related to whether permeability enhances and P- and S-wave velocities decrease -- due to reductions in sample stiffness -- with thermal shocking. Based on these findings, I expanded upon the experimental protocol to determine what the role of the number of thermal shocks is on transport and elastic properties. In other words, can the number of thermal shock cycles needed to enhance permeability be optimized according to lithology? This question encompasses both 1) how microstructure will impact the evolution of a thermal crack network and evolution of bulk transport and elastic properties and 2) is useful for applications such as enhanced geothermal systems, where sustainable stimulation practices require using minimal resources to achieve increased permeability. My approach was to repeat the thermal shocking protocol on the same samples to induce "cyclic thermal shocking". This protocol provides a key improvement over other cyclic thermal shocking experiments in that I again coupled my time-lapse rock physics measurements with time-lapse microstructural imaging. Additionally, my approach, unlike others, assessed how the stress-induced permeability and P- and S-wave velocities were impacted by thermal crack propagation. My results show that porosity remains relatively constant with cyclic thermal cracking, regardless of lithology. Additionally, a lithology's microstructure continues to control thermal crack propagation that in turn facilitates continued permeability enhancement to varying degrees. The extent to which permeability is enhanced depends on the thermal crack propagation mechanism, which is dictated by the lithology's microstructure. In particular, microstructures like irregular vuggy pores in carbonate will facilitate thermal crack widening, which corresponds to greater increases in permeability compared to lithologies, like granodiorite, which experience primarily crack lengthening along mineral grain boundaries. My results also show that continued thermal shocking reduces P- and S-wave velocities to lesser extents as the number of cycles increases. In summation, this thesis provides experimental observations for linking microstructural controls on the transport and elastic properties of tight rocks throughout thermal shocking and various stages of thermal crack propagation.

Copyright

© 2024 by Margariete GeorgeAlan Malenda

License

This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).

Each version has a distinct URL, but you can use this PURL to access the latest version.
https://purl.stanford.edu/rg468xb7144