This Is AuburnElectronic Theses and Dissertations

Enhancing Understanding of Mineral Surface Area in Fabricated and Real Sandstone




Salek, Md Fahim

Type of Degree

PhD Dissertation


Civil and Environmental Engineering

Restriction Status


Restriction Type

Auburn University Users

Date Available



Carbon dioxide (CO2) capture and storage (CCS) is a well-developed technology to mitigate climate change associated with increasing CO2 due to anthropogenic activities. Injecting CO2 in subsurface porous rock formation initiates a complex set of reactions that might gradually change the properties of the rock formation. Reactive transport simulations provide a powerful medium to understand the implications of this process and optimize its efficiency. However, a discrepancy is observed in the simulated mineral reaction rates from the laboratory with field observations due to various factors. Much of this variation can be attributed to the imprecise estimation of mineral reactive surface area, with estimated values spanning orders of magnitude. Image obtained mineral accessible surface area has been observed to simulate laboratory core flood experiments with more accuracy in recent studies. However, how pore connectivity affects mineral accessibility, to what extent the X-ray imaging recipe affects the quantification of porosity and mineral surface area, and how this surface area evolves during geochemical reactions is not entirely known. Additionally, the feasibility of resin-based 3D printing technology to replicate porous media to conduct geochemical kinetic investigations has never been studied. This work aims to improve our understanding of mineral accessible surface area and its evolution during geochemical reactions in 3D printed and real sandstone samples. The feasibility of using 3D printed reactive porous media is first considered. A porous structure extracted from a real sandstone sample was printed using resin mixed with reactive calcite to replicate rock’s reactive property. In the printed samples, porosity and surface area agree well with the real sample. Moreover, calcite dissolution during the batch experiment conducted on the printed sample validated the accessibility of calcite. Furthermore, to understand the impact of pore connectivity on mineral accessibility, seven sandstone samples of varying composition were imaged under a scanning electron microscope (SEM) and properties like porosity, mineral abundance, and accessibility were quantified from mineral maps created by combining backscatter electron (BSE) images with energy dispersive spectroscopy (EDS) data. Observed variations in accessibility for quartz, feldspars, and carbonate due to consideration of nano pore connectivity is within one order of magnitude, however, larger variations were noticed for clay. While imaging approaches hold promise to quantify the mineral properties, it might be dependent on how we capture the image. To understand the impact of 3D X-ray computed tomography imaging parameters on petrophysical property quantification, two sandstone samples were imaged in 3D at different resolutions, detector bin sizes, and projection numbers. It is observed that the porosity measurement of the Bentheimer sample is independent of the imaging parameters within the studied range; however, the accessible surface area quantification is significantly impacted by these parameters. Moreover, the highest resolution studied here (1.25 µm) was insufficient to capture majority of the pores for the Torrey Buff sample, which contains a substantial amount of clay. The final study focused on the evolution of accessible surface area during geochemical reactions. For this purpose, a core flood dissolution experiment was conducted on a sandstone sample by injecting CO2-saturated deionized water at elevated temperature and pressure. Image-obtained properties were utilized to model the experiment in a reactive transport simulation tool, CrunchFlow. 2D and 3D images were captured before and after the experiment to quantify the evolution of surface area and were compared with the simulation results.