Download files
Abstract
X-ray microtomography (micro-CT) provides a nondestructive way for estimating rock properties such as relative permeability. Relative permeability is computed on the fluid distributions generated on three dimensional images of the pore structure of a rock. However, it is difficult to numerically reproduce actual fluid distributions at the pore scale, particularly for a mixed-wet rock. Recent advances in imaging technologies have made it possible to directly resolve, thus capturing a large field of view for arbitrary wetting conditions. This thesis evaluates the relative permeability computations on imaged fluid distributions under water-wet and mixed-wet conditions.
Two-phase steady-state experiments are run to generate oil-water fluid distributions in a homogeneous outcrop Bentheimer sandstone core. The core is fully saturated with 0.4 M Sodium Iodide (NaI) solution. The saturated core is mounted in a specially designed flow cell which allows the flow experiment to be carried with the core mounted on the micro-CT facility. Then oil (Soltrol-130) and water (0.4 M NaI) phases are simultaneously injected into the core at various oil/water injection ratios. For each injection ratio, steady state pressure drop is noted and fluid distributions are imaged when equilibrium is reached. These imaged fluid distributions are used to compute image based relative permeability. While the measured pressure drops are used to calculate experimental relative permeability. Then a standard literature approach is used to change the core wettability to mixed-wet state. By repeating the water-wet core experimental procedure, image-based computations laboratory measurements of relative permeability are made.
For water-wet system, measured and computed relative permeability are in close agreement over the whole saturation range. This is a significant advancement compared to previous studies which underestimated oil relative permeabilities in a water-wet system due to snap-off. Saturation profiles generated from micro-CT images show capillary-end effect at low water saturations in water-wet. For mixed-wet case, capillary-end effect is greater at larger water saturations.
The agreement between computed and measured relative permeability for mixed-wet core is weaker than that of water-wet core. Analysis of the data shows that true steady-state conditions were not met during the mixed-wet conditions. Inability to match the experimental conditions in computations resulted in mismatch. This thesis shows that under water-wet conditions, relative permeability computations on imaged fluid distributions can provide reliable results. However, more experiments need to be performed under mixed-wet conditions to understand the mismatch between computations and measurements.