INVESTIGATION OF EFFICIENCY IMPROVEMENTS DURING CO2 INJECTION IN HYDRAULICALLY AND NATURALLY FRACTURED RESERVOIRS

The objective of this project is to perform unique laboratory experiments with artificial fractured cores (AFCs) and X-ray CT to examine the physical mechanisms of bypassing in HFR and NFR that eventually result in less efficient CO{sub 2} flooding in heterogeneous or fracture-dominated reservoirs. This report provides results of the second semi-annual technical progress report that consists of three different topics. In the first topic, laboratory experiments were performed on a Berea core to investigate the changes in rock properties and fluid flow under different stress-state conditions. A comparative study of different stress conditions was also conducted to analyze the effect of the various loading systems. The experimental results show that fracture permeability reduces significantly as the stress increases compared to matrix permeability. The hydrostatic and triaxial stresses have greater impacts on permeability reduction compared to applying stress in the uniaxial stress condition. Fracture flow dominates when the applied stress is less, however, the matrix flow rate increases as applied stress increases and dominates at high stress even if the fracture does not heal completely. In the second topic, the preliminary results of static imbibition experiments are presented as a precursor to image the saturation profiles using X-Ray CT scanner. The static and dynamic imbibition experiments have been done previously (Schechter et al, 2002). The imaging of imbibition experiment is underway to track the saturation profiles using X-ray CT scanner. Hence, no more conclusions are drawn from this study at this time. In the last topic, the modeling of fluid flow through a single fracture incorporating the effect of surface roughness is conducted. Fracture permeability is usually estimated by a cubic law that is based on the theory of hydrodynamics for the laminar flow between flat plates. However, the cubic law is too simple to estimate the fracture permeability correctly, because the surface of real fracture is much more complicated and rougher than the surface of flat plate. Several researchers have shown that the flow characteristics of an actual fracture surface would be quite different due to the effect of tortuosity, impact of surface roughness and contact areas. Nonetheless, to date, these efforts have not converged to form a unified definition on the fracture aperture needed in the cubic law. In this study, therefore, we show that the cubic law could still be used to model small-scale and field-scale data as long as it is modeled effectively, accounting for the effect of surface roughness associated with the fracture surface. The goal of this research is to examine the effect of surface roughness for flow through fractures and to effectively incorporate them into simulations with the aid of geostatistics. Since the research has been supported with experimental results, the consistency of the results enabled us to define a methodology for single fracture simulation. This methodology successfully modeled the slow rate and pressure drop from fractured core experiments, which were earlier not possible through parallel plate approach. Observations suggest that the fracture aperture needs to be distributed to accurately model the experimental results. The effect of friction and tortuosity due to surface roughness needs to be taken into account while modeling