Fracture aperture is usually estimated by cubic law, which assumes flow between two smooth parallel plates. However, many researchers have proved that the fracture aperture is not a smooth surface but rather has tortuous paths and roughness, and hence the flow behavior is different. Previous research showed that fracture aperture follows lognormal distribution. Nevertheless, there has not been any research conducted to validate the fracture aperture distribution with the change in stress conditions, which is common in fractured reservoirs. With the advent of X-ray CT scanner in the field of petroleum engineering, fracture apertures can be visualized and measured. Since there is no direct calculation for fracture aperture measurement from CT scanner data, a calibration curve needs to be established. We developed a calibration curve based on existing calibration techniques, which involves area integration of the fracture region to obtain a correlation between integrated CT numbers and the calibrated fracture aperture. Using this calibration curve, we obtained distribution patterns for fracture apertures along the length of the core for various stress conditions, from about six thousand fracture aperture measurements for each stress condition. The results show that aperture distributions still follow lognormal distribution under various stress conditions.

Fractured reservoirs have always been considered poor candidates for enhanced oil recovery. This is mainly due to the complexities involved in predicting performance in such reservoirs. A good understanding of multiphase flow in fractures is important to reduce oil bypass and increase recovery in these reservoirs. This paper presents CO 2 flooding experiments in homogeneous and fractured rocks with in-situ saturation and porosity measurements using an X-Ray CT scanner. We found that injection rates played an important role in the recovery process, more so in the presence of fractures. At high injection rates we observed faster CO 2 breakthrough and higher oil bypass than at low injection rates. But very low injection rates are not attractive from an economic point of view. Hence we injected viscosified water to reduce the mobility of CO 2 , similar to the WAG process. Breakthrough time reduced significantly and a much higher recovery was obtained. Saturation measurements were made from the CT scans and were found to be in good agreement with those obtained from effluent data.

Fractured reservoirs have always been considered as poor candidates for enhanced oil recovery. The fractures provide a pathway for injected fluids to channel through directly from injection to production wells. The interaction between these fractures and the reservoir rock matrix often determines the degree of bypassing during injection of CO2. The use of CO2 as a displacing agent through these reservoirs aggravates the problems of low sweep efficiency due to its high mobility. The microscopic displacement efficiency of CO2 is very high, but the overall displacement efficiency is often hindered by its high mobility that is largely the results of viscosity and density contrasts between the CO2 phase and the reservoir oil and brine phases. In this study, we performed CO2 injection experiments with different injection rates and utilized X-ray CT to determine the saturation distribution along the core and measure oil bypassed during CO2 process in fractured cores. We improved the CO2 sweep efficiency by controlling the CO2 mobility in the fracture. Water viscosified with a polymer was injected directly into the fracture, to divert CO2 flow into the matrix and delay breakthrough. Although the breakthrough time reduced considerably, water "leak off" into the matrix was very high. To counter this problem, a cross-linked gel was used in the fracture for conformance control. The gel was found to overcome "leak off" problems and effectively divert CO2 flow into the matrix. This experimental results increase the understanding of fluid flow and conformance control methods in fractured reservoirs. Introduction CO2 injection has been widely used for recovering oil from reservoirs due to its easy solubility in crude oil and its ability to "swell" the net volume of oil and thereby reduce oil viscosity by a vaporizing-gas-drive mechanism (Martin and Taber, 1992). The quantity of hydrocarbons that can be recovered from a reservoir is influenced by several characteristics of the reservoir including reservoir rock properties, reservoir pressure and temperature, physical and compositional properties of the fluid and structural relief, to name a few. However, the predominant factor in deciding the success of a CO2 flood is the reservoir heterogeneity. Highly heterogeneous reservoirs with variable lateral and vertical permeability characteristics can cause potential problems during CO2 injection. The injection gas tends to finger ahead into areas with high mobility ratios. This results in the gas forming preferential paths and "bypassing" large volumes of oil. Uleberg and Hoier (2002) suggest that the injection gas tends to flow in the highly permeable fractures, instead of the normally expected displacement path. These fractures are often responsible for early and excessive breakthrough of CO2, thus greatly affecting the economics of the project. In the recent years, there has been an increasing interest in the WAG process, both miscible and immiscible. The continuous CO2 injection process is an important process to identify displacement mechanisms but is not likely to be economic in practice unless significant recycling of gas is employed. Inherent in all gas injection processes is the lack of mobility and gravity control (areal and vertical sweep) necessary to sweep significant portions of the reservoir. Therefore, the replacement of high cost CO2 by a cheaper chase fluid such as water for horizontal displacements appears economically attractive. The WAG process involves alternate injections of small pore volumes (5% or less) of CO2 and water until the desired volume of CO2 has been injected. Since the microscopic displacement oil by gas normally is better than by water, the WAG injection combines the improved displacement efficiency of gas flooding with an improved macroscopic sweep by the injection of water. This has resulted in an improved recovery (compared to pure water injection) for most field cases.

A fracture is usually assumed as a set of smooth parallel plates separated by a constant width. However, the flow characteristics of an actual fracture surface would be quite different, affected by tortuosity and the impact of surface roughness. Though several researchers have discussed the effect of friction on flow, their efforts lack corroboration from experimental data and have not converged to form a unified methodology for studying flow on a rough fracture surface. In this study, we have shown an integrated methodology, involving experiments, stochastics and numerical simulations that incorporate the fracture roughness and the friction factor, to describe flow on a rough fracture surface. Laboratory experiments were performed to support the study and the flow contributions from the matrix and the fracture were matched through modified cubic law. Observations suggest that the fracture apertures need to be distributed to accurately model the experimental results. The methodology successfully modeled fractured core experiments, which were earlier not possible through parallel plate approach. A gravity drainage experiment using an X-ray CT scan of a fractured core has also validated the methodology. Introduction The search for hydrocarbons has been expanded into harder-to-evaluate formations, where potential and profitable hydrocarbon reserves are located. The prime candidate among those is the naturally fractured reservoirs, where large quantities of reserves are still left unexplored because of the complexities associated with a fractured reservoirs. Understanding the fluid flow characteristics of fractures is very important to model flow through fractures. This requires basic knowledge of flow from core studies. This research is aimed at studying fluid flow though a single fracture from simple core experiments and provide an effective methodology to simulate the flow behavior. Previous Research Efforts Early investigators based their idea that a parallel plate concept would be utilized to understand the concept of fluid flow through fractures. The first comprehensive work on flow through open fractures was by Lomize1. He used parallel glass plates and demonstrated the validity of the cubic law as long as the flow was laminar. He introduced the concept of defining the impact of surface roughness based on empirical data. Later he developed a flow regime chart that takes into account the effects of roughness and turbulent flow in open fractures. It was Snow 2 who used this concept to simulate real fractures. Iwai 1 conducted a comprehensive study of fluid flow through a single fracture and investigated the validity of the cubic law of fluid flow through a single natural fracture. One of the important features of his experiments was that the fracture planes had contact area as well as roughness.