The Gas-Assisted Gravity Drainage (GAGD) process was developed to take advantage of the natural segregation of injected gas from crude oil in the reservoir. It consists of placing a horizontal producer near the bottom of the reservoir and injecting gas using existing vertical wells. As the injected gas rises to the top to form a gas cap, oil and water drain down to the horizontal producer. Earlier experimental work using a physical model by Sharma had demonstrated the effectiveness of the GAGD process in improving the oil recovery when applied in water-wet porous media. The current research is an extension of that work and is focused on evaluating the effect of the wettability of the porous medium and the presence of a vertical fracture on the GAGD performance. The effect of the injection strategy (secondary and tertiary mode) on the oil recovery was also evaluated in the experiments. In the physical model experiments a Hele-Shaw type model was used (dimensions: 13 7/8 by 5/16 by 1) along with glass beads and silica sand as the porous media. Silanization with an organosilane (dimethydichlorosilane) was used to alter the wettability of the glass beads from water-wet to oil-wet. The experiments showed a significant improvement of the oil recovery in the oil-wet experiments versus the water-wet runs, both in the secondary and the tertiary modes. The fracture simulation experiments have also shown an increase in the effectiveness of the GAGD process.

The Gas Assisted Gravity Drainage (GAGD) process, currently being developed at LSU, is designed to take advantage of gravity to allow vertical segregation between the injected gas and reservoir crud oil due to their density differences. GAGD is recommended for use with CO2 gas. CO2 dissolves in oil and causes both swelling and viscosity reduction of oil. The GAGD process uses the existing vertical wells for CO2 gas injection, and a horizontal well near the bottom of the payzone for oil production. GAGD, as an EOR process, is not restricted to tertiary oil recovery only. In this research study, a visual glass model has been used to visually discern the mechanisms operative in the GAGD process. The model was also designed to fit different vertical well configurations. The model experiments have proven that GAGD is a viable process for secondary and tertiary oil recovery. Oil recovery in the immiscible secondary mode was as high as 83% IOIP and the oil recovery in the immiscible tertiary mode was 54% ROIP. The model has also shown that the gas injection depth may not have an influence on oil recovery as long as there is vertical communication between reservoir layers. Four different injection depths resulted in oil recovery values between 71% IOIP and 76% IOIP. The visual model experiments have also demonstrated that GAGD is applicable to naturally fractured reservoirs. The oil recovery in the fractured porous media was as high as 76% IOIP, which was higher than the average in homogenous porous media (73% IOIP). Additionally, the GAGD process was found to be viable for higher viscosity oils as well, where secondary immiscible oil recovery was 64% IOIP. Miscible secondary injection was performed by using naphtha as the oil phase and decane as the miscible gas phase to simulate the miscible GAGD process. The visual model has resulted in a microscopic sweep efficiency close to 100% in the miscible GAGD process. The visual model experiments have demonstrated three possible mechanisms responsible for high oil recoveries: Darcy-type displacement until gas breakthrough, gravity drainage after breakthrough, and film drainage in the gas invaded regions.

Minimum miscibility pressure (MMP) is an important optimization parameter for an enhanced oil recovery process involving Carbon Dioxide or hydrocarbon gas injection. Therefore an accurate experimental measurement is required to determine the MMP. The MMP for a gas-oil system is directly related to the interfacial tension between the injected gas and the reservoir crude oil. When CO2 gas contacts the reservoir oil at reservoir temperature, the interfacial tension between the fluid-fluid phases reduces as the miscibility is approached and the interface between the fluid-fluid phases eventually disappears at miscibility i.e. the interfacial tension becomes zero. Hence, a pressure condition of zero interfacial tension at reservoir temperature is the minimum miscibility pressure for a CO2-reservoir crude oil system. The Vanishing Interfacial Technique (VIT) technique to determine MMP is based on this principle. Therefore, this research project involves the measurement of gas-oil interfacial tensions for a CO2-live reservoir oil system at reservoir conditions using the pendant drop and the capillary rise techniques to determine the minimum miscibility pressure through the VIT technique. Gas-oil interfacial tension, being a property of the interface between crude oil and gas, is strongly affected by the compositional changes induced by the counter-directional mass transfer (vaporizing, condensing or a combination of the two) of the various components taking place between the CO2 and reservoir oil. This study hence investigates the mass transfer mechanisms involved in these dynamic gas-oil interactions responsible for miscibility development by performing detailed compositional analyses, and density measurements. All the measurements were carried out at different ratios of fluid phases in the feed mixture (both molar and volumetric) for various pressures at the reservoir temperature in order to also study the effects of the initial feed composition on IFT and the phase compositions. This study has experimentally demonstrated that the gas-oil interfacial tension measured at varying feed compositions (i.e., initial gas-oil molar and gas-oil volume ratios) at reservoir temperature, although showing different relationships with pressure, converged to the same endpoint of zero-interfacial tension or similar minimum miscibility pressures. The effect of gas-oil molar ratios and gas- oil volume ratios on the compositions of the equilibrium phases for this CO2-reservoir fluid system proved that the mechanism involved in the mass transfer of hydrocarbon components between the fluid-fluid phases was a condensing gas drive mechanism. This study has demonstrated that the MMP determined from the VIT technique is independent of the compositional path followed by the fluids during their continuous interaction prior to attaining mass transfer equilibrium.

Managed pressure drilling is an innovative technique to precisely manage wellbore pressure. It is particularly applicable for reducing the risk of a kick or lost returns when drilling with a narrow window between pore pressure and fracture pressure. The constant bottomhole pressure method of managed pressure drilling uses annular frictional pressure and choke pressure in addition to mud hydrostatic pressure to achieve precise wellbore pressure control. This project investigated alternative initial responses to kicks to determine which would be most effective and reliable under different well scenarios when applying the constant bottomhole pressure method of managed pressure drilling. Three different initial responses to a kick, 'shut-in the well,' 'apply back pressure' and 'increase mud pump rate' were studied using an interactive transient multiphase flow simulator. The kick scenarios were varied by changing the hole size, type of kick fluid, initial kick volume, pressure differential at the kick zone, and fracture injectivity index. No single best response was identified for the kick scenarios that were studied. Nevertheless, some conclusions were reached. The validity of these conclusions may be limited to the range of scenarios studied. 'Increasing mud pump rate' is advantageous when it increases bottomhole pressure enough to stop formation flow because it results in the minimum casing and shoe pressures. Therefore, it should minimize the risk of lost returns or surface equipment failure. However, it is unlikely to be successful in large hole sizes. The 'apply back pressure' response has a similar but smaller advantage versus the 'shut-in' option because circulation creates friction in the annulus. However, in cases where lost returns occurred, no reliable way of identifying the loss of returns and avoiding unintentional formation flow to the surface was defined. The 'shut-in' reaction generally results in the highest casing and casing shoe pressures. Therefore, it may be most likely to cause loss of returns before stopping formation flow and consequently causing an underground transfer with continuous influx. Nevertheless, it is probably the least likely to unintentionally allow formation fluid flow to the surface or to cause loss of significant mud volume downhole.

Mechanistic modeling of an underbalanced drilling operation using carbon dioxide has been developed in this research. The use of carbon dioxide in an underbalanced drilling operation eliminates some of the operational difficulties that arises with gaseous drilling fluids, such as generating enough torque to run a downhole motor. The unique properties of CO₂, both inside the drill pipe and in the annulus are shown in terms of optimizing the drilling operation by achieving a low bottomhole pressure window. Typically CO₂ becomes supercritical inside the drill pipe at this high density; it will generate enough torque to run a downhole motor. As the fluid exits the drill bit it will vaporize and become a gas, hence achieving the required low density that may be required for underbalanced drilling. The latest CO₂ equation of state to calculate the required thermodynamic fluid properties is used. In addition, a heat transfer model taking into account varying properties of both pressure and temperature has been developed. A marching algorithm procedure is developed to calculate the circulating fluid pressure and temperature, taking into account the varying parameters. Both single phase CO₂ and a mixture of CO₂ and water have been studied to show the effect of produced water on corrosion rates. The model also is capable of handling different drill pipe and annular geometries.

This study addresses oil bypassing caused by water invasion to wells in edge and bottom water-drive oil reservoirs a significant problem worldwide. It is shown that the amount of by-passed (not recovered) oil is significant and could be predicted analytically and reduced by modifying wells completion. A large statistical sample from the population of possible reservoir-well systems with edge and bottom-water has been created theoretically using several databases of actual reservoirs properties worldwide. Dimensional analysis allowed converting reservoir properties distributions into dimensionless group distributions. Then, the amount of by-passed oil was correlated with the dimensionless groups using designed experiments conducted on a reservoir simulator. The resulting correlations determine the percent amount of movable oil that could be recovered by the end of wells operation, when the water cut value reaches its maximum limit. They also show how operational parameters such as well spacing, penetration and production rate may affect oil recovery. From the sensitivity analysis, the end-point mobility ratio appears to control more than 55 percent of the oil bypassing process far more than other groups. The statistical results also show that half of the typical edge and bottom-water well-reservoir systems would have at least 17% or 25% of their movable oil bypassed, respectively. The effect of reservoir heterogeneity defined by permeability stratification has been studied for edge-water systems having transgressive, regressive and serrated depositional sequences with a Dykstra-Parsons coefficient of 0.75. Oil bypassing showed to be qualitatively more significant in the transgressive sequences. It was also found that the effect of reservoir heterogeneity is more significant for reservoirs with high end-point mobility ratios. Numerical reservoir simulation is also used to investigate improved recovery of wells completions of different penetration and dual-completed wells with segregated inflow from the top and bottom (water sink) completions. It appears that short completions perform best in reservoirs with large end-point mobility ratios produced at low rates by delaying water breakthrough and improving the amount of oil recovered per barrel of fluid produced. For most reservoirs with water drive, however, the results show that the best single completion strategy is the use of fully penetrating wells, since they improve the recovery rate. Dual well completions with water sink (DWS) enable independent (although synchronized) rate schedules at the two completions. This study offers a new method to operate DWS systems by using variable rates at the bottom completion for a constant production rate - with limited maximum water cut - at the top completion over the entire life of the well. The method provides better distribution of produced fluids, as it controls water saturation outside the well. When compared with conventional short completion, DWS well recovers oil faster and may also produce oil-free water for re-injection. However, a comparison with long single completion of similar length based on the same total fluid rate does not give different recovery rates but shows that DWS well operates at different pressure drawdowns and produces two streams of fluids having substantially different compositions. It is, then, concluded that the recovery performance of the two types of wells may be different if the basis for comparison is a maximum pressure drawdown rather than same total fluid rate.

Hydrocarbon resources from unconventional shale gas reservoirs are becoming very important in the United States in recent years. Understanding the effects of adsorption on production data analysis will increase the effectiveness of reservoir management in these challenging environments. The use of an adjusted system compressibility proposed by Bumb and McKee (1988) is critical in this process. It allows for dimensional and dimensionless type curves to be corrected at a reasonably fundamental level, and it breaks the effects of adsorption into something that is relatively simple to understand. This coupled with a new form of material balance time that was originally put forth by Palacio and Blasingame (1993), allows the effects of adsorption to be handled in production data analysis. The first step in this process was to show the effects of adsorption on various systems: single porosity, dual porosity, hydraulically fractured, and dual porosity with a hydraulic fracture. These systems were first viewed as constant terminal rate systems then as constant terminal pressure systems. Constant pressure systems require a correction to be made to material balance time in order to apply the correction for adsorption in the form of an adjusted total system compressibility. Next, various analysis methods were examined to test their robustness in analyzing systems that contain adsorbed gas. Continuously, Gas Production Analysis (GPA) (Cox, et al. 2002) showed itself to be more accurate and more insightful. In combination with the techniques put forth in this work, it was used to analyze two field cases provided by Devon Energy Corporation from the Barnett Shale. The effects of adsorption are reasonably consistent across the reservoir systems examined in this work. It was confirmed that adsorption can be managed and accounted for using the method put forth in this work. Also, GPA appears to be the best and most insightful analysis method tested in this work.

Gas Injection is the second largest EOR process in U.S. To increase the extent of the reservoir contacted by displacing fluids, gas and water are injected intermittently - water-alternating-gas (WAG) process, is widely practiced. This experimental study is aimed at evaluating the WAG process performance in short and long cores as a function of gas-oil miscibility and brine composition. This performance evaluation has been carried out by comparing oil recoveries between WAG injection and continuous gas injection (CGI). Miscible (2500 psi) and immiscible (500 psi) floods were conducted using Berea cores, n-Decane and two different brines, namely the commonly used 5% NaCl solution and another the multicomponent brine from the West Texas Yates reservoir. Each of the ten corefloods consisted of series of steps including brine saturation, absolute permeability determination, flooding with oil (drainage) to initial oil saturation, flooding with brine (imbibition) to residual oil saturation, and finally, tertiary gas injection to recover the waterflood residual oil. It was found that comparing tertiary gas floods only on the basis of recovery yielded misleading conclusions. However, when oil recovery per unit volume of gas injection was used as a parameter to evaluate the floods, miscible gas floods were found more effective (recovery 60-70% higher) than immiscible floods. The WAG mode of injection out-performed the CGI floods. At increased gas volume injection, the performance of miscible CGI flood, inspite of the high injection pressure, approached the immiscible floods. A change in brine composition from 5% NaCl to 9.26% multivalent Yates reservoir brine showed a slight adverse effect on tertiary gas flood recovery due to increased solubility of CO2 in the latter. While immiscible WAG floods in short cores donot show appreciable improvement over CGI immiscible floods, WAG recovery was 31% higher than 6-ft CGI floods. The results of this study prompted a new process by combining CGI and WAG modes of gas injection. Such a process was found patented and practiced in the industry. In addition to providing performance characteristics of the WAG process, this study has indicated directions for further research aimed at improving oil recovery from gas injection processes.

Although foam has been widely used in many scientific and engineering applications, the current understanding of foam rheology in pipes is still very limited because of its complex nature. This experimental study, for the first time, investigates the flow rheology of foams in pipes by placing a special emphasis on two distinct foam flow regimes. A wide range of experimental conditions are examined in this study, which include five different surfactant formulations (Cedepal FA-406, Petrostep CG-50, Stepanform 1050, Aquet TD-600, and Ultra-Palmolive), three different surfactant concentrations (0.1, 1, and 5 wt %), two different pipe diameters (0.5 and 1 inch nominal size stainless steel pipes), and two different filter opening sizes (50 and 90 micrometers) for upstream foam generation. The experiments revealed the following characteristics: (1) foam flow in pipes exhibited two different flow regimes called high-quality regime and low-quality regime, (2) the high-quality regime was characterized by unstable and oscillating pressure response which resulted from repeating free gas and foam slug, whereas the low-quality regime was characterized by stabilized pressure response which resulted from the flow of uniform and homogeneous foams, (3) different patterns of pressure contours were observed - the pressure contours were relatively steep in the high-quality regime but relatively gentle, or even almost horizontal, in the low-quality regime, (4) foam rheology in the high-quality regime was shear thickening to liquid injection velocity in all cases, and foam rheology in the low-quality regime was not consistent, and (5) the value of foam quality (fg*) that splits the two flow regimes was shown to be affected by experimental conditions such as surfactant formulations and concentrations. These observations imply that the rheology in the high-quality regime is primarily governed by dynamic mechanisms of lamella creation and coalescence during the flow, and the rheology in the low-quality regime is primarily governed by interactions between bubbles and/or interactions between bubbles and pipe wall. Therefore, the high-quality regime is likely to expand (or, the low-quality regime is likely to contract, equivalently) with a reduction in surfactant foamability. Implications of distinct foam behaviors in two flow regimes in practical applications are also discussed.

CO2 displacement is the most widely used EOR process, but poor sweep efficiency and large CO2 utilization rates are limitations to the economic and technical success of CO2 floods. Developing a methodology to maximize the sweep efficiency and minimize the CO2 utilization rate would greatly improve the economics of these fields. This thesis evaluates the sweep efficiency of a successful, late-in-life, continuous injection CO2 flood at the Little Creek Field, Mississippi. In this work, we evaluate several heterogeneity measures in terms of recovery efficiency and utilization rate. Core studies available from 41% of the wells in the field were used to compute various heterogeneity measures, and the resulting values were correlated with pattern-by-pattern recoveries and CO2 utilization rates. Weak correlation trends were found for most of the measures in terms of R2 values. However, there was still a trend confirming the idea that more heterogeneity corresponds to higher utilization rates and lower recoveries. Mapping of the well-by-well heterogeneity measures appear to show geological trends better than traditional maps of the basic parameters that make up the measures. These geological trends were then successfully used to adjust rock-types during reservoir modeling. Reservoir simulation was performed to understand the reservoir response to CO2 flooding and develop alternatives for sweep improvement. Continuous CO2 injection under certain alternate operations would help. The WAG process was effective in increasing the sweep efficiency of the reservoir for most of the cases studied by providing favorable mobility ratios and contacting more of the oil in the reservoir. The Gas-Assisted Gravity Drainage (GAGD) process was also evaluated. Solvent saturation profiles show that results are essentially consistent with the proposed GAGD theory. However, oil recovery was less than the best WAG cases, which is not surprising due to the high connate water saturation (0.56), relatively low thickness and lack of dip to the reservoir. Moreover, an increase in recovery could be realized more in the future for both the WAG and GAGD processes because CO2 contacted larger amounts of unswept oil in the reservoir compared to continuous CO2 flooding.