SPE Members
The problem of predicting recovery and pressure behavior during gas/surfactant solution flooding is too complex to be treated analytically with the current state of knowledge. In the early stages of such displacements, the gas and surfactants are macroscopically dissociated and segregated, each flowing in separate channels. The in situ generation of foam does not seem to initiate until a certain mass of surfactant has been injected. Frontal advance theories such as the Buckley-Leverett theory or the Dietz model assume the injected fluid to be single phase, homogeneous, and non-reactive. Since these criteria are not met by the flow of gas and surfactant solution, these theories fail to accurately predict the recovery performance of such displacements.
Based on the observations made during a series of experiments on a scaled-model, a semianalytical model is proposed to predict the oil recovery and pressure history from two-dimensional (x,z) systems undergoing gas/surfactant injection. The model is based on the concept of a combination flood, in which gas/foam and surfactant solution flow in separate gravity tongues. The gas/foam displacement is assumed to follow the Buckley Leverett theory, whereas the surfactant displacement follows the Dietz model. The displacements through these tongues are mutually dependent and, therefore, these theories are combined at each stage. The predictions from the model gave a good match with the experimental predictions from the model gave a good match with the experimental results for a fairly broad range of operating conditions. The semi-analytical approach adopted in developing the combination drive model has potential for application to certain other types of frontal behaviors.
The efficiency of a gas or steam injection process used to enhance oil recovery is lowered by gravity segregation and fingering. The shape and movement of the displacement front in such a process is governed by the mobility and density differences between the injected fluids and the reservoir fluids. The gravity segregation and fingering are thus caused by the lower density and much higher mobility of gases or steam compared to the reservoir oils.
In recent years considerable interest has been focused on developing some means of reducing the mobility of steam and gases to reduce their rate of channeling. The injection of foaming surfactants along with steam or gas is one of the methods studied, for it is well known that foams may have apparent viscosities as high as 100 cp. Laboratory studies have shown that it is possible to find foaming agents which are chemically stable at high temperatures (Al-Khafajil et al.), * and field tests using foaming agents have significantly improved the oil producing rates and the steam/oil ratios (Brigham et al., Dilgren et al and Ploeg and Duerksen).
Although these laboratory and field results are encouraging, it is clear that these tests have not been optimized. Nor could they be expected to be, for the basic flow mechanism of steam foam flow systems is not known.
Much of the petroleum engineering research on foam rheology and the mechanism of foam flow through porous media has been qualitative in nature, and no predictive methods have been developed to estimate the production performance of foam flooding. The general expectation of higher recoveries whenever a surfactant is added to a water-gas mixture drive, has not always proved true.
The work reported here is a first step in an attempt to determine the mechanism of foam flow in a porous medium in the presence of oil and water. The system studied is a low pressure presence of oil and water. The system studied is a low pressure low temperature displacement using nitrogen and surfactant solution, thus it does not include the effects of heat nor the vapor collapse which occurs with steam, but it does include the relative effects of the gravity and Darcy forces and the resulting oil recovery.
Based on the visual observations during these runs, a semianalytical model was formulated to predict the recovery and the pressure response during gas/surfactant injection under foaming pressure response during gas/surfactant injection under foaming conditions. The model presented is based on the combination-drive displacement observed in a large number of experimental runs. In this type of displacements, the injected gas and the liquid are mostly segregated and flow in separate gravity tongues. The displacement through the gas tongue is assumed to be governed by Buckley-Leverett theory and the displacement through the liquid tongue by the Dietz model. However, the displacements through these tongues are mutually dependent and the Buckley-Leverett theory and Dietz theory cannot be applied independently. The model couples these theories in a way that predicts the recovery performance under a wide range of operating conditions for the two dimensional system studied. An empirical relationship was also developed to predict the surfactant throughput required to initiate foam generation.
P. 325
