Abstract

Carbon dioxide (CO₂) flooding has come of age in the United States. CO₂ has been available in the past as a by-product of many industries, but it has not been seriously considered for wide scale application because of the availability of relatively inexpensive hydrocarbon gases. Miscible flooding with LPG and enriched gases has been practiced for some time; however, recently the cost of these materials has been prohibitive and has made them much less attractive for future flooding projects. As a result, CO₂ has now become a viable alternative for recovering additional oil without great economic cost.

At low reservoir presures, CO₂ flooding is usually an immiscible process resulting in improved oil recovery. Carbon dioxide dissolves in the oil, causes the oil to swell and lowers the viscosity of the oil. Under certain conditions of temperature and pressure, CO₂ may become first-contact miscible with the reservoir oil. This type of oil recovery process is very efficient, but the specific conditions required are seldom encountered. A more common situation is that of CO₂ multi-contact miscibility with the reservoir oil. In this case miscibility is established when, by means of successive contacts, CO₂ becomes sufficiently mixed with the intermediate components of the oil. This type of process can also be quite efficient in recovering oil.

Texaco U.S.A. has initiated two CO₂ miscible projects in South Louisiana. For the Bay St. Elaine CO₂ project, the injected CO₂ was combined with methane and normal butane (n-butane) to provide a solvent having the proper density and miscibility characteristics necessary to perform a gravity stable flood in this steeply dipping reservoir in a reasonable length of time. The design of the CO₂ solvent slug was the result of laboratory studies which included a Pressure-Volume-Temperature (PVT) study of the reservoir oil, slim tube testing of several possible CO₂ solvents, and long sand pack floods to verify the effectiveness of the selected co₂ solvent slug design. Additional design support was provided in the areas of pulse testing for improved reservoir definition and in residual oil saturation determination by means of pressure cores, log-injected log waterflood, and single well partitioning tracer tests.

The Paradis field project consists of two large scale tests in relatively low dip reservoirs. Laboratory studies were conducted to develop the PVT data for both of the reservoir oils. Slim tube tests were performed to investigate the range of carbon dioxide-nitrogen ratios in which multo-contact miscibility with these oils could be maintained. Several long sand pack floods were also conducted in order to investigate the effect on the recovery of residual oil of flooding rates higher than the critical velocity. Design support was again furnished to help plan and conduct the tests to define the continuity and the current residual oil saturations of each of the reservoirs.