Laboratory Evaluation of the CO2 Huff-N-Puff Process for Heavy Oil Reservoirs

Abstract

Abstract Laboratory tests were carried out to evaluate a CO2 huff-n-puff well stimulation process in a Lloydminster heavy oil reservoir. Two series of phase behaviour measurements were done on the reconstituted oil, each in a pressurization-then-liberation cycle. The first series was performed using CO2 as the carbonating gas. Some oil swelling occurred, and a large viscosity reduction was observed. CO2 was retained preferentially to CH4 during liberation and, consequently, reduced viscosities were maintained down to low pressures. The second series was carried out using a 1:3 mole ratio CH4/CO2, simulating the use of recycled gas in the huff-n-puff process. The results revealed that CH4 reduces the efficiency of CO2. Core floods were performed and included evaluations of gas/water and oil/gas relative permeabilities at reservoir conditions, and an assessment of the longitudinal distribution of CO2 and the effect of soak period. Poor displacement efficiency was obtained by rapid CO2 injection, resulting in a low free gas saturation which was too small to provide the required amount of CO2 for effectively reducing the oil viscosity. The presence of a low mobile water saturation resulted in preferential displacement of water and enabled a somewhat more uniform longitudinal distribution of CO2 along the core. Introduction Primary production of oil in heavy oil reservoirs is usually low when compared to that of conventional oil reservoirs. Furthermore, heavy oil reservoirs often do not respond well to waterflooding. Accordingly, enhanced oil recovery methods have to be employed to boost production. Various techniques have been proposed in the literature such as thermal methods, polymer and caustic flooding, and gas injection. Gas injection in heavy oil reservoirs is often applied in the huff-n-puff mode. Carbon dioxide is frequently selected as the stimulating gas because of its high solubility in the oil and its oil viscosity reducing characteristics. The work reported in this paper deals with laboratory tests carried out to characterize this process and help evaluate its applicability to an Alberta heavy oil reservoir. A brief literature survey is included in the next section. This is followed by a description of the phase behaviour studies, and then the core displacement studies. Finally, over-all conclusions are made based on the results obtained. Literature Survey There is an abundance of literature available on CO2 flooding of light oil reservoirs, whereas only a few papers have been published on CO2 stimulation of heavy oil reservoirs. For example, Reid and Robinson(l) reviewed the performance of the Lick Creek Meakin Sand Unit field project. This project consisted of a first phase in which producers and injectors were cycled. The second phase involved continuous injection of CO2, and the third phase consisted of alternate water-CO2 injection. Finally, water was continuously injected. They concluded that the project was successful and that the CO2/ water injection process was viable for relatively thin heavy oil reservoirs. Stright et al.(2) described a single well test for a reservoir containing bottom water (Grand Forks Lower Mannville C Pool, Alberta).