Abstract

Summary Two CO2 huff ‘n’ puff projects were conducted in the 4,900-ft [1495-m] Reservoir (BA) Sand Unit [4900' R(BA)SUI, Timbalier Bay field, Louisiana. This reservoir is a bottomwater-drive reservoir with a 26 degrees API [0.9-g/CM3] Oil gravity and 18% primary oil recovery. Before CO2 injection, both project wells were gas lifting more than 1,000 BFPD [160 M3/d fluid] with 99% water cuts. After CO2 injection, the production from each well increased to 200 BOPD [32 m3/d oil]. This paper discusses the CO2 huff ‘n’ puff process, specific reservoir characteristics, and project evaluation. Introduction When properly administered, the CO2 huff ‘n’ puff process provides quick payout with a low capital investment. These are important factors when oil prices are difficult to forecast. Timbalier Bay field was chosen to test the injection process because existing CO2 pipeline and facilities I would reduce the investment. The 4900' R(BA)SU was chosen for the huff ‘n’ puff because of its high residual oil saturation and relatively low API oil gravity. Process Discussion Limited references are available on the CO2 huff ‘n’ puff process despite voluminous publications on CO2 miscible and immiscible recovery. Khatib et al. summarized the evolution of immiscible CO2 injection in light and heavy crudes. Monger and Coma's research work concentrated on applying the CO2 huff ‘n’ puff process on light oils. From their laboratory work on cores at waterflood residual oil saturations and their data base of actual field test results, Monger and Coma concluded that residual oil can be displaced by the cyclic CO2 injection process. Patton et al. and Haskin and Alston have developed the only two correlations for estimating production responses from the CO2 huff ‘n’ puff process. Patton et al. defined two efficiencies to use in evaluating the success of a huff ‘n’ puff project. One efficiency is defined as the ratio of incremental oil produced to CO2 injected. This efficiency should range from 0.5 to 0. 8 STB/Mcf [2.8 × 10(-3) to 4.5 × 10(-3) stock-tank M3/M3], with a value of 1 STB/ Mcf [5.6 × 10(-3) stock-tank M3/M3] representing ideal conditions. The second efficiency is defined as the CO2 injection volume per foot of sand. This efficiency should range from 0.1 to 0.2 MMcf/ft [9.3 × 10(3) to 18.6 × 10(3) M3 /M]. Stright et al. Concluded that the cyclic process "does not appear to be a feasible recovery scheme" in a bottomwater-drive reservoir. When huff ‘n’ puff was applied, their subject reservoir was producing at much lower WOR's than the 4900' R(BA)SU project (2.19 vs. 70). The Timbalier Bay project is notable in that the incremental oil recoveries are greater than those predicted by the Patton et al. method and are recoverable from a bottomwater-drive reservoir with a low sweep efficiency. Throughout the literature is general agreement that the two principal recovery mechanisms in the immiscible CO2-injection process are oil swelling and viscosity reduction. The fractional-flow curves for the 4900' R(BA)SU illustrate how the viscosity-reduction and oil-swelling mechanisms cause production of incremental oil (Fig. 1). The procedure involved in generating these curves is as follows.The water/oil relative permeability is estimated from Stone's correlation.The Buckley-Leverett fractional flow of CO2-free oil is calculated with oil viscosity at reservoir temperature and pressure (Curve 1).The viscosity of CO2-saturated oil is estimated with Emanuel's equation.The new fractional-flow curve is calculated (Curve 2).The fractional-flow curves are overlaid. The viscosity reduction shifts the fractional-flow curve to the right, resulting in a reduced fractional flow of water for a given water saturation. Oil swelling reduces the water saturation, which also results in a decrease in the fractional flow of water. The CO2 swelling effect on the o"/CO2 mixture is estimated with the Peng-Robinson equation of state (PR-EOS). Because of the improvement in fractional-flow conditions, the reservoir is produced to a lower residual oil saturation.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call