Light Oil Recovery From Cyclic CO2 Injection: Influence of Low Pressures Impure CO2, and Reservoir Gas
- T.G. Monger (Louisiana State U.) | J.C. Ramos (Louisiana State U.) | Jacob Thomas (Louisiana State U.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- February 1991
- Document Type
- Journal Paper
- 25 - 32
- 1991. Society of Petroleum Engineers
- 4.3.3 Aspaltenes, 4.1.9 Tanks and storage systems, 4.2.3 Materials and Corrosion, 4.1.5 Processing Equipment, 5.2.1 Phase Behavior and PVT Measurements, 5.3.4 Reduction of Residual Oil Saturation, 4.6 Natural Gas, 1.6.9 Coring, Fishing, 5.6.4 Drillstem/Well Testing, 5.3.2 Multiphase Flow, 5.5 Reservoir Simulation, 5.7.2 Recovery Factors, 5.4 Enhanced Recovery, 5.4.2 Gas Injection Methods, 4.1.2 Separation and Treating, 5.4.1 Waterflooding, 5.5.8 History Matching
- 1 in the last 30 days
- 681 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
This paper is a laboratory and field investigation of the CO2 huff 'n' puffprocess for the enhanced recovery of light crude oil. Cyclic CO2 displacementresults with live and dead oils in watered-out Berea cores are presented.Sixty-five single-well cyclic CO2 field tests conducted in a pressure-depletedreservoir are evaluated.
Cyclic CO injection was originally proposed as an alternative to cyclicsteam for the recovery of heavy crude. Investigations of CO2 huff 'n' puffapplicability to the enhanced recovery of light oil began in 1984. Theliterature now provides laboratory core-flood results, field-test data, andnumerical simulations that demonstrate that cyclic CO2 injection is a viablemeans of improving light-oil recovery. The risk associated with implementingCO2 huff 'n' puff is considerably less than that associated with other EORmethods, which typically require much higher upfront investments and experiencemuch longer project lives. Cyclic injection may also be the only EOR option forsmall or discontinuous reservoirs because the single-well process does notdemand well-to-well displacement.
Most of the available literature on light-oil recovery by cyclic CO2injection addresses reservoir conditions that in this paper are callednear-miscible-i.e., at higher pressures that do not exceed the minimummiscibility pressure (MMP) but that afford dense CO2, ranging up to about 0.6g/CM. Information on the performance of CO2 huff 'n' puff at these conditionsis provided in a performance of CO2 huff 'n' puff at these conditions isprovided in a laboratory coreflood study, case histories, field-testevaluations, and a numerical-simulation paper. These reports concur that cyclicCO2 injection can recover incremental light oil, CO2 utilization varies,several oil-displacement mechanisms contribute to enhanced oil production,simulation can yield a successful history match, and available methods forpredicting incremental oil recovery show only modest agreement with actualfield data.
The available literature provides limited information for pressures thatexceed the MMP. One field test performed at miscible pressures that exceed theMMP. One field test performed at miscible conditions was evaluated, and theresults were favorable. Furthermore, cyclic corefloods at miscible conditionsshowed that waterflood residual oil could be displaced. Overall, laboratory andfield results suggest that miscible conditions will somewhat impair CO2utilization.
Very limited data are available on the performance of CO2 huff n' puff forlight-oil recovery at lower pressures-i.e., below the MMP and even below theCO2 vapor pressure. In the gaseous state, CO2 density is low, ranging down toabout 0.01 g/CM. Haskin and Alstoni evaluated field-test results for ReservoirL, which exemplified these conditions. CO2 density was about 0.06 g/CM, andreservoir pressure was significantly below the MMP. Each of four wells wastreated with a 4-MMscf [113 268 std m] CO2 slug. One well indicated mechanicalfailure, and the remaining three responded with an average CO2 utilization of7.2 Mscf/bbl [1282 std M/M]. The light-oil reservoir historymatched byDenoyelle and Lemonnier was also apparently an immiscible application. One wellwas treated with a 380-Mscf [10 760-std M] CO2 slug. The well responded with aCO2 utilization of 3.3 Mscf/bbl [588 std M/M]. Simulations performed to predictthe effect of increasing the amount of CO2 injected predict the effect ofincreasing the amount of CO2 injected indicated that process efficiencydeclined with increasing slug size.
This paper extends the literature with additional laboratory core-floodresults and evaluations of 65 single-well field tests. The primary aim is tohighlight the favorable performance of cyclic primary aim is to highlight thefavorable performance of cyclic CO2 injection for light-oil recovery inpressure-depleted reservoirs. The immiscible recovery of light oil by use ofgaseous CO2 is the focus of laboratory and field results. Information on theutili-zation of impure CO2 is also provided. Furthermore, this papeusescoreflood data at higher pressures to address the significance of an initialreservoir gas saturation. Claridge suggested that the presence of awell-distributed gas saturation should be a favorable factor. presence of awell-distributed gas saturation should be a favorable factor. Materials andMethods
Cyclic laboratory corefloods were performed to complement fieldtestevaluations. Watered-out cores were used to test the ability of CO2 huff 'n'puff to recover light oil that was undeniably tertiary.
Laboratory materials and methods were designed to examine cyclic injectionunder two sets of conditions, One set used stock-tank oil (STO) at roomtemperature and lower pressure (500 psig [3448 kPa]) to model pressure-depletedreservoirs like those in the field tests evaluated for this paper. Corefloodsat these conditions probed the immiscible recovery of light oil with gaseousCO2. The utilization of impure CO2 was also examined at these conditions. Theother set of conditions used reconstituted reservoir oil (RRO) at 130 degrees F[328 K] and higher pressures (2,100 to 3,600 psig [14.5 to 24.8 MPa]) to modelwaterdrive reservoirs similar to those in previously described field tests.Corefloods at these conditions previously described field tests. Corefloods atthese conditions explored the influence of free and solution gases on light-oilrecovery. The STO used for the lower-pressure corefloods was a 31.6 degrees API[868
kg/M] crude from the Appalachian basin, KY. For the live-oil displacements,Timbalier Bay STO was reconstituted with pure methane to a GOR of 400 scf/STB[72 std M/M]. This 31.2 degrees API [870-kg/M] Crude was also the STO used inan earlier laboratory study. 2 Table 1 lists physical properties of thesecrudes. The minimum purities of the CO2 and methane used were 99.5 and 99.97mol %, respectively.
The lower-pressure corefloods were performed with the same 5.77-ft [1. 76-m]-long, 2-in [50.8-mm] -diameter Berea sandstone consolidated core used in anearlier laboratory study. This Core had a PV of 756 CM and a porosity of 21.2%.The live-oil displacements were performed in a second core of the same diameterbut 6 ft [1.83 m] long. This newer core had a PV of 811 CM and a porosity of21.8%.
Fig. 1 is a schematic of the horizontal coreflood apparatus used for thecyclic CO2 displacements. Table 2 lists the conditions that characterized eachcyclic experiment. The procedures used to clean the core, to establish awaterflood residual oil saturation (ROS), and to perform the "huff ' and"puff' portions of the cyclic injection tests were published recently andare only summarized here.
In the cyclic displacements, the core modeled the wellbore vicinity. Theremainder of the reservoir was modeled by a transfer vessel that was preloadedwith brine, manifolded at the core outlet, and maintained at constant pressureduring the run.
|File Size||1 MB||Number of Pages||8|