Displacement Mechanism and Water Shielding Phenomena for a Rich-Gas/Crude-Oil System
- D.L. Tiffin (Amoco Production Co.) | H.M. Sebastian (Amoco Production Co.) | D.F. Bergman (Amoco Production Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- May 1991
- Document Type
- Journal Paper
- 193 - 199
- 1991. Society of Petroleum Engineers
- 4.3.4 Scale, 5.4.9 Miscible Methods, 4.1.2 Separation and Treating, 5.4.7 Chemical Flooding Methods (e.g., Polymer, Solvent, Nitrogen, Immiscible CO2, Surfactant, Vapex), 4.6 Natural Gas, 5.4 Enhanced Recovery, 5.4.2 Gas Injection Methods, 1.6.9 Coring, Fishing, 5.3.2 Multiphase Flow, 5.7.2 Recovery Factors, 5.2.2 Fluid Modeling, Equations of State, 6.5.2 Water use, produced water discharge and disposal, 4.1.5 Processing Equipment, 5.2.1 Phase Behavior and PVT Measurements
- 0 in the last 30 days
- 254 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
This paper presents an experimental study involving a series of hydrocarbonmiscible displacements of a crude oil. Compositional analyses of both flowingeffluent phases show that most of the oil was recovered by the condensing-drivemechanism. High mobile water saturations in the highly water-wet Berea coresused led to the same oil-trapping phenomena as noted for CO2 displacements.
Benham et al. and others described a process by which a rich injection gaswould displace a relatively lean oil by the condensing-drive mechanism. In thisprocess, the oil gradually becomes enriched by intermediate components thatcondense out of the injection gas. Eventually, the multiple-contacting processthat occurs in the reservoir enriches the oil to the point where it is misciblewith the injection gas, and a zone of contiguously miscible fluids from theoriginal oil to the injection gas forms. Fig. 1 is a ternary-diagramrepresentation of the fluid compositions that occur during this process. Thedotted line represents the overall composition path of the produced fluids, andthe solid curves are the composition of the coexisting produced vapor andliquid phases, which are joined by the tie-lines. The condensing-drive processcreated a transition zone of fluid compositions, which progress from reservoiroil through a two-phase region to a critical point and formally to theinjected-gas composition. The important feature of this diagram is that, as theliquid phase becomes more and more enriched, the coexisting liquid and vaporphases approach each other at a critical point near the injected-fluidcomposition. Recently, Zick proposed that not all rich-gas/real-oil systemscould develop miscibility by the condensing-drive mechanism. Instead, heproposed a condensing/vaporizing mechanism. In this mechanism, intermediatecomponents initially condense from the injection gas into the oil, as Benham etal. proposed. Alter the oil becomes saturated with the intermediates from theinjection gas, however, condensation ceases. Instead, vaporization of heavierintermediates from the enriched oil to fresh injection gas occurs, and themechanism changes to a vaporizing drive similar to that occurring during CO2floods. The injection gas, now enriched with heavier intermediates from theoil, moves ahead to contact fresh oil and begins the condensing process again.In this condensing/vaporizing process, thermodynamic miscibility might never beestablished even though oil recovery can be quite high. Fig.2 is a ternaryrepresentation of the condensing/vaporizing process. Here, the oil and gascompositions start to approach each other, as in the condensing process, but atsome point vaporization begins and the compositions begin to diverge. Thus, nocritical point is encountered and thermodynamic miscibility is not established.Studies reporting this vaporizing/condensing mechanism for a rich-gas floodwere based on computer modeling studies and laboratory mixing-cell experiments.The water-alternate-gas (WAG) injection process has been used in pilot andfield-scale enhanced-gas-drive floods to control the inherently poor mobilityratio between the injected gas and in-place oil. This injection technique isbelieved to improve the sweep efficiency of injected solvent, thereby improvingutilization of this relatively expensive fluid. WAG can lead to high mobilewater saturations, particularly in previously waterflooded reservoirs. Previousstudies, primarily emphasizing first contact-miscible (FCM) and multiplecontact-miscible (MCM) systems with CO2, where miscibility develops by avaporizing process, have shown that the in-place oil can be shielded from theinjected solvent by mobile water in water-wet porous media. This prevents theinjected solvent from contacting and displacing part of the residual oil inlaboratory systems. Shielding is either much less severe or nonexistent inoil-wet porous media. The purposes of this study were to determine thefollowing. 1. Whether the rich-gas displacement mechanism in a long corefloodis a classic miscible condensing process or whether a condensing/vaporizingprocess occurs. 2. The amount of oil recovered by vaporization, if it occurs.3. Whether high mobile water saturation in rich-gas floods results insignificantly different oil-trapping behavior in a water-wet core than MCM CO2or FCM systems. 4. The effect of long contact time on trapped oil recovery bydiffusion across the trapping water boundary.
Three types of displacement tests were conducted during this study:slim-tube tests and two types of long Berea core tests. Slim-tube tests wereconducted to measure the minimum miscibility pressure (MMP) with proceduresdiscussed elsewhere. Briefly, to obtain the MMP from a series of slim-tubedisplacements, oil recovery is plotted vs. run pressure. Recovery typicallyincreases with increasing pressure until a maximum recovery is obtained.Further increase in pressure results in little, if any, additional increase inoil recovery. This point of maximum recovery (or breakover torn in therecovery-vs.-pressure plot is the MMP. This point of miscibility has been foundto correlate with changes in sight-glass observations for MCM CO2 displacementsin which miscibility is developed by a vaporization or extraction process. Inone type of Berea core test, the effluents were collected at high pressure andthen analyzed after completion of the flood to study the rich-gas displacementmechanism. Fig. 3 is a schematic of the coreflood apparatus used to collectproduced fluids at test conditions. Each collected sample contains about 90cm3, or 0.05 PV, of fluid for subsequent PVT analyses. Additional corefloods tostudy water-shielding effects were conducted with a similar apparatus in whichthe produced fluids were flashed to atmospheric conditions rather thancollected at high pressure. More details on procedures for these 2-in. [5.1cm]-diameter Berea core tests are found elsewhere. All tests were conducted at135F [57.2C], and the corefloods were conducted at 2,000 psi [13.8 MPa]. Table1 presents compositions and fluid properties of the injection gas, recombinedreservoir crude oil, and brine used in all the tests. Also, Fig. 4 shows apressure-vs.-composition (bulk mole fraction solvent) phase diagram of thecrude oil and rich-gas solvent.
|File Size||523 KB||Number of Pages||7|