Influence of Texture on Steady Foam Flow in Berea Sandstone
- R.A. Ettinger (U. of California) | C.J. Radke (U. of California)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- February 1992
- Document Type
- Journal Paper
- 83 - 90
- 1992. Society of Petroleum Engineers
- 1.6 Drilling Operations, 5.6.5 Tracers, 5.1 Reservoir Characterisation, 1.6.9 Coring, Fishing, 4.1.2 Separation and Treating, 2.5.2 Fracturing Materials (Fluids, Proppant), 6.1.5 Human Resources, Competence and Training, 4.1.5 Processing Equipment, 5.2.1 Phase Behavior and PVT Measurements, 4.1.4 Gas Processing, 5.3.1 Flow in Porous Media
- 1 in the last 30 days
- 577 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
An understanding of texture, or bubble-size, evolution is paramount formodeling foam flow in porous media in that fine textures paramount for modelingfoam flow in porous media in that fine textures may impart large flowresistances. Bubble size, in turn, is determined by complicated lamellacreation and decay processes. We study, for the first time, the quantitativerole of bubble size in the steady flow of strong foam through a 0.8- m Bereasandstone. Inlet and outlet textures are determined from photomicrographs takenof bubbles flowing through specially designed visual cells. Concurrentmeasurements of pressure and liquid-saturation profiles by microwaveattenuation are acquired for gas velocities from 1 to 3 m/d and foam qualitiesfrom 70% to 90%. A simple, 1D foam population-balance model is outlined toquantify the observed flow and texture behavior. Agreement between the proposedmodel and the new bubble-size and flow data is adequate. The population-balancemethod proves to be a useful tool.
Foam shows promise as a general fluid for improving mobility control in EORprocesses. To optimize foam usage, however, knowledge of the rheologicalproperties of foam in porous media is necessary. From the outset of foamstudies in porous media, bubble-size distribution, or equivalently, foamtexture, was recognized as a very important factor controlling foam mobility.In spite of this recognition, relatively little work has been directed towardmeasuring pertinent bubble sizes. Marsden et al. apparently were the first toattempt bubble-size measurements characteristic of flow through porous media.They obtained mean sizes for foam exiting sandpacks of unreported length fordifferent foam qualities and for different surfactant types and concentrations.Their main finding was that bubble size, which was on the order of pore bodies,correlated directly with foam mobility, whereas quality did not. Surprisingly,they indicated that smaller bubbles compel a lower flow resistance in porousmedia. This finding is at odds with more recent work, which argues fordecreasing mobility with decreasing bubble size. More recently, Treinen et al.examined foam bubbles exiting sandpacks for different surfactantconcentrations, but at flow velocities close to 1 m/d. These researchers foundthat bubble sizes observed in visual cells were a factor of 10 larger thantypical pore sizes in the sandpacks. Hence, they were hesitant to correlatetheir flow results with texture. Friedmann and Jensen observed the bubbletexture exiting several different porous media, including steel-wool packs,sandpacks, and Berea sandstone. They used short cores and rather highvelocities up to several-hundred meters per day. Generally, at constantquality, increasing velocities decreased the bubble size. No concomitantpressure-drop data were given. Thus, to date, the few studies on bubble texturerelated to flow mobility in porous media have been mainly qualitative. Noeffort was directed toward modeling flow with the measured textures. This workattempts, as far as possible, quantitative evaluation of bubble texture and itsrelationship to foam-flow behavior.
Apparatus. Fig. 1 shows a schematic of the foam-flow apparatus. The porousmedium is a fired Berea sandstone 20 cm long x 10 cm wide x 1.2 cm deep. Thesandstone has an absolute permeability of 0.8 m and a porosity of 0.24. Topermit liquid-saturation measurements by microwave attenuation, the Berea slabis thin and is epoxied into a core holder constructed of 1.6-cm thickpolystyrene, a material that is nearly transparent to microwaves. polystyrene,a material that is nearly transparent to microwaves. Unfortunately, the fragilecore holder hindered measurements at high velocities, where large foam pressuregradients emerge, and also prevented application of a backpressure. During manyexperiments prevented application of a backpressure. During many experimentsfoam was pregenerated in a 5.1 x 2.5 x 2.5-cm, 0.5 m unfired Berea core. Foamtexture was observed in carefully designed visual cells placed just before andafter the 0.8- m Berea core. The surfactant solution is a degassed, salinesolution containing 0.83 wt% NaCl (Mallinckrodt, reagent grade) with 0.83 wt%active C(14-16) alphn-olefin-sulfonate surfactant (Bioterg AS-40TM, Stepan).The surface tension, measured by the Wilhelmy-plate method, is 33 mN/m, and thesolution viscosity is 1.0 mPas at ambient temperature. Foamer solution isdelivered with either a liquid-chromatogiaphy pump (Altex, Model 100A) or ahigh-pressure piston pump (ISCO, pump (Altex, Model 100A) or a high-pressurepiston pump (ISCO, Model #314). The nitrogen flow rate is controlled with amass flow controller (Brooks, Model #5850). Gas pressure profiles are obtainedwith four differential transducers (Validyne, Model DP-15) with 690, 520, 340,and 140 kPa diaphragms and located respectively, slightly upstream, 5.1, 10,and 15 cm from the core inlet. Signals from the transducers are sent to a10-channel demodulator (validyne, Model MCI), and the voltages are recordedcontinuously (Linear, Model 585). Liquid-saturation profiles are measured byscanning microwave attenuation, which, after calibration, detects the watercontent in the core by the Beer-Lambert absorption law. Scaring of thesandstone is accomplished with a stepper motor, and an IBM PC-XT is used forcontrol and data acquisition taking 30 saturation PC-XT is used for control anddata acquisition taking 30 saturation measurements at 0.51cm intervals. Moreinformation on the scaring microwave attenuator is available elsewhere. Despitethe long wavelength (i.e., 1.4 cm) of the microwaves, we were concernedinitially that the finely dispersed foam bubbles inside the core mightinterfere with the water-saturation measurements. To allay this concern, theattenuation was determined for a surfactant solution and a bulk foamedsurfactant solution containing the same amount of water: Measured attenuationwas identical for several different surfactant solutions and for severaldifferent foam textures, giving us confidence in the microwave technique. Toascertain the foam-bubble-size distributions, in-line viewing cells wereconstructed. These have the distinctive feature of allowing examination of asingle layer of bubbles. Each viewing cell consists of a cylindrical piece ofplexiglass with a thin disk of uniform thickness machined into one side. Thisside is then attached to a second, smooth cylindrical piece of plexiglass, andinlet and exit lines are drilled into the machined-out disk. It is crucial thatthe viewing area have a gap thickness that is less than the average bubblediameter. The foam bubbles are then deformed from their natural 3D polyhedralshape into 2D shapes that permit clear focusing under 20X magnification with a35-mm camera (Nikon Nikkormat) and attached bellows (Nikon, BP-4) and 50-mmlens (Nikkor). To messure a range of bubble sizes, two separate viewing cellswith different machined gap depths were constructed. The first cell, with aviewing region 0.25 mm deep and 3.1 cm in diameter, is used for the largerbubbles with diameters near 300 m. The second cell, which has a viewing region0.051 mm deep and 7.0 cm in diameter, is used to measure bubbles with diametersaround 100 m. Several photomicrographs of the foam flowing through the viewingcell are taken, and enlargement are made such that the bubble diameters rangein absolute size from 1 to 25 mm.
|File Size||1 MB||Number of Pages||8|