Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Drilling fluid selection and formulation (chemistry, properties)
Summary Several horizontal wells have been drilled in different sandstone formations in Sumatra. These formations have a typical permeability of 100 to 500 md, a low bottomhole pressure (BHP) of 450 to 750 psi, and a bottomhole temperature (BHT) of approximately 200ºF. The wells are completed with a perforated liner. The objective of the horizontal-drilling program was to increase oil recovery in low-permeability estuarine reservoirs. Some of the drilled horizontal wells did not perform to expectations, and an intensive study was undertaken to identify completion and stimulation opportunities to increase production. During this study, all aspects of the initial completions were examined and redesigned. The drill-in mud was reformulated to reduce the amount of polymer and increase the use of fine calcium carbonate to decrease lost circulation during drilling and to simplify the removal of filter cake during initial completion. Core tests were performed to identify the optimum fluid formulation, which dissolves the remaining filter cake but does not destroy the formation's natural permeability. A new way of removing the filter cake after completing the well was introduced using oxidizer technology. A new, true-fluidic oscillator (TFO) was used to remove near-wellbore skin (in conjunction with an improved acid system) for wells that have been producing for several months or years. The paper presents several case histories to discuss how completion and stimulation problems were systematically evaluated resulting in increased horizontal-well production. Introduction Oil- and gas-producing companies have been greatly interested in horizontal wells because their increased inflow area provides the potential to produce more oil and gas compared with vertical wells. The history of horizontal wells goes back as far as 1947.1 In the last two decades, the industry intensified efforts in exploring the potential of horizontal wells and overcoming many challenges that are particular to this type of completion. The advantages associated with horizontal wells have been identified by several authors and can be summarized as follows 2-4 :• Maximizing reservoir exposure. • Targeting multiple zones. • Exploiting thin pay zones. • Reducing drawdowns to minimize premature water and gas coning. • Improving production rates and increasing recoverable reserves. The three types of horizontal-well completions are openhole completions (with and without perforated or slotted liner), openhole completions with screens in place (with or without gravel pack), and cased-hole completions (also called stimulation completions).5 The decision of which completion to use depends on the specific reservoir characteristics.3 Cased-hole completions offer the advantage of simpler workovers and the option of specifically designed stimulation treatments. Furthermore, perforating the liner after cementing it in place can ensure that mud filtrate or invasion is bypassed. Consequently, these completions are more expensive and include challenges, such as obtaining an efficient cement placement along the entire interval for effective isolation and designing an effective perforating strategy. If sand control is an issue, then screens and prepacked liners are commonly used to avoid sand production (with or without gravel packing). Challenges in this case are associated with avoiding plugging the screen with mud and drilling solids and ensuring that the entire interval is producing to avoid hot spots, which could introduce local erosion of the screen.6 If the zone of interest is a consolidated formation that is not susceptible to formation collapse, then an openhole, or barefoot, completion becomes attractive. The disadvantages of a barefoot completion include the limited ability to perform workovers in certain areas in case water or gas breakthrough is observed. This becomes a significant problem in sandstone formations in which the high frictional force between the interface of the formation and coiled tubing does not allow coiled tubing to enter to a great depth. This challenge can be overcome by deploying slotted or preperforated liners plus external casing packers (ECPs). Optimum production results for openhole completions, with or without a preperforated liner, can be achieved by using a specifically designed drill-in fluid (DIF) then effectively removing the mud filter cake formed by the DIF. Both areas have been studied and discussed by Morgentaler et al.,7 Browne and Smith,8 and others. These studies found that 100% effective removal of the filter cake in horizontal openhole wells is not necessary because of the large inflow area. In fact, Browne and Smith concluded that if the permeability reduction is less than 70%, then the productivity of the wells does not fall significantly. Furthermore, they identified that the permeability of the mud filter cake has only to be increased to 0.1 md from approximately 10–5 to 10–8 md to achieve the same productivity as complete removal of the filter cake. Thus, 100% removal of the filter cake may not be a necessity and the optimum design should take this fact into account to identify a mud-removal design that meets the economic feasibility of the well. For horizontal wells that have been produced for a period of time and experience a production decline, the challenge is not to remove the mud filter cake but to identify where the damage is coming from and how to remove it. Common causes for production decline include depletion, scale buildup, paraffin and asphaltene dropouts, fines migration, and others. Thus, for older wells, the most important issue is identifying the damage and designing a treatment accordingly. For each particular well, an engineered solution should be designed involving problem identification, fluid-compatibility studies and core analysis, completion design (e.g., placement9 and unloading procedures), and more.
- North America > United States (1.00)
- Asia > Indonesia > Sumatra (1.00)
- North America > United States > Texas > Permian Basin > Yeso Formation (0.99)
- North America > United States > Texas > Permian Basin > Yates Formation (0.99)
- North America > United States > Texas > Permian Basin > Wolfcamp Formation (0.99)
- (29 more...)
- Well Drilling > Drilling Operations > Directional drilling (1.00)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Completion Installation and Operations (1.00)
Summary Drill-in fluids contain biopolymers such as xanthan gum, cellulose, and/or starch, along with bridging agents like sized calcium carbonate particles. These polymers are used to enhance the carrying capacity of the mud and to form a filter cake to minimize leakoff of the drilling fluids into the formation. Proper removal of drilling-mud filter cake is essential to minimize formation damage. The damage becomes more intense in tight formations, especially in horizontal gas wells in which the drawdown is not high enough to dislodge the filter cake. Another source of damage is the mud filtrate where fines migration and water blockage can occur in tight formations, especially in sandstone reservoirs. A thorough laboratory investigation of formation damage induced by drill-in fluids (water-based) was conducted. A mini-flow loop was used to assess formation damage induced by polymers present in the drill-in fluid. Various cleaning fluids were examined and their effectiveness in removing the filter cake was determined. These fluids included acidic brines, surfactants, mutual solvents, specific enzymes, and combinations of these fluids. The retained permeability of reservoir cores was determined for each cleaning system. This paper presents results obtained from the laboratory and recommendations made to remove drilling-mud filter cake from horizontal gas wells in a deep sandstone-gas reservoir. Introduction Wells drilled in the U reservoir have the potential to deliver a significant quantity of sweet gas and condensate, but can be damaged during drilling and completion. The bottomhole static temperature of this formation ranges from 280 to 305ºF. The reservoir depth is greater than 14,000 ft, and the initial reservoir pressure is nearly 8,500 psi.1 Well #1 is a vertical gas producer in the U reservoir. Petrographical studies showed that the formation has a fluvially dominated lithological structure in the U-B unit having massive- (Sm) and cross-bedded- (Sx) type sandstones. The horizontal permeability of the reservoir is very low, as seen from the core-analysis data. The poor permeability of the reservoir is attributed to the pore filling and pore throat, occluding quartz cements together with localized compaction. The low permeability is also ascribed to the narrow pore throats ranging between 1 to 3 µm. It has been documented by several researchers that formations with low and moderate permeability are susceptible to significant damage caused by drilling fluids.2 Return-permeability tests were performed with an indigenously designed "prototype" dynamic-mud-flow loop under simulated field conditions.2,3 Water-based drilling fluids are used for most of the wells in the well field. Solids from the drilling fluids can invade the formation and result in formation damage.2,4 Drilling fluids containing xanthan gum, starch, or polyanionic cellulose and sized calcium carbonate particles are used to limit the amount of fluid loss to the formation during the drilling operations. Better control of the drilling losses will maintain the integrity of the formation, reduce the risk to the production, and delay the requirement of stimulation and workover treatments. Most drill-in fluids consist of starch, cellulose, or xanthan gum, and bridging agents such as sized calcium carbonate or salt particles.5 The drilling mud used in the well field contained xanthan gum. Aqueous xanthan gum dispersions, prepared either by dilution of raw fermentation broths or by dissolution of xanthan powders, still contain insoluble particles and microgels.6 Some of the water-soluble polymers used in drill-in fluids invade the near-wellbore area and create skin damage.5 The use of specific enzymes to remove bio-polymers from the filter cake has increased over the last few years.5,7–10 Enzymes are proteins with active sites (i.e., they act like catalysts, which can promote a variety of chemical reactions).11 The normal activity of these enzymes depends on their environment. Abnormal conditions reduce their activity, while harsh conditions like high-acid concentrations and/or temperatures denature the enzymes. With enzymes acting as catalysts, they lower the activation energy of chemical reactions and accelerate the reaction rate. This new, growing remedial treatment involving enzymes is environmentally safe and can be applied over a wide range of pH values and temperatures up to 300°F.7 Unlike specific enzymes, acids and oxidizers are nonspecific reactive species that will react with most species present downhole.8 Because of their nonspecificity, they are being consumed by untargeted materials.8 The important characteristic of enzymes is that they target specific polymers. A given enzyme degrades certain substrates only.12 Cellulose, starch, and xanthan polymers are polysaccharides with different chemical structures. For this reason, various enzymes are needed to remove damage induced by each of these polymers. Xanthan gum is a biopolymer produced by a fermentation process that uses Xanthomonas campestris (a strain of bacteria). Xanthan gum is a heteropolysaccharide consisting of linear chains of D-glucose units that are bonded together by ß-1,4 glucosidic linkages and make the backbone structure of the polymer. A trisaccharide substituent attached to every other glucose unit of the backbone by a ß-1,3 glycosidic monosidic linkage.13 This side chain is responsible for the unique properties of this polymer, such as high viscosity at low polymer concentrations, high viscosity at low shear rates, a high degree of pseudo-plasticity, broad ionic-strength compatibility, and thermal stability. Resistance to a wide range of pH values and nonreactive tendency toward oxidizers and acids are also attributed to this side chain.14 The effect of ionic strength and pH on the apparent viscosity of various xanthan gum solutions was examined by Nasr-El-Din and Noy.15 Xanthan-specific enzymes degrade or cleave either the a-1,2 or ß-1,4 glycosidic linkages of the substituent and the ß -1,4 linkages of the backbone.16 The focus of this study is on core-flow tests with the water-based mud system used in the well field. Several core-flow tests were performed on reservoir core plugs. The main objectives of this study are to determine the retained permeability after mud circulation and acidic-brine (pH = 4) wash, the filtrate leakoff by the drilling mud and acidic brines, and the effect of specific enzymes on the return permeability.
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (0.95)
- Geology > Mineral (0.89)
Characterization of Partially Formed Polymer Gels for Application to Fractured Production Wells for Water-Shutoff Purposes
Sydansk, R.D. (New Mexico Petroleum Recovery Research Center) | Xiong, Y. (New Mexico Petroleum Recovery Research Center) | Al-Dhafeeri, A.M. (New Mexico Petroleum Recovery Research Center) | Schrader, R.J. (New Mexico Petroleum Recovery Research Center) | Seright, R.S. (New Mexico Petroleum Recovery Research Center)
Summary A laboratory study characterized partially formed chromium(III)-carboxylate/acrylamide-polymer (CC/AP) gels for water shutoff in fractures. These partially formed gels showed much lower effective viscosities during placement than comparable fully formed gels. During placement, leakoff rates through fracture faces were low for gelants and partially formed gels. During the first brine injection after gel placement, the pressure gradient required to breach the gel increased with the increasing polymer concentration. Most gel remained in the fracture and did not wash out. During brine flow through "wormholes" in the gel, stabilized residual-resistance factors (Frr) were large and increased with increasing polymer concentration. Introduction During this laboratory study, we characterized water-shutoff polymer gels of the type that are to be injected in the partially formed state into fractures that are connected to production wells. Findings of this study should also be relevant to other high-permeability anomalies that are connected to petroleum production wells. Other than fractures, these high-permeability anomalies could include solution channels, interconnected vugs, karstic features, joints, faults, rubble zones, and ultrahigh-matrix rock permeability. These features generally have permeabilities greater than two darcies. For water-shutoff applications in fractured production wells (i.e., during field applications), injected polymer gels are usually in the "partially formed" state during transit from the wellbore into the formation. For classical bulk-gel treatments applied to reservoir fractures, the injected polymer-gel solution should develop enough gel structure (including a microgel structure) to minimize detrimental gel-solution leakoff into the matrix-reservoir rock that is adjacent to the fractures. On the other hand, the gel should not be "fully formed" during placement because excessive injection pressures may be encountered. Use of partially formed gels permits manageable injectivities during placement and causes minimum damage to matrix rock when properly formulated, yet ultimately yields strong gels that will function as required to shut off water production. Gels must be in the partially formed state when injecting strong gels that will ultimately be "rigid and rubbery" in nature. The objective of this paper is to characterize the performance of polymer gels that are injected into fractures in the partially formed state. This study was intended to investigate the properties of gel that resides in the near- and intermediate-wellbore region and gel that is part of relatively small-volume gel treatments (i.e., treatments usually pumped in less than a day) that are applied to fractured-production wells. During the flooding experiments of this study that were conducted in 2-ft-long fractured cores, 40 fracture volumes (FVs) of gel fluid were injected as rapidly as possible (i.e., injected within about 7 minutes using a superficial velocity within the fracture of 16,600 ft/D). An explicit goal during this gel injection was to minimize time-dependent gel dehydration.1 For these particular experiments, the resultant mature gels residing in the fractures were 1.2 to 2.5 times more concentrated than the injected-gel formulation (as will be noted later in this paper). During the experiments conducted in 4-ft-long fractured cores, 80 FVs of gel formulation were injected at a superficial velocity of 4,130 ft/D within the fracture. Experiments in this paper addressed five objectives:Determination of the effective viscosity of partially formed gel formulations in fractures during gel injection. Estimation of damage to fracture-wall porous rock from gel-solution leakoff. Determination of the peak or critical pressure gradient at which the gel is first breached during brine injection after gel placement in a fracture. Determination of the stabilized (the equilibrium or final) Frr for water or oil flow through a gel-filled fracture. Characterization of disproportionate permeability reduction (DPR) during oil and water flow through gel-filled fractures.
- North America > United States > Wyoming (0.28)
- North America > United States > Texas (0.28)
- Asia > Middle East > Saudi Arabia (0.28)
- Geology > Rock Type > Sedimentary Rock (0.34)
- Geology > Geological Subdiscipline > Geomechanics (0.34)
- North America > United States > Wyoming > Bighorn Basin (0.99)
- North America > United States > Texas > Permian Basin > Yeso Formation (0.99)
- North America > United States > Texas > Permian Basin > Yates Formation (0.99)
- (23 more...)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Hydraulic Fracturing (1.00)
- Reservoir Description and Dynamics > Reservoir Fluid Dynamics > Flow in porous media (1.00)
- (3 more...)
Method for Characterizing Secondary and Tertiary Reactions Using Short Reservoir Cores
Ziauddin, Murtaza E. (Schlumberger) | Gillard, Matthew Robert (Schlumberger) | Lecerf, Bruno (Schlumberger) | Frenier, Wayne W. (Schlumberger) | Archibald, Iain (ChevronTexaco) | Healey, Duncan Stuart (ChevronTexaco International)
Summary A new technique for characterizing secondary and tertiary reactions during sandstone-matrix-stimulation treatments is presented. In the new technique, traditional experiments on short reservoir cores are supplemented with measurement of the effluent-element concentrations, batch-reactor experiments, and geochemical simulations to predict the extent of secondary and tertiary reactions in the reservoir treatment. Alternative methods of characterizing secondary and tertiary reactions, such as those using long-core flow tests and laboratory radial-flow setups, are reviewed. The new design technique is used in designing a treatment for a well in the North Sea. Details of how this technique was applied to the treatment design are presented. Post-treatment data from the well showed a successful matrix-treatment design. The production from the well increased by 1,400% immediately after the treatment. The 3-month stabilized production gain was 650%. Introduction Recent studies on matrix stimulation have strongly emphasized the importance of secondary and tertiary reactions in determining the success of matrix-stimulation treatments.1,2 However, the extent of these reactions under reservoir conditions is difficult to quantify. Several factors make the traditional acid response tests on short reservoir cores inadequate for characterizing secondary and tertiary reactions. First, secondary and tertiary reactions are slower than primary reactions, and so, much longer fluid-residence times in the core are required to observe these reactions. Second, linear flow along the axis in cylindrical cores is not representative of radial flow in a reservoir treatment. Third, cores used in the core tests may not be representative of the entire treatment interval. Fig. 1 illustrates the limitations of traditional core flow tests. Fig. 1a shows that a core plug is a small sample of the area of interest. For formations in which the mineralogy changes significantly in the pay-zone interval, a single core plug will not be representative of the entire treatment interval. Fig. 1b shows an example of how a traditional core flow test to evaluate two fluids on a short reservoir core can lead to erroneous conclusions. Shown in the figure are permeability profiles in a simulated reservoir treatment after injection of 50 gal/ft of acetic acid preflush, followed by 100 gal/ft each of 12/3 mud acid (12 wt% hydrochloric acid (HCl)/3 wt% hydrofluoric acid (HF) and an organic fluoroboric acid (i.e., a mixture of citric, boric, and hydrofluoric acids). In both cases, the reservoir was undamaged before the treatment. 12/3 mud acid provides good stimulation near the wellbore but causes damage deeper in the reservoir. The organic fluoroboric acid achieves a lesser stimulation near the wellbore but also causes less damage deeper in the reservoir than 12/3 mud acid. The post-treatment skin for the organic fluoroboric acid is -0.5, compared to a skin of 2 for the 12/3 mud acid. The organic fluoroboric acid is, therefore, a better fluid for the reservoir treatment. However, if these two fluids were evaluated with a traditional core flow test, 12/3 mud acid would have been erroneously selected because it provides better stimulation at the length scale of the core (˜ 4 in.). Therefore, core tests on short cores by themselves are inadequate for fluid selection for matrix treatments. For accurate evaluation of a proposed stimulation design, it is necessary to account for the formation damage caused by secondary and tertiary reactions, which are typically not observed in tests on short cores.
- North America > United States (1.00)
- Europe > United Kingdom > North Sea (0.48)
- Research Report (1.00)
- Overview (0.71)
- Geology > Mineral > Silicate (1.00)
- Geology > Geological Subdiscipline (0.91)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (0.35)
- North America > United States > Alaska > North Slope Basin > Kekiktuk Formation (0.99)
- Europe > United Kingdom > North Sea > Central North Sea > Moray Firth > Moray Firth Basin > Witch Ground Graben > Block 15/23a > Galley Field (0.99)
- Europe > United Kingdom > North Sea > Central North Sea > Moray Firth > Moray Firth Basin > Block 13/22a > Captain Field > Captain Formation (0.99)
- (9 more...)
Overview The rheology of polyacrylamide solutions and the effect of viscous-elastic fluids upon production equipment are described. Methods to solve the negative effects of viscous-elastic fluids on equipment are also introduced. The results of the modification(s) on the system and equipment show that the service life of equipment more than doubled, mechanical degradation of polymer decreased more than 50%, and energy consumption decreased markedly. Summary Most flow systems and equipment for the injection, production, and gathering of large amounts of fluid are designed for Newtonian fluids. Now, an increasing number of oil fields are on viscous-elastic flooding. After flooding by polyacrylamide (PAM) fluids, it was seen that the flow of this type of fluid in production and surface systems and equipment has its special characteristics, which in turn require special designs for the equipment and flow process. This paper introduces the actual performance of the production and surface flow systems and equipment, and points out the influence that the special rheologic characteristics, especially the viscous-elastic nature of the fluid, has on the system and the modification made. For example, there is a normal force acting on the beam pump sucker rods that causes pronounced eccentric mechanical wear on one side of the sucker rods and thereby very short service life for beam-pump wells. This normal force significantly lowers the efficiency of centrifugal pumps, increases the vibration and mechanical degradation in triplex pumps and the whole system, lowers the efficiency of maturing tanks and static mixers, etc. The modifications on the system and equipment and their results are shown. For example, the service life of beam pumps more than doubled; mechanical degradation of polymer in the system and triplex pumps decreased more than 50 and 70%, respectively; and the energy consumption of mixing equipment decreased 80%, among other results. Introduction China, especially Daqing oilfield, is implementing large-scale PAM flooding to increase the oil recovery. Very good technical and economical results1 have been obtained. However, when using equipment and flow processes suitable for Newtonian fluids, efficiency and online time is low, and vibration is serious. Results in the laboratory and experiences from the field show that this is caused by the viscous-elastic nature of the PAM fluid. Work was performed in the laboratory and field on the influences on equipment and flow processes by viscous-elastic fluids and on how to solve the problems that emerged. Some Rheological Properties, Flow Characteristics, and the Flow Field of Polyacrylamide FluidsThe viscosity of PAM fluids conforms to that of a power-law fluid2 (Fig. 1) From Fig. 1, it can be seen that PAM fluids, when tested by a HAAKE 150 rheometer, have viscous-elastic characteristics. From Figs. 1 through 3, it can be seen that the higher the concentration and molecular weight, and the lower the salinity of the fluid, the higher the viscosity and elasticity of the PAM fluid. From Figs. 4 and 5 it can be seen that PAM fluid has "extension viscosity" and "normal-stress differential" when it is flowing (pseudoplastic fluids do not). The higher the shear rate, the higher the extension viscosity and normal-stress differential. The relationship between resistance factor and Reynolds number and local resistance (pressure drop) and fluid velocity is shown in Figs. 6 and 7, respectively. It is similar to that of pseudoplastic fluids. From Fig. 6, it can be seen that when the Reynolds number NRe is less than 2,000, the relational curve of PAM fluid is similar to that of water (a Newtonian fluid), which has the relationship f=64/NRe. When NRe increases from 2,000 to 4,000 (the upper limit of this test on PAM fluid), the flow of PAM fluid is still very similar to laminar flow, while that of water is in the transition zone. For NRe>4,000, the flow of water is in its turbulent state. In this test, all resistance factors of PAM fluid were close to or lower than that of water. From Fig. 7 it can be seen that for this test, with pipe inner diameter of 50 mm and choke diameter of 12 mm, the pressure drop across the choke for PAM fluid is very close to or even slightly lower than that of water. The flow field of PAM fluid at "dead ends" is shown in Figs. 8 and 9. The parameters of the tests shown in Figs. 8 and 9 are pipe diameter of 65 mm; for PAM concentration of 1000 mg/L, the velocity in the pipe is 0.72 m/s; for PAM concentration of 5000 mg/L, the velocities in the pipe are 0.22 and 0.72 m/s, respectively. From Figs. 8 and 9 and other test results, it can be seen that the flow field is quite different from that of pseudoplastic fluids:When the PAM concentration is 5000 mg/L and fluid velocity lower than 0.41 m/s, there is no turbulence in the dead ends; when the velocity is higher than 0.41 m/s, there is turbulent flow in the "dead ends" and the turbulent flowlines are connected to the main flowlines. When PAM concentration is 1000 mg/L, in the range of velocity tested (0.22 to 0.72 m/s), there is always turbulent flow in the dead ends. The flowlines of this turbulent flow were separated from the main flowlines. A typical example of the flow field of PAM fluid in the mixing (maturing) tank is shown in Fig. 10. For this test, the mixing blades revolved at a speed of 100 to 320 rev/min, the ratio of the blade and tank diameter is 0.367 to 0.61, and the PAM concentration is 1000 to 5000 mg/L. Under these conditions, the flow field in the tank and the speed at the tip of the blade and nearby PAM-fluid flow speed were measured. From the flow-field tests, it could be seen that when the fluid in the tank is Newtonian or low-concentration PAM, then the following were true:The flow of fluids in the tank is axial (fluid flows down the axis and up the walls of the tank). When the medium is water, the speed of the fluid flow near the blade tip is approximately the same as the speed of the blade tip. The blind area (area with no fluid flow) of the tank is rather small.
- Asia > China > Heilongjiang > Songliao Basin > Daqing Field > Yian Formation (0.99)
- Asia > China > Heilongjiang > Songliao Basin > Daqing Field > Mingshui Formation (0.99)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production logging (1.00)
- Production and Well Operations > Artificial Lift Systems > Beam and related pumping techniques (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Chemical flooding methods (0.84)
Summary Foams provide a highly attractive alternative to conventional non- Newtonian fluids for various oil and gas applications because of their high viscosity and low liquid content. Foams have a long history of proven performance in well stimulations, drilling, acidizing, cleanout operations, and enhanced oil recovery (EOR). Foam apparent viscosity is used in hydraulic (pressure drop) calculations for various pumping operations. Thus, accurate rheological characterization is very important to predict foam viscosity. In this investigation, rheological experiments are carried out with guar gel and guar foam fluids with a 1/2-in.-pipe viscometer at 1,000 psi and temperatures ranging from 100 to 200°F. Guar, surfactant, and nitrogen (N2) are used as the gelling agent, foaming agent, and gas phase, respectively. Formulations were prepared and tested at several flow rates for qualities ranging from 0 to 80%. This study has shown that both guar gel and guar foam fluids exhibit behavior analogous to the power-law model fluid. The experimental data are used to develop new rheological correlations to predict power-law fluid-flow parameters and, thus, the apparent viscosity of the foam fluids. These new correlations are functions of temperature, foam quality, shear rate, and fluid concentrations and were compared with various published correlations. The available correlations in the literature have assumed that the foam's flow behavior index is similar to that of the liquid phase. This assumption does not support the basic physics that the increase in viscosity decreases the flow behavior index. Our results show that the previous assumption is invalid and leads to a prediction of extremely low apparent viscosity for foam fluids. In contrast to previous studies, we report high apparent viscosities of high-quality foam fluids. Introduction Foam has long been an important subject for scientific study because of either its applicability or its impairing effects in a variety of industrial oil and gas applications and chemical processes. The use of foam fluids is a highly attractive alternative to use of conventional non-Newtonian fluids. Foams possess many desirable properties that make them highly applicable as drilling fluids and find wide use in stimulation, wellbore cleanout, workover and remedial operations, gas-mobility control in improved oil recovery (IOR), acid diversion in well stimulation, and water shutoff and gas blocking in water-coning prevention and gas storage, respectively. Recently, the use of foam as a fracturing and underbalanced- drilling fluid has enjoyed rapid growth and recognition. On the other hand, formation of foams in chemical processes, such as oil/water separation, oil flotation, or distillation, causes problems by reducing process efficiency and operating control. Foams are a special kind of colloidal dispersion in which a gas [air, N2, carbon dioxide (CO2), etc.] is dispersed in a continuous liquid phase. Most of the foams used in the oil and gas industry contain a dispersed phase of more than 70%. The dispersed-phase volume fraction in foams is normally known as foam quality. The liquid in the foam always contains a surface-active agent (surfactant) that accumulates preferentially at the gas/liquid interfaces and stabilizes the films. The surfactant may be anionic, nonionic, or amphoteric. Foams are classified as non-Newtonian fluids because they exhibit shear-rate-dependent viscosity in laminar flow regime. Characteristically, foam viscosity is greater than the values for the liquid and gas phases. However, it is difficult to control foam behavior to achieve the advantages mentioned previously; fully understanding the mechanism governing foam behavior still remains a challenge. Bearing this in mind, an attempt has been made to achieve the following objectives in this work.Study the rheological characterization of guar foam using a pipe viscometer. Study the effects of quality, temperature, and fluid concentration on foam flow parameters. Develop rheological correlations for guar foam to predict the rheological parameters. To achieve the aforementioned objectives, this study was conducted in the laminar flow regime at a 1,000-psi confining pressure through a 1/2-in. pipe with a temperature range of 100 to 200°F. Gel concentrations of 20, 30, and 40 lbm/Mgal guar were used for foam qualities ranging from 0 to 80%. This work discusses in detail the fluids tested, experimental setup used, test procedure, and data-analysis technique followed and presents rheological results for guar foams in straight tubings. Literature Review A number of authors have studied foam rheology by use of capillaries and rotational viscometers. Mitchell studied aqueous foam in small capillary tubes and classified it as a Bingham plastic-type fluid with a strong dependence on quality. No wall slippage effect was observed with increased shear stress and tube diameter. Furthermore, the viscosity was dependent on shear rate at qualities greater than 54% but on quality only at very high shear rates. Saintpere et al. investigated the rheological properties of aqueous foams for underbalanced drilling. Gelled foams with polyacrylamide (PAA) and xanthan were investigated at ambient conditions using a parallel-plate geometry. It was found that the apparent viscosity of foam is a function of not only its quality but also its texture and the presence of polymers in the liquid phase. Their results confirmed that foam is a yield-pseudoplastic and can be represented by a Herschel-Bulkley model. Patton et al. found foams to be pseudoplastic and several orders of magnitude more viscous than the liquid phase, with drier foams having higher apparent viscosities than wet ones. As the capillary-tube dimensions decreased (both length and diameter), the measured apparent viscosities increased. Reidenbach et al. used a recirculating loop to perform experimental tests with aqueous and gelled water foams using N2 and CO2 as an internal phase. A Herschel-Bulkley model was used to describe gelled foams, and aqueous foams were characterized with a classical Bingham plastic model. They concluded that the flow behavior index, n, of foam fluids was the same as that of the liquid phase, while the foam consistency index is a function of the liquid-phase consistency index and foam quality. Harris and Reidenbach used a recirculating flow loop to investigate the rheology of nitrogen foam at 1,000 psi and various temperatures up to 300°F. Hydroxypropyl guar (HPG) was used as the liquid phase. They found that the foam behaved as yield pseudoplastic and could be described by the Herschel-Bulkley model. They assumed that the flow behavior index, n, of foam fluids is same as that of the liquid phase and that the consistency index is a function of liquid-phase consistency index and foam quality.
- Asia (0.68)
- North America > United States > Oklahoma (0.48)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring (1.00)
Summary The drag-reduction (DR) behavior of xanthan gum, partially hydrolyzed polyacrylamide, guar gum, and hydroxyethyl cellulose at various polymer solutions has been experimentally investigated with a full-scale coiled-tubing (CT) test facility that consists of 1-, 112-, and 238-in. CT reels. Data analysis revealed that the tubing diameter, curvature ratio, and polymer concentration are important factors affecting the DR in CT. The modified DR envelope is a useful tool for evaluating the DR performance of polymer solutions in CT. Introduction CT has found many applications in the petroleum industry, including drilling, cementing, wellbore cleanout, and hydraulic fracturing. However, the excessive friction-pressure loss because of the relatively small tubing diameter and because of the tubing curvature often limits the maximum obtainable fluid-injection rates. A recent experimental investigation indicates that the frictional loss in CT is significantly higher than in straight tubing. One way to increase the injection rate and reduce pumping costs is to use drag-reducing additives (or drag reducers). Typical drilling and completion fluids, usually polymer solutions, have been found to exhibit some drag-reducing property. Therefore, it is of practical importance to investigate the DR properties of these solutions in CT. Frictional pressure in turbulent pipe flow can be drastically reduced by adding small quantities of certain long-chain polymers to the solvent, such as water. This phenomenon is called DR. Credit is generally given to Toms for being the first to observe the phenomenon; therefore, DR is also called the Toms effect. Several references from petroleum literature indicate the importance and potential applications of DR to this industry. Savins reported pipe flow tests using a number of synthetic and natural polymeric materials and three tubing sizes, and factors affecting the drag ratios were studied. He also compared the test data with Dodge-Metzner friction-factor correlation and observed the "diameter effect." There have been several extensive reviews on DR, such as Lumley, Hoyt, Virk, and Berman. The paper by Virk addresses the DR fundamentals of dilute solutions of linear, random- coiling macromolecules in turbulent pipe flow. It covers broad areas of DR studies, including gross flow, mean velocity profile, turbulence structure, and mechanisms. Virk proposed the concept of a maximum DR asymptote and the DR envelope, which has proven to be a useful tool for evaluating the DR performance of polymer solutions. The additional difficulty in studying DR behavior in CT is caused by the different flow fields in CT flow. Because of the effect of centrifugal forces, a secondary flow of vortical form occurs in the CT cross section. Dean pioneered the theoretical study of Newtonian fluid flow in curved pipes. A similarity parameter, later called the Dean number [defined as NDN=NRe(a/ R)] was introduced to characterize flow in curved pipes. In 1929, White conducted experiments and confirmed the validity of the small Dean number series solution. Ito conducted flow tests with water in smooth, drawn copper tubing and developed empirical correlations for the friction factor of curved pipe flow. Srinivasan et al. also used smooth, drawn copper tubing in their experiments but used both water and oil as test fluids. Similarly, empirical friction-factor correlations were established for the laminar, transition, and turbulent flow regimes. Among the few reported studies of non-Newtonian fluids in CT, those by Mashelkar and Devarajan should be mentioned. They theoretically developed correlations for friction factors of non-Newtonian fluids in CT and checked these correlations against experimental data. Recently, two papers have reported results of frictional pressure losses in CT from full-scale experiments. Azouz et al. measured frictional pressure losses of linear and crosslinked guar gum and hydroxypropyl guar (HPG) in 112-in. CT. McCann and Islas' flow experiments were conducted with CT diameters of 1.75, 2, and 2.375 in. on a 98-in.-diameter reel. The proposed friction-factor correlation by McCann and Islas was generalized from a Newtonian model. In this study, extensive experiments have been conducted with a full-scale CT flow loop. This paper presents the experimental results and the DR performance characteristics of typical drilling and completion fluids tested in the CT. Major factors affecting DR performance are also discussed. Theory and Definitions Based on the experimental data of gross flow tests, the DR of each fluid can be calculated with Eqs. A-1 through A-4. The Appendix also describes Virk's envelope, which is useful for evaluating the DR performances of polymer solutions in straight pipes but is not applicable for CT. Because the friction factor for solvent in CT is higher than for the same solvent in straight tubing, the DR in CT should be evaluated by comparing the friction factor of the polymer solution in CT with the friction factor of its solvent in the same CT. Therefore, we modified Virk's DR envelope by replacing the Prandtl-Karman line with a Newtonian friction-factor correlation for CT, such as the Srinivasan et al. or Ito correlations, as presented in the Appendix. The modified DR envelope is illustrated in Fig. 1, in which the Srinivasan et al. correlation has been used for the zero-reduction line.
- Well Drilling > Drilling Operations (1.00)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Completion Installation and Operations > Coiled tubing operations (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Chemical flooding methods (1.00)
Summary A new model was developed to describe water leakoff from formed Cr(III)-acetate-HPAM gels during extrusion through fractures. This model is fundamentally different from the conventional filter-cake model used during hydraulic fracturing. Even so, it accurately predicted leakoff during extrusion of a guar-borate gel. Thus, the new model may be of interest in hydraulic fracturing. Contrary to the conventional one, the new model correctly predicted the occurrence of wormholes and stable pressure gradients during gel extrusion through fractures. Introduction This work was motivated by an attempt to understand the propagation of formed gels as they extrude through fractures during water shutoff treatments. Earlier works1-6 revealed that formed gels lost water during extrusion through fractures and that water leakoff controlled the rate of gel propagation. Leakoff was also known to control the rate of fracture growth during hydraulic fracturing 7-11 and during produced water reinjection.12-14 In the conventional view of hydraulic fracturing, the leakoff rate was determined by one (or a combination) of three mechanisms.7,8,10Propagation of the fracturing fluid front into the rock matrix (i.e., away from the fracture face). Reservoir fluid viscosity/compressibility effects. The formation of a filter cake associated with particulate matter that was suspended in the fracturing fluid. The latter mechanism may involve formation of a filter cake on the fracture surface (i.e., an external filter cake) and/or penetration of the particulates some distance into the porous rock (i.e., an internal filter cake). This paper focuses on leakoff that is dominated by the formation of an external filter cake in a fracture, first reviewing the conventional filter-cake model and then presenting experimental evidence that questions a key assumption of the conventional model. Next, an alternative model is presented, and then important differences in predictions from both are discussed. This paper also examines gel behavior in fractures as a function of temperature, composition, and brine injection after placement. Conventional Filter-Cake Model The widely accepted model of filter-cake formation was introduced by Carter.7,8,10 Assume that a particulate-laden fluid contacts a rock interface (i.e., a fracture face) and a pressure difference, Dp, exists between the fracture and the porous rock. As solvent (with viscosity m) flows into the rock at a velocity ul, the particulates form a filter cake of permeability k and thickness L. At any given time, the filtrate velocity (i.e., the leakoff rate) is given by the Darcy equation. Equation Gel Behavior in Fractures Before gelation, fluid gelant solutions can readily leak off from fractures into porous rock.15 After gelation, however, the crosslinked materials will not penetrate significantly into the porous rock.1-6 Thus, formed gels must extrude through fractures during the placement process. In other words, the crosslinked polymer moves through the fracture as a semisolid and does not penetrate past the fracture faces into the porous rock. The pressure gradients required to extrude gels through fractures are greater than those for gelant flow. Depending upon conditions, the effective viscosity of formed gels in fractures is typically between 10 3 and 106 times greater than those for gelants.1 However, useful gels do not show progressive plugging during extrusion through fractures. This point is illustrated in Fig. 1, in which 75 fracture volumes (3500 cm3 or 213 in.3) of Cr(III)- acetate-HPAM gel were extruded through a fracture (in a 48´1.5´1.5-in. Berea core) at a fixed rate of 2000 cm3/hr (4,130 ft/D). After gel breakthrough at the end of the 4-ft-long fracture, the pressure gradient was stable at 13.5 psi/ft. In other experiments with this gel, the pressure gradients sometimes showed greater variations than those illustrated in Fig. 1.2 These variations may result from temporary gel screenouts that form and break during the extrusion process. However, in general, the pressure gradients do not steadily increase with increased time and gel throughput. This behavior is necessary to propagate formed gels deep into a fracture or fracture system. Of course, the presence or absence of this desirable behavior depends on the gel and the extrusion conditions. Some gels show dramatic screenouts and progressive plugging.3 For Cr(III)-acetate-HPAM gels, a minimum pressure gradient was required to extrude the gel through a fracture with a given width. 1 Once the minimum pressure gradient was exceeded, the pressure gradient during gel extrusion was insensitive to the flow rate. For example, in a 0.04-in.-wide fracture,4 a 1-day-old gel with 0.5% HPAM and 0.0417% Cr(III)-acetate exhibited a pressure gradient that averaged 28 psi/ft for injection fluxes between 413 and 33,100 ft/D. The gel mechanical degradation was fairly small. For gel produced from the fracture at the highest rate, the elastic modulus was approximately 20% less than that of the original. In all cases, the physical appearance of the gel remained unchanged by passage through the fracture.
- Energy > Oil & Gas > Upstream (1.00)
- Materials > Chemicals > Commodity Chemicals > Petrochemicals (0.34)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Hydraulic Fracturing (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery (1.00)
Summary The paper describes a 2D model of acid fracturing that is more rigorous than current acid-fracturing simulators. The equations are discretized and solved along fracture length and across the width (in x and y directions), as opposed to a lumped formulation in the conventional model in which the y-solution is replaced by a mass transfer coefficient Kg. In the 2D model, correlations for Kg are not needed, and the model uses directly measurable data only. In addition, the model can represent transient effects (Kg is variable along time and fracture length), combined with dispersion and thermal effects. Two models, one for lab experiments and one for field design, were developed. The lab model was validated, with excellent agreement, against the results of Roberts and Guin and compared with the 1D approach on data typical for phosphoria dolomite. The results show large differences in regions of high and low leakoff. Acid spending predicted by the 2D model is also higher, even when the Kg values are comparable. Correspondingly, the field-design model predicts larger spending and fracture conductivity. An example of the field design and post-fracture history match from the Cottonwood Creek Unit, Washakie County, Wyoming, is presented. It shows that the new model is in better agreement with field data. Extensions to 3D modeling are also discussed. Introduction The majority of acid-fracturing modeling to date has been done with formulas that integrate the basic equations across the fracture width (y-direction) and solve the proppant-transport equations along the fracture length (x-direction). In the absence of a solution in the vertical (z-) direction, such formulas may be termed as 1D models. This approach reduces the dimensionality of the problem (and the computer work). On the other hand, it becomes necessary to introduce an apparent mass-transfer coefficient Kg (first proposed by Roberts and Guin), which relates the acid spending to the average concentration across the width. This coefficient must then be related to the measurable parameters that characterize the process (i.e., the effective diffusion coefficient De and the reaction rate constant k). Recent examples of such 1D models are found in the literature, and their comparison is presented in Li et al. The weak point in these models is the relation among Kg, the flow properties, and De. Usually, Kg is calculated from dimensionless correlations that depend on the flow regime. There are several problems with this approach.Most of the data used to construct the correlations has been interpreted with Terill's analytical solution (Nierode***), which assumes an infinite reaction rate and a known value of the diffusion (mixing) coefficient De. Li et al. have shown the correlations for De to be unreliable in the laminar-flow regime. The correlations are discontinuous at the boundary between the laminar and transitional flow and depend on different sets of variables in the two regions. As a result, the calculated Kg has a large discontinuity between the laminar and transitional region when plotted as a function of the Reynolds number NRe . While this discontinuity may be real, an approach that does not rely on correlations would be preferable. The correlations have been derived from data obtained with zero or small leakoff. As previously shown, there must be a relation between Kg and leakoff velocity vl that is not implicitly included in these correlations. Thus, the motivation of the 2D acid-transport solution is to eliminate the need for Kg correlations by a direct solution of the equations across the width. In this formula, the mass transfer depends directly on De, k, and the flow field. Romero et al. recently presented a 3D model, which is essentially an extension of their 2D x-z model to solve across fracture width. However, the main focus of their work was on the vertical aspects of the problem. The comparisons between models in Ref. 7 do not allow isolation of the effect of the rigorous y-direction solution from the effects of the vertical flow components. In contrast, this paper focuses on the more fundamental question: is it necessary to solve the equations across fracture width, and if so, what consequence does it have on forecasting fracture conductivity and well production? Once this question is answered, the effect of the vertical component of the solution, which is obviously very important, can be better understood. Formulation of the 2D Acid Transport For simplicity, we will consider an element of fracture with variable width but constant height, as shown in Fig. 1. It will be assumed that concentration C is defined as the acid/volume mass of the solution. To use concentration in terms of moles, C must be replaced by Mw every where. The equations describing the flow of acid in a channel of unit height under laminar flow are as follows. Equation of Continuity. The general equation within a channel of variable width is as follows.Equation 1 where A=the elemental area dxdy, which represents the change of control volume owing to the change of fracture width b with time (the actual form of this term depends on how the coordinate system uses accounts for the moving boundary; it is not important for the final form of the equations). The boundary conditions are as follows.Equation 2 where vl=the leakoff velocity. By integrating this across y and z, we get the following equation, which is solved by the geometry module of a fracture simulator (except for compressibility).Equation 3 where (x, t)=the average volumetric velocity, Ac(x, t)=the local cross-sectional area, and ql(x, t)=the leakoff rate on one surface of the fracture. Assuming constant width vertically, Ac and ql are given byEquation 4 where Hf=the total, and Hfl=the leakoff height.
- Geology > Geological Subdiscipline > Geomechanics (0.68)
- Geology > Rock Type > Sedimentary Rock > Carbonate Rock (0.34)
- North America > United States > Texas > East Texas Salt Basin > Cotton Valley Group Formation > Cotton Valley Group Formation > Bossier Shale Formation (0.99)
- North America > United States > Texas > East Texas Salt Basin > Cotton Valley Group Formation > Bossier Sand Formation > Bossier Shale Formation (0.99)
- North America > United States > Louisiana > East Texas Salt Basin > Cotton Valley Group Formation > Cotton Valley Group Formation > Bossier Shale Formation (0.99)
- North America > United States > Louisiana > East Texas Salt Basin > Cotton Valley Group Formation > Bossier Sand Formation > Bossier Shale Formation (0.99)
- Well Drilling > Pressure Management > Well control (1.00)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Hydraulic Fracturing (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production logging (1.00)
Summary Based on experimental results, a model was developed to quantify gel propagation and dehydration during extrusion through fractures. To maximize gel penetration along fractures, the greatest practical injection rate should be used. On the other hand, in wide fractures or near the end of gel injection, gel dehydration may be desirable to form rigid gels that are less likely to wash out after placement. In these applications, reduced injection rates may be appropriate. Significant advantages could be realized for gels made with a polymer that has the largest available molecular weight. Introduction Gel treatments were often applied to improve conformance and reduce water or gas channeling in reservoirs. During placement of conventional gel treatments, a fluid gelant solution typically flowed into a reservoir through porous rock and fractures. After the blocking agent was placed, chemical reactions (i.e., gelation) caused an immobile gel to form. In contrast, for the most successful treatments in naturally fractured reservoirs, the time required to inject large volumes (e.g., 10,000 to 37,000 bbl) of gel was typically greater than the gelation time by a factor of 100. Thus, in these applications, formed gels were extruded through fractures during most of the placement process. A need exists to determine how much gel should be injected in a given application and where that gel distributes in a fractured reservoir. These parameters critically depend on the properties of gels in fractures. Therefore, we have a research program to determine these properties and to characterize gel placement in fractured systems. Previous Experimental Work. Previous work demonstrated that gels do not flow through porous rock after gelation. This behavior is advantageous because the gel is confined to the fractures; it does not enter or damage the porous rock. Thus, after gel placement, water, oil, or gas can flow unimpeded through the porous rock, but flow through the fracture is reduced substantially. However, extrusion of gels through fractures introduces new issues that are not of concern during placement of fluid gelant solutions. First, the pressure gradients required to extrude gels through fractures are greater than those for flow of gelants. For a Cr(III) acetate hydrolyzed polyacrylamide (HPAM) gel, the pressure gradient required for extrusion varied inversely with the square of the fracture width (Fig. 1). In previous works, we demonstrated that a minimum pressure gradient was required to extrude a given gel through a fracture. Once this minimum pressure gradient was exceeded, the pressure gradient during gel extrusion was insensitive to the flow rate. This behavior was attributed to a strong slip effect exhibited by the gel. A second concern is that gels can concentrate (dehydrate) during extrusion through fractures. Depending on fracture width (see Fig. 2), this dehydration effect can significantly retard gel propagation (e.g., by factors of up to 50). Figs. 1 and 2 apply to a 1-day-old Cr(III)-acetate-HPAM gel at 41°C. This same gel was used for most of the experiments described in this paper. Specifically, our experiments used an aqueous gel that contained 0.5% Ciba Alcoflood 935* HPAM (molecular weight ˜5×106 daltons with a degree of hydrolysis of 5 to 10%), 0.0417% Cr(III) acetate, 1% NaCl, and 0.1% CaCl2 at pH=6. All experiments were performed at 41°C (105°F). The gelant formulations were aged at 41°C for 24 hours (five times the gelation time) before injection into a fractured core. We designate this gel as our standard Cr(III)-acetate-HPAM gel. In an earlier work, we showed that when large volumes of gel were extruded through a fracture, progressive plugging (i.e., continuously increasing pressure gradients) was not observed. Effluent from the fracture had the same appearance and a similar composition as those for the injected gel, even though a concentrated, immobile gel formed in the fracture. The concentrated gel formed when water leaked off from the gel along the length of the fracture. The driving force for gel dehydration (and water leakoff) was the pressure difference between the fracture and the adjacent porous rock. During gel extrusion through a fracture of a given width, the pressure gradients along the fracture and the dehydration factors were the same for fractures in 650 md sandstone as in 50 md sandstone and 1.5 md limestone (see Figs. 1 and 2). Model 1. Previously, a simple model (Model 1) was developed that correctly matched the behavior during gel propagation and dehydration in a fracture with dimensions of 48×1.5×0.04 in. and an injection rate of 12.2 in./hr (200 cm/hr). This model assumed the following.Gel in the fracture existed in one of two forms: flowing gel that had the same composition and properties as the originally injected gel and concentrated, immobile gel. The flowing gel wormholed through the concentrated, immobile gel. The Darcy equation was valid for water flow through gel with a gel permeability to water ratio, kgel. The driving force for gel dehydration (and water leakoff) was the pressure difference between the fracture and the adjacent porous rock. The average distance that water traveled through the gel to reach the matrix was half the fracture width, wf /2. For a given length of fracture, the rate of water entering the fracture (in the form of gel) minus the rate of water leaving the fracture (again, tied up as gel) equaled the rate of water leakoff through the fracture faces (water mass balance). No crosslinked polymer entered the porous rock. Any gel that concentrated (dehydrated) immediately became immobile (crosslinked polymer mass balance). At any point in the fracture, the gel permeability to water, kgel, was related to the average gel composition by Eq. 1.Equation 1 In Eq. 1, kgel had units of md when the gel composition, C/Co, was expressed relative to the composition of our standard gel (i.e., 24-hr-old 0.5% Alcoflood 935 HPAM, 0.0417% Cr(III)-acetate). Originally, Eq. 1 was simply an empirical three-parameter fit that allowed the model to correctly quantify the rate of gel propagation through a 48×1.5×0.04-in. fracture. Since the original development of this model, we found independent support for two of the three parameters in Eq. 1 (i.e., the 1.0 md coefficient and the -3 exponent for the concentration term). However, no quantitative basis was found for the third parameter, 0.00011 md. As a qualitative explanation for Eq. 1, we speculate that the concentration-dependent term accounted for progressive dehydration of the concentrated, immobile gel, while the constant term accounted for the dehydration of flowing gel in the wormholes. At a given point in the fracture, the flowing gel was continually replenished, representing a gel source with an unchanging concentration. Any flowing gel that dehydrated was added to the reservoir of concentrated gel. In contrast, the concentrated gel did not move and became ever more concentrated with time, so its average permeability to water continually decreased.
- North America > United States > Wyoming > Wertz Field (0.99)
- North America > United States > Alaska > North Slope Basin > Prudhoe Bay Field (0.99)
- Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)
- Well Completion > Hydraulic Fracturing (1.00)
- Reservoir Description and Dynamics > Improved and Enhanced Recovery > Chemical flooding methods (1.00)