Sanders, Aaron (Dow Chemical Co.) | Jones, Raymond Michael (Dow Chemical Co.) | Rabie, Arwa (Dow Chemical Co.) | Putra, Erwinsyah (Kinder Morgan CO2 Co. LP) | Linroth, Mark A. (Kinder Morgan Inc) | Nguyen, Quoc Phuc (University of Texas At Austin)
A new conformance solution has been developed to help producers reach and recover more oil from existing wells while simultaneously reducing operating expenditures. This new technology successfully addresses shortcomings associated with existing miscible gas EOR (enhanced oil recovery) processes by generating foams of supercritical CO2 fluid and water in the formation which alter mobility and improve vertical conformance. Ultimately, the process can improve the profitability of existing wells by enhancing reservoir sweep efficiency, thereby allowing access to previously bypassed oil. This report describes the technology development process and documents successful field trial results from Kinder Morgan's SACROC field.
Operators are demanding solutions that maximize recovery from existing reservoirs, capturing the high value of remaining oil and avoiding the risk and high costs associated with new exploration. A new conformance solution has been developed to help producers reach and recover more oil from existing wells while simultaneously reducing operating expenditures. This report describes an experimental field trial of a new technology to address shortcomings associated with existing miscible gas EOR processes by generating foams of supercritical CO2 fluid and water in the formation to alter mobility and improve vertical conformance. The history match and foam modeling work for the project are discussed, including the forecasted production curves derived from the optimization of the injection strategy.
A field trial was carried out in west Texas at the Scurry Area Canyon Reef Operational Committee (SACROC) field, which is owned and operated by Kinder Morgan LLC. The purpose of the program was to use surfactant injection in the CO2 phase to create a CO2-in-water emulsion or foamto improve vertical conformance, and create in-depth mobility control. The project was funded independently by The Dow Chemical Company and Kinder Morgan LLC and was not part of a government subsidized program or grant.
The SACROC field had an estimated original oil in place (OOIP) of 2.8Billion barrels, when it was discovered in 1948 (Mathis 1964). Primary production and a successful waterflood yielded 1.2B barrels of oil from the 1950's through the 1970's (Kane 1979). Aspects of the field's history have been covered previously including: field characteristics, CO2 project planning and design, reservoir and fluid properties, and this paper will only briefly discuss aspects as they are relevant to the current discussion (Simon and Fesmire 1977, Crameik 1972, Chaback and Williams 1988, Swulius 1986, Gill 1982). The SACROC field has been under CO2 flood since 1972. Production in the field had been on a steady decline prior to being purchased by Kinder Morgan in 2001, when its production had decreased to ~8500 b/d. Through the use of a miscible CO2 5-spot pattern injection program, Kinder Morgan was able to increase production to ~32,000 b/d by 2002 and is currently producing 28,000 BOPD) (Brnak and Petrich et al. 2006).
This paper establishes Low-Tension-Gas (LTG) as a method for sub-miscible tertiary recovery in tight sandstone and carbonate reservoirs. The LTG process involves the use of a low foam quality surfactant-gas solution to mobilize and then displace residual crude after waterflood. It replicates the existing Alkali-Surfactant-Polymer (ASP) process in its creation of an ultra low oil-water interfacial tension (IFT) environment for oil mobilization, but instead supplements the use of foam over polymer for mobility control. By replacing polymer with foam, chemical Enhanced Oil Recovery (EOR) methods can be expanded into sub-30 mD formations where polymer is impractical due to plugging, shear, or the requirement to use a low molecular weight polymer.
The proposed strategy is tested through low-quality, low rate co-injection of nitrogen and a slug/drive surfactant solution. Results indicate favorable mobilization and displacement of residual crude oil in both tight carbonate and tight sandstone reservoirs. Tertiary recovery of 75-90% ROIP was achieved for cores with 2-15 mD permeability. Consistent with successful ASP floods typically observed in high permeability rocks, a large oil bank was observed at the effluent before the production of Windsor Type III and Type II(-) microemulsion. High LTG tertiary recovery is contrasted with results from reference surfactant (no gas) flooding (28% ROIP tertiary recovery) and immiscible gas co-injection (no surfactant) flooding (28% ROIP tertiary recovery).
Additionally, high initial oil saturation was tested to determine process tolerance to oil and evaluate potential for application during secondary recovery. During flooding at initial oil saturation (1-Swi), LTG injection achieved recovery of 84% of OOIP with similar fractional flow, mobility, and other process attributes to those exhibited during LTG tertiary flooding. This reduces the risk that in-situ oil may cause unfavorable displacement due to destabilization of liquid lamellae which provide mobility control by creation of a dispersed gas phase (foam). Potential application at secondary recovery is suggested which would improve reserve capture and reduce high pressure gradients typically associated with flooding tight reservoirs.
To better understand the physical mechanisms which impact mobilization and displacement, early production of an elongate oil bank at reduced fractional flow of oil was shown to be an attribute of high crude oil relative mobility and low pore volume available to mobile oil. This should favorably impact economics during chemical flooding by accelerating production of an oil bank. Next, by application of salinity as a conservative tracer and oil material balance, gas saturation during LTG floods was calculated to be 18-22%. This is contrasted with gas saturation during co-injection of 5% and indicates that a large dispersed gas phase was present during LTG flooding and is consistent with stable lamellae production during flooding. Finally, by comparing effluent salinity profiles across floods, qualitative understanding of dispersion and macroscopic stability is developed. Plots show a reduction in dispersion for LTG flooding versus surfactant flooding, which indicates improvement in mobility ratios across the displacement fronts.
Macroscopic stability of displacement fronts was studied via pressure derived mobility ratios. Approximate parity of relative mobility of injected fluids was observed with respect to relative mobility of displaced water at true residual oil saturation and interpreted relative mobility of a formed oil bank. These results indicate that in-situ foaming was present which enabled mobility control, and that stable displacement of in-situ fluids was achieved during flooding.
The potential application of nanoparticle dispersions as formation stimulation agents, contrast agents or simply as tracers in upstream oil and gas industry requires knowledge of the flow properties of these nanoparticles. The modeling of nanoparticle transport in hydrocarbon reservoirs requires a comprehensive understanding of the rheological behavior of these nanofluids. Silica nanoparticles have been commonly used because of their low-cost fabrication and cost-effective surface modification. The aqueous silica nanoparticle dispersions show Newtonian behavior under steady shear measurements controlled by a rheometer as discussed by Metin et al. . The viscosity of nanoparticle dispersions depends strongly on the particle concentration and that this correlation can be depicted by a unified rheological model . In addition, during flow in permeable media, the variation of shear associated with complex pore morphology and the interactions between the nanoparticles and tortuous flow channels can affect the viscosity of nanoparticle dispersion. The latter is particularly important where the concentration of nanoparticles in dispersion may change because of nanoparticle adsorption on mineral/fluid and oil/water interfaces or by mechanical trapping of nanoparticles. In this paper, the flow of silica nanoparticle dispersions through different permeable media is investigated. The rheological behaviors of the dispersions are compared with those determined using a rheometer. We established a correlation between the nanoparticle concentration and dispersion viscosity in porous media for various nanoparticle sizes. The effects of pore structure and shear rate are also studied. We have confirmed that the concept of effective maximum packing fraction can be applied to describe the viscosity of aqueous nanoparticle dispersions in both bulk flow and flow in porous media with high permeability and regular pore structures but not at low permeability because of mechanical trapping. Our work provides new insight to engineering nanoparticle rheology for subsurface applications.
As much of the oil in the Akal field of the Cantarell complex is contained in the low permeability oil wet matrix, foam injection has been proposed as a method to control fluid mobility in the fracture, with the possible added benefit of transporting surfactant into the matrix so that additional oil could be liberated through a reduction of interfacial tension between oil and water (if this effect is significant for the surfactant in question). Presented in this paper is the work flow undertaken during an extensive study of all available laboratory experiments and pilot single well foam injection tests. Laboratory experiments ranged from simple water plus surfactant imbibition tests and surfactant flooding tests, to more complex foam flooding in split core experiments and co-injection of surfactant and gas for generation of foam in-situ. There were three field pilot single well foam injection tests that were included in this analysis that were of the huff-and-puff design.
This extensive analysis was done with the aid of numerical simulation that resulted in the development of a novel foam model that handles both mobility control and interfacial tension reduction effects, and is capable of simulating foam degradation, foam regeneration, and trapped foam phenomena. Previous foam models available in commercial numerical simulators were not capable of simulating all of these foam effects together. It is shown that with identical foam parameters, this model matches all laboratory core flood studies as well as the field pilot tests, showing that this foam model is capable of predicting foam performance in both laboratory and field settings. The foam components can be chosen to be defined as either gaseous or aqueous components and this choice is shown to affect the impact of capillary pressure on foam flow into the matrix.
Also discussed in this paper are details of how the foam behaves when injected into a gas saturated zone where the foam combines with in-situ gas, resulting in higher foam qualities than was injected. It is demonstrated that foam mobility control as a function of foam quality is an important aspect for matching field performance. The significance of correct foam density calculations is also discussed using field scale models.
The work done to match the many laboratory and field scale foam tests resulted in a significant improvement of the understanding of foam degradation, regeneration, permeability blockage, and flow in porous media and the phenomena responsible for generating incremental oil.
In mechanistic modeling of foam in porous media, reduced gas mobility is attributed to viscous resistance of flowing foam lamellas to gas flow, while gas trapping significantly modifies relative permeability. By using pore-network models representative of real porous media, we previously developed a relationship between flowing gas fraction and pressure gradient for strong foam (high lamella density). In this study, we expand our model to describe the effects of foam strength and pore-scale apparent gas viscosity models on both relative gas permeability and effective gas viscosity. Dimensional analysis in scaling of these two rheological quantities with pressure gradient and lamella density is discussed.
One of our important findings is that relative gas permeability is poorly sensitive to total lamella density while it is a strong non-linear function of flowing gas fraction, opposing to most of the existing theoretical models describing the effect of gas trapping on relative gas permeability. This is consistently observed for all the pore-scale apparent gas viscosity models. It is also found that effective gas viscosity increases exponentially with flowing lamella density. This result implies that the use of the commonly used apparent gas viscosity model for straight capillary tubes is not accurate for foam flow in porous media. In addition, shear thinning foam flow is more obvious at high flowing lamella density while Newtonian flow becomes significant at relatively low flowing lamella density. Furthermore, scaling of effective gas viscosity with flowing lamella density depends on how the later quantity is defined. Both empirical and mechanistic pore-scale apparent gas viscosity models give almost the same functional relationship between flowing gas fraction and pressure gradient. This would facilitate scaling of flow rate with pressure gradient and testing a range of shear-thinning and yield-stress behavior in a simple format. Our results necessitate the need for further improving the existing mechanistic foam modeling methods with focus on process upscaling.
Chen, Yunshen (U Of Texas At Austin) | Elhag, Amro S. (U Of Texas At Austin) | Poon, Benjamin M. (U Of Texas At Austin) | Cui, Leyu (Rice U.) | Ma, Kun (Rice University) | Liao, Sonia Y. (U Of Texas At Austin) | Omar, Amr (U Of Texas At Austin) | Worthen, Andrew (U. of Texas at Austin) | Hirasaki, George J. (Rice University) | Nguyen, Quoc Phuc (U. of Texas at Austin) | Johnston, Keith P. (U Of Texas At Austin)
Despite significant interest in CO2 foams for EOR, very few studies have reported stable foams at high temperatures and high salinities, which are often encountered in the Middle East and elsewhere. Stable CO2/water (C/W) foams at high temperatures up to 120 oC and salinities have been achieved with ethoxylated cationic surfactants. The surfactants were shown to stabilize C/W foams with high salinity brine with NaCl concentration up to 182 g/L at 120 °C, 3400 psia, and to form unstable dodecane/water emulsions with the 120 g/L NaCl brine solutions. Thus, the foams have the potential to provide mobility control to prevent loses of CO2 in high permeability regions, but simultaneously allow high permeability in the presence of residual oil. The surfactants are soluble in CO2 and thus may be injected in the CO2 phase to simplify the EOR process. The aqueous solubility of the surfactant at high temperatures is enhanced with the appropriate number of EO groups on the amine head group. Viscosities of high-pressure C/W foams (emulsions) formed with these surfactants were investigated by capillary rheology. These hybrid cationic/nonionic surfactants combine the high cloud points of ionic surfactants with high solubilities in CO2 of nonionic surfactants. Furthermore, the variation of the tail length and the degree of ethoxylation offer great flexibility for stabilizing CO2 foams for EOR at high temperatures and high salinities. Ethoxylated cocoamine exhibited lower adsorption on calcite than that on dolomite, given the presence of silica sites in the latter. High divalent ion concentrations in 22% total dissolved solids (TDS) brine contributed to the reduction of surfactant adsorption on silica sites in the dolomite powder.