Improved Steam-Foam Formulations: Concepts and Laboratory Results
- H.C. Lau (Shell Development Co.) | J.K. Borchardt (Shell Development Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- November 1991
- Document Type
- Journal Paper
- 470 - 476
- 1991. Society of Petroleum Engineers
- 2.4.3 Sand/Solids Control, 1.8 Formation Damage, 2.5.2 Fracturing Materials (Fluids, Proppant), 5.3.4 Reduction of Residual Oil Saturation, 5.7.5 Economic Evaluations, 5.5 Reservoir Simulation, 5.4.1 Waterflooding, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 5.4.6 Thermal Methods, 5.7.2 Recovery Factors, 4.3.4 Scale, 5.4.4 Reduction of Residual Oil Saturation, 5.4.7 Chemical Flooding Methods (e.g., Polymer, Solvent, Nitrogen, Immiscible CO2, Surfactant, Vapex), 5.6.5 Tracers, 5.2.1 Phase Behavior and PVT Measurements
- 1 in the last 30 days
- 242 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
Kern River pilot results show that a steam-foam formulation based on analpha olefin sulfonate (AOS) containing 16 to 18 carbon atoms in thehydrocarbon chain improves sweep efficiency and oil recovery of the steamdrivebut propagates relatively slowly and leaves the same residual oil saturation(ROS) as steam. Research has identified potentially superior formulations thatform stronger foams, propagate faster, and/or reduce ROS below that to steam inthe laboratory.
Pilot Results for Base-Case Formulation Pilot Results for Base-CaseFormulation Shell Development Co. has conducted two steam-foam pilots in theKern River field. Foam was generated by continuous injection of 50%-qualitysteam containing 0.5 wt% (active matter) AOS 1618 and 4 wt% NaCl in the aqueousphase and 0.6 mol% nitrogen in the vapor phase. This formulation serves as ourbase case in this paper. paper. Extensive monitoring of subsurface data showedthat the foam increased the apparent viscosity of steam by a factor of 20 to 60near the injecters and allowed steam to contact oil in the lower part of thereservoir. This resulted in improved vertical sweep and hence improved oilrecovery. The base formulation had several limitations. First, although foampropagated substantial distances from the injector, the propagation rate wasrelatively slow. Production data showed a 2-year lag between the start of foaminjection and the major production response. This delay probably resulted fromsurfactant production response. This delay probably resulted from surfactantretention caused by ion exchange between formation clays and surfactantsolution. Second, surfactant utilization was poor--about 15 lbm/bbl incrementaloil. And finally, postfoam cores showed that ROS to steam foam was the same asto steam (about 10% PV).
Opportunities for Improvement
On the basis of the above observations, we identified three opportunitiesfor improving process performance: faster surfactant propagation rate,increased foam strength, and ROS reduction. The first propagation rate,increased foam strength, and ROS reduction. The first two modifications areaimed at speeding the delivery of foam (and hence steam) to the lower portionof the reservoir. If foam growth is limited by surfactant transport, fastersurfactant propagation means faster foam propagation. A stronger foam, on theother hand, reduces the amount of gravity override. The third alteration (ROSreduction) obviously increases displacement efficiency. But it also speeds theoverall process by leaving less oil into which the surfactant can partition,thus increasing the rate of surfactant and hence foam propagation. In thispaper, we summarize some of our efforts to find foam formulations that exploitone or more of the above opportunities and that ideally result in the improvedperformance shown schematically in Fig. 1.
Methodology for Evaluating Steam-Foam Formulations
Fig. 2 summarizes the methodology we used to evaluate candidate steam-foamformulations for each target reservoir. As indicated, the general approachentails a series of experiments, followed by numerical simulation (usingresults from the experiments) to generate a production function and economicanalysis (based on the production function) to forecast foam projectprofitability. The production function) to forecast foam project profitability.The following information is needed for each step in the procedure.
Experimental Work. The principal materials needed for the experiments aresamples of reservoir sand and crude oil. We also need to know the compositionof both connate and produced water, reservoir porosity, and steamdrivetemperature.
Simulation Studies. Key field data required for the simulation studies arepetrophysical data (PVT, oil and water viscosities, and relative permeability),reservoir geology, and the steamdrive history. From the experiments, we needsurfactant propagation rate, foam strength, and ROS reduction as functions ofsurfactant concentration. These data are fed into a steam-foam simulatorcalibrated with respect to our Kern River pilots. Following a successful matchof the steamdrive history of the target reservoir, the simulator is used toforecast oil production resulting from foam injection.
Economic Analysis. Besides the production and injection schedules obtainedfrom reservoir simulation, chemical (surfactant, NaCl, nitrogen, etc.) andsteam costs, oil price, and incremental capital are needed for economicevaluation. The profitability obtained from the economic analysis not onlygives an indication of the viability of a particular formulation but alsoprovides feedback for further formulation work.
Types of Experiments
As indicated in Fig. 2, three types of experiments provided input data forthe simulations. Procedures for these experiments are described below. Not allthe experiment types were conducted for each formulation studied. When aformulation failed to perform significantly well in one type of experiment(e.g., surfactant propagation), a decision was made whether to proceed withother propagation), a decision was made whether to proceed with other type ofexperiments to minimize the number of experiments.
High-Flow-Rate Steam-Foam Experiments. These experiments measured both foamstrength and ROS at the high steam flow rates representative of near-injecterconditions. They were conducted in 1-ft-long, 1.5-in.-diameter sandpacks heatedin an oven. Experiments were run with both Ottawa and reservoir sands andusually gave similar results at the high flow rates. At the beginning of eachexperiment, the sandpack was at ROS to 50%-quality steam. A foam formulationalso using 50%-quality steam was then injected into the sandpack. Both thepressure rise inside the pack and oil production were monitored. After about 15PV of surfactant injection, production were monitored. After about 15 PV ofsurfactant injection, the experiment was terminated and ROS was determined bysolvent extraction.
Surfactant Propagation Experiments. These experiments measured thesurfactant propagation rate through reservoir sands in the absence of a vaporphase. They were conducted in 1-ft-long, 1-in.-diameter sandpacks heated in anoven. Temperature ranged from 212 to 350 degrees F over the various runs. Atthe beginning of each experiment, the sandpack was at Sorw to water (25 to 30 %PV). Synthetic connate water was used in the waterflood. Surfactant solutionwas then injected at a rate of 1.6 ft/D. The effluent was collected infractions and analyzed for surfactant, divalent ion, and chlorideconcentrations. The chloride concentration served as a tracer for the aqueousphase. Surfactant retention was obtained by integrating the effluent chlorideand surfactant curves and calculating the difference. Results were expressed asa normalized surfactant propagation rate, which is defined as Sw plus retentionin Kern propagation rate, which is defined as Sw plus retention in Kern Riverbase case divided by Sw plus retention in improved formulation. In using thisformula, we assumed Sw=0.30 in the reservoir.
|File Size||2 MB||Number of Pages||7|