Understanding the physics behind which CO2 and water displaces oil in porous medium is of upmost importance for the success of any miscible CO2 WAG injection project. Coreflood experiments have to be designed efficiently to provide relevant information about complex three-phase flow mechanisms and ensure precise measurements of key reservoir parameters such as trapped gas saturation, CO2 relative permeability curves and hysteresis effects.
This paper presents an experimental study to investigate the effect of miscible cyclic CO2 injection on three phase relative permeability by performing a series of coreflood experiments in horizontal carbonate cores with light oil from a Middle East field. These experiments are designed using innovative data acquisition techniques to ensure data quality and allow accurate determination of reservoir microscopic properties. Produced volumes are monitored to provide material balance of each phase at both reservoir and laboratory conditions with a clear separation of flashed oil, condensate and gas. A full compositional analysis is then performed, using gas chromatography and liquid analysis of flashed oil and condensate, providing essential information for recovery calculation. Differential pressure across the core is also monitored for relative permeability estimation. Finally, in-situ saturation profiles are acquired in real time during experiments, using an X-Ray scanner capable of generating the two spikes of energy necessary for three-phase calculations. Redundancy in measurements, along with rigorous accounting of measurement uncertainties, allows a successful cross-check of all acquired data.
Methods of saturation estimation are discussed. Results show that for miscible corefloods, recovery strongly depends on how swelling and stripping mechanisms are accounted for.
For the first time ever a novel biopolymer is being field tested in a mature onshore oil field in Northern Germany. This biopolymer - named Schizophyllan after its producing fungus - shows excellent viscosifying efficiency and high tolerance towards harsh reservoir conditions such as high temperature, salinity and shear. Hence, this biopolymer allows for application in reservoirs that could so far not be easily polymer flooded and extends the operating envelope for polymer flooding to temperatures up to 130°C at arbitrary salt contents. Furthermore, it provides an environmentally friendly alternative to the commonly used synthetic polymers.
To run this first field test petroleum engineers of Wintershall Holding GmbH, chemical research scientists and biotechnologists of BASF SE are closely cooperating in an integrated team. An existing biotechnological plant at the BASF compound in Ludwigshafen, Germany, was partly reconstructed into a production facility for the biopolymer providing sufficient material for the field pilot. In the oil field itself an existing production site was extended into an operation site also hosting the surface facilities for treatment of the produced water used for re-injection and preparation of the injected biopolymer solution.
Being trucked to site as mother solution the Schizophyllan is mixed on-the-fly with the treated, high-salinity reservoir brine and injected into a newly drilled injector in the project area surrounded by three production wells. A comprehensive surveillance programme was set up comprising regular microbiological sampling, pressure monitoring using permanent down-hole gauges as well as frequent analyses and production tests to monitor the progress of the polymer trial.
The paper presents further information on the biopolymer Schizophyllan, describes the preparation and the setup of the field trial and summarizes results and experiences from the first year of the two-year field trial.
A large amount of hydrocarbon fluids in shale formations are stored within the organic matters where the pore sizes are in nanometer scales. Inside these nanopores, the fluid-wall interactions play such an important role that can change the phase behavior and transport mechanisms of the fluids. However, current available equations of state do not account for such interactions.
This work focuses on modifying Peng-Robinson equation of state (PR-EOS) using simplified local-density theory. First of all, density profiles of the hydrocarbon fluids are calculated along the pore diameter for difference pore sizes. From the density profile one can distinguish the regions of adsorbed phase, transition phase and bulk phase of the fluids. The phase behavior and transport properties of the fluids such as viscosity and diffusion coefficient are then calculated in each region from the modified PR-EOS using the averaged fluid properties in that particular region.
Our results showed that depending on fluid composition, either single layer or multilayer adsorption is presented in those nanometer pores near the pore wall. The pore size range we focused on was from 20 to 2 nm. When the pore size gets smaller and smaller, the absorbed layers at opposite pore walls are merged into one layer and result in the absence of the bulk fluid phase in the center areas of the pores. In such case all the fluids in the pore are under influence of the wall. For a synthetic mixture of 75% methane, 20% n-butane and 5% n-octane, the results indicate that for smaller pores, the bubble point and dew point pressures of the adsorbed hydrocarbon fluids are 50 to 500 psi lower than the bulk values and the two-phase region shrinks. Our analysis also showed that fluid viscosity started to increase significantly by approximately one fold when pore size is below 5 nm.
As results show, the hydrocarbon fluids under confinement tend to behave similar to dry gas. This reduces condensate banking and near-wellbore permeability impairment in comparison to conventional approaches. It has several implications for prediction of reservoir and well performance by implementing it into a numerical reservoir simulation package.
Steam Assisted Gravity Drainage (SAGD) is a commonly used EOR/IOR method for improving recovery in heavy oil reservoirs. However, a continued research for a more energy efficient method has led to the development of an improved version called Expanding Solvent (ES)-SAGD, which has a potential to replace conventional SAGD method in some heavy oil reservoirs. The paper provides some insights into determination of the reservoir performance of ES-SAGD process using a semi-analytical method. The results of the semi-analytical model are compared to numerical simulation results. This model is then used for optimizing the solvent requirement while minimizing the steam injected.
The semi-analytical model is determined by combining Butler’s oil drainage analytical models and VAPEX solvent dilution effect. This approach, with some modifications, is validated by the use of a 2D simulation. Numerical simulation of a short length scale process like solvent dilution in oil sometimes may give less accurate result prediction due to inherent numerical dispersion issues. Dynamic gridding based on both temperature and concentration change criterion is specified for the simulation. In addition, the effect of temperature on fluid interfacial tension that in turn affects capillarity at the vapor/oil interface at edge of steam chamber zone is studied. The viability of such temperature effect as an additional heat transfer mechanism is also demonstrated through this model.
Results for oil production rate and some key operation parameters are in good agreement between the semi-analytical model and the numerical simulation, rendering this model suitable for performing solvents-screening studies. However, there are some discrepancies in results between the simulation and semi-analytical method that can be attributed to grid orientation effects. Temperature induced capillarity is demonstrated to be a viable additional heat and mass transfer mechanism that affects net oil production significantly.
Numerical simulation of the ES-SAGD process is strongly affected by numerical dispersion that can artificially enhance mass transfer mechanisms thereby affecting the production. Finely gridded models can obviate this problem, but, at the expense of an extended computation time. Therefore, the approach discussed in this paper, eliminates the above-mentioned issues while giving a good estimate of key reservoir performance parameters.
The existence of an optimum injection rate for wormhole propagation, and face dissolution at low injection rates during matrix acidizing are well established. However, little has been documented that describes how the presence of residual oil affects carbonate acidizing. This study demonstrates the impact of oil saturation on wormholing characteristics while acidizing field and outcrop cores under reservoir conditions (200°F). Knowledge of the effect of different saturation conditions on acid performance will contribute towards designing more effective acid treatments.
Coreflood experiments at flow rates ranging 0.5 to 20 cm3/min were performed to determine the optimum injection rate for wormhole propagation when acidizing homogeneous carbonate and dolomite reservoir cores, and low permeability Indiana limestone cores of dimensions 3, 6, and 20 in. length and 1.5 in. diameter. Absolute permeability of the cores ranged from 1 to 78 md. The study involved acidizing cores saturated with water, oil, and water flood residual oil using 15 wt% HCl. The viscosity of the crude oil used was 3.8 cP @ 200°F. CAT scans were used to characterize wormholes through the cores. The concentrations of the dissolved calcium and magnesium ions were measured using Inductively Coupled Plasma-OES and the effluent samples were titrated to determine the concentration of the acid.
HCl was effective in creating wormholes with minimal branches for cores with residual oil (Sor=0.4-0.5) at injection rates 0.5 to 20 cm3/min. Compared to brine saturated cores, water flood residual oil cores took lesser acid volume to break through. Besides, the wormholing efficiency of regular acid improved with increasing acid injection rates in the presence of residual oil. Cores with residual oil after water flood showed no face dissolution at low acid injection rates. This is evident from the fact that at low injection rate, brine saturated cores measured the maximum calcium concentration in the effluent samples while cores with residual oil the least. Conclusions from this work aids better designing of acid jobs by highlighting the impact of oil saturation on wormholing characteristics of acid while acidizing carbonate rocks.
Lee, J.J. (University of Pittsburgh) | Cummings, S. (University of Pittsburgh) | Dhuwe, A. (University of Pittsburgh) | Enick, R.M. (University of Pittsburgh) | Beckman, E.J. (University of Pittsburgh) | Perry, R. (GE Global Research) | O'Brien, M. (GE Global Research) | Doherty, M. (GE Global Research)
The only known CO2 thickener (a compound that dissolves in CO2 and increases it viscosity significantly when present in dilute concentration) is a fluoroacrylate-styrene random copolymer that is probably too expensive for commercial application. High pressure CO2 has also been thickened via the dissolution of high molecular weight polydimethylsiloxane (PDMS, silicone oil) or polyvinyl acetate (PVAc), but this strategy requires several wt% of the polymer and the addition of large concentrations of an organic solvent to the CO2 (e.g. 20% toluene + 80% CO2), which is also impractical for commercial use. Because the utilization of high molecular weight polymers no longer appears to be a viable strategy for affordably thickening CO2 at EOR conditions, we are assessing the use of novel small molecules that self-assemble into viscosity-enhancing supramolecular structures in dense CO2. Small molecules can actually increase fluid viscosity just as effectively as high molecular weight polymers when compared at similar concentrations. For example, tributyltin fluoride and hydroxyaluminum di(2-ethyl hexanoate) are remarkable thickeners of light hydrocarbons even when present at concentrations well below 1wt%. In this presentation, three types of novel CO2 thickening candidates are designed, synthesized and assessed for solubility in CO2 and viscosity-increasing capabilities. Each small molecule possesses a “CO2-philic” segment that promotes dissolution in CO2; the CO2-philic segments are low-cost oligomeric versions of CO2-soluble polymers. Three different types of slightly “CO2-phobic” functional groups known to promote intermolecular associations in hydrocarbon and/or aqueous systems are also included in the thickener structure. The foremost challenge in the molecular design is selecting the appropriate type and number of associating groups needed to enhance viscosity, while not rendering the compound insoluble in CO2. A variety of prospective CO2 thickeners have been synthesized and the solubility of these candidates in CO2, and their ability to thicken CO2, will be presented.
Mo, Di (New Mexico Institute of Mining and Technology) | Jia, Bao (New Mexico Institute of Mining and Technology) | Yu, Jianjia (New Mexico Institute of Mining and Technology) | Liu, Ning (New Mexico Inst-Mining & Tech) | Lee, Robert (New Mexico Recovery Research Cen)
Description of the Material- This paper presents a series of tests of nanosilica-stabilized CO2 foam for waterflooded residual oil recovery under 1,200 psi¬-2,500 psi, 20oC-60oC, and in different core samples of sandstone, limestone, and dolomite. Total oil recovery by the CO2 foam, pressure drop across the core, and core properties were measured under different condition. The effects of different factors on the CO2 foam performance for oil recovery were discussed.
Application- Reservoirs factors such as pressure, temperature, and reservoir lithologies are critical for CO2 flooding design. The experimental results in this paper will provide a complete understanding of those factors’ effects on CO2 EOR performance and optimize the CO2 flooding design.
Results, Observations, and Conclusions- Nanosilica-stabilized CO2 foam was observed to improve oil recovery after waterflooding. The total oil recovery was measured as 35.8% when 5 PV CO2/nanosilica dispersion flooded the core under 1,200 psi and 20oC. With a further increase in pressure to 1,500 psi, 2,000 psi, and 2,500 psi, the total oil recoveries increased to 37.0%, 37.3%, and 39.9%, respectively. However, as the temperature increased from 25oC to 60oC, the total oil recovery by CO2 foam decreased from 39.6% to 31.3%. Permeability of the core was observed to not change after each test. Further tests with limestone and dolomite revealed that the total oil recovery by nanosilica-stabilized CO2 foam decreased compared with the oil recovery in sandstone. For example, the total oil recoveries were measured as 32.0% and 24.1% for limestone and dolomite when 5 PV CO2/nanosilica were injected into the cores. The core permeabilities were also observed to decrease after the tests for both limestone and dolomite.
Significance of Subject Matter- The results of the study described in this paper can help us to understand the effects of different factors on the efficiency of CO2 foam flooding for oil recovery, and also provide effective and economic solutions for CO2 EOR flooding.
Description of the material
We conduct steady-state CO2-brine flow experiments in 2-foot-long and 2.8-inch-diameter Berea sandstone cores at 20 °C and 1500 psi. Four pressure taps drilled on a core allow both the total pressure drop and the pressure drop across five individual sections to be independently measured. Water saturation in situ is measured with Computerized Tomography x-ray scanning. The CO2 and brine relative permeabilities are directly calculated with Darcy’s law.
Accurate determinations of CO2-brine relative permeability are important for modeling potential CO2 storage. This two-phase relative permeability also helps understand and investigate unsteady-state CO2-brine-oil three-phase relative permeability, which is vital in estimating and designing CO2 injection to enhance oil recovery.
Results, Observations, and Conclusions
(1) The relative permeability to both brine and CO2 of the last section (downstream, 15-cm long) is significantly smaller than that of any of the middle three sections. This testifies that the capillary end effect makes the relative permeability under-measured at the end of a core sample. (2) The values of the middle three sections are very close to each other, which indicate the middle part of our core is free of capillary end effect. (3) The CO2 end point relative permeability is 0.4~0.8, which is much higher than recent published measurements. (4) The brine end point relative permeability during imbibition is about 0.08, which is close to literature data.
Significance of subject matter
The most common assumption for CO2-brine relative permeability is that it is likely to be similar to oil-brine relative permeability for water-wet rocks. But recent measurements of CO2-brine relative permeability have differed greatly from oil-brine relative permeability; particularly, the measurements show a very low CO2 end point relative permeability (kr,CO2=0.1~0.2) and a relatively high residual water saturation (Swr>0.4) ( Lee et al. 2010, Zuo et al. 2012, Akbarabadi et al. 2013 and etc.). We hypothesize that the differences are caused by large capillary end effects resulted from the very low CO2 viscosity. Our experiments have observed this capillary end effect and measured CO2-brine relative permeability data, which are free of capillary end effect and similar to oil-brine relative permeability.
The Bakken reservoirs represent a large untapped resource of oil. One of the challenges is extracting these crude oils from their low-permeability formations at economic rates. These reservoirs contain multiple pay zones, including some with carbonate (dolomite) lithology. Recent common practice is to drill horizontal wells and perform a series of large, multi-stage hydraulic fracture treatments. The fractures penetrate deeply into the reservoir and promote more efficient drainage of the oil.
This situation suggests that a surfactant technology designed to enhance oil recovery from fractured carbonate formations is a fit for these typical Bakken cases with Middle Layer complex fractured lithology. The concept of this technology is to incorporate appropriate surfactant formulations at a low dosage in the well stimulation fluids. If properly designed, such additives in the fracture fluids will penetrate into the highly oil-saturated matrix or natural fracture region and accelerate the extraction of the oil in place by rapid imbibition. This extracted oil can readily move from the matrix into the propped fracture system for production. Another benefit of the additive is its engineered property to leave the matrix or nature fracture face water-wet, facilitating oil movement during production.
This paper presents a study of a series of such stimulation fluid additives developed for enhanced oil recovery. Over 10 of special customized product blends were evaluated in laboratory for their effectiveness in increasing recovery of Bakken crude oil samples. Tests included compatibility with formation brine, surface tension and interfacial tension, wettability alteration, emulsion tendency, recovery factors from spontaneous imbibition with crude oil and formation brine, and compatibility with proposed fracturing fluids. These results show that more than one of these products improve recovery of Bakken crude oil by spontaneous imbibition from both outcrop limestone cores and from Bakken core material. The best of these products is recommended for field application.
The Guando field is a late Cretaceous sandstone with two distinct reservoirs with total gross thickness of approximately 1300 feet. Discovered in 2000, the field is Colombia’s largest oil discovery since the mid 1980’s. Due to the low initial reservoir pressure, water injection began almost simultaneously with field development. Water is injected selectively into the main reservoir´s main sands in order to optimize the distribution of injected water. However, oil recovery efficiency is challenged by reservoir heterogeneity, including natural fractures.
The first polymer gel conformance application in Colombia was implemented in 2008 in an effort to improve waterflood sweep efficiency. The objective of the gel treatments was to reduce the permeability in the most conductive natural fractures. The conformance pilot included two non-adjacent patterns. Initially, the design team considered isolating the upper reservoirs and treating only the lower reservoirs. However, in the final design both treatments were “bullheaded” with all zones open to gel injection. The treatment designs included several stages of varying gel concentrations and injection rates, which were modified in the course of each treatment application based on surface pressure response.
Post treatment oil response occurred within approximately two months and payout was achieved in less than six months, driving new candidate selection and treatment in 2013. The paper will discuss the reservoir diagnostics that led to the selection of polymer gel as the most effective solution to severe water channeling. Also, we will present plans for additional gel treatments with an emphasis on cost reduction and improved treatment designs.
Polymer gels have been successfully applied in naturally fractured reservoirs for almost two decades, however, reservoir characterization tools continue to evolve, providing more precise diagnostics that offer improved conformance treatment designs. This case history will provide an updated framework for operators considering chemical sweep improvement technologies as part of an integrated field management strategy.