Zhang, Yandong (Missouri University of Science and Technology) | Wei, Mingzhen (Missouri University of Science and Technology) | Bai, Baojun (Missouri University of Science and Technology) | Yang, Hongbin (China University of Petroleum) | Kang, Wanli (China University of Petroleum)
Enhanced oil recovery (EOR) processes are regarded as important methods to recover remaining oil after primary and secondary recovery. It is important to select the most appropriate EOR process among the possible techniques for a candidate reservoir. Therefore, EOR screening criteria have been constructed using available EOR data sets and serve as the first step to compare the suitability of each EOR method for a particular reservoir. Most screening criteria for polymer flooding are based on data sets from EOR surveys published biannually by the Oil & Gas Journal. These surveys missed significant polymer flooding parameters such as formation water salinity and hardness, polymer types and molecular weight, polymer concentration, reservoir heterogeneity, and so on. All of these topics are covered in this paper with data from relevant literature and records provided by oil companies in China.
Polymer flooding has been widely applied in China for over 20 years and a large number of pilot and field projects have been conducted. These projects include important information to quantify the development of polymer flooding as an EOR method. Nevertheless, most of them have been published in Chinese, and are not accessible to the global research community due to the language barrier. This paper represents an effort to collect all relevant information of polymer flooding from available Chinese publications and reports from all of the major oil companies in China. The primary objectives of this survey is to reveal EOR advances, to trace the development of the polymer flooding EOR methodology in China, and to benefit EOR worldwide.
This project collected information on 55 polymer flooding projects after reviewing nearly 200 publications in Chinese, including 31 pilot projects and 24 field projects from 1991 to 2014. A data set was constructed by collecting all relevant information for polymer flooding. Statistical analyses and graphical methods were used to analyze the whole data set. Box plots combined with violin plots were used to show the distribution and the range of each parameter. By defining and calculating lower and upper limits in box plots, special projects were identified and explained. Scatter plots, which show multiple parameters in one plot, were used to identify significant relationships among different parameters, especially for dependent parameters. This method overcomes some disadvantages of the range method, which is traditionally used for EOR screening. For example, using polymers with high concentration in low salinity reservoirs can lead to higher incremental oil recovery than in high salinity ones, and lower permeability usually correlates with the use of polymers with lower molecular weight. However, the traditional range method cannot show this relationship. Finally, comprehensive screening criteria for polymer flooding were updated based on information revealed in the field application projects.
This paper addresses two questions for polymer flooding. First, what polymer solution viscosity should be injected? A base-case reservoir-engineering method is present for making that decision, which focuses on waterflood mobility ratios and the permeability contrast in the reservoir. However, some current field applications use injected polymer viscosities that deviate substantially from this methodology. At one end of the range, Canadian projects inject only 30-cp polymer solutions to displace 1000-3000-cp oil. Logic given to support this choice include (1) the mobility ratio in an unfavorable displacement is not as bad as indicated by the endpoint mobility ratio, (2) economics limit use of higher polymer concentrations, (3) some improvement in mobility ratio is better than a straight waterflood, (4) a belief that the polymer will provide a substantial residual resistance factor (permeability reduction), and (5) injectivity limits the allowable viscosity of the injected fluid. At the other end of the range, a project in Daqing, China, injected 150-300-cp polymer solutions to displace 10-cp oil. The primary reason given for this choice was a belief that high molecular weight viscoelastic HPAM polymers can reduce the residual oil saturation below that expected for a waterflood or for less viscous polymer floods. This paper will examine the validity of each of these beliefs.
The second question is: when should polymer injection be stopped or reduced? For existing polymer floods, this question is particularly relevant in the current low oil-price environment. Should these projects be switched to water injection immediately? Should the polymer concentration be reduced or graded? Should the polymer concentration stay the same but reduce the injection rate? These questions are discussed.
Polymer flooding is a widely used commercial process with a low cost per barrel of produced oil, For this application, hydrolyzed polyacrylamide (HPAM) polymers are the most widely used type of polymer. In an era of low cost oil, it is becoming even more essential to optimize the polymer flooding design under realistic reservoir conditions. The objective of this research was to better understand and predict the behavior of HPAM polymers and their effect on residual oil saturation, in order to improve the capability of optimizing field design and performance. The corefloods were performed under typical field conditions of low pressure gradients and low capillary numbers. The polymer floods of the viscous oils recovered much more oil than the water floods, with up to 24% lower oil saturation after the polymer flood than the water flood. The experimental data are in good agreement with the fractional flow analysis using the assumptions that the true residual oil saturations and end point relative permeabilities are the same for both water and polymer. This suggests that for more viscous oils, the oil saturation at the end of water flood (i.e. at greater than 99% water cut) is better described as ‘emaining’ oil saturation rather than the true ‘esidual’ oil saturation. This was true for all of the corefloods regardless of the core permeability and without the need for assuming a permeability reduction factor in the fractional flow analysis.
Water-based polymers are often used to improve oil recovery by increasing displacement sweep efficiency. However, recent laboratory and field work has suggested these polymers, which are often viscoelastic, may also reduce residual oil saturation. The objective of this work is to investigate the effect of viscoelastic polymers on residual oil saturation in Bentheimer sandstones and identify conditions and mechanisms for the improved recovery. Bentheimer sandstones were saturated with a heavy oil (120cp) and then waterflooded to residual oil saturation using brine followed by an inelastic Newtonian fluid (diluted glycerin). These floods were followed by injection of a viscoelastic polymer, hydrolyzed polyacrylamide (HPAM).
Significant reduction in residual oil was observed for all core floods performed at constant pressure drop when the polymer had significant elasticity (determined by the dimensionless Deborah number,
Zhu, Youyi (State Key Laboratory of Enhanced Oil Recovery, Research Institute of Petroleum Exploration and Development, CNPC) | Fan, Jian (State Key Laboratory of Enhanced Oil Recovery, Research Institute of Petroleum Exploration and Development, CNPC) | Liu, Xiaoxia (State Key Laboratory of Enhanced Oil Recovery, Research Institute of Petroleum Exploration and Development, CNPC) | Li, Jianguo (State Key Laboratory of Enhanced Oil Recovery, Research Institute of Petroleum Exploration and Development, CNPC)
Chemical flooding technology is one of the effective enhanced oil recovery (EOR) methods for high water cut sandstone reservoirs with either medium and/or high permeability. Because of the small pore throat radius in the pore medium of low permeability reservoir, high molecular weight polymers cannot be injected in the low permeability reservoir. Therefore, many traditional chemical floodings (such as polymer flooding, alkali-surfactant-polymer (ASP) flooding and surfactant-polymer (SP) flooding) cannot be effectively applied in this case. Small-molecule viscoelastic surfactant (VES) has special rheological properties in porous medium. It showed both viscosified function and reduction of oil/water interfacial tension (IFT) performances under certain conditions, thereby providing the possibility of IOR/EOR potential application in low permeability reservoirs.
Most of reservoirs in Jilin Oilfield belong to low permeability reservoirs with permeability of around 50 mD in average. The recovery percent of reserves in Fuyu was only 23% by water flooding with water cut as high as 93%. A candidate EOR technique with chemical flooding has been proposed. Studies on VES flooding EOR methods targeting this reservoir condition were conducted. The rheological property, IFT property, viscosifying ability of VES and core flooding experiments of VES system were studied.
From VES screening experiment, a type of zwitterionic betaine surfactant with long carbon chain was selected. It showed viscosifying behavior, shear thinning property and low IFT performances at reservoir conditions. VES of EAB solutions showed a good viscosifying action at low surfactant concentration. Moreover, based on its shear thinning property under the wide shear rate conditions, VES exhibited a good injectivity performance. IFT between crude oil and formation water with EAB was 10-3-10-2 mN/m order of magnitudes. The results could be obtained at the concentration ranges of surfactants from 0.1wt% to 0.4wt%. Ultralow IFT (10-3 mN/m order of magnitudes) could be obtained in the presence of co-surfactants or alkalis (such as sodium carbonate). Core flooding experiments of VES flooding showed that the incremental oil recovery factors could reach up to 13%-17% over conventional water flooding at Fuyu reservoir conditions. Test results indicated that VES flooding might become a promise alternative EOR method for low permeability reservoir after water flooding.
In contrast to the complexity of ASP/SP combination system, VES flooding could avoid chromatographic effects in the reservoir based on their simple formula (single surfactant compound). This new chemical flooding technique might have a great potential for EOR application in the low permeability reservoirs.
It is now common knowledge among EOR practitioners that the combination of ferrous iron (Fe2+) and oxygen causes severe oxidative degradation to EOR polymers, resulting in a lowering of molecular weight and hence a loss of viscosity. During the design of polymer flooding projects, an important question is thus the acceptable levels of Fe2+ and dissolved oxygen that can be tolerated in injection water specifications. Furthermore, we would like to be able to predict the extent of degradation in the case of excess Fe2+ or oxygen ingress.
However, despite over fifty years of research and a general understanding of the degradation mechanism involved, quantitative prediction of the extent of degradation has proven elusive and dependent on the measurement protocol. This is likely due to the fastidious experimental protocols required to work under anaerobic or limited-oxygen conditions.
We examine existing protocols and demonstrate that experiments in which either Fe2+ or oxygen are the limiting reagent yield equivalent results when the stoechiometry of the Fe2+ oxidation reaction with oxygen is taken into account. Based upon these findings, a novel, easy approach is proposed to quantify polymer oxidative degradation as a function of either dissolved oxygen or Fe2+ content.
The limits of 225 ppb Fe2+ and 32 ppb dissolved oxygen are fixed for Flopaam 3630S in 6 g/l brine in the concentration range 500-1500ppm in order to ensure degradation of low-shear plateau viscosity does not exceed 10%. Higher levels will lead to severe polymer degradation. The influence of polymer concentration, temperature and salinity is also investigated. At last, evolution of redox potential and pH during Fe2+ oxidation are discussed.
There is a direct practical application of these finding for the design of surface facilities for polymer dissolution and transport and for the prediction of degradation in case of oxygen ingress. Moreover, a simple and easily performed protocol is proposed for the evaluation of polymer oxidative degradation.
Lee, Jason (University of Pittsburgh) | Dhuwe, Aman (University of Pittsburgh) | Cummings, Stephen D. (University of Pittsburgh) | Beckman, Eric J. (University of Pittsburgh) | Enick, Robert M. (University of Pittsburgh) | Doherty, Mark (GE Global Research) | O'Brien, Michael (GE Global Research) | Perry, Robert (GE Global Research) | Soong, Yee (US DOE NETL) | Fazio, Jim (US DOE NETL) | McClendon, Thomas R. (US DOE NETL)
CO2 miscible and immiscible displacements and hydrocarbon miscible floods are commonly plagued by low volumetric sweep efficiency, early gas breakthrough, high gas utilization ratios, and significant gas re-compression and recycle. Rather than addressing these problems via the water-alternating-gas (WAG) injection sequence that reduces gas relative permeability or the generation of gas-in-brine foams for reduced mobility, we propose increasing the viscosity of high pressure CO2 or NGL via the dissolution of dilute concentrations of thickening agents.
There are two strategies for increasing the viscosity of high pressure fluids; the dissolution of ultrahigh molecular weight polymers or associating polymers, or the dissolution of small molecules that self-assemble in solution to form viscosity-enhancing linear or helical supramolecular structures. Ideally a very small amount of the thickener will be required (roughly 0.1wt%) to elevate the CO2 or NGL viscosity to the same value as the oil being displaced (typically a 10-100 fold increase). Further, the thickened CO2 or thickened NGL should be a stable, transparent solution that does not require a heating/cooling cycle for viscosity enhancement to occur.
Thickener solubility and viscosity were determined over a 25-100oC range. Each of the three major NGL constituents (ethane, propane and butane) was thickened with an ultrahigh molecular polymer (commercial drag reducing agent), resulting in a 2-30 fold increase in viscosity at polymer concentrations of 0.5wt% or less. The polymer dissolved at the lowest pressure in butane and was most effective as a thickener in butane.
Three small molecule thickeners were identified for the NGL constituents; tri-alkyl-tin fluoride, hydroxyaluminum disoap, and a phosphate ester-crosslinker mixture. Remarkable viscosity enhancements were attained for propane and butane with the tri-alkyl-tin fluoride and aluminum soap; the crosslinked phosphate ester solutions exhibited modest viscosity increases. Only tri-alkyl-tin fluoride thickened ethane.
CO2 thickeners were assessed with a falling ball viscometer and pressure drop associated with flow through Berea sandstone. 4-5 fold increases in viscosity were attained with 1wt% of a high molecular weight polyfluoroacrylate. 3-4 fold increases in viscosity were attained with 1wt% high molecular weight polydimethyl siloxane, but a very large amount of toluene co-solvent was required. Although a remarkably effective small molecule thickener was designed for CO2 (100-fold increase at 1.3wt%), it required a heating/cooling cycle and a very large amount of hexane co-solvent.
We have identified the first polymeric and small molecule thickeners ever reported for ethane. Further, this study presents the largest viscosity increases ever reported for propane and butane with polymers and small molecule thickeners. We have presented the most effective polymeric thickeners for CO2 reported to date. This paper also summarizes numerous molecular architectures that are not viable for CO2 and highlights the most promising compounds that continue to be refined.
Solvent Aided-Steam Flooding (SA-SF) focuses on maximizing the oil production by reducing the economic and environmental challenges created by steam generation. However, the solvent selection is vital due to the interaction of solvents with asphaltenes. Moreover, the polar nature of asphaltenes also enables asphaltene-steam interaction which may result in emulsion formation. This study investigates solvent-asphaltene-steam interaction during SA-SF with low and high molecular weight asphaltene insoluble solvents.
Two different solvents were tested; n-hexane (E1 and E4) and a commercial solvent (CS) (E2 and E5) with four flooding experiments; two miscible flooding (E1 and E2) and two SA-SF (E4 and E5) experiments. Results were compared with steam flooding (E3) experiment. The performance evaluation of different enhanced oil recovery methods was accomplished by comparing the oil recovery rates. The asphaltene content of produced oil samples was determined by standard methods. The asphaltene-steam interaction was analyzed with microscopic images, and the water content of produced oil samples was measured by Thermogravimetric Analysis (TGA).
Even though similar cumulative oil productions were obtained by the end of E1 (n-hexane-flooding) and E2 (CS-flooding), the produced oil quality varied due to asphaltene and clay contents. While higher clay content was measured for E1, E2 had a lower quality, due to higher asphaltene contents. This finding is due to the heavy dearomatized hydrocarbons composition of the CS which ranges from C11 up to C16 and enables more asphaltene production. Even though, E5 yielded the highest liquid production among all experiments; the produced liquid was composed of emulsified oil. The solvent aided-steam flooding (SA-SF) experimental results, which have been conducted with n-hexane/steam (E4) and CS/steam (E5) injections, suggest that as the asphaltene content increases in produced oil samples, more hard-to-break emulsions are formed. The unusual stability of these emulsions can be attributed to the nature of the asphaltene present in the produced oil.
From the results presented, it is recommended the use of lower carbon number solvents to leave the larger amounts of asphaltenes in the reservoirs. The solvents differed in their interactions with the asphaltenes present in the oil and with the steam that has a direct impact not only on the quantity of oil produced but the quality as well. Hence, the wise selection of the appropriate solvent cannot be ignored during solvent aided-steam flooding processes.
Davidson, Andrew (Chevron Energy Technology Company) | Nizamidin, Nabijan (Chevron Energy Technology Company) | Alexis, Dennis (Chevron Energy Technology Company) | Kim, Do Hoon (Chevron Energy Technology Company) | Unomah, Michael (Chevron Energy Technology Company) | Malik, Taimur (Chevron Energy Technology Company) | Dwarakanath, Varadarajan
Low microemulsion viscosity is critical for the success of chemical EOR. Typical microemulsion viscosities are measured using a rheometer and are considered to be static measurements. Given that microemulsions have a propensity to show non-Newtonian behavior, static viscosity measurements are not scalable to dynamic viscosities observed in cores and hence difficult to scale-up to field designs using simulations. We present a technique to measure dynamic microemulsion viscosity using a modified two-phase steady state relative permeability setup. Such dynamic viscosities provide a more practical feel for microemulsion viscosity under reservoir conditions in the pores and allow for selection of low microemulsion viscosity formulations. A two-phase steady state relative permeability setup was used with continuous co-injection of oil and surfactant. A glass filled sand pack was used as a surrogate core and the injection fluids were allowed to equilibrate into the appropriate phases as determined by the phase behavior. For the rapidly equilibrating and low viscosity Winsor Type III formulations three phases are clearly observed in the sand packs. Using the phase cuts in the sand pack/effluent and the known oil and water viscosities, we can estimate the microemulsion viscosity. Both low and high viscosity formulations were tested in corefloods and oil recovery measured to illustrate the importance of low viscosity microemulsions for oil recovery. As expected, the low viscosity microemulsions correlated with higher oil recovery. In addition, the equilibration times to reach Winsor Type III microemulsions were also linked to better oil recovery. For the well behaved formulations that equilibrated in less than 2 days the static microemulsion viscosity correlated well with the dynamic viscosity. The modified steady state relative permeability setup can accurately estimate microemulsion viscosity and allow for better screening of surfactant formulations identified for field flooding. The dynamic microemulsion viscosities can also provide inputs for numerical simulation and better predict microemulsion behavior in the subsurface during field surfactant floods.
Research has discovered systems that can selectively flocculate mineral solids from a high molecular weight polymer flood matrix while leaving the polymer intact or alternatively achieving a viable total flocculation of the polymer in the produced fluids. Modified alkaline surfactant polymer (ASP) and standard polymer (P) flood systems were studied with findings obtained by controlled variations of both well-proven and non-prevalent chemical approaches. Results concluded that selectively removing the mineral solids from polymer-laden water produces reusable enhanced oil recovery (EOR) fluid.
EOR is a proven method to increase hydrocarbon yield from post-natural, stimulated, or standard flood driven reservoirs. Fluid produced from the reservoir contains the desired hydrocarbon and an aqueous phase. Previously considered a liability, properly treated, the aqueous phase can become an asset. Polymer floods have a proven history in EOR and, though complex in application, ASP also demonstrated EOR effectiveness in the laboratory. Most ASP approaches are currently in field trial stages. The produced fluid is subjected to hydrocarbon separation with the resulting aqueous system either treated for disposal or recycled into the system. The aqueous phase matrix is mainly composed of high molecular weight polymer, mineral solids, residual base, residual oil, and possibly surfactant. If the producer chooses disposal, the solids must be flocculated by a method balancing density, dewaterability, processability, process variability, and cost. However, if the producer opts to recycle the fluid for reinjection, steps must be taken to minimize polymer deviations requiring selective flocculation of all components with exception of the polymer. This undertaking is challenging as EOR polymers are also effective flocculants, therefore sensitive to standard coagulant and flocculant approaches. Utilizing controlled, standard methods and multivariable design of experiments, results were obtained for both total and selective flocculation.
Total flocculation systematically studies the influence of pH, inorganic, and organic coagulants in maximizing the treatment effectiveness. The same approach was successful for selective flocculation, however unique coagulants were applied. The selective flocculation process coagulated and separated the mineral solids, and left the high molecular weight polymer intact and the fluid matrix as viscous as prior to treatment. Effectiveness of treatments were determined using standard gravimetric and viscometric methods.
These discoveries will assist decision makers in determining whether total or selective flocculation is the most viable treatment for polymer based EOR, balancing environmental and economic aspects to pursue a desired treatment route. These methods, though targeting EOR, have practical applications for treatment of flowback and water produced from stimulation and potentially drilling operations as well.