Customizing Polymer Rheology for Application in Commercial Scale Polymer Floods

The availability of ~ 7 years of actual performance data for the ongoing field-scale polymer flood in South of the Sultanate of Oman provides ample opportunities to reveal the reservoir dynamics and its interplay with induced EOR mechanism. The paper focusses on analysis of the polymer pattern behaviour, underlying reasons for such response and key indicators to characterize pattern performance. Upsets in surface polymer injection facility leading to the phenomenon of WAP (Water-Alternating-Polymer) and it's impact on recovery is also assessed in context of actual field examples. The paper then illustrates how this information could be exploited to counter challenges faced in the field, enhance polymer pattern performance, optimize it's further expansion and de-risk any other future EOR development. A nested modelling approach has been employed, wherein models at different scales are generated, tailored to meet the objectives. High resolution 3D conceptual models are built in Shell proprietary tool PolyMoReS to calibrate the model response against the actual polymer pattern behaviour in the field, study the impact of mixing between polymer and water slugs in WAP type of recovery, and affirm the correct polymer rheology. Three segment models covering the field are created and history matched with the use of Stochastic Uncertainty Management. Attempts have been made to obtain history match (HM) on segment, pattern and well levels, with greater emphasis on polymer patterns capturing polymer oil response, water-cut reversals and polymer breakthroughs. Models are then complemented by Pressure Fall-Offs, tracer tests and PLTs to capture uncertainties in fracture growth and areal and vertical conformance. The HM model is then used to predict polymer performance. Significant insights into waterflood and polymer flood performance are gained, which help improve the pattern performances. Assessment of WAP with both conceptual Physics and field segment models demonstrate considerable deferment of oil. Capturing injector – producer connectivity has proven the most pivotal element in explaining polymer oil response and breakthrough. Models indicate that lower than expected incremental recovery and sharper decline of oil response in some patterns are related to the lower polymer mass injected, which in-turn could be attributed to many operational factors (e.g., polymer injection uptime, injection rate, low injection viscosity, WAP), and the presence of natural fractures or uncontrolled growth of induced fractures. The study also reveals optimization opportunity to reduce the volumes of back produced water. The paper presents a comprehensive multi-scale reservoir modelling study for a field with significant historical data of large scale polymer flood. Impact of WAP injection, reflecting the reality of interruptions in polymer flood due to operational upsets, on medium to long term polymer flood value is presented. Analysis of polymer patterns in the field demonstrates how different key indicators e.g., PUF (Polymer Utilization Factor) can characterize pattern performance throughout its life-cycle and answers questions, e.g., why some patterns behaved well, while others not.

High molecular polyacrylamides have been used widely in polymer flooding applications for improving oil production. The injected polymer is subjected to different shear rates and thus results in various performance. This paper integrates the behaviors of different laboratory experiments into a general workflow for evaluation of the impact of shear rate on oil production using numerical simulation. The impact of shear rates on the rheological property of a polymer solution was evaluated by injection of polymer solution into a core plug at different injection rates. The measured bulk viscosity and calculated apparent viscosity were fitted with different rheological models to identify flow behavior and the corresponding shear rate range. Shear rate at sand surface and shear rate profile from injector to deep formation were estimated using an inverted five-spot well pattern. Combination of the experimental results with the estimated shear rates was used to describe the flow behaviors of injected polymer solutions. The impact of shear-rate dependent viscosity on oil production was evaluated using a UTCHEM simulator. Laboratory experiments showed that polymer shear degradation occurs when the shear rate exceeds 4402 s−1 in a carbonate core plug with a permeability of 390 mD. The shear thickening was observed in the shear rate range of 90 to 4402 s−1. The injected polymer solutions flowing from injector to near producer experience four viscosity regimes in sequence, i.e., shear degradation, shear thickening, shear thinning and Newtonian. The scales of each regime are different. The critical shear rates upon mechanical degradation are closely related to perforation parameters including perforation size, density, and interval. When the polymer solutions is injected into a reservoir with permeability of 390 mD at a rate of 0.2 pore volume/year, shear degradation occurs within a radius of 0.2 m around the wellbore, and shear thickening behavior between a radius of about 0.2 to 10 m, and the regimes of shear thinning or Newtonian flow in the rest range over a radius of 10 m to deep formation. The numerical simulation results show that polymer viscosity loss leads to oil production reduction, and shear thickening has very little effect on oil production if no considerable injectivity declines. The paper provides a workflow for evaluating the impact of shear rates on oil production from laboratory experiments to well scale polymer flooding by numerical simulation studies.KeywordsPolymer floodingNumerical simulationShear rateRheological property

Summary The objective of polymer flooding is to improve macroscopic sweep efficiency and oil recovery by increasing the viscosity of the displacing water thereby decreasing the water/oil mobility ratio. One major parameter that determines polymer flood performance is therefore the in-situ polymer solution viscosity, which is dependent on several factors such as formation temperature, polymer concentration, shear rate and salinity (salt concentration). Rheological polymer properties such as bulk viscosity variation with polymer concentration for a given water salinity have been derived from laboratory measurements. Generally, a higher solution salinity yields less viscosity for a given polymer concentration. Such relationship between salinity and polymer solution viscosity needs therefore to be implemented in dynamic simulations to consider the mixing effect of different water salinities (injected, connate, aquifer, etc.) in order to obtain representative in-situ polymer viscosities for polymer flood evaluation. This paper describes the implementation of salt-dependent polymer viscosity functions in reservoir models and the evaluation of simulation results in order to provide answers related to the impact of salinity variations on polymer flood performance. Evaluation of results shows variation of low salinity and polymer flood fronts within the reservoir when a lower salinity (e.g. <1 g/l) viscous solution encounters the in-situ formation water at higher salinity (5 – 15 g/l), and how this leads to improved macroscopic sweep efficiency and oil recovery. Overall, low salinity viscous fluid injection resulted in a higher oil recovery (up to 6% incremental OOIP) given the same polymer concentration, or a similar recovery with lower polymer concentration (up to 50% less polymer consumption). The above results proved useful in evaluating the business case of using low salinity water (desalting using Electrodialysis Reversal (EDR) technology) for polymer solution preparation. Chemical savings and polymer concentration reduction are the main advantages for a polymer flooding project. Preliminary evaluations show that significant OPEX and CAPEX savings could be achieved by using EDR on an onshore field with moderate reservoir salinity, which is linked to increased operational efficiency and reduction of chemicals consumption.

Modeling dynamic fracture growth induced by non-Newtonian polymer injection

Improved polymer flooding injectivity and displacement by considering compositionally-tuned slugs

Application of polymer flooding in the revitalization of a mature heavy oil field

Accurate assessment of the potential of a nonisothermal polymer flood process for mobility control requires a model on the viscosities of reservoir fluids as a function of temperature. The temperature dependence of viscosity was included in the reservoir model and an extensive numerical simulation was conducted to examine the effects of temperature-related viscosity model on the reservoir performance during polymer flood in the reservoir with elevated temperature. A three-dimensional numerical model for fluid flow, mass transport, and energy balance is used to analyze the performance of the reservoir operating under polymer flood followed by waterflood. The nonisothermal scheme can be used as a quantitative tool to evaluate the comparative studies of different polymer flooding scenarios with respect to temperature dependence of fluids’ viscosities. Results of cumulative oil recovery and water-oil ratio (WOR) at the production well and injectivity at the injection well are presented for various types of temperature dependencies, reservoir temperature, and oil viscosities. The simulation scheme with thermal dependence of fluids’ viscosities allows a more quantitative prediction of reservoir performance. Significant improvement in predicted oil recovery and reduction in WOR is obtained for the case of including temperature dependencies of both water and oil because the reduction of oil viscosity is larger than that of brine viscosity. The inclusion of heat loss to over/underburden formations had limited effect on the oil recovery because of relatively small temperature difference under the conditions considered in this study. More rapid reduction of oil viscosity results in the considerable reduction in mobility ratio and significant increase of oil recovery at higher temperature. High oil recovery was obtained with small values of oil viscosity, but the improvement is not proportional to the reduction of oil viscosity.

Based on three established commercial test cases, this paper documents their polymer flood process performance, revealing the impact of reservoir properties and heterogeneity on incremental recovery and dynamic performance. The key reservoir management measures for improving polymer flood sweeping efficiency are illustrated, which are particularly important for the successful application of chemical EOR techniques in heterogeneous sandstone reservoirs. The three cases produced oil with similar fluid properties from sandstone reservoirs. They all adopted 5-spot well patterns, which provide excellent examples for studying the impact of reservoir property and heterogeneity on incremental recovery. The producing reservoir sands in the three pilot represent two types. The former is of high permeability and low heterogenity. The latter is characterized by low permeability, multiple flow units, thin individual sand layers and high inter-layer permeability contrast. The authors reviewed and compared changes in water-cut and production rate performance versus pore volume of polymer solution injected during the process, as well as the applied well spacing, polymer utilization, and ultimate incremental recoveries. The two high quality reservoir cases adopted 22 ac well spacing, achieved an incremental recovery of 12.5% and 13.7%, a low polymer utilization of 1.08 kg/BO and 1.79 kg/BO, respectively, and a maximum water-cut reduction of 24.4%. In the low permeability case, although a much smaller well spacing of 11 ac was adopted, it achieved a relatively low incremental recovery of 8.2% and a maximum water-cut drop of 5.2%, with a polymer utilization of 2.51 kg/BO. Incremental EUR of individual producer through polymer flood is 49 MBO in the low permeability case, compared with 130 MBO in the high permeability reservoir sands. The three cases proved the effectiveness of fit-for-purpose reservoir management and conformance improvement techniques in improving polymer flood sweep efficiency to achieve high incremental recovery. Such techniques include hydraulic fracturing of low permeability sands, optimization of polymer solution concentration and molecular weight to adapt to different sand units with varied permeability, optimization of injection process to maintain a uniform pressure system, adjustment of injection direction to sweep by-passed remaining oil, and, zonal injection to improve polymer solution intake in low permeability sand units. The discussed cases proved the technical feasibility and economic viability of polymer flood in sandstone reservoirs with varied heterogeneities and revealed the impact of heterogeneity on performance and incremental recovery. The reservoir management and conformance improvement efforts during polymer flood process provided valuable lessons and best practices for similar applications.

Polymer Flooding has been shown to increase oil production. The reason for increasing oil production is acceleration along flow paths but also flow diversion from higher permeability to lower permeability areas. Tracer tests performed in the 8 TH Reservoir in Austria prior, during and after polymer flooding show that the flow system dramatically changed. The connected volumes from injector to producer as well as the flow heterogeneity were influenced and substantial incremental oil produced. A number of tracer tests were performed in the pilot area of a polymer flood at various times. In addition, pressure data and polymer rheology was analyzed. The tracer results were used to calculate flow pattern, dynamic Lorentz coefficient and connected volumes. Pressure data were used in combination with geomechanical modelling to investigate the injection regime (matrix or fractures). The interpretation of the data was combined with the determination of incremental oil production based on simulation. The tracer tests reveal the dramatic changes in flow patterns, connected volume changes by more than a factor of three occurred and the Lorentz coefficient indicating the heterogeneity of flow changed by more than a factor of two. The injection regime changed from injection under matrix conditions prior to polymer flooding to injection under fracturing conditions during polymer injection and back to injection in matrix conditions during chase water injection. The reason for the changes in injection conditions is the near-wellbore viscoelastic rheology of the High Molecular Weight Polymers which were used. The growth of fractures leads to additional alteration of flow paths. The design of polymer flooding needs to take into account that flow paths are not only changed due to the reduction in water relative permeability resulting from polymer adsorbing to the rock and the increased viscosity of the injected fluid but also owing to changes in injection regime. The changes in injection regime might lead to early breakthrough of chase water as it might not flow along the same paths as the polymer solution.

In-depth gel treatment is a chemical EOR process used to improve the sweep efficiency from heterogeneous reservoirs with crossflow. However, if these reservoirs are saturated with viscous oil, polymer and surfactant flooding should be combined with in-depth gel treatment. Thus, in this study, a 3D model using the UTGEL simulator was built to model in-depth gel treatment combined with surfactant slug and polymer solution. The model was represented by one quarter of the five-spot pattern with eight layers where two thief zones are located in the middle of the model. The thief zones had a permeability of 1500 md with a total thickness of 20 ft, while the rest of the layers had a permeability of 100 md with a total thickness of 200 ft. The gel system consisted of a polyacrylamide/Cr(VI)/thiourea solution, which is considered an in-situ gelation system. Gelant solution was injected for 60 days when the water cut in the model reached 65%, followed by surfactant slug for 2 years, polymer solution for 3 years, and then post-water injection for the rest of the simulation time. The concentrations of the surfactant ranged from 0.01 to 0.2 wt.%, while the polymer concentration was 1,000 ppm. The injection rate was 1,070 bbl/day during all flooding and treatment processes. The results showed that it is imperative to implement surfactant with gel treatment to reduce the interfacial tension between water and oil phases and to alter the wettability of the reservoir rocks. Thus, gel treatment alone or gel followed by polymer was not as efficient as the injection of a surfactant slug. The results also showed that as the reservoir temperature increased, the overall performance of gel, polymer, and surfactant decreased. Therefore, the higher the temperature, the lower the recovery factor. The results also revealed the importance of viscoelastic behavior of the HPAM polymer solution where higher results for both water-wet and oil-wet conditions were obtained compared to shear-thinning behavior only. Moreover, the results revealed interesting behavior regarding the concentration of the surfactant, where the recovery factor increased as the concentration of the surfactant increased in oil-wet conditions. However, in water-wet conditions, the results were unpromising and unfavorable. Furthermore, the injection of surfactant directly after the gel treatment was more effective in improving the sweep efficiency than the injection of polymer directly after the gel treatment. Finally, as the salinity of makeup water and/or reservoir brine increased, the recovery factor decreased for both water and oil-wet systems. This is because, as salinity increased, the adsorption of both polymer and surfactant increased and the polymer viscosity decreased. Furthermore, the presence of divalent cations such as Ca+2 and Mg+2, would have a negative impact on overall treatment.

Low Salinity Polymer (LSP) flooding is one of the emerging synergic techniques in enhance oil recovery (EOR). Previous experimental studies showed an exceptional improvement in displacement efficiency, polymer rheology, injectivity, and polymer viscoelasticity. Nevertheless, when it comes to modeling LSP flooding, it is still challenging to develop a mechanistic predictive model that captures polymer-rock-brine interactions. Therefore, this study employs a coupled geochemical-reservoir numerical model to investigate the effect of water chemistry on polymer-brine-rock geochemical interactions during LSP flooding through varying overall salinity as well as the concentrations of monovalent and divalent ions. In this study, the MATLAB Reservoir Simulation Toolbox (MRST) was coupled with a geochemical interface module i.e., pH-Redox-Equilibrium in C programming language (PHREEQC), termed as IPHREEQC. The coupled MRST-IPHREEQC simulator enables simulating the effects on different parameters on polymer viscosity including the Todd-Longstaff mixing model, inaccessible pore volume, permeability reduction, polymer adsorption, salinity, and shear rate. For describing the related geochemistry, the presence of polymer in the aqueous phase was considered by introducing novel solution specie to the Phreeqc database. Using this coupled simulator, several geochemical reactions and parameters can be assessed including rock and injected water compositions, injection schemes, and other polymer characteristics where the focus of this work is on water chemistry. Moreover, different injection schemes were analyzed including low-salinity water, low-salinity polymer injection (1×LSP), and 5-times spiked low-salinity polymer injection (5×LSP) with their related effects on polymer viscosity. The results showed that polymer viscosity during low-salinity polymer flooding is directly affected by calcium (Ca2+) and magnesium (Mg2+) ions and indirectly affected by sulfate ion (SO42−) as a result of polymer-rock-brine interactions on a dolomite rock-forming mineral. Furthermore, the findings showed that monovalent ions such as sodium (Na+) and potassium (K+) have less pronounced effects on the polymer viscosity. However, the release of calcium (Ca2+) and magnesium (Mg2+) ions due to the dissolution of dolomite led to the formation of polymer (acrylic acid, C3H4O2) complexes and consequently, a pronounced decrease in polymer viscosity. In addition, the increase of sulfate ion (SO42−) concentration in the injected LSP solution affects the interactions between the polymer and positively charged aqueous species and leads to less polymer viscosity loss. Additionally, as a de-risking measure for LSP flood designs, estimating the effect of each ion can be highly useful step. The effect of cations is also related to charge ratio (CR), which renders it the key objective to determine the optimum CR ratio at which viscosity loss of LSP flood is avoided or at least minimal. The coupled simulator works as an integrated tool, which is sound, precise, and adaptable with the ability to encapsulate the reactions required for LSP mechanistic modeling. This paper is among the very few, which describe mechanistic geochemical modeling of the low-salinity polymer flooding technique. The coupled simulator provided new insights into understanding the mechanisms controlling LSP flooding. Based on the findings of this work, several successful low salinity-polymer field pilots can be designed.

The recovery efficiency by polymer flooding is 10% higher than that of water flooding for main reservoirs in Daqing Oil Field, and yearly oil production rate by polymer injection has continuously reached over 10 million tons for 8 years. However, to further enhance oil recovery, it is very important to know how many oil remained in the reservoirs after polymer injection as well as its distribution. The paper took use of logging data of 200 new drilling wells to make macro analysis on remaining oil distribution after polymer injection, and based on the data of coring wells, collected from the same well group in Western Fault of B1 Region, before and after polymer flooding, respectively, core membrane fluorescence technology was used to conduct the analysis on microscopic remaining oil distribution. The macro analysis results indicate that swept thickness of the reservoir increased 21.4% from 68.6% of water flooding to 90% of polymer flooding, and oil saturation reduced 11.9% from 52.8% to 40.9%, and displacement efficiency has went up 7.6 percentages. The residual oil distribution after polymer flooding is scattered in horizontal direction, while it is interleaved in vertical direction. Following polymer flooding, 10.6% unswept thickness is mainly distributed in thick oil layers, and most of remaining oil within thick oil layers lies in the interlayers and at the top of thick oil layers, which its effective thickness is close to 80%. From the macro point, remaining oil is mainly located in distributary line or some places where injection and production relation is imperfect or the phase is changed, and oil saturation located in distributary line is 4.4 percentages higher than that along mainstream line. The micro analysis results show that the behavior of remain oil distribution is similar to water flooding, which is mainly controlled by capillary force.

SummaryPolymer flooding can significantly improve sweep and delay breakthrough of injected water, thereby increasing oil recovery. Polymer viscosity degrades in reservoirs with high-salinity brines, so it is advantageous to inject low-salinity water as a preflush. Low-salinity waterflooding (LSW) can also improve local-displacement efficiency by changing the wettability of the reservoir rock from oil-wet to more water-wet. The mechanism for wettability alteration for LSW in sandstones is not very well-understood; however, experiments and field studies strongly support that cation-exchange (CE) reactions are the key elements in wettability alteration. The complex coupled effects of CE reactions, polymer properties, and multiphase flow and transport have not been explained to date.This paper presents the first analytical solutions for the coupled synergistic behavior of LSW and polymer flooding considering CE reactions, wettability alteration, adsorption, inaccessible pore volume (IPV), and salinity effects on polymer viscosity. A mechanistic approach that includes the CE of Ca2+, Mg2+, and Na+ is used to model the wettability alteration. The aqueous phase viscosity is a function of polymer and salt concentrations. Then, the coupled multiphase-flow and reactive-transport model is decoupled into three simpler subproblems—the first in which CE reactions are solved, the second in which a variable polymer concentration can be added to the reaction path, and the third in which fractional flows can be mapped onto the fixed cation and polymer-concentration paths. The solutions are used to develop a front-tracking algorithm, which can solve the slug-injection problem of low-salinity water as a preflush followed by polymer. The results are verified with experimental data and PennSim (2013), a general-purpose compositional simulator.The analytical solutions show that decoupling allows for estimation of key modeling parameters from experimental data, without considering the chemical reactions. Recovery can be significantly enhanced by a low-salinity preflush before polymer injection. For the cases studied, the improved oil recovery (IOR) for a chemically tuned low-salinity polymer (LSP) flood can be as much as 10% original oil in place (OOIP) greater than with considering polymer alone. The results show the structure of the solutions, and, in particular, the velocity of multiple shocks that develop. These shocks can interact, changing recovery. For example, poor recoveries obtained in corefloods for small-slug sizes of low salinity are explained by the intersection of shocks without considering mixing. The solutions can also be used to benchmark numerical solutions and for experimental design. We demonstrate the potential of LSP flooding as a less-expensive and more-effective way for performing polymer flooding when the reservoir wettability can be altered with chemically tuned low-salinity brine.

Polymer flooding is one of the most widely used chemical enhanced oil recovery methods due to its simplicity and low cost. To achieve high oil recoveries, large quantities of polymer solution is often injected through a small wellbore. Sometimes, the economic success of the project is only feasible when injection rate is high for high viscosity solution. However, injection of viscous polymer solutions has been a concern for the field application of polymer flooding.The pressure increase in polymer injectors can be attributed to (1) formation of an oil bank, (2) polymer rheology (shear-thickening behavior at near well-bore), and (3) plugging of the reservoir pores by insoluble polymer molecules or suspended particles in the water.In this paper, we propose a new model to history match field injection rate/pressure data. The pertinent equations for deep-bed filtration and external cake build-up in radial coordinate were coupled to the viscoelastic polymer rheology to capture important mechanisms. We selected radial coordinate in order to minimize the velocity/shear rate errors due to gridblock size in Cartesian coordinate.We used filtration theory and successfully history matched the field data. We performed systematic simulations and studied the impact of adsorption (retention), shear thickening, deep bed filtration, and external cake formation to explain the well injectivity behavior of polymer. The simulation results indicate that the gradual increase in bottomhole pressure during early times is attributed to the shear thickening rheology at high velocities experienced by viscoelastic HPAM polymers around the wellbore and the permeability reduction due to polymer adsorption and internal filtration of undissolved polymer. However, the linear impedance during external cake growth is responsible for the sharper increase in injection pressure at the later times.The proposed model can be used to calculate the injectivity of the polymer injection wells, understand the contribution of different phenomena on the pressure rise in the wells, locate the plugging or damage that may be caused by polymer, and accordingly design the chemical stimulation if necessary.

Polymer flooding is an important enhanced oil recovery (EOR) method with high performance which is acceptable and applicable on a field scale but should first be evaluated through lab-scale experiments or simulation tools. Artificial intelligence techniques are strong simulation tools which can be used to evaluate the performance of polymer flooding operation. In this study, the main parameters of polymer flooding were selected as input parameters of models and collected from the literature, including: polymer concentration, salt concentration, rock type, initial oil saturation, porosity, permeability, pore volume flooding, temperature, API gravity, molecular weight of the polymer, and salinity. After that, multilayer perceptron (MLP), radial basis function, and fuzzy neural networks such as the adaptive neuro-fuzzy inference system were adopted to estimate the output EOR performance. The MLP neural network had a very high ability for prediction, with statistical parameters of R2 = 0.9990 and RMSE = 0.0002. Therefore, the proposed model can significantly help engineers to select the proper EOR methods and API gravity, salinity, permeability, porosity, and salt concentration have the greatest impact on the polymer flooding performance.