IOR 2019 – 20th European Symposium on Improved Oil Recovery

Synergy between EOR processes has the potential for achieving more effective application. Previously, we demonstrated Smart water's slight negative-impact on polymer and chase brine injectivity, positive impact on polymer retention, and negligible impact on polymer acceleration. We also demonstrated polymer's positive effects on Smart water's surface potentials, contact angles, and oil recoveries. Herein, we further investigate this synergy focusing on polymer's possible effects on SmartWater spontaneous imbibition. We also revisit singlephase displacement observations with regard to polymer injectivity and retention using bench-top experiments. The main experimental program consisted of spontaneous-imbibition and sorption measurements. A supplementary experimental program included polymer filterability and adsorption tests to provide additional insights on this synergistic SmartWater/Polymer process.

Results showed the superiority of SmartWater as an imbibition fluid compared to the higher-salinity injection water. With polymer addition, the lower-salinity SmartWater upheld its superiority. Accounting for viscosity variations, sorption rates into powdered rock packs were actually very comparable with and without the polymer. In core plugs, and with both brines, polymer addition yielded further water imbibition above that obtained with polymer-free solutions. Still imbibition was much more effective with SmartWater even in the presence of the polymer. At the same polymer concentrations, polymer adsorptions onto the rock were comparable with both SmartWater and the higher-salinity injection water. However, at equivalent viscosities, polymer adsorption onto the rock was lower with SmartWater than with the higher-salinity injection water. These static adsorption results could provide an additional explanation to the substantially lower retention observed with the lower-salinity SmartWater in dynamic retention tests. At the same polymer concentration, the solutions in both brines exhibited comparable filtration ratios. However, for solutions of equivalent viscosities, the solution in the higher-salinity conventional injection water exhibited higher filtration ratio. This discrepancy with single-phase displacement results can be explained based on differences of the main mechanism contributing to permeability reduction (adsorption layer versus mechanical plugging).

In conclusion, the favorable SmartWater/Polymer synergy extends to spontaneous imbibition. Polymer solutions, in both SmartWater and injection water yielded additional imbibition in tertiary mode. However, with injection water the additional imbibition after polymer addition is much smaller than with SmartWater.

Hydrophobically modified polyacylamides (HMPAMs), also known as associative polymers, have high potential as EOR chemicals due to their specific chemical properties. However, the strength of hydrophobic interactions also leads to more complex flow behavior. Mostly, polymer injectivity is studied in linear coreflood experiments. However, in reality the polymer injection occurs in radial flow. The main difference is that linear flow has constant pressure gradient and thereby constant shear rate, while radial flow has a position dependent pressure gradient and therefore a shear gradient in the near-well region. For complex non-Newtonian polymers like HMPAMs this may have large implications for injectivity.

The influence of relatively weak hydrophobic interactions on flow properties was studied by radial injection of a semi-dilute solution of associative polymer under high injection rates. Radial flow was achieved by injection in a well placed in the center of a 30 cm diameter, 3 cm thick disc cut from Bentheimer outcrop sandstone. Bentheimer was chosen due to the high homogeneity of the rock. Pressure ports were positioned between the injection well and the rim to allow determination of the pressure decay curve during flow. Effluent was collected and analyzed for mechanical degradation.

The radial injection of the associative polymer showed shear thinning behavior at all rates, including at near-well equivalent rates. This means that this polymer can be injected with relatively low pressures and will gain viscosity as it propagates further from the well. Interestingly, this is different from the observations in linear core floods. Disruption and regeneration of weak hydrophobic interactions appears to be the main cause of the change in rheological behavior between linear and radial flow. The associative polymer showed excellent shear stability with only moderate viscosity reduction at extreme injection rates. The results confirm the need for in-situ measurements of polymer rheology not only in linear but also radial flow as input data for injectivity modeling. The experiments reveal that the associative polymer has different rheological behavior in bulk, linear and radial flow. The near-Newtonian behavior at near-well injection rates combined with building of viscosity at lower rates further from the well is beneficial for application in polymer flooding. This is to our knowledge the first systematic investigation of an associative polymer in radial flow. The results have major implications for polymer flood modeling and flood design, particularly for injections in vertical wells where injectivity is critical.

The present work presents the conceptual basis and experimental evaluation for a new technique to create insitu blocking foams in naturally fractured reservoirs by the injection of the foaming agent dispersed in a non-condensable gas stream. This work represents a further development of a previous paper (SPE-190219-MS; Ocampo et al, 2018) which presents a similar development for matrix dominated systems. Equivalent to previous work, the main objective is simplifying the operation and reducing costs for the deployment of EOR foams in gas injection based projects, and overcoming the limited reservoir volume of influence achieved by the surfactant alternated gas (SAG) technique.

An extensive and systematic experimental work was performed using fluids and low porosity naturally fractured rock representative of the Piedemonte area (Colombia, South America). The experiments were devised to investigate the effect of the dispersed chemical (surfactant) concentration and the gas velocity on the ability to create blocking foams at high pressure and temperature through the naturally fractured rock at a stress state representing open fracture conditions. The main physical mechanism behind this new technique is again the transfer of foamer droplets dispersed in the gas stream into the water present in the hydrocarbon reservoir. This transfer occurs because of the contrast in foamer concentration between the dispersed phase and the in-situ water. Results herein show reductions in gas mobility between 50% and 66%, along with increases in oil recovery factor between 10% and 34%, when the foamer chemical is dispersed in the gas stream, compared with the base gas flooding process performed on the described rock at residual oil and water saturations. This condition is obtained as far as the gas velocity is closed to rates equivalent to the velocities experienced near wellbore in the target reservoir, and the concentration of the active chemical is above 800 ppm (five times the concentration required to create blocking foam in the matrix system; Ocampo et al, 2018). Successful experiments with this new foam technique showed similar incremental recovery factors and stability periods as foams created by the SAG technique at higher chemical concentrations on the same rock fluid system.

The experimental results encouraged the progression and approval of a field pilot application of this foams technique this year in a Colombian Piedemonte gas condensate field characterized by the presence and dominance of the natural fractures both in the production and hydrocarbon gas injection performance.

Foam reduces gas mobility and can help improve sweep efficiency in an enhanced oil recovery process. For the latter, long-distance foam propagation is crucial. In steady gas-liquid flow, foam is generated in homogeneous porous media by exceeding a critical pressure gradient, which normally only happens near the wellbore. Away from wells, these requirements may not be met, and foam propagation is uncertain.

It has been shown theoretically that foam can be generated, independent of pressure gradient, during flow across an abrupt increase in permeability. This could dominate foam generation away from wells in layered or laminated geological formations and can improve the chances of success of a foam application. The objective of this study is to validate theoretical explanations through experimental evidence and to quantify the effect of permeability contrast, velocity and fractional flow on this process.

In this study, we validate theoretical predictions through a variety of experimental evidence. Coreflood experiments involving co-injection of gas and surfactant solution at field-like velocities were performed. Layered, consolidated and well-characterized sintered glass cores were used as the porous media. The permeability change in each core was analogous to sharp, small-scale heterogeneities such as laminations and cross-laminations. The experiments were carefully designed not to allow foam generation by mechanisms other than snap-off at the permeability boundary in the core. Local pressure gradient was measured at various locations and was used to identify foam generation and subsequent propagation through the porous medium. Additionally, X-ray computed tomography (CT) was employed to detect changes in phase saturation that accompany foam generation and subsequent propagation downstream. CT-based saturations measurements were also used to qualitatively chart the reduction in capillary pressure across the sharp permeability jump, supporting theoretical explanations behind this process. The effect of permeability contrast, superficial velocity and flowing gas fraction on this process was also investigated. For a given permeability contrast, foam generation was observed at higher gas fractions than predicted by previous theory ( Rossen, 1999 ). Conditions for propagation of foam were explored by successively performing experiments at lower velocities and higher gas fractional flows. Significant fluctuations in pressure gradient accompanied the process of foam generation, indicating a degree of intermittency in the generation rate - probably reflecting cycles of foam generation, dryout, imbibition, and then generation. The intermittency of foam generation was found to increase with decreasing injection velocities and greater permeability contrasts.

In the recent years, foam process has shown a growing potential in producing naturally fractured reservoirs. While much effort was devoted to fractured reservoir characterization and laboratory design of foaming chemicals, decision-making for field development goes through reservoir simulation. This work assesses foam flooding performance through dual-media reservoir simulations, to optimize injection strategy and foam key features such as stability, mobility reduction, adsorption and interfacial tension reduction.

Foam flooding may improve significantly oil recovery in fractured oil-wet reservoirs through a better fluid mobility control and an increase in viscous forces in fractures, improving the sweep efficiency. Additionally, foaming chemicals can yield ultra-low interfacial tension between oil and water, and thus enhance the oil mobilization. This study evaluates the sensitivity of foam flooding performances to foam model parameters and injection strategy at pilot-scale.

This sensitivity analysis is based on the variance decomposition (Sobol's index) of the response model (such as oil production), while a surrogate model is built to limit the number of simulations required to evaluate the objective function. Optimization are performed to constrain the process to be more efficient than gas and/or water injection.

Compared to other flooding processes such as gas and/or water injection, this work demonstrates how foam flooding can be optimized such that viscous gradients and gravity drainage act optimally to sweep oil in the matrix, but still maintain a sufficient injectivity to yield a maximum recovery. The foam process optimum is a trade-off between oil recovery maximization and costs minimization, which corresponds to a rise in viscous forces that force out the oil from the matrix while respecting the fracture pressure and limiting the amount of injected chemicals.

This sensitivity assessment brings new insights for pre-feasibility studies, in particular for the foaming formulation design. Specifically, as foam flooding recovery mechanisms interplay, impacts of chemicals adsorption, injection strategy and optimization parameters will strongly differ whether foaming chemicals achieve or not ultra-low interfacial tension. Hence, for a “low-IFT foam”, chemicals adsorption reduction will be crucial although negligible for a foam process without IFT reduction. Because of different mechanisms involved, the foam process efficiency mainly depends on viscous forces developed in the fracture network whereas the “low-IFT” effect mainly relies on the penetration of chemicals in the matrix, that facilitates oil recovery.

The optimization workflow demonstrated its ability to evaluate and to help the selection of the most appropriate process at pilot scale, according to the reservoir specificities.

Foam processes aim to improve the efficiency of gas-based injection methods through gases mobility control. They have been successfully applied in various EOR contexts: CCUS through CO2-EOR, steam injection for heavy oil reservoirs, and also in fractured reservoirs. The success of such processes depends on multiple factors, among which the interactions between the surfactants, the oil and the rock, play a key role. The purpose of this study is to provide initial answers by focusing on the influence of wettability and oil saturation on the behavior of CO2-foam flows.

A new coreflooding set-up is designed for ‘mesoscopic’ cores (2cm diameter) in order to conduct foam formulation screening and perform faster foam injection tests at reservoir conditions. This set-up was first validated by repeating experiments performed previously on classical corefloods with 4cm diameter cores. Similar results in terms of mobility reduction were obtained for the same operating conditions with a considerable reduction of test duration.

All experiments were performed with Clashach sandstones cores having approximatively 16% porosity and 600mD permeability. Two gas compositions have been studied: (1) a dense supercritical CO2 (density of 638kg/m3 at P=160bar, T=60°C) and (2) a non-dense gas mixture of CO2 and CH4. For each gas composition, four foam injection tests were carried out: two on water-wet rock samples, two others on crude-aged core samples, and for both in the absence and in presence of oil. Anionic surfactant formulations and gas were co-injected with a gas fraction of 0.7. Foam rheology was assessed by measuring foam apparent viscosity through a scan of interstitial velocities.

All the tests performed in dense conditions have highlighted the generation of strong foams, which present shear-thinning rheological behavior; the apparent viscosity decreases as a power law of the interstitial velocity. An influence of the wettability is observed on the foam apparent viscosity, which drops off by 30% in altered wettability rock samples. When samples were originally saturated with oil at Swi, the level of apparent viscosity remains globally unchanged but the kinetics of the initial formation of the foam is slower with oil than without. Foam flooding experiments are sometimes carried out simply in the presence of oil without taking into account the influence of wettability, which appears to be as important, if not more, than the oil saturation itself. These results will hopefully provide some guidance for future foam studies and raise awareness on the importance of these parameters.

After 6 years of continous polymer injection in El Corcobo Norte field, pilot's preliminar evaluation showed promising results. Although the evaluation is still ongoing, polymer technology economics look good enough to sustain a project expansion. Located in the Neuquén Basin, El Corcobo Norte field is an unconsolidated, underpressurized, strongly water-wet sandstone reservoir, producing medium-heavy oil from the Centenario Formation. Reservoir drive is waterflood, which has been implemented since the beginning of the field's development. Up to date the field has more than 650 producers and 350 injectors, mostly developed under 20 acres, inverted 7-spot patterns. Main challenge for this field's operation is sand production and chanelling issues that leave bypassed or undrained oil zones. Since 2008 EOR technologies were evaluated in order to increase ultimate recovery factor. As a results of this screening, polymer injection was chosen as the first candidate to test in the field. Polymer pilot design and execution was described in SPE 160078 (“Desing and Execution of a Polymer Injection Pilot in Argentina”) and pilot preliminar evaluation was presented in SPE 181210 (“Evaluation of a Polymer Injection Pilot in Argentina”). Based on the pilot's learnings, an expansion project was designed to maximize the use of the available capacity and upscale polymer injection as efficiently as possible, considering field's current operational conditions. The present article will focus on describing the upscale of the polymer pilot and the strategy to optimize the project's operation.

The main objective of this paper is show the design, implementation and results of a nitrogen + steam pilot implemented in a Colombian heavy oil reservoir. Given that the answer to the cyclic steam injection has declined significantly in some wells of the interest field due to the high number of cycles, a pilot project to assess the feasibility of injecting nitrogen accompanied cyclic steam was raised. The determination of optimum volumes and injection scheme by numerical simulation was performed in a sector model. Eleven schemes of steam plus nitrogen injection were evaluated. In all schemes, it was tested remained constant volume of injected steam, while the volume of nitrogen varied according to the scheme (pre-injection, post -injection, co-injection or a combination of the above). In all cases, it assumed the injection of nitrogen according to real drive capacity 1200 m3/hour (1017072 ft3/day).

The best case corresponds to start injecting only one day with nitrogen, followed by five days of co-injection and ending with a single nitrogen day. Under this scheme an incremental production of 5642 barrels of oil a trial period of six months, with average oil production of 53 BOPD in the same period and maximum rate of 142 BOPD. According to the simulation results, it was decided to implement the pilot steam + nitrogen injection, following the best injection scheme given above; that is, a day of pre-injection (nitrogen only), five days of co-injection (steam + nitrogen) and one day post-injection (nitrogen only). The results of the pilot show that oil production has increased compared to previous cycles, reaching similar results to the numerical simulation forecast. A methodology to implement steam injection enhanced with nitrogen is proposal in this paper, which can be applied in any field of heavy crude scheme developed under cyclic steam stimulation.

Previously proposed models of wettability change have not been tied to the chemistry that controls wettability but instead were driven by simplistic criteria such as salinity level or concentration of an adsorbed species. Such models do not adequately predict the impact of brine compositional change and therefore cannot be used to optimize brine composition. In this work, after testing proposed models in the literature on sandstones and carbonates, we propose a mechanistic surface-complexation-based model that quantitatively describes observations for ionically treated waterfloods. To the best of our knowledge this is the first surface-complexation-based model that fully describes ionic compositional dependence observed in ionically treated waterfloods in both sandstones and carbonates.

We model wettability change by directly linking wettability to brine chemistry using detailed colloidal science. Brine has charged ions that interact with polar acidic/basic components at the oil-water interface and rock surface and therefore oil/brine and rock/brine interfaces are charged and exert both Van der Waals and electrostatic forces on each other. If the net result of the forces is repulsive, the thin water film between the two interfaces is stable (i.e., the rock is water-wet) otherwise, the thin water film is unstable and the rock becomes oil-wet. Based on Hirasaki (1991) , we describe a ratio of electrostatic force to Van der Waals force with a dimensionless group, called “stability number,” where rock wettability is water-wet for values greater than one and oil-wet for values less than one. For sandstones, the zeta potentials of oil/brine and rock/brine interfaces become more negative/less positive by diluting or softening the brine and/or increasing pH. Similarly, for carbonates, dilution and/or sulfate enrichment of brine makes surface potentials more negative. Such brine modification can therefore be used to improve oil recovery.

We implemented the improved wettability change model in a comprehensive coupled reservoir simulator, UTCOMP-IPhreeqc, in which oil/brine and rock/brine zeta potentials are modeled using the IPhreeqc surface complexation module. We take into the account total acid number (TAN) and total base number (TBN) for the oil/brine interface and we use rock surface reactions for brine/rock surface potential modeling. Surface potentials obtained from the geochemical model are used to calculate the dimensionless group controlling wettability change, which is dynamically modeled in the transport simulator. The model is validated in sandstones and carbonates by simulating an inter-well test, and several corefloods and imbibition tests reported in the literature. For sandstones, we model Kozaki (2012) and BP's Endicott trial. For simple dilution in carbonates we model experiments by Shehata et al. (2014) and Yousef et al. (2010) . For enrichment with sulfate we model Zhang and Austad (2006) and for increasing total ionic strength via sodium chloride enrichment, Fathi et al. (2010a) .

Excess water production usually has negative influence on volumetric sweep efficiency, and as a result deteriorates the profitability of production operations. In the past half century, numerous techniques were extensively applied in frame of water shutoff technology including polymer-based macro- and microgels, emulsions, sols, crystalline compounds, etc. Surprisingly, the operators did not appreciated the use of silicates despite the fact that the silicates were proposed as chemical diverting agent and profile control already in 1922. In the past decades, the silicate technology has a revival because of their outstanding features, which is apparently superior to other gels. These facts are well proved by more than 100 well treatments carried out in Hungary, Serbia and Oman curing various problems including excess water production, water conning, leakoff, restriction of vertical gas migration in a collapsed well and profile control in water injection wells. Earlier, the pure silicate solutions were applied, however recognizing some weak points of the gel under HPHT conditions, substantial efforts were made to improve both the chemical systems and the injection protocol.

First, combining the silicates with polymers some fundamental deficiencies could be eliminated forming a double network in gel providing flexibility and much less sensitivity to fracture and syneresis at high-pressure gradient. Later, nanomaterials were added to silicate and silicate/polymer gels, which advanced nucleation providing faster, reliable and controllable gelation in case of sequential injection protocol. The detailed laboratory studies have shown that the gelation kinetics is also substantially modified because the polymer is a retardant, while the nanosilica is an accelerator influencing thus the setting time. In addition, inevitable advantages of the composite gel are that the technology is based on cheap, easily available, and environmentally friendly chemicals. The “green” method was first applied in an oil filed (Algyő 2, field, Hungary) already in 2014, and abundant information could be collected to summarize the results and lessons to learn. Application of silicate/polymer/nanosilica gel controlling the detrimental water production resulted in outstanding technical and economic results. In one-year basis, the water production decrease by 32,000 bbl, meanwhile the incremental oil production increased by 21,000 bbl. The net profit was 653,000 USD in contrast to the investment (OPEX) of 238,000 USD. Since the profitability of treatments was above the expectation, the operator plan to continue extending the advanced composite RCC method. The paper to be presented will give a detailed overview of the laboratory studies and pilot tests.