Wang, Shuoshi (University of Oklahoma) | Yuan, Qingwang (University of Regina) | Kadhum, Mohannad (Cargill, Incorporated) | Chen, Changlong (University of Oklahoma) | Yuan, Na (University of Oklahoma) | Shiau, Bor-Jier (University of Oklahoma) | Harwell, Jeffrey H. (University of Oklahoma)
While injection of CO2 has great potential for increasing oil production, this potential is limited by site conditions and operational constraints such as lack of proper infrastructure, limited cheap CO2 sources, viscous fingering, gravity override at the targeted zones, and so forth. To mitigate some of these common limitations, we explore alternative methodologies which can successfully deliver CO2 through gas generation in situ, with superior IOR performance, while offering reasonable chemical cost.
While dissolved easily in reservoir brine, urea is thermally hydrolyzed to CO2 and NH3 after equilibration under reservoir conditions. Therefore, given its exceptional compatibility with reservoir fluids, its CO2 producing capacity and reasonable cost benefit, urea appears to be a promising candidate for delivering CO2 to increase oil recovery. The in-situ gas generation requires single chemical slug, which can minimize the complexity of the injection system.
One-dimensional sand pack tests and core flooding experiments were operated at pre-set conditions: different API gravity oils were used, varying from 27 to 57.3. In addition, the reaction rates of the urea hydrolysis and urea solution PVT property were tested separately under reservoir conditions.
Most importantly, results of injecting urea solution (as low as 10 % solution) showed superior tertiary recovery performance (as high as 37.97%) are realized as compared to the most recent efforts at our group (29.5%) as well as similar in situ CO2 generation EOR (2.4% to 18.8%) approaches proposed by others.
The economic feasibility and operational advantages of this newly developed method were demonstrated in this work. In brief, results of this work served further as a proof of concept for designing in situ CO2 generation formulations for tertiary oil recovery at both onshore and offshore fields under proper conditions.
The Gas and Downhole Water Sink-Assisted Gravity Drainage (GDWS-AGD) process has been developed to overcome of the limitations of Gas flooding processes in reservoir with strong aquifers. These limitations include high levels of water cut and high tendency of water coning. The GDWS-AGD process minimizes the water cut in oil production wells, improve gas injectivity, and further enhance the recovery of bypassed oil, especially in reservoirs with strong water coning tendencies.
The GDWS-AGD process conceptually states installing two 7 inch production casings bi-laterally and completing by two 2-3/8 inch horizontal tubings: oil producer above the oil-water contact (OWC) and one underneath OWC for water sink drainage. The two completions are hydraulically isolated by a packer inside the casing. The water sink completion is produced with a submersible pump that prevents the water from breaking through the oil column and getting into the horizontal oil-producing perforations.
The GDWS-AGD process was evaluated to enhance oil recovery in the heterogeneous upper sandstone pay in South Rumaila Oil field, which has an infinite active aquifer with a huge edge water drive. A compositional reservoir flow model was adopted for the CO2 flooding simulation and optimization of the GDWS-AGD process. Design of Experiments (DoE) and proxy metamodeling were integrated to determine the optimal operational decision parameters that affect the GDWS-AGD process performance: maximum injection rate and pressure in injection wells, maximum oil rate and minimum bottom hole pressure in production wells, and maximum water rates and minimum bottom hole pressure in the water sink wells. More specifically, Latin hypercube sampling and radial basis neural networks were used for the optimization of the GDWS-AGD process performance and to build the proxy model, respectively.
In the GDWS-AGD process results, the water cut and coning tendency were significantly reduced along with the reservoir pressure. That resulted to improve gas injectivity and increase oil recovery. Further improvement in oil recovery was achieved by the DoE optimization after determining the optimal set of operational decision factors that constrains the oil and water production with gas injection. The advantage of GDWS-AGD process comes from its potential feasibility to enhance oil recovery while reducing water coning, water cut, and improving gas injectivity. That gives another privilege for the GDWSAGD process to reach significant improvement in oil recovery in comparison to other gas injection processes, such as the Gas-Assisted Gravity Drainage (GAGD) process, particularly in reservoirs with strong water aquifers.
Aqueous foam has been demonstrated through laboratory and field experiments to be a promising conformance control technique. This study explores the foaming behavior of a CO2-soluble, cationic, amine-based surfactant. A distinguishing feature of this surfactant is its ability to dissolve in supercritical CO2 and to form Wormlike Micelles (WLM) at elevated salinity. Presence of WLM led to an increase in viscosity of the aqueous surfactant solution. Our study investigates how the presence of WLM structures affect transient foam behavior in a homogenous porous media (sand pack).
Sand pack foam flooding experiments were performed with two aqueous phase salinities: low salinity (15 wt. % NaCl) associated with spherical-shaped micelle and high salinity (20 wt. % NaCl) associated with WLM. We compared the onset of strong foam propagation and foam apparent viscosity buildup rate between the two salinity cases. The effect of WLM presence in transient foam behavior was investigated for co-injection and water-alternating-gas (WAG) injection strategies. In all foam flooding experiments, the surfactant was delivered in the CO2 phase.
Strong foam was generated in all foam flooding experiments, with an apparent foam viscosity of at least 600 cp for co-injection and 200 cp for WAG floods after five total injected pore volumes. The observed strong foam indicated that the delivery of surfactant in the CO2 phase was successful and that the surfactant molecules partition to the water phase in the sand pack. In comparison to the low salinity cases, the high salinity foam floods associated with the presence of WLM led to better foam performance. We observed an earlier onset of strong foam propagation as well as a higher apparent viscosity buildup rate. Better foam performance at higher salinity may be attributed in large part to the presence of WLM structures in the foam liquid phase. Entanglement of these WLM structures may have led to in-situ viscosification of the foam liquid phase and an increase in disjoining pressure between foam films. Both phenomena may have reduced the rate of foam film coalescence.
WLM structures behave similarly to polymer molecules. Our study may offer evidence that WLM is a valid alternative to polymer as an additive to enhance foam conformance control performance. Some potential advantages of WLM over polymer include: Delivery of surfactant in the gas phase (to alleviate the injectivity issue typically associated with high viscosity polymer-surfactant solution), resistance to extreme temperature and salinity, and reversible shear degradation.
Andersen, Pål Østebø (Dept. of Energy Resources, University of Stavanger) | Lohne, Arild (The National IOR Centre of Norway, University of Stavanger) | Stavland, Arne (The National IOR Centre of Norway, University of Stavanger) | Hiorth, Aksel (Dept. of Energy Resources, University of Stavanger) | Brattekås, Bergit (Dept. of Energy Resources, University of Stavanger)
Capillary spontaneous imbibition of solvent (brine bound in gel) from formed polymer gel into an adjacent, oil-saturated porous medium was recently observed in laboratory experiments. Loss of solvent from the gel by spontaneous imbibition may influence the blocking capacity of the gel residing in a fracture, by decreasing the gel volume, and may contribute to gel failure, often observed in water-wet oil fields. Formed gel cannot enter significantly into porous rock, which has important implications for spontaneous imbibition: the gel particle network itself is not imbibed, and remains close to the rock matrix surface, while gel solvent can leave the gel and progress into the matrix due to capillary forces. Polymer gel is an inherently complex fluid and modelling of its behavior is, as such, complicated. Accurate description and quantification of gel properties and behaviour on the laboratory scale is, however, necessary to predict the performance of gel placed in an oil field, particularly in fractured formations. In this work, we present an original modelling approach, to simulate and interpret spontaneous solvent imbibition from Cr(III)-Acetate HPAM gel into oil-saturated chalk core plugs. A theory describing solvent flow within a gel network is detailed, and was implemented into an in-house simulator. Simulations of spontaneous imbibition from gel was performed, and compared to free spontaneous imbibition of water. A good overall match was achieved between experiments and simulations on the core scale, which validates the proposed gel model.
All Faces Open (AFO) and Two Ends Open - Free Spontaneous Imbibition (TEOFSI) boundary conditions were used in the experiments, and formed the basis for simulation. Spontaneous imbibition occurs at the core end faces that are open to flow and exposed to gel (different for the two boundary conditions). The gel surrounding the core was discretized and included as a part of the total grid to capture transient behavior. The surrounding gel is treated as a compressible porous medium where the gel's polymer structure constitutes the matrix having constant solid volume while the gel porosity is a function of pore pressure. The gel permeability is modelled as function of gel porosity using a Kozeny-Carman approach. The flow equations for the gel and core domains were solved simultaneously by implementing the proposed description into the core scale simulator IORCoreSim. Two properties were identified to control the transport of water from gel into the adjacent matrix: the permeability and compressibility of the gel. The flow of water from the gel was observed in simulations to occur in a transient manner, driven by the coupled gradients in gel fluid pressure and gel porosity, where the gel porosity initially decreases in a layer close to the core surface due to reduced aqueous pressure. Gel porosity continued to decrease in layers away from the core surface; the propagation rate was controlled by two main gel parameters: (i) Gel compressibility controlled the pressure gradient within the gel network, and the amount of water transported from the outer part of the gel towards the core surface to balance the pore pressure. (ii) Gel permeability limited how fast water could flow within the gel at a given pressure gradient, thus increasing the time scale of the overall imbibition process.
Alkaline-surfactant-polymer (ASP) flooding is an effective technique to improve oil recovery. It has been applied typically after a water flood. Recently, there has been a successful field test where an ASP flood was conducted after a polymer flood. Is the ASP flood after a polymer flood more effective than an ASP flood after a water flood? It is difficult to conduct this experiment in exactly the same location in a field. The goal of this study is to answer this question in a laboratory heterogeneous quarter 5-spot model. A heterogeneous quarter 5-spot sand pack of size 10″ × 10″ × 1″ was constructed. Two sands with a permeability contrast of 10:1 were packed into a 2D square steel cell. An alkali-surfactant formulation was identified that produced ultra-low interfacial tension with the reservoir oil (27 cp). In one experiment (WF-ASP), waterflood was conducted first followed by the ASP flood. In a second experiment (PF-ASP), polymer flood was conducted first followed by the ASP flood. The ASP formulation and slug size were kept the same. Secondary water flood of the heterogeneous quarter 5-spot recovered 22% OOIP. Post-waterflood ASP flood recovered 32% OOIP additional oil with a cumulative (WF-ASP) oil recovery of 54%. Secondary polymer flood of the same heterogeneous quarter 5-spot yielded 50% OOIP. Post-polymerflood ASP flood recovered 32% OOIP additional oil with a cumulative (PF-ASP) oil recovery of 82% OOIP. The water flood and the subsequent ASP flood swept a large part of the high permeability region and a small part of the low permeability region. The polymer flood swept all of the high permeability region and most of the low permeability region. The subsequent ASP flood swept the polymer-swept regions. These experiments demonstrate that the polymer flood - ASP flood combination is more effective than the water flood - ASP flood combination.
We present the first comprehensive experimental evaluation of CO2 EOR in organic rich shale. Experiments in preserved core demonstrated the potential of CO2 to extract the naturally occurring oil in organic rich shale reservoirs, whereas tests in re-saturated core plugs were used to compute accurate recovery factors, and evaluate the effect of soak time, operating pressure, and the relevance of slim-tube MMP on recovery. 18 core-flooding experiments were conducted in sidewall cores from different shale plays.
The cores re-saturated with crude oil, were first cleaned by Dean-Stark extraction, and submitted to porosity and compressibility determination. The re-saturation, confirmed by CT-scanning, was attained by aging the core plugs at high pressure for two to four months. In all experiments, glass beads surrounding core samples were used to simulate the proppant and physically recreate in the laboratory a hydraulic fracture connected to the shale matrix. The slim-tube MMP was measured with CO2, and core-flooding experiments were performed below, close to, and above the MMP. The displacement equipment was coupled to a medical CT-scanner that enabled us to track the changes in composition and saturation taking place within the shale cores during the experiments. Continuous CO2 injection and huff-and-puff were evaluated using soak time from zero to 22 hours. Fixed reservoir temperature was used in all the experiments.
Recovery factors ranged from 1.7 to 40%. The wide variation was the result of different experimental conditions for pressure and soak time. Both operational parameters were found to significantly affect the recovery. Increasing soak time at constant pressure consistently resulted in significant increase in recovery. The increase varied from 78 to 464% for different pressures and oil composition. Similarly, increasing operating pressure at constant soak time resulted in significant increase in recovery factor from 44 to 338% depending on soak time and oil composition. Unlike the typical response during CO2 EOR in conventional rocks, in organic rich shale, further pressure increases beyond the slim-tube MMP continued to increase the recovery factor significantly. In all runs, almost all oil recovery occurred within three days from the start of the experiment, and in all huff-and-puff tests the highest rate of recovery was observed in the first cycle, implying oil recovery with CO2 is a fast process, in comparison to oil re-saturation of the samples which occurs at a significantly slower rate.
This investigation demonstrates CO2 EOR is a technically feasible method to extract significant amounts of crude oil from organic rich shale reservoirs and it provides operational understanding of how to manage pressure and soak time to maximize recovery. The recovery factors obtained in this investigation, in the context of the vast reserves of crude oil contained in organic rich shale, can sustain a second shale revolution and further capitalize oilfield infrastructure.
Mukherjee, Biplab (The Dow Chemical Company) | Patil, Pramod D. (The Dow Chemical Company) | Gao, Michael (The Dow Chemical Company) | Miao, Wenke (The Dow Chemical Company) | Potisek, Stephanie (The Dow Chemical Company) | Rozowski, Pete (The Dow Chemical Company)
Steam injection is a widespread thermal enhanced oil recovery (EOR) method to increase oil mobility. The introduction of steam heats the reservoir, ultimately lowering oil viscosity and in turn enhancing heavy oil recovery. In the steam injection process, recovery of oil is limited by steam channeling due to reservoir heterogeneities. Early breakthrough implies that there is a large consumption of steam and incomplete reservoir drainage. Injection of surfactant with steam and a non-condensable gas such as nitrogen can generate foam
In this paper, a systematic approach to screen surfactants for field applications at high temperature is presented. A feasibility test was conducted with the surfactant formulation (HSF-X) at target reservoir conditions to understand the thermal stability and adsorption behavior of the surfactant. Investigation found that the thermal decomposition and adsorption of the surfactant on sandstone rock under static conditions was mimimum at 200°C. In core flood testing conducted using silica sand and natural sandstone cores, foam generated by injecting N2 and HSF-X surfactant solution was able reduce steam mobility between 40 to 100 times at 100°C and 10 to 15 times at 200°C more compared to steam mobility in the absence of the foam. Finally oil recovery experiments at 200°C using silica sand cores indicated the ability of the HSF-X surfactant to foam in the presence of oil and enhance recovery of oil (a +20% increase in the original oil in place (OOIP) was observed).
In-situ upgrading (IU) is a promising method of improved viscous and heavy oil recovery. The IU process implies a reservoir heating up and exposition to temperature higher than 300°C for long enough time to promote a series of chemical reactions. The pyrolysis reactions produce lighter oleic and gaseous components while a solid residue remains underground. In this work, we developed a numerical model of IU based on lab experiences (kinetics measurements and core experiments) and validated results applying our model to an IU test published it the literature. Finally, we studied different operational conditions searching for energy-efficient configurations.
In this work, two types of IU experimental data are used from two vertical-tube experiments with Canadian bitumen cores (0.15 m and 0.69 m). A general IU numerical model for the different experimental setups has been developed and compared to experimental data, using a commercial reservoir simulator framework. This model is capable to represent the phase distribution of pseudo-components, the thermal decomposition reactions of bitumen fractions and the generation of gases and residue (solid) under the cracking conditions.
Simulation results for the cores submitted to 370°C and production pressure of 15 bar, have shown that oil production (per pseudo-component) and oil sample quality were well-predicted by the model. Some differences in gas production and total solid residue were observed with respect to laboratory measurements. Computer-assisted history matching was performed using an uncertainty analysis tool on the base of the most important model parameters. In order to better understand IU field-scale test results, the Shell’s Viking pilot (Peace River) was modeled and analyzed with proposed IU model. The appropriated grid-block size was determined and calculation time was reduced using the adaptive mesh refinement technique. The quality of products, the recovery efficiency and the energy expenses obtained with our model were in good agreement with the field test results. Also the conversion results (upgraded oil, gas and solid residue) from the experiments were compared to those obtained in the field test. Additional analysis was performed to identify energy efficient configurations and to understand the role of some key variables, e.g. heating period and rate, the production pressure, in the global IU upgrading performance. We discuss these results which illustrate and quantify the interplay between energy efficiency and productivity indicators.
When EOR by polymer flooding is applied in the offshore environment, polymer can be expected to face long reservoir retention times of up to several years. The stability and viscosity retention of the polymer needs to be evaluated throughout this time. Copolymers of acrylamide (AM) and acrylamide tertiary butyl sulfonic acid (ATBS) are a potential polymer choice for relatively high temperature and salinity reservoirs. In this study, it was evaluated how the long term stability of AM-ATBS copolymers can be predicted by accelerated aging tests at elevated temperature.
The main cause of viscosity loss in AM-ATBS copolymer solutions is the hydrolysis reaction turning AM and ATBS groups into acrylic acid (AA). This hydrolysis was studied in strictly anaerobic conditions at 70-120°C. Besides temperature, the impact of ATBS content in the copolymer (15-55 mol%), pH, salinity and divalent ion concentration were studied. In addition, some tests were done with copolymers of ATBS and AA to elucidate the effect of neighboring monomer on ATBS hydrolysis. The reaction progress was followed by 13C NMR and viscosity measurements.
Over a range of conditions, it was observed that AM-ATBS copolymers aged at different temperatures but at otherwise equal conditions, had similar AA/AM/ATBS ratios at equivalent degrees of hydrolysis. Additionally, the shapes of viscosity retention vs. time curves were corresponding in the studied temperature interval after being normalized into the same time scale. It is proposed that AM hydrolysis is the rate determining step, resulting in a similar hydrolysis pathway over a range of temperatures. AM hydrolysis reaction rates were fitted into an Arrhenius function to estimate acceleration at elevated temperatures.
The data generated in this study helps to predict how the results from accelerated tests (e.g. at 100-120°C) correlate to stabilities at lower temperature (e.g. actual reservoir temperature at 70-80°C). Specifically, a practical overview of expected time of stability (viscosity retention >80%) has been generated for AM-ATBS copolymers in seawater.
The objective of this research was to develop a model to predict the optimum phase behavior of chemical formulations for a given oil based on the molecular structure of the surfactants and co-solvents. The model is sufficiently accurate to provide a useful guide to an experimental testing program for the development of chemical EOR formulations. There are thousands of combinations of surfactants and co-solvents that could be tested for each oil, so even approximate predictions are very useful in terms of reducing the time and effort required for testing and for prioritizing the chemical combinations to test that are most likely to yield ultra-low IFT at reservoir conditions. The effects of changing molecular structures (e.g. swapping head groups, swapping hydrophobes, increasing the length of hydrophobes, increasing the number of PO and EO groups, adjusting the ratios of surfactants) are shown. The variables with the greatest impact on the optimum salinity and solubilization ratio were identified, and methods are proposed to shift the optimum salinity and the optimum solubilization ratios in any desired direction. The structure-property model was developed and tested using a large dataset consisting of 684 microemulsion phase behavior experiments using 24 oils. The chemical formulations used 85 surfactants and 18 co-solvents in various combinations. Both optimum salinity and optimum solubilization ratio (and thus IFT) are modeled whereas other models have focused almost exclusively on the optimum salinity. Predicting the optimum solubilization ratio is actually of more value because of its relationship to IFT. The models include the effects of co-solvent partitioning, soap formation and the molecular structure of both the surfactants and co-solvents.