Barnes, Julian Richard (Shell Global Solutions) | Smit, Johan (Shell Global Solutions International ) | Smit, Jasper (Shell Global Solutions International ) | Shpakoff, Greg (Shell Global Solutions) | Raney, Kirk H. (Shell E&P Co.) | Puerto, Maura (Rice University)
Accurate laboratory screening of surfactants for their ability to give ultra-low interfacial tensions in oil/brine systems is important as a pre-cursor to laboratory core flow tests and surfactant flooding processes in the field. Screening is usually judged by visualisation of middle-phase micro-emulsions in oil/brine systems. Three laboratory methods are described which enable the phase behaviour of oil/water systems containing surfactants to be more safely visualised and measured in glassware at higher temperatures. Higher temperature test conditions result in significant vapour pressures from crude oil and water, and some glass tube test methods currently used in the industry may not be appropriate from a laboratory safety standpoint. The new methods have been verified in our laboratories for higher temperature use and provide useful screening methods for higher temperature reservoirs (up to 150°C).
In enhanced oil recovery (EOR) with surfactants, the mobilisation of residual oil saturation is achieved through the crude oil/water interfacial tension (IFT) being reduced to sufficiently low level that the capillary number is large enough to overcome capillary forces and allow the oil to flow1. Due to the well-established relationship between the micro-emulsion phase behaviour and IFT2 it is common in the industry to screen surfactants and their formulations for low IFT through laboratory-based oil / water phase behaviour tests3, 4. This approach works well for tests carried out at room temperature and slightly higher, but at higher temperatures (appropriate for screening of hotter reservoirs) there may be safety issues. Specifically, conventional sealed glass tube test methods may be problematic from a laboratory safety standpoint at higher temperatures due to higher vapour pressures from water and crude oil.
This poster presents the results of the evaluation of surfactants using improved phase behaviour experimental methods for very high temperatures (up to 150°C). Optimal salinities and solubility ratios have been measured at high temperatures. An oral paper to be presented at the Improved Oil Recovery symposium, Tulsa, April 2008 discusses the results of these phase behaviour phase tests5.
2. Phase test methods suitable for higher temperatures
The following three glassware methods have been developed for phase behaviour tests at high temperatures. The method principle in each case is to measure the volumes of water, oil and any emulsion phases at particular test temperatures as salinity is increased causing a transition in phase behaviour from Windsor type I to type III to type II. Alternatively, different surfactants and combinations of surfactants can be screened for particular crude oil and brine conditions. In the case of tests where salinity is varied, the volumes of oil, water and micro-emulsion phases give the oil and water solubilisation parameters (i.e. volumetric ratios of oil/surfactant and water/surfactant) are plotted against salinity. The intersection of these ratios gives the optimal salinity and the solubilisation parameter at that salinity. The phase test methods give additional qualitative information about the micro-emulsions formed including their colour, appearance and an indication of their viscosity.
Enhanced-Oil Recovery (EOR) for asset acquisition or rejuvenation involves intertwined decisions. In this sense, EOR operations are tied to a perception of high investments that demand EOR workflows with screening procedures, simulation and detailed economic evaluations. Procedures have been developed over the years to execute EOR evaluation workflows.
We propose strategies for EOR evaluation workflows that account for different levels of available information. These procedures have been successfully applied to oil property evaluations and EOR applicability in a variety of oil reservoirs. The methodology relies on conventional and unconventional screening methods, emphasizing identification of analogues to support decision making. The screening phase is combined with analytical or simplified numerical simulations to estimate full-field performance while maintaining rational reservoir segmentation procedures.
This paper fully describes the EOR decision-making procedures using field case examples from Asia, Canada, Mexico, South America and the United States. The type of assets evaluated includes a spectrum of reservoir types, from oil sands to light oil reservoirs. Different stages of development and information availability are discussed. Results show the advantage of flexible decision-making frameworks that adapt to the volume and quality of information by formulating the correct decision problem and concentrate on projects and/or properties with apparent economic merit.
Our EOR decision-making approaches integrate several evaluation tools, publicly or commercially available, whose combination depends on availability and quality of data. The decision is laid out using decision-analysis tools coupled with economic models and numerical simulation. This allows integrated teams to collaborate in the decision making process without over-analyzing the available data. One interesting aspect is the combination of geologic and engineering data, minimizing experts' bias and combining technical and financial figures of merit rationally. The proposed methodology has proved useful to screen and evaluate projects/properties very rapidly, identifying whether or not upside potential exists.
We present a sharp-interface mathematical model of CO2 migration in saline aquifers, which accounts for gravity override and capillary trapping. The major differences with respect to previous investigations is that we model the shape of the plume during the injection period, we account for regional groundwater flow during the post-injection period, and we introduce rigorously the reduction in water mobility due to trapping of the CO2.
The model leads to a nonlinear advection-diffusion equation, in which the diffusive term is due to buoyancy forces, not physical diffusion. The three key dimensionless groups are the mobility ratio between the injected and initial fluids (M), the gravity number (Ng), and a newly defined trapping coefficient (G). For the case of interest in geological CO2 storage, in which the mobility ratio is much smaller than 1, the solution is largely insensitive to the value of the gravity number. Under these conditions, the mathematical model can be simplified to a hyperbolic model.
We present a complete analytical solution to the hyperbolic model that includes the injection, early post-injection, and late post-injection periods. Despite the fact that the solution involves the interaction of a sharp imbibition front with a drainage rarefaction front, it admits a closed-form expression. The main outcome of the analytical developments presented here is a formula that predicts the ultimate footprint on the CO2 plume, and the time scale required for complete trapping. Both a formula that predicts the ultimate footprint on the CO2 plume, and the time scale required for complete trapping. Both quantities depend strongly on the shape of the plume at the end of the injection period, which must - therefore - be modeled.
A second application of the analytical solution is a formulation for upscaling the capillary trapping coefficient from the laboratory scale to the basin scale. Explicit expressions are given for the megascopic and gridblock-effective trapping coefficients, as functions of the local trapping coefficient, the mobility ratio, and the grid resolution. Although the expressions derived are based on a one-dimensional sharp-interface model, we anticipate that they will have broader applicability to injection scenarios with unfavorable mobility ratio and dominated by gravity override.
Wassing, B. (Petroleum Development Oman) | van Wunnik, J. (Petroleum Development Oman) | Warrlich, G. (Petroleum Development Oman) | Lamki, A. (Petroleum Development Oman) | Johnson, T. (Petroleum Development Oman) | Gittins, J. (Petroleum Development Oman) | Riyami, M. (Petroleum Development Oman)
This paper was prepared for presentation at the EAGE 14th European Symposium on Improved Oil Recovery held in Cairo, Egypt, 22-24 April, 2007. It was selected for a "Best of Cairo?? presentation at the 2008 SPE/DOE Improved Oil Recovery Symposium.
The subject field is a fractured carbonate reservoir located in the Ghaba Salt Basin in Oman containing heavy, viscous oil. The field was discovered in 1972 and has been producing since 1976 through a process of mainly fracture depletion and natural water-influx. Without the application of an EOR technique the recovery will be low as the processes that displace oil from the matrix reservoir rock are adversely affected by the rock and fluid properties, particularly the high oil viscosity. Following the success of the steam-assisted GOGD test in an analogous nearby field, the field is being screened for feasibility of this thermal recovery process. In the steam-assisted GOGD process the fracture network plays a dual role : as distribution system of crestally injected steam to heat up the oil in the matrix and as offtake point deeper down in the column where the oil collects in the fracture oil-rim, after having drained predominantly through the matrix. Critical data on the fracture network and thermal PVT properties are collected and integrated in dynamic models to assess the oil drainage rate under steam injection.
In modeling the SA-GOGD process, care needs to be taken to correctly represent the effect of unevenly heated matrix blocks on the oil viscosity and hence drainage rates. Oil close to the steam filled fractures will heat up more quickly than in the center of the matrix blocks and the former will start to drain faster than the latter. Ignoring this effect and using viscosity values based on averaged temperatures and established viscosity-temperature relationship, as is usually done in dual permeability simulation, will underestimate the SA-GOGD drainage rates. This underestimation will increase with a larger difference between the cold and hot oil viscosity and is also more relevant to early times (economics). A new method has been developed to conserve the 1D-drainage rate of unevenly heated matrix blocks based on an effective or "pseudo?? viscosity correlated with gridblock-averaged temperature.
Horizontal and multi-lateral wells have become increasingly important and represent a growing percentage of production wells. They are used to maximize the well to reservoir contact and improve oil recovery in a cost efficient manner. This is especially true for offshore fields where these wells are used to drain large areas with limited platform capacities. Commonly, a horizontal well trajectory undergoes undulations that may result in special wellbore flow dynamics. In addition, technologies such as intelligent completions can be used to regulate flow from various perforation intervals or producing laterals.
Our recent field studies required the simulation of special wellbore dynamic behavior specific to horizontal/multi-lateral wells. It is a significant challenge to capture such behavior in a simulation model. This paper covers the following issues of horizontal/multi-lateral well simulation.
The simulation of these issues requires special techniques, such as the use of a discritized well model. A discritized well is represented by multiple segments along the path of the wellbore. This option provides flexibility in the control of appropriate parameters by segment to properly simulate the different issues that arise with horizontal/multi-lateral wells. In addition, local grid refinement is often needed to accurately capture the well trajectory and detailed fluid flow and pressure profiles.
This paper discusses the previously mentioned horizontal/multi-lateral well fluid and completion dynamics and the approach used to simulate them. Simulation results for successful actual field studies as well as semi-synthetic examples are presented.
Dong, Huanzhong (Daqing Research Institute) | Fang, Shufen (Harbin Engr U.) | Wang, Dongmei (Daqing Research Institute) | Wang, Jiaying (Daqing Oilfield Co. Ltd.) | Liu, Zhen Lin (Daqing Research Institute) | Hou, Weihong
This paper describes successful experiences employed during polymer flooding at Daqing that will be of considerable value to future chemical floods, both in China and elsewhere. Based on laboratory findings, new thoughts have been developed that expand conventional ideas concerning favorable conditions for mobility improvement by polymer flooding. Particular advances integrate reservoir engineering approaches and technology which is elementary for successful application of polymer flooding. These include: (1) Considering permeability differential among the oil zones and interwell continuity, optimizing the oil strata combination and well pattern design. (2) The injection procedures and injection formula are the key points when designing a polymer flood project. These points include:profile modification is needed before polymer injection and zone isolation is of value during polymer injection, higher molecular weight of the polymer used in the injected slugs, large polymer bank size, higher polymer concentrations and injection rate based on the well spacing and injection pressure. (3) Characterizing the entire polymer flooding process in five stages, with its dynamic behavior distinguished by the water cut changes. Additional techniques involved with reservoir engineering should also be considered, such as dynamic monitoring using well logging, well testing, and tracers. Effective techniques are also needed for surface mixing, injection facilities, oil production, and produced water treatment.
Continuous innovation and effective response to new challenges must be a priority during polymer flooding. New directions and opportunities at Daqing will (1) explore the feasibility of polymer flood application in poorer ("third-class??) strata, (2) to identify new polymers to suit portions of the reservoir with higher temperatures and higher water salinities, and (3) continually see improvements in our approach to polymer flooding.
We examine buoyancy-driven multiphase flow when the less dense phase is placed below the other phase in a heterogeneous domain. After generating geostatistical realizations of permeability, we apply the Leverett J-function so that each grid block has a drainage curve (Pc vs Sw) physically consistent with its permeability. The behavior of the displacement front depends strongly on the correlation structure of the heterogeneity and upon the magnitude of the mean entry pressure. This behavior is of particular interest for assessing the degree of immobilization of anthropogenic CO2 injected into an aquifer.
In a relatively homogeneous domain, capillarity is a second-order effect. It damps the instability of the rising CO2 front and smooths the shape of the plume. As the heterogeneity of the aquifer increases, capillarity begins to dominate buoyancy. Regions with smaller permeability that would readily conduct single-phase flow can completely block rising CO2, simply because the capillary entry pressure in these regions is somewhat larger than in neighboring regions. These local capillary barriers prevent CO2 from rising and cause it to move laterally. The disruption can be so extreme that above-residual saturations of CO2 become trapped below these barriers. These local accumulations respond differently when the top seal of the aquifer is breached. Thus we distinguish them as a new mode of CO2 trapping, dubbed "capillary trapping.?? Overall, in some regions the CO2 follows preferential flow paths determined by the spatial correlation of permeability, while in others capillarity determines the flow path. Though the displacement front is much less uniform, the extent of dissolution trapping remains significant.
In recent years, the use of pore-scale network models has greatly advanced our understanding of solution gas drive processes by accounting for the complex dynamics operating at the microscopic scale. Moreover, it has also been demonstrated that a pore-network model, when suitably anchored to core material is able to provide both qualitative and quantitative descriptions of relative permeability and hydrocarbon recovery. In contrast, many so-called "experimental?? depletion drive relative permeabilities are not measured directly but are generally obtained by history-matching laboratory production data with reservoir simulators, often resulting in very low gas relative permeabilities that are difficult to explain from a physical viewpoint. Although pore-scale network models have been successfully used in the past to match raw production data, the steady-state relative permeabilities calculated from such models commonly predict much slower gas saturation build-up than that found experimentally. Some previous authors have related this low gas saturation build-up to the difference in the definition of critical gas saturation between reservoir simulators and pore-network models. However, the dentritic nature of gas-cluster topology in network models, especially in the presence of other forces, such as gravity or strong viscous pressure gradients, clearly suggests that significant anisotropy may exist in relative permeability due to the balance of forces at this scale.
In the present work, we describe how the naïve process of scaling up steady-state relative permeabilities obtained from pore-scale network models to the laboratory scale may contribute significantly to the difficulty in history-matching experimental production. By considering the influence of the various forces (capillary, gravity, viscous) on the topology of the growing gas clusters and by accurately incorporating anisotropic network-model relative permeabilities, we show that high gas saturation build-up, consistent with experimental observations can be obtained from reservoir simulation.
A number of commercially available polymers have been tested for enhanced oil recovery based upon viscosity, filterability, and surfactant compatibility, and chemical and thermal stability testing has been carried out with some of these as well. Several high molecular weight polymers exhibited high viscosities at salinities up to 170,000 ppm NaCl and with greater than 17,000 ppm CaCl2 present. Polyacrylamide polymers hydrolyze at high temperatures and beyond a certain point are subject to precipitation by calcium. If calcium concentrations can be kept below about 200 ppm, the use of polyacrylamide polymers is feasible up to reservoir temperatures of at least 100 °C. For higher concentrations of calcium, copolymers including AMPS moieties should be considered. Calcium tolerance can be improved with sodium metaborate or by using copolymers of acrylamide and sodium 2-acrylamido-2-methylpropane sulfonate (AMPS). The stability problems at elevated temperatures in the presence of iron can be mitigated by the use of chemicals such as sodium dithionite and sodium carbonate. The polymers tested did not lose viscosity after 220 days of aging at 100 °C with dithionite present.
van den Hoek, Paul Jacob (Shell) | Al-Masfry, Rashid Ahemed (Petroleum Development Oman) | Zwarts, Dirk (Shell Intl. E&P BV) | Jansen, Jan-Dirk (Delft University of Technology) | Hustedt, Bernhard (Shell International E&P) | van Schijndel, Luc (Shell Intl. E&P BV)
It is well established within the Industry that water injection mostly takes place under induced fracturing conditions. Particularly in low-mobility reservoirs, large fractures may be induced during the field life. This paper presents a new modeling strategy that combines fluid-flow and fracture-growth (fully coupled) within the framework of an existing ‘standard' reservoir simulator. We demonstrate the coupled simulator by applications to five-spot pattern flood models, addressing various aspects that often play an important role in waterfloods: shortcut of injector and producer, fracture containment, reservoir sweep. We also demonstrate that induced fracture dimensions can be very sensitive to typical reservoir engineering parameters, such as fluid mobility, mobility ratio, 3D saturation distribution (in particular, shockfront position), positions of wells (producers, injectors), and geological details (e.g. flow baffles). The results presented in this paper are expected to also apply to (part of) EOR operations (e.g. polymer flooding).
Water injection will generally result in rapid injectivity decline unless it takes place under induced fracturing conditions. This is illustrated in Fig. 1-2 1-2, comparing matrix injection of fine-filtered seawater (Fig. 1 1) with fractured injection of heavily contaminated production water (Fig. 2 2). In the former case, regular acidizations are required to keep up well injectivity (in spite of the high water quality), whereas in the latter case, injectivity remains constant over years (in spite of the low water quality). However, important risks associated with waterflooding under induced fracturing conditions are related to potential unfavorable areal and vertical sweep. These risks can be managed if one has a proper understanding of dynamic induced fracture behaviour as a function of parameters such as injection rate, voidage replacement, reservoir fluid mobility and reservoir / injection fluid mobility ratio . In order to enable building and using such an understanding as part of field development planning and of reservoir management, we developed an ‘add-on' fracture simulator to our existing in-house reservoir simulator.
In the past, several attempts were made to address the coupled problem of reservoir simulation and induced fracture growth. Common approaches can be grouped into fully implicit simulators (Tran et al.) where both fluid flow equations and geomechanical equations are solved at the same time on the same numerical grid, and coupled simulators (Clifford et al.) where a standard, finite-volume reservoir simulator is coupled to a boundary-element based fracture propagation simulator. Both approaches are not standard and currently not used in the industry mainly because reservoir models need to be purpose-built, and numerical stability is questionable.
Our approach, as briefly described in , uses a ‘standard' reservoir simulator, thereby enabling reservoir engineers to model induced fracturing around injectors using their ‘standard' reservoir models (sector, full-field). Moreover, our specific methodology of coupling induced fractures to the reservoir via special connections helped to eliminate most of the numerical instabilities that are generally encountered in the coupled (reservoir flow)-(fracture growth) problem.