Abstract

C-CORE of St. John's, Canada, has established CIRUS - Consortium for Industrial Research in the Use of Space whose mandate is to provide benefits to industry for the energy and environment sectors. Research to date has focused on enhanced oil recovery and contaminant transport by the study of fluid physics in microgravity. Three experiments performed by CIRUS members in ground-based or parabolic flight programs have been chosen for further development. These experiments are combined in a Get Away Special (GAS) container which will fly on board NASA's space shuttle. The development program for the GAS container is entitled MIRROR - Microgravity, Industry Related Research for Oil Recovery. Research projects in the MIRROR program include the study of diffusion coefficients of crude oil (DCCO), foam stability in the absence of gravity drainage and capillary flow in porous media. This paper describes the development and potential benefits of the DCCO and foam stability projects.

INTRODUCTION

In 1990 a task force was convened by the Commercialisation Office of the European Space Agency to examine the potential for the industrial use of space. The result was a report defining RADIUS (Research Associations for the Development of Industrial Use of Space) and requests for proposals for the establishment of RADIUS programs from ESA members and associates. C-COPE of St. John's, Newfoundland in Canada was awarded the honor of forming the first RADIUS for its proposed consortium dedicated to energy and environmental research. C-CORE subsequently established CIRUS - Consortium for Industrial Research in the Use of Space - whose mandate is to provide benefits to industry in the energy and environmental sectors by performing fluid physics experiments in reduced gravity. CIRUS research so far has focused on oil recovery science and contaminant transport. The use of a reduced gravity environment in performing the experiments allows all factors affecting fluid flow to be isolated and quantified without the masking effects of gravity. The resulting data are then used in empirical modelling. Members of the consortium are : C-CORE, St. John's, Canada; Microgravity Research Center, Brussels, Belgium; Norwegian Geotechnical Institute, Oslo, Norway; Microgravity Advanced Research and Support Center, Naples, Italy. CIRUS nodes include the Geotechnical Research Centre (McGill University, Montreal, Canada), the Petroleum Recovery Institute (Calgary, Canada) and ANS Technologies (Fredericton, Canada). One of the main activities of CIRUS at present is the development of three experiments scheduled to fly on board NASA's Space Shuttle. This program, entitled MIRROR (Microgravity, Industry Related Research for Oil Recovery), has obtained support from the oil industry in Europe and Canada. The research projects combined into the MIRROR program are the study of diffusion coefficients of crude oil (DCCO), foam stability in the absence of gravity drainage and capillary flow in porous media. The MIRROR program will provide long-duration, high-quality reduced gravity conditions for the performance of advanced experiments on a space shuttle platform called GAS (Get Away Special). This paper will describe the experimental development and potential benefits of two of these projects - DCCO and foam stability.

DCCO

As part of a concerted effort to aid in the understanding of the mechanics of crude oil extraction and reservoir characterisation, the Microgravity Research Center (MRC), in conjunction with ELF Aquitaine in France, is performing two distinct investigations into mass transport phenomena in crude oils. These experiments are the first stages in a large program aimed at studying diffusion in crude oil. The SCCO program (Soret coefficients of crude oil) was initiated in 1993 at MRC and has over the past two years been managed by CIRUS with significant support from ELF. A second part of the program is a study of diffusion coefficients of crude oil (DCCO) Lack of understanding of in-situ characteristics of crude oils is a major issue for predicting the natural hydrocarbon extraction conditions. Present modelling methods are based on pressure-temperature equilibrium diagrams and on gravity differentiation (the static model). The escalating costs of exploration are forcing oil companies to develop models which predict reservoir production more accurately. Thermodynamic laws are being used increasingly to estimate quantities and composition of reservoir hydrocarbon fluids but a detailed knowledge of fluid behavior at reservoir temperature and pressure conditions is required. Diffusion is evident in all fluid mixtures (or solutions) for which the concentration distribution is not homogenous. In systems with a concentration gradient, the amplitude of the resulting mass flux depends on the chemical species of the system. The ratio between the mass flux and the concentration gradient is described by the diffusion coefficient - a simple, linear relation for binary mixtures. However, there is a lack of diffusion data on most mixtures of interest to the oil industry. For multi-component mixtures, the formalism is more complicated and the interaction theory of all chemical species present in the mixture is still not developed. Much work is being performed presently to understand species behavior in systems far from concentration equilibrium and to determine the diffusion process in multi-component mixtures. The use of numerical models in oil reservoir characterisation requires reliable values for the relevant diffusion coefficients. Most authors stress the fact that there exist only few experimental values for diffusion coefficients of crude oil and as a result the use of the models is difficult and often provides inaccurate results. To complicate matters further, the process of thermodiffusion is also evident and contributes to mass flux within the system. Thermodiffusion (Soret diffusion) is proportional to the magnitude of the temperature gradient and also depends on the chemical species present. When modelling mass flux in liquid systems both diffusion processes need to be considered. In most cases where reservoir field measurements are performed natural convection is included as a mechanism responsible for the observed concentration profiles. However, when a static model approach is used in numerical correlation unexpected results are obtained. Buoyancy- driven convection is well understood so that more accurate estimates for the diffusion processes in such multi-component systems are required. The coupling between natural convection and the diffusion processes is termed thermogravitation. This mechanism is believed to be capable of efficiently redistributing the hydrocarbon species in a porous medium over a geological time scale. To better understand this dual process, it is necessary to understand each element separately. The DCCO program provides the opportunity to measure diffusion coefficients of a fluid in ternary fluid mixtures. This experiment will complement the Soret coefficient investigation by providing data needed for complete understanding of species distribution in crude oil reservoirs. In addition to the rationalisation of a convection-free environment, there are other fundamental reasons for using a reduced gravity environment for performing the DCCO experiment. The ultimate application of the experimental data is the characterisation of liquids in a porous medium. Therefore, the effects of such a medium must be considered. A porous matrix can be defined as a network of interconnected small channels, of various widths, with two main descriptive parameters - porosity and permeability. The relatively small size of the pores in the porous medium of an oil reservoir mean that capillary phenomena and surface wetting are important. If the saturating fluid is a mixture of two immiscible liquids, such as oil and water, another difficulty is introduced. On earth, fluid stratification is expected to cause heavier liquids to fall while the light components rise to the top. However, the interaction of capillary forces, wetting and gravity segregation result in unquantifiable convective mixing and fluid distributions which are beyond intuitive forecasting. In a microgravity environment, gravity segregation is negligible and wetting characteristics will reflect only the capillary behavior of the porous medium/multi-component liquid system. In this condition, even if complete understanding of the phenomena is difficult, it is known that the process depends upon simple thermodynamic factors like pressure, surface tension and wetting angles without competition from external gravity forces. It has been demonstrated that to perform isothermal diffusion and thermodiffusion coefficient measurements near the critical demixation point, microgravity conditions are mandatory. Previous experiments show important discrepancies in the dependence of these diffusion coefficients with respect to near critical conditions. This may be due to residual convection.induced by density Gradients [Montel 1994 and Faissat 1994]. The method for the DCCO instrument is to place the liquid to be measured in a transparent quartz cell and illuminate it with a coherent plane wave beam (diode laser). An identical beam passes through a reference cell. Recombining these beams in a special prism causes them to interfere with one another due to phase differences resulting in an interference pattern which contains information about refractive index variations in the beam optical paths. The DCCO apparatus initially envisioned for the MIRROR program was a modification of one used by MRC on a TEXUS sounding rocket experiment. The apparatus used for the TEXUS experiment was based on a spatial differential interferometer using a helium neon laser as its illumination source. Although the baseline for the MIRROR DCCO experiment at first used this same design, substituting a diode laser for the He-Ne laser, it has since been shown that measurements of refractive index in three-fluid mixtures gives rise to ambiguous results in determination of the liquid quantities. To avoid these ambiguities, two different laser wavelengths are used. A Fizeau interferometer is used to help reduce external disturbances on the interferograms (Figure 1 is not included in this version of the document). The diode laser wavelengths are in the visible (635 run) and near infrared (740 run). Interferograms are recorded by a photographic camera loaded with film sensitive to the visible and near infrared radiation. At a given measurement time, one picture is taken at 635 run with the other laser turned off. Then the first laser is turned off, the second turned on and a second picture is taken at 740 nm. The use of photographic emulsions instead of a CCD camera will allow better resolution. Nine sample cells are used with different concentrations of ternary liquid mixtures. The cells are mounted on a rotation table and sequentially analysed by rotating the next sample cell into the optical path every 5 hours. Photographs are taken at each wavelength every 8 minutes. The quartz sample cell is divided in half, separated by a thin stainless steel gate. On one side is the diffusing fluid while on the other is a three-fluid mixture. The diffusing fluid is chosen from N-dodecane, tetrahydronaphthalene and isobutylbenzene, which have similar fluid behavior to that of crude oil light fractions. For each cell, the mixture is one of various ratios of these three fluids. After the sample cell is rotated into the optical path, the gate is withdrawn, starting the diffusion of the fluid through the mixture. In summary, the industrial problem that is being addressed in this project is improvement of the thermodynamic models which predict oil reservoir conditions and corrections to irreversible processes such as gravitation and thermal forces. These models are used by oil companies for both exploration and production. The industrial objective in improving the models is enhancement of oil recovery by better prediction of extraction conditions. The origin of the SCCO/DCCO experiment series stems from research performed at ELF Aquitaine in France, indicating, the importance of isothermal diffusion and thermodiffusion in the prediction of potential reservoir yield. The ability to predict accurately the position and condition of a hydrocarbon reserve enables a company to allocate the proper amount of resources to the project under consideration. The result is efficient use of technology and manpower coupled with more accurate financial forecasting. If the prediction models are sufficiently accurate a detailed financial audit of the project, even before it has started, will determine the level of investment the company will require for full exploitation of the site. In varying instances this could result in the decision to continue with extraction or move on to other sites.

FOAM STABILITY

As part of an initiative based in Canada, the Petroleum Recovery Institute (PRI) of Calgary, Alberta, has been working with CIRUS on foam stability experiments in the absence of gravity drainage. This project is an extension of the considerable amount of research performed at PRI for industry during the past decade. The objective of the project is to study surfactant foam stability in the presence of hydrocarbons. This information is useful for the development of oil recovery techniques that use foam injection technologies. The research techniques used to study foams are constantly challenged by the forces of gravity which mask other phenomena of interest, such as surface elasticity. The use of a reduced gravity environment allows foam stability to be measured without any contribution from gravity-induced drainage. Foams have long been of great practical interest because of their widespread occurrence in everyday life. In soaps, plastics and even metals, the ability to control foams has resulted in a wide range of industrial products. Foams in petroleum may not be as familiar but they have a similarly widespread, long-standing, and important function in this industry [Schramm 1994]. In the hydrocarbon resource industry, foams can be applied at all stages in the petroleum recovery and processing activities including oil well drilling, reservoir injection, oil well production and process plant foams. An interesting point to note is that in the petroleum industry foams may be desirable or undesirable depending on the context in which they are used. In many instances naturally-developed foams hinder the production of an oil well, while in other instances the purposeful introduction of foams will aid in reservoir yield. Thus, there is a two- fold benefit in controlling their occurrence and properties. All reservoir oil recovery processes are aimed at displacing crude oil from the rock pores in which it resides and driving that oil to a well from which it can be recovered. Primary recovery, in which oil flows naturally due to pressure in the reservoir, and secondary recovery, in which water (typically) is injected to raise reservoir pressure and cause flow, still leave behind large amounts (up to 70%) of oil in place. Improved oil recovery methods, usually chemical in nature, are needed to get at this large fraction of oil remaining in a reservoir. There are two main factors involved in improving oil recovery. One is the areal sweep efficiency, which is a function of fluid mobilities. Addition of polymer to injection water is one way to improve the sweep efficiency. The second factor is microscopic trapping, in which capillary forces cause trapping of oil in capillaries of the rock. In water-wet rock this causes oil to be trapped in the larger pores. Mobilisation of the trapped oil depends on altering the balance of viscous to capillary forces, which is reflected in the well-known capillary number concept. In a typical case one has to increase the capillary number by three or more orders of magnitude to significantly improve oil recovery over levels achieved by water flooding. There are substantial limits to what can be achieved by increasing injected fluid velocity with pumps or by increasing fluid viscosity with polymer additions. Dramatic improvement can be achieved by reducing interfacial tension through the injection of surfactants but this is a relatively expensive process that requires optimisation. There are a number of improved oil recovery processes presently under development, most of which involve some combination of adjustments to sweep efficiency and/or microscopic displacement efficiency. CIRUS' proposed foam stability research is aimed at answering questions about the mechanisms important to the stability of foams that are being developed for the purpose of improving the sweep efficiency of gas injection processes. This is connected to CIRUS' surface and interfacial tension research which is aimed at answering questions about mechanisms involved in the dynamic interfacial tension lowering observed in chemical (surfactant, alkali/surfactant or alkali/surfactant/polymer) systems that are being developed for the purpose of improving microscopic displacement efficiency. These aspects actually converge in the case of some of the foam systems recently patented by PRI, which can both improve sweep efficiency and increase incremental oil recovery by microdisplacement. As with knowledge to be gained from the other MIRROR experiments, the work performed in the foam stability studies is applicable to the development of useful numerical models. However, unlike the other projects, these studies can result in the development of new products, such as surfactants, which can be used in the process of oil recovery. This type of product will become increasingly important as current oil reserves are used and the more difficult reserves are targeted for exploitation. Some of the world's largest reserves are also the most difficult to exploit due to the type of matrix within which the hydrocarbon is found. As in the case of the Tar Sands of Alberta, the hydrocarbon species are integral with a matrix of a thick, unconsolidated, -ranular medium, from which fluid extraction is very difficult and expensive. By developing advanced methods of extraction the oil industry will be able to exploit this type of reserve with an estimated worth on a global scale equivalent to billions of dollars. A foam structure can always be formed in a liquid if bubbles of gas are injected faster than the liquid between the bubbles can drain away. Even though the bubbles coalesce as soon as the liquid between them has drained away, a temporary dispersion is formed. An example is the foam formed when bubbles are blown vigorously into a viscous oil. Such a foam is comprised of spherical, well-separated bubbles. When two or more bubbles come together, coalescence occurs rapidly without detectable flattening of the interface between them, i.e. there is no thin film persistence. It is the adsorption of surfactant at the gas-liquid interface that promotes thin film stability between the bubbles and lends a certain persistence to the foam structure. Here, when two bubbles of gas approach, the liquid film thins down to a persistent lamella instead of rupturing at the point of closest approach. The spherical bubbles become transformed into a matrix of polyhedral foam cells, almost but not quite regular dodecahedra. The arrangements of films coming together at equal angles results from the surface tensions, or contracting forces, along the liquid films. In carefully controlled environments, it has been possible to produce surfactant-stabilised, static bubbles and films with lifetimes on the order of months to years. Most foams having any significant persistence contain a foaming agent which is needed to reduce surface tension and to form a protective film at the bubble surfaces that acts to prevent coalescence with other bubbles. In determining, foam stability one considers stability against two different processes: film thinning and coalescence (film rupture). In film thinning two or more bubbles approach closely; the liquid films separating them thin, but the bubbles do not actually touch each other and there is no chance in total surface area. In coalescence, two or more bubbles fuse together to form a single larger bubble. In foam terminology, the thin liquid film(s) rupture, reducing the total surface area. Foams are thermodynamically unstable (we use the term stable to mean relatively stable in a kinetic sense) so it is important to distinguish the degree of change and the time scale. The stability of a foam is determined by a number of factors involving, both bulk solution and interfacial properties: gravity drainage, capillary suction, surface elasticity, bulk and surface viscosity, electric double layer repulsion, dispersion force attraction and steric repulsion. Immediately after foam generation, there is a tendency for liquid to drain from the bubble interfaces under the force of gravity. The draining liquid flows downwards through the interior of the foam lamellae. Eventually the gas bubbles will no longer be approximately spherical and relatively planar lamellae will separate polyhedral-shaped bubbles. At this point the capillary forces become competitive with the forces of gravity. Interfacial tension causes a pressure difference to exist across a curved surface, the pressure being greater on the concave side (i.e. on the inside of a bubble). There is a pressure variation between the Plateau borders in a foam, where the radius of curvature is relatively small, and in the more laminar part of foam lamellae, where the radius of curvature is relatively large. Liquid flow in response to the pressure difference just described is known as Laplace flow (or capillary flow) and is a mechanism for film thinning and possible rupture. However, a restoring. force can be present through the Gibbs-Marangoni effect. A foam film must be somewhat elastic to be able to withstand deformations without rupturing. The surface chemical explanation for film elasticity comes from Marangoni and Gibbs. If a surfactant stabilised film undergoes sudden expansion, then the expanded portion of the film must have a lower degree of surfactant adsorption than unexpanded portions because the surface area has increased. This causes an increased local surface tension which provides increased resistance to further expansions. Unchecked, further thinning would ultimately lead to the rupture of the film. A local rise in surface tension produces immediate contraction of the surface. The surface is coupled by viscous forces to the underlying, liquid layers. Thus, the contraction of the surface induces liquid flow in the thin film from the low tension region to the high tension region. The transport of bulk liquid due to surface tension gradients is termed the Gibbs-Marangoni effect and provides a resisting force to film thinning. This kind of resisting force only exists until the surfactant adsorption equilibrium is re-established in the film, a process that may take place within seconds or over a period of hours. in thick films this can take place quite quickly. However, in thin films there may not be enough surfactant in the extended surface region to re- establish the equilibrium quickly, requiring diffusion from other parts of the film. The restoring. processes are then the movement of surfactant, along the interface, from a region of low to hi-h surface tension, and the movement of surfactant from the thin film into the now depleted surface region. Thus, the Gibbs-Marangoni effect provides a force to counteract film rupture, but may be significant mainly for either rapid deformations or for stabilising very thin films. The first requirements for effective foam stability are surface tension lowering and surface elasticity. Other considerations include bulk and surface viscosity, electric double layer repulsion, dispersion force attraction and steric repulsion. A greater elasticity tends to produce more stable bubbles. If the restoring force contributed by surface elasticity is not of an appropriate magnitude, then persistent foams may not be formed due to the overwhelming effects of the gravitational and capillary forces. The object of the CIRUS experiments is to eliminate gravitational forces to enable investigators to piece together the underlying phenomena which make up this complicated process. The principle behind the foam stability instrument is simple - mix various target fluids with surfactants or oil, combine the mixture with nitrogen gas, send it through a foam Generator and view the resulting foam in a transparent cell. The fluids are contained in syringes that inject precise amounts of liquid under computer and stepper motor control. The fluids are mixed and passed into a measured stream of nitrogen gas through a Berea disk into a thin, glass-walled chamber. The foam in the chamber is allowed to stabilise and then the half-life (the time it takes for the foam to decrease to half its original size) is measured. The fluids will be an aqueous sodium chloride solution; a light to medium gravity crude oil; I to 2% by mass polyacrylamide in water also containing sodium chloride and surfactant; two different surfactants such as sodium dodecyl sulfate. The foam will be photographed using a camera similar to the one used for the DCCO experiment. Each foam sample will be monitored over a 2-3 hour period and 7 different foam samples (varying combinations of fluids and surfactants) will be examined.

SUMMARY

Three experiments are being developed for a microgravity environment, designed to examine the diffusion and capillary flow behavior of fluids and stability of foams of interest to the oil industry. These experiments will be incorporated in a GAS canister to be flown on NASA's space shuttle. The data from these experiments will assist in the development of numerical models and chemical products intended to improve oil recovery from reservoirs.

Acknowledgements

The authors would like to acknowledge the European Space Agency, the Canadian Space Agency, the Government of Newfoundland, the Natural Sciences and Engineering Research Council of Canada and ELF Aquitaine for their support of the MIPROR projects. The Canadian Space Agency should also be singled out for obtaining the GAS flight opportunity for the MIRROR program.

References

Faissat, B., K. Knudsen, E.H. Stenby and F. Montel (1994) "Fundamental Statements about Thermal Diffusion for a Multicomponent Mixture in Porous Medium", in Fluid Phase Equilibria, 100: (209-222).

Montel, F. (1994) "Importance de la Thermodiffusion en Exploration et Production Petrolifere", in Entropie a30 (184/185): 83-93.

Schramm, L.L. Ed. (1994) Foams: Fundamentals and Applications in the Petroleum Industry, American Chemical Society; Washington, DC.

Citation

Hart, D., Hansen, N., Legros, J-C., and Schramm, L.L. (1997) Microgravity, industry related research for oil recovery: Space Technology and Applications International Forum (STAIF 97), Second Conference on Commercial Development of Space, Albuquerque, NM, AIP Conference Proceedings 387, Part Two, pp. 761-766, Jan. 26-31.

 Home Page    Reprint Ordering / More Information