Geomechanics for the Thermal Stimulation of Heavy Oil Reservoirs

Yanguang Yuan, SPE, Bin Xu, SPE, and Baohong Yang, SPE, BitCan G&E Inc.

Copyright 2011, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Heavy Oil Conference and Exhibition held in Kuwait City, Kuwait, 12–14 December 2011.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
This paper will cover both caprock integrity and reservoir deformation, drawing from our years of experience in working with the heavy oil/oilsands industry in Alberta, Canada. Theoretical principles are described, analytical derivations made and field examples given, all to help illustrate the fundamentals and summarize the learnings in proactive utilization of geomechanics to enhance the reservoir performance and proactive consideration of geomechanics to ensure the caprock integrity. Topics include dilation tendency and fracturing behaviour in the oilsands, major geomechanical work components for the caprock integrity analysis/design, mini-frac tests and nonlinear coupled thermo-hydro-mechanical processes.

Introduction
Canada plays an important role in the global energy industry and the oilsands resource in Alberta, Canada is an increasingly important component to this energy supply. It has recently over-taken the conventional crude in the total production output. Oil sands can be classified as an extra heavy oil reservoir primarily made up of a mixture of sand, water and bitumen. Bitumen is an extremely heavy oil with a viscosity reaching up to 106 cp at in-situ reservoir temperatures and therefore must be liquefied via steaming before it can be produced. According to the Alberta government, using currently available technology and under the current economic conditions, there are 170.4 billion barrels of recoverable oil in the oil sands deposits of Northern Alberta. Moreover, there are 315 billion barrels of potentially recoverable oil.

Approximately 80% of the oil sands are recoverable through in-situ production with only 20% recoverable by mining. A variety of commercial in-situ production methods are currently available, including: (1). Primary recovery where sands and heavy oil are produced together; (2). Conventional enhanced oil recovery methods such as water flood or polymer flood; (3). Cyclic steam stimulation (CSS); (4). Steam assisted gravity drainage (SAGD). Recovery factors for CSS and SAGD are the highest and therefore, they are the most common in-situ bitumen/heavy oil recovery processes in the industry. They are also the focus in the current paper.

Both CSS and SAGD inject steam into the reservoir, heating the bitumen and reducing its viscosity enabling it to flow freely. CSS injects the steam at a high pressure, usually around the original reservoir fracture pressure. These periods of high pressure steam injection are followed by production cycles on the same wells when no steam is injected. SAGD injects the steam at a lower pressure but the steam injection is continuous and a growing steam chamber is formed in the reservoir. While reservoir engineering focuses mainly on the fluid phase, an issue that was often overlooked is the inevitably altered structure of the rock matrix caused by the steam injection. The overburden rock can be impacted as well. Therefore, deformation and potential failure of rock materials, which falls into the domain of geomechanics, becomes an important subject.

In CSS and SAGD, two primary geomechanical issues need to be addressed: reservoir deformation and caprock integrity. Nowadays, the industry proactively utilizes geomechanics to enhance the thermal operation. One primary example is to create a fracture in the oilsands reservoir so that hydraulic conduits are formed. These conduits then serve as staging areas for the controlled injection of steam in the reservoir. Timing the stimulation of properly spaced hydraulic fractures allows for more efficient oil recovery. In this case, geomechanics can create profits for the industry and is highly sought after. On the other hand, the industry is obligated to ensure safe caprock integrity. In these situations, geomechanics may appear to be a undesired burden. However, maintaining an intact caprock is far more important than immediate reservoir performance. As one senior geologist said, the hydrocarbon is an important resource, but we have a far more important resource for mankind — clean groundwater supply (Personal communication, 2008). Fortunately, the industry has taken a proactive stance and the government enforces a vigilant work program on the caprock integrity issue. The following description presents further examples to illustrate the above points.

Two major objectives are planned for this paper: to present the theoretical and practical evidences to support the importance of geomechanics and to introduce the optimum work flow for studying geomechanical issues. First, a brief introduction on the reservoir geology sets the stage to understand the unique geomechanical properties and in-situ conditions in the oilsands. Then, an overview is given about deformation and fracturing behaviour in the oilsands. Subsequently, attention turns to the caprock integrity, explaining the geomechanical work program needed. The next two sections are relevant to both reservoir and caprock: operational and quality-control considerations for mini-frac tests and nonlinear thermo-hydro-mechanical coupling. A summary and discussion is given at the end.The examples are drawn mostly from our own experience although some

The examples are drawn mostly from our own experience although some literatures are also cited for the further support. For decades, geomechanics has been promoted for the in-situ oil sands development in Alberta. Excellent work has taken place in both industry and academia. Initially, it was driven by the need to determine the optimal placing for steam injection. Then, the caprock integrity issue came into play as large-scale commercial production began and unforeseen stresses were placed on reservoirs and their caprocks. The literature to be introduced below is not extensive. Other publications of equal importance exist but are not referenced here as the objective of this paper can be achieved through limited citation.

A general description on the reservoir geology
The following description about the regional and reservoir geology, although brief, is warranted in order to explain the unique geomechanical properties possessed by the oilsands. It will also help appreciate the unique opportunities and challenges met in the in-situ oilsands developments. Most of the following summaries are based on a government publication by Flach (1984).

The oilsands reservoir rocks in Alberta were deposited during the Cretaceous period in the upper and lower Mannville Formation. There are three major oil sand areas in Alberta, namely Athabasca, Cold Lake and Peace River. The Athabasca oil sands area is the largest deposit and holds the most reserve. Most of the recent industrial activities are concentrated in this area and thus, more relevant details are presented below:

  1. The bitumen-bearing formation in the Athabasca area is called McMurray. It is a sequence of uncemented sands and shale that were accumulated in incised valleys on a pre-Cretaceous unconformity. From its bottom upwards, the McMurray deposits were formed in a fluvial environment initially. Such an environment then gave way to estuarine and finally transgressed by a marginal-marine environment. This relatively energetic depositional environment gives rise to a reservoir that is heterogeneous both vertically and laterally. For example, the presence of shale stringers or inclined heterolithic stratification (IHS) in most reservoirs can stop the steam chamber from rising and prevent the bitumen reserve above from being heated and
  2. The Clearwater Formation, a layer of marine shale and sandstone, constitutes the regional caprock to the underlying McMurray reservoirs. Its thickness varies from a few meters to tens of meters. More Cretaceous deposits are also formed above the Clearwater. But their presence and thickness vary greatly across the region. The Cretaceous deposits may be absent in some areas due to the paleo-topography and more importantly to the glacial movement in Quaternary period. The McMurray formation overlies unconformably on carbonate rocks of Devonian age, including limestone and calcareous shale.
  3. The hydrocarbon source, migration and trapping mechanisms for the oilsands reservoir are not well understood. It is generally believed that microbial processes caused the extra heavy and highly viscous bitumen currently present in the reservoirs. The bitumen viscosity at the reservoir temperatures is of million-order centi-poses. The water phase exists in the reservoirs generally as a film around the sand grains. It is thus immobile or initial mobile water saturation is small. Thus, the initial water injectivity is very small if without rock deformation and failure.
  4. In addition, little energy is stored in the reservoirs. It is generally held that McMurray formation is under-pressured although the concept of pore pressure is less clear in such a fluid phase made of highly viscous bitumen and connate water. Therefore, in its virgin state there is essentially no drive mechanism to produce the bitumen from the formations. For the drive, reservoir production relies on the gravity in SAGD or compaction of the dilated sands in CSS. The dilation is induced during the injection cycles.
  5. Note that in all the areas, a continental glacial ice sheet as thick as 2 to 3 kilometers was present during the Quaternary period (e.g., Clark, 1980). This is equivalent to a minimum of 20 to 30 MPa of vertical overburden weight exerted on the Cretaceous deposits. The impact of such glaciation and its subsequent deglaciation on the mechanical properties of both reservoir sands and caprock shales is far-reaching. For example, the Cretaceous materials are all over-consolidated in the geomechanical terms. Strain-weakening and post-peak dilation behaviour dominates under low confining pressures. The glaciation also caused the diagenetic process that altered the grain-grain contacts in the oil sands from the initial tangential to the interpenetrative types currently observed. To a certain degree, pressure solution and crystal overgrowth also occurred in the sands. As will be shown later, such a unique grain-grain contact structure renders the oilsands with a significant dilation tendency that can be an advantage for the reservoir recovery if used properly.
  6. The impact of glaciation-deglaciation sequence should be also felt in the in-situ stress field although the exact details are still unknown. Other post-deposition events such as localized Devonian salt solution, tectonic subsidence and fault re-activation may have also taken place in the areas and may also vary the local in-situ stresses.

Deformation and failure behaviour of the oilsands
Fundamentally, four major challenges are present for the in-situ oilsands development: (1) No or little initial reservoir injectivity due to the extremely viscous bitumen and connate water phase; (2) No or little reservoir drive for the fluid production; (3). Variable lithological sequence in the reservoir; and (4). Perservation of the caprock integrity. Geomechanics, through the mechanisms of deformation and failure of the oilsands or caprock rocks, can help in overcoming all these challenges. This section focuses on the oilsands.A number of laboratory test programs were completed on the mechanical properties of the oil sands. It was discovered that the oilsands materials have a unique interlocked grain-grain contact structure (Dusseault, 1977; Dusseault and Morgenstein, 1977; 1979). Due to the glacial loading in the Quaternary period, the grain contacts in the oil sands have been altered from the typical tangential type to an interpenetrative nature. A large number of grain contacts in the oil sands exhibit long and concavo-convex nature, resulting in an even larger contact area (Barnes, 1980; Barnes and Dusseault, 1981).

A number of laboratory test programs were completed on the mechanical properties of the oil sands. It was discovered that the oilsands materials have a unique interlocked grain-grain contact structure (Dusseault, 1977; Dusseault and Morgenstein, 1977; 1979). Due to the glacial loading in the Quaternary period, the grain contacts in the oil sands have been altered from the typical tangential type to an interpenetrative nature. A large number of grain contacts in the oil sands exhibit long and concavo-convex nature, resulting in an even larger contact area (Barnes, 1980; Barnes and Dusseault, 1981).Consequently, the dilation tendency of the oilsands is significant (Agar, 1984; Kosar, 1989; Wong et al., 1993; Samieh, 1995). This is especially true under low confining pressures. Figure 1 compiles the dilation angles measured from the laboratory

Consequently, the dilation tendency of the oilsands is significant (Agar, 1984; Kosar, 1989; Wong et al., 1993; Samieh, 1995). This is especially true under low confining pressures. Figure 1 compiles the dilation angles measured from the laboratory tests, and shows that the dilation increases with a decreasing confining. It can reach up to 52º at a confining pressure of 200 kPa. The interlocked grain contact structure causes a greater disturbance once the grains are sheared and rolled over each other. This is responsible for the high dilation angles observed.The need to increase the reservoir injectivity naturally calls for hydraulic fracturing stimulation in in-situ oilsands developments. Indeed, this has been subject to a detailed scrutiny. Armed with the laboratory findings described above, a number of simulation efforts were undertaken to determine the impact deformation and fracturing processes in the oilsands have on its in-situ recovery (e.g., Ito, 1984). Many field tests were then carried out and detailed monitoring efforts were used during the tests. For example, post-test

The need to increase the reservoir injectivity naturally calls for hydraulic fracturing stimulation in in-situ oilsands developments. Indeed, this has been subject to a detailed scrutiny. Armed with the laboratory findings described above, a number of simulation efforts were undertaken to determine the impact deformation and fracturing processes in the oilsands have on its in-situ recovery (e.g., Ito, 1984). Many field tests were then carried out and detailed monitoring efforts were used during the tests. For example, post-test coring was done through the fracture path (e.g., Chinna and Agar, 1985) or post-test image log was done on the fractured wells (e.g., Kry et. al., 1992). In other tests, tilt meters were deployed both on the surface and downhole (e.g., Kular et. al., 1988). Pressure and temperature monitoring was used in almost all the tests (e.g., Kry, 1989). In recent years, many mini-frac tests were completed in the reservoir. These tests primarily aim to measure the fracture closure pressure; but some were executed as a full-field geomechanical test to understand the fracturing process in the reservoir (Yuan and Fung, 2008; Xu et al., 2010).

A recent paper by Yuan et al. (2011a) summarizes the major available laboratory and field data, and analyzes the fracture initiation and propagation processes. It concludes that the fracturing in the oilsands is a combination of shear-induced dilation and tensile parting. Dilation refers to a geomechanical phenomenon where the volume of the materials increases when sheared even though the overall stress condition is still compressive. When the dilation occurs, sand grains are still in contact with each other although their original contacts are disturbed by the grains rolling over each other. In the tensile parting, sand grains are detached from each other and no more contacts are present. Figure 2 illustrates the difference between dilation and tensile parting.

In the oilsands, dilation plays a predominant role. It happens first during the high pressure injection. The dilation increases the injectivity in the dilated area and permits further pressure increase. As the injection and pore pressure continue to increase, the effective stress state can eventually reach tension. Given the uncemented nature and thus nearly zero tensile strength in the oilsands, tensile parting ensues as soon as the effective stress state becomes tensile. Away from the high-pressure sources, the transition from start of the dilation to onset of tensile parting can be slow and gradual. The details are shown in Figure 3.Figure 3 also compares the simulated evolving stress state in the oilsands vs. in a hard rock formation during the hydraulic fracturing process. The large cohesion in the hard rock makes the stress state reach the shear failure significantly later than in the oilsands. When the shear failure state is reached, the stress state is already tensile. Although the tensile strength of the hard rock prevents the immediate tensile failure, the shear failure happens in a tensile state1.

Figure 3 also compares the simulated evolving stress state in the oilsands vs. in a hard rock formation during the hydraulic fracturing process. The large cohesion in the hard rock makes the stress state reach the shear failure significantly later than in the oilsands. When the shear failure state is reached, the stress state is already tensile. Although the tensile strength of the hard rock prevents the immediate tensile failure, the shear failure happens in a tensile state1.The combined nature of shear dilation and tensile parting for fracturing in the oilsands is significant in effecting the hydraulic fracturing stimulation. The fracture created is a zone with relatively large thickness. In the fracture zone, micro tensile cracks are present, but they do not connect with each other to form a continuous path. Instead, these micro cracks are embedded in the shear-dilated medium that still belongs to a disturbed continuum. As a result, the fracture zone in the oilsands still possesses certain friction or resistance to the flow, but it disturbs a much wider area. All of these characteristics are different from the conventional tensile fracture created in the hard rock. In the latter, a fracture has two continuous walls separated by a certain aperture. The cubic law is typically used to describe its flow capacity. Figure 4 depicts the difference.

The combined nature of shear dilation and tensile parting for fracturing in the oilsands is significant in effecting the hydraulic fracturing stimulation. The fracture created is a zone with relatively large thickness. In the fracture zone, micro tensile cracks are present, but they do not connect with each other to form a continuous path. Instead, these micro cracks are embedded in the shear-dilated medium that still belongs to a disturbed continuum. As a result, the fracture zone in the oilsands still possesses certain friction or resistance to the flow, but it disturbs a much wider area. All of these characteristics are different from the conventional tensile fracture created in the hard rock. In the latter, a fracture has two continuous walls separated by a certain aperture. The cubic law is typically used to describe its flow capacity. Figure 4 depicts the difference.The fracturing

The fracturing behaviour described in the above is utilized in the CSS operation. It is an excellent example of how to proactively utilize geomechanics to enhance reservoir performance. For example, a fracture zone is formed during the

Contact us for the full extent of our papers, we typically reply within one business day.

 

REFERENCES

Agar, J.G., 1984, Geotechnical Behaviour of Oil Sands at Elevated Temperatures and Pressures, Ph.D. thesis, Dept. of Civil Eng., Univ. of Alberta, Canada.

Alberta Energy and Utilities Board, 1999, Decision 99-22: Imperial Oil Resources Limited Cold Lake Production Project Mahkeses Development. September 1999, 70 p.

Barnes, D.J. 1980, Micro-fabric and strength studies of oil sands. M.Sc. Thesis, University of Alberta, Canada.Barnes, J.D. and M.B. Dusseault, 1981, The influence of diagenetic

Barnes, J.D. and M.B. Dusseault, 1981, The influence of diagenetic microfabric on oil sands behaviour. Can. J. Earth Sci., 19, 804-818.Beattie, C.I., Boberg T.C. and G.S. McNab, 1991, Reservoir simulation of cyclic steam stimulation in the Cold Lake oil sands. SPERE, 6(2), 200-206.

Beattie, C.I., Boberg T.C. and G.S. McNab, 1991, Reservoir simulation of cyclic steam stimulation in the Cold Lake oil sands. SPERE, 6(2), 200-206.Butler, R.M., 1986, The expansion of tar sands during thermal

Butler, R.M., 1986, The expansion of tar sands during thermal resovery. Journal of Canadian Petroleum Technology. Sept.-Oct., 1986. p.51-56.

Chhina, H.S. and J. Agar, 1985, Potential use of fracture technology for recovery of bitumen from oil sands. Presented at the 3rd Int. Conf. On Heavy Crude and Tar Sands, Long Beach, California, July 22-31, 1985.Clark, J.A., 1980, The reconstruction of the Laurentide ice sheet of North America from sea level data: methods and preliminary results, J. Geophys. Res., 85, 4307-4323.

Clark, J.A., 1980, The reconstruction of the Laurentide ice sheet of North America from sea level data: methods and preliminary results, J. Geophys. Res., 85, 4307-4323.Dusseault, M.B., 1977. The Geotechnical Characteristics of Athabasca Oil Sands. Ph.D. Dissertation, Dept. of Civil Eng., Univ. of Alberta, Canada. 472 p.

Dusseault, M.B., 1977. The Geotechnical Characteristics of Athabasca Oil Sands. Ph.D. Dissertation, Dept. of Civil Eng., Univ. of Alberta, Canada. 472 p.Dusseault, M.B.

Dusseault, M.B. and Morgenstein, N.R., 1977, Shear strength of Athabasca oil sands. Can. Geotech J., 15, 216-238.Dusseault, M.B.

Dusseault, M.B. and Morgenstern, N.R., 1979, Locked sands, Q. Journal Engineering Geol., 12, 117-131.Flach, P.D., 1984, Oilsands Geology — Athabasca Deposit North, published by Geological Survey Dept. Alberta Research Council, Edmonton, Alberta, Canada. 31 p. with additional 9 maps.

Flach, P.D., 1984, Oilsands Geology — Athabasca Deposit North, published by Geological Survey Dept. Alberta Research Council, Edmonton, Alberta, Canada. 31 p. with additional 9 maps.Ito, Y., 1984, The introduction of the

Ito, Y., 1984, The introduction of the microchanneling phenomenon to cyclic steam stimulation and its application to the numerical simulator (Sand Deformation Concept), SPE Journal, 24 (4), 417-430.Kosar, K.M., 1989, Geotechnical Properties of Oil Sands and Related Strata. Ph.D. thesis, Dept. of Civil Eng., Univ. of Alberta, Canada.

Kosar, K.M., 1989, Geotechnical Properties of Oil Sands and Related Strata. Ph.D. thesis, Dept. of Civil Eng., Univ. of Alberta, Canada.Kry, P.R., 1989, Field observations of steam distribution during injection to the Cold Lake reservoir. In “Rock at Great Depth”, ed. By V. Maury and D. Fourmaintruax, p. 853-861.

Kry, P.R., 1989, Field observations of steam distribution during injection to the Cold Lake reservoir. In “Rock at Great Depth”, ed. By V. Maury and D. Fourmaintruax, p. 853-861.Kry, P.R., Boone, T.J., Gronseth, J.M., Leshchyshyn, T.H.

Kry, P.R., Boone, T.J., Gronseth, J.M., Leshchyshyn, T.H. and Reinke, R.F., 1992, Fracture orientation observations from an Athabasca oil sands cyclic steam stimulation project. CIM paper no. 92-37, presented at CIL 1992 Annual Technical Conference in Calgary, June 7-10, 1992.Kular, G.S.,

Kular, G.S., Chhina, H.S., Best, D.A. and MacKenzie, W.T. 1988, Multiple hydraulic fracture propagation in oil sands. SPE 17534, presented at SPE Rocky Mountain regional Meeting, Casper, WY May 11-13, 1988.Plahn, S.V., Nolte, K.G., Thompson, L.G. and Miska, S., 1997, Quantitative investigation of the fracture pump-in/

Plahn, S.V., Nolte, K.G., Thompson, L.G. and Miska, S., 1997, Quantitative investigation of the fracture pump-in/flowback test. SPE Production & facilities, Feb. 1997, 20-27.

Samieh, A.M., 1995, Modeling of Athabasca Oil Sand. Ph.D. Dissertation, Dept. of Civil Eng., Univ. of Calgary, Canada.

Smith, R.J., Alinsangan, N.S. and S. Talebi, 2002, Microseismic response of well casing failures at a thermal heavy oil operation. SPE 78203-MS presented at SPE/ISRM Rock Mechanics Conference, 20-23 October 2002, Irving, Texas.

Smith, R.J., et al., Cyclic steam stimulation below a known hydraulically induced shale fracture. Journal of Canadian Petroleum Technology, 2004. 43(2): p. 39-46.

Wong, R.C.K., Barr, W.E., and Kry, P.R., 1993, Stress-strain response of Cold Lake oil sands, Canadian Geotech. Journal, 30,220-235

Xu, B., Yuan, Y. and Wong, R.C.K. 2010, Modeling of the hydraulic fractures in unconsolidated oilsands reservoirs. Presented at the 44th US Rock Mechanics Symposium and 5th U.S.-Canada Rock Mechanics Symposium held in Salt Lake City, UT, June 27–30, 2010.

Yuan, Y. and G. Fung, 2008, Well injection tests and geomechanical history-matching for in-situ oil Sands development, CIPC 2008-194 presented at Canadian International Petroleum Conference held in June, 2008 in Calgary.

Yuan, Y., Yang, B. and B. Xu, Fracturing in the Oil-Sands Reservoirs. CSUG/SPE 150293 presented at 2011 CSUG/SPE Canadian Unconventional Resources Conference held in Calgary, Canada, 15-17 November 2011.

Yuan, Y., Xu, B. and C. Pamlgren, Design of Caprock Integrity in Thermal Stimulation of Shallow Oil-Sands Reservoirs, CSUG/SPE 150293 presented at 2011 CSUG/SPE Canadian Unconventional Resources Conference held in Calgary, Canada, 15-17 November 2011.