Print this page
Click to view article in PDF format
Injection of Acid Gas (CO2/H2S) into a Devonian Pinnacle Reef at
Zama, Alberta, for Enhanced Oil Recovery and Carbon Sequestration*
Steven A. Smith1, James A. Sorensen1, Anastasia A. Dobroskok1, Bill Jackson2,
Doug Nimchuck2, Edward N. Steadman1, and John A. Harju1
Search and Discovery Article #40355 (2009)
Posted February 27, 2009
*Adapted from extended abstract prepared for oral presentation at AAPG Convention, San Antonio, TX, April 20-23, 2008.
1Energy and Environmental Research Center, University of North Dakota, Grand Forks, ND([email protected])
2Apache Canada, Ltd, Calgary, AB, Canada
Since December 2006, a stream of acid gas (approximately 70% CO2 and 30% H2S) has been injected into a Devonian pinnacle reef structure in the Zama oil field in northwestern Alberta, Canada. The injection has been conducted at an average rate of approximately 750 mcf (thousand cubic feet) of acid gas per day, which includes approximately 15 tons of CO2 per day. The project includes a variety of efforts focused on examining the effects that high concentrations of H2S can have on enhanced oil recovery (EOR) and carbon sequestration operations, particularly with respect to monitoring, mitigation, and verification.
Research activities are being conducted at multiple scales of investigation in an effort to predict and ultimately verify the fate of the injected gas. Geological, geomechanical, geochemical, and engineering data are being used to fully describe the injection zone, overlying seals, and other potentially affected strata. Validating the integrity of the anhydrite sealing formation and determining the nature of potential geochemical and geomechanical changes that may occur because of acid gas exposure are primary goals of the research. Challenges in dealing with acid gas as a miscible fluid for EOR and sequestration have been identified and examined. Lessons regarding the use of acid gas for EOR and sequestration may be widely applicable, as the exploitation of deeper sour gas pools increases throughout the world.
|
|
Carbon dioxide (CO2) capture and storage (CCS) in geological media have been identified as important mechanisms for reducing anthropogenic greenhouse gas emissions currently vented to the atmosphere. Several means for geological storage of CO2 are available, such as in depleted oil and gas reservoirs, in deep saline aquifers, in CO2-flood enhanced oil recovery (EOR) operations, and in coal seams for enhanced coalbed methane recovery. Studies in CO2 capture, transportation, storage, and monitoring, mitigation, and verification (MMV) have been, and continue to be, conducted to support for the deployment of large-scale demonstrations. Understanding the fate of the injected CO2 is an important aspect of the emerging CCS technology. MMV activities are critical components of geological storage locations for two key reasons. First, the public must be assured that CO2 geological storage is a safe operation. Second, markets need assurance that credits are properly assigned, traded, and accounted for. Integrated geological and hydrogeological characterization programs that include the analysis and modeling of petrophysical, geochemical, and geomechanical properties of sinks and seals are technologies that can document the movement of the injected gases and detect potential leakage from the storage unit.
The Energy & Environmental Research Center (EERC), through the Plains CO2 Reduction (PCOR) Partnership, one of the U.S. Department of Energy (DOE) National Energy Technology Laboratory’s Regional Carbon Sequestration Partnerships, is working with Apache Canada Ltd., the Alberta Geological Survey (AGS), and Natural Resources Canada (NRCan) to determine the effect of acid gas (H2S and CO2) injection for the simultaneous purpose of disposal, sequestration of CO2, and EOR. The injection process, and subsequent hydrocarbon recovery, is being carried out by Apache Canada Ltd., AGS has developed baseline characterization data, and the EERC is conducting MMV activities at the site. The MMV activities have been designed in such a way as to be cost-effective and cause minimal disruption to ongoing oil production activities, yet provide critical data on the behavior and fate of the injected acid gas mixture and provide warning in the event leakage from the reservoir should occur.
The field validation test, conducted in the Zama oil field of northwestern Alberta, Canada (Figure 1), will evaluate the potential for geological sequestration of CO2 as part of a gas stream that includes high concentrations of H2S (20% to 40%). The results of this project will provide insight regarding the impact of H2S, in conjunction with CO2, on sink integrity (i.e., seal degradation), MMV techniques, and EOR success within a carbonate reservoir. Monitoring activities are focused on cap rock integrity, wellbore leakage, and spillpoint breach in the near-pinnacle environment.
As part of the EOR scheme, acid gas is being injected into the top of pinnacle reef structures (a process referred to as “top-down” injection) that have been depleted of oil through primary and secondary (waterflood) production techniques. Incremental oil is produced from a second well in the reservoir completed near the base of the reservoir. A third well that formerly penetrated the production zone within the pinnacle but was subsequently plugged off and recompleted into a shallower stratigraphic horizon is being used to monitor fluid chemistry and pressure (Figure 2).
The acid gas used in this project is obtained from the Zama gas-processing plant and injected into the reservoir at a depth of approximately 4900 feet (1500 meters). Approximately 12,000 tons of acid gas was injected between December 2006, when injection began, and March 2008. Injection is expected to continue for up to 15 years. Over the 4-year life of the project, between 40,000 and 60,000 tons of acid gas is expected to be injected into the pinnacle. Some recycling of this gas will occur through the EOR process, but it is anticipated that most of the injected gas will remain in the injection zone resulting in the sequestration of as much as 42,000 tons of CO2 in this single pinnacle. With over 800 pinnacle reef structures in the Zama oil field, the potential for CO2 sequestration through EOR activities is significant.
The development and execution of effective MMV operations are a critical element in conducting large-scale injection projects. Successful MMV activities will result in data sets that 1) verify that injection operations do not adversely impact human health or the environment, and 2) validate the sequestration of greenhouse gases for the purpose of monetizing carbon credits from geological storage if such a market were to be developed. There is a broad range of technologies and approaches that can be, and in some cases have been, applied to CO2 sequestration projects of various scales around the world. Early geological sequestration research and demonstration projects deployed MMV strategies that were developed based on a lack of knowledge about the effectiveness and utility of many of the applied technologies. The absence of knowledge required early projects to gather as much data as possible using a wide variety of techniques. In particular, a desire to “see” the plume of injected CO2 led to a strong emphasis on the use of geophysical data, especially 3-D and 4-D seismic, to monitor the plume. While the use of geophysical-based approaches and techniques in early projects yielded valuable results that are essential to the development of geological sequestration as a CO2 mitigation strategy, their high costs of deployment and often limited ability to identify CO2 in many geologic settings may render them as being the exception rather than the rule when it comes to developing MMV plans for future projects.
If the deployment of large-scale CO2 injection for geological sequestration is to become widespread, then MMV activities must be cost-effective. The use of existing data sets to develop background and baseline conditions should be maximized wherever possible. The use of invasive or disruptive technologies should be minimized not only to reduce costs, but also to limit the inadvertent development of leakage pathways through new monitoring wells. Where sequestration is associated with EOR operations, it is also important that MMV activities have minimal impact on commercial injection and production operations. MMV activities need to be coordinated and integrated as much as possible with ongoing and planned oil field operations. An emphasis on the collection of reservoir dynamics and monitoring well data (including the use of tracers) in conjunction with routine well operation and maintenance activities can, in some geological settings, be an appropriate and cost-effective strategy for MMV. An emphasis on cost-effectiveness and integration with routine oil field activities was the driving philosophical basis for developing the MMV plan that has been implemented at the Zama oil field.
The following techniques are being employed to monitor the effects of acid gas injection at the Zama site. The preinjection state of each of these parameters has been determined either by currently available historical field data or field activities conducted in 2005 and 2006 to acquire new baseline data:
Geomechanical Characterization
A suite of activities focused on geomechanical characterization have been performed to confirm the mechanical integrity of the reservoir and cap rock system. Historical analytical work was examined, including wireline log data that provided information on dynamic elastic properties and stress regimes; and data that allowed for the correlation of static-to-dynamic elastic properties and geomechanical simulation. Hydrogeological evaluations have shown that the reef does not appear to be in communication with adjacent reefs and can be considered a closed system (Buschkuehle, 2007); thus a significant buildup in pressure can be experienced during the injection period. Pressure fluctuations of the reservoir during waterflood or acid gas injection activities, and the irregular shape of the structure (Figure 3) can cause stress concentrations, the ultimate consequence of which may be caprock failure. Geomechanical testing and modeling will help establish the thresholds and integrity of the system, particularly the transition from reservoir to caprock, when subjected to injection of acid gas at pressures exceeding the in situ conditions. Additional laboratory tests have been conducted, including compression and sonic tests. Compression tests yield information on strength, static and dynamic elastic properties, compressibility, and stress-dependent permeability while sonic tests provide data on compressional and shear wave velocities. These data sets, along with data collected in subsequent tests that are ongoing, will ultimately form the basis for developing numerical models that will be used to assess the long-term integrity of the reservoir/cap rock system. Initial results of the geomechanical studies of the rock system in the Zama oil field indicates that both reservoir and cap are comprised of rocks that are generally characterized by high mechanical strength, dolomite and anhydrite, respectively. In general, the completed tests to determine the elastic properties of these rocks have confirmed that both reservoir and cap rock have very high mechanical strength and can sustain high stresses without experiencing significant deformations. Results of the tests also indicate that the peak strength of these rocks is much higher than the original reservoir pressure of 2100 psi (14.5 MPa). This is a positive indication that failure of the cap rock should not occur under permitted operating conditions. Geological and Hydrogeological Characterization
To evaluate and predict the long-term migration of gases injected into geological formations, an in-depth knowledge of the target injection zone and surrounding area is critical. An evaluation of the geological province, fluid flow regimes, and water quality for the area that encompasses the Zama sub-basin of the Alberta Basin was completed in June 2007. Figure 4 illustrates the geographic area that includes the Alberta Basin and highlights the regional-scale study area in the northwest corner of Alberta. This evaluation includes a brief history of the Zama oil field, a detailed accounting of the basin-scale geology and hydrogeology, structural setting and tectonic framework, and water chemistry for the Zama sub-basin. A discussion of the larger Alberta Basin is also included as it provides context with regard to hydrostratigraphic units and larger regional flow systems that pass through the smaller basins it contains.
Results of the investigation will aid in the verification of this site as an appropriate candidate for CO2 sequestration or acid gas disposal scenarios. Stratigraphically, the injection zone is well contained between massive anhydrite and shale packages that will ultimately slow and/or prevent the migration of leaked gas, should it occur. Because of the number of pinnacles in the region, the Zama sub-basin is analogous to an upside down egg crate (Figure 5). As fluid and gas migration takes place over geologic time, it is likely to travel through “inter-pinnacle” region and become trapped in pinnacles along the flow pathway. Gas migrating upwards along wellbores may be introduced to a minimum of two zones of porosity above the Keg River and may be carried along through these flow systems, each already containing “sour” hydrocarbons, and acted upon by dissolution in formation water, dispersion, residual gas saturation and mineral trapping. It is important to note that the process of gas migration in these formations is acting at geological time-scales on the order of tens of thousands to hundreds of thousands of years to move through the system and is unlikely to reach the surface.
New core was collected from a well in the vicinity of the pinnacle in March 2007. The new core is approximately 55 feet long and includes portions of the Muskeg Formation (anhydrite caprock) and the Keg River (pinnacle reservoir). This core will be used to evaluate the transition zone from caprock to reservoir rock. Additional core will be collected in 2008 from an area of the Slave Point Formation in the Zama field that has been exposed to high concentrations of high pressure acid gas. All cores will be evaluated with respect to geomechanical, geochemical, and mineralogical characteristics. The results of these core analyses will provide a basis for developing accurate models that can be used to predict the effects that large-scale acid gas injection can have on reservoir and caprocks.
Geochemical evaluations are being conducted to generate data that will hopefully provide further confidence in the establishment of the Zama field as a preferred site for sequestration activities. Geochemical models are being created using existing water quality, petrographic and mineralogical datasets. These datasets will be supplemented by ongoing field and laboratory activities. Results of this work will provide an understanding of the interaction of acid gas mixtures, formation waters, and carbonate rock systems that contain hydrocarbons and will be used in future demonstrations of this type.
Acid gas injection for the combined purpose of EOR, disposal, and CO2 sequestration is proving to be a technology that can bridge the time gap between implementation of small scale CO2 sequestration demonstration projects and full scale injection for mitigation of greenhouse gases currently vented to the atmosphere from large industrial sources. Research activities are being conducted at multiple scales of investigation in an effort to validate predictions of the ultimate fate of the injected gas. Geological, geomechanical, geochemical and engineering work is being used to fully describe the injection zone and adjacent strata. Certifying the integrity of the cap rock is a critical research area with additional tests being completed on the reef to determine the nature of potential geochemical and geomechanical changes that may occur due to acid gas exposure.
Preliminary study of the rock system in the Zama oil field indicates that both reservoir and caprock are highly favorable for sequestration of CO2 as a component of acid gas. This gives every indication that failure of the caprock should not occur under normal operating conditions. Even if failure of the anhydrite cap were to occur, fluid flow processes, dissolution in overlying formation waters, and mineralizing reactions would act to slow the movement of gas and would likely prevent any migration to the surface. Over the four years of this project it is likely that up to 42,000 tons of CO2 may be sequestered. With over 800 pinnacle reef structures of similar size in the Zama field the potential for sequestration of large volumes of CO2, as part of an acid gas stream, is significant. Apache Canada Ltd., 2003, Resource application for approval to implement an enhanced oil recovery scheme in the Zama Keg River F Pool using acid gas as a miscible flooding solvent: EUB Guide 65 Schedule 1, March 31, 2003. Davies, G.R. and S.D. Ludlam, 1973, Origin of laminated and graded sediments, Middle Devonian of Western Canada: Bulletin, Geological Society of America, v. 84, p. 3527-3546. Buschkuehle, M., K. Haug, K. Michael and M. Berhane, 2007, Regional-Scale Geology and Hydrogeology of Acid-Gas Enhanced Oil Recovery in the Zama Oil Field in Northwestern Alberta, Canada, Client Report for the PCOR Partnership. |
